my a paper sa hee ese ~ Pea we yet = wists aecr we ee Pes Seg CSTE Sse = 2 +S Lied sotoe ryaes we Wish es Sea * =O brie haipasas +s oa ee cate oe oe 5c Ase SERPS aS ores aes Stas ee Paget Oy eas Sears << Broach ee ieee" ned p vat se dae ges asi ere Jay Sain oe Epes oye See a a vs WAS tim thee wed ey bie ons o> e Poe Aad yee d ry ae Ces eee Pe gee memes +s, ’ vaca , ht See rs a Se wey : aoe seth) 2 re a pacar hes PROCEEDINGS rr OF THE American Philosophical Society HELD AT PHILADELPHIA FOR PROMOTING USEFUL KNOWLEDGE VOLUME LVII 1918 RS MOCcxxvi_ YA PHILADELPHIA THE AMERICAN PHILOSOPHICAL SOCIETY 1918 PRESS OF ERA PRINTING COMPANY CONTENTS e Pace. Interrelations of the Fossil Fuels, III. By JoHn J. STEVENSON. 1 The Archeological Significance of an Ancient Dune. By MN St PAIR ec be oe hls ce vn deb ei ew sce 49 American Sanitation in the Philippines and its Influence on the eet ey VOC Cs PUMIORE on soc eee eee ccc ees 60 A Critical Survey of the Sense of Hearing in Fishes. By G. pes a a bo nin ss oa Senko ne etnes 69 The Syriac Dialogue “Socrates.” By WM. RoMAINE NEWBOLD. 99 ‘Biochemical Studies of the Pitcher Liquor of Nepenthes. By I RII TE oo cae Sine -b win soc vb boo sleie ds 112 Twin Hybrids from Crosses of Gnothera lamarckiana and fran- _ eiscana with CG. pycnocarpa, in the F, and F,. By Grorce Oe I os ho vk s ww eh ee nese sccaguretes 130 The Art of George Catlin. By Epwin Swirt Batcu ........ 144 Parasitism among the Red Alge. By WuLLIiAM ALBERT gg oc nG ody ca ca we bee esceweedees 155 An Annotated Translation of the Part of Schweinitz’s Two Papers Giving the Rusts of NorthAmerica. By J.C. ARTHUR Ee Rt el os eve bee ce va eee eee e's 173 A Study of Some Ant Larve, with a Consideration of the Origin and Meaning of the Social Habit among Insects. By Wu- ee CTO WV a dine ve tay ee ee eee ss 293 Development of Magnetic Susceptibility in Manganese Steel by Prolonged Heat Treatment. By CHartes F. Brush ....... 344 Preliminary Notes on Some New Species of Agarics. By Geo. RT Gy ERS i te a ee 354 The Genus Galerula in North America. By Geo. F. ATKINSON. 357 Organization, Reproduction and Inheritance in Pediastrum. By R. A. Harper ...... Se es a ee oa ge eos 375 iv CONTENTS Page. Symposium on Food Problems in Relation to the War: I. Physiological Effects of a Prolonged Reduction in Diet on Twenty-five Men. By Francis G. BENE- DICT Gis se aeinb a pe ee as st cso Cee eee ae ee 479 II. Food Conservation from the Standpoint of the Chem- istry of Nutrition. By H. C. SHERMAN .......... 491 III. Some Economic. Aspects of the American Food Sup- oly. By J. Russet. Smira . .. 22.9555 4 501 On the Number of Spiral Nebulae. By Heper D. Curtis .... 513 The Nayades (Freshwater Mussels) of the Upper Tennessee Drainage. With Notes on Synonymy and Distribution. By Me IOTMANN 200 .050..0.....0 0000s eee une een 521 Brief Notes. By BENJAMIN SMITH LYMAN ............. He remarks that each dirt bed may represent a notable period of time; 2 to 3 feet of vegetable soil is the only product of very old tropical forests. The Kimmeridge Clay, at base of the Upper Oodlite, contains, according to Phillips,* a highly bituminous shale, which is utilized as fuel at Kimmeridge on the Purbeck coast. As shown in cliffs near that place, the clay, finely laminated and grayish-yellow, with remains of plants and animals, passes gradually into a bituminous shale, which is dark brown, lusterless, slightly calcareous and burns with a smoky flame. Lyell® states that this sometimes becomes an impure coal and that in Wiltshire it resembles peat. Plant remains are rare and the bitumen may be due, at least in part, to animal matter. ; The coal at Brora in Scotland belongs to the Great Oodlite or highest division of the Lower Oodlite, which in that region is a mass of sandstones and shales. The seams at Brora are thin, but one of them was worked many years ago for local use. This petty area was described by Murchison, whose measurements are (1) fossil shells, marine, quartz grains, carbonaceous matter, all cemented by - calcareous material, passing downward into a mass of compressed leaves and stems, in turn becoming shaly coal, 2 feet, 7 inches; (2) coal resembling jet, divided midway by a parting of pyritous, in- 21. C. Russell, “Second Expedition to Mount St. Elias,” 13th Ann. Rep. U. S. Geol. Survey, 1893, Part I. p. 14, Pl. XII. 3C. Lyell, “ Elements of Geology,” 6th ed.. New York, 1866, pp. 391-393. 4J. Phillips, “ Outlines of Geology of England and Wales,” Part I., 1822, pp. 127, 128. ; 5C. Lyell, “ Elements,” p. 394. 4 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. durated clay; in burning it gives off the characteristic odor of imperfect coal; the powder is brown, 3 feet, 3 inches to 3 feet, 6 inches. This is one of the few localities in Britain where coal is present in workable thickness, but the coal is inferior and no longer of even local importance. Miller® has given some notes concerning the Odlite conglomerate of Eigg, one of the Hebrides. The Scuir of Eigg is described as a mass of igneous rock resting on a pile foundation, composed of pine stems, laid crosswise. These stems — of Pinites eiggensis are transported material; they are so numerous near Helmsdale that the people collect them and burn them into lime. The tree was as abundant on the mainland of Scotland as the Scotch fir is at present. It was of slow growth but attained gigantic size. Witham’s study of the structure proved it very different from that of the Carboniferous conifers. The wood abounds in turpentine vessels or lacune of varying size, which are well defined, the minutest detail of structure being distinct. Occa- sionally Miller found a thin streak of brilliant lignite, resembling that of Brora, but in every case it was only the bark of a tree. The Lower Odlite in Lincolnshire, according to Morris,’ has soils of vegetation with well-defined underclays. In one section, bituminous clay, 18 inches thick, rests on “gray clay with vertical stems and roots descending from the overlying bed.” Another sec- tion shows the bituminous band only 6 inches thick with 7 feet of underclays containing vertical stems. At Dane’s Hill the root-bed is only 9 inches, but at Aunby Cutting, he saw two bituminous clays, of which the upper contains lignite and impure coal. Each has its root-bed below. The Inferior Odlite, at base of the Lower Odlite, has some coal in Yorkshire. Phillips* recognized two groups of sandstones and shales along the coast. The lower consists of white to yellow sand- stones and shales, with irregular seams of bad coal; the plants are cycads and ferns but equisetiform remains are in the upper layers, standing vertically as if in place of growth. A thin irregular seam 6H. Miller, “ The Cruise of the Betsy,” Boston, 1862, pp. 51-55, 71. 7J. Morris, “On Some Sections in the Odlitic District of Lincolnshire,” Quart. Jour. Geol. Soc., Vol. 9, 1853, pp. 326-331. 8 J. Phillips, “ Geology of Yorkshire,” 2d ed., London, 1835, Part I., pp. 8-10, 65, 66, 173, 174. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 5 of coal near the top has been mined at some places. The higher group contains thin irregular coal seams; one, 8 inches thick, rests on 4 feet of grit holding carbonaceous markings and at 80 feet lower .a white sandstone, associated with coal, has similar “coal pipes.” Coal seams are present throughout northeastern Yorkshire and occa- sionally become thick enough for mining, but the coal is not good. The flora consists preéminently of ferns, but cycads and conifers are abundant. Fox-Strangways and Barrow® have given additional details re- specting the east coast of Yorkshire. A‘ section on Gresthope Bay, where the Middle Estuarine Series of the Lower Oodlite consists of thin-bedded sandstones and shales, shows (1) black coaly shale, o feet, 3 inches; (2) soft, white sandstone with rootlets, 1 foot; (3) gray shale, 5 feet; (4) sandstone and shale, 3 feet, 6 inches; (5) black shale, 1 foot, 6 inches; (6) fine laminated sandstone, 1 foot, 6 inches; (7) fine laminated shale, 6 feet; (8) false-bedded sandstone, with irregular patches of coal, plants, pyrite and car- bonized wood, 21 feet. The last rests on the Millepore Series, in which rippled sandy shales occur; the impure coal at top of the section rests on the sandy floor into which the plants thrust their roots. The Lower Estuarine series is exposed at many places between Whitby and Scarborough, where it underlies the Millepore Series. A section at Blea Wyke shows a thin coal seam roofed by 30 feet of dark shale and resting on 2 feet of underclay, below which is ferruginous sandstone, 12 feet, containing great numbers of erect stems, allied to Equisetites and often 5 feet high. Two other seams, 2 and 3 inches thick and separated by 2 feet of soft sandstone, are at 18 feet below the top seam. The lower one rests on 6 feet of dark shale overlying 24 feet of false-bedded sandstone. In the Hawsker District, a coal seam, 4 inches, is at only 3 feet above the Dogger and the intervening shale contains roots. The Dogger in this district has vertical stems of Equisetites. The Middle Estua- 9C. Fox-Strangways, “The Geology of the Odlitic and Cretaceous Rocks South of Scarborough,” Mem. Geol. Surv., 1880, p. 5; “ The Geology of the OGdlitic and Liassic Rocks to the North and West of Malton,” Memoirs, 1881, p. 8; the same and G. Barrow, “ The Geology of the Country between Whitby and Scarborough,” 1882, pp. 31, 32. 6 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. rine Shale Series at Cloughton Wyke has vertical Equisetites in sandy shale and, at base, a false-bedded sandstone as in the area south from Scarborough. Judd” has given the section of a pit at Ufford, Northampton, in the Lower Estuarine sands, which shows a thin seam of lignite, below which are 3 feet of purplish clay and 3 feet of sand, both of which contain plant remains in vertical position; he considers that the manner of occurrence indicates that the plants are im situ, and that they were embedded by quiet deposition as they stood. Ken- dall™ states as result of study of clays along the Yorkshire coast, that every coal seam examined by him rests on a root bed. The resemblance of the Estuarine Series to the Carboniferous Coal Measures has been emphasized by several observers; the resemblance to those of the Cretaceous is equally marked. The deposits were laid down in shallow water at many horizons within the Odlite. Ramsay’? and his associates observed that, in their district, there is much false-bedding in both the Great and the In- ferior Odlite as well as in the Forest Marble, which has many frag- mentary fossils in its sandy layers. Even the deposits containing marine forms frequently give evidence of deposition in shallow water. Scrope’® reported that many layers of the Forest Marble Beds (Great Odlite) in the neighborhood of Bath are rippled and that they show impressed footprints of various types. Those layers contain rolled fragments of shells, corals, echini, etc., and exhibit the characteristic features of a shore deposit. According to Lyell, rippled bands of Odlite are known in broad areas and are utilized for roofing. The Lias of England is without coal, though at some localities jetified wood is abundant. The soils of vegetation in Yorkshire were described by Conybeare and Phillips :** Conybeare stated that 10 J. W. Judd, “ The Geology of Rutland,” Mem. Geol. Survey, 1875, pp. 104, 105. : 11 P, F. Kendall, in letter of May 27, 10917. 12 A.C. Ramsay, W. T. Aveline, E. Hull, “Geology of Parts of Wilt- shire and Gloucestershire,” Mem. Geol. Survey, 1858, pp. 10, 12, 14. 13G. P. Scrope, “On the Rippled Markings of the Forest Marble Beds,” Proc. Geol. Soc., Vol. 1, 1834, p. 317. 14 W. D. Conybeare, “ Outlines of the Geology of England and Wales,” 1822, p. 272; J. Phillips, “Geology of Yorkshire,” 1835, p. 66. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 7 gigantic reeds are in the cliffs near High Whitby. They appear to have been rooted in a bed of shale or slate-clay and their remains protrude into a sandstone, 5 feet thick. Those which are erect retain their shape, but prostrate stems are compressed. The tops seem to have been broken off and the woody matter has disappeared, there being only sandstone casts. Phillips gave more of detail. He reports that a Lias sandstone near Whitby contains great numbers of cylindrical plants like Equiseta, which are erect. They were broken off above and in some cases do not reach to the top of the bed. They are broken off below but commonly pass to the lower surface of the bed and, at times, the lower joints reach into the underlying shale. The conditions have led some to regard these plants as in situ, but Phillips prefers to believe that they were floated down and that they were kept vertical by the weight of their roots. The writer is compelled to dissent from this explana- tion. If the trees had been floated down stream, they would not remain vertical, even though it be conceded that the weight of their roots would keep them vertical while floating. As soon as the roots had touched bottom, the current, gentle or strong, would push the stem down stream. “Snags,” only too familiar in our western rivers, invariably point down stream. Grand’ Eury was able to determine the direction of currents in St. Etienne coal basin by means of “snags” enclosed in the sandstones. Murchison, in a brief note referring to the observations by Phillips, stated that he had discovered another locality at the same horizon, but 40 miles away and well inland. At both localities, the stems of Equisetum columnare are in the normal position and appear to be rooted in the black shale. The only fossil accompanying these plants is a fresh- water bivalve. France—tThe Jurassic deposits of France contain some thin seams of coal, which rarely have more than local importance. de Serres*® reported upon the coals of Aveyron, belonging to the Lower Odlite. The mines are on the plateau of Larzac within an area of 15 R. I. Murchison, “On the Occurrence of Stems of Fossil Plants in Vertical Position, etc.,” Proc. Geol. Soc., Vol. 1, 1834, p. 391. 16 M. de Serres, “ Des houilles séches ou stipites des terrains jurassiques, etc.,” Bull. Soc. Geol. France, t. 16, 1859, pp. 97-99, 104, 105. 8 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. not more than 60 by 200 kilometers. The only workable seam is extremely variable. The greatest thickness is in the group of mines known as Nuejols, where the seam is 70 to 80 centimeters. The center of the area is on the summit of the plateau, where, in two mines, the thickness is but 45 centimeters. The decrease continues toward the north, there being only 12 to 15 centimeters at 10 kilometers north from La Cavalerie. The lower part of the coal group is largely calcareous and the limestones have both marine and freshwater forms. The coal rests directly on black shale; the roof is similar but more carbonaceous and, at times, has a wood-like structure; it is at most 12 centimeters thick and is combustible. The coal yields a very fair coke with imperfect metallic luster. The lenticular form of the seam is distinct, for the thickness de- creases in all directions from La Cavalerie. , Austria—The Jurassic coals of Upper Austria belong to the Grestener beds at the base of the Lias. They have been described by Lipold*? and his associates. Hertle, in his notes upon the mining area of Bernreuth on the eastern side of this region, states that Czjzek’s profile shows a marine limestone between two coal seams, which contains Mytilus, Pleuromya and Pecten, and a sandy shale in the same section has Ammonites. Sphzrosiderite concretions as large as half a cubic foot are fossiliferous. These calcareous de- posits were not exposed at the time when Hertle made his examina- tion, but he saw a sandy shale with Pholadomya and Mytilus. The coal seam, which is mined, is 3 feet thick and rests on an underclay containing remains of plants. The coal looks like good coal but it has 42 per cent. of ash. Near Gresten, according to Rachoy, the coal seams are in a sand- stone group. One tunnel cut seven streaks of coal, one to 12 inches thick, while a shaft passed through 16 seams, 3 inches to 3 feet thick. The roof and floor are sometimes clay and sometimes sandstone. The thickest bed yielded a good caking coal with less than 4 per cent. of ash; the dip is about 20 degrees. Plant remains are poorly preserved but marine fossils occur in fine condition. At 17M. V. Lipold, G. v. Sternbach, J. Rachoy and L. Hertle, “ Das Kohlen- gebiet in den nordlichen Alpen,” Jahrb. k. k. Geol. Reichsant., Band 15, 1865, pp. 29-61. > STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 9 Hinterholz, Rachoy found dips of 40 to 60 degrees and one seam, 4 feet 6 inches thick, was mined. The coal yielded 66.3 per cent. of high-grade coke, used in iron-making. v. Sternbach’s section near Grossau is (1) shale, 6 inches; (2) coal and shale, 1 foot; (3) clay shale, 1 foot; (4) coal, 3 feet; (5) shale, 6 inches; (6) sandstone, 6 feet; (7) carbonaceous shale, 6 inches; (8) coal, 6 inches; (9) carbonaceous shale, 6 inches; (10) sandstone, 1 foot; (11) shale, not measured. This, like many others, closely resembles typical short sections in Cretaceous and Carboniferous coal measures. The workable seam, Number 4, has lenses of shale, so that not more than three fourths of the output is clean coal. The roof is black shale but the floor is fine to coarse sandstone. The dip is from 55 to 60 degrees and the coal seams are extremely variable; but the variations seem to be due only in part to serious disturbance. In the Pechgraben area, v. Sternbach saw 6 well-defined coal seams as well as numerous streaks of coal in the great Franz-Stollen, where the dip is 40 to 50 degrees and the rocks as well as the coal are much shattered. The sandstones have been broken into great wedges, which interlock with similar wedges of shale. The coal seams are thin and often are distorted ; but they show variations, which clearly are not due to disturbance of the stratification. The third seam, where first opened at the out- crop, consisted of numerous streaks, one to 3 inches thick; it was prospected for a considerable distance in the hope that these streaks would unite; eventually the mass became 4 feet thick but about one half of the shale still remained. The sixth seam is 9 feet thick in the tunnel, where it has 5 clay partings, in all 3 feet. But this seam, resting on shale with plant remains, is variable; in another tunnel the thickest seam is only 16 inches, while in another it is from 3 inches to 2 feet. One cannot determine in the strongly disturbed area whether the seams are lenticular or not, but there are consider- able areas, in which according to the diagrams, there was little dis- turbance and the succession is normal; in these the lens-form is distinct. The coal is somewhat inferior, having 17.2 per cent. of ash. This Pechgraben coal, according to v. Giimbel,’* shows woody 18 C, W. v. Giimbel, “ Beitrage, etc.,” 1881, p. 160. 10 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. structure distinctly after treatment with Schultze’s solution; even the minute details can be recognized. The whole region of the Lias, except locally, is much disturbed, dips of 80 degrees being by no means rare, but the coal throughout contains a high percentage of volatile combustible matter and yields a strong coke. The Grestener deposits are very largely sandstone. No freshwater fossils were noted by any of the observers but there is abundant evidence of repeated invasions by the sea; the marine mollusks belong to off-shore types. Hungary.—The importance of Liassic coals in Austria, where land conditions became pronounced, prepares one for the great de- velopment farther east in Hungary. The coal-bearing formation belongs to the Lower Lias and, according to Hantken,’® the coals are — as important to Hungary as the Carboniferous coals are to Eng- land, Belgium, France and Germany, the seams being thick and the coal good. There are five important districts: Doman-Resicza, Steierdorf-Anina, Berszaszka, Fiinfkirchen-Uralja and Neustadt- Torzburg ; the first three are in the Krassoer Comitate between 39 and 40 degrees of Longitude and between 44 and 45 degrees of North Latitude and are near the Serbian border; the fourth is near the 36th meridian and the 46th parallel, while the fifth is in Tran- sylvania, close to the border of Roumania. In the Doman-Resicza district the Lias rests on deposits of Dyas age and the dip is from 30 to 9o degrees, at times overturned. Two seams, 40 meters apart, are intercalated in the sandstone mass. The thickness of each is from nothing to nearly 3 meters and the variation is as marked along the strike as along the dip. Each has clay as floor and roof, so that the coal is apt to be dirty. The Lias sandstone in the Steierdorf-Anina district rests on Dyas. It is 160 meters thick, light in color, is almost clean quartz sand with some mica and little clay or cementing material. There is about 10 meters of other rock, including the coal seams. These thicknesses, according to Hantken, are averages only, for all por- tions of the section, especially the coal seams, are variable. Eleven coal horizons were seen, of which 5 have workable seams, one to 4 19M. Hantken, “ Die Kohlenflétze, etc., der Ungarischen Krone,” Buda- pest, 1878, pp. 44-118. ; STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 11 meters thick. Immediately above the lowest seam is a laminated sandstone, carbonaceous and containing many plants of swamp types. The upper part of this sandstone, floor to the second seam, is somewhat argillaceous and holds vertical plant remains resembling roots. The coal seams consist ordinarily of several benches, some of them good, but others worthless. Kudernatch’s section at one locality shows (1) upper bench, clean coal, 0.713; (2) earthy, im- pure coal, a mixture of Faser and bright coal, locally known as “Brand,” 0.552; (3) middle bench, clean coal, 1.025; (4) coal and shaly coal, 0.053; (5) lower bench, clean coal, 1.394; (6) impure coal, not mined, has steel-like luster, 0.154; total, 3.891 meters. The coal is in bright and dull laminz, but the bright predominates. The Hangendflétz also has the Stahlband as faux-mur. The roof and floor of all the seams are shaly sandstone with remains of plants. In the lower coal group, ferns predominate, in the upper group, cycads are abundant. These groups are separated by 97 meters of barren measures and Hantken is inclined to regard the upper one as belonging to the Middle Lias. About 74 meters of bituminous shale overlies the sandstone mass and contains streaks of coal as well as layers of iron ore. Some portions of this shale yield 3 to 7 per cent. of crude oil, from which paraffin and illuminat- ing oil are obtained. The Liassic in this district is apparently of freshwater origin ; the variations in thickness of the coal seams are due in very small part to compression, as is evident from the many illustrations given by the author. Grand’Eury, in the memoir already cited, states that the coals at Anina and Bregeda rest on soils of vegetation. At Bregeda, where the coal is anthracitic, the mur and partings have many roots in place, some of them spreading out under the coal and much divided, while others are erect and cross several layers of the shale. At Anina, where the coal is fat, woody roots are in the mur and herbaceous roots in the partings. The greatest thickness of coal is in the small area near Finf- kirchen, where the coal group, consisting of alternating sandstones, marly shales, clay shale, coal seams and layers of iron ore, rests on Rhaetic beds and underlies the marine Middle Lias. It is about 800 meters thick. Not less than 180 coal horizons have been recognized, 12 -STEVENSON—INTERRELATIONS OF FOSSIL FUELS. with 25 to 28 workable seams. The thicknesses vary greatly and, at times, the rapid increase of earthy matter renders an important _ seam worthless. Mining operations are extensive and the horizons have been correlated closely. The succession of the thicker seams from below upward is — I. and II., 24 to 36 inches, mostly unworkable ; III., IV., VI., 36 to 48 inches, one half to three fourths good coal; VII., VIII., [X., 24 to 30 inches, occasionally too thin for working ; X., 18 to 24 inches, a hard coal; XIII. to XIX., 12 to 24 inches, light, caking coal ; XX., 20 to 60 inches, coal similar to the last; XXII., o to 60 inches, often absent ; XXIV. to XXVIII., 20 to 24 inches, coal is hard. Almost the whole of the formation was crossed in the tunnel at Vasas, where 174 seams were crossed in 717 meters. The total thickness of coal is 52 meters, but one half of it is unavailable be- cause the seams are too thin or the coal is impure. 39 seams, with thickness of somewhat more than 14 meters, are marked as con- taining dirty coal. The mining districts are Fiinfkirchen, Szaboles and Vasas. In the Fiinfkirchen district, dips are 30 to 50 degrees and seams less than one foot are rarely mined ; those of more than 2 feet are usually divided by partings. The mass, numbered XI.and XII. in the Vasas tunnel, consists of (1) clean coal, 0.40; (2) shale, 0.25; (3) clean coal, 0.40; (4) carbonaceous shale, 0.45; (5) clean coal, 0.48; (6) carbonaceous shale, 0.05; (7) clean coal, 0.25; (8) carbonaceous shale, 0.20; (9) clean coal, 1.00; (10) dirty coal, 0.60; (11) clean coal, 0.60; (12) shale, 0.05; (13) clean coal, 0.40; (14) carbona- ceous shale, 0.20, resting on sandstone. The roof is shale contain- ing mullusks. Other seams are double or triple and the partings are clay or carbonaceous shale. Of the 512 beds of rock cut by the Vasas tunnel, 8 contain marine fossils; three of them being in the highest portion, 70 meters, a transition to the overlying Gryphea beds. Many marine mollusks have been obtained from the roof of coal III., the floor of XVIII. and the partings of XIII. and XXII. These are Ostrea, Gervillia, Panopea, Lima and other off-shore STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 13 genera. Faux-mur and faux-toit are common. The dip is high and the coal is tender; of that mined at Fiinfkirchen, only one per cent. is lump, pieces as large as a man’s head ; 20 per cent. is coarse, 20 millimeters or larger, while the remaining 70 per cent. is “dust”; volatile in this district is about 18, but in the Szabolcs district it is about 23. The coal is black, tender and in great part caking. The gas is low in illuminants. The flora of Fiinfkirchen consists of ferns, cycads and lycopods, some of which seem to persist throughout the Lower Lias. Leaf- bearing beds seldom overlie coal seams. In the western part of the southern area, that of Neustadt- Térzburg, the coal-bearing Lias rests on crystalline schists and con- sists of brown, argillaceous, micaceous sandstone, which, through _increasing content of plant remains, becomes darker and finally passes into carbonaceous shale, containing streaks of coal. The roof is a quartzose sandstone without trace of plants. In the eastern division, the schists are not reached and the coal group, consisting of sandstones, marls and coal seams, rests conformably on lime- stones. There seems to be but one coal seam, one to 2 meters thick, but the region is so broken by folding and faulting that a detailed section cannot be obtained. Hantken called attention in the first edition of his work to the presence of roots in the floor of coal in the Steierdorf region; Zincken,”® soon afterward, noted that near Kola, in the Steierdorf district the same horizon yields abundance of roots in vertical posi- tion. Gothan,”* having seen the root-bearing underclays associated with Jurassic coal seams on the Yorkshire coast of England, thought wholly probable that similar clays might be present in the Fiinf- kirchen area. His examination was successful though, owing to physical conditions, it covered only a portion of the district. At one locality, he found under coal VII. a characteristic underclay with irregular branching coaly markings, varying in direction and 20C. F. Zincken, “ Erganzungen zu die Physiographie der Braunkohle,” Leipzig, 1878, p. 159. 21W. Gothan, “Untersuchungen iiber die Enstehung der Lias-Stein- kohlenflotze bei Finfkirchen (Pecs), Ungarn,” Sitz. k. preuss. Akad., VIIL., IQIO, pp. 120-143. 14 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. wholly resembling roots. At another, he discovered a rhizome with : its rootlets, which made the relations of the other markings clear. “Through such horizontal rhizomes, the analogy of the Mesozoic underclay with the Carboniferous Stigmaria-beds and the recent or sub-recent reed beds is the more marked.” Roots are rarely recog- nizable in freshly exposed rock but they are sufficiently distinct after slight weathering. Gothan removed the débris for some meters at several horizons and in one day he found well-marked underclays with roots, associated with 8 coal seams. In all, he un- covered such clays under 12 seams. Spitzbergen.—Nathorst** has described a sandstone group mid- way in his Jurassic of Spitzbergen. It contains coal seams and freshwater mollusks. A coal seam is exposed on the south side of Cape Bohemian, underlying sandstone and resting on shale or shaly sandstone. Leaves abound above the coal, Ginkgo, Baera, with cycads and some ferns, and Elatides is under the coal. Bituminous sandstone with plant impressions and a seam of coal was seen at another locality, where, somewhat higher in the section, there is a soft clean sandstone with the same plants as well as freshwater mollusks, Lioplax and Unio; but still higher is a deposit with fossil wood and marine mollusks. The same group was seen on the shore of Van Keulen Bay, where the lower portion contains some thin coal seams and some clay ironstone. Siberia.—Coal seams of Mesozoic age are present in extensive areas within Siberia. Their place in the column had not been de- termined when the description cited was prepared.** They were taken to be Jurassic, but they may be in part Rhaetic. In the region between the Yenisei and Irkutsk rivers, the coal- bearing portion of the Jura, 60 to 90 meters thick, consists essen- tially of sandstone with subordinate beds of conglomerate and shaly clay. Fat and dry coals are here and boghead is not rare. A small area, about 10 kilometers square, of the freshwater Jura, near 22 A. G. Nathorst, “ Beitrage zur Geologie der Baren-Insel, Spitzbergens, und des Konig-Karl-Landes,” Bull. Geol. Inst. Upsala, Vol. X., 1910, pp. 362, 363, 365-369. 23“ Apercu des explorations géologiques et miniéres le long du Transsi- berien,” publié par le Comité Géologique de Russie, 1900, pp. 68, 86-92, 97, 179, 182, 190, 197, 199. ead ae a i Bas a eee ee - ™ ¥: ‘ gy et Ge Pe en en tI Ree ee ty Spe STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 15 Tcheremkhovo in the government of Irkutsk, shows about 65 meters of friable sandstone, in which are 3 coal seams from a half meter to nearly 3 meters thick. The coal is good, caking and has from 3 to Io per cent. of ash, though occasionally it has more, in one case, 25 per cent. The sulphur is low. The Transsiberian railroad crosses the brown coal basin of the Middle Tchoulym River between the cities of Mariinsk and Artinsk. In this basin, embracing not less than 7,000 square kilometers, the rocks, almost horizontal, are sands, argillaceous sands, gravels, sandy or plastic clays, freshwater limestones and coal. The mass is 260 meters thick and contains numerous lenticular seams of brown coal. These have small areal extent, the largest being 2 or 3 kilometers long by a kilometer wide, but the maximum thickness in the lenses is from 2 to 6 meters, though in some instances it is far greater, 14 meters at one locality. The coal ordinarily rests on clay, with an intervening faux-mur, and passes upward into a friable coaly ma- terial, resembling peat, on which rests clay or sand. This brown coal is excellent, that from the mined portions of the lenses having barely 3 per cent. of ash, and the quantity in this field is said to be “colossal.” Another area of brown coal was seen on the Upper Tchoulym River, but the quality is inferior, there being at times as much as 30 per cent. of ash. In other areas, farther west, some seams of brown coal are very thick. In the extensive region along the Angora River, the coals approach boghead in their general fea- tures and they have from Io to 34 per cent. of ash; but some of the seams yield excellent caking coal. No Mesozoic coal is reported from the Transbaikal region, where Jurassic deposits seem to be wanting; farther east, in the Upper Amur Basin, some coal seams were observed, which are thin and of no economic importance ; but on the divide between the Amur and the Zéia Rivers, Jurassic beds occupy a vast area and consist of gray or greenish sandstones with conglomerates and coal seams. Excellent coal has been obtained from a seam on the Grande-Bira River. In the eastern provinces, rocks were found similar to those of central Siberia, with several seams of coal, one to 2 meters thick and yielding anthracitic as well as caking coals. 16 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. New Zealand.—According to Hutton,™* conditions favoring ac- cumulation of coal existed in New Zealand at several horizons, but only during brief periods. The seams are nowhere thick enough to repay mining. One, 6° feet, is merely carbonaceous shale with numerous streaks of coal. Alaska.—The Jurassic area of northeastern Alaska was ex- amined by Collier,2> who made a reconnaissance survey of the Cor- win formation, probably Upper Jurassic, between meridians 163 and 165, beyond the 69th parallel. The formation is present at 100 miles farther east and notes by other explorers lead to the belief that it extends far inland; Collier’s studies were confined to the Arctic coast line. Lithologically, the formation consists of thinly bedded shales, conglomerates, sandstones and coal seams. Shales predominate, more or less calcareous, gray brown to black and vary — from mere mudstone to sandy shale. The sandstones and con- glomerates are few and seldom exceed 10 or 12 feet. The pebbles are of quartz and chert, the largest being about 4 inches in diameter. The thickness is at least 15,000 feet and the coal area within the district covers not less than 300 square miles. Mining operations were insignificant and the studies were made almost wholly upon outcrops. The coal seams appear to be in two groups, Corwin, above, and Thetis, below, separated by a great thickness of barren measures. The highest seam in the Corwin, 4 feet, 6 inches and without parting, is enclosed in black shale or shaly sandstone. Some thin beds and impure coals were seen in the interval, 1,000 feet, to the next workable seam, which is 5 feet thick and divided by two thin partings of clay. Its roof is shaly sandstone and the floor is hard clay. The next seam, 500 feet lower, is the Corwin, which was opened many years ago, but the opening was inaccessible at the time of Collier’s visit, being covered by a great snowdrift. This seam, about 1,000 feet above the bold conglomerate of Corwin Bluff, is said to be 16 feet thick, of which 7 feet are practically clean coal, 24F, W. Hutton, “Geology of Otago,” Dunedin, 1875, pp. 99, 100; Geol. Survey of New Zealand, Reps. for 1873-74, p. 36. 25 A. J. Collier, “Geology and Resources of the Cape Lisburne Region, Alaska,” U. S. Geol. Survey, Bull. 278, 1906, pp. 27, 28, 37-40. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 17 the rest being so badly broken by partings as to be worthless. The interval to the conglomerate of Corwin Bluff is filled with shale, holding at least 8 coal seams. The cliff could not be reached and the thicknesses of only three could be estimated: 4, 12 and 30 feet. _An irregular seam underlies the conglomerate and rests on sand- _ stone; it is in pockets but the coal is good in spite of the distortion. Lower seams were seen in the next 1,000 feet, not distorted, as they are in soft shale, which took up the strains. Below the lowest seam of the Corwin is a series of barren meas- ures, about 8,000 feet, in which only thin streaks of coal were seen; this overlies the Thetis group, which is reached at 6 miles east from Corwin Bluff and its highest seam is known as the Thetis, 6 feet thick, and opened many years ago. Ten seams were found in the succeeding 700 feet of shale, only two of which are likely to prove important. This necessarily imperfect record suffices to show that the quan- tity of Jurassic coal in northwestern Alaska is enormous. Some cannel is said to have been found in the Corwin group, but Collier _ Saw none. Jura-TRIASSIC. Generally speaking, it may be said that where the succession is complete, there is always a portion of the column, which is de- batable ground, and there is difficulty in determining the boundary between formations. In some cases, unconformities due to folding or to erosion offer evidence on which to base a final determination ; but such areas, though large in square miles, often mark only local adjustments, similar to those observed within formations, but which no one regards as important. Occasionally, the matter is compli- cated by lack of fossil remains. Such is the condition in a great part of Australia, where it has not been possible to divide satis- factorily the great mass between the Permo-Carboniferous and the Cretaceous. Jack and Etheridge*® recognize a Trias-Jura system in Queensland, which Jack has divided into the Ipswich and Burrum formations. 26R. L. Jack and R. Etheridge, Jr., “ Geology, etc., of Queensland,” 1892, PP. 300-39, 366. PROC. AMER. PHIL. SOC., VOL. LVII, B, JANUARY 30, 1918. 18 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. The Ipswich or newer formation occupies an area of about 12,000 square miles in southeastern Queensland and consists, for the most part, of fine conglomerates, grits, sandstones and shales with seams of coal and beds of fireclay. At a few miles south from Brisbane, W. H. Rands saw a mass of coal and carbonaceous shale, — 12 to 13 feet thick. The “best” coal is in the lower portion; some pieces of hard bright coal, free from shale partings, contained 24 per cent. of ash. A seam in the same neighborhood shows (1) coal, 4 inches; (2) shale, 2 feet, 2 inches; (3) coal with bands of shale, 1 foot, 6 inches; (4) good, hard coal, 5 feet, 3 inches; (5) shale, 3 inches ; (6) fireclay, 2 inches; (7) shale with bands of coal, 5 feet, 6 inches; (8) fireclay, 1 foot, 4 inches; (9) coal, 2 feet; (10) black band, 8 inches; (11) coal, 5 feet, 3 inches; (12) hard sandstone, 2 inches ; (13) coal, 4 inches; in all, 24 feet, 11 inches with somewhat more than 14 feet of coal aside from the thin bands of coal in the thick shale division. A piece from No. 11 had 19 per cent. of ash. A shaft in this district cut one foot of cannel and, lower down, a seam of hard and bright bituminous coal, but the sample from it contained 31.61 per cent. of ash. The coals in this district are much broken by partings and high in ash; yet, there were times and places during and in which conditions favoring accumulation of clean coal ex- isted; for a piece taken from a thin bench at one locality showed only 2.5 per cent. of ash. The type district, that of Ipswich, about 30 miles west from Brisbane, was examined by A. C. Gregory. The best seam near Ipswich is 5 feet, 6 inches thick and contains 3 to 4 feet of coal, of which the best contains about 11 per cent. of ash. Beyond Brisbane River, the seams contain comparatively little coal and that is usually poor ; but one of them becomes 4 feet, 6 inches at one locality and its coal has barely 9 per cent. of ash. This seam, however, deteriorates in all directions. This variability characterizes seams throughout the district ;a thin seam may be disseminated ina mass of coaly shale, 20 to 30 feet thick. Gregory ascertained that the quality of coal bears some relation to its distance from the northern margin of the field, the ash increasing in that direction—that is, toward the border of the great valley in which the coal measures accumulated. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 19 3 Going westward from Ipswich along the railway to Toowoomba, one reaches lower members of the formation, which have seams of ' cannel at many localities; the associated rocks are sandstone and shale. A thick seam on Blackfellow’s Creek shows (1) coal, 1 foot; (2) fireclay, 1 foot; (3) coal, 6 feet; (4) white clay, 1 foot; (5) coal, 1 foot; (6) clay shale, 1 foot; (7) coal, 1 foot; (8) a thick bed of fireclay and shale. The coal in all benches is hard cannel, so that the whole of it is available. There are some seams of bi- tuminous coal, caking, in this region but they are thin. Near Clifton - station on this railway, a shaft cut three seams at 60, 80 and 100 feet from the surface. The lowest is a rich, very hard “oil coal”; the middle seam yields good caking coal and the highest, 4 to 5 feet thick, consists of bright bituminous to dead-black “oil coal,” all being hard and tough. From the description, it would appear that the cannel is in lenses within the bituminous coal. The town of Warwick is on an outcrop of sandstone, which holds a great quan- tity of fossil wood, usually replaced with iron ore. The coal be- tween Warwick and Walloon is mostly cannel, which yields a high percentage of gas or of oil and paraffin. The only mollusk recognized is Unio. Vertebraria is in the un- derclay of a coal seam near Tivoli. From various horizons, there were collected 11 genera of ferns, 4 of cycads and 5 of conifers. The Burrum formation or lower portion of the Trias-Jura is exposed in a continuous area of about 3,000 square miles as well as in some small areas. Not much development had been attempted prior to 1892, owing to lack of railroad communication; but com- paratively extensive operations were under way near Howard, about 150 miles north from Brisbane. There W. H. Rands measured a section of 1,015 feet, representing the top fifth of the formation and _ containing 6 seams, 1 foot, 8 inches to 5 feet thick, which were a mined. These coals are of good quality, low in ash, are caking and 4 _ yield a good gas for illuminating. The coal seams generally are ir- _ regular. The fauna is scanty, a few specimens of Corbicula and of Rocellaria have been seen. The flora is almost equally scanty and is represented by a few fragmentary specimens belonging to 4 ‘genera of ferns, 2 of cycads and one conifer. 20 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. In New South Wales,”* strata seen on the Clarence River Dis- trict have some insignificant streaks of coal and the flora has Juras- sic affinities. The rocks are conglomerates, sandstones and shales, and the coal seams are unimportant. The Wianamatta beds, about 700 feet thick, are older and more argillaceous. Entomostraca occur in the upper layers. The coal seams are, at most, only a few inches thick. The Hawkesbury series, resting on the Permo- Carboniferous, is about 1,000 feet thick and consists of yellowish- white sandstone with a few beds of shale and conglomerate and some streaks of coal, without economical importance. The sand- stones show much false bedding, usually directed toward the north- east, but reversal of the currents is evidenced by occasional inclina- — tion in the opposite direction. Contemporaneous erosion of the sandstones is proved by old channel-ways filled with gravel and angular bowlders are not rare. Wilkinson thought the false bed- ding due to currents in shallow water; but he cites J. E. Tennison- Woods, who asserts that the peculiar structure is evidence that these sandstones are a wind-blown ‘formation. Plant leaves and frag- - mentary stems as well as remains of fishes are in both formations - but no remains of marine animals had been discovered. The later studies of the New South Wales geologists make it clear that the relations of the Wianamatta and Hawkesbury to the Triassic are very close. TRIASSIC. The term Trias is of German origin; on much of the continent, the system is triple or was recognized as triple, being divided into the Keuper, Muschelkalk and Bunter. In later years, the Rheetic or Infra-lias has been taken to be more closely allied to the Trias, so that now the divisions are four. Within Great Britain, Rhetic, Keuper and Bunter have been recognized, but the Muschelkalk or limestone division seems to be wanting. The several formations consist of conglomerates, shales, marls and sandstones ; the Bunter in considerable areas passes down- ward gradually into the Permian. Rock salt and gypsum are in the 27 C, S. Wilkinson, “ Notes on the Geology of New South Wales,” Dept ° of Mines, Sydney, 1882, pp. 53-55. ey ee ee STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 21 Upper Keuper. The whole mass is apparently without coal. The sandstones in very many cases are false bedded, suggesting wind- drift structures; footprints abound at numerous localities. The general features have led some English geologists to believe that the Trias of that country was formed during desert conditions. On the continent, coal was formed during the Upper Keuper as well as in the Lower or Kohlenkeuper. There appears to be none in the Mus- chelkalk or Bunter sandstone. Salt and gypsum are in the marls of Upper Bunter, footprints are numerous in the Middle, while the sandstones of the Lower Bunter are usually false bedded and foot- prints are abundant. As in England, the Bunter of north and cen- tral Germany passes gradually downward into the Permian. Sweden.—Coal is present in the Rhetic of Sweden. Hebert”® states that at Ramloesa, 4 or 5 kilometers southeast from Helsing- borg, he measured a section of somewhat more than 240 feet, con- sisting mostly of black shale, with a streak of coal, 2 centimeters, at the top, and another, 3 decimeters, at the base. The latter was mined. The shales associated with this coal yielded no plant re- mains to Hebert, but other collectors had obtained specimens, which are in the museum at Lund. Plant impressions were seen in a sandstone, midway in the section. Plant structure is distinct in the coal. Geikie,2® summarizing results obtained in this region by Nathorst, E. Erdmann and G. Lindstrom, says that the area of these Rhztic beds is about 250 square miles. They have been divided into a lower, freshwater group, containing workable coal seams, and an upper, marine group with only poor coal but abun- dant marine organisms. Clay ironstone occurs in the lower group and beds of fireclay underlie the coal seams. France—Servier® described the Keuper area of the Vosges, northeastern France. The Upper Keuper is triple; variegated marls on top, dolomitic limestone in the middle; the lower division is (1) variegated marl, 1.50; (2) micaceous sandstone, more or less 28E. Hebert, “Notes sur les grés infraliassiques de Scanie (Suede),” Bull. Soc. Geol. France, IL, Vol. 27, 1870, pp. 366-376. 29 A. Geikie, “ Text-Book of Geology,” 3d ‘ed., 1893, pp. 870, 871. 30M. Servier, “ Notes géologiques sur les mines de houille de Norroy (Vosges),” Bull. Soc. Ind. Min., t. IV., 1858-59, pp. 384-398. 22 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. argillaceous, with abundant impressions of plants and animals, 3.30; (3) laminated shale, argillaceous, with leaf impressions, 0.05 to — 0.15; (4) coal, 0.25 to 1.00; (5) black-brown carbonaceous shale, with great abundance of vegetable impressions, ill-preserved but remarkably like modern swamp plants, such as reeds and ferns, 0.50; (6) marly shales with plant impressions and Posidonia, 0.90; (7) shaly sandstone with remains of plants, 0.50; (8) silicious fetid limestone, 3.00; total, 10.40 meters. This rests on the Lower Keuper, mostly variegated marls with gypsum and salt. The dolo- mite marks a notable change in conditions; below it, the deposits are micaceous and sandy, with abundant remains of plants and animals; but above it, marls predominate and remains of any sort are rare. The coal seam varies greatly in thickness; at times it bifurcates, at others it disappears. These variations are not due to disturbance as the dip is less than 3 degrees. Kidneys of dark calcareous iron ore are in the coal, now concentrated under the roof but again scattered throughout the seam. Pyrite is abundant. Where thickest, the seam is triple, showing (1) upper bench, variable, con- sisting at times of alternating bright and dull laminations, when the coal is rejected as it burns badly and is not reduced to ash; commonly, however, it is brilliant black and an excellent fuel; (2) middle bench, not always present; its coal is glossy black, is almost uniform, burns well and is reduced to red ash; it encloses vegetable remains, some of them root-like; (3) lower bench, has brilliant black coal, yielding a brown powder. The quality and thickness improve toward the north. At the south, near La Marche, Romain and Talliancourt, it is replaced with clays containing great numbers of tree stems. Mining begins farther north near la Rouville and Croinville, where the thickness is 0.15 to 0.30; at Norroy, it becomes 0.40 to 0.80, but at Gemmalaincourt and Parey it is 1 meter. Fragments of shale, quartz and sandstone with rounded angles occur occasionally in the coal. The lenticular form of the seam is distinct. The same horizon has been recognized at widely separated locali- ties in France, though coal is rarely present. Rouville,** describing 81 P. de Rouville, Comptes Rendus, t. 48, 1857, pp. 696-608. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 23 the Trias of Aveyron and Herault, states that the abundant coaly __ impressions of plants in Keuper beds had induced many to search for coal but without success. Grand’Eury saw some coal in the Keuper of Nice which contains stems of Equisetites and in the shale are rootlets of these plants. At Gemmalaincourt, he found Equtse- tites roots in the underclay but bark and seeds are in the coal. Germany.—The Lower or Kohlenkeuper contains at many places in Germany the Lettenkohle, which Credner*? describes as a car- bonaceous clay, filled with plant remains and at times passing into impure coal. Near Siwierz in Poland there are 3 beds 30, 50 and 80 inches thick. Sandberger** published records of numerous sec- tions of Triassic deposits obtained in Unterfranken of Bavaria. He offers no comments, but the records suffice to prove irregularity of deposit. Two Lettenkohle sections near Wurzburg exhibit the triple structure. That obtained between Wurzburg and Rothen- dorf shows at top of the middle division a yellow fine-grained sand- stone with many erect roots, while at base of the division is a zone with abundant remains of plants. This rests on the Hauptsand- stein of the lower division, part of which is diagonally bedded. No coal is present. In the other section, between Wurzburg and Schweinfurt, plant remains abound in both the upper and the middle division, but coal is wanting ; the top layer of the Hauptsand- stein is argillaceous sandstone with many roots, while at the base is a fine-grained sandstone with irregular layers of pulverulent coal. A section of the Krainberg gives these details respecting the middle division: (1) clay shale, with Lettenkohle at base, 3.66; (2) shale, 0.15; (3) sandstone with roots, 1.18; (4) clay shale, 1.18; (5) Lettenkohle and plants, 0.70; (6) ochreous limestone, 1.32; below which to the base are sandstone, ochreous limestone, clay shale and sandy shale, all apparently without coal. The sandstone, No. 3, is the root bed for plants which produced the impure coal above it. The conditions seem to have been much the same at all localities where Lettenkohle exists. The coal is irregular in occurrence and 32H. Credner, “ Elemente der Geologie,” 8te Aufi., 1897, p. 535. 33 F. vy. Sandberger, “ Die Lagerung des Muschelkalk- und Lettenkohlen Gruppe in Unterfranken,” Verh. Phys. Med. Gesells. Wiirzburg, Band XXVL, 1893, PP. 200, 205, 206. 24 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. usually is of little value. The influx of foreign matter into the petty swamps was too great to permit accumulation of clean coal; but root-bearing underclays and soils of vegetation without coal are characteristic features. v. Giimbel®* states that Lettenkohle from Guildorf in Wiirtemberg and Schweinfurt in Franken gives a weak brown tint to solution of caustic potash; it is easily decomposed by Schultze’s solution and woody structure is distinct in the residue. Rhetic coal from near Bayreuth reacts to Schultze’s solution as does Lettenkohle. Many layers of this coal appear to consist almost wholly of pollen exines. Austria—An important area of Triassic coals is in Upper and Lower Austria; these belong to the Lunzer beds of the Upper Keuper. A less important area is in Siidtirol, where coal is in the Wengener beds at the base of the Keuper. = The Upper Keuper area was studied by Lipold** and his associates. The Triassic deposits are in the interior of the northeastern Alps and they have suffered more from disturbance than have the Liassic beds of that region. Lipold reports that near Baden, on the eastern side of the area, the coal and shale are so crushed and intermingled that definite sections cannot be made and that all attempts to obtain merchantable coal have failed. No mollusks were seen but Calami- tes arenaceus and Pterophyllum longifolium are not rare. Hertle found only unimportant seams in the Lunzer sandstones near Ramsau; but in Kleinzell, where the sandstone is much dis- torted, 3 thin seams were seen, all marked by extreme variations in thickness, which seem to be due to compression during folding. At Lilienfeld on the Traissen River, the dip is from 40 to 70 degrees and the coal seams, being between sandstones, have been distorted seriously. The thickness of one seam varies from one inch to 9 feet within a short distance. The Lunzer sandstone is distinctly of — freshwater origin in this district, but it is between the Opponitzer above and the Goslinger below, both of them calcareous and con- taining marine fossils. The workable coal seams, 4 and 2 feet thick, 34C, W. v. Giimbel, “ Beitrage, etc.,” p. 160. 85 M. V. Lipold, G. v. Sternbach, J. Rachoy, and L. Hertle, “Das Kohlen- gebiet in den Nordostlichen Alpen,” Jahr. k. k. Geol. Reichs., Band 15, 1865, pp. 62-1590. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 25 are in a mass of black shale, about 70 feet thick. Egquisetum columnare is abundant in the black shales and beautiful specimens have been obtained from the roof of the upper seam. This seam, 4 feet thick where first seen, is from 3 to 24 feet. Occasionally it divides into two or more benches, of which only one is persistent. The coal of Lilienfeld and Kleinzell yields 72 to 74 per cent. of good coke and ash is from 8 to 14 per cent. in the raw coal. Hertle examined the area near Kirchberg on the Pielach River where the black shale mass has 4 seams of coal. This shale is 40 feet in one tunnel, 48 to 60 in another, while in a third it is not less than 100 feet. In one tunnel, the middle seam is 72 feet above the lower one; followed westward, the interval becomes 50, 30 and 18 feet. A similar convergence is that of the middle and upper seams, which actually unite with increased thickness. These rela- tions existed before disturbance occurred. Dips in this district are from 40 to 70 degrees. The coal throughout is tender and caking, giving 67 per cent. of good coke; but the ash is high, averaging 15.8 per cent. In the Rehgarten area, the coal is cleaner, having only 9 per cent. of ash. Here distortions of the rocks are few but other troubles are encountered ; the seams thin away and frequently they pass into carbonaceous shale. Hertle’s descriptions make it clear that the seams are lenses, sometimes joined by carbonaceous shale, but at other times wholly separate. The “horseback” seems to be a feature here, as in the older as well as in the newer coals. One tunnel reached sandstone, with no admixture of clay or coal, at 480 feet from the mouth. It was pushed through the rock and again reached the coal. The lower seam at Loichgraben yields a good coal, but that from the upper seam has 52 per cent. of ash, though it looks like excellent coal. Rachoy found plant-bearing shales as roof of coal seams near Lunz and he says that, near St. Anton, a bituminous limestone is the roof in some mines. The coal at several localities is good but at others the ash is very high, while the coal externally resembles the best in the district. This area is on the westerly side of the Lunzer region and, in most cases, the seams are thin. Zincken** states that plant-bearing shales are the roof at many 36 C. F. Zincken, “ Erganzungen, etc.,” 1878, pp. 110, III. 26 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. localities ; that the coal seams are distinctly lenticular in some of the important districts, and that the coal is caking at some places, ~~ non-caking at others. The Wengener beds are at base of the Keuper and rest on ‘the Muschelkalk. Keyserling*? found coal within these beds, west from Cordeville Valley on the southeasterly slope of Mt. Cordai in south Tyrol. The rocks are alternating tuff sandstones, red, green and brown clays and marls, interrupted by beds of limestone. All yield so readily to the weather that a detailed section cannot be made. The Hauptflotz, locally regarded as “ workable,” is from 4 to 5 deci- meters thick and is well exposed in the bed of a stream, where it rests on dark limestone; elsewhere, it is frequently enclosed in clay and sandstone. The coal is laminated, some of it resembling brown coal but other portions are much like stone coal. The transforma- tion is so far advanced that no trace of organic structure can be recognized by the naked eye, but the mode of occurrence convinced the author that it was derived from water-loving plants. The quan- tity of pyrite is remarkable. Coal rarely occurs at this horizon. The Lunzer horizon was recognized by Lipold** in Carniola (Krain) who saw near Idria coaly shale with streaks of coal, but he could discover no definite seams. Hungary.—Hantken*® reports that in the Fiinfkirchen region of Hungary a sandstone formation, 620 to 950 meters thick, underlies the Liassic coal complex conformably. Its coals appear to be local and in most cases they are too thin to be mined. Fossils are not abundant; at one locality, Zamites, Palissya and Thawmatopteris have been collected; another yielded Cardinia and Acrodus. This assemblage is accepted as evidence that the mass is of Rhetic age. United States—tTriassic deposits of the Atlantic border extend in detached areas from Massachusetts to North Carolina. No coal of economic importance has been discovered north from Virginia, though thin streaks have been observed in Massachusetts, Rhode 87H. G. Keyserling, “Ueber ein Kohlenvorkommen in den Wengener Schichten der Siidtiroler Trias,” Verh. k. k. Geol. Reichs., Jahrg. 1902, pp. 57-61. 38M. V. Lipold, Jahrb. k. k. Geol. Reichs., Band 24, 1874, p. 445. 39 M. Hantken, “ Die Kohlenflétzen, etc.,” pp. 104, 105. Ee SE Sn een ee a ve i li eR STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 27 Island and Pennsylvania. McCreath*® states that P. Frazer had found coal in Triassic beds of York County, Pennsylvania, but neither he nor Frazer in his York County report gives a descrip- tion of the deposit. According to McCreath, the coal is deep black, with pitchy luster, brittle and with conchoidal fracture. The proxi- mate composition is: Water at 225° F., 4.310; volatile, 18.482; fixed carbon, 74.358; sulphur, 0.528; ash, 2.322. There is no tendency to cake and the gases burn with non-luminous flames. The dried coal absorbs water with great avidity, so that within a few hours it re-absorbs about 63 per cent. of the water originally present. - The important region known as the Richmond coal field is reached at a little way north from-the James River in Virginia. Mining operations were begun a century ago and for many years they were on extensive scale. Irregularities in the seams and the many faults made mining costly and the local coal was displaced by anthracite from Pennsylvania. Operations now are unim- portant. Fontaine,** in the introduction to his descriptions of fossil plants obtained in the Richmond and adjacent areas, gave a synopsis of the relations. The Triassic rocks occupy several areas in a belt extending from Rhode Island to South Carolina. The most westerly area, termed the Palisade, is almost continuous from the Hudson River across New Jersey, Pennsylvania and Maryland to about 75 miles southwest from the Potomac River in Virginia; it is without coal. The small area of Buckingham County, Virginia, is east from the last and like it is without coal. The Dan River area, still farther east, is in Virginia and North Carolina; it has some coal in the latter state. The Cumberland (Farmville) area is small but has some coal seams of local importance. The Richmond, 30 miles east from the Cumberland, is the last in Virginia, but the Deep River, still farther east, is in North Carolina and extends to the South Carolina border. Red beds prevail in the western areas but they are insignificant in the Cumberland and Richmond areas. Fontaine recognized three 40 A. S. McCreath, Second Geol. Survey Penn., Report MM, 1879, p. 103. 41 W. M. Fontaine, “The Older Mesozoic Flora of Virginia,” U. S. Geol. Survey, Mon. VI., 1883, pp. 1-7, 12-16, 32, 45, 79. 28 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. distinct groups in the Virginia areas: the upper group, consisting of a loose granitic sandstone or sandy shale, containing no coal but much lignite, resembling jet; silicified wood is not rare; a middle group, coal-bearing, with a large proportion of black shale; a lower group, sandstone and shale. The sandstones of the lower group are not easily distinguished from the underlying granitoid gneiss and are 100 to 600 feet thick in the Richmond area. The middle group is 100 to 200 feet thick in the same field, where it usually has two thick seams of coal—but the number, thickness and quality vary greatly. At many places the roof is a plant-bearing shale; Equi- setum rogersi is usually associated with Macroteniopteris and its — ‘casts are present in the coal. Schizoneura occurs in the underclay of the main seam. The plants described by Fontaine are conifers, cycads, equiseta and ferns. Shaler and Woodworth*? applied names to Fontaine’s groups; the Chesterfield or upper group is 2,500 feet thick and consists of sandstone above, shales below; the Tuckahoe, equivalent to the middle and lower groups, consists of the coal measures, 500 feet, more or less, sandstones and shales, 0 to 300 feet, and bowlders, o to 50 feet. The Richmond field was discussed many years ago by geologists, who studied it when the mines were still in operation.* It is well — 7 to summarize the statements of each observer as the conclusions reached by them have been regarded as not in agreement and they appear to be in some respects contrary to those reached by observers who have studied the region since mining operations practically ceased. | Taylor reponed that the deposits occupy a narrow trough, which deepens so rapidly toward the median line that coal mines are pos- 42 N. S. Shaler and J. B. Woodworth, U. S. Geol. Survey, roth Ann. Rep., Part II., 1890, p. 423. 43 R. C. Taylor, “Memoir of a Section Passing through the Bituminous Coal-Field near Richmond, Virginia,” Trans. Geol. Soc. Penn., Part L, 1835, pp. 275-207; W. B. Rogers, “ Reprint of Annual Reports on Geology of Vir-— a ginia,” 1884, pp. 62-69; “On the Age of the Coal Rocks of Eastern Virginia,” Reps. Amer. Asso. Geol. and Nat., 1843, pp. 298-316; C. Lyell, “A Second Visit to the United States of. North America,” 2d ed., London, 1850, pp. 281-287. -STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 29 _ sible only on the eastern and western margins. The maximum a thickness of coal, as far as can be ascertained, is near the middle _ of the eastern border, whence it thins toward the north and the south. The coal in all mines, of which Taylor gives measurements, is near the base of the section and rests on the granite or is separated from it by, at most, a few feet of shale. The overlying rocks, for about 400 feet, as cut in shafts on both sides of the trough, are mostly grits, sometimes conglomeratic, with interstratified gritty micaceous, carbonaceous or argillaceous shale. : ; q In the northeastern part of the trough, he saw two seams, 5 and 3 to 4 feet, separated by 10 or 12 feet of slate and about 10 feet from the granite, there being a thin seam in the latter interval. On the northwestern side, the seams are 30 feet apart and are 6 to 16 and 4 to 8 feet thick. These are said to unite farther north. The lower seam, of rather inferior quality, rests on the granite. On the eastern border, the Chesterfield shaft shows (1) coal shale, 6 feet, 10 inches; (2) coal, 5 feet, 6 inches; (3) coal shale, 3 feet; (4) coal, 1 foot, 6 inches; (5) hard grits, 2 feet, 6 inches; (6) shale and thin coal, 2 feet, 6 inches; (7) coal, 7 to 40 feet; (8) granite. The lowest coal has some variable partings. The sections on this side of the trough are much alike; but the coals, 4 and 6, are not always present and not infrequently some shale was seen between the coal and the granite. As the mines had been worked extensively prior to Taylor’s visit, he had opportunity to examine considerable spaces from which the coal had been removed so that the underlying granite surface was exposed. Not rarely a boss of granite rose through the lower _ division of the seam; in such cases, the work was usually abandoned ; 4 _ but occasionally a drift was carried around the boss and entered a body of coal, filling a hollow, 50 or 40 feet deep. There is no parallelism between top and bottom of the seam. The roof is irregular, rising and falling, and the depressions sometimes reach the floor, but they never conform to the irregularities of the granite . surface. In spite of these irregularities, the lamination of the coal _ is wholly undisturbed. The lower part of the seam is less clean __ than the upper, but the coal is fat and coking throughout. Rogers was studying the region at the time of Taylor’s visit. 30 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. His report, published in 1836, contained a brief statement which adds important observations while confirming those made by Taylor. He discovered that the overlying sandstone group apparently over- laps the coal measures and that the lowest coal seam is separated from the granite in most cases by only a few feet of shale. The coal thickens toward the center of the basin and, as a Anes: the higher seams are the best. In the Midlothian and several adjacent mines, there is ample evidence to prove that the coal accumulated in saucer-shaped basins to the thickness of 40 or 50 feet, while on the eminences of the same floor it is thin. On the south side of the James River, the River pit was abandoned when the granite floor rose almost to the sandstone roof. Near Tuckahoe, on the north side of the river, the coal was found central in a small, isolated, cup-like depression. This coal rose gently in all directions from the shaft and thinned from 5 to 2 feet toward the edges of the shallow basin. This is several hundred feet in diameter and its strata vary little from the original nearly horizontal position. “Everything lends countenance to the opinion that the surface of the primary rock, previous to the deposition of carbonaceous matter, was a valley of rolling outlines, occupied by hollows and elevations, causing the first layers of matter, which were thrown down, to be deposited in greater thick- ness in some places than in others. As the lowest coal seam is separated from the crystalline rock by only a very few feet of shale and in some cases by none at all, it appears likely that the distri- bution of the coal was made unequal in thickness from the very commencement.” | In his later memoir, discussing the relation of the plant remains, Rogers stated that the most abundant plants are Equisetum columnare, Teniopteris, and a large species of Zamites. These occur in vast numbers immediately upon the coal or interlaminated with it. They are accompanied by Calamites, Pecopteris and Lyco- podites. The Equisetum is so abundant, at times, as to give a coarse coal consisting of alternate laminations of coal and shale with occa- sionally 30 laminations to the inch. Ferns are rare, aside from the great Teniopteris. The only animal remains are those of fish and some teeth supposed to be reptilian. The fish remains are in dark STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 31 shale associated with the coal; but scales along with teeth and plant impressions were seen at times in the upper part of the coal itself. Rogers saw nothing answering to the Stigmaria-clay of the Car- boniferous. These descriptions by Rogers make clear that a faux- toit is the ordinary feature ; while the presence of animal remains in the coal indicates existence of pools on the swamp surface. Under- clays are in this field, but they do not hold Stigmaria, for Lepido- dendron and Sigillaria had become extinct. Lyell visited this field in 1845. He was much impressed by the fact, already noted by Rogers, that stems are found so often erect and compressed vertically ; he could think of no reason to doubt that the greater number of such plants, in beds above and between coal seams, and which he saw at localities miles apart, had grown where they are now enclosed in sand or mud. The great coal seam rests at times directly on the granite, but at others is separated from it by an inch or two of shale. He was inclined to think that the absence of deposits between the coal and the granite may be due to disturb- ances, which were considerable, as shown by the extensive faults. j Mining operations ceased at nearly all localities about 50 years ago and the old mines, abandoned, soon became inaccessible. A long interval passed before new studies were made and few* of these dealt with details respecting the coals. Fontaine’s detailed stratigraphical work was done near Clover Hill in the southeastern part of the field, where some work was going on at the time of his examination. There he found thick deposits between the coal and granite and assigned to them a thickness of 100 to 600 feet. Clif- ford stated that in outlying districts of the Richmond basin there is only one coal seam, usually of great thickness and separated from the granite by a thin bed of shale, often not more than a few inches. This refers to the northern part of the field. It should be noted here that the earlier observers regarded the benches of coal as separate seams. 44 W. M. Fontaine, “ The Older Mesozoic Flora of Virginia,” U. S. Geol. _ Survey, Mon. VI., 1883; W. Clifford, “Richmond Coal Field, Virginia,” _ Trans. Manch. Geol. Soc., Vol. XIX., 1888, p. 320; I. C. Russell, “ The Newark System,” U. S. G. S. Bull. 85, 1892, pp. 38-40, 63; N. S. Shaler and J. B. ; _ Woodworth, “Geology of the Richmond Basin, Virginia,” 19th Ann. Rep. _ U.S. G.S,, Part IL, 1899, pp. 423-426, 429, 483. 32 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. Russell asserts that the coal seams of the Richmond basin are irregular and greatly disturbed by faulting. They are not continu- ous though they are approximately at the same horizon. He regards. them as overlapping lenses, individual deposits thinning away. A thin seam in one mine may be the important one in another. As to the interval to the granite, he cites O. J. Heinrich, who in 1879 re- ported that at Midlothian the coal is at 570 feet from the granite; 4 also Fontaine, who in 1883 stated that the interval at Clover Hill is 250 feet. Russell suggests that the luxuriant subtropical vege- tation of these Triassic lowlands has its nearest modern analogue in the fern forests of New Zealand. The ground must have been covered with ferns, above which rose equiseta and the great ferns with palm-like leaves ; cycad forests with pines of Araucarian type covered the upland. Shaler and Woodworth report that the lower barren beds, aden : 4 lying the coal measures, are not always recognizable with certainty ; “4 sometimes the barrenness may be due to lack of coal accumulation __ at the locality, but there are places where the coal group is fully developed and where a considerable thickness of barren rocks was seen. These authors offer no explanation of the origin of the bowlder beds occasionally observed at the base of the section. They consist of granitic bowlders with a partial bedding of reddish gritty sandstone. Plate XXI. of the report illustrates well the disin- tegration of the granite, which proceded deposition of Trias in this basin. This is remarkably similar to conditions observed by the writer in central France between Aurillac and Decazeville, where such disintegration is shown at many places. In the Decazeville basin, this preceded the deposition of the Coal Measures and the accumulations were mistaken for deltas by several observers. Some have supposed that the great variations in thickness of the Richmond seams were caused by pressure during disturbance; but there appears to be no reason for resort to this explanation. Such swelling and contraction of seams is certainly common enough in disturbed regions, but there the structure of the coal is changed; it is exceedingly tender or it is rolled into flakes like pastry. But in the Richmond basin the lamination, according to Taylor and % according to observations by the writer, is undisturbed in locali- STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 33 4 that the varying interval between the coal and the granite floor is likewise a result of disturbance. This suggestion may, perhaps, E prove good for some localities but to the writer it seems unnecessary to resort to that hypothesis; Rogers’s suggestion is far better, that _ the deposits were made on an irregular surface. This accords with _ the conditions observed in North Carolina as well as in Virginia. = Two Triassic areas are in North Carolina; the Dan River, at the _ northwest, is without coal in Virginia but has some irregular de- a posits in North Carolina; the Deep River, at the southeast, begins q near the Virginia line and extends as a narrow strip southwest- _ _wardly into South Carolina.‘ / 4 Emmons’s section in the Deep River area shows a triple struc- E ture: Upper red sandstones and marls ; Coal measures, slates, shales and drab sandstones ; Lower red sandstone with conglomerate at _ base. The red rocks, wanting in the Cumberland and Richmond _ areas of Virginia, reappear here on the southeasterly border. The _ middle group, about 1,200 feet thick, has fine-grained sandstones _ which frequently are rippled; the coal seams are few and very ir- _ tegular but some of them have been opened. Russell states that at _ Egypt a shaft reached, at 422 feet from the surface, a coal seam a showing (1) black shale; (2) coal, 2 feet; (3) black band, 1 foot, 4 4 inches; (4) coal, 1 foot, 1 inch; (5) slate, 6 inches; (6) coal, 7 _ inches. Another seam, 25 feet lower, has black band roof and floor 7 and is one foot thick ; the upper seam has black shale roof and floor. Both are irregular in thickness and Russell asserts that there is no ' feason to suppose that they are continuous in any considerable area. __. The coals are indefinite within the Dan River area. Emmons _ reports that, near Leakesville in northern part of the area, a coal _ seam shows: (1) coal, semibituminous, 2 to 3 feet; (2) micaceous "shale, 2 feet; (3) coal, shaly, 1 foot, 6 inches. This is very near _ the base of the coal group. The lowest rock at the southern extrem- a 45E. Emmons, “Geological Report of the Midland Counties of North _ Carolina,” 1856, pp. 228, 230, 235, 256, 257; I. C. Russell, Bull. 85, p. 41; W. a C. Kerr, “ Report of the Geological Survey of North Carolina,” Vol. L., 187s, _ p. 143; R. W. Stone, “ Coal on Dan River, North Carolina,” Bull. 471-B, 1812, _ BP. 5, 6, 16. a PROC. AMER. PHIL. SOC., VOL. LVII, C, JANUARY 30, 1918. 34 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. ity, near Germanton, is a conglomerate of angular fragments of granite and gneiss, containing roots of silicified tree-stems penetrat- ing and branching in the deposit. The stems are very abundant just above the conglomerate, so abundant as to suggest that they are remains of an ancient forest. Most of them are prostrate and occa- sionally one finds the roots: converted into lignite. The great a abundance of stems near Germanton in Stokes county impressed — - Kerr, who says “the public road being in a measure obstructed by the multitude of fragments and entire trunks and projecting stumps of a petrified Triassic forest ; and similar petrifactions are abundant in the Deep River belt, occurring in this as in the other among the sandstones near horizons of the coal.” . Stone’s examinations led him to assign a thickness of about q 7,800 feet to the deposits within the Dan River area, where the mass rests on Archean gneiss. The zone of carbonaceous shale with _ coal is 250 feet thick and just below it, at about 1,000 feet from the base, is conglomerate with subangular fragments, whichis absent from the northern portion of the area. The roots and bark of the — silicified stems within this mass in some cases have been converted q into lignite. Shafts have been sunk in many places but usually only _ black shale has been found. At one place, 37 inches of such shale — with much coal was found. The Leakesville deposit is insignia 7 and its area is but a few square rods. 4 Triassic rocks are exposed in very many localities west from | the 105th meridian to the coast but they appear to be without coal — in both the United States and in the Dominion of Canada. a Mexico.—But coal is present in Triassic deposits of the Santa — Clara field on the eastern border of Sonora, Mexico. Dumble*® — has given brief notes respecting the locality. The Rhetic age of — the deposits was recognized by Newberry and Fontaine after study of the plant remains. The region has been disturbed greatly by ig- q neous rocks, which have metamorphosed the coals. The heavier a sandstones are uniform and are moderately coarse conglomerate q grits, which have a few fragments of silicified wood and occasional — imprints of stems. The shales and finer sandstones are excessively 4 46E, T. Dumble, “Triassic Coal and Coke of Sonora, Mexico,” Bull. 4 Geol. Soc. Amer., Vol. X., 1900, pp. 10-14. e STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 35 variable, sandy shales change abruptly into coarse massive sand- stone or into clay shale. The shales generally are rich in well- preserved remains of plants, which, according to Fontaine, are allied to those of Virginia and North Carolina. The more massive slates hold silicified stems and branches of shrubs, while the finer- grained sandstones have tree-trunks up to one foot diameter. No false bedding was observed in the sandstones. Coal seams are numerous, each prominent slate bed having one or more; but in all cases these are irregular. Near San Marcial, southeast from the area of detailed examination, much work had been done on supposed anthracite, which proved to be only black slate; but at localities north and northwest from the mining center two seams are known, 8 and Io feet thick. Much of the thicker beds is composed of coal with concentric structure, “shelling out into eggs of greater or less hardness.” The coal has been affected by the igneous rocks and usually it is a hard anthracite, though occasionally it is coke. In two im- portant openings on the coke, igneous rock is the roof; in another, it is the floor; but other pits show no igneous rock anywhere near the coke. In one seam of anthracite, there are pockets of coke near the middle, while in a seam of coke pockets of anthracite were found at the bottom. In several beds divided by partings, coke prevails in some benches, anthracite in others. The proximate composition of the anthracite is: Water, 4 to 8; volatile, less than 5; fixed carbon, 76 to 85; ash, 4 to 8 per cent. SoME CHEMICAL FEATURES OF THE COALS. Coals of various grades are present in the Jura and Trias. _ Lignite and bituminous coal are present in the Lower Odlite of Great Britain within practically undisturbed rocks and at nearly the same horizon; while high-grade bituminous coal prevails in the Lias of Austria and Hungary, where the rocks have suffered severe disturbance. The Lower Lias of Siberia yields high-grade bitumi- nous in the Tcheremkhovo and Grande-Bira fields but typical brown coal in the great Tchoulym region, where the strata are little dis- turbed and the rocks are only slightly consolidated. The Jura- 36 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. Triassic coals of Queensland are high-grade bituminous as are ares : from the Upper Trias of Austria and Virginia. Cannel has been reported from the Jura of Alaska, but tt Collier saw none. Cannel, however, is certainly present in the Steierdorf- Anina field of the Hungarian Lias. Hantken has given the proxi- mate analysis of two samples from the Hauptflétz, which show — DERBER 265 i ei aera e ees 1.10 Volatile .....cicsesses tee soe S307) 2.60 55.98 a WOR poe s kh tat caw ne see eiie 14.67 Fixed carbon ......:ssecsue «1 Gag % 22.00 44090 The cannel is evidently in lenses, as at other localities this seam has only bituminous coal. In the same field, the Middle(?) Lias has a great mass of black shale, portions of which yield from 3 to 7 per cent. of crude oil, from which paraffin and illuminating oil are obtained. In Siberia the Lower Lias coal of the Angora River field is mostly of boghead type, while in the Ipswich or upper Jura- Trias of Queensland cannel or “oil coal” is present in a large area. Jack has given three analyses of the material: Ash. Volatile. Fixed Carbon, Blackfellows Creek... .........20..-0.026. 17 56 er Clifton; top Sean ys sc se ce ee cess es | 10 53 46 Clifton, lower sata: 2. vias bese sess | 16 55 44 The seams at Clifton are separated by a considerable interval, which holds a seam of caking bituminous coal. Cannel prevails in the Walloon district where some of it is rich, that at Jimbour yielding about 37 gallons per ton. The coals vary greatly in tendency to cake. Collier reports that none of the Alaska coals tested for him gives a coke. His samples, however, were collected mostly from outcrops, where leaching had been energetic during a long period. “Crop coal,” even in the Connellsville region of southwest Pennsylvania, yields only a wretched coke. In Austria and Hungary many seams have caking coal but that from others is non-caking. In Siberia, the coals of the Tcheremkhovo and Grande-Bira fields are caking but that of the great Tchoulym field gives only pulverulent coke. The Jura-Trias coals of Queensland are caking in some instances, non-caking in STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 37 others. The Upper Trias coals of Austria are usually caking and _ those of the Richmond area are always so. Apparently no relation exists between proximate composition and tendency to cake. No reference to the presence of resins in coals of the Jura or Trias is made in any of the works to which the writer has had access. One observation by Witham, cited by Miller, bears upon the subject. In studying silicified stems of Pinus eiggensis from the Lower Odlite of Scotland, he discovered that the wood abounds in 9 _ turpentine vessels or lacunz, well defined and varying in size. Mineral charcoal (Fusain, Faserkohle) is a characteristic fea- ture throughout. At times, it forms thin partings in seams, but at others it is an important constituent of thicker partings, where its abundance suggests that the partings are merely residues from a considerable mass of peat. Occasionally it is in lumps, embedded in the coal or in a clay parting. Spheerosiderite or clay ironstone is reported by all except a very few observers. It is present in the coal, in the underclays, and is scattered in the other rocks, while occasionally it is in layers of varying thickness. At times, it replaces the stems of trees or frag- ments of wood. Black band layers, associated with seams of coal, have been reported from the Ipswich formation of Queensland and from the Rhetic of North Carolina. The Jurassic coals of Great Britain are lignite or very low grade bituminous. No analysis of the coals in France is available. The analyses of the Austrian coals, as officially given, are incomplete and afford no information for comparisons ; but the coals are clearly _ _ high grade bituminous, for that of many seams is caking. Hantken ___ has published many analyses of the Jurassic coals in the Steierdorf- _ Anina and Finfkirchen areas, and Nendtwich made a number at a much earlier date. The Steierdorf-Anina samples have as proxi- mate composition : The low percentage of ash makes evident that the analyses are of specimens supposed to represent the average best coal from the mines. This, however, is unimportant here. The upper Liegendflétz is separated from the higher bed by about 300 feet of rock. No marked tendency to decrease of volatile downward is recognizable 88 | STEVENSON—INTERRELATIONS OF FOSSIL FUELS. Water. © Ash, Volatile. Fixed Carbon. Hangendflétz............. 1.04 1.72 33-77 66.23 2.50 1.75 36.59 63.41 I.55 3.10 32.47 67.53 ; Hauptflétz ..... Bere ca ey la 1.28 35.04 64.06 1.88 2.07 30.41 69.59, 2.10 7.20 39-04 61.06 1.90 1.95 34.98 66.02 1.70 2.21 30.22 67.78 I. Liegendflétz............ 2.25 2.56 32.23 67.7 2.25 16.78 23.28 96.72 2.25 2.56 29.22 AGRE 3 1.85 12.88 42.73 Lye | femeos II. Liegendflétz........... 2.05 4.19 34-64 65.36 1.85 3-44 30-77, | 69.23 1.75 5.65 41.86 : Bt c in this series. The samples from the lower seams, containing the high volatile, must be considered as consisting in part of cannel. | : The analyses of the coals from the Fiinfkirchen area, published by Hantken, are ultimate ; reduced to pure coal, as were those sc the Steierdorf area, they are: 3 Seam. Water. Ash, Carbon, Hydrogen. Oxygen. — Me BE sree T-50: <3 24.93 89.6 4.5 pens. Y EV eres I.10 13.03 81.6 4.3 | allie: Vik gece 1.10 5.10 88.1 4.6 sae 2 | BG) CRE ere et oe 1.58 7.80 91.7 4:3 3-9 BOS oe aan 1.80 15.77 93-7 4.5 1.7 : He 3-20 II.64 77.7 4.7 S525, SV Eekisee es I.00 13.67 93-4 4.6 ° BOs Bel " 5-44 7.28 85-7 4:3 98 SX 1.60 15-45 82.9 4.2 12.87. MMTV eases 2.70 9.85 86.0 4:7 a e The order is ascending. The sulphur is from 1.07 to 6.88 per cent. ; but in the great mass of seams XI. and XII. it does not exceed 2. 50. In IV. at Vasas there is but 1.23 but at the Colonie mine it is 6.88. The thickness of the seam and the proportion of sulphur are not in relation ; some thin beds have little, others much. The coal of XII. yields a great quantity of illuminating gas, that from three other . seams about two thirds as much, while that from others is much > less. The two analyses for XIV. and for XVI. are from different localities, but only a short distance apart. The local conditions , STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 39 differ little but the oxygen-content at Szabolcs is very much greater than at Colonie. The proportion of oxygen has apparently no rela- tion to the depth below the surface. _ The analyses by Nendtwich** show as a rule less ash in the Steierdorf coals than in those of Fiinfkirchen but the oxygen is -somewhat less. : The Lower Jurassic coals of Siberia, according to analyses given by the Comité géologique, show extreme contrasts. I. and II. are _ from the Tchoulym field and III. is from the Grande-Bira area. Water. Ash. Carbon. Hydrogen. Oxygen. 11.68 1.56 69-92 6.13 23-95 16.66 2.28 69.73. | 7-12 23-15 2.35 12.00 81.5 5-5 12.7 The brown coal obtained in other districts is much inferior, as ash is very high. The Jura-Trias coals of Queensland are bituminous throughout. _ Jack reports only proximate analyses but these suffice to show the great difference in conditions: IpswicH Group. ) Water. Ash. Volatile. Fixed Carbon. ET ARSE I.00 24.35 ° 32-42 67-57 SE Rea ge! 19.00 27-7 72.2 erates 1.32 31.61 29.3 70.6 nie = Ok ES Oe 2.02 22.8 30.8 69.1 patna s 1.32 19.70 29.3 70.6 Pipe eros e 8.10 2.50 43-29 56.70 SRE A aie 16.00 42.81 57-1 Burrum Group. Seg pate Tene 2.50 2.50 32-00 i 68.00 OS waakuiea eis 2.00 8.00 31.25 68.75 iin hee Cane 2.25 2.10 30.47 69.53 aaa dct 2.75 3-25 29.50 70.50 The Ipswich coals throughout are very high in ash, the specimen VI. being picked from a thin band; all the coals except VI. and 47C. M. Nendtwich, “Ungarns Steinkohlen, etc,” Haidinger’s Berichten, Band IV., 1848, pp. 18, 21, 30. 40 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. VII. are coking; the high volatile in these last, so greatly beyond 4 that of other coals within the same little area, suggests that perhaps they contain some cannel. The Burrum specimens are all froma _ very small area, where mining has been carried on extensively and they are from only two seams. II. and III. are from the bottom and top of the Lapham or most important seam. The ash is low | throughout, showing that, in this area at least, the conditions were favorable to the accumulation of clean coal. All of the seams yield good caking coal, though they differ in the hardness; that from several seams is hard shipping coal whereas that from others, espe- cially that from one, is tender and therefore inferior as a steam coal. There is nothing in the structure to explain this difference as the seams are separated by a small interval. The Lapham coal yields 10,200 cubic feet of gas per ton, with 14.73 candle power; this is the result of a trial lasting for 20 months. The Triassic coal of Norroy, France, was analyzed by Regnault, — who obtained 19.20 per cent. of ash. The ultimate composition of the pure coal is: Carbon, 77.23, hydrogen, 5.39, oxygen and nitrogen, 17.37. Servier asserts that the specimen was not fairly representa- — a tive and gives the results of a proximate analysis by himself: Mois- ture, 10.00, ash, 9.20, volatile, 42.4, fixed carbon, 57.5. This he — regards as a fair average composition. He thinks it is a transition from brown to stone coal but the distillate is alkaline, not acid. _ The Upper Triassic coals of the Richmond basin are all of high- grade bituminous quality, are caking and for many years they were used in the manufacture of illuminating gas in New York, Phila- delphia and other large cities. The available analyses are those re- ported by W. B. Rogers,** which represent the average of the coal as observed at the more important localities. Twenty-two analyses were made. The ash is below 6 per cent. in all except 7 and ex- ceeds I1 per cent. in only 3. The volatile in pure coal varies from 30 to 40 per cent., south from James River, and from 25 to 35 in mines north from that river. Much of the basin is broken by dikes which in some portions have converted the coal into coke; but there are anomalies not due to the influence of igneous rock. Analyses of samples from the bottom, middle and top of the thick 48 W. B. Rogers, “ Report of Progress for 1840,” reprints, pp. 532-535. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 41 a mass in one shaft show 40, 30 and 31.7 per cent. of volatile in q the pure coal, with 10.82, 5.10 and 9.52 of ash. In a shaft, north 7 from James River, the 4 divisions of the coal show a difference of 4g about 4 per cent. in volatile, while the ash is 5.20, 22.20, 9.80 and q 22.60 in the several divisions. Stone, in his report already cited, has given analyses of the coal at Leakesville in the Dan River area of North Carolina. The seam is an insignificant lens but is apparently the most important deposit in that area. It is in two benches separated by only 2 feet of micaceous shale but the composition is very different. The | Water. Ash. | Volatile. ' Fixed Carbon. Rroperbench. ..... 5.606.225 11.67 | 9.65 38.6 | 61.3 Ieowerbench:..... 20.5255. | 5-35 20.27. °.} 12.8 87.1 lower bench is anthracitic and the upper bench is a high-grade bituminous. The sulphur in both is at most little more than a half _ of one per cent. The ash is very much higher at most of the North ' Carolina localities, occasionally reaching 39 per cent. in “ best coal.” a It may be well to gather the notes respecting ash as presented _ inthe several analyses. The conclusions at best can be merely tenta- tive because analyses, in almost all cases, appear to have been those _ of hand specimens supposed to represent the average of the seam _ as shipped: and there are comparatively few showing the com- q position of coals not regarded as fit for working. It is sufficiently _ clear that conditions were not the same in all portions of the area occupied by any seam or during the time of its accumulation. 4 In the Jurassic region of Austria, the coal of one seam near q _ Bernreuth, though externally resembling good coal, has 42 per cent. ; __. near Gresten, the same seam has only 3.9 per cent. The ash is low _ at Hinterholtz but at Grossau it rises to 10 per cent. At Pechgraben _ the average of all analyses is 17. These in all cases are from coals which are mined. No attention was paid to other seams be- cause they are “dirty.” Similar conditions exist in the Triassic region of Austria. Near Kleinzell, the highest seam has 14 per _ cent.; near Lilienfeld, the good coal, with little more than 7 of ash, ___is in the middle seam; near Kirchberg, the coal mined has from 42 STEVENSON—INTERRELATIONS OF FOSSIL FUELS, 15.8 to 19.9 of ash; but near Rehgarten, the same beds yield a coal, — with only 7.8 per cent. of ash; at Loichgraben, the lower seam has good coal, while that from the upper seam, though in appearance equally good, has 52 per cent. Coal'is mined near Gossburg, which contains upwards of 30 per cent. of ash. Rachoy has shown clearly that in both Jura and Trias a seam varies greatly in this respect in different portions of its area. There are numerous coal seams in the Steierdorf-Anina area of Hungary, but only 5 of them are workable—each of these in limited spaces. They are divided into benches, some containing good coal, the others worthless. Samples of good coal from the highest two have from 1.28 to 7.26 of ash, while those from the third have from 2.56 to 16.78. The fourth seam shows less variation, the percentage being 3.44 to 5.65. Within the Fiinfkirchen region, 174 seams were crossed by the tunnel at Vasas, with a total thickness of 52 meters. Thirty-nine of them, 14 meters thick, are “dirty” and worthless; of the 28 seams, which are workable in areas, large or small, at | least one third become at times too impure to be mined. Hantken - gives 26 analyses; 5 show between 16 and 20; 7, from 12 to 15; 4, from 10 to 12 and only 5 have less than 6 per cent. of ash. All of these are from mines in full operation. The brown coal of the Tchoulym field in Siberia has at most only 2.28 of ash in the samples analyzed but, apparently, the same ~ horizons in the North Tchoulym area yield coal with more than 3 per cent. The Ipswich seams of Queensland have from 19 to 31 per cent. of ash, while the Burrum coals are all remarkably free — mineral matter, the highest percentage being only 8. The analyses of specimens from the two benches of a coal seam in the Dan River district of North Carolina show 9.65 in the upper bench and 20.27 in the lower. The best coal in the area has only 5 to 6 per cent., but other samples of “best coal” contain from 20 to 39 per cent. Samples taken by the early students in the Richmond basin were all from the mines then in operation. The lower division of the great seam is usually described as much in- ferior to the higher portions. In most cases, the samples appear to have been chosen from the better portions, for the ash rarely exceeds STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 43 5 per cent.; but in two mines the samples represent different parts of the great seam and the contrast in conditions is marked; at one _ mine, the ash content in the several parts is, ascending, 10.82, 5.10 q and 9.52; in another, the percentages are 5.20, 22.20, 9.80 and 22.60. There is little of detailed information respecting variations of coal in different portions of lenses, as analyses have been made only of coals supposed to be worth mining. But incidental references 3 abound, which show that, toward the borders, ash increases until _ the coal becomes worthless. SUMMARY. : The areas of Jura and Trias, containing coal in economic quan- _ tity, are utterly insignificant, when compared with those in which _ the systems are exposed ; but there are many localities in which coal - accumulated during brief periods and amid unfavorable conditions. _ The odlite coals of Britain and a few spots on the continent of _ Europe are of inferior quality, merely local and almost without _ interest. Elsewhere the useful deposits are in the lower part of the Lias and in the highest divisions of the Trias. The Jurassic ' above the Lias and the Triassic below the Keuper may be regarded _ as barren. 4 The associated rocks are as in the later periods. The Oolite _ coals of England are intercalated in sands; the Jurassic coal of _ Spitzbergen is confined to the Middle or sandstone division, as de- fined by Nathorst; the Grestener or coal-bearing Lias of Austria is composed of sandstones and clays; the same conditions prevail __ in the Liassic coal areas of Hungary and Siberia; the Jura-Trias of - Queensland and New South Wales are almost wholly sandstone ; the "supper Trias in Austria and Hungary is sandstone with intercalated _ shale. But the Jura in Alaska is almost wholly shale and the Upper _ Trias in some small areas has little sandstone. Freshwater fossils, in rocks associated with coal seams, have been oberved in England, Siberia, Spitzbergen, France and Quuensland. The structure of the rocks is evidence of, at most, shallow water and in some cases it is very suggestive of eolian agency. False bedding is reported from England, Australia, Germany and North Carolina and ripple marks 44 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. are common features at many places. Sandstones and shales fre- quently contain logs of wood, in such relations as to leave little room for doubt that they are simply stranded material, There is, however, ample proof that the sea invaded many places where coal was accumulating. The Lower Odlite of England has beds with great abundance of fragmentary marine shells; the Liassic sandstone of Austria and Hungary includes layers with many marine mollusks of littoral types; Ammonites was found at one locality, but that does not militate against the conclusion that the water was shallow—if the shell be not drifted, it shows that the genus could exist in shallow water; the Rhetic of Sweden is freshwater below, but has marine shells in the upper portion, where the coal seams — are very thin and impure. The lower beds of the Jura-Trias in Queensland have yielded a few specimens of offshore mollusks. The incidental references to beds with marine fossils do not enable one to determine the extent of areas covered at one time or another by salt or brackish water; but in the Fiinfkirchen district of Hungary such beds, though few in number, are present in the roof, floor or — even partings of several coal seams, recalling the conditions ob- served in southwestern Utah, within the Benton, near-base of the Upper Cretaceous, where a coal seam between beds of marine lime- stone has freshwater mollusks in a parting. In any event, these deposits suggest that the areas in which they exist were lowland, close to the ocean level. The shallowness of the water cover during their deposition is so evident that one may well conceive that the — invasions were due to diversions of drainage, to shifting of channels of large streams. How readily such shifting of channel ways may change conditions in a plain country is shown by Featherston- haugh’s*® statement that, in one area, the Arkansas River broke through its banks and converted 30,000 acres into swamp land, kill- ing all the trees. Still more remarkable illustrations exist on the broad plains bordering the Paraguay and other rivers in South © America. Many times in sections of coal-bearing rocks, marine deposits are in contact with those of land origin or are separated from them by an inch or two of fine sediment. 49G. W. Featherstonhaugh, “Geological Report of Examination of the Elevated Country between Missouri and Red Rivers,” Washington, 1835, p. 84. STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 45 The lenticular form of coal seams is as distinct in the Jura and Trias as it is in later periods. It is characteristic of Jurassic coals in Great Britain, France, Austria, Hungary, Siberia and Queensland, as well as of Triassic coals in France, Austria and the United States. _ Direct reference to this feature is not made in some of the earlier 4 reports as, at the time the studies were made, the bearing which _ the form of coal seams has upon the problem of their origin was _ not recognized. But in every area the varying thickness of coal j seams is emphasized ; the frequent passage of coal into carbonaceous _ shale is noted; the presence of coal seams in some vertical sections and their absence from others attracted the attention of all observers. The lenses may have considerable area but often they are small; they may be thick or thin. Those of the Tchoulym field of Siberia have small superficial extent, rarely exceeding a few square kilo- meters and they are rarely connected, but their thickness is so great that the Russian geologists speak of the total quantity of coal in this district as “ colossal.” References to contemporary erosion are rare in the reports. _ Wilkinson has recorded instances of filled channel ways in the _ Triassic of New South Wales and Hertle has described an in- teresting “ horseback” in a Triassic seam near Rehgarten in Austria. The irregularities in the roof of coal seams in the Richmond field, as described by several observers, have much resemblance to “ horse- backs,” but the mines in which they were seen were abandoned half _ ‘a century ago, so that one cannot determine whether or not these __ irregularities are due to trenching of the coal seams. Soils of vegetation have been reported from England and the | United States, but, if they be present elsewhere as one should think probable, observers have failed to make note of them. In such soils one finds vertical stems of plants, rooted apparently in place of growth but not associated with seams of coal. The Purbeck “dirt beds” of southern England have stumps of conifers and cycads rooted in carbonaceous clay. Mantell states that the conifer ____ Stems have lost their bark and have a weatherbeaten surface like that of posts set between tides. They resemble the stumps exposed | _ above the Yahtse gravels, as described by Russell. Stems of the Purbeck conifers were snapped off at 3 or 4 feet from the ground 46 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. and they lie prostrate in intervals between the rooted stumps. Henslow saw, at the Portland locality, root-shaped cavities descend- ing into the rock underlying the dirt bed. LEquisetiform plants in vertical position and rooted in place of growth occur at several horizons in the Lower Odlite and the Lias in Yorkshire. Calamites and Equisetum, in erect position, are found in beds above and below seams of coal at numerous localities within the Richmond field. These ancient soils, with erect stems in place, would seem to indi- cate land surfaces at various times during deposition of the coal- bearing deposits, As in the newer formations, the roof may be sandstone, shale or limestone; it may contain marine or freshwater forms, At a Brora in Scotland, it is a mass of marine shells with quartz sand and carbonaceous materials, bound together by a calcareous cement; it passes downward into coarse coal—a faux-toit. Marine shells are present in the roof of at least one seam in the Fiinfkirchen district and bituminous limestone rests on the coal at some localities near St. 4 Anton in Austria. The ordinary roof is sandstone or shale, one 4 or the other predominating in different areas; not infrequently itis sandstone in one mine but shale in another nearby. Finely lami-— - nated sandstone is not rare. Roof shales are often very rich in plant remains, leaves being especially well-preserved, as though they had been lifted gently from the surface of the bog by muddy water. - The sandstone roof of the Lettenkohle in Unterfranken is an old soil, containing erect roots. Frequently, the passage from good coal to roof is gradual anit this is equally true of the passage from coal to the floor, there being distinct faux-toit and faux-mur; but, at times, the passage is abrupt. Occasionally, the character of the coal changes in such — manner as to suggest that one portion of the seam sank below drain- age while the other remained above it; the “ Kimmeridge coal” in the typical area is merely a rich carbonaceous shale, whereas in Wilt- _ shire it resembles peat. In the Tchoulym field of Russia, the burial must have been abrupt, for the upper portion of the coal is very peat-like at some localities. Coal seams, more than 2 feet thick, are rarely single, but are divided into benches by partings of sandstone — or clay, often containing much mineral charcoal. These vary much 2 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. 47 in thickness. The interval between seams XI. and XII. in the Fiinfkirchen area is from zero to 72 feet; similar, though less marked variations are recorded from other localities. Ordinarily, the partings appear to be of freshwater origin, but occasionally one _ contains marine forms of immediately offshore types. The char- acter of the coal differs greatly, many times, in the several benches; some yield excellent coal, but that from others is worthless; that _ from one bench may be caking, that from another may be non- ' caking; that from one bench may be richly bituminous while that _ from another may be almost anthracitic. Coal of Jurassic and _ Triassic age is usually so far advanced in chemical change that iden- _ tifiable plant structure seldom appears in the coal itself until after treatment with Schultze’s solution. But Grand’Eury states that, at _ Nice, Equisetites is present in the coal, recognized by its form, _ though all trace of structure had disappeared. The Keuper coals _ of the Vosges contain bark and seeds, while Rhetic coal from Bayreuth has many streaks which appear to consist wholly of pollen exines. In the Rhetic of the Richmond field Equisetum is abundant in coarse coal. But treatment with Schultze’s solution brings out evidence of vegetable tissue from all the coals examined. x The floor is as variable as the roof, being clay, shale or sand- _ stone. Limestone is reported from only two localities described _ in works consulted by the writer. Within several counties of Eng- _ land the floor of the Lower Odlite coal or coaly shale is usually _ clay or fine-grained more or less clayey sandstone and it contains 3 many roots, which, in at least one locality, clearly descend from the _ overlying coaly shale. A calcareous floor in the Causses of France _ holds roots, which are well defined. Lipold and his associates give no details respecting the floors of Austrian coal seams but the _ presence of plant remains is recorded incidentally for many locali- _ ties. The presence of roots in floors is a familiar phenomenon in the Lias of Hungary; in the Steierdorf-Anina district, they are described as vertical, often branching, and they are associated with plants of several types. According to Grand’Eury, roots, both woody and herbaceous, are abundant in underclays and partings. The condition is similar in the Fiinfkirchen area, where, according to Gothan, the underclay proved to be a root-bed in every locality 48 STEVENSON—INTERRELATIONS OF FOSSIL FUELS. at which the floor could be studied. Vertebraria has been recog- nized in underclays of Queensland; Equisetites roots are in under- clays of the Vosges as also at Nice, where the plants seem to have supplied material for the coal. The underclays of Lettenkohle in Unterfranken are root-beds. The coals of the Richmond field, ac- cording to Rogers, have nothing answering to the Stigmaria-clays q of the Carboniferous; but the underclays are present. They carry no Stigmaria, for the gigantic Lepidodendron and Sigillaria had disappeared; but Fontaine has shown that Schizoneura is present in the floor of the main bed. The flora has been studied in all of the important areas. In the Upper Odlite of southern England, ferns, conifers and cycads are the prevailing types; the Lower Oodlite of Yorkshire contains ferns as the preéminent feature though conifers and cycads are abundant ; Equisetum is common above the coal horizon, at which ferns and conifers prevail. Conifers, cycads and some ferns from Spitzbergen have been described by Nathorst. The Ipswich or upper division of the Queensland Jura-Trias has 11 species of ferns, 4 of cycads and one of Equisetum; ferns prevail in the lower por- tion of the Lias within the Steierdorf-Anina area and cycads in the upper; but in the Finfkirchen area, the flora consists chiefly of ferns, cycads and lycopods. LEquisetum is extremely abundant in — the Trias of Austria and Calamites and Pterophyllum were obtained at many places; the Trias of Hungary has yielded cycads, Palissya and some ferns, but collections have been small, as the coal is unim- portant. The beds in the Atlantic coast areas of the United States contain cycads, reeds and ferns—the last being few in species but extremely abundant in individuals. That the coal-bearing deposits were laid down on an undulatiese surface is well shown in the Liassic areas of Hungary. Within the Torzburg area, the underlying rock is crystalline schist; in the Steierdorf area it is Dyas but in that of Fiinfkirchen it is Trias. A similar condition is distinct in the Trias of Virginia and North Carolina. In the Richmond field, the interval between the lowest coal seam and the granite varies from a few inches to 600 feet, while in the Dan River basin of North Carolina it is more than 1,000 feet. ! THE ARCHZZOLOGICAL SIGNIFICANCE OF AN ANCIENT DUNE. By CHARLES C. ABBOTT, M.D. (Read December 7, 1917.) When solid rock is before us, its history is readily traced, its place in geological sequence determined and its characteristics, _ lithological and mineralogical, determined beyond dispute, but, when _ this and associated rocks are reduced to a coarse powder or sand and _ carried by water or borne by wind hither and yon, it is with diffi- 2 culty that the earlier chapters of the record of its career can be _ deciphered. As words accumulate as books due to the winds of 2 doctrine, so, ridges, hillocks and undulating plains are formed when _ the wind gains access to the sand and rearranges the same en masse __as the stable fixtures of the region determine. These are transient, _ necessarily, every shifting of the wind changing the scenery, but traces of some of these phenomena have, by lucky chance, survived "every vicissitude and it is possible to discover what remains of a one-time dune that was shaped by the winds blowing over a desert- like plain, and at a time when the ocean water filled the adjacent river valley and the tributary brooks were filled with brackish water due to the inflowing tide, and this at a point now fifty miles inland. In other words, here in the valley of the Delaware River, at the head of tide water, but where the salt or brackish water now never reaches, is what remains of a dune that formed on the bank of a _ small creek, now diminished to a brook that itself is reduced almost to the vanishing point during the drought of mid-summer, but has been known to resume its former importance as the result of a cloudburst or of a protracted but less impetuous rainfall. _ So changed now are all the conditions of a few thousands of years ago, that it seems hopeless to reconstruct the surrounding ‘country at the time the dune was formed. This is a task, however, PROC. AMER. PHIL. SOC., VOL. LVII, D, JANUARY 31, 1918. 49 50 ABBOTT—ARCHAZOLOGICAL SIGNIFICANCE OF that should never be beyond the capabilities of an archeologist. Such reconstruction is not a side-stepping from facts to fancy, but a confirmatory demonstration relating to the discovery of artifacts. The plat under consideration is the low-lying termination of a long, rectangular field, mostly at a considerable elevation above the brook that drains a tortuous valley of about five hundred acres. The “dune” area is not distinguishable, at present, from the field of which it forms a part, but until recently was noticeable because of a slight conical elevation, which was the more prominent when not covered with vegetation. The surface soil is a shade lighter than that of the field of which it is a continuation and of finer grain, as indicated by clouds of dust that rise from it when a breeze passes, but which is not forceful enough to equally affect the surrounding surfaces. 4 This plat, the “dune,” if such it be, was deforested about 1770 and since 1800 has been more or less continuously under cultiva- tion. The region hereabout, a plain of thousands of acres, has undergone marked changes since the influx of European settlers, less than three centuries ago. Before that time, it was distinctly — one of hills and hollows that have now disappeared. Lands deeded — as “swamp” and “meadows” have now lost every vestige of such Pp y conditions as these names imply, and only a slightly undulating 4 surface marks what was once highly diversified. This change is ae due, unquestionably, to deforesting and subsequent cultivation, for a sandy soil, unprotected by vegetation, is necessarily the sport of a the elements. Wind and rain attack it viciously at times or play- fully, if we may so call it, but the surface is affected by the lightest — a breeze and the gentlest rain. I have known a strong March wind in 2 seventy-two hours to build up a ridge of sand, a hundred yards long, twelve yards wide and seven feet high; leaving a depression in an : adjoining field, from whence the sand was carried deeper by the — subtraction of this “dune,” of about sixty thousand cubic feet. — Again, on August 24, 1877, there was a “cloudburst” here that materially altered the surface in places, although practically all the area was protected by growing crops or weeds, yet tons of sand and gravel were washed from upland fields and carried to the meadows, © and a gully through which the present brook flows—the natural out- AN ANCIENT DUNE. 51 " let for ordinary drainage—was deepened until the clay underlying the coarse glacial gravel was exposed. The ordinary rains of a season falling on ploughed ground may not have a marked effect, but the rains of centuries tell another story. A section of this dune, cut to a depth of about six feet, shows e present surface soil of about eight inches and of a brownish color inclining to yellow, and beneath, a thick deposit of yellow sand, or brownish-yellow, which is very compact, of uniform size of grains and without any trace of stratification. This, more than ught else, suggests that it is a wind-blown deposit from some near-by point and subsequently compacted by pressure of an over- lying stratum and the slow infiltration of moisture. _ This deposit, during severe winters, is frozen, nearly if not quite, _ its whole depth, but this and the subsequent disappearance of frost r not affect its structure, except to gradually render it more 2 _ compact. 4 _ This yellow sand is but a continuation of the present surface - soil, the difference in color being due to stain from the yearly coat- _ ing of broadcasted barnyard manure and the decomposition of _ vegetation. 3 The deposit merges into a greenish sand, of somewhat coarser _ grain and looser texture, which in turn rests upon coarse gravel and this on the clay—possibly pre-glacial. _ This uniform structure of the deposits is not confined to the - entire area or the conical hill to which reference has been made. _ Near the center of it we meet with irregular bands of red clay of varying thickness and of more varying length. This clay is in no sense a continuous deposit of this material as when the Raritan or ensauken clays were laid down, nor are they derived directly from them. I should say they are due to gradual infiltration when the texture was looser than at present and so a deposit of argillaceous _ character from overlying sand and the actual surface carried to _ where the more compact sand checked the water—summer rains and melting snows—and the particles held in suspension were arrested. This once started, the water would be held here, later, until all for- eign matter was deposited. Even now, there is very little sand to be ound here, that is even approximately clean, and the sand of this 52 ABBOTT—ARCHAZOLOGICAL SIGNIFICANCE OF dune clouds the water when thrown into it. This method of band- formation is beautifully shown, where extensive exposures, i. e., of hundreds of square feet, have been made. Deep-lying bands can here be directly connected with the surface, the lines or channels of infiltration or inflowing still to be seen. 3 Sand is never found where it originated as sand. It is, a the day of its origin, a wanderer until conditions finally imprison it, as with this dune, which was derived from the surrounding area, whether wind or water left it where it is. This, of course, necessi- tates an open country, for the wind cannot reach the sand when protected by vegetation. It was an open country and a coastal plain in a very literal sense, and it was such at so remote a date, not geo- logically, but as we measure history, that the water of the brook nearby was salt or brackish, as evidenced by the presence of a marine conchological fauna—Mya, Ostrea, Cardium, and unde- terminable fragments. Also four valves, broken, of an Anadonta or Unio ; all of which brings us face to face with an antiquity worthy of consideration, as there is also an archeological interest in this dune, in that it contains traces of man’s handicraft. : Or, are these traces of man, intrusive objects? The plat under consideration was part of an extensive forest less than two centuries — ago; the tree-growth being largely oak, with some chestnut, maple, birch, sour gum, hickory and sassafras. The undergrowth was — largely greenbriar (Smilax rotundifolia), with some ampelopsis and grape. The ordinary semi-aquatic growths of to-day fringed the brook and in such effective fashion that the water was hidden except in winter. see The annual deposit due to decay of such vegetation is greater ; than the erosion and has therefore gradually raised the surface of the brook’s surroundings; not measurably perhaps, but certainly to some extent during many centuries. That at one time there was an open brook with characteristic fauna is certain, as even now, when the stream is at freshet stage, the mud minnow (Umba pygmosa) and crayfish (Cambarus diogenes) come from — the reaches of the brook nearer the river, where the conditions for aquatic life are favorable throughout the year. It is, however, the one-time tree-growth to which attention should be called, and not AN ANCIENT DUNE. 53 without reason, particularly to the oaks. These are of slow growth, yet reach to the largest dimensions. It is within a short distance from this dune that, until 1869, there stood a white oak (Quercus alba) which was twenty-seven feet in circumference, three feet from the ground. It was, unquestionably, at least one thousand years old,* but we have no warrant for assuming that this ancient tree was the ancestor of all the white oaks. On the contrary, this forest, when at last felled by the settlers, who sacrificed all beauty to their god, utility, was the remote descendant of a primeval forest growth which began to flourish who shall say when? a If this locally known “ Pearson Oak” was the remote descend- ant, as is logically certain, of Quercus I. of the reign of Oaks, or the last of a long line of forest monarchs, then we must ascribe to the _ forest floor or that soil which in slow course of time accumulated _ during the period that the “dune” and its surroundings were for- a ested, an antiquity which removes it from the remotely historic to a strictly pre-historic time. I know of no means of determining when the forest age was ushered in, except that we view it from the point of physical geog- raphy, if not really a geological standpoint. The forest growth would not start until the condition of soil was favorable, and, in _ this instance, a change from a coast-line condition to an inland one and strictly fresh-water upland, quite uninfluenced by the ocean 2 tides. How long then was this change in taking place? Also, when, _ while an herbaceous flora was in its prime, did tree-growth begin? Was it not until the lesser, annual growths had flourished long enough to spread a thin soil due to decomposed vegetation over the _ old “dune”? Grasses, more pretentious flowering plants, perma- nent shrubbery, might well have had a long day of their own, before _ the overshadowing tree-growth began to encroach on their domain, and there is not a particle of evidence forthcoming that the reign of oaks was not a period of several thousand years. The result of the forest growth is the formation of “ black soil,” as it was called by Peter Kalm,? and how long it took a foot or more of it to accumulate is problematical. It was never a period of wholly renee AMIE ERNST 2 BE 4 1See Annual Report, Smithsonian Institution, 1876, p. 260. Sith Travels into North America,” London, 1770, Vol. II., p. 10. 54 : ABBOTT—ARCHAZOLOGICAL SIGNIFICANCE OF undisturbed accumulation. Through every woodland tract pe trickles a little brook, and often a stream of considerable width and. depth broke the monotony of a forest floor. These would neces- sarily prevent a uniform accumulation of each season’s foliage, a large proportion being carried away, for I have often seen currents of air lift and bear away the dead leaves in a forest and deposit them far from the trees from which they fell. Matted dead leaves are bulky, but when such leaves have lost their identity and become. dust, the result is an addition to the accumulating soil not thicker. than a sheet of tissue paper. Adding to this the decay of fallen tree trunks, we must still admit that the growth of a forest’s black soil is a matter of centuries ; that it is one of Nature’s slow processes. It was on this floor that the Lenni Lenape dwelt, and for how many generations I think no one will presume to deal in figures. He came and little do we know of his career, save that it was not one of such bestial savagery as has been asserted. The variety of artifacts i fashioned by him is evidence of this. a When, in 1678-80, the English settlers began in grim de to convert the wilderness into a garden, or destroy beauty in the interests of utility, the forest floor began rapidly to disappear. Where the surface is undulating, and I have seen but slight acreage that might be called “a dead level,” the forest floor, when exposed to the weather, is washed or blown off and not worked or washed into the underlying sand. The result of rain, if not violent, is to compact sand and steadily lessen its penetrability. This leads us 0 a consideration of the suggestion, so frequently made, of the i in- trusion of objects from the surface: that a grooved axe or polished celt or broken pot or other distinctly “Indian” possession had by chance sunk from the surface where it was lost or intentionally left and reached to a considerable depth in the yellow sands be- neath and since its passage, all trace of the track of its intrusion become obliterated. Such disturbance always leaves behind it in- eradicable traces. The yellow sands, whether laid down by wind or water action, become arranged in such a way that, if disturbed, no rearrangement on the same lines is practicable. Those who have trenched in such deposits intelligently known instantly when a spot has been disturbed since the original deposition. Nature has not AN ANCIENT DUNE. : 55 the power to repair the damage so that it can deceive the observant eye and skilled hand of the experienced archeologist. When a “ for- eign” object is discovered in the yellow sands, it is recognized at once as part and parcel of the containing bed. If, again, intrusion were possible, why is not pottery found at all depths to which un- doubted artifacts occur? Why not the familiar surface finds that - collectively we call “Indian relics”? I have gathered probably fifty thousand such -objects, and have lingered so long over sand _ banks and gravel beds that I feel entire confidence in the message they hold out to me, and this intimate association has a significance that is not within the experience of the casual observer, who too frequently is the victim of preconceived ideas. To decipher a sand- bank requires patient labor and constant association and, above all, endless comparisons of one point of view with another. Without this, the digging of a single trench and the gathering of a few score “traces” of man’s presence is what the hint is to a practical demonstration. Unfortunately the hint has often led to the most _ grotesque conclusions, and the fact of man’s antiquity been hidden 3 by an array of assertions to the contrary, not one of which has an iota of warrant. I assert without fear of successful contradiction that “Indian” relics do not occur in the yellow sands underlying the forest floor. The character of the disappearance of the forest floor by rain- wash or wind at once demonstrates that such traces of a people whe _ were forest-dwellers as stone artifacts and pottery would ultimately be left scattered over the surface of the underlying formation, upon _ which the forest floor had been built up. The greater part of such traces, as axes, celts, spear-points, steatite vessels and pottery would prove too heavy for the gentle action of rain or wind, and a storm’s cataclysmic action would only bury such objects with abundant evi- _ dence of how they were buried, so no confusion need arise. As it is, _ we find, on the one hand, the relics of the historic Indian with traces of the country’s Colonial period, and, on the other, with such traces of the precursor of this forest-dweller as were left upon the surface of the ground when the forest floor began to accumulate, or earlier. No one hesitates to separate the pennies of the English kings from Indian arrow-points, although found together ; but upon what basis, 56 ABBOTT—ARCHAZOLOGICAL SIGNIFICANCE OF we are asked, is a distinction to be drawn between a chalcedony knife or elaborate gorget and a rude basalt or argillite point with which it is now associated? It has been denied that any such dis- tinction could be drawn’ and it is curious and significant to know — that the vehemence of this insistence is in direct proportion to igno- rance of the locality. The conditions that here obtain are favorable — to preservation of traces of the sequence of events; very generally — they are absent. : , : To eliminate doubt, when a trench is opened, or where any digging is done, other than systematic trenching, the present sur- * face—in no sense an “Indian” surface—for a reasonable depth is _ not admitted to be demonstrative as to the age or origin of the arti- — facts found therein. This zone of doubt I have considered to be the — topmost six inches, after removal of the twelve inches of surface that has been continually disturbed by cultivation. Assuming the forest floor to have been twelve inches thick—I have found it con- siderably more in some localities—then an artifact found some ten ~ inches below the zone of doubt would have been forty inches below — the surface if the forest floor still existed. When, then, we con- sider that this dune, treated by others, as well as by myself, at dif- _ ferent points, have exposed pebbles, and some too large to suggest eolian origin as to locality, shells, marine and fresh-water, frag- ments of bone and artificially produced chips of basalt and sal esi and a few of chert, and completed artifacts. In my own experience, the position of every object, as exposed, — suggested—demonstrated?—that it had not slipped down any crevice, but always with the long axis of its diameter horizontal. This, I believe, is the experience of those who have examined the deposit. Several long, narrow points of basalt or argillite have been recovered and every one was as described, as to position; the deposit suggesting, by reason thereof, water action and the points floated or rolled to their position when found. However this may be, the fact remains, that the dune, assumed to be post-glacial, has a geological antiquity and that it contains traces of man that reach back to the time of. its formation. A few words in conclusion concerning the sand deposits of the neighborhood. As familiar to all, the unimpressionable rock is our . AN ANCIENT DUNE. 57 ‘standard of stability and its opposite, the so-called, ever-shifting sands. It is to be noted, however, that extremes are always giving _ fise to misleading impressions. Rocks are not so resistant as to _ merit the term “ eternal” and many a bed of sand has withstood the _ changes that time has wrought about them for unnumbered cen- turies. This has not been duly considered by those geologists who feel at home among the rocks where Nature presents a decipherable script, but omits it where only sand has been accumulated. Because of this extra demand for exertion in solving geological problems, the natural history of sand has been neglected or grotesquely misrep- resented. There is a wide distinction between quick-sand, dunes, and the 4 long level reaches of a sand deposit due to aqueous and not eolian transporting force, that has been shut from the light of day and little affected by the rain that reaches it or frost that penetrates the earth’s mat sufficiently to congeal its moisture. There is, too, a vast difference between a sand that has been washed until nearly pure silica and sand with sufficient clay to produce a more or less marked cementation of the mass. Hence it follows that there is a great difference in degree as to the penetrability of a deposit of sand, the clay rendering it resistant in proportion to its presence. Having considered the dune as such and derived from the imme- diately adjoining fields, the archzological interest now shifts to a locality about nine hundred yards west of the dune and trenches opened by Messrs. Skinner and Spier. There is in this locality an area of some one thousand acres where sand underlies the present surface soil. It varies considerably, as sand, and suggests that since the original deposition it has had a varied experience, the same agen- a cies not affecting the whole area. Thus, it has given rise to various _ Opinions as to its age and origin, the judgment based upon a single point of examination. That this sand area was at one time the sandy bottom of a shallow arm of the sea is probably true, if not demon- strably so, as I believe, and so gives a clue to the age of the artifacts contained therein ; the sequence of event being—as suggested by the late Prof. N. H. Winchell, after an examination of the locality with Mr. Volk and myself, Aug., 1913, and this suggestion he main- tained with greater confidence after an exhaustive study of arti- 58 ABBOTT—ARCHAZOLOGICAL SIGNIFICANCE OF facts from here, surface-found, from the sands and from the im- plement-bearing drift gravels, declaring that the changes wrought in the surfaces of these worked stones could only be explained by a a submergence in sea water.’ Illustrative of this, July 4, Mr. Albert Moyer, of New York City, and myself, made a section of these sands, and he had the good fortune to expose, by careful paring down of the exposure, a series of objects, all of which are, I think, of artificial origin. The — surface soil, twelve inches in depth, was carefully removed and the — sand underlying for several inches was considered a zone of doubt and nothing found therein was accepted as indicative of antiquity. Beneath ‘this, the sand was of lighter color, and only moderately compact, but increased in density and where really resistant to the trenching tool, the artifacts were found. Among them was a minute fragment of pottery. This was a little disconcerting, for I have never seen potsherds from these sands and Mr. Volk informs me, he has never, in his many years’ experience, found any traces of pottery, even of the rudest pattern. I can only conclude that pre- Wisconsin man was acquainted with the rudiments of the ceramic — art. _ Probably more effective than rain in changing the conditions of a deposit of sand is the action of frost upon it. This, of course, refers to rain as absorbed and not as a transporting agency. I have known the soil and sand at this point to be frozen to a depth of four feet, while the lowermost of the artifacts found by Mr. Moyer was forty inches below the surface and twenty-two in the compact, un- disturbed, clay-cemented sand, so it becomes evident that during many winters of each century since the sand was in its present position and under present conditions, these objects have been frost- bound and then liberated by the springtide warmth. This periodic condition of frost appears only to affect physically but not disturb or displace the containing bed. The upheaval of the surface of a field is only soil deep in its disturbance of the contained pebbles and artifacts. They may be lifted up and let down, but the relative position of these objects, each to the other, is not materially changed 3 See Winchell- Abbott Correnspondence—unpublished—at Peabody Mu- seum, Cambridge, Mass. AN ANCIENT DUNE. 59 and there is no significant inhumation of a surface-lying specimen. A notable example of this non-disturbance is shown in caches of chert or basalt blades, which have lain undisturbed near the sur- face, just as placed by aboriginal man, until cultivation of the soil or other interference by man brought them to light. Again, were frost an inhuming agent, how is it that stone mortars weighing from ten to fifty pounds do not gradually sink with each winter’s freezing and thawing? I have passed over these relic-bearing fields when I sank “knee-deep” in the mud, but this pressure of the foot was a ‘matter of twelve or fifteen inches actually and explained by the 4 _ weight of my body, but a mortar or even a stone axe of five or six _ pounds ought gradually to sink deeper and deeper in sands when the _ frost has melted, but we never find them at any such depth as the lower compact sand of the “yellow drift,” except perhaps in some deep pit, the definition of which is clearly shown by the dark discol- % oration and unmistakable boundary line. But, there, at such sig- _ nificant depth, we do find rude basalt and argillite artifacts of incon- siderable weight, usually less than an ounce, yet as distinctly the _ output of man’s skill as the most elaborate production of the his- toric Lenape. a The single explanation of the presence of the characteristic arti- ___ facts of the yellow sands lies in the suggestion that they are as old as __ the containing bed and were made at the time or earlier than its _ deposition as now obtaining, be the agency of distribution either wind or water. That these traces of early man are intrusive objects is simply impossible, and this applies equally to the palzolithic im- _ plements of the Kansas gravel, through which the Wisconsin Ice- __age floods have washed the present channel of the Delaware River. AMERICAN SANITATION IN THE PHILIPPINES AND ITS INFLUENCE ON THE ORIENT. (Read December 7, 1917.) By VICTOR G. HEISER, M.D. Sanitation is constantly becoming more exact. America’s work in the tropics has contributed greatly to that end. The public is beginning to realize that science is rapidly reaching the point at which the proper expenditure of definite sums of money may be 4 counted on to produce proportionate reductions in the morbidity and _ mortality rates. Every dollar wisely invested should produce an appreciable improvement. It does not necessarily follow that great sums of money are required. We are all aware of the marvelous results which were obtained by our health department in Panama. It is not so well known that equally striking results on a far greater scale were obtained in the Philippines. In Panama the cost has been given as approximately $3.38 per capita per annum. In the Philip- pines the cost was about 20 cents, and the results were obtained under civil conditions without the use of military force or extra- ordinary powers. The entire cost of the sanitation was defrayed by the revenues of the Philippine government. When it is remem- bered that the Filipinos are among the lowest taxed people on earth, — it will be apparent that it should be quite possible to achieve in coun- tries with greater resources even better results than were accom- plished in those far-away islands. Since my return to America it has been a great shock to me to find this country in many respects far behind in health accomplishment and to discover that sanitary procedures which have been in force and have been producing good results for many years in the Philippines are now only gradually coming into use and are being heralded as among the most modern -and recent advances. : Soon after the occupation of the Philippines, a board of health was organized under Army General Orders No. 15, under the 60 HEISER—SANITATION IN THE PHILIPPINES. 61 authority of which army officers did good work and made an excel- lent beginning in reducing the ravages of certain diseases which they _ found very prevalent. This work was largely concerned with pro- tecting the health of the troops and was chiefly confined to the city of Manila. When the civil regime began, in addition to deplorable sanitary conditions resulting from centuries of neglect, the newly created civil board of health found itself confronted with a severe outbreak of plague in Manila and in a number of the provinces. To add to these difficulties, the board of health had scarcely opened its offices before there began one of the severest epidemics of cholera that has been known in modern times. In a little more than a year it num- bered over 300,000 victims, of whom 150,000 or more died. When the civil board of health began its work 40,000 persons were dying annually from smallpox. Beriberi in jails and public institutions was responsible for a large number of deaths. There was no governmental provision for the insane, and more than 3,000 of these unfortunate individuals were without adequate care. The sanitary condition of the prisons throughout the islands left much to be desired. With the exception of the water system which was available for a part of Manila, and possibly a few other minor in- a stallations, there was not a reservoir, pipe line, or artesian well for _ the seven or eight million people of the entire archipelago, and even _ the water for Manila was known to be grossly polluted. The burial _ Of the dead was not properly regulated. In making new interments, the bones of those who had been previously buried were frequently cast out to bleach in the sun or were thrown upon a bone pile. The _ city of Manila, with its population of over 200,000, had no sewer system. Disease-carrying human discharges found their way into _ esteros or canals or were deposited directly on the ground, causing serious soil pollution. _ Sections of Manila varying in population from 5,000 to 25,000 were built up with houses so closely crowded together that there was no room for streets and alleys. Entrance and egress, in many in- ‘stances, had to be made by passing under the houses. As most of _these crowded sections were built over tidal flats, the difficulties of the situation can well be imagined. 62 HEISER—AMERICAN SANITATION There were no adequate building laws, and, as a resis too , free quently the case in Oriental countries, small dark interiors with no light or air were the rule. Street cleaning was most indifferently carried out. Large quantities of garbage and other filth accumu lated in the back yards and upon the streets. Tuberculosis was prob- ably responsible for at least 50,000 deaths per annum, and no gen- eral education measures were in operation with a view to eran” the people how to combat disease. There was no food law in the modern sense. Perishable provi- sions were sold under insanitary conditions. The vilest class of food products was often shipped into the country, There was pra tically no inspection of animals before slaughter, neither were there suitable slaughter-houses. Dysentery soon caused sad havoc among the American troops and among those who came in civil capacities. Subsequent investigation showed that the native population also suf- fered severely from this cause. Se Hospitals for the masses, with modern operating rooms ‘and surgical equipment, were practically unknown. Persons died on every hand with diseases which could have been relieved by ordi- nary medical procedures. It was not uncommon to find victims horribly deformed by conditions resulting from injuries or disease that could have been cured without deformity if skilled attention and facilities had been available in the beginning. There were perhaps a half million persons living in a wild state, for whom there was no medical relief. Seana In the days prior to American control, maritime quarantine was often conducted upon a basis of graft. Naturally the result of such lax methods was the introduction from nearby foreign countries of dangerous communicable diseases such as plague, cholera, and small- pox. More than 5,000 lepers were at large throughout the Philip- pine Islands. A few hundred were cared for by charity, but there was no attempt to segregate lepers with a view to avoiding the danger of infection or bringing the disease under control in the entire archipelago. Malaria, likewise, prevailed in hundreds of towns and there was no quinine with which to combat it. Imita- , tion quinine pills were frequently sold at fabulous prices in the a IN THE PHILIPPINES. 63 stricken districts, and the people had no means of relief or redress from this intolerable condition. It would be a pleasure to state that all the evil conditions men- 2 tioned above, as well as others, have been remedied, or relieved. & This, however, is not the case. At best, in the time which has _ elapsed and with the funds available, it has been possible only to _. make a good beginning. Much ridicule was cast upon the efforts of _ the American government to better the sanitary conditions of an Oriental population. It had been a fairly well-established rule in _ other countries, in dealing with dependent peoples, to permit the masses to live as they would and to direct efforts at sanitation largely _ toward the benefit of Europeans. This policy, of course, was not in accordance with the views of the people of the United States, happily for the residents of the Philippine Islands, who are now enjoying most of the benefits available to the residents of Europe or America. _ The American sanitarian had much to learn, and in the beginning __ his efforts were further hampered by the passive opposition of the bulk of the population. j The first campaign against cholera was not as successful as could have been wished, but it paved the way to attacking future outbreaks _ With greater result. It soon became apparent that nothing was to ___ be gained by the use of force. Methods of codperation and of win- ning the confidence of the people were rapidly substituted for more _ drastic measures in controlling the disease. Early efforts to combat _ plague, also, did not meet with complete success, although better results were obtained than with cholera. In dealing with plague, not _ only Filipinos but Chinese and other races had to be considered. _ Efforts to bring the foreigners to the ways of the twentieth-century _ hygiene often would have been ridiculous had the outcome not been so tragic. In brief, it may be stated that the American policy has been to bring about a sanitary regeneration of the Philippine Islands, not in spite of the Filipinos, but with their codperation and assistance. One of the first steps was to organize some 300 boards of health throughout the islands, with Filipinos in charge. In many cases the officials who composed these boards were brought to Manila and 64 HEISER—AMERICAN SANITATION given a course of instruction in modern sanitation and hygiene. This resulted in efficient codperation. It is but natural that a people should resist health measures which they believe are enforced by the governing power for the purpose of making them miserable, un- happy, and uncomfortable. When it became apparent that cholera seldom occurred among Americans who drank only boiled water and ate only cooked food served hot, these practices soon had imitators among the better-class Filipinos and from them gradually spread to the masses. Vaccination had been practiced in the Philippines for several centuries, but was never done in a systematic manner so as to reach all the population. The result was that a favorable soil for small- pox remained, and unvaccinated individuals were constantly at- tacked. Over 10,000,000 vaccinations were made in the Philippines, | without the loss of a life or limb. As province after province fell into line, the disease disappeared in the wake of the vaccinators, so that the number of deaths was reduced from 40,000 per annum to a few hundreds. The Island of Culion was set aside for a leper colony. The con- struction of a modern town was begun. When it had proceeded — sufficiently far, the collection of lepers was started. More than 4,000 now find their home on the island, thus giving America the distinction of having the world’s largest leper colony. A laboratory. for the study of leprosy has been established, in which every effort is made to find and use remedies believed to be efficacious in the treat- ment of the disease. Considerable success has been had through the administration, by the hypodermic method, of a chaulmoogra-oil mixture. A number of apparent cures have taken place. Most of the Oriental countries are now giving this treatment a trial and cures have already been reported from many of them. In Manila, a modern water system has been constructed at a cost of approximately two million dollars. The water is obtained from an uninhabited watershed, an improvement which has resulted in a reduction of approximately 800 deaths annually. Water has also been made available in many sections of the city not previously supplied. Ata cost of another two million dollars, a modern sewer system was installed. IN THE PHILIPPINES. 65 Hundreds of artesian wells have been bored in different parts of the islands. In many sections in which artesian-well water is exclusively used, the death rate has fallen one half. Beriberi, which in former days caused frightful mortality in jails and public institutions, has been brought under control through a governmental order which prohibits the use of polished rice in public institutions. This fact is gradually coming to the attention of the masses, and there is reason to hope that in the future the number of deaths from beriberi among the general population will be con- siderably reduced. Modern sanitary market buildings constructed of reinfotced con- crete have been built all over the archipelago. These have been a great factor in the cleanly and economical distribution of food and at the same time an important source of municipal revenue. It is the frequent comment of travelers that Manila is one of the cleanest cities of the world. The streets are swept daily. Garbage is collected every night. Largely as a result of these two measures, Manila is almost a flyless city. Plague has been eradicated. By making available safe water and by active educational propaganda, the spread of amebic dysentery has been cheeked. Laws are now enforced for the proper laying-out of cemeteries, and for proper burials. Streets and alleys have been cut through the congested districts of the city. Many thousands of residents have been re- moved from low swampy lands to higher sites. Modern, danger- ous-communicable-disease hospitals have been built in Manila and elsewhere, and the people, educated to an appreciation of such in- stitutions, now willingly avail themselves of their use. The govern- ment has built a hospital for the insane, where at least the more violent cases and those urgently in need of care can receive atten- tion. A large general hospital, with a capacity of 350 beds, has been built in Manila. It is one of the most modern in the Orient. A nurses’ training school, with over 300 young Filipino students, men and women, is in successful operation; its graduates are already rendering most important service. A medical school, with modern laboratories and the latest equipment for teaching by whole-time in- structors who are specialists in their respective branches, was organ- 66 _ HEISER—AMERICAN SANITATION ized in 1906. It has a five-year course and its graduates are assum- ing positions of medical responsibility. An anti-tuberculosis society has been formed, and an active educational propaganda is in progress. : A hospital has been established at Baguio for incipient cases of tuberculosis, and sanatoria are being conducted in Manila and other, places. at Manila now has the most complete set of sanitary ordistigaels ut s any city in the world, and in many directions greater sanitary prog- — ress has been made than elsewhere. No doubt many of the coun- | tries in the Orient feel themselves compelled to join the movement for modern sanitation instituted in the Philippines. They well — understand that the crystallized opinion of the world demands more J and more that conditions in other Oriental countries must be made. to compare with those of higher standard. Before the lepers of the Philippines were segregated, scarcely any Eastern country had segre- gated lepers. The maritime quarantine practices of the Philippines — are being emulated, and agreements are being entered into between the different countries: for the control of dangerous commntnedtle: | diseases. ee Largely through, the efforts of the medical men of the Philip- pines, the Far Eastern Association of Tropical Medicine was or- ganized. This bringing together of the medical profession of the various countries has resulted in the promotion of good will and the interchange of ideas, all of which has been mutually beneficial. Instead of viewing the medical men of the Philippines with sus- picion, their brethren of other countries now meet with them in full fraternity. The influence that this has had in promoting better understanding and progress can scarcely be estimated. Previous to America’s advent in the Orient, fraternizing among the officials in the different countries was scarcely known. Each remained in ~ his own little sphere and much labor and effort were wasted in solving problems which had already been successfully met in other lands. Now there is free interchange of ideas and the knowledge gained in one country is available in a very short time in others. The death rate in Manila was reduced from 46.83 in 1904 to 23.18 in 1914. This means a saving of over 5,000 lives per annum. IN THE PHILIPPINES. 67 The total reduction throughout the islands is more than 60,000 lives a year. The death rate among the civil employees steadily _ declined and in 1915 was 3.88 per thousand per annum. It is small wonder that results such as these, achieved entirely under civil regime and with the limited revenues of the islands, should commend themselves to other countries, and it is a fact that the achievement of American sanitation in the tropics has produced a profound impression. When the Rockefeller Foundation, through its International Health Board, entered the field, it found the world in a receptive mood toward American methods. The conception that the establishment of public-health agencies can be stimulated through hookworm control has won rapid ac- ceptance and has already a good record of achievement. It is realized that if public-health measures are to be successful they must be brought about upon the demand of the people rather than be imposed upon them. To gain this end much effort has been expended in bringing home to the people of tropical countries the practicability of curing hookworm disease. This disease is one of the few over which the medical profession exercises complete control: first, its cause is definitely known; second, a person afflicted with it can be cured with certainty ; third, its prevention is completely practicable. Furthermore, when meas- ures to prevent soil pollution are carried out, other intestinal af- fections such as typhoid, cholera, and dysentery, largely disappear. Thus, the improvement made in public health more than justifies: the money spent in hookworm control. Even more important is the interest awakened in the people. The work in connection with the relief and control of hookworm infection comes very close to the home life. It causes the speedy substitution of rosy cheeks for pale anemic faces, a result that can be understood by the most ignorant of the community. Moreover, credulity is not strained by being asked to believe in bacteria which can not be seen and which too often are regarded as mythical. The worms which are expelled are plainly visible to the naked eye. Hookworm measures, then, are capable of creating a genuine in- terest in public health in the masses, who, quickened to a reali- 68 HEISER—SANITATION IN THE PHILIPPINES. zation of the possibilities from control measures, soon demand relief from other preventable diseases, Health officers are sought and their work is welcomed instead of being regarded as an trusion upon personal liberty. The interest which was awakened by the achievements of the American sanitarian is being followed up by the International Health Board through codperation with the governments of many countries. In the East alone, codperative © measures have been carried out in Egypt, India, Ceylon, Straits Settlements, Seychelles, Fiji, Papua, Siam, Java, Australia, and negotiations are in progress for the further extension of the work. — Thus the United States, a nation that was almost entirely ignorant of tropical sanitation when it entered upon its war of 1898, is now gradually assuming a position of importance in this remarkable field. The establishment of educational institutions has followed hand in hand with the sanitary work, so that in the future the natives of the Philippines may have the knowledge to. achieve health results for themselves. An important outcome of America’s entrance into the field an tropical sanitation is the reflex stimulus which has been produced in the United States. We are emulating in our own country the wonderful achievements which we ourselves have helped to ac- complish in the tropics. But the greatest effect has been to the world at large. The impetus which sanitation in the Orient has received during the past few years has contributed greatly to the well-being of mankind, and America’s efforts, which have been made largely through altruistic motives, have added no small share. — RocKEFELLER FouNDATION, New York. A CRITICAL SURVEY. OF THE SENSE OF HEARING IN . FISHES. By G. H. PARKER. z It was the opinion of many ancient writers that fishes could hear. _ Thus Aristotle in his “History of Animals,” Book IV., Chapter 8, after having stated that fishes possess no evident organs of hearing, declared that nevertheless they must hear, for they flee from loud noises such as those made by the oars of a trireme. Aristotle added further that fishermen were careful to avoid making a noise with their oars or their nets when they perceived many fishes col- lected together, and he concluded that it was evident from these considerations that fishes have a sense of hearing. Among the Latins Pliny in his “Natural History,” Book X., Chapter 89, stated that though fishes were without ears, yet it was quite certain that they could hear, for it-was a well-known fact that in some fish-ponds, the fishes were called to their food by the clap- ping of hands and that in the fish-ponds of the Emperor they came each kind in response to its name. Thus, notwithstanding that these older writers sometimes confused dolphins and other cetaceans with true fishes, they had from unquestionable sources abundant evidence upon which to base their opinions. - The credit of having discovered, contrary to the belief of such authorities as Aristotle and Pliny, that fishes really possess internal ears, seems to rest with Casserius (1610). This discovery was quite in keeping with the opinion of the times as may be inferred from the conversation between Venator and Piscator in that delight- ful repository of ancient fish lore, “ The Complete Angler.” In the first edition of this classic (1653, p. 128) Walton makes Venator put the question to him “But Master, do not Trouts see us in the night?” And to this query Walton, in the guise of Piscator, replies, “Yes, and hear, and smel too, both then and in the day time.” Whereupon he adds an account of an experiment by Sir Francis PROC. AMER. PHIL. SOC., VOL. LVI, F, JUNE 14, 1918. 69 70 PARKER—A CRITICAL SURVEY OF Bacon to show that sound is easily conducted through water and he concludes with the statement that this experiment “has made — me crave pardon of one that I laught at, for affirming that he knew — Carps come to a certain place in a Pond to be fed at the ringing of a Bel; and it shall be a rule for me to make as little noise as I can when I am a fishing, until Sir Francis Bacon be confuted, which I shall give any man leave to do.” In the second edition of “ The Complete Angler” (1655, p. 175) Piscator, who seems to have pon- dered the matter of fish hearing in the two years since the first edi- tion appeared, added the following final touch. “All the further use that I shall make of this, shall be to advise Anglers to be patient, — and forbear swearing, lest they be heard, and catch no fish.” In the eighteenth century the ears of fishes were studied by such workers as Klein (1740), Geoffroy (1780), Hunter (1782), Monro (1785) and others. Hunter (1782, p. 383), in commenting on the function of the ears of fishes, makes the following statement: Thus Hunter confirmed the opinion of previous investigators, who were further supported by what was learned of the structure of the fish ear by a host of later workers including such men as Comparetti (1789), Cuvier (1805), E. H. Weber (1820) and espe- cially G. Retzius (1881), whose monumental work on the ears of vertebrates may be said to have completed a chapter in our knowl- edge of this sense organ. Retzius (1881) has reported very fully on the structure of the “ As it is evident that fish possess the organ of hearing, it becomes ‘unnec- essary to make or relate any experiment made with live fish which only tends to prove this fact; but I will mention one experiment, to shew that sounds affect them much, and is one of their guards, as it is in other animals. In the year 1762, when I was in Portugal, I observed in a nobleman’s garden, near Lisbon, a small fish-pond, full of different kinds of fish. Its bottom was level with the ground, and was made by forming a bank all round. There was a shrubbery close to it. Whilst I was laying on the bank, observing the fish swimming about, I desired a gentleman, who was with me, to take a loaded gun, and go behind the shrubs and fire it. The reason for going behind the shrubs was, that there might not be the least reflection of light. The instant the report was made, the fish appeared to be all of one mind, for they vanished instantaneously into the mud at the bottom, raising as it were a cloud of mud. In about five minutes after they began to appear, till the whole came forth again.” -THE SENSE OF HEARING IN FISHES. 71 ears of no fewer than forty-eight species of fishes. The completely differentiated internal ear of one of the higher fishes consists of a utriculus (Fig. 1, «) with its three semicircular canals and a sacculus (sc) with its appended lagena (/g). The utriculus is ordinarily sc Fic. 1. Fic. 2. Fic. 1. Left Ear of the European Perch, Perca fluviatilis, lateral view, showing the three otoliths; a, asteriscus; |, lapillus; lg, lagena; s, sagitta; Sc, sacculus; «, utriculus. After Retzius. 5 Fic. 2. Left Ear of the European Perch, Perca fluviatilis, median view, showing the sensory patches; c, crista acustica; I, lapilla acustica lagene; n, _ macula acustica neglecta; s, macula acustica sacculi; “, macula acustica utric- uli. After Retzius. connected with the sacculus by the utriculo-saccular canal. The sense organs in this type of ear reach a maximum number of seven: a crista acustica in the ampulla of each of the three semicircular canals (Fig. 2, c), a macula acustica (“) in the utriculus and a second one (s) in the sacculus, a macula acustica neglecta (#) in the utriculus, and a papilla acustica (1) in the lagena. No fish is ‘known to possess a papilla acustica basilaris cochlee or organ of Corti, which makes its first appearance in certain amphibians and is found in all higher vertebrates. Three otoliths are commonly pres- ent in the ears of the higher fishes: a large one, the sagitta (Fig. 1, S), on the macula acustica in the sacculus, a smaller one, the asteris- cus (a), in the lagena, and a still smaller one, the lapillus (7), on the macula acustica in the utriculus. 72 PARKER—A CRITICAL SURVEY OF Some fishes show considerable divergence from the plan of struc- ture just laid down. Aside from amphioxus, which possesses no — ears at all, the cyclostomes exhibit the simplest and probably the most primitive type of this sense organ. In these fishes each ear consists of a single sac with never more than two semicircular canals corresponding very probably to the anterior and the posterior ver- tical canals of the higher vertebrates. There are three sense organs, a crista acustica for each of the two canals and a macula acustica communis on the wall of the sac. In all higher fishes each ear-sac — is double, as already described, consisting of a sacculus and a utricu- - lus with its three semicircular canals. This type of ear possesses ordinarily the seven sense organs already enumerated, the macula acustica neglecta being, however, occasionally absent. In the elas- mobranchs the utriculus and sacculus of a given ear communicate freely with each other through a relatively large opening. In the teleosts and other higher fishes a narrow tube, the utriculo-saccular — canal, may connect these two parts, or they may be quite disconnected _ and separate. Of the thirty-three species of teleosts reported on by — Retzius, eleven possessed a well-developed utriculo-saccular canal, two showed traces of it, and twenty were without the least sign of it, though in embryonic stages they presumably possessed it. These are the chief facts in the comparative anatomy of the ears of fishes. As the terminology shows, these organs were regarded as organs of hearing and this opinion was the prevailing one among scholars of the last century. It has been more or less tacitly assumed in the more important text-books of that period such as Owen (1866), Wiedersheim (1883), Gegenbaur (1898), and others. | The first noteworthy opposition to this opinion came from de Cyon (1878). This investigator, in his study of the function of the semicircular canals in vertebrates, made the observation (p. 93) that lampreys did not respond to sounds and that after their internal ears had been removed, in itself a relatively simple operation, they ex- hibited great disturbances in locomotion. These disturbances were to be observed seven weeks after the operation and were presum- ably permanent. De Cyon, therefore, concluded that the ears in this primitive fish were concerned with responses to spacial relations and had nothing to do with hearing. This opinion was supported THE SENSE OF HEARING. IN FISHES. 73 _ by the fact that the ear of this fish was unprovided with a cochlea, -that organ which is present in the ears of the higher vertebrates and is especially concerned with hearing. | Some seventeen years later and apparently without knowledge of _ de Cyon’s results, Kreidl (1895) undertook the study of the func- tion of the fish ear. His work was carried out on the goldfish (Carassius auratus) and with much care and many precautions. Normal fishes in a carefully guarded aquarium were found not to _ respond to sounds produced in the air, or even in the water itself, _ though the creatures did react to a blow on the cover of the aqua- rium. Fishes poisoned slightly with strychnine were more sensitive and, though they did not respond to a bell or whistle sounded in the air nor to a metallic rod made to vibrate in the water, they did respond to the tapping of the rod, to the clapping of hands, and to _ the report of a pistol. After the removal of the ears, the equilibrium of these fishes was greatly disturbed, as was to be expected from the previous work of Loeb (18914, 18915), Lee (1892, 1893, 1894) _ Kreidl (1892), and Bethe (1894, 1899), but the animals showed no change in their responses to sounds. Kreidl (1895, p. 464), there- _ fore, concluded that it could not be shown that the goldfish hears and that the responses that this fish exhibits to sound-waves were _ dependent upon a specially developed skin-sense. The year following, Kreidl (1896) carried out some simple but _ conclusive experiments at Krems where large numbers of trout and . other fish were bred for market purposes and where the fish were said to come for food at the sound of a bell. Kreidl showed that _ when the bell was rung by an unseen person, the fishes failed to assemble and that the real stimuli that caused them to come to- _ gether was the sight of the keeper and the vibration of his tread. Thus Kreidl was confirmed in his view that fishes, including both _ goldfishes and trout, do not hear. Kreidl’s papers were soon followed by one from Lee (1898), who tested a number of species of fishes by subjecting them to the sounds __ of the human voice, the clapping of hands, and the striking of stones both above and under water. Though the fishes tested proved to be very sensitive to the jarring of the tank in which they were and to concussions on its walls, they did not respond to sounds produced 74 PARKER—A CRITICAL SURVEY OF as already described and Lee (18098, p. 138) concluded that fishes do not possess the power of hearing, in the sense in which that. term is ordinarily used, and that the sole function of their ears is equi- librium. ? These conclusions were not supported by the work of Parker 19034, 1903b) on Fundulus heteroclitus. Recognizing the possi- bility that sound might stimulate not only the skin and the ear but — also the organs of the lateral-line system, three sets of Fundulus were tested. One set was entirely normal. A second set was pre- pared by cutting the roots of the fifth and seventh nerves, the lateral- line nerves, and the spinal cord a short distance behind the skull, thus rendering inoperative the lateral-line organs and the organs of touch on the whole surface of the fish except in the region imme- diately about the pectoral fins. In this set the ears were left intact. In the third and last set the eighth nerves were cut, thus eliminat- ing the ears, while the receptivity of the skin was not interfered with. These three sets of fishes were subjected to sound siinaieene = a large aquarium. The sound was generated by plucking a bass- — viol string attached to the wooden end of the aquarium and so ar- ranged that its vibrations were transmitted directly through the — wood to the water of the aquarium The normal fishes responded by z pectoral-fin movements in 96 per cent. of the trials. The fishes in — which the skin had been rendered insensitive, though greatly re- duced in their powers of locomotion by the operations they had un- dergone, nevertheless responded in 94 per cent. of the trials, Fi- — nally the fishes in which the ears had been eliminated responded in only 18 per cent. of the trials. It was, therefore, concluded that sounds called forth responses in Fundulus by stimulating not only the skin but also the ears, in other words, that this fish hears. To remove any doubt as to the nature of the stimulus, an electrically driven tuning-fork of the rate of 128 complete vibrations per second was made to replace the bass-viol string on the wooden end of the aquarium. When the fork was in vibration, its base could be brought in contact with the wall of the aquarium and withdrawn at will. If this operation was carried out with a motionless fork, no response from the fishes was to be observed, but when the fork was THE SENSE OF HEARING IN’ FISHES. 75 in vibration normal fishes and fishes in which the skin was insensi- tive responded quite regularly with fin movements whereas those in which the ears had been eliminated showed no reactions. Hence there seemed to be no doubt that the ear of Fundulus was stimu- lated by tones. In view of the discrepancy between the results of Kreidl and _ those of Parker, Bigelow (1904) was led to retest the goldfish. Three sets of fishes were prepared corresponding to those that had _ been used in Fundulus by Parker. These sets were subjected to _ the tones from an electrically driven tuning-fork led into the water _ in which the fish was by bringing the base of the fork into con- tact with the wooden side of the aquarium. Normal fishes re- sponded in 78 per cent. of the trials. Fishes with insensitive skins but normal ears reacted in 80 per cent. of the trials. While fishes ' in which the eighth nerves had been cut gave no responses what- q soever to the tone of the fork. These results agreed in the main with what had been obtained by Parker in Fundulus, but disagreed _ with Kreidl’s results on the goldfish. Bigelow, therefore, sought _ for the grounds of this disagreement. For this purpose he re- _ peated exactly Kreidl’s procedure in preparing the fishes and in- _ stead of eliminating the ear by cutting the eighth nerve, he re- _ moved this organ by opening the skull and withdrawing the semi- 4 circular canals and the attached parts of the ear as Kreidl had done. On testing such goldfishes, they were found, as Kreidl had asserted, __ to respond to tones as normal fishes do, but on dissecting them, it a was discovered that by this method only the utriculus had been taken out with the semicircular canals and that the sacculus, uninjured ' and intact, had been left behind. It was, therefore, clear that _ Kreidl’s operation removed only part of the ear and that the portion left behind was the very part most likely to be concerned with hear- ing. Thus the discrepancy between Kreidl’s work and that of Parker and of Bigelow was cleared away. Following these results came a series of papers that were in part favorable to the opinion that fishes could hear and in part opposed to this view. Of those in opposition the first was by Korner (1905). This author tested twenty-five kinds of fishes that had become 76 PARKER—A CRITICAL SURVEY OF accustomed to life in aquaria.*_ The source of sound was a “cri- cri,” a child’s toy consisting of a slightly deformed metal key which on being depressed gave forth a momentary high-pitched, penetrat- ing sound. This sound was made under water at a distance of 30 to 60 centimeters from the fish and was in no instance followed by a response. Korner (1905, p. 126), therefore, concluded that sa ing was an unproved function for the ears of fishes. Marage (1906) was also unable to get any responses from seven species of fishes subjected to synthetic vowel sounds led into the water through a rubber tube closed by a thin rubber diaphragm. Six of these fishes (Gobio fluviatilis, Anguilla vulgaris, Esox lucius, Tinca vulgaris, Cyprinus carpio, and Leuciscus rutilus) were tested — in confined water and one (Alburnus lucidus) in the open. Briining (1906) noted that stickelbacks in an aquarium were not disturbed by the clapping of hands even when this was done close to the top of the water and that fishes in a pond did not respond to a cry — though they were startled by the tread of the observer on the bank. Maier (1909) installed under water in an aquarium an electric bell so wired that it could be controlled from outside. With this — device he tested eleven species of marine fishes (Gadus morrhua, — Clupea harengus, Ammodytes lanceolatus, Trigla gunardus, Cottus — scorpius, Rhombus maximus, Solea vulgaris, Pleuronectes platessa, é P. flesus, P. limanda, and Raja clavata) and twelve species of fresh- — water fishes (Cyprinus carpio, Alburnus lucidus, A. bipunctatus, — Idus melanotus, Gobio fluviatilis, Barbus fluviatilis, Rhodeus amarus, Anguilla vulgaris, Macropodius sp., Anabas sp., Osphromenus sp., and Girardinus sp.). To the sound of the bell no reaction of any kind was given by any of these fishes and Maier (1909, p. 394) concluded that they possessed no powers of hearing. Nevertheless he was surprised to find in connection with another line of experi- 1 The fishes tested by Korner (1905, p. 123) were as follows: Abramis blicca, Cobitis fossilis, Gasterosteus pungitius, Idus melanotus, Petromyzon fluviatilis, Rhodeus amarus, Betta pugnax, Callichthys fasciatus, Carassius — auratus, and two varieties, Chromis multicolor, C. tristramus, Eleotris sp. — Gambusia affinis, Geophagus brasiliensis, Girardinus candimaculatus, Haplo- chilus panchax, Heros fascetus, Pecilia mexicana, Polyacanthus viridi-auratus, Saccobranchus fossilis, Tetragonopterus sp., Trichogaster fasciatus, and T. lalius. THE SENSE OF HEARING IN FISHES. 77 mentation that the American catfish, Amiurus nebulosus, regularly took fright when he whistled. On testing this fish further Maier ‘was completely convinced that it responded to sounds. It was, however, the only fish of those examined by him that so responded. 4 Bernoulli (1910) tested fresh-water fishes in theit natural sur- _ roundings with the sounds given out by a submerged electric bell and with shrill whistling. Three species (Salmo fario, Anguilla a vulgaris, and Lucioperca sandra) were subjected to the sound from the bell and two (Salmo fario and Thymallus vulgaris) to whis- tling. In no instance was there a response. Haempel (1911) also used the sound from a submerged electric bell and a shrill whistle as stimuli for fishes. Five species of fresh- water fishes were tested (Cyprinus carpio, Scardinius erythroph- _ thalmus, Gobio fluviatilis, Trutta fario, and the Zwergwelse= _ Amiurus). None of these fishes reacted to the sounds used except ' Amiurus which regularly responded to both the sound of the bell a and to whistling. On removing the ears from a specimen of q Amiurus and allowing the wounds to heal, the animal lost all re- sponse to the sounds employed. Haempel (1911, p. 325), there- q fore, concluded that while members of the Salmonide and Cypri- 4 nidz cannot be said to hear, the Siluride and particularly Amiurus __ must be admitted to possess powers of hearing. __ -In consequence of the results of Maier (1909) and of Haempel (1911) Korner (1916) was led to investigate hearing in Amiurus. _ This fish was subjected to various kinds of shrill whistling, includ- a ing that from an automobile whistle, to a series of musical tones, __ to the notes of a scale sung by the human voice, and to the sounds from a “cri-cri.” To none of these stimuli was there the slightest _ response. Korner (1916, p. 263) was unable to explain his nega- _ tive results with Amiurus as compared with the positive outcome q of the tests made on the same fish by Maier (1909) and by Haempel (1911). 4 The papers that have thus far been summarized support in general the conclusion that most fishes do not hear. Those that follow have yielded evidence of an opposite kind. Piper (1906a, 1906b) prepared the ear and the eighth nerve of the pike and of EAS RITE NRPS PTR TAI TI PII, meet Sete RR PETRIE OS St MET 78 PARKER—A CRITICAL SURVEY OF the eel so that he could demonstrate a demarcation current on th parts. On producing sounds in the water in which the prepara- tions were, an action current was identifiable that lasted as long the sound did. Such a current was also produced by tapping the walls of the containing vessel, but it did not result from a noise- less jarring of the preparation, nor from a stirring of the water — around the preparation. From these results Piper (1906a, p. 296) concluded that fishes responded to sounds by means of their ears. — Parker (1909, 1911a) attempted to ascertain if there was any evidence for hearing in the dogfish, Mustelus canis, which, as previ- ous study (1903a, p. 62) had shown, was not responsive to ordinary — sound vibrations in water. It was found, however, that if the wooden wall of a tank containing a dogfish was struck by a heavy — swinging pendulum, the dogfish within would respond by a sudden — jump forward or at least by a waving of the posterior edges of the pectoral fins. The pendulum consisted of a bob weighing 3,800 grams and a suspending wire, the whole apparatus having a length — of 260 centimeters. This device was calibrated so as to strike the - wall of the tank with a momentum of 83,600 centimeter-gram-second _ units or more. The minimum stroke was taken as unity and strokes — of greater magnitude could be conveniently delivered up to about five times that of the assumed unit. Normal fishes when swimming freely in the water occasionally responded by pectoral-fin movement to a stroke of magnitude 1 and invariably to a stroke of 1.5. After their eighth nerves had been cut, they did not respond to a stroke of less than 3 and invariably only to one of 4. To ascertain if this reduction in sensitivity was due to the operation they had suffered, a second set, in which for other purposes the optic nerves had been cut, were tested with the pendulum. These fishes responded regu- larly to a stroke of magnitude 2. To eliminate the skin and lateral- line organs, the fifth, seventh, and lateral-line nerves were cut, the spinal cord destroyed up to the neck region and the skin around the pectoral fin cocainized. Notwithstanding the extent of their prep- arations, these fishes responded by movements of the pectoral fins to strokes of the pendulum of magnitude I to 1.5. Without question their ears were receptive for these vibrations. Parker, therefore, — ‘ THE SENSE OF HEARING IN FISHES. 79 concluded that though dogfishes are not responsive to ordinary musical tones, they do possess hearing. _ Tests carried out by Parker (1910a) on Ammocetes by the _ same means as those used with the dogfish yielded similar results. _ This fish is sensitive to sound not only through the skin but also a through the ears. E Parker (1910b) also studied the ears of Cynoscion. In this fish, as in many other acanthopterygians, the sacculus and the utriculus _ are entirely separate structures, there being no utriculo-saccular canal. Cynoscion, after having been in a large wooden tank for some time, became adjusted to its new environment and when the _ side of the tank was tapped vigorously, it responded by a slight E forward spring. The utriculus and semicircular canals were then - destroyed through a small incision on the top of the head, leaving _ the sacculus intact. Such fishes showed at once disturbed equilib- rium, after which they recovered their upright position. On having blinders put over their eyes, however, they swam with great irregu- q larity. Thus both eye and ear are involved in their responses for equilibrium. During all these tests, however, they reacted as normal _ fishes do to taps on the wall of the tank, showing that the destruc- tion of the utriculus and semicircular canals had not interfered with _ their responses to sounds. It was found impossible to reverse the operation just described and destroy the sacculus leaving the utric- _ ulus intact. But by forcing a strong pin through the paper-thin __ bone between the roof of the mouth and the sacculus, it was possible " to fix the large otolith of the sacculus, the sagitta, firmly against the "outer or non-nervous wall of the sacculus and thus prevent its inde- 4 pendent motion. Fishes treated in this way were only occasionally 4 responsive to taps on the wooden wall of the tank. If a normal fish and one with the sagitte pinned down were tested in the same tank, the greater responsiveness of the normal individual was easily _ noticed. Although the experiments on Cynoscion leave open the _ question of the extent to which the skin may participate in sound _ reception, they show very clearly that the sacculus of the ear, as contrasted with the utriculus, has a well-defined part in this activity. Meyer (1910), whose work was chiefly concerned with the capac- 80 PARKER—A CRITICAL SURVEY OF ity of fishes to associate, showed that goldfishes could be taught to g for food to one or another part of an aquarium depending on th sounding of a high- or a low-pitched bell, a result favorable rather than otherwise to the opinion that goldfish hear, Without knowledge of the work of Haempel (1911) and q Korner (1916) Parker and Van Heusen (1917) undertook the stud of the responses of Amiurus to sound and other mechanical stimuli, They were influenced in this by the hardiness of Amiurus and y the observation of Maier (1909) that this fish responded to. whistle. As in Parker’s former experiments, attempts were made to eliminate the ears, the lateral-line organs, and the skin. In two of these operations new methods were devised. In excluding t 2 ear nothing better was found than cutting the eighth nerve. After the operation the necessary incisions on the head quickly healed and the fishes lived well. Following the tests, fishes that had been thus operated upon were dissected to ascertain that the eighth nerves, had actually been cut, an almost invariable result. In the elimina : tion of the lateral-line organs those of the trunk were rendered inoperative by cutting the lateral-line nerves near the gill clefts and those of the head by destroying individually the forty-eight or of that region. This was done by means of an electric depilating needle. Histological examinations of the spots thus treated showed in the preliminary tests the complete destruction of these organs. Finally, the skin was rendered non-receptive by painting it with 20 per cent. solution of magnesium sulphate, which was allowed act for five minutes. The skin of a fish so treated remained insensi- tive to mechanical stimulation for an hour to an hour and a half. In preparing fishes for experimental tests they were always previously blindfolded by having a pair of thin leather goggl shaped shields placed over the eyes and held there by a few stitch taken in the skin. Because of its gregarious habits Amiurus w always tested in pairs, single fishes being much less satisfactory fe experimental work than two. In accordance with the states of © their sense organs eight groups of fishes were used: first, nor fishes with skin, lateral-line organs, and ears intact; second, fishes Ri with skin and ears intact but lateral-line organs eliminated; third, — THE SENSE OF HEARING IN FISHES. 81 fishes with skin and lateral-line organs intact but ears eliminated; 4 fourth, fishes with only skin intact; fifth, fishes with only lateral- tine organs and ears intact; sixth, fishes with only ears intact; seventh, fishes with only lateral-line organs intact; and eighth and last, fishes with none of the three sets of sense organs intact. The fishes were tested in an aquarium of glass and stone, measur- ing 75 cm. by 35 cm. by 40 cm. This was supported on an in- flated bicycle tire that rested on a table each leg of which pressed on a mass of excelsior wood chippings spread on a tile which in turn had under it a pad of rubber 1.8 cm. thick. The whole apparatus was set up on the concrete floor of a basement room in the labora- tory and proved to be remarkably free from extraneous vibrations. Of the several kinds of sounds to which the fishes were sub- jected, that from a watchman’s whistle? blown vigorously in the air gave most striking results. Of the four classes of fishes in which the ears were intact all responded with clearness and cer- tainty by swimming at once from the upper surface of the water into deeper positions in the aquarium. Those in which the eighth nerve had been cut did not respond at all to the whistle, though they responded to other stimuli, such as currents of water, water dropped on the surface of that in the aquarium, and pendulum strokes on the wall of the aquarium. Incidentally it may be mentioned that the "currents of water and the drops of water proved to be stimuli for the skin only, but that the strokes of the pendulum affected not only the skin but also the ear (compare Table I., Parker and Van Heusen, 1917, p. 472). : Another means of stimulating Amiurus consisted in a-series of tones from a telephone submerged in the water of the aquarium. This telephone was enveloped in a tightly stretched thin rubber bag. By means of a piece of apparatus consisting of a series of seven alternating-current generators with their armatures on a common shaft driven by a ten-horse-power electric motor, currents of 43, 86, 172, 344, 688, 1,376, and 2,752 cycles per second were produced. By appropriate switches any one of these could be thrown into the 2 The sound produced by this whistle consisted of at least two elements: a low vibration probably due to the rapid oscillation of the small ball con- tained in the whistle, and a shrill piping note. 82 PARKER—A CRITICAL SURVEY OF telephone which then yielded a tone of corresponding pitch. tones were of a musical quality and were accompanied by harr Thus the fishes in the aquarium could be subjected to any one of the seven tones from 43 to 2,752 vibrations per second without ht least mechanical jar or disturbance. To be perfectly sure operation of the telephone had no effect upon the fish through the sound it produced, its vibrating plate was re which it was operated in the aquarium as in the ordir Under these circumstances no responses of any kind were obtai1 from the fishes. The electromagnetic field and such other incid disturbances necessarily introduced by the telephone were thus shown to be ineffective as stimuli. The reactions of Amiurus to the tones from the telephone are given in the following table: TABLE I. RESPONSES oF Amiurus To TONES AT OcTAVE INTERVALS FROM 43 70 27 2 COMPLETE VIBRATIONS PER SECOND. Each number represents the number of responses in ten trials, five. : each of two fishes. Rae's Pitch of Tones in Compe brations oat Second. Conditions of the Fishes. : 2) 318 1. Normal: skin, lateral-line organs and ears functional. . 10. ORF 2. Ears functional; skin and lateral-line organs eliminated| 10; 8 | 7 3. Skin functional; ears and lateral-line organs eliminated) 6 4 | 3 4. Skin, lateral-line organs and ears eliminated......... o'olo Heusen (1917, p. 477) concluded that Amiurus is more gene stimulated by tones of low pitch than by those higher in the that both the ears and the skin are effective as receptors for tones, but that the ears have a wider range than the skin. results completely confirm Haempel’s conclusion that Amiurus can hear. The judgments that from time to time have been passed in thes two lines of evidence have been almost as diverse as the eviden THE SENSE OF HEARING IN FISHES. 83 _ itself seems to be and much has naturally depended upon the mo- . mentary phase of the subject. Lang (1903), after an extended ac- count of the relations of the otocysts of invertebrates and the ears of vertebrates to equilibrium, concluded on the basis of Kreidl’s experiments that there is no great likelihood that fishes hear, but that experiments should be tried on fishes that have differentiated structures for the production of sounds. Blochmann (1903, p. XCVIL.) on similar grounds also doubted if fishes could hear. Hen- sen (1904) reviewed the work of Zenneck (1903) and of Parker (19034) and concluded from their results that fishes do hear, a conclusion that was justly criticized by Bezold (1904, p. 159), who pointed out that Zenneck’s results might be explained on the assump- tion that the skin was stimulated. Somewhat later Zacharias (1906) in a popular article concluded on the basis of the work of Kreidl and of Korner that fishes could not hear and misstated (1906, p. 373) entirely the results of Zenneck and of Bigelow which he claimed supported this conclusion. Two years later Korner (1908) declared that conclusive experimental evidence to show that fishes hear had not yet been produced, but he felt that it was not im- possible that they possessed a certain degree of audition. In the same year Edinger (1908) pointed out the relation of sensory reac- tions to central nervous structures and stated on the basis of Piper’s : work that with fishes it was rather a question of what did they hear _ than did they hear. Willem (1913, p. 1247), on the basis of the evi- _ dence already cited, argued in favor of hearing. Watson (1914, p. _ 393), after reviewing the more important statements pro and con on _ the question of fish hearing, summed the matter up in the sentence: - “It seems very difficult to reach any conclusion in the face of such contradictory evidence.” ‘ In attempting to sift what has been thus far advanced on the ' problem of fish hearing, it is natural to begin with the query of _ what would constitute hearing in a fish. Both Kreidl (1895) and ” Lee (1898) have discussed this question in the light of their own experiments. Kreidl (1895, p. 461) has pointed out that it is not in accord with ordinary usage to speak of hearing as any sensory dis- _turbance produced in an animal by a vibration propagated through ' 84 PARKER—A CRITICAL SURVEY OF the surrounding medium. Such disturbances, as has long heat known, may stimulate the organs of touch as well as the ear. Kreidl, therefore, rightly maintained that these disturbances. ‘must be shown to stimu” te the ear before they can be said to be stimuli for hearings Lee (1898, p. 138) has also emphasized the impor- 3 tance of regarding hearing “in the sense in which the term is ordi- narily used.” It seems, therefore, fair to conclude that any dis- turbance that can be said to produce hearing through the human ear may also be said to call forth hearing in a fish provided it can be shown to act through the ear and not simply hoe wing or ay other such receptive surface. Pos ee The human ear is normally stimulated by a grea Frasiety: of sounds, some in the nature of tones and others in the nature of noises. We hear not only the tones of a tuning-fork, but the less pure tones of musical instruments, and of the voice as well as an 4 immense array of very irregular disturbances, difficult to describe — 4 from a physical standpoint and classed generally as noises. Perhaps a among the most extreme of these are explosive noises such as are 4 produced by the clapping of hands, the discharge of firearms and a so forth. All of these we certainly hear, for they affect us chiefly q through the ear and their inefficiency as stimuli for the deaf is well, known. a When they are extreme, they produce what we commonly apie 4 of as shock or concussion and there has been a tendency on the part of some workers (Bateson, 1890, p. 252) to regard the shock as distinct from the sound. From a physical standpoint there seems to be no grounds for this assured distinction. The powerful dis- turbance that emanates from the midst of an explosion is not made up of sound and shock or concussion, but is a single complex dis- turbance which when it strikes our bodies may stimulate ears, skin, and even other sense organs. In so far as it affects our ears, how ever, we must admit it as a stimulus for hearing. Kreidl (1895, p. 459) has pointed out that sounds with shock quality are more effec tive as stimuli for fishes than ordinary tones are, and the expe! mental work of later investigators goes far to substantiate this clusion. Nevertheless, for reasons already given, this state of THE SENSE OF HEARING IN FISHES. 85 affairs does not militate against the use of this class of sounds as - stimuli for the ear. It is, therefore, entirely appropriate to use such _ sounds in testing hearing in fishes, but the experimenter must show : beyond a doubt that they do stimulate the ear, otherwise evidence _ derived from such tests fails to touch the problem. The test for 2 hearing in fishes is the proved presence of a response mediated by _ the ear and dependent upon some vibratory physical disturbance in the water which disturbance may vary from the extreme regularity _ of a pure tone to the extreme irregularity of a noise such as the report of a gun or other like explosion. In discussing hearing in fishes, Lang (1903, pp. 44, 48) ex- pressed the opinion that these animals probably possess through the _ ear a sense of trembling (Erschtitterung, Erzitterung) rather than g one of true hearing and that this sense of trembling is a forerunner a of hearing. In distinguishing the sense of trembling from that of _ hearing he states that in the former the pressure waves are per- _ ceived as a series of more or less distinct and separate entities, 4 whereas in hearing the impression is more homogeneous. This distinction is one that pertains to sensation and, therefore, it can hardly be made the basis of experimental tests in fishes. It, more- over, implies that we cannot be said to hear sound vibrations whose _ note is so low that the single beats fail to fuse. But that we hear these beats as well as we do tones is beyond dispute and Lang’s dis- - tinction, therefore, is in reality without support. Something of the same view has been expressed by Bernoulli (1910, p. 639) who, _ however, assumes the receptor for such beats to be the skin not the ear. Lang (1903, p. 48) and a few other workers have also intimated _ that hearing is a process that probably cannot be carried out in water, but is necessarily associated in some way with the air. A little thought, however, will show that this position is quite unten- _ able, for watery fluids bathe the end organs of the internal ears of _ all vertebrates whether they be inhabitants of the air or of the water. If fishes hear, sounds normally reach their ears much more _ simply and directly than in the case of air-inhabiting forms, for _ such disturbances pass at once through fishes’ bodies and require no By Be, a PROC. AMER. PHIL. SOC., VOL. LVII, G, JUNE 14, 1918. 86 PARKER—A CRITICAL SURVEY OF translation from an air medium to a water medium as they do. air-inhabiting vertebrates. When, therefore, as I happens, a fish takes up with a temporary residence in the air i should not be expected to be very responsive in this situation to sounds. This seems to be the case with Periophthalmus phya which often deserts the water for the shore and which, when in th air, is apparently quite deaf even to the report of a shotgun (John- stone, 1903, p. 300). It is only after the development of some fc of translating apparatus, such as an ear-drum and a middle ear, tha it would be fair to expect such animals to show much response to sounds in the air. Organs of this kind characterize the ears of air- inhabiting vertebrates and represent a means of overcoming an auditory obstacle which fishes have not had to meet, for, as has just been made clear, there is not the least ground for assuming that from a physical standpoint water-inhabiting animals find — im- pediment to hearing. . It is a well-known fact that sounds produced in the air pence water to only a very slight degree and, conversely, that sounds gen- erated in the water pass out into the air only to a correspondingly limited extent. The ordinary surface between air and water is an — excellent reflector of sound. Parker (19110, p. 4) found that even the loud noise from a motor boat was only faintly heard by an ob- server who dove close to the boat and Watson (1914, p. 393), when under four feet of water, was unable to hear the report of a re- volver discharged in the air overhead. It is, therefore, not surpris- ing that Fundulus, though very sensitive to sounds, did not respon to the report of a saluting charge of two pounds of gun-powder ex- ploded from a six-pound howitzer until the fish was within thir ty feet of the muzzle of the gun when to the human ear the sound w deafening (Parker, 1911b, p. 8). These conditions were fully a preciated by Bateson (1890, p. 251) when he remarked apropos o certain tests on pollack: “As might be expected, none of the fishes were seen to take notice of sounds made in the air.” Such sounds, as has already been shown, fail in large part to enter the water, being mostly reflected from its surface back into the air. ; It is probably due to this circumstance, rather than that fishes ~ do not hear, that the tests of a number of investigators who used 4 THE SENSE OF HEARING IN FISHES. 87 sounds generated in the air yielded negative results. Kreidl’s (1895, p. 458) inability to stimulate goldfishes by bells and whistles may thus be explained as well as Lee’s (1808, p. 137) failure to get responses to the human voice, ciapping of hands, and striking to- gether of stones. This may also have been the case with the ex- periments of Marage (1906), notwithstanding the care with which _ a translating diaphragm was used, and it seems quite certainly to _ have been true of Bernoulli’s observations (1910, p. 643), accord- ing to which Lucioperca failed to respond to a pistol shot from a boat at the distance of two kilometers. When fishes in water do respond to sounds made in the air, as in the case of Amiurus (Maier, 1909; Haempel, 1911; Parker and Van Heusen, 1917), it must be taken as evidence of very unusual sensitiveness. As a rule such re- sponses are not to be expected, for, as already stated, sound in the air enters water to only a very slight degree. The production of sounds by fishes is not without its bearing on the question of fish hearing. Kreidl (1895, p. 463) appreciated this side of the problem when he argued that “ Die Thatsache, dass es auch Fische gibt, die Tone hervorzubringen im Stande sind, welche méglicher Weise den Zweck haben konnen, as Lockmittel zu _ dienen, lasst immerhin die Moglichkeit zu, dass bei diesen Species 3 bereits eine geringe Ausbildung des GehGdrorganes stattgefunden _ hat.” The importance of testing such species was emphasized by Lang (1903, p. 48). In the seventh volume of the “Cambridge _ Natural History,” Bridge (1904, pp. 355-365), after remarking _ that “contrary to popular belief sound-producing or vocal organs are by no means uncommon in fishes,” gives an extended account of the various means that fishes possess for the production of sounds. In some instances the sounds produced by them are unquestionably _ accidental accompaniments of other types of activity, but in other _ cases the sounds are dependent upon such differentiated mechanisms _ that it is impossible to attribute these emanations to accident. One instance alone will suffice. Of the fishes studied by Parker (1903a, p. 48: 1910b) Cynoscion produces a deep drumming sound audible when the fish is in the air to a distance of at least fifty feet. This sound is produced only by the males (Smith, 1905, p. 377) and Tower (1908) has shown that it results from the vibratory action 88 PARKER—A CRITICAL SURVEY OF of a special muscle on the abdominal organs and partict a the air-bladder. The females not only do not drum, but not possess the drumming muscle. This condition of high ization, which is doubtless connected with the breeding habits Cynoscion, is common to many of the sound-producing fishes in makes it impossible to.agree with Kérner (1905, p. 103) in di ing all such cases as of accidental nature. Though it i that fishes produce sounds that are in some way serviceable to then but that they themselves do not hear, it is very unlikely that such the case and the occurrence of instances of unisexual sound prodt tion, as in Cynoscion, strongly suggests the presence of the — hearing rather than the reverse. It is reasonable to suppose that if fishes hear, sey will « some form of response to sounds. If it could be demonstrated no fish responds to sounds of any kinds, it would be highly i 1pro ably that fishes heard. Several investigators have thus tested fishes and, without reference to skin or ear, they have attempted to ascer- tain whether in fact fishes respond to sounds at all. Such inves- tigations are fundamentally important for the problem at hand bu as already explained, they do not allow of a discrimination between touch and hearing. Bateson (1890, p. 251) noticed that to vibrations from blasting pouting scattered, sole, plaice, and turbots buried themselves, and congers drew back a few inches. To a bl on the aquarium wall pollack made an obvious response. Kre (1896, p. 585) stated that Salmo iridens was stimulated by the vibra- tion from the human footfall. Zenneck (1903) found that Leuc cus rutilus, L. dobula, and Alburnus lucidus swam away from electrically driven bell immersed in a stream. Parker (1903a, p 62) showed that mackerel (Scomber scombrus) and menhz (Brevoortia tyrannus) responded to the vibration of a cord app’ to an aquarium. Lafite-Dupont (1907) found that, except for elasmobranchs (la Roussette, la Torpille), the other fishes tested (1 Grondin papillon, la Vieille, le Mulet, la Sole) were responsive a stroke on the side of the containing vessel. Parker (1912) foun that certain fishes, Tautoga, Stenotomus, Menticirrhus and Sphe- roides, avoided the end of an aquarium at which blows were deliv- — ered by a swinging pendulum, that Prionotus gathered near this THE SENSE OF HEARING IN FISHES. 89 source of sound, and that Fundulus, though much disturbed by the ‘sound, tended to go neither toward the source nor away from it. These positive results show that many fishes respond to noises or even tones, but they do not throw light on the question of the par- ticular sense organ concerned and consequently it cannot be stated whether they are due to stimulation of the ears or of the integu- mentary sense organs or of both. As opposed to this line of evidence several investigators have re- ported lists of fishes that are said not to respond to sounds in any way. As already noted in an earlier part of this paper, Korner (1905) recorded twenty-five kinds of fishes none of which re- sponded to the sounds from a “cri-cri.” This is certainly a for- -midable list. When Korner learned through the work of Maier (1909, p. 394) and of Haempel (1911, p. 325) that Amiurus reacted to a whistle blown in the air as well as to sounds from a submerged electric bell, he undertook to test this fish with a variety of whistles, the human voice, and other sound-producing devices including the “cri-cri.” His results were completely negative (K6rner, 1916, pp. 263, 267), and he confessed his inability to explain the conflict be- tween this outcome and the results of Maier and of Haempel. Parker and Van Heusen (1917) have shown not only that Amiurus is receptive to sounds but that, in respect to this stimulus, it is an exceedingly sensitive fish. Their method of work throws some light on Kérner’s results. When Amiurus was to be tested by them for response to sound, blindfolded individuals were put into a large aquarium. Here they appeared to settle themselves quickly near the bottom and to assume in a short time a condition in which it was reasonable to carry out tests. But in this state they seldom responded to sounds and it was only after they had been some hours, or better a day or so, in the aquarium that they really arrived at 2 condition of responsiveness. After this period they began to desert the bottom and to swim in the upper water, and in this state they were most responsive to sound. When thus swimming near the top, a blindfolded Amiurus would immediately descend to the deeper water in response to a very slight finger-tap on the slate wall of the aquarium. It was. only under these conditions that Parker and Van Heusen obtained responses to a whistle or to sounds from the 90 PARKER—A CRITICAL SURVEY OF telephone. If the hand of the experimenter was held in the ac water, be it ever so carefully done, the Amiurus immediately scended to the deeper parts and responses to the more delicate of stimuli were completely inhibited. Hence Korner’s method operating a “cri-cri” by hand under water could have had no result that that of rendering the fishes quite unresponsive and would have been surprising if he had obtained anything but nega tive results. As this responsive phase of Amiurus seems to have. en- tirely escaped Kérner’s attention, it is natural that he should. ‘ have failed to observe the reaction of this fish to whistles, and | t other sound-producing devices. Hence so far as Amiurus is cerned Kérner’s negative results, as contrasted with those of \ (1909), of Haempel (1911) and of Parker and Van Heusen (1 are quite clearly due to defective technique and as this technique was also the basis of his tests of the twenty-five kinds of fishes fir reported by him as without hearing, it follows that these tests no longer be regarded as valid and that Korner’s statements based upon them are, therefore, without weight. Another source of error in the testing of fishes for hearing As the assumption that their only form of response to sound is From the time of Aristotle this has been known to be a typical re- sponse, but that it is the only method of reaction to sounds is far from true. Kreidl (1896, p. 585) in his experiments at the fish basins in Krems got evidence that certain fishes would approach ¢ center of vibratory disturbance and Parker (1912, p. 103) showed that Prionotus, which produces a loud grunting noise, approaches sound center rather than retreats from it. Thus, though fi under most experimental conditions commonly are put to flight by sounds, they occasionally may do the reverse and under more nat- ural conditions this may be a much more usual form of respons than has been suspected. But whether fishes approach or avoid ¢ source of sound, their responses in such activities are chiefly throu their fins. It is, therefore, not surprising that in experimental tests — sound, and particularly slight sounds, call forth very characteristic fin movements. As these movements follow with such regularity on the application of this stimulus, to deny them as a sign of effec-_ tive stimulation is to ignore that very feature which may be of prime — THE SENSE OF HEARING IN FISHES. 91 _ importance in the determination of an experimental result. Hence it is not surprising that Haempel’s outcome on Cyprinus, Scardinius, _ Gobio, and Trutta should have been negative, for he states (1911, p. - 320) at the outset that movements of the pectoral fins, of the caudal fin, and of the respiratory apparatus, however called forth, are not accepted by him as evidences of sound stimulation. To any one familiar with the responses of fish such a declaration must seem to __ say the least, arbitrary and condemns without further ado any nega- ¥ tive results that its author might claim. Such movements are often _ most characteristic and significant and they call for close scrutiny _and careful observations. Although they can be seen clearly and beyond question when the fishes are in aquaria, they would very _ probably escape attention when these creatures are at some distance _in open water. In consequence it seems doubtful if negative results _ recorded under these conditions (Bernoulli, 1910, p. 640) can be said to be well grounded. From the observations of Parker and Van Heusen (1917, p. _ 477), it is clear that Amiurus is by no means equally responsive to tones of different pitches. It responded with greatest certainty to tones of 43 complete vibrations per second, and with less and less certainty to succeeding octaves up to 688. It failed entirely to re- _ spond to the two tones above 688, namely 1,376 and 2,752. It is, _ therefore, clear that Amiurus is much more receptive to tones of a low pitch than to those of a high pitch. Since most of the sounds produced by fishes are of low pitch, being described usually as -croaking, grunting, or drumming sounds, it is probable that fishes are adapted chiefly to this class of tones. It is, therefore, not im- _ possible that many tests that have yielded negative results may have done so because the tones employed were too high in pitch for the fishes. This may have been the case in the sound from the “cri- cri” employed by K6rner (1905, 1916) and with that from the elec- tric bells used by Maier (1909), by Bernoulli (1910), and by Haem- pel (1911). If the sounds thus produced were out of range for the fishes, it is not to be expected that they would react. All such _ tests, therefore, that have yielded negative results are open to this objection until doubt on this point has been removed. Thus the negative evidence of practically all the recent workers on this sub- 92 PARKER—A CRITICAL SURVEY OF ject is thrown under suspicion and it, therefore, remains to discuss this problem from the standpoint of the few cases of poalline er dence. ce These few instances cover a considerable range of fishage “They begin with Ammocetes which is apparently not responsive to ordi- nary noises (de Cyon, 1878, p. 93) though it will react by a winking movement of its oral hood and by curving its body when the wall of its aquarium is struck by a swinging pendulum. After cutting the eighth nerves, these responses can be called forth only by ee: stroke at least three times as strong as in the previous instance, thus showing that the ear is decidedly more sensitive to this stimulus _ than the other receptors in the body, very probably anee in a the skin (Parker, 1910a, p. 470). Mustelus exhibits conditions very similar to those in Ameena It is not responsive to tones (Parker, 1903a, p. 62) and to ordinary noises (Lafite-Dupont, 1907), but it reacts with a sudden jump forward or a quivering of the pectoral fins to a pendulum stroke on the wall of its aquarium (Parker, 1909, 19114, p. 48). On cutting the eighth nerves, three times the former stimulus was required to : call forth the response previously noted. This fin-movement re- mained normally elicitable in fishes whose skin had been desensitized by combined nerve-cutting and treatment with cocoaine, but disap- peared entirely from them on cutting their eighth nerves. Thus Mustelus is responsive through the ear, and less so through the skin, to the noise produced by a stroke on the wall of its aquarium. Among teleosts three cases call for consideration: Fundulus, Carassius and Amiurus. The grounds for concluding that Pundulus (Parker, 1903a) and Carassius (Bigelow, 1904) hear have already been briefly stated in the earlier part of this paper. Each fish re- sponds by at least fin-movements to the tones of a tuning-fork an to other sounds. These responses cease in part or wholly on cutting the eighth nerves. They are not greatly reduced by very extensive nerve-cutting through which much of the skin can be rendered insensitive. The responsiveness of the fishes under these condition: shows that the operation of cutting the eighth nerve cannot be re-— garded as the occasion of the decline in sensitivity of the particula ‘ group in which this operation was carried out but that this decline THE SENSE OF HEARING IN FISHES. 93 must be ascribed to the loss of the ear as a receptor. Hence the futility of the objection that the cutting of the eighth nerve involves in itself serious inhibition. Watson (1914, p. 394) has urged against these results the criticism that the sound-producing apparatus “used by Parker and by Bigelow,” an electrically driven tuning- fork, “is open to the severest kind of criticism.” No further com- ment is made on this point and the reader is left in uncertainty of what should have been used except for the remark (p. 394) that it is strange that Parker did not repeat Bateson’s experiment of tapping stones under water. Such comments as these show a very imperfect appreciation’of the conditions under which tests on fish hearing can be carried out, for it is extremely doubtful if anything of value could be obtained by Bateson’s procedure whereas that so severely condemned yielded position results. Hence there appears to be no good grounds to oppose the conclusion that both Fundulus and Carassius hear. | Notwithstanding Korner’s negative results (1916) the unusual responsiveness of Amiurus as shown by Maier (1909), Haempel (1911), and Parker and Van Heusen (1917) is beyond doubt and Haempel’s tests of a fish from which the ears had been removed is strongly indicative of hearing. This conclusion is abundantly con- firmed by the much more extensive experiments of Parker and Van Heusen already summarized. The fact that these investigators used a submerged telephone as a source of sound and avoided much of the nerve-cutting previously employed in eliminating lateral-line organs and the skin has removed practically all of the assumed ob- jections to the earlier work of Parker. They confirm, beyond doubt, Haempel’s conclusion that Amiurus can hear. The part of the fish ear concerned with hearing has not yet __ been determined with certainty. The condition seen in many of the _ higher fishes in which the two chief parts of the ear, the utriculus . and the sacculus, are completely separated, suggests at once differ- ent functions for these parts. And the fact that in the goldfish the animal still responds to sounds after the removal of the utriculus and its appended canals (Bigelow, 1904) offers the natural sugges- tion that in this fish hearing is associated with the sacculus. This view is supported by Parker’s observation (1910) that when the 94 PARKER—A CRITICAL SURVEY OF large otoliths in the sacculi of Cynoscion are pinned off ag non-nervous walls of these organs, responses to sounds larg not affect hearing. These observations support Piper’s conclu (1906a, 1906b) based on experiments involving what were w doubt the saccular otoliths. Thus, the sacculus, rather thar utriculus, seems to have to do with hearing in fishes. p. 378) to the effect that in those sciznid fishes that make noises the otoliths from the sacculi are exceptionally large, in Menticirrhus, a scieenid which does not drum, they are | seems probable that in the ears of the higher fishes where t utr and sacculus are well differentiated, the sacculus has to hearing and the utriculus with equilibrium. | The tearing of this conclusion on the functional interpreta’ of the parts of the internal ear in the higher vertebrates must t -f obvious. It points at once to the macula acustica saceuli as a sible organ of hearing. Whether, in mammals, for instanc means of distingaishians between the presence or absence of s including possibly its intensity. In this primitive way fishes ably hear, for it is unlikely, since they lack a cochlear organ, they respond in any differentiated way to differences of tone. Their hearing is probably to be compared to the vision of the tote color-blind, rather than to that form of vision in which colors discriminated. But the fish ear is not only primitive in itself; it exhibits in i various conditions several grades of proficiency. In not a si THE SENSE OF HEARING IN FISHES. 95 primitive fish, cyclostome or elasmobranch, has the ear been shown to be a receptor for what may reasonably be called tones. The ears of these lower fishes are stimulated only by relatively loud noises : such as have been shown to be effective stimuli for the skin. In the higher fishes, the teleosts, the ears are not only stimulated by noises of the kind just mentioned, but they are stimulated by much less intense sounds and sounds more in the nature of tones. In this respect they mark a great advance over the condition found in the lower fishes, a condition probably phylogenetically earlier. From this standpoint it is maintained that fishes from the cyclo- _stomes to the teleosts have been shown to have, in varying degrees, powers of hearing. While it is easy to agree with Haempel (1911, Pp. 325) that Amiurus can hear, it is quite impossible to accept his _ further conclusion that “unter den Siisswasserfischen einzig und -allein den Welsen die Fahigkeit des Horens zukommt.” That these fishes are the only ones that hear is so unnatural a conclusion that it, carries with it its own refutation. Harvarp UNIVERSITY, April, 1918. BIBLIOGRAPHY. : Aristotle. 1883. History of Animals. Trans. R. Cresswell. London, 8vo, . Xx ++ 326 pp. Bateson, W. 1890. The Sense-organs and Perceptions of Fishes; with Re- marks on the Supply of Bait. Journ. Marine Biol. Assoc. United Kingdom, N. S., Vol. 1, pp. 225-256. Bernoulli, A. L. ror0. Zur Frage des Horvermdégens der Fische. Arch. ges. fe Physiol., Bd. 134, pp. 633-644. Bethe, A. 1894. Ueber die Erhaltung des Gleichgewichts. Zweite Mitteilung. . Biol. Centralbl., Bd. 14, pp. 563-582. ——. 1899. Die Locomotion des Haifisches (Scyllium) und ihre Beziehungen zu den einzelnen Gehirntheilen und zum Labyrinth. Arch. ges. Physiol., Bd. 76, pp. 470-493. Bezold, F. 1904. Weitere Untersuchungen ueber “Knochenleitung” und Schallleitungsapparat im Ohr. Zeitschr. Ohrenheilkunde, Bd. 48, pp. 107-171. Bigelow, H. B. 1904. The Sense of Hearing in the Goldfish Carassius oe: auratus L. Amer. Nat., Vol. 38, pp. 275-284. Blochmann, F. 1903. K6nnen die Fische héren? Jahresh. Ver. varterland. a Naturkunde Wirttemberg, Jahrg. 59, pp. xcv—xcvii. _ Bridge, T. W. 1904. Fishes. Cambridge Natural History, Vol. 7, pp. 139-537. Briining, C. 1906. Versuche iiber das Hdren der Fische. Natur und Haus, Jahrg., 14, pp. 312-313. 96 PARKER—A CRITICAL SURVEY OF Casserius, J. 1610. Pentestheseion. Francofurti, fol., 354 pp. G only through the reference in Retzius, 1881, p. 38.) Comparetti, A. 1789. Observationes anatomice de aure interna com Patavii, 4to, lvi + 396 pp., 3 tab. Ee Cuvier, G. 1805. Lecons d’anatomie comparée. Tome II. Les organes sensations. Paris, 8vo, xvi-+ 697 pp. de Cyon, E. 1878. Recherches expérimentales sur les Functions des semi-circulaires et sur leur Réle dans la Formation de la Nc YEspace. Ann. Sci. Nat., Zool., Ser. 6, Tome 7, Art. 8, 96 pp. — Edinger, L. 1908. Ueber das Héren der Fische und anderer niederer Ve tebraten. Zentralbl. Physiol., Bd. 22, pp. 1-4. Gegenbaur, C. 1808. Vergleichende Anatomie der Werbelthiere, Ba Leipzig, 8vo, xiv + 978 pp. Geoffroy, E. L. 1780. Herrn Geoffroys Abhandlungen von ‘ders Gehér- werkzeuge des Menschen, der Amphibien und Fische. Leipzig, 14+ 146 pp., 5 Taf. (Known only through the reference in 1881, p. 38.) i Haempel, O. 1911. Zur Frage des Hérvermégens der Fische. Tnternat, Rev. ges. Hydrobiol. Hydrograph., Bd. 4, pp. 315-326. x Hensen, V. 1904. Ueber das Horen der Fische. Muenchener med. W ct 1 . schr., Jahrg. 51, p. 42. ie Hunter, J. 1782. Account of the Organ of Hearing in Fish. Phil. T Roy. Soc., London, Vol. 72, pp. 379-383. : Johnstone, J. 1903. Report on the Marine Fishes. N. Annandale and he Robinson, Fasciculi Maylayenses, Zodlogy, Part 2, pp. 293-302. Klein, J. T. 1740. Historie piscium naturalis promovende missus primus lapillis eorumque numero in craniis piscium, cum przefatione: de i auditu. Gedani, 4to, 35 pp., 6 tab. é Korner, O. 1905. K6nnen die Fische héren? Beitrage zur Ohrenheilk und Festschrift gewidmet August Luce, pp. 93-127. —. 1908. Ké6nnen die Fische héren? Ber. Senckenberg. Nabsrfare Gesellsch. Frankfurt am Main, Jahrg. 1908, pp. 110*-111*. ee ——. 1916. Ueber das angebliche Hérvermégen der Fische, inbesondere Zwergwelses (Amiurus nebulosus). Zeitschr. Ohrenheilkunde, Ba. 7 PP. 257-272. e: Kreidl, A. 1892. Weitere Beitrige zur Physiologie des Ohrlabyrinthes, Sitz.-Ber. K. Akad. Wiss. Wien, math.-nat. Cl, Bd. ror, Abt. 3, is ae ee ——. 1895. Ueber die Perception der Schallwellen bei den Fischen. ges Physiol., Bd. 61, pp. 450-464. —. 1896. Ein weiterer Versuch ueber das angebliche Ho6ren Glochenzeichens durch die Fische. Arch. ges. Physiol., Bd. 63, 581-586. Lafite-Dupont, J. A. 1907. Recherches sur l’audition des Poissons. Comp Rend. Soc. Biol., tome 63, pp. 710-711. d Lang, A. 1903. Ob die Wassertiere héren? Separatabdruck Mitth. Na at wiss. Gesellsch. Winterthur, 1903, 55 pp. Lee, F. S. 1892. Ueber den Gleichgewichtssinn. Centralbl. Physiol., Bd. | pp. 508-512. ‘ec eine THE SENSE OF HEARING IN FISHES. 97 Lee, F. S. 1893. A Study of the Sense of Equilibrium in Fishes. I. Journ. ; Physiol., Vol. 15, pp. 311-348. —. 1804. A Study of the Sense of Equilibrium in Fishes, Part II. Journ. Physiol. Vol. 17, pp. 192-210. 1898. The Functions of the Ear and the Lateral Line in Fishes. Amer. Journ. Physiol., Vol. 1, pp. 128-144. Lenz, E. 1906. Hoéren die Fische? Wochenschr. Aquar.-Terrar.-kunde, = Jahrg. 4, pp. 158-160. (Not accessible.) _ Loeb, J. 1891a. Ueber Geotropismus bei Thieren. Arch. ges. Physiol., Bd. = 49, PP. 175-189. _ —. 1891b. Ueber den Antheil des H6rnerven an den nach Gehirnver- a letzung auftretenden Zwangsbewegungen, Zwangslagen und assoziirten Stellungsanderungen der Bulbi und Extremitaten. Arch. ges. Physiol., Bd. 50, pp. 66-83. Maier, H. N. 1909. Neue Beobachtungen iiber das Hérvermégen der Fische. _ Arch. Hydrobiol. Planktonkunde, Bd. 4, pp. 393-397. Marage, E. 1906. Contribution a l’étude de l’audition des poissons. Compt. Rend. Acad. Sci., Paris, tome 143; pp. 852-853. | Meyer, M. 1910. Ergebnisse von Versuchen betreffend den Gehdrssinn der _ Fische. VIme Congrés Internat. Psychol., Genéve, 1909, pp. 731-732. Monro, A. 1785. The Structure and Physiology of Fishes. Edinburgh, fol., : 128 pp., 44 tab. Owen, R. 1866. On the Anatomy of Vertebrates. Vol. 1, Fishes and Rep- os tiles. London, 8vo, xlii-+ 650 pp. Parker, G. H. 19030. Hearing and Allied Senses in Fishes. Bull. United e States Fish Comm., 1902, pp. 45-64, pl. 9. —. 1903b. The Sense of Hearing in Fishes. Amer. Nat., Vol. 37, pp. - 185-204. » —- 1909. The Sense of Hearing in the Dogfish. Science, N. S., Vol. 20, p. 428. —. 1910a. The Function of the Ear in Cyclostomes. Science, N. S., Vol. 31, DP. 470. —. 1910b. The Structure and Function of the Ear of the Squeteague. - Bull. United States Bureau Fisheries, Vol. 28, part 2, pp. 1211-1224, pl. 122. —. 1gita. Influence of the Eyes, Ears, and other Allied Sense Organs on the Movements of the Dogfish, Mustelus canis (Mitchell). Bull. United States Bureau Fisheries, Vol. 29, pp. 43-57. —. 1g11tb. Effects of Explosive Sounds, such as those Produced by Motor Boats and Guns, upon Fishes. Report United States Comm. Fisheries, ts 1911, Document No. 752, 9 pp. —. 1912. Sound as a Directive Influence in the Movements of Fishes. 8 Bull. United States Bureau Fisheries, Vol. 30, pp. 97-104. Parker, G. H., and A. P. Van Heusen. 1917. The Reception of Mechanical Stimuli by the Skin, Lateral-line Organs and Ears in Fishes, especially | in Amiurus. Amer. Journ. Physiol., Vol. 44, pp. 463-480. | Piper, H. 19062. Actionsstréme vom Gehdérorgane der Fische bei Schallrei- - zung. Zentralbl. Physiol., Bd. 20, pp. 293-297. 98 PARKER—HEARING IN FISHES. Piper, H. 1906b. Ueber das Hérvermégen der Fische. Muer Wochenschr., Jahrg. 53, p. 1785. Pliny. 1887. Natural History. Trans. J. Bostock and H.'T. Riley, 8vo, 6 vols. Reinhart, H. 1913. Vom Gehoér der Fische. Wochenschr, Kunde, Jahrg. 10, pp. 358-359. (Not accessible.) oe Retzius, G. 1881. Das Gehérorgane der Wirbelthiere. I. Das der Fische und Amphibien. Stockholm, 4to, xii + 222 pp., Richard, P. 1916. Sind Schallreize adiquate Reize fur den Vor gangapparat? Zeitsch, Biol., Bd. 66, pp. 479-509. a Roth, W. 1910. Aus dem Sinnesleben der Fische. I. Héren die Wochenschr. Aquar.-Terrar.-kunde, Jahrg. 7, ‘PP. 430-432. cessible. ) Scott, T. 1906. Observations on the Otoliths of some Tel Twenty-fourth Annual Report Fisheries Board, Scotlan 48-82, pls. 1-5. Smith, H. M. 1905. The Drumming of the Drumfishes (Scisenidse’ ence, N. S., Vol. 22, pp. 376-378. Tower, R. W. 1908: The Production of Sound in the Drunt hes, Robin and the Toadfish, Ann. New York Acad. Sci., Vo 1 180, pls. 6-8. Walton, I. 1653. The Complete Angler. London, 8vo, 16 + 246° ——. 1655. The Complete Angler. London, 2d ed., 12mo, 24+ 35 Watson, J. B. 1914. Behavior. New York, 8vo, 439 pp. ie Weber, E. H. 1820. De aure et auditu hominis et animalium. — aure animalium aquatilium. Lipsiz, 4to, 134-+ 34 pp., Io t Wiedersheim, R. 1883. Lehrbuch der vergleichenden Anator belthiere. Jena, 8vo, xvi-+ 905 pp. Willem, V. 1913. Les origines de l’audition chez les Vertébrés,, Sci. Acad. Roy. Belgique, 1913, pp. 1231-1259. ; Zacharias, O. 1906. Ké6nnen die Fische héren oder nicht? Oe6es Fischerei-Zeitung, Jahrg. 3, pp. 371-374. a3 Zenneck, J. 1903. Reagiren die Fische auf Téne? Arch. ges. Phy: 95, PP. 346-356. THE SYRIAC DIALOGUE “SOCRATES.” A Stupy 1n Syrian PuiosopnHy.? By WM. ROMAINE NEWBOLD. (Read April 23, 1914.) In 1858 Paul de Lagarde published in his “ Analecta Syriaca” _ a short dialogue entitled “Socrates.” The only known copy is _ found in that precious Nitrian codex of the British Museum (Add. _ 14658) which also contains, besides other philosophical works, the _ only existing texts of the Bardaisanian “Book of the Laws of the Countries ” and “ The Oration of Melito before Antoninus Czsar.” Notwithstanding the unusual intrinsic interest of the “ Socrates,” _ it has been, so far as I have been able to ascertain, quite ignored _ since its publication. I have seen no translation of the text nor any 3 discussion of the problems which it presents except a brief and mis- 2 leading statement by Renan (in Duval, “La Littérature Syriaque,” F p- 270).? It has not been republished and the orginal! edition is now | difficult to obtain. a 1This paper was read before the American Philosophical Society in _ April, 1914, but publication was deferred in anticipation of the appearance of Mr. Mitchell’s second volume (see note 5) which was promised for Septem- ber of 1914, but was prevented by the outbreak of the war. As this now seems to have been indefinitely delayed, I have decided to publish my tentative conclusions. 2Soon after this paper was written my friend, Mr. Robert Pierpont lake, brought to my attention V. Ryssel’s paper “Der pseudosocratische _ Dialog iiber die Seele,” in Rhein. Mus., N. F., Vol. 48, pp. 175-05. Ryssel gives a translation, suggests some emendations and adds a few footnotes _ but does not attempt a systematic interpretation. He thinks it a translation _ from the Greek and attributes the translation to Sergius of Ras‘ain (d. A.D. - 536), who was the translator of other texts in the same volume. Whether the _ dialogue was originally written in Greek or Syriac is a question upon which _ I have not formed a definite opinion, but I am inclined to think it was _ Syriac. The style as a whole is singularily idiomatic and the occurrence of Greek words and constructions is not conclusive evidence to the contrary in a work obviously imitated from Greek models. The atmosphere is purely 99 100 NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” In extent it is not very long, occupying only nine pages octavo. In form it professes to be a dialogue between Socrates and an anxious inquirer named Herostrophus or Erostrophus.? But contains little in the way of dialogue, the greater part of the book being occupied by a discourse in which Socrates answers Hero- strophus’s questions, cae The author’s conception of Socrates bears no resemblance to th Socrates of Plato and but little to the Socrates of Xenophon. ‘He is indeed an oriental sage whose utterances are received as oracles by his admiring hearers, and, ees he expresses his views wit modesty—the only trace of the “irony” of the historical Socrates— he nevertheless feels that their homage is justified. “O young man,” he says to Herostrophus in one passage, “not in vain and not ee nought have you come to me to hear my words.” | eee The ostensible theme is the nature of the soul, but in the course of the discussion Socrates reveals the outlines of a system of phi losophy which is of no little interest to the student, not because of its intrinsic value, but because of the light which it throws into so dark corners of the history of thought. The elements of this system are those same Platonic, Aristotelian and Stoic ideas which long before the beginning of our era had become the common property of the races that shared the Hellenistic culture. The syncretistic sys- tems into which they were wrought by sundry thinkers are known by many names—Alexandrian, Hermetic, Gnostic, Neo-Pythagorean and Neo-Platonist—but all possess many features in common. Some of these the system of the “Socrates” also presents, but. peculiar interest lies in the fact that the familiar elements are com- bined in novel fashion. I have indeed been able to discover but one other system which is closely akin to it, that of the Syrian ph losopher Bardaisan, who was born in Edessa A.D. 154 and died 222. It has long been known that Bardaisan exerted no little inflt oriental, the borrowed ideas are treated as no Greek would treat them ar the conspicuous absence of the technical terms of philosophy common | Syriac from the fourth century onward suggests an early date of compos tion. Whatever the original language, I think it probable that the author vy a Syrian; certainly not a Greek. 3 Ryssel suggests that the name should be read “ Aristippus,” for which see no good reason. NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” 101 ence during his life and that the Christian Church which he founded endured for five or six centuries after his death, but, until recently, little has been known of his ideas. The discovery and publication _of Theodore bar-Koni’s* account of his system and of some hitherto unknown works of Ephraim’s® have thrown a flood of new light upon him and Mr. Mitchell promises that his second volume, which _ will appear in the course of a few months, will contain still more _ valuable information. With the aid of this new material one may recognize in the Syriac “Socrates” a work certainly of the school _ of Bardaisan. Whether it is from his own hand or not is another question. 4 The accounts of Bardaisan’s philosophy which we possess are so inconsistent that it is necessary to determine which are and which are not trustworthy. The most exended are those of Ephraim, _ Theodore bar-Koni, Moses bar-Cepha, Moses the Syrian, the Fihrist, and Shahrastani. Of these the first two are the oldest, are in sub- _ stantial agreement and probably are derived from the same docu- a ment. That of Moses bar-Cepha is akin to Theodore’s but contains gy Manichzan elements not found in him; in the later documents these _ elements become still more pronounced. I think it quite certain that these later versions represent the teaching of the Bardaisanian Church after it had been for centuries in contact with the. closely related system of Mani. It is quite possible that the system known _to Ephraim and Theodore had alsd been more or less contaminated : _ by the same influences, having been exposed to them for more than a hundred years. The “Book of the Laws,” which is the oldest a authentic Bardaisanian document, unfortunately gives no definite in- _ formation upon the points of interest. Ephraim and Theodore 4g therefore must be regarded as the only trustworthy authorities. The “Socrates” represents bodies as composed of four “ele- ments” or “powers”—earth, wind, fire and water. Bardaisan ‘posited five “ powers ” or “existents,” ithye, out of which bodies are Oe ee Va 4 Pognon, “Les Coupes Mandaites,” 1898; Addai Scher, “ a Script- orum Christianiorum Orientalium,” Vols. 65-66, 1912. 5“ St. Ephraim’s Prose Refutations of Marcion, Mani and Bardaisan,” - edited from a palimpsest MS. of the British Museum by C. W. Mitchell, Vol. I, 1912. PROC. AMER. PHIL. SOC., VOL. LVII. H, JUNE 14, I918. 102 NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” — Pi hated wind, water, fire and darkness. The elements “Socrates” are those traditional in Greek philosophy, e th the Persian term wind is substituted for “air”; those ; daisan are Persian and are identical with those of the | But Ephraim says (“ Adv. Haer.,” 41, Vol. IL, p. 532, Vatican-ede tion) that Bardaisan regarded darkness as “nothing and capable o nothing.” It could not therefore have been in his system, as it was in the Persian and Manichzan, the active principle of evil, but was rather a mere negation, analogous to the Aristotelian substrat a doctrine which Ephraim repeatedly ascribes to Bardaisan. It ; be compared to that fifth something which the “ Socrates” "speaks of as “that which was undifferentiated and unknown” or an knowable,” although this is not explicitly termed an element. According to the “ Socrates” the animal soul or life is on pounded out of the four elements; its nature depends upon the proportions in which the elements are combined and in partic upon the amount of fire present. The animal soul then consists of four parts. At least some human souls consist of the animal plus the rational soul. The latter has three parts, “ Greatness, “Power,” and “Goodness,” which are the first three manifesta- tions of the “Original Root,” a term which in this work is clearly | equivalent to “God.” Thus the rational soul is divine and t human souls which contain it are composed of seven elements. or parts. Ephraim’s statements about Bardaisan’s theory of the sor have hitherto presented insurmountable difficulties, all of which ¢ dis unconsciousess, could contribute nothing to soul. Elsewhere, (“Second Discourse to Hypatius,” p. 8, 5 sqq., Mitchell), he says that the soul consists, according to Bardaisan, of seven parts. ; cording to the “Socrates” the first of these statements is true 0 the animal soul, the second of the union of the animal with the rational soul. Ephraim then describes the dependence of the soul’s character upon the proportions of the components precisely as is : NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” 103 done in the “Socrates,” and adduces two arguments against the theory. First, the souls of angels and devils are unchangeable, whereas this theory makes them changeable. Second, the sun is unchangeable. This second objection seems on the surface quite irrelevant, and there is nothing in Ephraim’s text to indicate why he | regarded it as an argument against Bardaisan’s theory. Turning to the “ Socrates” one finds that the author devotes nearly one fourth of the dialogue to drawing a parallel between the changes which the soul undergoes during life and those through which the sun passes in completing his daily and annual course. The “Socrates” uses for “God” the term “ Original Root” or “That Power.” The first manifestation of the Root is “Great- ness” (rébhiitha), which is probably here equivalent to the Greek péyebos, a term used in geometry for “extension” and by some Gnostics as a designation of God. It is complementary, so to speak, to “ ZZon,” the more common Gnostic term for God, which means “duration,” duration and extension being conceived as the primary expressions of the divine essence. The relation between this “Original Root” and space (athra@) is conceived in the “ Socrates ” as very intimate indeed. “That Space” is said in one passage to “be” the Root, but more commonly space is conceived as anterior to the Root. Thus we are told that “ Greatness, Power, and Good- ness” “cannot exist in Place but (only) where they have Space that they may be kept in righteousness, refined and pure.” And again, “ Because the Greatness of this Power is vast [ Power], there- fore is it in the compass of the All and outside the All, and there exists no empty Space wherein is nought of it.” Thus God is placed in space. Ephraim repeatedly charges Bardaisan with making _ space superior to God and placing God in space, and in one passage (“ Against Hypatius,” IV., p. 133, 1 Mitchell) he says: “Greater are the praises which Bardaisan uttered concerning space than those which he uttered concerning the God who is in the midst of space.” In another (ibid., p. 132, 42): “ Therefore the Greatness which the Teachings give to space, the Teachings of Truth give to God.” It is noteworthy that the name “ God ” is sedulously avoided _by the author of the “Socrates”; it does not occur even once. 104 |= NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” __ From this mystical theory of the soul’s relation to her : the rational element is identical in all human souls which : it and since it is still in organic union with their common R knowledge. It is by virtue of this community of knowin truth spoken by one receives the assent of another. “ But y Herostrophus,” says Socrates, “have not come to ask or inquire of ‘ me aught which is not your own. If it be of speech that you would “ inquire of me, it is in you and is yours; if of sight, it is ‘yours ; if, Q again, of hearing, it is in you. For no man shall see Good except : him in whom it is and no man speaks speech unless it be implanted in him nor hears unless it is in him. That through which the ey: sees and the ear hears and speech speaks, [and] these three wh I appear distributed among sense-organs (lit. parts, 7. e., pépia)— -theit Root is one. And all these things which I have said to you, He strophus, if you should see them as I do and hear them as I and you do (lit., such you are)—the Root is one. And if agair in another Space you should hear this discourse which you are hear- ing from me, know that this is the Root of which is no Space emp where it is not. For we who abide in the Root are like the branches % of a tree, some in the east, some in the north, some in the south, a some in the west, but the remainder is One Root.” Ephraim criti- cizes (“ Against Hypatius,” V., p. 159, Mitchell) a precisely similar doctrine: “ According to it one soul has no need of another soul to : learn or teach . . . because the knowledge of their essence is eq if, as they say, the essence of all souls is one. If Teacher ; Teaching (pupil?) are from one Root and both are clad in flesh . how does one go astray and another teach? ... If there recollection in all the Root, there is no error in all the Essence. And the sons of this Essence (i. e., those who share in the Essence —how does one fail and another succeed? Their essence is not the same.” Ephraim rarely mentions the name of the author he is controverting and in this passage he does not mention Bardaisan. But in these sermons he is criticizing only Marcion, Bardaisan and Mani and as there is no reason for ascribing the doctrine in question NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” 105 to either of the other two, it is probably to be referred to Bardaisan. And the language and imagery strongly suggest the “Socrates.” The author of the “ Socrates” draws from his theory the legitimate conclusion that the rational soul only is immortal; the animal soul and the body both perish at death. That Bardaisan denied the orthodox doctrine of the resurrection of the body is agreed by all our authorities. _ The agreement between the cosmology of the “Socrates” and that of Bardaisan is not so exact as that subsisting between these 0 groups of psychological theories, but they present nevertheless ‘some notable points of analogy. _ Theodore bar-Koni describes Bardaisan’s system as follows (a few additions have been made from Ephraim and Moses bar- Cepha) : From eternity God and the five elements coexisted in per- fect peace. God was above all, Darkness below all; in between ‘were the other four elements disposed in the same plane—Light the east, Wind in the west, Water in the north, Fire in the south. Then the Wind blew by chance and beat upon and agitated the ele- 1ents ; a smoke not born of Fire gathered (Ephraim has: Darkness ‘crossed its border upwards). Then God sent an utterance of Thought which arrested the Wind and a Wind from on High quieted them in part. The confused portion was then separated from the others and from it the world was made. _ According to the “Socrates” there existed from eternity with “That Power,” z. e., God, a something which you may call at pleas- ‘ure soul or fire or nature (¢vo1s). Since the author has just shown _ at length that soul consists of the four elements, fire being only the _ chief among them, and since he immediately proceeds to refer to “the four elements of the powers,” i. e., the four elementary prop- erties, as existing, one may infer that the eternal existence of soul implies that of the four elements. The scheme is therefore analo- ‘gous to that of Bardaisan, save that the latter has the Persian term instead of the Greek Air. “Socrates” proceeds: “ When that Power wished to exist : i. se) in purity all by himself he therefore commanded that Body should become (or, come into being as) the organization of the PEELE PAIR epee re eh ef NTIS MA TEENS FMT 106 NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” — whole world. And it was spread out® (?) by him upon tha was undivided (or undifferentiated) and unknown.” The motive for the construction of the world here ascribed to x desire to purify himself of soul or matter—occurs, so far a nowhere else. It is clear from Ephraim’s discussions that ’ of no motive assigned by Bardaisan. — In the next stage, according to the “ Socrates,” “by hi he agitated her (the soul) and separated her from himself. all the things which are now seen were (or, became) commingled one with another without form.” The Word or Logos aéeiti 80: e here represented as the cause of chaos—another strange ‘idea, in the systems influenced by Stoicism it is usually the functi ) the Logos to transform chaos into the cosmos. Bardaisan’ agrees with the “Socrates” in recognizing an origination —which is an unusual trait—but attributes it to chance. Eph has preserved (“ Against Hypatius,” III., p. 69, 40 Mitchell) very words of his source—“ At that very time a cause came by chance and the Wind dashed against the Fire.” According to Theodore, the Wind is checked by “an utt of Thought,” the agitation quieted by a Wind (1. e., mvedpa. ) on high, and the portion of the elements which is still con removed from the others and made into the material uw i These steps are not described by the “Socrates.” They are b- viously derived, as Theodore himself remarks, from the Valentini Gnosis. According to the “ Socrates,” “it was his will that ches ( should be divided and should coin and constitute bodies (pagn a animate bodies) out of the four elements of the powers and acco ing to the number of those seven Governors and Servants of hi 6 The MS. has as the verb an Ethpeal or Ethpaal perfect from the psq. According to Payne-Smith, this verb is used only in Pael and E and in the active means “make easy,” “expound,” “translate.” These id cannot be fitted into the above context. Ryssel translates und ganz und , erleichtert (d. h. von fremden Substanzen befreit) werde zu etwas Unthet baren und Unerkennbaren. I have emended it to read ’ethpéSet; comp: Eusebius “ Theophania,” p. 12, Lee, where the same verb is used to describ the activity of the Logos in the universe—“ Throughout the universe spread (pésat) himself, above in the height, below in the deep, himself, in- corporeal, he extended (méthah).” NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” 107 One may note in passing that the word here used to denote the planets, which I have translated “Governors” (madhbérane), is used in the same sense in the “ Book of the Laws.” ‘The “ Socrates,” then, ascribes the organization and constitution f the universe to the soul, not to the Logos. Ephraim (“ Adv. Haer.,” III., lines 102-110, 125-134) rails bitterly against Bardaisan _ for denying the orthodox doctrine that the Logos constructed the __ universe and asserts that he represented Wisdom as acting as God’s agent in the work of creation. Thus the soul of the “Socrates” corresponds to Wisdom in Bardaisan’s system. The cosmology of the “Socrates” resembles, therefore, that of -Bardaisan in several important features, especially in denying the orthodox doctrine of creation out of nothing (the words “create,” creation,” “creature” do not occur in it at all) and representing world as made out of eternally existing elements ; in recognizing fifth something, Darkness, which corresponds to the Aristotelian ‘irst Matter; in describing the origin of chaos; in regarding the world-process as essentially the resolution of chaos into cosmos; in regarding evil as nothing but the unregulated conflict of eternally existing and opposed attributes. But the two systems differ in other important features, and it is quite certain that the “ Socrates ” s not the source from which Ephraim and Theodore drew their nformation about Bardaisan’s cosmology. Postscript—tIn view of the uncertainty of accomplishment which in these troubled times attaches to all activities not contributing to the war, I have decided to add to the above paper a brief statement of the conclusions which I had reached when it was written but ~ withheld in anticipation of the new evidence promised by Mr. Mitchell. Limitations of space will permit only brief reference to _ the sources, but those that are interested will have no difficulty in __ verifying them. 2 _ Bardaisan wrote dialogues against Marcion and others (Euseb., “Hist.” IV., 30) and many other works, some of which may well have been dialogues; the “Socrates” is one of these. It is the source from which Ephraim drew his knowledge of Bardaisan’s theory of soul ; it manifests in conspicuous degree the “ patience and es 108 NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” polite answers to every man” for which Bardaisan’s disc praised him (Philoxenus ap. Cureton, “ Spicilegium,” p. v) cosmology is closely akin to and i in no point inconsistent with : of the chief source. That the “Oration of Melito” (Cureton, op. cit., pp. 22- Syr.) is also the work of Bardaisan is rendered extremely proba by its close affinity to both the “ Socrates” and the “ Book of | Laws.” The “Socrates” teaches that knowledge of God is impli in every man by virtue of his relation to the “ Original Root.” the “Oration” a similar doctrine is both stated in general terms (e. g., compare “Or.,” p. 30, 15, with “S.” p. 161, 16) and | rectly applied to the Emperor, e. g. (p. 25, 24): “ But thou, a fre intelligence and cognizant of the truth, enter into thyself;” (p. 27 14) “Know thyself and thou shalt know God;” (p. 29, 16) “But thou, feeble man, within whom He is and without whom He is and above whom He is.” Its relation to the “ Book of the Laws” i: even closer; with the latter it insists upon the doctrine of free will and makes extensive use of material drawn from the history anc customs of foreign nations in the same curious and characteristic way. The Emperor to whom it is addressed is without doubt Cat calla. He was generally known during his life as simply “ Ante ninus”’; he spent the winter of 216-17 in Edessa; he was wont seek out and consult astrologers, and Bardaisan, who had been an intimate friend of the late king Abgar IV., must have been brought into touch with him. The dialogue on Destiny which Bardaisan dedicated to him, which M. Nau is right in distinguishing from “Book of the Laws” (“Le Livre des Lois,” pp. 11-12), and the “Oration,” were probably both among the results of the personal relation thus established. One may also recognize in some of the bold characterizations of the “Oration” leading traits of Ca calla; compare, for example (25, 25), “if they clothe thee in the fashion of a woman remember that thou art a man” with Dio Cassius’s description of the effeminate appearance which Caracalla affected while in the East—his removal of his beard at Antioch (Dio, 77, 20) and the barbaric long-robed costume of his own designing which he wore in Mesopotamia (Dio, 78, 3, 3); com- — NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” 109 pare also (“Orat.,” p. 27, 26) “Therefore thou rollest thyself upon the ground before demons and shadows and askest vain petitions from one that hath nought to give” (and also p. 29, 25) with Dio’s account (77, 15, 5-7) of Caracalla’s vain efforts to recover his health by assiduous devotion to the gods. As regards the bearing of these conclusions upon the hypothesis which I suggested some years ago (“Bardaisan and the Odes of Solomon” in the Journal of Biblical Literature, 1911), that the Odes” were written by Bardaisan, I can only say that no infer- ences can be drawn until the “Odes” have been given much more careful study than they have yet received. They are certainly more deeply tinged with Valentinian ideas than are the above three -works—much more deeply, indeed, than I supposed when I pub- lished my first study of the problem—and many of them are as yet but imperfectly understood. I may, however, remark that the ref- erences to persecution in the “Odes” (e. g., 5, 8, 29) would be perfectly appropriate to the situation above supposed; especially would the perplexing allusions of the 29th be at last intelligible. The boldness of the “Oration” must have inflamed Caracalla’s avage temper to the highest degree and it was no doubt after its delivery that Apollonius, “the friend of Antoninus” (Epiph., Haer.,” 56), demanded of Bardaisan that he renounce his faith, and received an uncompromising refusal. Epiphanius says that on that occasion Bardaisan very nearly attained the rank of a con- fessor. It is probable that he would have attained the still higher rank of a martyr if Caracalla had not been assassinated, April 8, 217, while making a trip from Edessa to Harran. Compare with this situation the language of the 29th Ode: 4 He has raised me from the depth of hell and from the mouth of death has drawn me; 5 I have brought low my enemy and He has acquitted me by His grace. ... lac a eh ial Keni ale in ee i 7 He showed me His sign and guided me by His light; He gave me the rod of His power 8 that I might subdue the thoughts of the peoples, to bring low the prowess of warriors, 110 |NEWBOLD—THE SYRIAC DIALOGUE “SOCRATES.” _ 9 to make war by His word, to take victory by His might. 10 The Lord cast down my enemy by His word Bes and he became like the chaff which the wind carries of, Another scrap of evidence, from a quite different source, points | in the same direction. The sixth Ode contains a simile which has caused no little perplexity : 7 For a stream went forth and became a river great and broad, 8 For it overwhelmed everything and shattered and brought (them) to the Temple. 9 And the restraints of men could not restrain it nor the arts of them that restrain waters. This river, it appears, is the Gospel. But why does the Gone bring its conquests to the “Temple”? What is the “Temple”? And why this curiously specific allusion to the hydraulic engineers? In the Edessene Chronicle one finds a contemporary account of a flood which devastated Edessa A.D. 201 (BO, L., 390-91). A spring within the palace grounds overflowed and inundated th | palace. “While the wise men were inlore what they should do to the flood of water which was increasing” a heavy rain came on during the night, the river Daisan overflowed its banks and formed a deep lake which finally overtopped the west wall of the city and poured over the battlements. King Abgar ordered the sluice-gates to be opened, but it was too late—the wall collapsed, the flood destroyed the palace “and the waters swept away everything bef them, the fair and beautiful buildings of the city, everything nea the river southward and northward, and they also made an on- slaught on the temple of the congregation of the Christians.” U: fortunately the word used’ does not indicate the amount of damage to the “temple” of the Christians, but from the very fact of ambiguity, following as it does unambiguous words, and from th order of the narrative one may infer that the damage fell short ry complete destruction. Manifestly, this is precisely the situation depicted in the Ode—the building used by the Christians of Edessa 7 srhw may signify any amount of injury from a mere attack upon to- total destruction. THE SYRIAC DIALOGUE “SOCRATES.” 111 Ey eee rship was popularly known as their “temple,” a great hydraulic engineers had striven in vain to control car- hed, as a symbol of the triumphal progress of the Gospel which »s through the world, overcoming all obstacles, and brings its ves into the Church. _ Unrversity oF PENNSYLVANIA, April 3, 1918. BIOCHEMICAL STUDIES OF THE PITCHER LIQUO NEPENTHES. By JOSEPH SAMUEL HEPBURN, A.M, MS., Pa.D. (Read April 14, 1917.) ins the pitcher of N epenthes gradually develops, a ig secreted and occupies the lower portion of its cavity. After operculum, or lid, has opened, insects are attracted by the ne which is secreted by glands. The nectar glands are found on outer surface of the pitcher, more abundantly on the inner surface the lid, and on the inner edge of the corrugated rim that surro the margin of the pitcher. The insects, thus attracted, are tempte down into the pitcher, and pass to a richly glandular zone with smooth surface—the so-called detentive surface—on which they | their footing, and are precipitated into the liquor. The << then digested by the liquor. = Two theories exist as to the manner in which digestic mele each theory is supported by experimental evidence. Hooker,? Tz von Gorup and Will,? Vines,* Goebel,® Clautriau,® and Fenner? ¢ cluded from their researches that the digestion is due to an er Site 1 Hooker, Nature, 1874, X., 366-372. Report of the Forty-fourth Mee of the British Association for the Advancement of Science, 1874; Notes Abstracts of Miscellaneous Communications to the Sections, 1875, pp. 102- 2 Tait, Nature, 1875, XII., 251-252. 8yon Gorup and Will, Sitzungsberichte der physikalisch-meiieanas Societit zu Erlangen, 1875-6, VIII., 152-158. Ber. der deut. chem. a ' 1876, IX., 673-678. - 4Vines, Jl. of Linnean Society, Botany, 1877, XV., 427-431. Annet Botany, 1897, XI., 563-584; 1898, XII., 545-555; 1901, XV., 563-573. 5 Goebel, Pflanzenbiologische Schilderungen, 1893, II., 186-193. 6 Clautriau, Mémoirs couronnés et autres mémoires, publiés par Acad Royale des Sciences, des Lettres et des Beaux Arts de Belgique, Collectic in 8°, 1899-1900, LIX., third memoir, 56 pages. 7Fenner, Flora oder allgemeine botanische Zeitung, 1904, XCIILI., 33. 434 (especially pp. 358-363). 112 : HEPBURN—THE PITCHER LIQUOR OF NEPENTHES. 113 present in the pitcher liquor. On the other, hand, Dubois* and Tischutkin® concluded from their experiments that the digestion is due to the activity of microédrganisms. A third factor to be con- sidered is the autolysis produced by the tissue enzymes of the cap- tured insects. _ In the present research, the proteolytic enzyme of the pitcher liquor and the bacteria, which occur in. opened pitchers, have been studied separately. The following species and hybrids of Nepen- thes supplied material for the research: ampullaria, atrosanguinia, Chelsonii, Claytonii, Dominii, Dyeriana, gracilis, Hamiltoniana, Hen- ryana, Hookeriana, Mastersiana, mixta, Morganiana, paradise, Raffiesiana pallida, rufescens, splendida, Wittei. The plants were grown in the Nepenthes House of the University of Pennsylvania. PROTEASE OF THE PITCHER LIQUOR. _ In the study of the protease of the pitcher liquor, pitchers were ways selected prior to opening. They were closely watched and the mouth of each pitcher was closed with absorbent cotton as soon the lid opened; the entrance of insects was thereby prevented, possible contamination of the pitcher liquor by the tissue en- nes of the digested prey was entirely excluded. The digestion experiments with the pitcher liquor were made in vitro in the pres- ice Of a bactericide; bacterial action was thereby excluded. The teolysis, which was observed, was, therefore, due to enzyme Liquor from both non-stimulated and stimulated pitchers was udied. The experiments on liquor from non-stimulated pitchers ere carried out as soon as possible after the opening of the pitchers. When liquor from stimulated pitchers was desired, recourse was had to mechanical stimulation by chemically inert substances. some experiments, the glands of the inner wall of the pitcher rere stroked repeatedly with a camel’s hair brush, and the cotton ug was then inserted in the mouth of the pitcher; the liquor was emoved for study on the following day. In other experiments, eral round solid glass beads, such as are used in fractionating _ § Dubois, Comptes rend. de Academie des Sciences, 1890, CXI., 315-317. ®Tischutkin, Botanisches Centralblatt, 1892, L., 304-305. 114 HEPBURN—BIOCHEMICAL STUDIES OF columns, were inserted into the newly opened pitcher; the plug was introduced, and the pitcher and its contents were sl thoroughly at intervals during one or more days, taking care. wet the cotton and thereby lose liquor; the liquor was -“ moved and used in digestion experiments. The volume of liquor secreted by a single pitcher was s ah so small that liquor could not be obtained from the same pitcher — both before and after stimulation. Since two pitchers rarely n tured on the same plant at the same time, it was impossible to ma a comparative study of the liquor from both non-stimulated | stimulated pitchers of the same plant. While the differences, to individual plants, could not be entirely eliminated, the prob was attacked by several methods for the study of proteolysis, a number of experiments were conducted according to each mi the results obtained by all the methods lead to gi same gé conclusions. Either sodium fluoride or trikresol was used as a bacberierde all the experiments reported below. When sodium fluoride used, sufficient solid fluoride was added to the mixture of pi liquor and substrate to render the final concentration of s fluoride one per cent. When trikresol was used, a sufficient vo of a two per cent. aqueous solution of trikresol was added to r the: final concentration of trikresol 0.2 per cent.—a concent their experiments with proteases. Whenever the mixture of pite cl liquor and substrate was diluted to a definite volume, the trib solution was added before the dilution to the final volume was n Unless otherwise stated, the temperature of incubation 37° In each experiment, a blank or control experiment was out with pitcher liquor which had previously been boiled, cooled to the temperature of the room; the control experiment carried out in exactly the same manner, in all other respects, as determination proper. The control was always compared with tl determination proper; and due allowance was thus made for the 10 Graves and Kober, Jl. Am. Chem. Soc., 1914, XXXVI., 751-758. ‘THE PITCHER LIQUOR OF NEPENTHES. 115 é action of any thermostable catalyst present in the pitcher | sor, and also for any action of the reagents on each other. The following reactions for the detection of a protease were 1. The formol-titration of Sorensen. 2. The digestion of: (a) Carmine fibrin. (b) Edestan. (c) Protean of castor bean globulin. (d) Ricin (Jacoby). 3. The cleavage of glycyltryptophane. FORMOL-TITRATION. Composite samples of liquor, obtained from several pitchers, used, a different sample for each experiment. The indicator phenolphthalein. TABLE I. ForMOL-TITRATION AFTER DIGESTION. Total g Volume} Volume) Period a wees ot ar Di of Formol-titration Liquor from Substrate. of Sub-| Pitcher] gestion | Incuba-| after Deduction strate, | Liquor,| Mix- | tion, of Blank, Gram. | Cc, ture, | Days. : Cc. timulated Ovalbumen....| 0.05 | 10.0 | 10.0 4 | 0.00 cc. 0.1 N NaOH Nahrstoff 0.25 1° 12-5 | 3755 4 |0.00 cc. o.1 N Heyden* NaOH : Witte peptone..|; 0:25 | 12.5 | 37-5 4 | 0.00 cc. 0.1 N e ; NaOH 4 d pitchers.| Ovalbumen....) 0.05 | 15.0 | 15.0 3 | 0.15 cc. 0.05 N i : NaOH if Wibea SOS eee 0.05 | 15.0 | 15.0 14 |} 0.45 cc. 0.1 N 18 NaOH k Ovomucoid....) 0.05 55° 55-5 6° | 0.10 C.c.:0.1-N i NaOH i Nahrstoff 0.15 | 15.0 | 30.0 Sc GA er Ors Heyden* NaOH Witte peptone... 0.25 | 25.0 | 50.0 A125 Ce. ON, NaOH DP Acouting to Gotschlich (Kolle und Wassermann, Handbuch der enen Mikroorganismen, 2 Auflage, 1912, IL., 102), Nahrstoff Heyden 1 is nixture of different albumoses. 116 HEPBURN—BIOCHEMICAL STUDIES OF In the first series of experiments, reported in detail in Ta the pitcher liquor was permitted to act on a given substrate time stated. Undissolved protein was then removed by fil and was washed on the filter; the combined filtrate and ings were made neutral in reaction. Any precipitated metapro was separated by filtration, and was washed on the filter ; the bined filtrate and: washings were neutralized, if necessary half their volume of a neutral, 40 per cent. formaldehyde was then added, and the carboxyl groups of the amino acids immediately titrated with standard sodium hydroxide solution. — should be noted that no acid was added in the digestion exp: pel of Table I. = tion during the digestion. The details are cee in Table IL TABLE Il. ForMOL-TITRATION AFTER DicESTION oF EDESTAN. Volume | Total ee Volume} Weight |of 0.1 V| Volume Volume] Period |tra ; _of of |Hydro-| of | of Di- | of Incu-| De Liquor trom Pitcher | Fdestin,| chloric | Water, | gestion | bation, | ¢ b Liquor, | Gram. | Acid, | C.c. |Mixture,| Days. | C. C.c. Cc, City Roe Non-stimulated pitchers...... II.5 | 0.100| 15.0 | 23.5 | 50.0 28 | 8.0 | 0.036 5-4 | 11.6 , 25.0 at of: Stimulated pitchers.......... 8.0 | 0.050! 7.5 | 17.5 | 33-0 254k In both series of experiments the liquor from stimulated pi cl invariably digested the substrate, liberating compounds which of the nature of amino acids, and responded to the formol- a The liquor from non-stimulated pitchers did not digest the s with the production of such compounds, for the formol-t invariably was zero. ' : DIGESTION OF CARMINE FIBRIN. Two sets of experiments were made using carmine fibrin as substrate. In both series, the temperature of incubation was of the room. The liquor from a separate pitcher was used in Bik aha Pe nl este by gee ome THE PITCHER LIQUOR OF NEPENTHES. 117 “liquor, obtained from several pitchers, were used. gents. The time required for solution of the substrate and details of each experiment are given in Table ITI. TABLE III. Dicestion or GELATINOUS CARMINE FrsrIn By PitcHER Liguor IN THE PRESENCE OF 0.2 Per Cent. HyprocHLoric AcIp. t Vuene of Volume of Solution of Substrate. k oe cm tole. Gz meee in Complete in : : ours. Hours. ita b mulated pitchers... . 1.50 0.10 13 e *4.75 0.25 31 4 ted pitchers........ 1.50 0.10 ; 13 E 4.00 0.50 15 24 ’ 4.00 0.50 48 144 *4.75 0.25 26 In the final series of experiments, 0.2 gram of carmine fibrin weighed out into a separate tube for each experiment. This es falls into three groups. In group A, the carmine fibrin was len in its tube in a solution containing 0.2 per cent. hydrochloric and 0.2 per cent. trikresol, then was transferred to another and immediately used in a digestion experiment. In groups and C, unswollen carmine fibrin was used. No hydrochloric acid as used in group B; in the other two groups, sufficient 0.6 per . hydrochloric acid was added to render the final concentration that acid 0.2 per cent. Trikresol was used as a bactericide in all e groups, adding sufficient of its 2 per cent. aqueous solution to roduce a final concentration of 0.2 per cent. The details of these iments are recorded in Table IV. | PROC. AMER. PHIL. SOC., VOL. LVI, I, JUNE 14. 1918. 118 . | HEPBURN—BIOCHEMICAL STUDIES OF | TABLE IV. TimME REQUIRED FOR DIGESTION OF 0.2 Gus OF Cine BY THE PitcHeR Liquor. s Volume & in C.c. | Time Required for Com- 8 Liquor from of plete Solution of ro) Pitcher Substrate, Liquor, A .| Stimulated pitchers...... S56 48 hours 2.5 72 “ 2.0 Co Rea ak rrr I.0 5 ks RO Sons B .| Non-stimulated pitchers. .| 1.0 | Substrate absolutely is unattacked after in- 2.5 cubation for 70 days 3.0 3-5 5:5 Time required for marked digestion of substrate C .| Non-stimulated pitchers..| 2.5 16 hours I.0 + Baie had absolutely no digestive action on that substrate in tl of that acid. DIGESTION OF EDESTAN. The solution of edestan was prepared by dissolving o. edestin in 15 c.c. of o.1 N hydrochloric acid, previously dih water to a volume of 25 c.c. The details of the various ments are given in Table V. After incubation of the mix pitcher liquor and edestan solution, it was rendered neutral to nolphthalein by addition of 0.1 N sodium hydroxide solution. — In the experiments with liquor from stimulated pitchers precipitate formed in the determination proper on neutralizatior showing that the digestion of the protean edestan had adva ace beyond the metaprotein stage. In the experiments with li from non-stimulated pitchers, a precipitate always formed in - THE PITCHER LIQUOR OF NEPENTHES. 119 tion proper, but it always was decidedly less voluminous nan the precipitate in the corresponding control experiment ; there- ee digestion of the substrate had occurred, though less rapidly than meen liquor from stimulated pitchers was used. G5 EMEA IT Se SEI TABLE V. DIGESTION oF EDESTAN BY THE PITCHER Liguor. Si EE RR ER Volume! Total | Volume! of | Volume| Volume! Period ae : of |Edestan of | of Di- | of Incu-| Precipitate on Neutral- Liquor from Pitcher | Solu- | Water, ion | bation, |_,ization of Experiment re tion, ee Nag Days. Proper After Incubation. cose Cc; Ce: : ‘Non-stimulated pitchers...) 1 2 2 5 8 | Precipitate one half : te as great as in i blank F 4 I ° 5 13 | Precipitate one half t as great as in : blank ; ed pitchers....... 20 25 5 50 14 | No precipitate f I 2 2 5 8 | No precipitate & t _ DIGESTION OF THE PROTEAN OF CASTOR BEAN GLOBULIN. A 2 per cent. solution of castor bean globulin in 5 per cent. dium chloride solution was used. This solution was filtered, if , then mixed with the pitcher liquor ;0.1 N hydrochloric , acid was next added; a cloudy precipitate of the protean derived from the globulin formed. The presence in the pitcher liquor of proteolytic enzyme, active in the presence of hydrochloric acid, was shown by the digestion or solution of the protean, the cloudy precipitate gradually becoming less dense, and finally disappearing completely. Liquor from a separate pitcher was used in each ex- periment ; the details are recorded in Table VI. The protean was usually dissolved by the liquor from both non-stimulated and stimu- lated pitchers. 7 cama eas dca li aia Cee alae DIGESTION OF JACoBY’s RICIN. The reagent was prepared by dissolving 1 gram of Jacoby’s _ficin and 1.5 grams of sodium chloride in 100 c.c. of water, and filtering, if necessary. The test was carried out by mixing I c.c. of pitcher liquor and 3 cc. of ricin solution, adding 1 c.c. of 0.56 cent. hydrochloric acid, then incubating. A cloudy precipitate 120 HEPBURN—BIOCHEMICAL STUDIES OF TABLE VI. ; DIGESTION OF THE PRoTEAN or Castor BEAN GLOBULIN BY THE PIT Total Volume| Volume | Volume Volume of of ~ lofo.r V| Volume) of pj- Liquor from Pitcher |Globulin| Hydro-| of Liquor,| Solu- | chloric | Water, aT RS 5. | Hons | Acid Cielo ae nag C.c, Ce? Cc. Non-stimulated pitchers...| 0.6 2.0 0.5 1.9 5.0 2.5 2.0 0.5 0.0 5.0 1.0 2.0 0.5 1.0. 4.5 1.0 2.0 | 05 1.0 4-5 Stimulated pitchers....... 2.5 2.0 0.5 0.0 5.0 0.5 4.0 1.0 4.5 | 10.0 diventea or dissolved. In one experiment, Miquor from. a non-stimulated nreciphaes ‘was nasndee digested in two days. from both non-stimulated and stimulated pices exert teolytic action on the protean. CLEAVAGE OF GLYCYLTRYPTOPHANE. Liquor from stimulated pitchers was permitted to act on gl tryptophane in the presence of toluene as a bactericide; I0 c., pitcher liquor and 2 c.c. of an ageous solution of the dipeptide called Ferment diagnosticum) were used. In one experir THE PITCHER LIQUOR OF NEPENTHES. 121 vage of the dipeptide had not occurred after digestion for nine ys in the incubator. In a second experiment, after digestion for twenty-one days in the incubator, followed by seven days in the ‘room, a distinctly positive reaction for free tryptophane was ob- tained by the bromine and acetic acid test. Hence, liquor from stimulated pitchers apparently hydrolyzed glycyltryptophane, pro- vided the period of incubation was sufficiently long. BACTERIA OF THE PITCHER Liguor. _ The bacteriological study was made in collaboration with E. intard St. John. Unopened and opened pitchers were studied _ Unopened Pitchers——Sterile scissors were used in cutting the plant tissues. The prolonged midrib or tendril, which carries the er, was severed; the top portion of the pitcher was rapidly passed through the flame and was immediately cut off. The cut e of the pitcher was then flamed rapidly; and the liquor was thdrawn at once with a sterile pipette, and plated on plain nu- trient agar. After incubation for four days at 37° C., the plates ere examined for bacterial growth. Twelve pitchers were studied in this manner, the liquor from pitcher being plated separately. Colonies invariably failed to develop on the plates; hence the liquor in unopened pitchers was c. _ Opened Pitchers—Partly opened pitchers, which had not been invaded by insects, were used in two experiments. The liquor tom each of these pitchers contained a goodly number of bacteria hich grew on plain nutrient agar at 37° C. _ All the other experiments were conducted on liquor from open, active pitchers in which insect remains were present. Total Count.—With each of five pitchers, several successive di- lutions of the liquor were sown on plain nutrient agar, and the plates were incubated at 37° C. for four days; the colonies were then counted. The number of bacteria per c.c. of pitcher liquor as low as 48,000 in one pitcher, and as high as 8,000,000 in another pitcher; the other pitchers gave values: 450,000, 1,200,000, and 1,900,000, respectively. 122 HEPBURN—BIOCHEMICAL STUDIES OF Several of the colonies were studied separately (a) by s smears, and (b) by transfers to lactose bile salt broth microorganisms were rod-like, and a few of them contain none of the transfers developed gas ; therefore it may be conch that members of the family Bacteriacee, other than the aérogenes group were present. The liquor in an old pitcher, which was becoming brov top, gave a count of 104,000 bacteria per c.c. Liquefaction of Gelatin—tThe liquor from each of two was sown on nutrient gelatin; the bacteria grew and ee r liquefied the gelatin in forty-eight hours. : Tests for the Colon-aérogenes Group.—In two experi | c.c. of liquor from a single pitcher was sown in lactose bil bouillon ; on incubation at 37° C., gas developed within seventy-tw hours, showing the presence of organisms of this group. — posite sample of liquor, collected from several pitchers, re in this medium in several successive dilutions, the greatest ¢ being 1: 10,000; gas developed in even the greatest dilution seventy-two hours, therefore at least 10,000 organisms of the 0) aérogenes group were present in each c.c. of the liquor. — Certain special media were used in the study of the poe tions. A stock solution of inorganic salts, containing mapa ferrous, potassium, chloride, sulphate, and phosphate ions, prepared as directed by these authors, and was used in the 1 diz The liquor from each pitcher was studied separately in. all experiments described below. The temperature of incubation : always 37° C. Bs Production of Tryptophane and Indol from Protein.—Pre obtained from aleuronat, gave a purple color with glyoxylic < and sulphuric acid, and therefore contained a tryptophane group. medium, containing 0.4 gram protein, 20 c.c. 0.1 N sodium hy- droxide solution, and 80 c.c. of the stock solution of inorganic sz ts, was prepared and sterilized. One c.c. of pitcher liquor was so v 11 Crabill and Reed, Biochem. Bull., 1915, 1V., 30-44. THE PITCHER LIQUOR OF NEPENTHES. 123 in 10 c.c. of the sterile suspension of protein. In one series of eight experiments, indol had not been produced after incubation for ten days. In another series of eight experiments, neither free trypto- i phane nor indol was present after incubation for twelve days. _ Digestion of Proteins by the Bacteria.—The proteins used were: casein, egg albumen, unswollen carmine fibrin, edestin, ricin (Ja- coby), protein (prepared from aleuronat). The media were pre- pared by addition of 2 per cent. of agar and approximately 1 per _ cent. of one of the proteins to the stock solution of inorganic salts, and were then sterilized. The proteins, therefore, served as the sole source of carbon and nitrogen for the bacteria. These media were used in plating experiments, sowing I c.c. of pitcher liquor in each plate. Whenever proteolytic bacteria were present in the _ pitcher liquor, their colonies gradually digested and dissolved the suspended particles of protein over which they grew. The plates were examined at intervals until drying of the media rendered further observation useless. The bacteria grew and colonies de- veloped on the vast majority of the plates, as may be seen by refer- ence to Table VII., which gives certain details of these experiments. TABLE VII. DIGESTION OF ProTeIN Mepra By BAcTERIA, PRESENT IN LiIguor oF OpeN PITCHERS. eee a . Drape. | Newest * | Growth Oc-| Digestion |Total Period Protein Used. ments. lments Show.| cared in Begun Be- | of Incuba- Plated. | ing Growth. Days. tween Days.| tion Days. 3 3to9 3 3 z 3 to5 i2 See albumen... .. 2.22.22: 15 II 12 to 14 Sarnioe fibrin. .2.. 2... ...- 6 6 5 to9 9 MET Se to So 4 4 3to9 12 MR re Sa) siete the hae 3 2 3to5 |Byothday 12 Protein (from aleuronat). .... | 8 8 3 3to5 12 _ The suspended egg albumen was not even partially digested ; possi- _ bly this was due to the presence of non-coagulable ovomucoid, and its utilization by the bacteria as a source of carbon and nitrogen. The other suspended proteins were gradually digested, the digestion becoming more marked as the period of incubation increased ; how- ever, complete digestion and disappearance of the suspended parti- cles had never occurred by the end of this period. 124 HEPBURN—BIOCHEMICAL STUDIES OF Sufficient liquor was not always obtained from a singl to permit plating on all six media. However, a general was noted that, when the microdrganisms present in liquor grew on one of these media, they grew on all the usually exerted a visible digestive action on all the prot Production of Basic Compounds.—A study was als concentration equal to that of he asparagin. These media always sterilized by the discontinuous method. Plating ex: expe ments were made, sowing I c.c. of pitcher liquor i in each plate. production of basic compounds by microorganisms growing on th media was indicated by a red color of the medium beneath surrounding the colony. Sterile plates were always poured trols, to be used in determining the changes in color of the ¢ expe ments proper. oe The number of experiments plated on each of these met Glycocoll rosolic aGi@ a@ai. ooo... is cc ce eae 7 experiments Acetamide rosolic acid agar...........s.s0s 7 experiments. — Asparagin rosolic acid Gar... .......5...<5 22 experiments. | Ammonium lactate rosolic acid agar......... 7 experiments. — The period of incubation was from ten days to a fortnight plates being inspected at intervals until further observation 1 rendered useless by drying of the media. Colonies developed in save one of the experiments, the sole exception being a plate glycocoll rosolic acid agar. The colonies were usually apparent the third day; though, in a few instances, they grew so slowly t they became apparent only at the tenth day. The organisms alw THE PITCHER LIQUOR OF NEPENTHES. 125 _ produced an alkaline reaction, 7. e., a reddening of the medium. _ This red coloration gradually spread from the colonies as centers over the entire plate, and was quite marked, as a general rule, by the third or fifth day. In those experiments in which colonies developed but slowly, the red coloration was noted by the tenth day. Hence the microdrganisms produced basic-compounds from the substrates. During the last portion of the period of observation, from the tenth to the fourteenth day of incubation, ofttimes the _ medium changed in reaction and became acid to rosolic acid, never- theless the colonies themselves remained alkaline. Included in the above experiments was a group in which the liquor from each of seven pitchers was plated on all four rosolic acid agars—glycocoll, acetamide, asparagin, and ammonium lactate. The bacteria grew and produced an alkaline reaction at about the same rate in all four media. In these experiments, a record was also kept of the odor of the cultures; quite frequently the plates were characterized on the third day by an odor recalling that of ammonia or amines; this odor was rarely present on the tenth day. Hence the microdrganisms were able to utilize glycocoll, aceta- mide, asparagin, and ammonium lactate, which formed their sole source of carbon and nitrogen, and were able to produce basic nitrogenous compounds from these substrates. 2 IN ED PEER OSPR met GENERAL SUMMARY. The following conclusions are based on the chemical studies: Using the formol-titration, it was found that liquor from non- stimulated pitchers lacked proteolytic power, while liquor from stimulated pitchers produced proteolysis of a number of substrates: ovalbumen, fibrin, ovomucoid, Nahrstoff Heyden, and Witte pep- tone. In the presence of very dilute hydrochloric acid, edestan was digested by liquor from stimulated pitchers, but not by that _ from non-stimulated pitchers. Carmine fibrin was not dissolved by liquor from non-stimulated pitchers in the absence of acid, but was digested and dissolved by the liquor from both non-stimulated and stimulated pitchers when 0.2 per cent. of hydrochloric acid was present in the reaction mixture. e 126 HEPBURN—BIOCHEMICAL STUDIES OF Edestan was digested by the pitcher liquor in the pr en “e very dilute hydrochloric acid; liquor from stimulated pitche duced a more rapid digestion than did liquor from non-s' pitchers. and stimulated pitchers in the presence of very dilute hy acid. a“ Liquor from stimulated pitchers apparently prod of glycyliryptophane, when the period of incubation was st long. Liquor from non-stimulated pitchers exerted proteoly: on only in the presence of acid, failing to produce proteolysis in ‘tl absence of acid. Liquor from stimulated pitchers exerted proteolytic 4 both the presence and the absence of acid. The manner in which stimulation causes the pitch ‘tig 101 acquire active .proteolytic power is a field for further res Stimulation may produce a change in the hydrogen ion conce tion’? and thus render the reaction favorable for the actiy | protease already present in the pitcher liquor; or it may | : activation of a zymogen already present; or it may give rise to increased secretion of protease by the glands of the pitcher. In the presence of acid, certain substrates—especially edest were digested by liquor from stimulated pitchers more rapidly by liquor from non-stimulated pitchers. The following conclusions are drawn from the bacteriole y studies : . Liquor taken aseptically from unopened pitchers was sterik 12 Since this paper was presented, an abstract of a monograph by Je Hempel entitled “Bidrag til Kundskaben om Succulenternes Fysiolo (Copenhagen, H. Hagerup, 1916, 147 pp.) has appeared in Physiological stracts, 1917, II., 146 (issued in May, 1917). The following quotation is tz from this abstract. “The sap of the stimulated pitcher of Nepenthes gi values for the hydrogen ion concentration greater than 10-7, but unstim pitchers give no definite value.” THE PITCHER LIQUOR OF NEPENTHES. 127 _ A goodly number of bacteria were present in the liquor of pitchers which had partly opened, but had not yet been invaded by insects. The bacterial content of the liquor of open active pitchers, which contained insect remains, was quite high—from 48,000 to 8,000,000 per c.c. of liquor; the organisms were rods. These bacteria lique- fied gelatin, and formed colonies on solid media (agar) in which the sole source of carbon and nitrogen was either a protein or a simple organic compound. They usually digested the protein (casein, egg albumen, carmine fibrin, edestin, Jacoby’s ricin, protein from aleu- ronat), but at an exceedingly slow rate. They decomposed the _ simple nitrogenous organic compounds (glycocoll, acetamide, as- paragin, ammonium lactate), frequently producing an odor like that of ammonia and amines, and always imparting an alkaline reaction to the medium; this reaction subsequently changed to acid in some ‘experiments, but the colonies always remained alkaline. The bac- teria did not produce either tryptophane or indol from aleuronat ‘protein, when sown on a medium in which the latter was the sole source of carbon and nitrogen. Organisms of the colon-aérogenes group were present; on the average, each c.c. of pitcher liquor con- tained 10,000 organisms of this group. _ The chemical and et studies taken together lead to these conclusions. The protease of the pitcher liquor is the chief factor in the diges- tion of the insects in the pitcher. The bacteria, which occur in the liquor of opened pitchers, play merely a secondary part, as is shown by the slowness with which they digested proteins. The bacteria and the Nepenthes plant live in symbiosis ; the bac- teria obtain their food from the digested insects and assist, to a limited extent, in the digestion of the insects. ___. The tissue enzymes of the insects, of course, may assist in the _ digestion by causing the insects’ tissues to undergo autolysis. NoTE ON THE BIOCHEMISTRY OF THE PITCHER LiQguoR oF SARRACENIA. Closely related to the family Nepenthacee, with its single genus Nepenthes, is the family Sarraceniacee, consisting of three 128 HEPBURN—BIOCHEMICAL STUDIES OF genera: Sarracenia, Darlingtonia, and Heliamphora.** The ing experiments were made on the pitchers of two species o cenia—flava and minor—by Frank M. Jones, E. Quintard St and the author, and are mentioned in this place, since they ine that the digestive action in the pitchers of Sarracenia is — to a proteolytic enzyme. 2 Liquor was obtained from unopened pitchers of Sa rrace flava, growing in their native habitat, and was used in test-tube e: periments. The liquor digested edestan in the presence dilute (less than 0.1 per cent.) hydrochloric acid, and rapidly gested carmine fibrin—swollen or unswollen—in the presence 0.2 per cent. hydrochloric acid, 0.2 per cent. of trikresol being use as a bactericide. Liquor was also obtained from open pitchers; it had been diluted by rain water, but rapidly digested carmine fibri in the presence of hydrochloric acid and trikresol, the conbenttrat of these reagents being that just stated. By means of culture experiments, it was determined that contents of unopened pitchers of Sarracenia flava and Sarrace minor (Sarracenia variolaris) were bacteriologically sterile. The contents of opened pitchers of these species, which tained insect remains, were also studied bacteriologically. The te for the liquefaction of gelatin was conducted as directed by Ri the medium was liquefied ; and both motile and non-motile roc microOrganisms were recovered from the resulting cultures. contents of these pitchers were also plated on certain of the gs; agar media described in the preceding pages. The bacteria on casein agar and on protein (from aleuronat) agar, and dig these proteins, but at an exceedingly slow rate. The bacteria < grew on the various rosolic acid agars—glycocoll, acetamide paragin, and ammonium lactate—changing the reaction of the n dium to alkaline, and producing an odor of ammonia or of b on prolonged incubation the reaction changed to acid. Organis of the colon-aérogenes group were found to be present by t reaction with lactose bile salt bouillon. 13 Macfarlane, Engler’s Pflanzenreich, 1908; IV., 110, Sarraceniacez, — Heft; and IV., 111, Nepenthacee, 36 Heft. 14 Rivas, Jl. Am. Med. Assoc., 1908, L., 1492-1495. THE PITCHER LIQUOR OF NEPENTHES. _ 129 These ecerinents indicate that in Sarracenia, as in Nepenthes, protease of the pitcher liquor plays the leading role in the ion of the a insects. The bacteria of the pitcher liquor _ The author desires to record his deep indebtedness to Dr. John M. MacFarlane, director of the botanic garden and laboratory of this university, for the material and facilities placed at his disposal - the prosecution of this research, during 1914-1916. Boranrcat LABORATORY, _ UNIversity oF PENNSYLVANIA. TWIN HYBRIDS FROM CROSSES OF CENOTHE! MARCKIANA AND FRANCISCANA WITH @. — PYCNOCARPA, IN THE F, ANDF,. (Prates I-IV.) By GEORGE F. ATKINSON. (Read April 13, 1917.) crosses of these with other species of CEnothera. Among the s: which were used more attention has been given to Ginothera marckiana and CZ. franciscana.* Seed of the former was obt from de Vries, of the latter from H. H. Bartlett. at Reciprocal crosses of these two species with Enotheeas n gave results which indicated that twin hybrids were produced ir F,. The plants were grown as annuals in 1915, so that the obs tions were made on summer rosettes and on the mature p The number of individuals in some of these crosses was few. broad leaves of @. nutans, however, resembling in general and size those of @. lamarckiana and franciscana, made an an of the results more difficult and uncertain than in the recipr crosses when . pycnocarpa was used, since its rosette leaves: narrow and deeply cut over the proximal half. f This paper, therefore, treats only of the reciprocal crosses CE. lamarckiana and franciscana, with CZ. pycnocarpa. The poll tions for the reciprocal crosses were made during the season 1914. The seeds were sown in March, 1915, transplanted to flats pots in April, and then transplanted to the garden in June. season was quite rainy until the latter part of August and in 1 These cenothera studies were undertaken more from the morpholog standpoint than from that of plant breeding. ? Enothera franciscana Bartlett, Rhodora, 16, 35, 1914. This species, C. lamarckiana, is one of the large-flowered, open pollinated species. 130 ATKINSON—TWIN HYBRIDS. 131 tember. A very few of the plants did not pass beyond the rosette _ stage, and as there were a number in which stem development began late in the season, it was possible to connect the types of rosettes ' with the types of mature plants. RECIPROCAL CROSSES OF CENOTHERA LAMARCKIANA AND PYCNOCARPA. Cultures of 1915, Annual. __ Gnothera lamarckiana X G. pycnocarpa (No. 99).—Including _ those individuals which did not advance beyond the rosette stage there were between 80 and go plants in the F,. There was a dis- tinct splitting into two types, 7. e., twin hybrids were formed. In certain respects these twin hybrids agree with twin hybrids obtained by de Vries (1913) in crosses of . lamarckiana with certain other species. In certain characters they resemble one of the parents but are modified by the other parent. I shall speak of them as the pyc- _ mocarpa type and the lamarckiana type, but there are such a number of strong contrast characters in the two species that the names of the types might with equal reason be reversed, depending on the form character chosen to represent the type. In this case the _ deeply cut feature of the rosette leaves, present in pycnocarpa, serves _ to mark the pycnocarpa type, while the nearly plain, or slightly _ toothed feature of the rosette leaves of lamarckiana serves to indi- _ cate the lamarckiana type. Pycnocarpa Type; Rosettes—A rosette of this type obtained in the 1915 cultures is represented in Fig. 1, Pl. 1. The pycnocarpa character, cutness of the basal half of the leaves is clearly seen, though they are not so deeply cut as in the rosette leaves of the _ parent (see Atkinson, 1917, p. 228, Fig. 13). The rosette is strongly q _ modified, however, by the lamarckiana characters, convexity and _ crinkledness of the leaves, and the leaves are a little broader than those of pycnocarpa. Pycnocarpa Type; Mature Plant—There were 54 mature plants of this type in the culture. The width and edge character of the leaves come from pycnocarpa. They are long, narrow, more or less furrowed, and rather strongly toothed, more so over the base, as in pyenocarpa. The leaves are rather crowded and drooping, but are 132 ATKINSON—TWIN HYBRIDS. more or less crinkled as in lamarckiana, The stem tubercles red. The calyx bud is rather robust, about 3 cm. long by ; 8 mm ‘stout, tapering slightly, and then abruptly at the base of the tips, ar there is considerable red in longitudinal bars. The flowers large, up to 6.5 cm. broad; petals 25-30 mm. long and 30-35 broad, broadly obovate and emarginate, overlapping or just | clo the gap. The pod spikes are rather dense; the pods 3-3- 5 cm. ne and 6-7 mm. broad, often with red bars. os Lamarckiana Type; Rosettes—No fully developed silseltas 0: the lamarckiana type appeared in this annual culture. ‘The type rosette, however, is shown in Fig. 1, Pl. L., from the F, culture. As a whole it resembles neither Jamarckiana nor pycnocarpa. analysis of the rosette, however, reveals a combination of lamare. i ana and pycnocarpa characters. The edge character of the leav toothedness, is that of lamarckiana, while the narrowness crinkledness and furrowedness are those of pycnocarpa. Lamarckiana Type; Mature Plants ——There were 35 mature plants of this type in the culture. They agree with those of the lamarc iana type in the reciprocal cross. The plants (annuals) were 100- 110 cm, high. The stems are green, with red tubercles, a few tubercles also on the young ovaries. The foliage is rather green. The leaves are rather narrow and long, over the midd é part of the stem 15-17 X 3-3.5 cm., the edge only slightly toothed The lower bracts are leaf like, sessile, up to 13 cm. long by 3.5 broad. The calyx buds are usually green, but sometimes wi flush of red in spots, 3-3.5 cm. long by 8 mm. stout at the tapering gradually to the apex or somewhat abruptly contrac the base of the tips. The buds and flowers are intermediate in between the two parents; petals 32 X 30-40 mm. The pod sp is lax. ‘ (Enothera pycnocarpa X CE. lamarckiana (No. 98).—There w approximately 400 plants in this culture. There were 360 of the lamarckiana type, which agree in all respects with the lamarckias ! type of the reciprocal cross. The remaining 39 plants presen two types of flowers, 34 with large flowers and a long fruiting spi 60-70 cm. long; bracts, especially the lower, large and leaf-like, flush of red observed on the calyx bud, except rarely, the leay ATKINSON—TWIN HYBRIDS. 133 were strongly toothed, pod spike lax, and tubercles on the stem green. There were three plants distinctly of the pycnocarpa type with small flowers, but no red color was observed. All these 39 plants probably belong to the pycnocarpa type so far as the rosettes are concerned, as well as edge character of the leaves. 2 In all of the hybrid types it has been observed that the color is . very variable, especially that of the calyx buds. Also it has been - observed that there is a variation in size of the flowers, some of the _ plants having small flowers, others large. I have suspected that 3 there was further splitting of the types in regard to flower size, but it was impossible with the amount of other work on hand to study a flower size, or color behavior in either the F, or F, cultures. The studies have been confined largely to the rosettes, exclusively so in _ the F,. It would be of interest, however, to study carefully flower and color behavior in the F, and F, hybrids. 3 One of the very marked differences between the two hybrid types ' is the length and general habit of the fruiting spike in the annual 4 forms, and this feature in each twin is in strong contrast to the _ fruiting spikes of the parents. In the pycnocarpa type the density & of the foliage and broader leaves, or lower bracts, contrast well 3 with the open foliage and narrow leaves, or lower bracts, of the _ lamarckiana type. The relative width of the leaves, and their crinkled, or noncrinkled character, parallels these features in the _ rosettes of the twins. The differences in length of the fruiting _ spike were very striking in this annual culture. While this feature has not been carefully analyzed in biennial cultures, I doubt if _ the variation in biennial cultures is so great. Even in annual cul- 3 tures I am inclined to believe that the results would vary to some __ extent with the season, and the time of the year when stem develop- 7 g ment began. In the reciprocal cross the same features were pre- E 4 sented in the fruiting spikes of the twin hybrids. Cultures of 1916: Biennial. In the autumn of 1915 seed was harvested from protected plants, of parents, and twin hybrids of the reciprocal crosses except the -pycnocarpa type of lamarckiana X pycnocarpa. The seed was planted in pans during April, 1916, transplanted to 2 inch, or 214 PROC, AMER, PHIL. SOC., VOL. LVI, J, JUNE 25, 1918. 134 ATKINSON—TWIN HYBRIDS. inch pots in May, and from these set out in the garden in Seoie Of the 75-100 seedlings of each hybrid type of the reciprocal cross, 5° were transplanted to the garden. A few (12-15) of the par pycnocarpa were grown for comparison, and about 150 of the par lamarckiana. The object in growing them as biennials was to obtain the well-developed autumnal rosettes. In all of the cultures the rosettes were large, well developed and remarkably uniform except for an occasional mutant from the pycnocarpa type, and a abe bind tants from the parent /amarckiana. . The F, of the pycnocarpa type (pycno X lam) No. +. ue pycnocarpa character, the cutness of the leaves over the basal half is very pronounced, though it is not so strong as in the parent pycnocarpa. The convexity and crinkledness inherited from la marckiana is very striking. The width of the leaves is greater than in the lamarckiana type. The type of rosette is well —— in the F, of the reciprocal cross (PI. I, fig. 1). The F, rosettes of the pycnocarpa type of the reciprocal cross (lam X pycno) were identical with those of the pycno X lam crc and no photograph was made, but the form of the rosette of this : is shown in PI, I, fig. 1. While this hybrid type has been called the pyenoeuel type, rosettes really show more of the lamarckiana character than th do of the pycnocarpa character. If these twin hybrids were to named now, I should reverse the names because of the prepon¢ ance of lamarckiana characters in the rosettes of the pycnocarpa type, and the preponderance of pycnocarpa characters in th marckiana type. But as all my notes, numbers and marks on negatives correspond tothe names here employed, it does not see wise to change at this time. In this connection it is of interest t note that the pycnocarpa type throws occasional mutants. rosettes of the mutants which have thus far appeared are of we types. One appears to be a dwarf of the true lamarckiana. 1 other has very narrow, furrowed leaves, resembling in some resp ect E the lamarckiana type of these twins, but is much smaller and the leaf edge rather strongly toothed the entire length. The pycno: carpa type of twin, in addition to presenting a predominance ot ATKINSON—TWIN HYBRIDS. 135 lamarckiana character in its rosettes, appears also to have the mutat- _ ing constitution of lamarckiana. _ The F, rosette of the lamarckiana type (pycno X lam.)—The complex of characters expressed in the rosette, derived from both parents, are such that it resembles neither parent. Still the pre- ponderance of characters expressed in the rosette of the lamarckiana type of twin comes from pycnocarpa. These are the narrow, fur- _ rowed, noncrinkled, repand leaves, while the edge character comes from lamarckiana. The leaves are light green. It is a striking rosette and the uniformity throughout the entire row was remark- able (see Fig. 2, Pl. I). The F, rosette of the lamarckiana type of the reciprocal cross (lam X pycno).—lIts resemblance to the rosette of the Jamarckiana type of pycno X lam is remarkable. The only observable difference is that there is a slight buckling of some of the leaves, a pycnocarpa character. In both of these /Jamarckiana types the uniformity of the rosettes in the row was remarkable. ReEcIPROCAL CRosSES OF CENOTHERA FRANCISCANA AND PYCNOCARPA. Cultures of 1915; Annual. Cnothera pycnocarpa X CZ. franciscana (No.: 100).—In this culture there were between 170 and 180 individuals. The majority reached maturity, but there were a few of the more tardy ones which formed autumn rosettes. Several of the latter began stem develop- ment.and advanced far enough so that the rosette types could be correlated with the types presented by the mature plants in which the mature rosette stage was omitted. In the F, the progeny splits into two distinct types, corresponding to twin hybrids in the sense of _ deVries. j . Pycnocarpa Type; Rosette—tThe rosette of the pycnocarpa type _ is shown in Fig. 4, Pl. IL, at the right. The pycnocarpa character _ is shown in the strongly toothed, partially cut margin of the basal _ portion of the leaves. This rosette is not so large nor so well fur- __nished with leaves as those of the F, since the few plants of 1915 annual culture which did not form stems, were belated and did not form such fully mature rosettes. 136 ATKINSON—TWIN HYBRIDS. Pycnocarpa Type; Mature Plants—There were 13 mi; nearly mature plants of this type in the culture. They were : mately 114 m. high (about 130 cm.). All of the red color the sunburn comes from the franciscana parent, as well as the size of the flowers. The stems are green with a rather strong of red in streaks, but parts exposed to the sun usually become « red. Red tubercles are present on the stem, branches and o The lower branches over the base of the stem are 50-60 cm. The flowering branches occur from the middle upward. The over the middle portion of the stem are nearly like those o carpa, long, narrow, furrowed, drooping and strongly serrate the edges, but the midveins are pink. The foliage is dark The calyx buds are 3-3.5 cm. long by 6 mm. broad at the slightly tapering from the base to the base of the tips, then abru or gradually, and with considerable red color in longitudinal ban When open, the flower spread is 4.5-5 cm. The petals are ob o strongly emarginate, pale lemon yellow, 20 mm. long by 25 m broad. The stamens overlap and nearly reach the tips of the st mas, or do not quite reach the base of the stigmas. The pods ¢ 3.5 cm. long by 6-7 mm. broad at the base, tapering oaieny evenly to the apex. Franciscana Type: Rosette—An F, rosette of the fron type is shown in Fig. 5, Pl. III, left hand, a belated plant 1915 culture. The rosettes of the franciscana type are very dit cult to distinguish from those of franciscana itself. In fact not believe that I could distinguish them. It is clearly distir however, from its twin, the pycnocarpa type of the same cross, the right hand, Fig. 6, Pl. III]. The leaves of this rose narrower than the mean, for rosette leaves of C. francisca for this twin hybrid franciscana type, but as this rosette is a b plant the leaves have not reached their full sizes. _ Franciscana Type: Mature Plants—There were 153 plants of this type in the cross. They measure approximately | high (90-120 cm.). The stems are green with occasionally a fai tinge of pink over the older portion. Red tubercles are presen the stem, branches and ovaries. The fruiting spike is 40-50 long, the bracts long, green, persistent, the lower ones up to 9 ATKINSON—TWIN HYBRIDS. 137 long by 3 cm. broad. The calyx buds are 3-3.5 cm. long by 7-8 mm. broad at the base, tapering gradually to the apex, usually an abundance of red color in the calyx, sometimes with only a faint tinge. The open flowers are 4.5—-5 cm. broad, the petals narrowly obovate to cuneate with gaps between them at the base, 20-25 mm. long and broad in the larger flowers. The stamens do not quite reach the base of the stigma, and the stigmas are more slender and longer than in the pycnocarpa type. The flowers are more delicate d wilt earlier than those of the pycnocarpa type. Pods 4-4.5 cm. ¢ by 7 mm. stout at the base. The leaves are exactly like those the parent franciscana, narrow, long, only slightly furrowed, _ toothed on the edges, not so drooping as in the pycnocarpa type, plane, dvein white, foliage pale green contrasting strongly with the dark green foliage of the pycnocarpa type. The fruiting spikes in these annual F, cultures also show a dis- . t splitting into two types in regard to the length of the spike or _ fruiting axis. (nothera franciscana X pycnocarpa (No. 1o1).—The reciprocal cross, nothera franciscana X pycnocarpa gives also a F, progeny hich splits into two hybrid types. These types are identical with hose just described from the cross pycnocarpa X franciscana. There were 102 plants of the F,, 90 of these belong to the francis- cana type and 12 to the pycnocarpa type. Here as in the reciprocal cross a preponderance of the progeny is of the franciscana type. PU NTTRALRLL iy UR IIE YET WA IY PR AE VIMO RTE 2 ete bor Tue F, GENERATIONS. From the F, progeny of the crosses between Cnothera francis- ana and pycnocarpa, seed was saved and sown from three of the hybrid types, the pycnocarpa type, and franciscana type of twin from the F, of @. pycnocarpa X franciscana; and from the pycnocarpa of twin in the reciprocal cross. The cultures were carried along parallel with those of the F, described above, from crosses between C. lamarckiana and pycnocarpa. They were grown as biennials, and the rosettes were mature and fully developed in the Pycnocarpa type F, No. 157 (pycno X fran).—There were 50 138 ATKINSON—TWIN HYBRIDS. plants in the F, of this culture. The pycnocarpa type is n in the F,, but splits in the F, into two types, the pycnocarpa ° and the franciscana type. Of the 50 rosettes in the culture, 13° of the pycnocarpa type, and 37 of the franciscana type types which result from the splitting of the pycnocarpa type F, are shown in Fig. 7 (==157.33), Pl. IV., rye and in Fig. 8 (= 157.10), Pl. IV., franciscana type. = Franciscana type F,, No. 158 (pycno X fran). —There wer rosettes of this type in the F,, all of the franciscana type. is no splitting of the franciscana type in the second generatio the fluctuating variations in the rosettes are quite marked. — variations relate to the size and shape of the leaves, and p: the fluctuating variations of the leaves in the rosettes of the Enothera franciscana. oa Of the reciprocal cross (Cénothera franciscana X pycnoc only the pycnocarpa type twin was grown in the F,. There w plants, most of them developed into mature, autumnal rosettes in a few premature stem development checked rosette develop Splitting in the second generation occurs here also. There wer rosettes of the franciscana type and 4 of the pycnocarpa type. Fluctuating Variations in the F, Hybrid Types——tIn the generation in all of the hybrid types of the crosses between franciscana and pycnocarpa the fluctuating variations as shown the rosettes were very marked. These variations were studied carefully in the franciscana types. It was possible to re several groups into which the principal variations could be ass although the limits of variation in the several groups were not ¢ cut. The groupings are as follows: Series No. 157, F, of pycnocarpa type (pycno X fran). I splitting between the 50 plants in this culture there were 15 of pycnocarpa type and 30 of the franciscana type. The groups variation in the franciscana type, with the number of rosettes each group, are as follows. 1. Leaves medium broad, dark green, white-veined, 3. 2. Leaves broad, dark green, crinkled, pink-veined, 17 (6 of the rather narrow leaved). 7 3. Leaves narrow, dark green, white-veined, 4. ATKINSON—TWIN HYBRIDS. 139 4. Leaves broad, dark green, white-veined, 5. Leaves narrow, pale or yellowish green, pink-veined, 6. 6. Leaves narrow, dark green, pink-veined, 2. _ Series No. 156, F; of pycnocarpa type (fran X pycno). In the splitting between the 43 plants in this culture there were 4 of the pycnocarpa type and 39 of the franciscana type. The groups of variation in ‘the franciscana type, with the number of rosettes in each group, are as follows: 1. Leaves broad, dark green, crinkled, pink-veined, 4. 2. Leaves broad, dark green, white-veined, 16. Leaves medium broad, dark green, white-veined, 15. 4. Leaves narrow, pale green to yellowish, pink-veined, 3. 5. Leaves medium broad, pale green, white-veined, I. _ Series No. 158, F, of franciscana type (pycno X fran). There _ is no splitting of the franciscana twin in the F,. There were 67 plants in the culture. The groups of variations in the rosettes of the second generation of the franciscana twin, with the number of rosettes in each group, are as follows: 1. Leaves broad, dark green, pink-veined, 9. 2. Leaves broad, dark green, white-veined, 14. 3. Leaves narrow, dark green, white-veined, 13. 4. Leaves medium broad, dark green, white-veined, 25. 5. Leaves narrow, pale green or yellowish olive green, 3. 6. Leaves narrow, pale green, pink-veined, 2. Considerable fluctuating variations were presented by the rosettes of the pycnocarpa types in the F, of the reciprocal crosses. Parallel fluctuations undoubtedly occur in the rosettes of the F, of the two - types of hybrids in the reciprocal crosses, but as the F, plants grown were nearly all annuals the number of rosettes was not sufficient for a study of these variations in the first generation. Since the rosettes of CEnothera pycnocarpa are very uniform, the pronounced fluctuating variations in the hybrids of the crosses be- tween pycnocarpa and franciscana are traceable to the constitution of @. franciscana, for the fluctuating variations in the rosettes of the franciscana hybrid type are parallel with the fluctuating varia- ons in the rosettes of franciscana itself. In the 1916 cultures, 135 settes of CZ. franciscana were grown in the garden as a parallel 140 ATKINSON—TWIN HYBRIDS. culture. Five or six different groups were recognized, but record was kept of the number of rosettes which could be ass : to each group. The groups are as follows: . Leaves broad, dark green, plain, white-veined. . Leaves broad, dark green, crinkled, white-veined. . Leaves narrow, dark green, red-veined. . Leaves, broad, dark green, red-veined. . Leaves medium broad, dark green, white-veined. , These marked fluctuating variations in leaf form, rep : ir one of the features in the constitution of CEnothera franei. which is inherited in its hybrid progeny, marks this species as favorable one for stimulating great fluctuating variations in hybrids from crosses with other species, indicated not only by th crosses of Cinothera franciscana* and pycnocarpa described here but also by the great fluctuating variations resulting from crosse between Cinothera franciscana and C. biennis as descritied: by Davis (1916). oe The marked fluctuating variations in the twin hybrids it reciprocal crosses between C. franciscana and G2. pycnocarpa mi be interpreted by some as indicating that two distinct hybrid 1 were not present, but that the two forms represent merely a s wide range of fluctuating variation. That this interpretation is valid is shown by the fact that the franciscana twin type, alth variable, is fixed, it does not split in the second generation nor in i fluctuations does it produce typical pycnocarpa twin forms; ie the pycnocarpa twin type splits in the second generation i two types. Further evidence that the interpretation given, in this pape t the results of these crosses, so far as the production of twin hybrid: and one-sided splitting is concerned, is admissible, is found in th very close genetic relation which C2nothera franciscana bears (Enothera hookeri (see Bartlett, 1914, p. 33). Reciprocal cros of CG. hookeri with-@. lamarckiana, or with certain of its muta nb wW DN 8’ The seed of CEnothera franciscana which I have used came from D H. H. Bartlett in the winter of 1914, from a series of cultures which he continued for a few years, and as I understand it has the same pedigree as” the seed employed by Dr. Davis in his interesting crosses with CG. biennis. — ATKINSON—TWIN HYBRIDS. 141 give twin hybrids in the first generation, with splitting of one of the ‘twins in the second generation (de Vries, 1913, p. 131). The re- sults of reciprocal crosses of CE. franciscana with @. pycnocarpa are a close parallel, and indicate that the genetic constitutions of CE. franciscana and C. hookeri are very similar. SUMMARY AND CONCLUSION. The history of the progeny of the crosses of (Enothera lamarcki- ana and ©. franciscana with CG. pycnocarpa belongs in the series of some of the most interesting phenomena of hybridization known in the CEnotheras, and discovered by de Vries (1907, 1908, 1909, 1913). These phenomena are, first, the production of twin hybrids in the first generation of a cross, the two hybrids being fixed in the 1 generation and continuing to reproduce themselves in the F, and succeeding generation ; and, second, the production of twin hybrids ‘in the F, with one-sided splitting in the F., and succeeding genera- tions. In the second case one of the twin hybrids of the first gen- eration is fixed in the F,, the other splits in the F,, into two types, like the twins of the F,, one of which is fixed while the other splits in the F, and so on. In the crosses of (Enothera lamarckiana and . franciscana with C. pycnocarpa, the cultures have not been carried beyond the second generation. But the existence of this peculiar phenomenon of inheritance in certain crosses among the evening primroses has been so thoroughly demonstrated by de Vries for several succeeding generations that there can be no reasonable doubt that it applies also to the behavior of the crosses here de- scribed, for succeeding generations. _ _ ___In the reciprocal crosses of @. lamarckiana with G@. pycnocarpa _ the twin hybrids are fixed in the first generation, and “ breed true ” in the second, and in all probability, in succeeding generations. Each of the twin hybrids is very uniform, at least in the rosette stage, and shows a minimum of fluctuating variations. The pyeno- -carpa twin type is a physiological homozygote but its fundamental _heterozygotic constitution is now and then manifested by the salta- tory production of lamarckiana forms, and also of forms approach- ing pycnocarpa, or the lamarckiana twin type. The reaction system 142 ATKINSON—TWIN HYBRIDS. established in the F, zygote which produces the pycnocarpa twin type, is in a very high percentage of cases stable in the F, and th following generations. But occasionally other reaction sys usually dormant, are activated, resulting in /amarckiana, and forms. The lamarckiana twin type presents also a minimum fluctuating variations in the rosette stage, though it appears to probable that it is a physiological homozygote and carries. in latent, or inactive condition the other factors of both parents w ic are not manifest in the phenotype. : In the reciprocal crosses of Ginothera franciscana with a. ye nocarpa, only one of the twin hybrids (franciscana type) is fixed in the first generation* The other (pycnocarpa type) having “hybrid constitution” splits in the second generation into two which are like the twins of the F,. The pycnocarpa type with probability would continue to split in the same way in succ generations. When the reaction systems of C. franciscana and | pycnocarpa meet in egg or F, zygote, a new reaction system tablished combining the factors of the two parents. In the uni certain factors of one or the other parent preponderate but influence is modified more or less by the homologous factors. the new reaction system established in the F, zygote is not the in all the zygotes. As the two different parent reaction system 7 meet in the egg, one or the other of two new reaction systems is or- ganized, and chance seems to determine which one of these work systems is established in a given zygote. The reaction system o lamarckiana twin is stable, that of the pycnocarpa twin is unsta (and heterozygotic). The twin hybrids in these crosses display the organization of their reaction systems in the F, zygote wha have termed “ selective dominance” (Atkinson, 1917, p. 253). _ I wish here to express my appreciation of aid given by Mr. E E. Stork, assistant in botany, in writing notes in the field from m dictation, and for making some of the photographs. 4 The franciscana twin probably carries the pycnocarpa factors also, in a subordinate or permanently latent condition. If so, it is a physiologi homozygote. If it were possible to introduce a splitting factor into tk franciscana twin by an appropriate cross, and cause the pycnocarpa character a to reappear in some of the progeny, it would indicate the fundamental hetero- 3 zygotic constitution of the franciscana twin. ATKINSON—TWIN HYBRIDS. 143 3 t b a Geo. F. 1917. Quadruple Hybrids in the F, Generation from (£nothera nutans and CEnothera pycnocarpa, with the F. Genera- tions, and the Back and Inter-Crosses. Genetics, 2, 213-260; Figs. 1-15. See also Twin Hybrids from C£nothera lamarckiana and franciscana when crossed with CEnothera pycnocarpa. Sci- ence, N. S., 46, 22, 1917. Bartlett, H. H. 1913. Systematic Studies on CEnothera, III. New species from Ithaca, N. Y. Rhodora, 15, 81-85. 1914. Systematic Studies on CEnothera, IV. CEnothera franciscana and . venusta, spp. novv. Rhodora, 16, 33-37, Pl. 107, 108. Davis, B. M. 1916. Hybrids of CEnothera biennis and (£nothera franciscana in the first and second generations. Genetics, 1, 197-251, Figs. 1- 26. See also 1916. CEnothera neo-lamarckiana, hybrid of O. franciscana Bartlett x O. biennis Linnzeus. Am. Nat., 50, 688-606. saat T. H., and Clausen, R. E. 1917. Mendelian Factor Differences versus Reaction System Contrasts in Heredity. Am. Nat., 51, 31-46, 92-101. de Vries, H. 1903. Die Mutationstheorie, 2,xvi+752 pp. Leipzig: Veit & Co. 1907. Twin Hybrids. Bot. Gaz., 44: 401-407. 1908. Uber die Zwillingsbastarde von CEnothera nanella. Ber. d. deutsch. bot. ges., 26A: 667-676. 1909. On Triple Hybrids. Bot. Gaz., 47: 1-8. 1911. Uber doppeltreziproke Bastarde von CEnothera biennis L. und €. muricata L. Biol. Centralbl., 31 : 97-104. - I913. Gruppenweise Artbildung, unter spezieller Beriicksichtigung der Gattung CEnothera. viii+ 365 pp. Berlin: Gebr. Borntraeger. THE ART OF GEORGE CATLIN. By EDWIN SWIFT BALCH. (Read April 19, 1918.) Within the past decade, a number of American painters transferred their Lares and Penates from Europe and the e United States to Arizona and New Mexico. They have don thi because it has dawned on them that the American Indian southwestern states offers a splendid opportunity to put on c subjects virgin in form and color. About a dozen pictures of zona and New Mexico Indians by these painters were in this al exhibition of the Pennsylvania Academy of Fine Arts. The aim these painters is undoubtedly artistic, but their works have an portant scientific attribute, namely that they record ethno subjects and in time will form a grand series of illustrations of th appearance and the customs of a few tribes of the original tants of America. About this movement, Mr. Edgar L. H Director of the School of American Research, Museum of Mexico, Santa Fe, on the 4th of January last, gave a most inte ing account to the American Philosophical Society. the Indians of to-day. In the United States, they can record t appearance and the doings of the Indians of the desiccated reg of the southwest, whom one may call generically the Pueblo Indiz : and even those Indians have had their costume affected by tha the White Race. But they cannot record the historical Neo Indian. For the Indian of the Allegheny forest, of the Plains, of the Rocky Mountains, the Indian of the deer, the bison and the grizzly bear horizon, is a thing of the past. In his genuine nat ‘trappings, he can never be painted again. Fortunately for ethnology and for the history of the natives 0: America, a handful of painters of by-gone days have left us so 144 BALCH—THE ART OF GEORGE CATLIN. 145 drawings and paintings which form a precious record of our copper-colored predecessors before they had become largely Euro- -peanized. A number of these paintings are in the United States National Museum in Washington, a number are in the American Museum of Natural History in New York, a few are in the Harvard _ University Peabody Museum and others are scattered through- out the country. Many of these paintings are portraits, usually not of any great art merit. As works of art they will doubtless be _ greatly surpassed by some of the pictures now being painted. But 4 as ethnological data they are exceedingly important and will always hold their own. , One of these painters of Indians was J. O. Lewis, who painted a good many portraits of chiefs of various tribes, Sioux, Winne- bagoes, Chippewas, etc., and who published a portfolio of colored lithographs of them.t Many of his models were garbed in a hybrid European dress and the lithographs are too poor to render ac- -curately the heads. _ Another painter of Indian portraits was C. B. King. Some of his paintings were reproduced by colored lithography in McKenney’s and Hali’s book? and historically they are of importance. The frontispiece of the book by P. Rinetisbacher (in text Rhinedes- bacher) is an interesting picture of an Indian dance. One amateur artist who portrayed sporadically Indians was Cap- — tain Sully, U. S. A., son of the portrait painter Thomas Sully. Ina lecture before the Historical Society of Pennsylvania, on January 14, 1918, Mr. Henry Budd stated that Captain Sully while on frontier duty made some sketches of Indian life. Captain S. Eastman, U. S. A., made a number of drawings of Indian scenes, which were engraved for Schoolcraft’s great work.* Some of these were from his own sketches, apparently made while _he was on active service along the frontier. But some of his draw- ings were from sketches by other persons, Schoolcraft himself, Lt. 1J. O. Lewis, “ The Aboriginal Portfolio,” Philadelphia, 1835. 2 Thomas L. McKenney and James Hall, “ History of the Indian Tribes of North America,” Philadelphia, 1836, 1842, 1844. 3 Henry R. Schoolcraft, “ Historical . . . Information . . . Respecting the History . . . of the Indian Tribes,” etc. Illustrated by S. Eastman, Capt. U. _ S.A, Philadelphia, 1851. 146 BALCH—THE ART OF GEORGE CATLIN. Col. J. H. Eaton, U. S. A., R. H. Kern, Esq, ete. It is not prising therefore that the engravings are somewhat monotonous handling and lack to some extent realistic detail. The dra’ doubtless were much better than the reproductions, but these theless have saved a great deal which has now passed mea (Ol: life of the American Indians and Eastman’s work will remain ar manent contribution to American Ethnology. oe But by all odds the most important of all painters of oe Auees ican Indians is George Catlin. Catlin was a man of many activiti a great traveler and something of an explorer, an ethnologist, ; geologist, a voluminous writer, but above all a painter. About travels and his views on geology and ethnology, his own writings offer all necessary data to a student; of his art, numerous engra d reproductions are accessible. But of his art, from a painter’ Ss of view, and of his rank as an artist, no critical study, to my kno edge, has yet been made. And to fill this lacuna by a tech examination of the paintings of this remarkable man is the obj of this paper, which although it appears in my name, is really case of joint authorship. For my wife studied the Catlin pict in New York and Washington with me and many of the observat 0 and ideas here presented are hers. I had the pleasure of meeting Catlin on one occasion many y ago in Europe, I think in 1871. He was then traveling about hibiting his collection of pictures. I went to see these and lucky enough to find him in the gallery where they were an have a long talk with him, and I remember him as a most inter ing and friendly old man, who loomed up to my boyish eyes hero. 5 ‘es Catlin was born in 1796 in the Wyoming Valley, Pennsylvania His boyhood was passed principally in hunting and fishing. he grew up he studied law, but soon grew tired of this and wen Philadelphia, where he started as a painter, without teacher or viser. After several years, one day a delegation of Indians f the “Far West,” arrayed in their native dress, happened to pa through the city, and this event determined the course of Catlin life. He dreamed of nothing but painting Indians and he carri out his dream. BALCH—THE ART OF GEORGE CATLIN. 147 _ He started in 1832 and wandered all over the plains as far as the Rocky Mountains, living with the Indians for nine or ten years and all the while painting their portraits and making pictures of all the different phases of their life. For some years after this he was _ occupied in exhibiting these pictures in America and Europe, and also in writing and publishing several important books.* In the _ “fifties” he traveled extensively in South America, principally in q E the regions of the Orinoco and the lower Amazon, where his brush _ omce more was ceaselessly busy. After this again he wrote nu- _ merous valuable contributions to the knowledge of the Indians of North America and South America and also traveled about exhibit- ing his collection. After his death, the greater part of his pictures ortunately passed into the possession of the Smithsonian Institution Washington and of the American Museum of Natural History New York City. Almost all of the hundreds of pictures painted by Catlin are of the same size, about 19 by 25 inches. Almost all are painted length- wise, not upright. In his more elaborate compositions he covered _ the entire surface. But in many cases he painted an oval picture, _ which he framed with a black line. He may have used the oval _ shape because he recognized either consciously or unconsciously that the eyes do really see an oval rather than a rectangle ; or because e thereby avoided certain difficulties in filling corners; or he may have found that the oval shape sometimes assisted the composition ; S omabeeicresaiadans inet okalntami-ckes Meena a a A 6a a a ad Sail eR al = r 4 4 Catlin’s most important publications are as follows: _ “Catlin’s North American Indian Portfolio,” London, 1844. _ “Letters and Notes on the Manners, Customs and Condition of the North American Indians,” New York, 1844. __ “Tilustrations of the Manners, Customs and Condition of the North American Indians,” London, 1845-1848. ; “Life Amongst the Indians,” 1861. “Tlilustrations of the Manners, Customs and Condition of the North American Indians,” London, 1866. “Last Rambles amongst the Indians of the Rocky Mountains and the _ Andes,” New York, 1867. _“O-Kee-Pa: a Religious Ceremony and other Customs of the Mandans,” Philadelphia, 1867. _ “Catlin’s Notes of Eight Years’ Travel and Residence in Europe,” New _ York, 1867. - “North and South American Indians. Catalogue,” etc., New York, Baker and Godwin, 1871. 148 BALCH—THE ART OF GEORGE CATLIN. or in some cases he may have saved time which in painting. speed in the wilds must frequently have been precious. ns Most of Catlin’s pictures are on prepared paper of a light brown, which often helps a good deal as an undercolor, occa remaining untouched. The pictures, as a rule, are light 1 bright in tone; there are few brilliant lights and few deep they are usually in a high middle, somewhat dull, register. a Catlin’s palette is limited but complete. All the essential cc are on it. The bright colors are used most sparingly and on small touches and accents. There is certainly white lead. Y. ochre is much used. A little bright yellow, which may be Nz yellow. Light red. A few touches of two bright reds, all surely vermilion and rose madder. One bright blue, which al certainly is cobalt. In one or two cases, in night effects, t seems to be some darker blue, possibly indigo. Brown, prot Vandyke and umber, is a good deal used. Black is occasional ployed and sometimes in night effects pure or nearly pure. T is much dull, usually light green in Catlin’s pictures: this ma be a mixture rather than a pure pigment. es The method of Catlin in laying on the paint is of interest. a paint is thin and smooth. It is all applied evenly in one thin without retouches. There is no impasto; there are no rep His work might almost be called tinted drawing rather than ing. There are two explanations of this mode of work. One them is that it was to a great extent the method then in use. © painters covered their canvas with a slick surface of paint, which all roughnesses and ridges were removed. The other probably is the great difficulty Catlin must have had in carrying terials and paints with him. He must have opened his col his palette in the smallest possible amounts, and made every of paint do as much covering as possible. One of the curiosities of the Catlin collections at Washin and at New York, is that there are no sketch books, no rowan : ings, no slips of paper with pencil or chalk marks or blots of y color. Catlin does speak in connection with his first bison ae of making drawings of an old bull from his horse in his sketch b sole and in another place he writes of altering the finished portrait of BALCH—THE ART OF GEORGE CATLIN. 149 dissatisfied Indian with water colors; but no such sketches in either pencil or water color, as far as I know, have come down to us. All his works are small finished pictures, which Catlin carried as far forward as he knew how. And considering how well understood his pictures are as a whole, it is astonishing how much detail he gets into his figures and their accessories. But while this multiplicity of detail always takes its position in the whole, as a result his pictures do not carry any great distance; they are best looked at close by. ‘Some of his detail is minute and delicate. Details on the dresses in his portraits are beautifully carried out; there is the greatest delicacy of touch. And it is of ethnographic significance that all ‘the decorations he depicts on the clothing of the figures or on the teepees are always square or rectangular decorations, such as one sees on the Lewis and Clarke skin robe in the Harvard University Museum. Catlin drew well; not academically but accurately. His portrait heads and full-length portraits may be ranked as fair examples of the style of portraiture in vogue in America in the first half of the nineteenth century. Had he continued painting portraits at home he would doubtless have earned a comfortable competency. And while, of course, Catlin never painted any pictures of architecture requiring linear perspective, his pictures always have the correct artistic perspective which all good landscape painters obtain through intelligent drawing. A strong point of Catlin is his splendid sense of proportions. He got the natural proportions of figures. His figures are utterly unacademic. He was not preoccupied with Greek or modern Euro- ‘pean conventional canons of what a human should be. His nudes are nudes, the real thing ; they have much the feeling of the French primitives of the Middle Ages. Catlin merely tried successfully to make humans look like what they are, and one feels that nobody looked over his shoulder and told him he was not right. __ While all of Catlin’s models are copper-colored with straight black hair and sometimes are daubed over with red ochre or other colored earth, nevertheless there are two variations in type. One of _ these, which is most apparent in the portraits, has features very _ similar to the Americanized European whose ancestors came over _ PROC. AMER. PHIL. SOC., VOL. LVII, K, JUNE 25, I918. 150 BALCH—THE ART OF GEORGE CATLIN. early in Colonial times and this type resembles the thin gaunt ican type of to-day. The other type, which is most apparent in incident pictures, resembles the Mongol type, both in the faces in the figures, which are decidedly squat. The latter type suggests the faces in Aztec or Maya art. Mr. Huntington W former assistant Secretary of State, tells me that he observed types among the Indians of South America, one on the high A the other in the hot forest lowlands east of the Andes. App Catlin observed something of the kind among the Indians | « Northern Plains. | Color and also values, that is light and shade, Catlin een very realistically. He never attempted to solve any artistic problem in color nor in light and shade; he simply painted his subjects strai; forwardly and quickly as well as he could. He was absol sincere in trying to render what he saw. In the real sense fe) word, therefore, his works are genuine realistic impressions. they have not a semblance of so-called impressionism. His values and colors are always an attempt to present as nearly as he could a scene in nature. His color is sober. Evidently he tho much of local color and little of artistic color schemes. There no decorative quality in his work. The true function of decor tive painting is to make patterns of lines and patches of color decorations, not to represent or imitate nature. And there are 1 line patterns nor patches of color work in any of Catlin’s pict What he does get in his coloring is a most remarkably fait rendition of the colors of nature. ges The accurate rendition of the colors of nature is shown fe in some of Catlin’s pictures of South American forests. In he shows great nerve in tackling the, from our usual pictorial : point, utterly unpictorial subjects of the swamps and jungles, color might be called a vegetable green monochrome. He s the sogginess, the pestilential malarial character of these American swamps in a wonderful way. His forests give the pression of forests, his trees really look like trees in a forest, n more so than does much of the more learned work of the European painters, then for instance, the forests of some of Barbizon men or of some of the impressionists. And he succe BALCH—THE ART OF GEORGE CATLIN. 151 _ largely because he is not afraid of covering a canvas with a mass _ of green, and because that green does imitate closely the color of a mass of green leaves. Values Catlin always tried for and usually got very fairly. It is _ partly because his numerous detail is in value and stays in place that _he gives the impression of simplicity and a look of out of doors. He often suggests most successfully distance and atmosphere, as for instance in a picture, now in the American Museum, of some snow mountains, probably the Andes, in which the mountains seem miles away. Some of his skies also,-especially at sunrise and sunset, have not only color and light, but most delicate values.. To chiaro- scuro, that is to an artistic arrangement of light and shade or _ values, he paid less heed. He sought values and sometimes ob- tained arrangements of light and shade which are most artistic, but it seems always as if it were the subject which bore them in itself, rather than that he was searching for them. _ While there is little striving after effect in Catlin’s work, still sometimes he painted some memory effects most successfully. Among his wuvre are a certain number of night effects, forerunners _ of our modern nocturnes but not just a dark blue smudge like some of these. They are painted with a generous use of black. There is lots of detail in them: the more you look into them the more you see. Two of these nocturnes in the American Museum may be instanced. One is a camp fire under pine trees which is excellent in composition and in which the pine trees are really drawn. The other is a South American river with some men standing over a lot of captured turtles and a number of women running up waving torches with the most splendid action and motion. Evidently an inborn gift for composition was one of Catlin’s artistic attributes, for he received as little outside artistic influence as any painter ever did, yet each of his pictures shows a distinct power of composing every subject. He had the dramatic instinct, he knew how to place on canvas a scene he had observed so as to make it into a picture. In some of his works, he renders the ap- pearance of a crowd, of a multitude of animated beings, whether Indians, or bison, or peccaries, in a way few painters have done. In his pictures of Indian games, one feels as if there were hundreds 152 BALCH—THE ART OF GEORGE CATLIN. of Indians before one; in his bison hunts, the bison herds over the prairies by the thousand. It is largely Catlin’s power « composition and selection which makes these pictures successful indeed almost invariably his pictures have good a sometimes they have really seiner composition. painted on the spot, as soon as seen and in their ie ment, but they could only have been done from memory, as’ quality which enabled him to get so much life in his work. For humans and animals have both action and motion: they are a they stand plumb on their feet, they walk, they run, they ju they have none of the arrested motion of certain academic : His groups of figures render the movements of the grou feel the way each group is moving. Except in his” portraits, humans are never posing. There is no rigidity in his work. one weak spot in regard to motion is that he painted some of galloping horses and bison with the incorrect open-scissor aa which no white race man ever discovered was wrong, until i taneous photography obliterated it from art. It is the matter and not the manner of Catlin’s pictures, wont which is of supreme importance. The paramount value of tures lies in the subjects and in the fact that the subjects are han realistically. His pictures are extremely original through their ; jects and they are absolutely individual because the subjects pealed to Catlin and were motives to him. There is nothing i tic about his pictures; they are not imaginative; they are | realism. His Indians are not the Indians of romance nor of warped mentality of hostile whites; his Indians are the real Catlin is a great illustrator-painter. He painted endless incider of the life of the American natives realistically and accurately. - painted his pictures of the wild Indians while actually living < them, with the scenes which he was painting, the real history of Indians, actually being enacted before him. And the result is tt BALCH—THE ART OF GEORGE CATLIN. 153 : Catlin, as no other artist, makes the Indians a part of their surround- ings, a part of the wild life of the plains, of the forests, of nature; he makes them a living part of their environment. His pictures place before us the Indians in the chase, in the dance, in the tepee, in fact in all the incidents of life. He shows us in an unexcelled _ way how people who lived by hunting with stone weapons obtained their livelihood ; and he makes it clear that killing bisons and grizzly bears was anything but child’s play to a man armed with a stick tipped with a pointed stone. _ Catlin looked at the Indians with a friendly eye. He lived with them for years, he admired them as models and as characters, in- leed one might say that he loved them. The usual idea that the ian is a lazy, good-for-nothing individual, who lets his squaw k and slave for him, is really a libel and is dispelled by Catlin. [t is formed from the Indians on reservations, who received their beef and blankets from government agents. When the Indian was corralled and the bison exterminated, the Indian’s occupation was gone. The real Indian provided meat and skins for his family ; food d the materials for clothing and teepees. To obtain meat and skins from deer, bison and grizzlies with a flint-headed arrow was ough for any man; it took his time and strength. When he hunted day after day and week after week and year after year, in good and bad weather, in sunshine and sleet, in cold and heat, he considered and he considered rightly, that he was entitled to have his food and his clothing prepared for him at home. He did not go downtown to deal in finance, nor did he stand up in a store to sell millinery, but in his native conditions he was just as much a business man as any American of to-day, and just as much entitled to find a good dinner at home in the evening with his dress clothes laid out nicely brushed, as our hardest worked lawyer or physician. It is a godsend for the history of the American Indians that Catlin was never taught to draw, that he lacked the opportunity of studying and learning to paint like everyone else. If he had been trained in the schools of the day, probably he would have developed the what might be termed rather grandiloquent style of some of the so-called Hudson river school. Fortunately he did not. For as a result of being self-taught and of living most of his life in the wilder- 154 BALCH—THE ART OF GEORGE CATLIN. ness, Catlin’s painting is truly individual, it is unlike anyone a sure test that he had real underlying art powers. His pictures not founded on tradition and therefore perhaps have a certa primitive look; indeed Catlin more than any American might | re called a oR piece The wie of psi would not see th n: is probably accurate to say it is partly those very naiveté make it so good. a Catlin’s position among artists is unique. He deren his with almost no pecuniary reward, to delineating the deeds and artistic beauties of a vanishing race. His pictures are the g record of our displaced predecessors. His incident pictures painted directly on the spot, either from the Indians posing fo: or from memory immediately afterwards. He painted hundre such incident pictures from occurrences he actually saw. N. else has done anything of the kind except most sporadically. one could do it now. For all these scenes have disappeared the face of the earth. Anyone in the future, artist or layman, \ wants to see how our Indians, untouched by white civilization, tually lived and appeared, must turn to Catlin. In the coming turies the Indians more and more will amalgamate and fuse their conquerors and the more they do, the greater value will : tists attach to the wonderful records which Catlin has left Of copper-colored men who once ranged and roamed in wild an restrained liberty from Alaska to Tierra del Fuego. 2 PARASITISM AMONG THE RED ALG&. By WILLIAM ALBERT SETCHELL. (Read April 19, 1918) _ The question as to when a particular plant is, or is not, a parasite often difficult to answer, although in many cases the parasitic re- ion is readily to be inferred because of certain morphological peculiarities and also because of the apparent dependence of the (parasite) upon another (host) in the matter of nourishment. fundamental conception is, of course, that the parasite draws upon the host for materials of greater or less metabolic value, . but, as to amount and extent, is difficult of demonstration and may be inferred largely from various indications of a morphological ie A _ There are also to be considered in connection with parasites, _ especially among the thallophytes, epiphytes and endophytes. A rue epiphyte uses the plant upon which it grows only for mechan- cal support. Its metabolism is independent of that of the plant on which it grows. It is conceivable that even those epiphytes which penetrate the tissues of the supporting plant do so only in a mechan- 1 way, although it seems probable that penetration is usually as- sociated with the establishment of metabolic relations. In case of the endophyte, a variety of relations seems to exist between it and the plant it inhabits. Epiphytes exist both among cormophytes and thallophytes. EEndophytes are always thallophytes and they may grow entirely within the body of another plant or only ‘partially so. Some algal endophytes only penetrate between the layers of the outer walls while others descend deep among the cells of the plants they inhabit. It seems very possible that there may be parasitic relations between many, or most, of the endophytes and their “ hosts.” _Epiphytes are numerous among the red alge and, while no exact enumeration has been made, it seems safe to say that, at least, half 155 156 SETCHELL—PARASITISM AMONG RED ALGAE. © : of the known species are epiphytes. It is very desirable that ‘ should be investigated as to their relations to the plants on they grow. Where epiphytes are constantly observed on a sing even on a few closely related plants it is to be suspicioned, at that there may be some, even if slight, parasitic relation. Pol phonia fastigiata (Roth) Grev., e. g., is an epiphyte whose a! occurrence on Ascophyllum nodosum (L.) Le Jolis has ticed. R. J. Harvey Gibson (1891, p. 132) states that “ The ment of the epiphyte to Ascophyllum is very intimate” an farther that “root filaments given off from the base of th penetrate deeply into the tissue of the host and wander amongst cortical cells and medullary hyphe.” A similar penetration of “host” usually takes place from the base of Pterosiphonia Wo. (Harv.) Falkenb. into the tissues of the Laminariacez it grov upon, as observed by N. L. Gardner and myself. Callitham Lejolisea Farlow penetrates deeply into the tissues of the nod the Amphiroa on which it is always found. Clara K. Leavitt ( p. 294) describes the penetration of Microcladia californica and also of an unnamed species of Callithamnion into the fronds “Callymenia Phyllophora J. Ag.” The list will undoubted! considerably extended after carefully examining other “ epi It remains a question as to whether such forms are to be con as parasites or not. The Polysiphonia is of low stature, but hardly be considered as reduced. The Pterosiphonia and Mi cladia are often found in very much reduced forms but not as a and they show little, if any, loss of color. | Certain forms of penetration of the plant upon which th clusively are to be met with occur in the cases of Placo Binderi J. Ag. and Ceramium codicola J. Ag. Both species on Codium and possess rhizoidal filaments which differ from an the regular outer structures and which penetrate, at least, bet the utricles of the Codium. This does not seem to indicate parasitism. | Among the marine species of Chantransia (or Acrocheti are some epiphytes which are confined to a single “ host” and whic are partially, at least, parasites (cf. Rosenvinge, 1909, p. 82 The plants of this genus are all small, whether growing on rock, SETCHELL—PARASITISM AMONG RED ALG. 157 true superficial epiphytes, or as wholly or partially endophytes. one of these species has been observed to actually attack the of the host, and that is Chantransia cytophaga Rosenvinge, ; growing in the fronds of Porphyra umbilicalis. This species seems certainly to be reckoned among the parasites. Of endophytes or endozoic species there are a few reds. Be- ides species of Chantransia or Acrocheium alluded to above, there are Schmitziella endophlaa Bornet & Batters and Rhodochorton Beer onacenen Magnus. These are simply within the outer mem- , the former of Cladophora, the latter of Sertularia, one of Biron. Somewhat more deeply, and also partially, endo- is Rhodochorton subimmersum Setchell & Gardner, whose n filament is totally included and whose short, erect tetraspo- gia-bearing branchlets are emersed at their tips. There are sev- similar endophytes to be found among the red alge which may slightly parasitic, but it is difficult to determine this with ex- In contrast with the various epiphytes and endophytes, such as se mentioned above, are those red alge which seem undoubtedly » be parasites. In placing these among parasites, three criteria of robable parasitism have been considered, viz., penetration, reduc- of thallus and loss of color. It has been considered that at the first two ought to be present and in very evident form to stitute evidence of parasitism, while the last may or may not be ticeable. _ The undoubted parasite enters the host plant, as a rule, by rhizoidal filaments, or more solid haustoria, which penetrate beyond superficial assimilating cells into the conducting tissues. It ally also, establishes more or less conspicuous pit connections be- tween its haustoria and the cells of the host. _ The reduction of the thallus or vegetative plant body varies much the different parasites. As mentioned above, some dwarfing akes place even in certain plants which it seems best to consider, for the time being, at least, as epiphytes, but in those usually reck- oned as parasites, dwarfing is extreme and usually accompanied y a greater or less condensation of the thallus, resulting in tuber- growths of greater or less extent. Taken in connection with 158 SETCHELL—PARASITISM AMONG RED ALGAE. penetration, extreme dwarfing or condensation of the thallus ma taken to indicate true parasitism of greater or less — 7 same time the reproductive organs present little if any metam phosis. In color, parasites vary from little if any loss of pigment ei where none is present. Loss of color is always associated w extreme penetration and very considerable dwarfing or pene Hon It is only within the last thirty years that the fact has realized that there exists a group of peculiar and undoubted p sites among the red alge, although a few cases were noted bef that time. Probably the first reference to such a parasite is that Lyngbye in noting that certain tubercles on “ Spherococcus Brodi (Phyllophora Brodiai) called by him “ Chetophora membranifo and which had been considered to be the nemathecia of the pla on which they were found, were no part of the “ Spherococcus’ but belonged to a parasite. This was in 1819 (p. 11, pl. 3, f. B, 3, In 1834 Lyngbye describes this plant in more detail, naming : Chetophora subcutanea (1834, pl. 2135), but saying nothing nitely as to its being parasitic. Kuetzing, in 1843 (p. 177, pl f. IV), re-described and named it Actinococcus roseus, Seem unaware of the earlier description of Lyngbye. The species, " now bears the name Actinococcus subcutaneus (Lyngb.) Rosenv later became the object of a considerable discussion and differ of opinion as to its exact nature (cf. Schmitz, 1893, etce.). It now recognized as a true parasite by most phycologists. 2 The second parasite belonging to the red alge to be reco 1 was Ricardia Montagnei described by Derbes and Solier in 18 209, pl. 1). This plant, usually assigned to the Bonnemaisoni forms ovoid red bladders of larger or smaller size, on the tip species of Laurencia. Its basal portion occupies the apical pit the branches of the Laurencia and penetrates into its tissue (c manns, 1905, p. 326, f. 586). In 1874-75, Reinsch published his “Contributiones ad Alg lo giam and Fungologiam” and in this he described a number of 1 bercles found on various red alge which be believed to be para members of the same group. Among the tubercles thus describ by Reinsch, some are undoubtedly simply warts or pathologic OU SETCHELL—PARASITISM AMONG RED ALG. 159 growths of the plants on which they are found, but some are un- _ doubtedly parasites. Such are some of the plants belonging to his ; genera Choreocolax and Syringocolax, but Reinsch did not find any cystocarps or other organs of frutification on any of his specimens. _ The work of Reinsch produced little immediate effect and the basis of truth in it was not recognized until some years had passed. 1877, however, there appeared the first convincing description illustration of a parasitic red alga. In this year, Solms Lau- bach published his paper on Janczewskia verruceformis which pos- sessed all the characteristics convincing of parasitism, viz., deep etration of the host plant (Laurencia obtusa), reduction of the Ilus to a tubercle and color varying from dark red, through ange to pale yellow. In addition to these, it showed cystocarps, antheridia and tetrasporangia. In the case of Janczewskia, as de- ibed by Solms, there can be no doubt either as to the parasitism as to the nature of the parasite. It is to be noted here also, that e parasite is very closely related to its host and is the first of this of parasite to be described among the red alge. Between the years 1876 and 1889, little was done to further our owledge of parasitism among the red alge. McNab (1876) re- ded the finding of Choreocolax polysiphonie Reinsch near Dublin and speaks of its tubercular thallus and penetrating rhizoidal por- mn, but mentions no reproductive organs of any kind. Bornet, in 78 (pp. 97-99, pl. 50, f. 1-8) described on Jania rubens Lamour a’ ty distinct parasite which he named Melobesia Thureti, noting at it had been described by Harvey as early as 1849 (pl. 201) asa nd kind of tetrasporangial conceptacle of Corallina squamata Park. In 1881, Solms Laubach (p. 57, pl. 1, f. 5, pl. 3, f. 12) de- ibed a second parasitic Corallina which he named Melobesia de- nans growing on and in Jania natalensis Harv. and whose apices distorts through the action of its penetrating rhizoidal filaments. pebius, also, described a Ceramiaceous parasite on Centroceras ERROR RT MIRC the genuine study of the parasitic red alge. Farlow (1889, p. 6) le known the tetrasporangia of Choreocolax Polysiphonie 160 SETCHELL—PARASITISM AMONG RED ALG, — Reinsch. Schmitz and Reinke (cf. Reinke, 1889, p. 28) de: Harveyella mirabilis (Reinsch) Schmitz et Reinke (Choreocoi mirabilis Reinsch) with its antheridia and cystocarps, and Schmitz, in his “ Systematische Uebersicht der bisher b Gattungen der Florideen,” enumerated eight distinct genera o sitic red alge, viz., Actinococcus Kuetzing, Ricardia Derbes « Choreocolax and Syringocolax Reinsch, Janczewskia Sols sporium Moebius, Harveyella Schmitz et Reinke and Choreo Schmitz. The last genus was created to contain the - lo Thureti Bornet. ae The enumeration of Schmitz, together with the « tetrasporangia, antheridia and cystocarps in J anczewskia and in sporium, the antheridia and cystocarps in Harveyella, the cystoca and tetrasporangia in Choreonema, and the tetrasporangia in Ch colax were convincing as to the existence of real parasites and 1 search for more. In 1891, Richards described the cystocarps as well the tetrasporangia of Choreocolax Polysiphoni@ Reinsch. In 1 Batters made known Gonimophyllum Buffhami with its cystoca and tetrasporangia. In 1892, Schmitz published a discussion o tubercular growths on various red alge, with a view to disting ing those which are true parasites from those which are warts or galls. In 1893 Heydrich created the genus Pleur idium, a Rhodomelaceous genus parasitic on one of the F (Fucodium) in New Zealand. It is dwarf, penetrating and vided with antheridia, cystocarps and tetrasporangia. In th year, Schmitz published his very memorable paper on Actinoco After a full discussion, Schmitz distinguished between Actinoce and the true nemathecia of Phyllophora and enumerated four sp of Actinococcus besides two for Colacolepis and two for Sterro: He also considered the placing of these genera, deciding upon Gigartinacee, the same family to which the hosts belong, rather the Squamariacee, where Actinococcus had previously bee: signed by J. G. Agardh. The Actinococcus paper of Schmitz provoked considerable ¢ cussion. Darbishire (1894) who had been investigating the sp of Phyllophora very carefully, held that the so-called Actinoc species were the true nemathecia of Phyllophora and repeated this SETCHELL—PARASITISM AMONG RED ALG. 161 5. Later, in 1899, however, Darbishire, after the death of hmitz in 1894, published the results of further investigation which resulted in finding the germinating stages of Actinococcus n the antheridial cavities of the Phyllophora and came to the same int of view as Schmitz. This point of view was also confirmed the investigations of Gomont (1894). In 1894 Kuckuck described the tetrasporangial plant of Harvey- mirabilis (Reinsch) Schmitz et Reinke, under the name of horeocolax albus, and in 1895 Batters published the genus Callo- lax of Schmitz, a genus whose single species is parasitic on Callo- yilis, which is very closely related to itself. In 1896-97 appeared those parts of Engler and Prantl’s Natuerlichen Pflanzenfamilien” dealing with the red alge and th whose preparation Schmitz had long been busy. The genera parasitic red alge were worked ’over either by Schmitz himself, by Hauptfleisch or, in case of the Rhodomelacex, by either Schmitz or Falkenberg. The total number of genera of parasitic d alge was increased from the eight detailed in 1889 to nineteen, e five genera proposed as new in this work belonging entirely to. ie Rhodomelacee and parasitic on other members of the same mily. Since 1897, additions have been made to the list of both genera and species of parasitic red alge. One genus already proposed, ., Callocolax, was not included in the Engler and Prantl account. is makes twenty genera known up to the close of 1897. Rosen- vinge added Ceratocolax in 1898 (p. 34) and in the same year Foslie (1898, p. 7) created the genus Chetolithon to receive the Melobesia deformans - Solms. Falkenberg (1901) in his monograph of the Rhodomelacez, published more detailed descriptions and figures of the various parasitic genera of this family previously proposed 1 Schmitz and by himself, but added no new genera. In 1905, Setchell and Lawson (cf. Setchell, 1905, p. 7) proposed the genus Peyssonneliopsis and in 1910 Setchell and Wilson) cf. Wilson, 1910, p. 81) proposed the genus Gracilariophila, which from their point of view is a Gracilaria-like genus parasitic on a Gracilaria (cf. how- ver Eddelbiittel, 1910, p. 230, 231, and Svedelius, 1911, p. 220—). n 1913, Yendo (p. 283) described and figured a most interesting RS ATEN RETIN EMME I Nw I TIN” TE schism 162 SETCHELL—PARASITISM AMONG RED ALG, * new genus of parasitic red alge belonging to the Rhodom which he named Benzaitenia. The single species is paras other Rhodomelacee. Finally, M. A. Howe (1914, p. 90) has mé known the genus Lobocolax, which he refers doubtfully to “ Nemalionacee ” (Helminthocladiacee Auctt.). The single species forms tubercles on Prionitis decipiens (Mont.) J. Ag. To summarize the genera thus far proposed of parasitic alge, there were nineteen recorded by Schmitz and Fale 1b , 1897, and six have been proposed since that time. In addi may be stated that there are four additional genera as abe scribed in the collections of the writer. The total number of g known to the writer, therefore, amounts to twenty-nine. These all reduced or condensed as to the thallus, penetrating and a ently forming protoplasmic connections with the host plants varying from full deep red to pure shining white. aaa In regard to species, the number assigned thus far to these and fairly certain, number about fifty and it seems best to. list of these arranged by families and to indicate in connection each its host or hosts, in order that the basis for further diseu may be made clear. In this list there have been included only species which seem fairly certain as representing definite and di: parasites. All decidedly doubtful species are omitted. An placed against those species which are credited to hosts among red algz but not of the same family as the parasite and a + agai those parasitic on Phzophyceze. The rest are parasitic on red of the same family as themselves. HELMINTHOCLADIACE, *1. Lobocolax deformans Howe (1914), on Prionitis deci bin Mont. J. Ag. GELIDIACE2, *2. Choreocolax polysiphonie Reinsch (1874-75), on Polysip fastigiata (Roth) Grev. *3, Choreocolax tumidus Reinsch (1874-75), on Ceramium at Cystoclonium purpuracens (Huds.) Kuetz. 2 *4. Choreocolax cystoclonti Kylin (1907), on Cystoclonium p purascens (Huds.) Kuetz. : _ SETCHELL—PARASITISM AMONG RED ALG2. 163 Harveyella mirabilis (Reinsch) Schmitz et Reinke (1889), on _ Rhodomela. Harveyella pachyderma (Reinsch) Batters (1902), on Graci- laria confervoides (L.) Grev. GIGARTINACEZ. “Actinococcus subcutaneus (Lyngb.) Rosenvinge (1893), on Phyllophora Brodigi (Turn.) J. Ag. and P. imterrupta (Grev.) J. Ag. . Actinococcus aggregatus Schmitz (1893), on Gymnogongrus Griffithsie (Turn.) Mart. . Actinococcus pelteformis Schmitz (1893) on Gymnogongrus norvegicus (Gunn.) J. Ag. and G. crenulatus (Turn.) J. ae ». Actinococcus latior Schmitz (1893), on Gymnogongrus dilata- tus (Turn.) J. Ag. Actinococcus mollis M. A. Howe (1914), on Gymnogongrus _ disciplinalis (Turn.) J. Ag. Actinococcus Chiton M. A. Howe (1914), on Gymnogongrus linearis (Turn.) J. Ag. Colacolepis decipiens Schmitz (1893), on Phyllophora Heredia (Clem.) J. Ag. Colacolepis incrustans Schmitz (1893), on Phyllophora nervosa _ (DC.) Grev. . Sterrocolax decipiens Schmitz (1893), on Ahnfeldtia plicata (Huds.) Fr. Sterrocolax crasstor Schmitz (1893) on “ Gymnogongrus fas- : tigiatus var. crassior Ruprecht.” Ceratocolax Hartzti Rosenvinge (1898), on Phyllophora inter- rupta (Grev.) J. Ag. Callocolax neglectus Schmitz (1894), on Callophyllis laciniata (Huds.) Kuetz. SPH ZROCOCCACEZ. Gracilariophila oryzoides Setchell et Wilson (1910), on Graci- _ laria confervoides (L.) Grev. 164 20. 2; 134. . Ricardia Montagnei var. gigantea Farlow, on Lau . Janczewskia verruceformis Solms (1870); on . Janczewskia tasmanica Falkenb. (1897), on Lawrencia . Janczewskia moriformis Setchell (1914), on . Janczewskia Gardnerii Setchell et Guernsey (1914) 0 on . Janczewskia Solmsii Setchell et Guernsey (1914), on i" . Microcolax botryocarpa (Hook. et Harv.) sme . Pleurostichidium Falkenbergu Heydrich (1898)) . Colaconema pulvinatum Schmitz (1897), on Vidalia ’ . Colacodasya inconspicua Schmitz (1897), on Polystah . Colacodasya verruceformis Setchell et McFadden (191 : . Stromatocarpus parasiticus Falkenberg (1897), on SETCHELL—PARASITISM AMONG RED ALG, DELESSERIACE. Gonimophyllum Buffhami Batters (1892), on ‘N laceratum (Gmel.) Grev. BoNNEMAISONIACE2. Ricardia Montagnei Derbes et Solier (1856) on Loure tusa (Huds.) Lamour. RHODOMELACEZ. tusa (Huds.) Lamour. (Mert.) Grev. purpurea Harv. . Janczewskia lappacea Setchell (1914), on Cha te . Q Harv. cia spectabilis R. & R. cia subopposita (J. Ag.) Setchell. Strebdocladia neglecta Schmitz. phora chondrophylla (R. Br.) Harv. (Suhr.) J. Ag. Heterosiphonia. Chondria. Haplodasya Reinboldii Falkenberg (1897), on Cystopho 7 : troflexa (Labill.) J. Ag. phonia virgata (Ag.) Spr. - SETCHELL—PARASITISM AMONG RED ALGZ. 165 Tylocolas microcarpus Schmitz (1897), on Lenormandia spec- _tabilis Sond. Benzaitenia yenoshimensis Yendo (1913), on Chondria crasst- _ caulis Harv. and Laurencia paniculata J. Ag. CERAMIACE. Syringocolax macroblepharis Reinsch. (1874, 75) on Gelidium cartilagineum (L.) Gaill. 39. Episporium centroceratis Moebius (1885), on Centroceras . clavulatum Mont. SOQUAMARIACE. Peyssonneliopsis epiphytica Setchell et Lawson (1905) on _ “ Meredithia californica J. Ag.” CoRALLINACEZ. Choreonema Thureti (Bornet) Schmitz (1889), on Jania rubens Lamour and Corallina squamata E. & S. Chetolithon deformans (Solms) Foslie (1898), on Jania natal- ensis Harv. addition to the species listed above and which have been ed from the species published as being certain or very nearly ere have resulted from collections, chiefly by N. L. Gardner, the Pacific coast of North America, some nine additional but as unpublished species, together with four additional genera, also yet unpublished, as noted previously. All of these species are sitic on genera closely related to themselves. To make this : evident and, at the same time, to make known the distribution ong the families of red alge of the new genera and species, a ef 1 resumé of these as yet unpublished species is appended. In hzerococcacez, a new species of Gracilariophila has been found on wlaria Cunninghamit J. Ag.; in Rhodymeniacee, three new of a single species each on Fauchea laciniata J. Ag., Rhody- Palmetta (Esp.) Grev.?, and on Plocamium coccineum .) Lyngb., respectively ; in Delesseriaceze, two new species of ophyllum, one on Nitophyllum Ruprechtianum J. Ag. and an- C, AMER, PHIL. SOC., VOL. LVI, L, JUNE 21, 1918. See Ea teases, 2? sande ea an a ihe 166 SETCHELL—PARASITISM AMONG RED ALGE, __ other on a species of Delesseria, as well as a single species of a new genus on Neuroglossum Andersonianum J. Ag.; and in Rhod melacee, two new species of Stromatocarpus on Pterosip Baileyi (Harv.) Falkenb, and on another species of the same gen’ respectively. These new genera and species are in a fairly advanced . stage of preparation towards description and illustration. In summarizing the distribution of these distinctive par it genera, among the red alge, we find the results as follows: m an Helminthocladiacee ........... I genus with I species. Gelidiacee ....... pals Benen .2 genera with 5 species. © Cipartinaces: soci caer sinks 5 genera with 12 species. mpricerocbccntem 47.45 65 ssee es I genus with 2 species. eu BOC YMCHIBCRR soos Sass sos 3 genera with 3 species. Drelesseriacee ces saws ss 2 genera with 4 species. Bonnemaisoniace®:............. I genus with 2? species. Rhodomelacete 7 otoy 0a. .ss ss 9 genera with 17 species. CHUAMIECOR Co iis se tek 2 genera with 2 species. — Squamanacese srs 6. I genus with 1 species. COPAIUGACOR iy sas at so as 2 genera with 2 species. This summary shows clearly the extent of the distribution parasitic genera and species through the group of the red alge the fact that they are, thus far, known only from eleven a twenty-one families into which the group is usually divided. 1] shows that of these eleven families, two, viz., Gigartinacez ( : genera and I2 species) and Rhodomelacee (with 9 genera and species) contain one half or over of the known genera and Besides the literature dealing with the strictly systema and describing, for the most part new species, there are a few which attend mostly to other matters connected with the pa red alge. Such, for instance, is the paper by Nott (1897) di ing the finding of certain parasitic red alge on the coast of - nia, and a similar paper by the writer published later (cf. Setchell, 1905). Sturch (1899) published the results of a careful study in’ the structure, development, and nature of the parasitism of Har- veyella mirabilis Schmitz and Reinke and its systematic position. 1905, Oltmanns (p. 319 et seq.) discussed parasites among the < SETCHELL—PARASITISM AMONG RED ALG#. 167 arly describing and illustrating the parasitism of Ricardia, i ococcus, Harveyella, Janczewskia, Stromatocarpus, Choreo- na (“ Melobesia Thureti Born.”) and Chetolithon (“ Melobesia deformans Solms”). The penetration, reduction of the thallus and natural relationship to the host plant are all dealt with. This is the only general discussion of the parasitism of the red alge thus far published. In 1910 Eddelbiittel published a general account of asitism among the red alge, with special reference, however, to reocolax and Harveyella. This account dealt particularly with systematic position of Choreocolax and Harveyella, advocating emoving them from the Gelidiaceez, where they had been placed Schmitz, and placing them among the Gigartinales, as Sturch 9, p. 98) had advocated. He also discussed Gracilariophila shell and Wilson (1910), suggesting placing it near, if not uniting with, Choreocolax. I am unable to agree with this latter view ecause of the very different structure of the cystocarp in the two era, the essentials of which, viz., the different shape and ar- gement of the spores, Eddelbiittel did not mention or seem to der in his discussion. From all the previous consideration, two things, at least, seem in. First, there are approximately twenty-nine genera and fifty- species of undoubted and peculiar parasites among the red -. Doubtless there are many of similar character to be discov- 1. In fact, there is knowledge, but of unsatisfactory character, existence of a number of such. Doubtless also, there are possibly many, at least partial parasites among the various epiphytes” and “endophytes,” so numerous in the group. Seoond, there is an overwhelming restriction in the matter of itism on other red alge and even on other members of the ie family. Of 51 parasites enumerated above, 41 are fairly cer- nly parasitic on another member of the same family. This is a ttle over 80 per cent. of the whole number. Of the remainder, 8 little less than 16 per cent. are parasitic on red algz not of the family (t. e., practically 96 per cent., therefore, parasitic on 3 red alge), while only two or a little less than 4 per cent. are 2 > on alge (brown) other than red. Abhough acquainted with a far less number of cases Batters 168 SETCHELL—PARASITISM AMONG RED ALG. — (1892, p. 66 and 1895, p. 317), Schmitz (1893, p. 390) and Olt- manns (1905, p. 334) have all spoken emphatically of the fact that so many of the parasitic red alg are restricted to near relati i as hosts. Oltmanns further remarks (loc. cit.) that no satisfactory | ex planation of this can be brought forward. The suspicion has" produced, especially earlier in the progress of our knowledge, that some of these parasites, especially some species of Actinococcus, are really parasitic tetrasporangial generations of the hosts they inh Darbishire (1899, p. 264) has voiced this suspicion and has s his opinion that while this is not impossible, it is not very prob The probability, as it seems to the writer, is, however, that the various parasites, or some of them, may have originated in close connection with their hosts by some mutation decreasing the chlorophyll tent or power in one or other of the different forms of spore. . an inducement to increase the power of penetration and possi protoplasmic connection between a spore (tetraspore or carpospore) S germinating in position might, it would seem probable, initiate f sitism on the parent plant, and this parasitic tendency incre penetration and dwarfing, might, therefore, be inheritable. There is one case known which seems to be such a case line with such action. This is the condition found in Agard) ell tenera (J. Ag.) Schmitz (““Rhabdonia tenera” J. Ag.) by Oster hout (1896). It was noticed that the tetrasporangial plants in many cases, numerous short bristle-like branches or prolifera projecting at right angles to the main stem and. branches, but the antheridial and cystocarpic plants are always destitute of Examination showed that these peculiar branchlets are usua theridial, while full-sized antheridial plants are rare. Some however, the bristly short branchlets have tetrasporangia or cys carps and the three kinds of reproductive bodies are some: ! borne on branchlets side by side. Some of the branchlets sterile “but in the majority of cases they bear reproductive a before they are more than a quarter of an inch in length.” (O hout, loc. cit., p. 420.) It was found by Osterhout that the zonate tetrasporangia di in regular fashion forming what seem to be four tetraspores arranged serially. These spores then sometimes divide further SETCHELL—PARASITISM AMONG RED ALG#. 169 ue divisions and as this continues the contents of a tetrasporan- act as a whole, producing penetrating rhizoidal filaments below regular, though much dwarfed, Agardhiella-frond above. The filaments penetrate even into the region of the medullary of the parent plant and establish “secondary connections ” the making. There exist full-sized plants of Agardhiella tenera all three sorts, viz., antheridial, cystocarpic and tetrasporangial. here exist also dwarf plants, parasitic on, but arising from, the rasporangial plant. While these dwarf plants, and they are very h reduced and simple, are largely antheridial, yet, according to Jsterhout,, all three kinds of dwarf plants, viz., antheridial, cystocar- and tetrasporangial, may exist side by side, all parasitic on and obably arising from the same full-sized tetrasporangial plant. In ler that our knowledge of this interesting and seemingly very sig- nt case may be more complete, it is very desirable that culture made from the spores (both carpospores and tetraspores, if ob- able) of the dwarf plants. It is very desirable that this be dertaken by some investigator who has access to abundant growths these plants. The very similar species, Agardhiella Coulteri vy.) Setchell, of the California coast has not been observed to , nduce dwarf plants (or bristly proliferations) from the tetra- generation. In summarizing, then, the aim of this paper, it may be said to intended to indicate how general is the extent of the parasitism _ of the parasitic genera and species of red alge upon their near rela- tives and to draw attention to the similarity of these cases and the se of the production of a dwarf parasitic generation from the asporangia of Agardhiella tenera and with the hope of suggesting _ the probability of their origin. BIBiioGRAPHY. s, E. A. L. 1892. Gonimophyllum Buffhami: A New Marine Alga. cane Journ. of Botany, Vol. 30, pp. 65-67, PL. 310. "1895. On Some New British Marine Alge. Annals of Botany, Vol. 9, pp. 307-321, Pl. 11. anias A Catalogue of the British Marine Alge. Suppl. Journal of Botany, Vol. 40, pp. 1-107. , Ed. 1878. Consult Thuret et Bornet, 1878. 170 SETCHELL—PARASITISM AMONG RED ALG Darbishire, Otto Vernon. 1894. Beitrag zur Anatomie und En geschichte von Phyllophora. Bot. Centralblatt, Vol. 369. 1895. Die Phyllophora-Arten der westlichen Ostsee deutschen Wiss. Meeresunters. herausgegeb. v. d. Kom. z. Unters, « Meere in Kiel u. d. biolog. Anstalt a. Helgoland, : n¢ Vol. 1 (pp. 38, 48 text-figures). 1899. On Actinococcus and Phyllophora. Annals of Botany pp. 253-267, Pl. 15 and text-fig. 1-7. Derbes. 1856. Description d’une Nouvelle Espéce de Floridée un nouveau Genre, et Observations sur quelques Sci. Nat., bot., Ser. 4, Vol. 5, pp. 209-220, Pl. 14. Eddelbiittel, H. 1910. Ueber die Kenntniss des parasitaren als “ Parasiten ” bekannten Florideen, inbesondere de Choreocolax Reinsch und Harveyella Schm. et Rke. Zeitung, Vol. 68, Abth. 2, pp. 186-191, 226-232. Falkenberg, P. 10901. Die Rhodomelaceen des Golfes von Neapel | Angrenzenden Meeresabschnitte. Fauna und Flora d S von Neapel, 25 Monographie. (Pp. xvi and 754, 24 and text fig.) Farlow, W. G. 1889. On Some New or Imperfectly ree Al United States. 1. Bull. Torrey Botan. Club, Vol. 3 : Pl. 87, 88. ae Foslie, M. 1808. List of Species of Lithothamnia. Kgl. Norske Selskabs Skrifter, 1898, No. 3. 1906. Revised Systematical Survey of the Melobesiez. Videnskab. Selskabs Skrifter, 1900, No. 5. Gibson, R. J. Harvey. 1891. Notes on the Histology of Polysip lonié tigiata (Roth) Grev. Journal of Botany, Vol. 29, pp Pl. 304. a Harvey, William Henry. 1849. Phycologia Britannica, Vol. 2 (op 121-240). Heydrich, F. 1893. Pleurostichidium, ein Neues Genus der Rhodo: Ber. deutsch. bot. Gesell., Vol. 11, pp. 344-348, Pl 16. 1906. Die systematische Stellung von Actinococcus Kitz. i Ve pp. 71-77, Pl. 5. . Howe, Marshall Avery. 1914. The Marine Alge of Peru. ew Botan. Club, Vol. 15 (pp. 185, Pl. 66, text-fig. 44). ~ Kuckuck, Paul. 1894. Choreocolax albus n. sp., ein echter Schmarotze den Floridien. Sitzungsber. d. Acad. Berlin, 1894, pp. PLO Kuetzing, Friedrich Traugott. 1843. Phycologia Generalis oder Anat Physiologie und Systemkunde der Tange (pp. xvi and 4 80). SETCHELL—PARASITISM AMONG RED ALG#. 171 Harold. 1907. Studien ittber die Algenflora der Schwedischen West- kiiste. Inaug. Diss., ee, pp. iv and 288, Pl. 7, map and 41 text-fig. Clara K. 1904. Observations on Callymenia phyllophora J. Ag. Minnesota Botanical Studies, Ser. 3, Part 3, pp. 291-296, Pl. 44, 45. H. Christ. 1819. Tentamen Hydrophytologie Danice (pp. xxxii F and 248, Pl. 70). 1834. Chetophora subcutanea, in Hornemann (J. W.), Flora Danica, a Vol. 12, fasc. 36, p. 8, Pl. 2135, f. 2. Fadden, Mabel Effie. 1911. On a Colacodasya from Southern California. Univ. Calif. Pub. Bot., Vol. 4, pp. 143-150, Pl. 19. ab. 1876. Exhibition, New to Ireland, of the New Parasitic Rhodo- sperm, Choreocolax Polysiphonie Reinsch. Quart. Journ. Mic. Science, N. S., Vol. 16, p. 336. M. 1885. Ueber eine neue Epiphytische Floridee. Ber. deutsch. bot. Gesell., Vol. 3, pp. 77-80, PI. 7. ott, Charles Palmer. 1897. Some Parasitic Floridee of the California oe coast. Erythea, Vol. 5, pp. 81-84. ons, Friedrich. 1905. Morphologie und Physiologie der Algen, Vol. 2 (pp. 443, text fig. 468-617). out, Winthrop J. V. 1896. On the Life-History of Rhabdonia tenera J. Ag. Annals of Botany, Vol. 10, pp. 403-427, Pl. 20, 21. old, Th. 1899. Meeresalgen von Investigator Street (Siid-Australien). 2 Hedwigia, Vol. 38, pp. 39-51. J. 188. Algenflora der westlichen Ostsee deutschen Anteils. Sechster Bericht der Komm. zur Wissenschaftlicher Untersuch. der deutschen Meere in Kiel, 17-19 Jahrg., I. Heft., pp. 1-101, 8 text-fig. and 1 chart. h, Paul Friedrich. 1874, 75. Contributiones ad Algologiam et Fungo- logiam (pp. xii and 103, Pl. 36, 61, 18 and 9). Herbert Maule. 1891. On the Structure and Development of Choreocolax Polysiphoniz, Reinsch. Proc. Amer. Acad., Vol. 26, pp. 46-63, with plate. envinge, L. Kolderup. 1893. Grgnlands Havalger. Meddelelser om Grgnland, III., pp. 765-981, Pl. 2, 57 text-fig. 1898. Deuxiéme Mémoire sur les Algues Marines du Groenland Meddel- ‘ elser om Grdnland, XX., pp. 1-125, Pl. 1, 25 text-fig. ke The Marine Alge of Denmark. Kgl. Danske Vidensk. Selsk. Skrifter, 7 Raekke naturvidensk, og Mathem. Afd., VIL, 1. Fr. 1889. Systematische Ubersicht der bisher bekannten Gattungen ema der Florideen. Flora, Vol. 72, pp. 435-456, Pl. 21. 892. Knolichenartige Auswiichse an den Sprossen einiger Florideen. Botan. Zeitung, Vol. 50, pp. 624-630. 1893. Die Gattung Actinococcus. Flora, Vol. 77, pp. 367-418, Pl 7 (also text-figures). 172 SETCHELL—PARASITISM AMONG RED. 1897. Rhodophycee, in Engler und Prantl, Die na familien, I. Teil Abth. 2. (1896, 1807 with P. P. Falkenberg). @ Schmitz et Reinke. 1889. See under Reinke, 1889. Setchell, William Albert. 1905. Parasitic Floridee of Califo or Notarisia, Ser. 16, pp. 59-63. as. 1914. Parasitic Floridee, I. dad Calif. Pub. Bot, Ve Pl. 1-6. Solms-Laubach, M. le Comte H. de. 1877. Note sur le Jas n ’ Floridée Parasite du Chondria obtusa. Mem. _ Cherbourg, Vol. 21, pp. 209-224, Pl. 3. 1881. Die Corallinenalgen des Golfes von Neapel und der A An Meeresabschnitte. Fauna und Flora des ba Monographie. (Pp. 64, Pl. 3.) Sturch, Harry H. 1899. Harveyella mirabilis (Schmitz and of Botany, Vol. 13, pp. 83-102, Pl. 3, 4. ee Svedelius. Thuret (Gustave) et Bornet (Edouard). 1878. Etudes Fuye iii and 105, Pl. 51). Be Wilson, Harriet L. 1910. Gracilariophila, a New Parasite on fervoides. Univ. of Calif. Pub. Bot., Vol. 4, pp. 7: Yendo, K. 1913. Some New Alge from Japan. Nyt Ma videnskaberne, Vol. 51, pp. 275-288, Pl. 13, 14. N ANNOTATED TRANSLATION OF THE PART OF ‘SCHWEINITZ’S TWO PAPERS! GIVING THE RUSTS OF NORTH AMERICA... By J. C. ARTHUR anp G. R. BISBY. (Read April 13, 1917.) Lewis David von Schweinitz was elected to membership in the nerican Philosophical Society in 1817, one hundred years ago. Te was at the time a resident of Salem, North Carolina, a talented : n of forceful character, secretary of the Moravian Missions of . orth America, and with one important botanical work to his credit. n 1805 there had been published in Leipzig a volume describing the i about Niesky,? a town of Saxony (later of Prussia), being the : product of teacher and pupil during Schweinitz’s four years’ az ecourse. The plates of the volume, with more than a hundred res, were drawn, engraved and colored by Schweinitz, and much the text bears the impress of his labor and judgment. After five years of college teaching subsequent to his gradua- n, and five additional years in the ministry, he returned to nerica as general agent of the Moravian church in the Southern states, and became the pioneer mycologist of the New World. He s the only mycologist in the United States who added materially to the literature of mycology during the half century following his ognition by the American Philosophical Society. His magnus is, which was truly a colossal work for the times, no less a work in a systematic account of the known fungi of North America, 1 The papers referred to are the following: “Synopsis fungorum Caroline superioris secundum observationes,” Schriften Nat. Ges., Leipzig, 1: 20-131. 1822. The rusts on pp. 65-75. _ “Synopsis fungorum in America Boreali media degentium secundum ob- vationes,” Trans. Amer. Phil. Soc., Il, 4: 141-316. 1832. The rusts on P. 208, 209, 290-297, 306-314. dg 2 Albertini & Schweinitz, “Conspectus fungorum in Lusatie superioris o Niskiensi crescentium,” pp: 376, pl. col. 12. Lipsiz, 1805. PROC. AMER, PHIL. SOC. LVII. M, JULY 16, IQI8. 173 174 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S with nearly 4,000 species and 250 genera, was presented to the tific world through the Transactions of the American Philosot Society, having been transmitted to the Society April 15, 1831 issued in printed form about a year later. It is usually spoken “Synopsis of North American Fungi” from the SCORE one at the top of the pages. : It seems, therefore, especially fitting that on ‘the centennial versary of Schweinitz’s election to membership, the Society s take cognizance of his eminent and invaluable services to sc encouraged and.aided as they were by the Society’s approval. No second attempt has followed Schweinitz’s effort to a full survey of the fungous flora of North America until rec when the “ North American Flora,” to include all classes of | from the highest to the lowest, was projected and supported New York Botanical Garden. In this work the fungi are to ten or more imperial octavo volumes, and the text is to be s by many specialists. One volume is to contain the Uredin rusts, and its preparation has been intrusted to the senior w this article, aided by the junior writer and other mycologist pursuance of this work it has become necessary to know defi the extent of the contribution to the subject made by Schweini amount so considerable in fact that his name is encountered systematic student of the American rusts in much the same’ the name of Linnzus is encountered by the student of the flo plants. The result of the detailed examination of the speci the Schweinitz herbarium, now deposited with the Acaden Natural Sciences of Philadelphia, and the interpretation of h lished account in the light of this study of his original mate presented to the Society in the following annotated translation the Latin into English of that portion of Schweinitz’s works ] ing to the rusts. In Schweinitz’s day the rusts were not recognized as a dis and sharply defined group of fungi, as they now are, but some extent classed with other fungi occurring on living or lang ing hosts. They are all of microscopic size, but usually pr some characteristic discoloration or hypertrophy of the subs which aids in making them noticeable. In a few instances PAPERS GIVING RUSTS OF NORTH AMERICA. 175 changes in the host amount to conspicuous alterations that attract the casual observer, as in the case of “cedar apples,” and all the more so because the distortions are often accompanied by brilliant coloration. For the study of these small objects Schweinitz was dependent upon lenses of poor definition and no considerable magnification. His chief instrument was undoubtedly the pretentious one now in the possession of his grandson, the eminent oculist of Philadelphia, Dr. Geo. de Schweinitz.* This is still in almost or quite as good condition as when purchased probably some time prior to 1817. It was evidently one of the best instruments to be had at that period. As was pointed out in an early paper pertatning to the rusts, the first published on the subject by the senior author,* a magnification of dry spores amounting to seventy-five diameters will give an ap- pearance answering to the most detailed parts of Schweinitz’s diag- noses. It is considered by Shear & Stevens,’ who kindly loaned to the writers during the preparation of this paper copies of their manuscripts embodying results of researches pertaining to Schwei- tz’s scientific labors and collections, that Schweinitz had to deal with a greater handicap than low magnification in his microscopic work. They find that the lack of spherical and chromatic correc- tion of the lenses and the poor illumination must have resulted in decidedly inferior definition. _ But in many cases it is clear that Schweinitz drew up the descrip- tions of his new species without making use of this instrument. He doubtless had some form of hand lens, although considerable inquiry has failed to reveal any present trace of such a glass. Even a simple hand lens seems not to have been used at times, and in neral much dependence was placed upon the gross appearance and the changes wrought in the host. _ It would be interesting to know what facilities in the way of books were possessed by Schweinitz. Probably his botanical 8 The instrument was kindly loaned by Dr. de Schweinitz for display before the Society at the presentation of this paper, and is illustrated by ‘Shear and Stevens in Mycologia for July, 1917. __ * Arthur, “ The Interpretation of Schweinitzian and Other Early Descrip- tions,” Amer. Nat., 17: 77-78, Jan., 1883. 5 Mycologia, 9: 195, 1917. 176 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S library was not large, but what works it contained can only be ferred. There are no records of books having been given to Academy of Natural Sciences of Philadelphia or to the Ame: Philosophical Society, and no such books are now in possession his descendants. In a letter to the senior author, dated Decemt 2, 1916, the Rev. Dr. Paul de Schweinitz, secretary of missions the Moravian Church now living at Bethlehem, Pa., says that grandfather who died in 1834 “left four sons, the oldest of w [Emil] was only eighteen. The presumption would naturally that when his widow died twenty-four years afterward [in 1 his [botanical] books would have been divided among the sons. I do not recall seeing any in my father’s library. My fa [Robert] was the last of his four sons to die.” The widow to) third son, Mrs. Edmund de Schweinitz, is still living in Philade phiz and graciously received Dr. C. L. Shear and the senior author the evening of February 5, 1917, but did not recall having ever any of Schweinitz’s botanical books. It is probable that the current works of Pursh, Micha N ; Darlington, Bartram, Torrey, Barton, Muhlenberg,and other Ame botanists of the time were at his disposal in studying the flowe plants. Of these doubtless Barton’s “Flora of Philadelp (1818), but above all Muhlenberg’s “ Catalogue” (1813, 2d edi in 1818) and Torrey’s writings were in constant use. Alt Amos Eaton, of Yale College, published a “‘ Manual of Botan 1818, with successive editions until 1840, it does not appear to been his guide in matters of nomenclature. There were no American works on fungi at the time Sch was most active in preparing his important contributions. Na he brought to this country the knowledge and many of the which had aided in making the “ Conspectus of Fungi about Ni prepared by himself and his teacher, Albertini, a work of sta value. In that work, as well as in the Carolina list he foll Persoon very closely as his model, and did not think it advisab attempt any marked deviation from what he considered an aut! tative nomenclature and systematic arrangement. In 1825 treatment of the Hyphomycetes and Gymnomycetes for Willden edition of the “ Species Plantarum” became available, and rece’ Schweinitz’s full indorsement. PAPERS GIVING RUSTS OF NORTH AMERICA. 177 Among the innovations introduced by Link and adopted by iweinitz in his later work was the use of the genus Ceoma to in- de what had before passed under the genera Uredo, Zcidium, ridermium, etc. These older genera were only half ingested, however, and a sort of double generic name was made, that is, the _ genus and subgenus were used together: it was Ceoma (Uredo), oma (Zcidium), etc. But this proved too clumsy for general use,and we find Schweinitz constantly reverting to the older nomen- ature in his comments, as under 2887, Ceoma (cidium) lumi- m, he speaks of “this AEcidium,” not of this Ceoma, or of this @oma (Z:cidium). Link’s genus Ceoma never found much sup- ort, and eventually fell into disuse, although the older application the name as a genus coérdinate with Uredo, Zcidium, etc., is © still in favor, these names in the most modern usage constituting form-genera. In the list of species placed by Schweinitz at the end the volume, as those first detected by him in America, he lists cidium, Ceratites and Peridermium with initial rank, each with oma as subgenus, leaving Ce@oma as a genus to include only the one subgenus, Uredo, thus indicating some revolt, or at least incli- ion to deviate from Link’s method. That the form of name ven in the final list was no careless indexing but the conclusion of ture judgment seems certain from the use of one of these names the description of 2932, P. investita, where he speaks of “ Zcidium phalitatum,” the name in the final list, and not of Ceoma Gnaph- unt, as given in the body of the work under 2873. _ Another unfortunate innovation by Link faithfully adopted by ‘Schweinitz was the change of specific names having the form of a oper noun, usually in the genitive singular, to the form of an ad- tive. Thus Zcidium Galii became Ceoma galiatum, A. Ber- is became C. berberidatum, A. Viole became C. violatum, and added many more, i. ¢., Ceoma pyrolatum, C. hepaticatum, C. i ‘atum, C. dracontionatum, C. houstoniatum, C. pedatatum, C. ratitatum, C. helianthatum, C. trachelifoliatum and eighteen or w enty more, all of them again listed under “ Ecidium (Czoma)” t the end of the volume. These changes with few exceptions were de under the genus Ceoma. Link changed a few specific proper 178 ARTHUR-BISBY—TRANSLATION OF SCHWEINITEE: names under the genus Puccinia from the singular to the pl thus P. Galii became P. Galiorum, P. Pruni-spinose became P. ; norum, P. Viole became P. Violarum, etc., and in this was imite ite to some extent by Schweinitz as in the change of Puccinia Hel to P. Helianthorum. All these changes were with the clear intent of making ie more accurately and fully represent the facts pertaining to species. It was an attempt to carry out the idea that still pe from pre-Linnzan times, that the name should embody some acteristics of the thing named, and in so far as a binomial name mitted, be descriptive. It was logical, consequently, to bring name down to date, and upon ascertaining that the rust on P. was not confined to one species of Prunus, as at first supposed, occurred on more than one, to change the name from Puccin Pruni-spinose to P. Prunorum, and similarly so for other The same result was even better attained by using a generalized jective form for the specific name. It must be borne in mind DeCandolle’s dictum that the first name given to a species w only legitimate name and should not be changed because fou be inappropriate had only been stated in 1813, and had receiv general adherence, certainly not by German authors. : Along with the belief in descriptive names went the pi idea of the nature of species. Species were treated as cor This accounts for Schweinitz’s insistence that when Link t one of Schweinitz’s species to another genus and also specific name in accordance with reasons just stated, or any it is Schweinitz and not Link who should be cited for the new of the name. Schweinitz established Acidium Caladii, and changed the name to Ceoma (A:cidium) aroidatum, yet Schw places his initials after the latter name to indicate that it species (i. e., his concept), and not Link’s species. And so it that the names first published by Link, Ceoma Iuminatum, Puc aculeata, Podisoma macropus, and many others, founded Schweinitz’s earlier descriptions of species differently named, followed by the initials of Schweinitz in his later work. : The collection of Schweinitz’s fungi at his death in 1834, left to the Academy of Natural Sciences of Philadelphia. PAPERS GIVING RUSTS OF NORTH AMERICA. 179 specimen was preserved in a paper packet, made by folding over the _ sides of a sheet of paper until they touched or somewhat over- lapped, then folding over the ends in the same manner and in the same direction. On the back of the packet an autographic record was made in ink. When a change was necessitated in the label by the adoption of Link’s nomenclature, or for other reasons, in many cases the packet was not discarded, but refolded inside out and the data replaced in the new form on the back. This conservative prac- tice, doubtless adopted merely as a convenience in handling, has given a chronological record that has often proved of much value when studying the original material, as showing changes in Schwei- nitz’s views regarding the best form of the name or the identity of the material. The packets were of no uniformity in size, but varied from about three by six centimeters or smaller up to six by ten cen- timeters, and a few still larger. _ Some thirty or forty of these packets were placed loosely in large envelopes, folded in a similar manner to 22 by 38 centimeters from heavy steel-blue paper, and a list of the species inclosed writ- ten on the back. Three to five of these envelopes according to bulk ere put into a pasteboard portfolio of the same size and seven or ht centimeters in thickness, and tied with tape, the back being ered with the consecutive number and the genus represented. e whole collection was contained in 39 portfolios, making a series shelf volumes in outward appearance resembling a set of the odern bound fungi exsiccati. All the fungi were placed in one ies, the European, North American and Surinam specimens _ being intermixed. The part of Schweinitz’s work on North American Fungi with hich this paper has to deal is with the exception of eight species prised under the two genera Ceoma and Puccinia. The mate- 1 under Ceoma, both American and European, occupies the five envelopes in portfolio no. 38, and embraces 243 packets, of which considerably more than half are now empty. The material under uccinia occupies two of the envelopes in portfolio no. 39, and embraces 84 packets, more than half being empty. Altogether der Ceoma and Puccinia 178 collections are European, 130 being out specimens, 18 are from Surinam, 3 without specimens, and 180 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S — a 131 are North American, 60 without specimens, making a ‘total 327 packets, of which 193 are empty. So far as the North American material in the portfolios is cerned, it is only the surplus after a suitable part had been remo for mounting. The Schweinitz collections representing his. on the North American Fungi, were mounted by Dr. Ezra Michet mostly during the years 1856 and 1857. As pointed out by Shea Stevens (Mycologia, 9: 337. 1917) the packages of fungi an mounting material were sent by the Academy of Natural Sci of Philadelphia to Dr. Michener, the work being done at his hon New Garden, Pa. Even at that time some of the packets were er as in a letter to Rev. M. A. Curtis Dr. Michener says: “I have grieved to find a number of the envelopes either missing or em They were doubtless essentially in the same condition when came into the possession of the Academy some twenty years befa From a letter written to Dr. John Torrey by Schweinitz sh after his return from Europe in 1819 we learn that he had tal a full set of specimens illustating his new species together wi list of his American fungi abroad with him and left them with Schwagerichen at-Leipzig. This was the North Carolina list prin not long afterward at Leipzig under the editorship of Dr. Schwag ichen. It is not known whether or not these specimens are y existence. Taking out this set may have nearly or quite exhat his supply in some instances. Specimens were also sent to no - than fourteen individuals and herbaria according to Shear Stevens,° among them being his correspondents at Upsala, | : Edinburgh, Paris, Berlin, Vienna and elsewhere, which dou drew heavily upon his material at times. So far as concerns the part of the collections examined by writers it seems that Schweinitz was usually in the habit of mz but a single collection to represent a species and when he observe the same species in another locality he merely added the new 1 ity on the outside of the packet. In a few cases he preserved lections, made by himself or sent to him by others, illustrating ferent hosts, as of 2826 Ceoma (Uredo) Solidaginis. Occasit ally he appears to have replenished an exhausted packet by a : 6 Mycologia, 9: 333, 1917. PAPERS GIVING RUSTS OF NORTH AMERICA. 181 collection as under 2930 Puccinia Asteris, the packet says “on Aster us” but contains only material on A. cordifolius. In rare instances he may have placed a second collection of what he believed ‘be the same form in a packet still having some of the original collection. In most cases, however, the specimens now to be found in the packets appear to represent Schweinitz’s first American col- lection of that form. And so it comes around that when a species had first been found in North Carolina and subsequently found in Pennsylvania or elsewhere the material preserved to represent it generally is the North Carolina collection. This is a most fortu- é situation, as the specimen is thus the type for the earlier of Schweinitz’s names, when a change was made in the latter work. present priority rules require the use of the earliest specific mame which in the present connection is a name usually much to be ferred for its brevity and aptness. __ The fungi from North America in the portfolios as presented by Schweinitz to the Philadelphia Academy were labelled in ac- dance with his work on North American Fungi, and in large part tituted the basis for that work. Under the genera Ceoma and uccinia only one North American specimen occurs not mentioned in his published account. It is labelled “ Acidium Dircatatum Ind.,” d must have been collected upon his visit to Hope, Indiana, hale went to organize a church. This was in the summer of 1831 and doubtless too late to have the name placed in his manuscript. The packet contains three leaves of Dirca, 5 by 7.5 cm., 4 by 8.5 cm., 5 by 6 cm., the last with part of each end removed, each leaf :.. a single small group of excia. Besides the specimens which Schweinitz cited abroad, and those sent to his European correspondents as mentioned above, Many were sent to his American correspondents, and especially to intimate friend, Dr. Torrey. The last were finally given by Torrey either to Curtis and are now in the Herb. Curtis at Harvard University, or to Berkeley, and are now in the Kew Herbarium. After the collection came into possession of the Philadelphia Acad- emy portions of specimens were removed by Curtis for purpose of Study during a seventeen-day visit in 1851 (Shear & Stevens, yeologia, 9:335), part of which were transmitted to Berkeley. 182 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Not long afterward the Academy arranged with Dr. Michener to” place the collection in a more secure and accessible form, Cu having been largely instrumental in bringing this about. In mounting the collection a representative portion, or all hie the material was scanty, was taken from each packet and glued to” uniform slips of white writing paper 8 by 10 cm., on which the number, name, and source were written as given in the North Amer- ican Fungi (see cut under no, 2881). In some cases the material was placed in paper packets that were glued to the slips. These mounts were consecutively arranged by pinning them to the inner page of folded sheets of brown paper, and the sheets placed in heavy board portfolios. The portfolios, 12 altogether, are 26 by 36 cm. and tied with tape. There are 85 mounts under the genus: Ceoma, of. which five are smuts, and some others belong to non- uredinalean species, as stated under the several numbers in the sys tematic account which follows. There are in addition 6 mounts representing rusts, two under Spheria, one under Seiridium, o1 e under Gymnosporangium, and two under Podisoma. The whole genus Puccinia is unrepresented. When the senior author was preparing to make his first visit to the Academy for the purpose of examining some of the types in. the Schweinitz collection, he learned from Mr. W. C. Stevenson, Jr. (in letter dated Oct. 19, 1898), a member of the Academy, that part of the mounted collection had disappeared. Few persons had been critically interested in rusts in the recent years, and was easy to ascertain that none of them had knowledge of whereabouts of the missing specimens. No one then belonging the Academy could give any information. It was generally lieved that the missing sheets would eventually be found in the herbarium rooms of the Academy. However, a subsequent sear failed to bring the missing material to light. The researches Shear & Stevens regarding the history of the Schweinitz f have shown quite conclusively (Mycologia, 9:340. 1917) that material representing nos. 2905-2946 embracing Puccinia and some subsequent genera, was mounted by Michener and that the mounted part must have disappeared later. The original packets are still their envelopes in the portfolios. Fortunately there is some m PAPERS GIVING RUSTS OF NORTH AMERICA. 183 terial of Schweinitz’s forty-two numbers under Puccinia in the autographic packets and also in other herbaria. Dr. Farlow states that 32 of these numbers are represented in the Herb. Curtis at Harvard University and Dr. Shear writes that there are 37 in the Michener collection at Washington. The senior author has consulted the part of the Schweinitz col- lection containing the rusts a number of times between 1899 and 1917, for a few hours or a few days each time, as other duties de- manding a visit to Philadelphia or nearby cities permitted. The first visit of three hours’ duration was on Feb. 17, 1899, and a second one of about the same length of time on Aug. 4, 1900. At this second visit the impossibility of satisfactorily deciding upon the identity of many of the collections without better microscopic facili- ties and more time than could be hoped for while in Philadelphia was forced into prominence. A bit from an ample specimen, such as would furnish a few spores for examination under the micro- scope, could be carried away when the need was great, without a feeling of having done harm to this precious historical collection, but many specimens were too meager for such liberties. About a score of specimens of the unmounted material were selected at this time which most needed study and a request left to have them sent to Lafayette, Indiana, for more careful examination. But the authorities of the Academy had become wary, their attention having been called recently to the mysterious hiatus in the mounted set, including the important genus Puccinia, and had decreed a general ban on all loans. It was not until 1915 that the regulations were so far modified that the privilege was obtained to study these speci- Mens microscopically for a few days in April of that year at the laboratory in Lafayette. During the four days of December 28-31, 1903, many hours were spent in consulting the collection, at which time the senior _ author was assisted by Dr. Frank D. Kern, and again much study was given the collection during the five days of December 28, 1914, to January 1, 1915, assisted by Dr. F. D. Fromme. The senior author also consulted the collection on February 5-12, and April I-14, 1917, Dr. C. L. Shear being present part of the time during the April period and giving valuable assistance in interpreting the 184 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S data. A few hours of study were also given on other dates — now definitely in mind. In order to verify and complete the r of information secured in this fragmentary manner the authori of the Academy, upon presentation of the situation by Dr. Witme: Stone, the acting curator, most generously transmitted all of po folio 38 and 39 of the original set, and the final portfolio of the mounted set. These were received in Lafayette, Ind., the latter par of April, 1917, and returned the latter part of February, 1918, exactly the same condition as when received. Owing to this valuable opportunity for verification it is believed that the statis given in the following account are accurate within the limits of o nary error. ie It has been the privilege of the senior author to examine mez collections of micro-fungi, and he can say advisedly that Schweinitz collection shows great care in its labelling and arrang ment, and considering the vicissitudes of practically a hundred years in which the requirements of correspondents, the need of transm ting specimens for examination, the later consultations by visitin, mycologists, the ravages of insects and the accidents incident handling by attendants, is in a remarkably good state of pres tion. The packets would have been somewhat more secure, if. had been folded after the modern manner by overlapping the e more and folding the ends in the reverse direction from that of + sides. But as it is, there is little evidence that specimens have lost out, or intermixed to any harmful extent. To insure fu protection and facilitate examination in the future the senior at in February, 1917, after consultation with Dr. Shear? and Dr. © mer Stone, placed each packet still containing any material, in the seven large gray envelopes marked Ceoma and Puce whether American or foreign, into small manila envelopes wrote the name on the front. Of the 140 numbers in the N American list under the genera Ceoma (exclusive of the subger Albugo and Ustilago), Puccinia, Phragmidium, Podisoma and G4 nosporangium, 103 are represented at this date by specimens in { 7 Dr. Shear, of the Bureau of Plant Industry, Washington, D. C., the senior author are members of a committee from the American Phytopza _ ological Society to give whatever assistance may be possible in the prese: ; tion of the Schweinitz Herbarium. PAPERS GIVING RUSTS OF NORTH AMERICA, 185 collection at Philadelphia, either in the original autographic packets or mounted. Of the additional species of rusts, two under the genus Spheria and two under Seiridium, there are three represented _ by specimens. The careful and conscientious work of Schweinitz is further evident in the identification and naming of his material. This can be shown by examination of the species which Schweinitz consid- _ ered to be new, and to which he attached his initials. In the North Carolina list there are 45 such species under the genera 4cidium, _ Uredo (exclusive of the subgenera Albugo and Ustilago), Puccinia and Gymnosporangium, and of these only one was wholly misun- _ derstood, nine are still accepted under the full names given by _ Schweinitz, twenty-one still have the same specific name but are placed under other genera and fourteen only have the name wholly suppressed under synonymy. In the North American list there are 88 names followed by the initials of Schweinitz under the genera Caeoma (exclusive of the subgenera Albugo and Ustilago), Puc- cinia, Phragmidium, Gymnosporangium and Podisoma. . Only four of these species were misunderstood and erroneously placed, while twelve are still accepted as named, twenty-four still retain their specific ndmes under other genera, and forty-eight have the whole name relegated to synonymy. The discarding of over half of the new names found in the later work is largely due to Schweinitz’s replacement of earlier names by others conforming to Link’s new methods, as already explained, which made them untenable accord- ing to the present requirements of priority. The above showing is _ as good as can be found in most lists of rusts by recent mycologists, _ so rapid are the mutations in nomenclature of this group of fungi. ‘In general it shows that Schweinitz made comparatively few mis- takes in the identification of his material, and in naming tried very commendably to follow the most progressive and authoritative methods as then understood. At the present time the two or three dissimilar stages which many rusts exhibit are included under one ‘name, while formerly they were placed under separate genera. This in large part accounts for the 125 numbers in Schweinitz’s North American list, now known or believed to represent rusts, having shrunken to go species as at present classified. fa iin i ia Aan i 186 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ'S 2 The very large part of the material, which was the fout n¢ of Schweinitz’s two works, especially of the portions relating | to tl rusts, was secured by himself. He collected over a radius of thirty miles or so about Salem, North Carolina, and probably over even wider radius about Bethlehem, Pennsylvania, the two localities America where he resided. A very few collections were made upc his trips to more distant points, and some specimens were to him by his correspondents, especially by Torrey and Halsey New York, and Collins of Philadelphia, while a few were hand him by friends whose names appear at times upon the pz particularly Detwiler and Denke. The earliest biographical account of Schweinitz is that by 1 Wa R. Johnson, read before the Philadelphia Academy of Science May 12, 1835, a little more than a year after his death. It has the source of information for many later sketches, notably those Morgan,® Kellerman,® Shear,’° Harshberger,"t and Lloyd.%? — writers have added various facts, obtained from Schweinitz’s scendants, especially Gore,* Youmans,'* Lehman,"* and Shear Stevens.?® ie The three articles by Shear & Stevens were the result of tended researches regarding the history of Schweinitz’s collect of fungi, his methods of work, and the present disposition of specimens. Manuscript copies of the last two papers, as well one on Ezra Michener (Bull. Torrey Botanical Club, 44:54 Dec., 1917) by the same authors, were generously loaned t writers while this article was in preparation. Most of the 1 of various kinds referred to by the several authors have also. at the disposal of the writers. They have also consulted the 1 : 8 Bot. Gaz., 9: 17-19, 1884. 9 Jour. Myc., 2: 31-34, 1886. 10 Plant World, 5: 45-47, 1902. 11“ The Botanists of Philadelphia,” 127-132, 1899. 12 Mycological Notes, No. 44, 1916. 13 Jour. Elisha Mitchell Sci. Soc., 3: 9-25, 1886. 14 Pop. Sci. Mo., 44: 833-840, 1804; and “ Pioneers of Science in ica,’ 167-175, 1896. 15 The Wachocia.Moravian, 13142: 4-6, 1904. 16U, S. Dept. Agric. Bull. no. 380: 1-82, Jan., 1917; Mycologia, 9 204, 333-344, July, Nov., 1917. PAPERS GIVING RUSTS OF NORTH AMERICA. 187 script works of Schweinitz and the letters (amounting to 237) from his correspondents deposited at the Philadelphia Academy of Nat- _ ural Sciences, the letters from Schweinitz to Torrey (35 in number) at the New York Botanical Garden, and the letters from corre- spondents in the possession of his grandson, Dr. Geo. de Schweinitz, _ of Philadelphia. Some of his biographers say that during the latter years of his life he used de in place of von in his name. It is quite certain that after his death his sons and their families used the French form of the name, as their descendants do at the present time. His cor- respondents addressed him variously. By German friends and _many others the address used was Herr von Schweinitz, or by a few of them Baron von Schweinitz, while a less number used de Schweinitz. His intimate American friends, Torrey and Darling- ton, both of English descent, invariably used von. All of the Schweinitz letters to Torrey at the N. Y. Bot. Garden are signed Lewis D. v. Schweinitz; they extend from June 24, 1820, to May 2, 1832. His published writings bear this form of his name on their title pages, except when made to conform to the Latin. The initials invariably used on his packets of fungi and other collections were LvS. When used in print to indicate authorship they were written Lw.S. In the North Carolina list the abbreviation was Sw. There were doubtless reasons why he might have favored a change in the family name, either out of consideration for his wife, who was of French ancestry, or because of his dislike to Prussia, which at the Congress of Vienna in 1815 had acquired a third of Saxony, including that part where the ancestral home was situated and where his youth had been passed. But it is quite probable that e himself did not adopt the new form. The botanical work of Schweinitz was made the avocation of a busy life largely devoted to religious duties and churchly service. He ‘was imbued, nevertheless, with the most thoroughly scientific spirit. His monographic work upon the very difficult genera, Carex, Viola, __and Spheria, was of the highest order. He eschewed the easy as- , sumptions too rife in his day, and believed that a scrutiny of facts outweighed all plausibilities. What may be designated as his scien- tific creed is given in the preface to the Conspectus by Albertini & 188 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S- Schweinitz, which was doubtless written by Schweinitz. It ref especially to the study of fungi, and as translated Bd os (Memoir, p. 25) reads: “A solid basis to this department of botanical science must be laid, : on a sandy foundation, on the varying freaks and fancies of the mind, bu on a perpetual daily and nightly employment of microscopic observation, diligent and oft-repeated examination of the whole history of the fur tribes, a careful perusal of authors, a comparison of their respective nyms, and above all, by the observation of living naturé herself, as unfolds her rich abundance in the recesses of forests, lawns and agin observation which must be continued from day to day, and from year.” The following account includes a translation of that potions Schweinitz’s two works pertaining to rusts, given in the order the later one, together with a record of material still remaining represent them, and with comments by the writers. It has been pared with a view of making this monumental work more avail to students, especially students of American mycology. Following the main body of the work all of the species 0 mentioned by Schweinitz are arranged in systematic order cordance with present ideas of classification. The accepted ni used for the hosts are generally those of Britton & Brown’s “ I trated Flora,” 2d edition, or of Small’s “ Flora of the South United States,” 2d edition. . A serial list is then given of all the numbers in Schwe ir “North American Fungi” with which this account deals, with corresponding numbers from the North Carolina list in paren _and in a-parallel column the name or fact which the study o material has disclosed. An index of hosts and another of fur appended for convenience of reference. The microscopic and bibliographical work carried on in ¢ tion with this study of the Schweinitz material pertaining to 1 during the eighteen years since the work has been in progress been done in large part at Lafayette, Ind., in the laboratories agricultural experiment station of Purdue University. More a dozen of those associated in the laboratory work during thi period have taken part in the study, and to them, and to a n of correspondents credit is accorded for material aid. To PAPERS GIVING RUSTS OF NORTH AMERICA. 189 horities of the Philadelphia Academy of Sciences the gratitude the authors and of every scientific person interested in this sub- ject is due in unstinted measure. Under Mr. Stewardson Brown, _ Curator of the herbarium, and Dr. Witmer Stone during Mr. Brown’s absence, every facility that the Academy could offer has been placed freely at the disposal of the authors. Rusts oF NortH AMERICA RECORDED BY SCHWEINITZ. The arrangement is that in Schweinitz’s Synopsis Fungorum in America Boreali. Additions to the translation of the original text are in square brackets. The general serial number is followed by the species number under each genus. As stated by Schweinitz on page 144 of his work “species preceded by an asterisk are those not recorded in the ‘Synopsis Fungorum Caroline Superioris.’ Species with L.v.S. added were fest described by me either in my previous work or in the present one.’ After the complete record for each number the corresponding record in his “ Synopsis Fungorum Caroline Superioris,” if there one, is given in parentheses. Following the English version of Schweinitz’s words is a state- ment of the material to represent the number as it occurs in the weinitz Herbarium at the Philadelphia Academy of Sciences at the present time, the data on the packets being copied exactly as to Spelling, capitals, punctuation, etc. Finally come comments by the uthors. 1474. 329. S[phzria] epiphylla, L.v.S., Syn. Car. 130, F. 258, not in Penn- sylvania. (130. [Sphzria] epiphylla Sz. S. cespitose, blackish brown, shining, the pulverulent receptacle yellowish, spherules without ostioles, obovate, very minute, crowded, arranged cespitosely or fasciculately. It grows in an unusual place, namely, upon still growing leaves of Galega virginica. Scattered, on the upper surface of the leaves, punctiform, or oblong or linear, less than a line in diameter. Recep- tacle arising from the altered substance of the leaf, pulverulent, yel- lowish or brownish. Spherules globose, minute, obovate. At a younger stage subpellucid.) Represented by two leaflets mounted, each about 2.5 cm. long, ‘PROC, AMER. PHIL. SOC., VOL. LVII, N, JULY 16, 1918. 190 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S | and by the original packet, empty, labelled on the front “ Sphi epiphylla LvS ... Salem .. .,” a portion not being legible, on the folded sa “ Spheria epiphylla Salem.” While Dietel was making a study of the genus Ravenelia, received a fragment of the original Schweinitz collection, sent Lagerheim from the Herb. Fries, from which he was enabled transfer the species to that genus (Hedwigia, 33:27. 18 although he points out that the true nature of the fungus already been detected (Farlow & Seymour, “Host Index,” 30. 1888). The name is now generally written Ravenelia é phylla (Schw.) Diet. *1487. 342. S[pheria] canaliculata, L.v.S., of the same group [as the pi ceding species, 1486], but abundantly distinct, Bethlehem, leaves of the involucres of Cyperus, found on the dorsal s S. covered, dark, composed of series of perithecia situated b the striz of the leaves, parallelly confluent on a pitch black so that the spot-appears beautifully canaliculate; rather Ostioles thick, punctiform. On the margin occur subs: subrotund, applanate perithecia. In the middle, moreo pitch black spots are sometimes sterile—and, it may be no spot is frequently interrupted at intervals of a quarter of so that the unaltered substance of the leaf comes into view. Represented by a mounted specimen, consisting of a po ‘ion five leaves, originally six, one having become detached and each portion about 5 cm. long and 6 or 8 mm. broad, well st with uncovered uredinia and covered telia. The original contains two small pieces of leaf, and is labelled “ Sph. ca LvS in Scirpi involucr.” It was evidently first labell graminis,” as the word “graminis” has been crossed out. The true character of this fungus was first pointed out by | heim (Tromsé Mus. Aarsh., 17:51. 1895), from the study original autographic specimen in the Fries Herbarium. It i called Puccinia canaliculata (Schw.) Lagerh., and is a wide American species. | *Species preceded by an asterisk are those not recorded in the “ Fungorum Caroline Superioris.” PAPERS GIVING RUSTS OF NORTH AMERICA. 191 V. GYMNOMYCETES (Envtopnyt and TuBERcULaRINI Fries). Serres I. ENTOPHYTZ#. Genus 211. COMA. a Subgenus UREDo. 1. Ustilago. _Note—tThe six species under this heading nos. 2811 to 2816 are belonging to the Ustilaginales, and are therefore omitted. : 2. Rubigines (Orange-yellows). 7. 7. C. U. Rubigo, Lk. n. 9. Halsey from New York, on cereals. RT LEFT NLL ee v The packet is labelled “ Uredo tecta Halsey,” and again “Czoma rubigo Newyork Halsey.” Both leayes appear to be those of wheat (Triticum vulgare Vill.), nd are well covered with large, scattered, oblong uredinial sori. _ The name was correctly applied by Schweinitz in the sense in h it was first employed by De Candolle and others of the times. ers a number of species, however, and the one represented by collection is Puccinia graminis Pers., in its uredinial stage, now y called P. poculiformis (Jacq.) Wettst. 8. C. U. linearis, Lk. n. 8, Syn. Car. 464, on leaves of cereals, Salem, Bethlehem, and everywhere. (464. 6. [Uredo] linearis. Fairly common on grain.) epresented by portions of four narrowly linear leaves, each 8 to 10 cm. long, loose in a mounted packet, bearing a few ttered uredinial sori. The original packet is labelled inside edo linearis Sal,” and outside “Czoma (Ured) lineare Salem.” The compound microscope easily shows the rust to be the edinial stage of Puccinia Poarum Niessl now more often referred epiphylla (L.) Wettst. It is characterized by peculiar capi- 192 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S © tate paraphyses. The host is the common Kentucky blue -Poa pratensis L. It is a species not found on other grasses or grains, although uredinia of similar gross appearance are found on © both, and were all given the same name by older mycologists. Pro = ably the original portion of the material on cereals was removed | Schweinitz, —- only the part on meadow grass. *2819. 9. C. U. rimosum, Lk. n. 14, rather rare on Scirpus near — Jersey. ae Represented by one 5 cm. mounted piece of a terete cttleny five similar pieces, 3 to 5 cm. long, in the original packet, which labelled “Czeoma (Ured) rimosum in Scirp acut. spec. imper ob bonas pertus. Hope Jersey.” The host is undoubtedly Sci; lacustris L. (S. acutus Muhl.), the plant that Schweinitz took i to be. The smooth surfaces of the culms show a few quite | rifts, 5-15 mm. long, but no spores or fungus of any kind. 1 rifts may have been interpreted by Schweinitz to be the “ acery rimis longitudinalibus parallelis positis” of Link’s description he has entered on his packet that he had an “ imperfect specimen account of marked perforations.” Link’s Ce@oma rimosu: 3 ' however, founded upon a fungus on Juncus acutus from and could not have been the same as an American fungus on S Lagerheim in his study of the rusts in the Herb. Fries (1. c. 67) has erroneously added “ Uredo rimosa Schwein.” as a sy of Puccinia obtecta Peck, a rust that occurs on both Scirpus p (the host in the Herb. Fries from New York), having stems, and S. lacustris, having terete stems. Had this rust present Schweinitz would probably not have referred it to species, because of the slight resemblance which it bears to description. - *2820. 10, C. U. Andropogi, L.v.S., on leaves of Andropogon ave Bethlehem; rare and related to C. longissimum, from differs particularly by an evident purple spot. C. spots much elongated, narrow, purple. Sori much el parallel, narrowed, longitudinally erumpent from the raised dermis. Spores at last loosely scattered, globose, rufo-fu PAPERS GIVING RUSTS OF NORTH AMERICA. 193 Represented by parts of two leaves, about 5 cm. long, and of wo others, 7 cm. long, all 5 to 8 mm. wide, mounted, and in the original packet five similar pieces with some fragments, all bearing an abundance of brown uredinia and a few telia. The packet is labelled inside “Czoma (Ured) Rubigo Lk in Androp. avenacei fol Beth 1820,” and outside “Czoma (Ured) Andropogi LvS.” The host is evidently Andropogon avenaceum Michx., as stated, ow often referred to Sorghastrum nutans (L.) Nash, and the rust proves to be Puccinia virgata Ellis & Ev., a species not at all related to P. Andropogi Schw., no. 2911. . i. C. U. Iridis, L.v.S., frequent on withered leaves of Iris virginica, Bethlehem. i C. related to C. Lilii; spots yellowish, sori roundish oval, not circinate but scattered; at first covered with the epidermis, rather elevated. Spores numerous, somewhat pedicelled, fulvo-fer- rugineous, at length scattered. Spores never turn black as in C. Lilli. Represented by two well-preserved pieces of leaves mounted, one being 1 by 6 cm., and the other 1.5 by 7 cm., and two pieces much : eaten by insects, in the original packet, and all well covered with edinia. .°The packet is labelled “ Puccinia Iridis LvS Beth,” with word Puccipia crossed out and “Czoma (Ur)” substituted. There is an empty duplicate packet labelled in a similar way. _ The rust is a common one of both hemispheres for which the ac- cepted name is Puccinia Iridis (DC.) Wallr. In America, east of the Rocky Mountains, only uredinia have been found. Although the host is called Iris virginica, a linear-leaved species, both because these leaves are especially wide, and because no rust is known on species, the host must be J. versicolor L. 12. C. U. Smilacis, Lv.S., Syn. Car. 471, Link n. 22, and Bethlehem on leaves of Smilax. (471. 13. [Uredo] Smilacis Sz. U. peridia variably flexuose, minute, grouped, often concentric, dark brown, the spore-mass luteo-fuscous. Frequent, on leaves of Smilax rotundifolia, seated on yellowish spots.) Represented by one piece of leaf 3 by 4 cm., cut from a leaf of ‘obably twice the size, and mounted. It is thickly covered with 194 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S uredinia. The empty, original packet is labelled inside € Smilacis S, rotundifol Sal,” and outside “ Caoma C<— milac LvS in S. rotundifol Salem.” oe 2916, very common in the southeastern states on various ; pe Smilax. 2823. 13. C. U. Labiatarum, Lk. n. 34, Syn. Car. [as] U. Cinna Bethlehem on species of Pycnanthemum. (469. 11. [Uredo] Clinopodii Sz. U. orbicular, somewhat inflated, yellowish. oe Frequent in autumn on the leaves of Clinepodiie incanum, Related to U. Menthz.) Represented only by an empty packet, which is labelled “Uredo Clinopodii In Pycnanth. Salem,” and outside “ (Ur) Pycnanthemi LvS C. clinopodii Salem.” Without Schweinitz had the uredinia of Puccinia Menthe Pers., on Koei incana (L.) Kuntze, of which the preceding names are syno He accepted Link’s disposition of his new species as a §) under Link’s name for all the common mint uredinia. 2824. 14. C. U. Ipomee, [L.v.S.,] Syn. Car. 468, Lk. n. 38, not Penna (468. 10. [Uredo] Ipomcee Sz. U. rather small, sparse, not confluent, bright red. Frequent on the lower surface [of leaves] of Ipomeea 1 trile Related to U. Tussilaginis. ) Represented by three cordate leaves, 3 cm. long, moun : covered beneath with uredinia and telia, and two smaller leaves tached to a slender stem, in the original packet, bearing a f The packet is labelled inside “ Uredo Convolvuli Salem,” afte: “Tpomeee ” written above Convolvuli, and outside “ Czoma ( Ipomee LVS in Ip. pandur. Salem.” The rust is an excellent example of Coleosporium Ip (Schw.) Burr., showing uredinia and telia, and the host is doubtles I. pandurata L., which was at first confused by Schweinitz with tt more southern species, J. triloba. Although Schweinitz incide: tal omitted his initials as author of the specific name in accor with his custom in other similar instances, L. v. S. should be ac ' PAPERS GIVING RUSTS OF NORTH AMERICA. 195 while the combination with Ceoma was first made by Link, it based entirely on Schweinitz’s account in his Carolina list. . 15. C. U. Elephantopodis, L.v.S., Syn. Car. 467, Lk. [n.] 54, only in Carolina. (467. 9. [Uredo] Elephantopodis Sz. U. rather large, sori depressed, sparse, circular, bright yellow. On leaves and stems of Elephantopus tomentosus, very frequent in the autumn. Related to U. farinosa. Older sori leave Peziza- like hollows in the leaf.) cket i is labelled inside “ Uredo Rleshantopodis Salem,” an out- _ side “Czoma (Ur) Elephantopodis LvS Salem.” _ The rust is now called Coleosporium Elephantopodis (Schw.) iim. As indicated for the preceding number Schweinitz adds his name to the Czoma combination as author of the species although le combination was first made by Link. This was in accord with the opinion then held that the author’s name was attached to the species as a voucher for the concept as expressed by the original description, and not for the technical formation of the name as ap- plied toa particular specimen, according to present usage. 2826. 16. C. U. Solidaginis, L.y.S., Syn. Car. 472, common, and Pennsylvania. (472. 14. [Uredo] Solidaginis Sz. U. compact, closed, red, linear, sometimes long. Very frequent, almost all large Asters. Solidagos, Vernonias; related to U. pustulata.) Represented by four original packets, and mounted material from — of them. Two smooth lanceolate leaves, probably of Solidago rotina Ait., showing purple discolorations, are mounted, evidently taken from the empty packet marked “1 Coma (Ured) Solidagi- um LvS in maculis purp.” A duplicate packet, also empty, is labelled “2 Czoma (Ured) Solidagini LvS.” The other mount mnsists of about two thirds of a smooth, lanceolate leaf with entire margin, probably of Solidago sempervirens. It was doubtless taken rom the empty packet labelled inside “Uredo (Ecidium) ovale Nyk Halsey,” and outside “Czeoma ovale Halsey Nyk,” with the 196 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S word “ovale” crossed out and “ Solidaginis” substituted. — was a correspondent living in New York. The fourth orig | packet is labelled inside “Uredo Solidaginis in Vernonia nove- boracens Beth,” and outside “Caoma (Ur) Solidaginum LvS Salem & Beth.” The packet contains the larger part of f lanceolate leaves, each fragment about 18 mm. wide and 7 cm. | Three of these leaves are yellowish and are doubtless Solid altissima, and may have been obtained at Salem, the fourth greenish with sparse, colorless hairs, and is doubtless S. rugosa, at may have been obtained at Bethlehem. The inclusion of Verne may after a time have been considered erroneous, and the lea removed. Ty ae All the seven leaves representing this number show uredinia Coleosporium Solidaginis (Schw.) Thiim., one of the commor of rusts in the eastern states. The unusual abundance of ma preserved to illustrate this number was doubtless due to its b encountered frequently in the fields on many hosts. | 2827. 17. C. U. Terebinthinacee, L.v.S., Syn. Car. 473, not in 2 Pennsy (473. 15. [Uredo] Terebinthinacee Sz. U. aggregated, almost solid, pustulate, closed, bictoialtn durated, orange red, rather large. Frequent on the lower surface the very thick leaves of Silphium terebinthinaceum. ‘Related populina. N. B. They [i. e. the Rubigos] occur on almost all autu: plants of the class Syngenesis, as on Helianthus, Aster, So etc., etc. As to the Rubigos, which ones constitute distinct : spe it is most difficult to decide.) , Represented only by an empty packet, labelled inside “ terebinthinacee in Silph terebint Salem,” and outside “ (Ur) Silphii terebinthinaci LvS. Salem.” The rust is undoubtedly Coleosporium Terebinthinacee (S Arth., and the host Silphium terebinthinaceum Jacq. Schweinitz’s observation that it is difficult to decide upon systematic distinctions among orange-yellow uredinia remains lat true at the present day. ¥2828. 18, C. U. Helianthi Ly.S., rather rare on leaves of H. gigan Bethlehem. PAPERS GIVING RUSTS OF NORTH AMERICA. 197 C. spots obscure. Sori clustered, naked, pulvinate, flavo-rubrous, at first rather solid, finally sprinkled with the minute orange red spores. ‘ _ Represented by parts of two small, lanceolate leaves. The smaller one, about 4 cm. long, is mounted, and is doubtless Helian- thus giganteus L. The other, about 7 cm. long, is half in the orig- ‘inal packet, which is labelled “Czoma (Ured) Helianthi LvS in ‘Helianth gigant. Bet,” and half mounted. It is possibly H. strumo- sus L. A similar leaf, 4 cm. long by 1 cm. broad, and evidently part of the latter collection, is in the Michener Collection at Wash- ington, now belonging to the U. S. Department of Agriculture. _ The leaves all show many telia and a few uredinia, of what is now called Coleosporium Helianthi (Schw.) Arth. It is not an abundant species, but is widespread. *2829. 19. C. U. Anemonis, L.v.S., on under surface of [leaves of] Anemone quinquefolia, Bethlehem, rare. C. spots yellowish, rather large, sori roundish, dilated, slightly ele- vated, spores pale. _ Represented by a compound trifoliate leaf about 4 cm. broad d long, thounted, having plenty of pale round uredinial sori be- veath. The original packet is labelled inside “Uredo anemones,” and in another place “Czoma Anemonis quinquefolie Bethl,” while side it reads “Czeoma (Ur) Anemonis quinquefo LvS Detwyler thi H.” As no such rust has been collected since on the host named, there s been much speculation regarding its identity. Not until the “ior author’s recent visit to examine the Schweinitz material at the Philadelphia Academy did the solution of the enigma become ident. It was then noticed that this so-called Anemone leaf is sparsely sprinkled with long colorless hairs, which remind one of those on Osmorrhiza. Comparing this leaf with material for no. I and no. 2851, which had previously been determined as Os- ‘morrhiza, left no doubt that all were the same host. On this host s Puccinia Pimpinelle (Str.) Mart. (P. Osmorrhize C. & P.), th the uredinia of which this material exactly agrees. This instance illustrates the danger in collecting too small speci- 198 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S — mens, mere fragments. The large decompound leaves of the | growing Osmorrhiza could not be mistaken in the field for the | wind-flower, but the trifoliate tip of one of the large leaves whe isolated might well be supposed to be the whole leaf of a — D 2830. 20. C. U. Campanularum, Lk. 44, on C. amplexicaulis, Sym. Coe. and Bethlehem. (465. 7. [Uredo] Campanule. Rarely occurs on Campanula per- foliata. ) No specimen or packet is in the collection to represent number. The host is one on which there is no other record rust, although a species of Coleosporium does occur on the related genus Campanula as now understood. It is highly pro however, that Schweinitz had some fungus not a rust. The p’ nt now known as Specularia perfoliata (L.) A. DC. — foliata L., C. amplexicaulis Michx.). 2831. 21. C. U. Onagrarum, Lk. 32, Syn. Car, Circaez, 466, and Bethlehe (466. 8 [Uredo] Circee. Here and there on the lea) Circea Canadensis. ) There is no material or packet at Philadelphia to represen number, which is unfortunate, as no common rust exactly an the requirements of the record. The names employed for the are of a European species, not known in America. Uredo Cir was established by Albertini and Schweinitz in their work Lusatian fungi for the uredinia of what is now called Pueciniast Circee (Schum.) Schrét. The only rust on Circea in this is Puccinia Circee Pers., which is so very unlike the one jt ferred to that it seemingly could not have been mistaken foi Although P. Circee possesses no uredinia, yet the young telial are pale and in gross appearance might be so considered. record in both publications appears to parallel the correspa records of P. Circee under no. 2938, and the most reasonable pretation appears to be that Schweinitz mistook the young of P. Circee Pers. for a Uredo. PAPERS GIVING RUSTS OF NORTH AMERICA. 199 22. C. U. miniata, Lk. 84, Syn. Car. 463, Salem and Bethlehem. (463. 5. [Uredo] miniata. Frequent but only on Rosa pauciflora.) _ Represented by a mounted rose leaf, 7 cm. long, consisting of five leaflets, and the original packet containing one smaller com- d leaf and a number of leaflets, all similar. There are large, irregular sori on rachis and midribs and annular, pustulate sori on the blades, all ecia. The packet is labelled outside “Czoma (Ur) miniata Salem,” and added later “& Bethl & Herrnht.” MHerrnhut the place where Schweinitz studied in Saxony. The material apparently is that gathered at Salem, N. C., and the addition of two other localities to the packet indicated the collector’s ield observations, and not his actual addition to the collection. The 10st name of Rosa pauciflora is given in Muhlenberg’s “ Catalogue ” as synonymous with R. carolina L., the name now in use, which is doubtless the species Schweinitz found the rust on. The rust proves to be the aecia of Earlea speciosa (Fries) Arth., formerly called hragmidium speciosum Cooke. Telia of this species were placed Schweinitz under the genus Seiridium, no. 3084. The species is not known in Europe, and the selection of Persoon’s name, Uredo iniata, has proven unfit, although at the time the two forms could not well have been separated. The transfer of the species to the genus Ceoma was first done by Schweinitz, not by Link. *2833. 23. C. U. ruborum, Lk. 86, frequent, Bethlehem. Represented by no mounted specimen, but by some ten leaflets n the original packet, which is labelled inside “Czoma ruborum, Uredo (Rubigo) Rubi In Rub id horti mei fr Oct. 1824,” and out- = “Czoma (Ur) Rubi Idei Bethl in hort.” The largest of the aflets is about 6 by 7 cm., and all are pale tomentose beneath, with aced by the similar native form, R. strigosus Michx. The rust is the uredinial stage of Kuehneola Uredinis (Link) Arth., a common cies on various raspberries and blackberries, but whose affinities = only been recognized within the last few years. The telial 200 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S stage is white, and the name, Phragmidium albidum, is often a plied. Link’s name of Ceoma ruborum belongs to another rust. — 2834. 24. C. U. Potentillarum, Lk. 87, Syn. Car. 461, frequent on Potent canadensis, [and in] Pennsylvania. (461. 3. [Uredo] Alchemille. I am certain it is the same as ti on Alchemilla. Here and there on the leaves of Potentilla ¢: densis, living through the winter.) Represented by a mounted packet containing loosely a bit stem and five leaves of the host mentioned. Three of the small leaves show primary uredinia above, and two larger leaves sh« secondary uredinia beneath, the sori being numerous. An p' original packet is labelled “Czoma (Ur) Potentillz canadens Li Sal & Beth.” —e It was quite natural for Schweinitz to think this one a f of Uredo Alchemille, both from the gross appearance of the i inclusive name, C. Poteanilaren. The rust is now known to wholly different, and is called Frommea obtusa (Strauss) Arth., more commonly, Phragmidium Potentille-canadensis Diet., _ Kuehneola obtusa (Str.) Arth. 2835. 25. C. U. Agrimoniz L.v.S., usually wholly covering the lower su of Agrimonia, wrongly [referred] to U. Rose, Syn. Car. 462. C. spots becoming yellowish. Sori minute, confluent, spores b fully reddish orange, finally losing their color. : (462. 4. [Uredo] Rose. I do not doubt that it is the s occurs very frequently on Agrimonia Eupatoria in a never on roses with us.) Represented by three terminal leaflets, mounted, each nearl cm. long, and by fragments of three compound leaves in the packet, which is labelled “Caoma (Ur) Agrimonie LvS All of the leaflets are abundantly covered with sori. The rust is the characteristic uredinial stage of Puccinias Agrimonie (Schw.) Tranz., which occurs in Europe and Asia, not so common there as in America. The host appears to be 4 monia parviflora Soland. PAPERS GIVING RUSTS OF NORTH AMERICA. 201 896. 26. C. U. Filicum, Lk. n. 101, on Aspidium, from New York, com- municated by Dr. Torrey. Represented by about 4 cm. of the terminal part of a frond, mounted, and by parts of one or more fronds of uniform appear- ance in the original packet, which is labelled “Coma Filicum orrey Nyk in Asp. obtus,” and in addition “U. polymorph in Asp. dryopt.,” with a number of German localities and names of German collectors. Probably the additions to the inscription on the packet do not indicate collections, but only memoranda. : _ The rust occurs in rather large, covered, blistery sori, on the under surface of the fronds, and is the uredinial stage of Hyalopsora Aspidiotus (Peck) Magn. The host is evidently Phegopteris Dryopteris (L.) Fée, the Aspidium obtusum of Muhlenberg’s “Catalogue,” and the collection was probably made in the Catskill mountains, as Dr. Torrey lived for a time at West Point, N. Y. The rust is not known outside of North America. It is a moun- _ tainous form, the type collection being found by Peck in the Catskill *2837. 27. °C. U. Teucrii, L.v.S., very rare on leaves of Teucrium virginicum, Bethlehem. C. spots obsolete. Sori densely crowded into semblance of a spot, effused, beautifully red. Spores very small, very red, almost scarlet. Represented by one leaf, oblong, 3.5 by 7 cm., mounted, and by _ the empty packet, labelled inside “ Uredo Teucrii in fol Teucrii canadens. Salem,” and outside “ Cazoma (Ured) Teucrii LvS. Naz.” The leaf shows a number of rusty-looking spots, still finely purplish ted, which the microscope reveals to be due to a Hyphomycetous fungus, having small oblong to linear-oblong spores, and in nowise related to the rusts, of which there are none known on Teucrium 1 America. __ This material has been examined by Dr. C. L. Shear, who states that it is identical with Cercospora racemosa E. & M., a species founded upon a collection made by the senior author in Tawke Sep- tember 27, 1882. It is a somewhat common fungus extending from : Atlantic coast to Kansas and Nebraska. The name should be- 202 ARTHUR-BISBY—TRANSLATION OF ScoWEnTeaS ng come, in accordance with the rules of priority, Cercospora Tev (Schw.) comb. nov. . 2838. 28. C. U. Azalex, L.v.S., Syn. Car., 470, [as U.] minima, frequent leaves of Azalea nudiflora, Bethlehem and Salem. C. spots obsolete. Sori on the lower surface, at first solneatea cone : shaped, minute, orange, finally effused. Spores very min losing their color, and unequal, pyriform, with mestis 6 fe ‘ intermixed. 2 (470. 12. [Uredo] minima Sz. U. very minute, punctiform, pale orange, sparse, subconic. Frequent on the lower surface of the leaves of Azalea : flora.) Represented by a mounted leaf 2 by 6 cm., thickly covered the lower surface with uredinia corresponding to the deserig 0 and by an empty packet labelled inside “Uredo farinosa B minir in Azalea nudif Salem,” together with the later name “Cz minimum,” written above, and on the outside “ Cazoma (NOG Az LvS. Beth & Sal.” ' The rust is the uredinial stage of Pucciniastrum (Schw.) Arth., as reported in the “ North American Flora” 7: 1907, a name now believed to be synonymous with P. My -(Schum.) Arth., a rust occurring upon various species of | cinium, as well as on Azalea nudiflora L., and other Ericac hosts. 3. Fuscentes and Nigredines (Browns and Blacks). 2839. 29. C. U. Ari virginici, L.v.S., Syn. Car., [as U.] Caladii, 480, Lk, 21. It is not Caladium but Arum on which this is frequer found and in Pennsylvania. (480. 22. [Uredo] Caladii Sz. U. punctiform, solitary, seated on large yellowish spots, | spore-mass fuscous. Frequent on the under side of the leaves of Caladium. at first closed, at length scattering the spores.) Represented by a 3 cm. square portion, cut from a large lea mounted, showing uredinia scattered over the surface, and by empty packet labelled inside “ Uredo Caladii Salem,” and outsid “Ceoma Ari virginici LvS. n. Caladii Salem.” | PAPERS GIVING RUSTS OF NORTH AMERICA. 203 The rust is the uredinial stage of Uromyces Caladii (Schw.) Farl., the xcial stage being given under nos. 2860 and 2861, and the telial stage under no. 2946. Doubtless Schweinitz was right in ‘thinking the host to be Arum virginicum L., now known as Pel- tandra virginica (L.) Kunth, and not Caladium [sagittifolium utt.], although the fact can not now be verified. Both hosts occur in North Carolina, but only the former in Pennsylvania. 2840. 30. C. U. Spermacoces L.v.S., Syn. Car., [under] Puccinia, 502, Lk. n. 57, elegant. Spores not septate, and Philadelphia. (502. 17. [Puccinia] Spermacoces Sz. P. subquadrate, dark chestnut-brown, spores globose, simple, pedicel very long, filiform. Frequent on leaves and stems of Spermacoce. Breaks through the epidermis in the form of a square. Spores fuscous, irregularly globose, pointed or blunt, without septum. Pedicel ten times longer, hyaline. By pressure the epidermis is separated from the square mass as a continuous membrane in which a cellular structure is not to be seen under lenses having a focus of half a line, and a very thin vesicular substance escapes.) Represented by two small fragments of stem with leaves and uit, placed loose in a mounted packet. The original empty packet s labelled inside “ Diceoma Spermacocis Salem,” and on the out- side “ Czoma Spermacocis IvS. Sal.” The rust is chiefly the telial stage of Uromyces Spermacoces Schw.) M. A. Curt., common throughout the southern states, and the host is undoubtedly Diodia teres Walt. (Spermacoce diodina Michx.). It is interesting to trace the change in view, in the interim be- tween the publication of the two papers, regarding the systematic position of forms with dark teliospores, which we would now cal! _Uromyces. In the North Carolina paper of 1822 Schweinitz di- vided the genus Puccinia into “A, spores distinctly bilocular,” and B, spores globose with septum inconspicuous,” evidently following the example of DeCandolle in the Flore Francaise (2:224) of 1805. Under the latter division Schweinitz placed two species of Uro- Wees, with the septum described as absent or not conspicuous, re- spectively. Evidently there was a feeling that these forms with an certain septum and globoid spore belonged with those species of 204 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Puccinia having elongated spores and an evident septum. Late the idea of a possible septum was abandoned, and it was necessa to place these dark, globoid, non-septate forms under the all clusive genus Uredo, in spite of their apparent relationship to Puc- cinia. Still later systematists placed them in the genus Uromyce. but recently the opinion has been growing that the earlier met! of DeCandolle and Schweinitz better indicates their true relati ship. The mention of the kind of lens used in these studies helps to explain why the question of the presence of a septum should ha remained uncertain. Even without knowing the degree of defini- tion, doubtless far less than that of modern hand lenses, it is clea that the magnification left much to be desired. 3 *2841. 31. C. U. Cherophylli, L.v.S., on leaves of Cherophyllum or Myr Claytoni, Pennsylvania. C. spots obsolete: sori rounded, sparse and aggregated, even sa what confluent, finally uncovered by rupturing the epider Spores effused, globose, from tobacco-like to black, shining. a Represented by a compound leaf of three leaflets, each abomt. cm. long, mounted, showing uredinia and telia, and an empty pz labelled inside “ Uredo cherophylli, N. Beth Detwyler,” ana r side “Caeoma (Ur) cherophylli LvS prope Beth Detwyler.” The rust proves to be Puccinia Pimpinelle (Str.) Mart. Osmorrhize C. & P.), and the host to be Osmorrhiza, in all pr ability O. Claytoni (Michx.) Clarke (Myrrhis Claytoni Michx suggested by Schweinitz. The material is essentially — that of nos. 2829 and 2851. *2842. 32. C. U. Hyperici, L.v.S., on stems of an unidentified Hype: rare in Carolina; not the same with C. hypericorum, Lk. C. spots on the pilose-strigose stem, purple: sori sparse, acum ovate, bullate, elevated, surrounded by the Bi epid Spores fuscous purple, becoming effused. Represented by a much branched stem, without leaves, but eleven seed pods, mounted, having uredinia sparingly distri over the stem, and by an empty packet labelled “Czoma Hyperici LvS. Salem.” PAPERS GIVING RUSTS OF NORTH AMERICA. 205 The rust is the uredinial stage of Uromyces Hyperici-frondosi aw.) Arth., and the host is some species of Hypericum, not yet entified, but which doubtless can be. Schweinitz was right in hinking his material quite different from Ceoma hypericorum Link, 2843. 33. C. U. Heucherz, L.v.S., Lk. 79, Syn. Car. 479, not in Pennsylvania. 3 (479. 21. .[Uredo] Heuchere Sz. - U. seated on orbicular, yellowish spots, peridia subconcentric, crowded, dark chestnut brown, spore mass dark fuscous. Here and there on the leaves of Heuchera Americana and vil- losa. Peridia at first closed; at length scattering the spores, minute. Related to U. Anemones.) Represented by part, about 4 by 5 cm., of a large leaf, mounted, having small, hypophyllous, pulvinate, brown sori, and by an empty = packet labelled inside “Uredo Tiarella Heuchere Salem,” and on ide “ Czoma (Ur) Heuchere LvS. Salem.” The spores are oblong, two-celled, and smooth whether examined vet or dry. The rust is now called Puccinia Heuchere (Schw.) etel. The mounted leaf appears to bé that of Heuchera amer- _ icana L., but the rust is known to occur on many species, and may ell have been seen by Schweinitz on H. villosa Michx. The systematic position of the species must have been deter- mined by Schweinitz from the gross appearance alone. This would unt for its inclusion in the subgenus Uredo, and for the omis- on of spore characters in the description. : B44. 34. C. U. apiculosum, Lk. [n. not] p. 90, on Phaseolus, Bethlehem, Syn: Car. 478. (478. 20. [Uredo] flosculosorum. Conspect. fung. On Kuhnia, Eupatorium, and other composites. (Czomurus Link.) ) : ' _No specimen or packet remains to represent this number. Two ‘typographical errors occur in the entry. The asterisk should be | omitted, and the reference to Link’s work should read n. go, and not “p. 90,” the reference beng to the number of the species and ot to the page. The name Uredo flosculosorum was established by Albertini and Weinitz (Consp. Fung. Nisk. 128) and they named as hosts _ PROC. AMER. PHIL. SOC., VOL. LVII, 0, JULY 17, 1918. 206 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Prenanthes, Leontodon and Hieracium, all Cichoriaceous composi while here Schweinitz has extended the use of the name to duaceous composites, and even legumes. Link at the place reduced this name to a synonym, together with twenty-two 0 under his inclusive species, C. apiculosum. The species has no val in the modern sense, being a concept supported only by sup 2 characters, and represented by an incongruous mixture of sp 2845. 35. C. U. appendiculosum, Lk. 91, Syn. Car. 477, and Bethe : (477. 19. [Uredo] appendiculata. Common on Phaseolus and ¢ Pisum sativum. (Czomurus.) ) A record in the North Carolina list that is not accounted fc the later one may be entered here, as it is the same rust, althou placed by Schweinitz under Puccinia and erroneously referred t name belonging to another species of rust. (490. 5. [Puccinia] Avicularie 68 Fabe. Not infrequent Phaseolus. ) Represented by a mounted packet loosely containing three I lets of the garden bean (Phaseolus vulgaris L.) and two leaflets garden pea, while the original packet labelled “Caoma (Ured) pendiculos Beth,” has one leaflet of bean and two of pea. The leaflets are well covered beneath with uredinia. The pea le are discolored with spots but have no rust; furthermore, no has ever been found in America on the garden pea, Pisum a Schweinitz mistook the spots for a common European rust, he naturally expected to find under the same conditions here as it Europe. The rust on the leaflets of Phaseolus, the common bean, Uromyces appendiculatus (Pers.) Fries. The European rust Pisum is a different species. The specimen preserved dow tle: represents no. 477 of the Carolina list, showing the uredinial sta; of the rust, while no. 490 of the same list is unrepresented by a lection, and as it was placed under Puccinia, doubtless had refe to the telial stage of the same rust. ) PAPERS GIVING RUSTS OF NORTH AMERICA. 207 “2846 36. C. U. punctuosum, Lk. 93, Syn. Car. 474, [as U.] scutellata, also ee Bethlehem on Euphorbia hypericifolia. (474. 16. [Uredo] scutellata. More or less frequent on Euphorbia hypericifolia.) Schweinitz had an entry in his Carolina list, which is nowhere _ referred to in the later one. It can be entered here, as it is the same rust, although he placed it under his section “ Rubigo.” (459. 1. [Uredo] Euphorbie. Not rare on leaves of Euphorbia maculata.) Represented by some four pieces of branched stem about 3 cm. long, with leaves, more or less fragmentary, inflorescence and mature seeds, showing a few, scattered uredinia, placed loose in a mounted _ packet, and by an original packet, containing a few similar frag- - ments, labelled “Czeoma (Ured) punctuos in Euphorb hypericif _ Beth.” Another original packet containing fragments of branched stems about 2 cm. long, with leaves and inflorescence, butenot mature seeds, was first labelled “Caoma (Ur) Euphorbie hypericif non scutellat Sal & Bet,” then the specific name was cancelled and “ punc- _ tosum” substituted. The latter packet doubtless represents the Salem collection and the former one the collection from Bethlehem. _ There is no material or packet for the collection on E. maculata. : The rust is Uromyces proéminens (DC.) Pass., showing vary- ing proportions of uredinia and telia. In the interim between his __two lists Schweinitz had ascertained that the European name used ‘in his earlier list, “U. scutellata,’ applied to another rust which he had not foundin America. The hosts are Chamesyce Preslii (Guss.) _Arth. (Euphorbia Preslii Guss., E. hypericifolia having recently been ascertained to be a more southern species) and Chamesyce maculata (L.) Small (Euphorbia maculata L.). 2847. 37. C. U. Leguminosarum, Lk. 92, Syn. Car. 476, [as U.] Vicia, on Vicia Faba, Bethlehem and Salem. (476. 18. [Uredo] Vicie. Fabe. On the stems of Vicia Faba. __ There is no mounted specimen or original packet to represent _____ these entries. 208 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S If a rust were really present, as there may have been, i Uromyces Fabe (Pers.) DeBary, which is occasionally found ot the English bean, V. Faba, in America, but is more common Of native species of Vicia and Lathyrus, cia 38. C. U. Lobelie cardinalis, L.v.S., rather rare on the under of leaves of Lobelia cardinalis, Nazareth. C. spots obsolete, sori effused-confluent, not elevated, or su: by the epidermis. Spores effused, pedicelled, co 1.5 cm. wide, mounted, and by an empty ‘paces labelled : inside “Uredo Lobeliz cardinal,” and on the outside 7 “¢ (Ured) Lobelia Cardinal Lvs. Beth.” Hyphomycetous fungus, Cercospora effusa (B. & C.) E. cE a ; #2840, 39. C. U. Thalictri, L.v.S., very rare but beautiful, on eas Thalictrum cornuti, Bethlehem. Be C. spots none. Sori pulvinate, roundish—a line or more in dia widely aggregated, somewhat surrounded by the epid Spores rather large, and from chocolate to fuscous. Represented by part of a leaf, 1.5 by 2 cm., mounted, and an empty packet labelled “ Ceoma (Ured) Thalictri LvS. Naz.” The leaf is thickly and evenly covered with round, brown bearing 2-celled, and a few 1-celled, teliospores of the characte form belonging to Polythelis Thalictri (Chev.) Arth. (P: Thalictri Chev.), on Thalictrum polygamum Muhl. (T. 6 Auct.). *2850. 4o. C. U. brunneum, L.v.S., on leaves of an unknown plant from collection of Mr. Collins, Philadelphia. C. spots yellowish, on the upper surface of the leaf. Sori app! irregular in form, variously confluent. Spores minute, bre fuscous, at first conglutinate. Represented by an oblong leaflet, about 3.5 cm. long, appa leguminous, mounted, and by an empty packet labelled “ (Ured) brunea in fol exot Collins.” . The leaf bears reddish-brown spots on the upper surface, tt - PAPERS GIVING RUSTS OF NORTH AMERICA. —= 209 origin being obscure. The microscope shows no evidence of my- celium, and the spots are probably not due to a fungus. This con- clusion has been confirmed by Dr. C. L. Shear of Washington, D. C. *2851. 41. C. U. Chelidonii, L.v.S., very rare. On leaves of Chelidonium sent from New York. C. spots yellowish. Sori irregular in form, clustered, conffuent. Spores rather large, fuscous and black, oval, loosely scattered. Represented by an angularly ovate leaf, incised, 3 by 5 cm., hav- ing characteristic white hairs, especially on the veins, mounted, and _ by an empty packet labelled inside “ Uredo Chelidonit Halsey NYk,” and outside “Czoma (Ured) Chelidonii LvS NewYk Halsey.” The error in mistaking Osmorrhiza for Chelidonium was pointed out by Dr. W. G. Farlow in the preface to his “ Host Index of Fungi,” 1888. The mounted fragment of leaf bears two small groups of brown sori on the under surface, rather pulverulent, having both uredinio- spores and teliospores present, identical with Puccinia Pimpinelle (Str.) Mart. (P. Osmorrhize C. & P.), and essentially the same as nos. 2829 and 2841. It is a curious result of too credulously ac- cepting the first impression of the identity of a host that led Schweinitz three times to describe the same rust from the same host, as if representing three independent species on three wholly unlike and unrelated hosts. 4. Albugo. Note—Two numbers are given under this heading, both true representatives of the accepted Phycomycetous genus Albugo, and they are, therefore, omitted here. . 5. Sporidiis inequalibus (spores unequal). *2854. 44. C. U. gyrosum, Lk. 105, on leaves of Rubus Ideus, Bethlehem. = Represented neither by specimen nor packet. There is, how- ever, an original packet labelled “Czoma (Ur) gyrosa Reb. in Rub Id. Kunze,” and a similar one in the Herb. Curtis at Harvard Uni- versity. This collection shows a few small fragments of raspberry 210 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S — leaves, bearing pycnia and ecia of a Phragmidium on their w surface. It is possible that this European material represents the entry, inadvertently made for North America. It seems more pro able that Schweinitz found a rust at Bethlehem, which he consi ered the same, but for which there is now no specimen, If so, host was probably the European red raspberry, at that time mu 4 cultivated in American gardens. In that case the rust may have — been the ecial stage of Phragmidium imitans Arth., although Schweinitz nowhere records the more striking telial stage. — The exact status of the record necessarily remains uncertain. ; #2855, 45. C. U. cylindricum, Lk. 108, on Populus italica, Bethlehem. Represented by a 5 cm. square portion, cut from a large, | leaf, mounted, and by a few small fragments in the original pack: which is labelled “Caeoma (Ur.) cylindrica populina Bet.” The fragments of leaf are well besprinkled with uredinia, the microscopic examination shows essential similarity to the dinial stage of Melampsora Meduse Thiim., the common Ame! rust on various species of Populus. The host may well be the bardy poplar (Populus dilatata Ait.), as stated, —_— no : collection on this host has come to hand. 5 gies *2856. 46. C. U. epiteum, Lk. 112, on leaves of Salix nigra, over near! whole tree, Bethlehem. be Represented by two short stems with respectively two and t attached leaves and three unattached leaves placed loosely in mounted packet, and by small fragments of a young stem and leav in the original packet, which was at first labelled “ Uredo epit in Salici nigri Beth,” then the word “epiteum” crossed out “ Saliceti” substituted, and afterward the first wording res All the leaves are covered beneath rather sparingly with ure The collection is the first to be recorded for the very com ‘ American form on various willows, Melampsora Bigelowii Th The spores are noticeably small and thin-walled for the species The willow rusts are yet imperfectly understood. The host clearly Salix nigra Marsh. z PAPERS GIVING RUSTS OF NORTH AMERICA. 211 8. Subgenus ZcIDIUM. aa #2857. 47. C. A. Convallariatum, Lk. 114, on leaves of Smilacina racemosa, Bethlehem, very rare. _ Represented by a mounted specimen of the middle part 4.5 cm. long, of a 2.5 cm. wide leaf, bearing beneath about ten small groups of circinating zcia, and by an empty packet labelled “ Acid Con- vallariatum Salem.” a The rust is an hetercecious form, without doubt, and is usually considered to be the cial stage of Puccinia Majanthe (Schum.) A. _ &H,, occurring in both Europe and America on Phalaris and other __ grasses, but the genetic connection has not been fully established for _ the American material. s We must assume that “Salem” on the original packet was an @ error for “ Bethl,” in view of the printed record, which is starred a and does not mention Salem. % The host was doubtless as stated, Vagnera racemosa (L.) ~ Morong (Smilacina racemosa Desf.) - 2858. 48. C. A. Uvulariatum, L.v.S., Syn. Car. 453, hardly C. Alliatum as re- ” ferred by Link, n. 116, for it differs in having spots rather small, never exceeding a fourth of an inch, also in being white. (453. 24. [Ecidium] Uvularie Sz. A. orbicular, white, delicate, peridia excentric, circinate, white, spore-mass white. Here and there on the leaves of Uvularia perfoliata. Peridia crowded in concentric circles, none in the center itself. Similar to A. Allii ursini, but the color in that is yellowish.) Represented by the proximal half of two perfoliate leaves at- tached to the 2.5 cm. stem, mounted, one of the leaves bearing a _ Single, rather diffused group of zcia, and also by an empty packet, __ labelled on the inside “ A&cidium circinatum Rhig In Uvularia perfol _ & Polygonatum Salem,” and on the outside “ ZZcidium Uvulariatum LvS. Salem.” This rust has the same uncertain status as the preceding one, but is generally considered the zxcial stage of Puccinia Majanthe (Schum.) Arth. Schweinitz’s name was changed on p. 309 of his later work to Zcidium (Ceoma) uvulariatum. 212 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S 2859. 49. C. A. Smilacinatum, L.v.S., Syn. Car. 452, Lk. 117, not yet met. with in Pennsylvania. (452. 23. [A®cidium] Smilacis Sz. A. wart-like, convex below, concave above, yellow-red, p copious, spores white. Here and there on leaves of Smilax rotundifolia and leunifolia’ Very distinct. Making thick, conic-cylindric warts on the under side of the leaf. These warts are somewhat truncate and on the pulvinate- truncate part covered with sunken peridia, two lines to a quarter of an inch wide and two or three lines high. Spores white, rather large, oval, vesicular.) ; Represented by a nearly round leaf, 5 cm. in diameter, mounted, bearing one group of ecia, and by an empty packet labelled inside “ ZEcidium Smilacis In S. rotundifol & al Salem,” with the addition “Czoma Smilacinatum,” and on the outside “ A2cidium ——— tum LvS Salem.” This is the ecial stage of Puccinia Smilacis Schw., a rust dus is widely distributed in the southern states, and tropical America. The ecia are rarely collected, and so far have been reported only from North and South Carolina. The name was changed by Schweinitz to Zcidium (Ce@oma) smilacinatum on page 309 of his later work. 2860. 50. C. A. Aroidatum, L.v.S., Syn. Car. 457, [as A.] Caladii, on virginicum, Salem. (457. 28. [Aécidium] Caladii Sz. : A. simple, on very extended areas, peridia rufous-yellow, sphz form, spore-mass orange. Frequent in some years on the midrib of the leaves and stems of Caladium sagittefolium; it kills the plants. The cle peridia resemble Sphzrias). ae Represented by the middle part, 3 cm. long, of a 5 cm. w leaf, with over-mature xcia along midrib and large veins, now ez by insects, and by an empty packet labelled inside “ A¢cidium i In Calad. Salem,” with the later addition “ Ceoma aroidatum,” < on the outside “ A£cidium Caladiatum LvS. Salem,” with the s sequent addition “ Aroidat.” This is the cial stage of Uromyces Caladii (Schw.) Farl., . on Peltandra virginica (L.) Kunth (Arum virginicum L.), see also no. 2839. The name Ceoma Aroidatum should have been credited PAPERS GIVING RUSTS OF NORTH AMERICA. 213 to Link, n. 118. Schweinitz changed the name to 4cidium (C@oma) aroidatum on page 309 of his later work. *2861. 51. C. A. Dracontionatum, L.v.S., frequent on leaves and petioles, and also on the scapes of Arum dracontium, Bethlehem. Not the same as the preceding. Also Salem. C. spots pale, widely scattered over the leaf, occupying nearly the whole of it. Pseudoperidia large, scattered irregularly in dense clusters on the spot. Spores orange color. Represented by a much broken leaf, 3 by 5 cm., mounted, thickly covered beneath with large xcia, and by a packet labelled inside “ ZEcidium Dracontii In Aro Dracont Salem,” and on the outside “ 7Ecidium Dracontiatum LvS Salem,” containing a few very small fragments of leaf, showing ecia. ; The differences noted by Schweinitz between this collection and the preceding one are now ascribed to the influence of the host, and the form is referred to Uromyces Caladit (Schw.) Farl, the host being Muricauda Dracontium (L.) Small (Arum Dracontium L., Arisema Dracontium Schott.). The name of the rust was changed to Zcidium (C@oma) dracontionatum on page 309 of his later work. <» *2862. 52. C. A. rubellatum, Lk. n. 120, rather rare on various species of Rumex, Salem and Bethlehem. Spots generally sterile. It is evident that Schweinitz should have cited here the follow- ing similar entry in his North Carolina list, and have omitted the asterisk. (433. 4. [4Ecidium] Rumicis. Frequently seen as spots on Rumex and Grossularia; but the fungus is very rarely perfect.) ite No specimen or packet remains to represent these records nor is ___ there any in the Herb. Curtis at Harvard University. Both entries _ are without doubt founded upon errors of observation. Rumex leaves are often spotted from the action of fungi imperfecti which could easily be mistaken for the small zcia not uncommon on this _ host in Europe. The mention of Grossularia was doubtless in con- _ formity with Persoon, who thus associates these hosts. 214 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S 2863. 53. C. A. Lysimachiatum, Lk. 126, Syn. Car. 438, absolutely the In Pennsylvania, generally on L. racemosa. : (438. 9. [Afcidium] Lysimachie Sz. A. diffuse, pale, rather small, epiphyllous, peridia crowded, tate, spore-mass somewhat flesh colored. On leaves of Lysimachia quadrifolia and stricta, wales: chance two species exist; for the one on quadrifolia is not pale, r tinged with a red color. It makes a rather small spot on the ! surface of the leaves.) etna Represented by a somewhat torn leaf, 1.5 by 4 cm., bearing neath a rather diffuse, compound group of old ecia, and by an « packet labelled inside “ AScidium Lysimachie in L. quadrifol and on the outside “ A°cidium Lysimachiatum in Le “ Salem.” Schweinitz’s statement, “absolutely the same,” doubiie re to a note in Link’s work as to the identity of American and Et pean material, which mycologists still hold in general with Sch nitz to be one, although Link was too uncertain about the matter to accept Schweinitz’s name as a basis or even as a synonym of C. Lysimachiatum, founded upon Schlechtendahl’s C. Lysimac which was published two years later than Schweinitz’s name. ’ fungus i& now accounted the ecial stage of the Carex rust, ust called Puccinia limose Magn., a widely scattered but rather loc species, recently given the name P. /ysimachiata (Link) Kern, tk being already a P. Lysimachie of Karsten, 1879. Both spot and ecia on the mounted leaf still appear reddish, stated by Schweinitz for L. quadrifolia. The two names, L. . Ait. and L. racemosa Lam., are now considered synonyms of L. te restris (L.) B.S. P. . 2864. 54. C. A. Pentstemoniatum, L.v.S., Syn. Car. 449, Lk. p. 47, only b- served in Carolina. (449. 20. [A®cidium] Pentastemonis Sz. A. orbicular, rather small, dense, purple, yellow beneath, f white, congested. Not infrequent on leaves and stems of Pentastemon hirst Distinct species. Two lines broad. Peridia large for the size of plants. Spores yellow-brown, simple, vericulose.) Represented by an original packet, containing three fragmen PAPERS GIVING RUSTS OF NORTH AMERICA. 215 _ leaves and a small portion of a stem, now in rather poor condition, and showing only a few ecia on one of the leaves, labelled inside “ cidium Pentstemonitis Salem,” and on the outside “ Acidium Pentstemoniat LvS Salem.” Although there is no mounted speci- ‘men there are pin marks where one may have been attached. The rust is common in the eastern United States, and is the zcial stage of no. 2911, Puccinia Andropogonis Schw., as proven by cultures first made by the senior author in 1899 (Bot. Gaz., 29: 272), and subsequently repeated a number of times. The southern Pent- stemon, corresponding to the northern P. hirsutus, is P. australis Small. Schweinitz changed the name of the rust to £cidium (Caoma) pentstemoniatum on page 309 of his later work. 2865. 55. C. A. Apocynatum, L.v.S., Syn. Car. 448, Lk. n. 135, not yet [seen] in Pennsylvania. (448. 19. [Ecidium] Apocyni Sz. A. orbicular, very large, orange, pale below. Peridia arranged in a few concentric circles, somewhat fuscous. On leaves of Apocynum cannabinum in the mountains. Spots delicate. Peridia when closed from yellow to chestnut-brown or somewhat fuscous, when open with a pale, lacerate margin. Spores simple, white.) Represented by a mounted specimen of the middle part, 4 cm. long, of a 3.5 cm. wide leaf, bearing beneath two groups of «cia, centrally placed on dark spots 7 mm. across, and by a packet con- taining a small part of a leaf, showing no fungus. and labelled “ ZEcidium Apocyniatum in Apocyn. pubes. Salem.” This rust is not much better understood than in the days of Schweinitz. Only six other collections are known to the writers, which have come from Delaware, New Jersey, District of Columbia and North Carolina. It is probably a hetercecious form, but no suggestion has been made regarding the alternate host. The name was written 4cidium (Ce@oma) apocynatum by Schweinitz on page 309 of his later work. 2866. 56. C. A. Convolvulatum, L.v.S., Syn. Car. 454, very frequent also in Pennsylvania on C. panduratus. (454. 25. [Ecidium] Ipomez-pandurane Sz. A. very large, bullate, depressed above, white, peridia flexuose, rather large, elevated, ruptured by a slit, spore-mass cinereous- golden-red. 216 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Frequent on the leaves of Ipomcea (Convolv.) pandurana, | idia thick, the loose epidermis larger than in almost any Xcidia, cept cornutum and cancellatum. Spores rather small, oblong.) Represented by a mounted stem, 7 cm. long, and part of leaves, and by an original: packet, containing ample material, labelled inside “ AZcidium Ipomez in pandurata & lacunosa Salem, and on the outside “ A2cidium Convulvuliat LvS. Salem & tig Conv. pandurat.” ce The fungus is certainly and wholly Albugo I pomea-ianee (Schw.) Swingle one of the Peronosporales, and not a rust. 7 name was changed to Aicidium (C@oma) convolvulatum at page 309 of the later work. 2867. 57. C. A. Compositarum, Lk. n. 139, and frequent in Pennsylvania, a Prenanthis on Krigia, Salem, Syn. Car. 434. — 8 Eupatorize, Bethlehem, frequent on E. purpureum. (434. 5. [4Ecidium] Dandelionis Sz. Why not merely a variety of Acidium prenanthis, to w it is very similar? Spores subglobose, without septum and ped- icel, chestnut-brown. On leaves and stems of Tragopogon Dan- delion. Rare.) Represented by an original packet, containing a few very small fragments of a leaf with many ecia, and labelled inside “ Aéci Eupatorie maculate Bethl,” and on the outside “ Acidium Eupa- toriatum LvS Beth,” with “compositatum” written above. Th is no packet for the other entry, and no mounted material for either, although there is indication that there may once have been a mount where pin marks now show. The xcia on Eupatorium are doubtless to be assigned to th widespread rust, Puccinia Eleocharidis Arth., very common b north and south, the uredinia and telia being on various species a Eleocharis, and the zcia on various species of Eupatorium, in ing both E. maculatum L. and E. purpureum L. As the fragm in the original packet shows the leaf to be smooth above with mint sparse pubescence beneath and not at all scabrous, the host is dou less E. purpureum and not E. maculatum, the conclusion evidentl reached by Schweinitz. The identity of the form on Krigia is somewhat uncertain. The PAPERS GIVING RUSTS OF NORTH AMERICA. 217 color of the spores fits well the uredinia of Puccinia Pyrrhopappi Syd. (P. Krigie Syd.), the only known collection on Krigia having been made by Dr. B. L. Robinson at Asheville, N. C., Aug. 2, 1893, on K. virginica Willd. But that form of rust has scattered sori, and not clustered as in an 4cidium. Schweinitz thought the fungus not unlike 4cidium Prenanthis Pers., and fortunately there is a specimen of this species in the Schweinitz collection, which had been received from Kunze. It consists of a smooth, thin, deltoid leaf, some 5 or 6 cm. across, which bore a single cluster of ecia, most of which has now disappeared. It is clear, nevertheless, that Schweinitz must have had an ecidioid fungus on the Krigia. The only known form on Krigia with clustered sori having “ chestnut- brown” spores is that of the short-cycle species which at another time and on another host Schweinitz called Puccinia maculosa (see no. 2922). The teliospores germinate at maturity in the sorus, and placing some of them under such magnification as Schweinitz prob- ably used, gives the appearance of “spores subglobose, without septum and pedicel.” The host was well known to the contemporaries of Schweinitz, and commonly called the “small dandelion” (see Muhlenberg’s Catalogue, p. 71). It was considered closely related to Prenanthes. The latest form of the name is Adopogon Dandelion (L.) Kuntze. *2868. 58. C. A. Hieraciatum, L.v.S., here and there on the leaves of H. . paniculatum and maculatum, Bethlehem. C. spots deep purple, widely effused. Pseudoperidia circinate, on the center of the spot, margins beautifully fimbriate, spores orange. 32... Represented by 5.5 cm. of a lanceolate leaf, 2 cm. wide, denticu- __ late, slightly pubescent beneath, having two groups of zcia, and by an empty packet labelled “ ZEcidium hieraciatum Lv Hieracii panicu- lat Beth.” The host is correctly named, for the leaf exactly matches the leaves of a phanerogamic specimen collected by Schweinitz at Salem, N. C., now in the herbarium of the Philadelphia Academy, which is without question H. paniculatum. : The name of the rust was changed by Schweinitz to Zcidium | ‘a (Ceoma) hieraciatum on page 309 of the same work. The rust is 218 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S undoubtedly identical with a widespread species, having telia on Carex and ecia on many Cichoriaceous hosts, but it has not been reported by any other collector on Hieracium paniculatum, 1 species has generally been called Puccinia patruelis Arth., Schweinitz’s specific name is much older and should heehee used, making the name P. hieraciata (Schw.) comb. nov. No otheds collections of P. hieraciata, either of zcial or telial stages, are known with certainty east of Michigan and Indiana, but it is not im- probable that the species occurs sparingly in the eastern mncntalnde #2869. 59. C. A. Erigeronatum, L.v.S., rather rare but ample on E hetero- phyllus, Bethlehem. C. spots very large, yellowish, rather thick. Pseudoperidia | dens¢ and irregularly scattered, elevated. Spores yellowish, + Represented by the major part, 5.5 cm. in length, of two leaves, 2 and 3 cm. broad respectively, the smaller showing four groups 0: cia, and the larger many ecia thickly and evenly grouped ov an area 2.5 cm. across, and also by an empty packet labelled insi “ Acidium Flosculosorum Salem,” later added below “in Solid: ) Erigeron, Aster,” and still later added above “ Coma asteratum,’ and finally “erigeronatum,” and also labelled on the out “ 7Ecidium compositat Erigeronatum LvS Bethl.” ue The host, which has been compared with phanerogamic sf mens, is certainly Erigeron annuus Pers. (E. heterophyllus Muhl.), and the rust is the common one on this host, being the zcial s ag of Puccinia Asterum (Schw.) Kern, and belonging to the phys logical race represented by the name Puccinia Caricis-Erigero Arth., as proven by cultures. Schweinitz changed his name cidium (Ca@oma) erigeronatum on page 309 of the same — 2870. 60. C. A. Asteratum, L.v.S., Syn. Car. 444, Lk. 143, common, espec on A. paniculatus. Link does well to join with this C. daginis, Syn. Car. 446, and C. Verbesinz, 445. But C. Heli does not belong here. (444. 15. [AEcidium] Asterum Sz. A. effuse, confluent, very delicate, pale, purplish, peridia gated, immersed, spore-mass white. Here and there on leaves and stems of smooth leaved Aves Spores rather large, vescicular, globose or oblong, simple.) PAPERS GIVING RUSTS OF NORTH AMERICA. 219 (445. 16. [£cidium] Verbesine Sz. A. oval, rather thick, small, pale reddish yellow, peridia few, prominent, white. Frequent on Verbesina, Sigesbeckia, and others. Spots four lines in diameter. Spores simple, very small, pale, margins of the peridia entire.) (446. 17. [Ecidium] Solidaginis Sz. "A. effuse, rather large, peridia scattered, minute. Frequent on stems of Solidagos before flowering. Similar to the preceding. ) Neither specimens nor packets remain to represent these entries. -Schweinitz was right in putting the Solidago xcia with those on Aster. They go with the Aster-Solidago-Erigeron-Carex combina- tion lately passing under the name, Puccinia extensicola Plowr., along with the preceding number, one belonging to the physiological race, Puccinia Caricis-Asteris, and the other to that of P. Caricts- _Solidaginis as abundantly indicated by cultures. The present ac- cepted name is Puccinia Asterum (Schw.) Kern. He was also right in excluding A. Helianthi-mollis, hete given under the subsequent number; but he was wrong in retaining A. Verbesine. The Verbesina xcia belong with the autcecious rust _ Puccinia Verbesine Schw. (see no. 2925), a rust which is common throughout the southern states. All collections of this species ap- pear to be on V. occidentalis (L.) Walt., which doubtless was the host of Schweinitz’s no. 445. No rust has yet come to hand on Siegesbeckia (Actinomeris), and the inclusion of the name must have been due to an assumption not supported by collections. Schweinitz claimed authorship of this species, hence places his initials after the name, although Link was the first to write it in this form, as Schweinitz was well aware. The name was written by _ Schweinitz cidium (Ceoma) asteratum on page 309 of his later work. 2871. 61. C. A. Helianthatum, L.v.S., Syn. Car. 450, frequent on H. mollis. Rare in Pennsylvania. (450. 21. [Ecidium] Helianthi mollis Sz. A. oblong, thick, whitish, peridia congested, pale, spores oblong. Frequent on the under side of the leaves of Helianthus mollis; 220 ARTHUR-BISBY—TRANSLATION OF SCHWEINITE® hairy. Spores under the microscope yellow-fuscous, vesicular old pellucid, white.) Represented by a lanceolate, very tomentose leaf, 4.5 cm, and part of another similar leaf, both mounted, showing s groups of «cia. An empty packet is labelled inside « Aeidiut Helianthi mollis Salem,” and outside ‘‘ A=cidium helianthatum * on Helianthi molli Salem.” The name was changed by Schweinitz to Acidium ‘(Cac helianthatum on page 309 of his later work. This collection resents the basis for the earliest name to be applied to any part the cycle of the American sunflower rust which is generally ¢ Puccinia Helianthi Schw. A less convenient, but technically m correct name, therefore, is P. Helianthi-mollis (Schw.) comb. *2872. 62. C. A. Trachelifoliatum, L.v.S., here and there on the lea Helianthus trachelifolius, Bethlehem. C. spots broadly effuse, yellowish or rufous, confluent, large. doperidia very densely aggregated in the center, as if ere and appressed to each other, and hence somewhat angular, erately elevated; margin not fimbriate. Spores yellow, decolored. : Represented by parts of two originally large leaves, 3 and 4 broad respectively, mounted, and by three broken leaves and ma fragments in the original packet, which is labeled “ A2cidium H anthi trachelif.” The leaves bear a number of groups of The fungus is the cial stage of the common sunflower Puccinia Helianthi-mollis, and the host, so far as the sp shows, is as given by Schweinitz. The name was chang Schweinitz to Zcidium (C@oma) trachelifoliatum on page the same work. | *2873. 63. C. A. Gnaphaliatum, L.v.S., striking and very common in the 1 autumn on leaves (under side), also on the woolly stems Gnaphalium polycephalum, Bethlehem. C. hypophyllous, at first cloaked in the wool of the leaves and s! Spots more or less effuse, yellowish. Pseudoperidia only a but densély approximate, very often even single, very long, : very white, cylindric, apex fimbriate. Spores orange yellow: is related to C. Pini in the form of the peridium. Represented by two stems, each 6 cm. long, and many rae PAPERS GIVING RUSTS OF NORTH AMERICA. 221 Mectcs, loose in a mounted packet, and by two original packets, one containing a stem 12 cm. long, and a few leaves, labelled “ AEcidium Gnaphalites LvS 1828,” and another containing a _ leaves labelled “Czoma AEcidium Gnaphalitum LvS. spec. exim.” The collection shows a few xcia. . _ The host is without doubt G. obtusifolium L. (G. polycephalum Michx.), and the rust is the ecial form of what has commonly been called Puccinia investita Schw. (no. 2932), but owing to the prior- ity in position of the present specific name, should be called P. gnaphaliata (Schw.) comb. nov. The name was changed by Schweinitz to Zcidium (Ce@oma) gnaphalitatum on page 309 of the same work. 4. 64. C. A. Clematitatum, L.v.S., Syn. Car. 447—and collected in Penn- sylvania—a good species. (447. 18. [£cidium] Clematitis Sz. A. pale red, peridia congested, few. On younger leaves of Clematis Virginiana, Bethany. A waded species ?) _ Represented neither by a specimen nor a packet. In his Caro- lina list Schweinitz was in doubt about the validity of his species, later felt assured, and consequently added “a good species” in his later list. There can be no question, however, that the fungus is one identical with the well-known Zcidium Clematidis DC., and which has now been proven by cultures in both Europe and Amer- ica to be the cial stage of Puccinia Clematidis (DC.) Lagerh. .(P. Agropyri Ellis & Ev.). __ The variable use of ¢ and d in forming a suffix was not un- common among the earlier mycologists, where in recent years d only is employed, thus the spelling “ Clematitis,” instead of Clema- tidis, etc. Schweinitz changed the name to Zcidium (C@oma) clematita- m on page 309 of his later work. 2875. CA. Rammienestien, Lk. [n.] 150. Frequent, Carolina (Syn. Car. 440) and Pennsylvania on various species of Ranunculus, e. g. R. abortivus and others. PROC. AMER. PHIL. SOC., VOL. LVII, P, JULY 17, 1918. . 222 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S (440. 11. [AE=cidium] Ranunculi (abortivi). Frequent on the r radical leaves, almost devoid of spots.) _ Represented by three radical leaves of Ranunculus abortivus 2.5 cm. broad, mounted, well covered beneath with ecia, and empty packet labelled inside “ AEcidium Ranunculi nitidi S and outside AEcidium ranunculiat Ran abortivi Sal & Bet.” Schweinitz was correct in his first list in considering this ft distinctive, and in error later in assigning it to Link’s inch species. It occurs only in America, and in the eastern United only on Ranunculus abortivus, being the ecial form of FP Eatonie Arth. *2876. 66. C. A. Cimicifugatum, L.v.S., very beautiful, rather rare on leav of Cimicifuga racemosa, Bethlehem. Where found almost leaves are infested. C. spots large, orbicular, yellow, bullate. Pseudoperidia on n th surface, concentric, very long, cylindric, apex at first then subfimbriate. Spores orange, becoming white. = Represented by parts of three leaves, each part about long, mounted, showing considerable groups of very long cy nd peridia, and by an empty packet labelled inside “ A®cidium . near Easton on Delaware very rare,” and on the outside ‘ Pee Acteeatum LvS Bethl,” with Actezeatum crossed out i * fugatum” substituted for it. IS This imperfectly known rust is even at the present sia form. It is probably hetercecious, and may belong to some rust. Schweinitz changed the name to Acidium (Com fugatum on page 309 of the same work. *2877. 67. C. A. Hibisciatum, L.v.S., on leaves of Hibiscus ni lehem, cultivated, not rare. C. spots orbicular, yellowish, confluent. Pseudoperidia i but densely scattered, delicate, yellow. Spores not comp loose, yellowish. Represented by one obliquely triangular-ovate leaf, 3 by 5 mounted, having many groups of zcia, and by an empty labelled “Czeoma ZEcidium Hibiscatum LvS in H. militaris E The rust is the ecial stage of Puccinia hibisciata (§ PAPERS GIVING RUSTS OF NORTH AMERICA. 223 ‘Kellerm. (P. Muhlenbergie Arth. & Holw.), on Muhlenbergia and other grasses, as repeatedly proven by cultures. Schweinitz changed the name to Zcidium (Ceoma) hibisciatum on page 309 of the same work. #2878. 68. C. A. Hepaticatum, L.v.S., scarcely C. quadrifidum, Lk. n. 152. Here and there on degenerate leaves, i. e., not trilobate, but nearly reniform and multilobed, of Anemone hepatica, Bethlehem. C. spots entirely wanting; the leaf, nevertheless, on which it rests degenerates. Pseudoperidia very large, broad, the margin ex- actly cleft into four parts, revolute, the lobes broad, brown. Spores fuscous-brown. Occupying the whole leaf. __ Represented only by an empty packet labelled “ Aécidium He- paticatum Bethlehem, 24.” _ It is probable that the failure to recognize this rust as the cidium quadrifidum DC., found on Anemone in Europe, was 7 largely due to the peculiar distortion of the leaf produced by the fungus in the case of Hepatica. The form on both Hepatica and Anemone is the ecial stage of the plum rust, Tranzschelia punctata _(Pers.) Arth. (Puccinia Pruni-spinose Pers.), and is on the common liverleaf_of the eastern states, Hepatica Hepatica (L.) Karst. (H. triloba Chaix., Anemone Hepatica L.). The combination Zcidium _(Ce@oma) hepaticatum is made by Schweinitz on page 309 of the same work. 2879. 69. C. A. Geraniatum, Lk. 156, on leaves of Geranium maculatum and G. carolinianum. Exactly identical with the European. Syn. Car. 443. (443. 14. [4&cidium] Geranii maculati Sz. A. diffuse, hypophyllous, thickened, red, peridia dense, broad, smooth on the margin, spores yellow. Frequent and large on leaves of Geranium maculatum. On the upper surface of the leaves it makes a diffuse spot. Peridia densely aggregated. Spores simple, globose, cellular under the microscope, yellow-fuscous; some are united in pairs as if compound, and very rarely are furnished with a pedicel.) Represented by the central part of a leaf, 2 by 3 cm., mounted, wing one large group of xcia, and by an empty packet labelled side “Acidium Geranii maculat Salem,” and on the outside “ ZEcidium Geraniatum LvS G. maculat Salem.” 224 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Schweinitz’s inclusion of Geranium carolinianum as one of hosts must have been a hasty generalization. A specimen of thi plant in the phanerogamic herbarium at the Philadelphia Academy of Sciences, obtained by Schweinitz at Salem, shows that he familiar with the plant, but no zcia are known to have ever collected on the species, or on any American Geranium with sim leaves. z Although Schweinitz adopted Link’s name, yet Link hesitated place the American rust under his species, and properly so as ti has proven. Link’s form is a stage of Uromyces Geramii (DC. Otth & Wartm., an entirely different rust. The Schweinitz form is the zxcial stage of Puccinia Poly amphibii Pers., as established by cultures in both this country Europe. Recently some European mycologists have consid that the American form of this widespread species should be tre: as distinct from the European form. But it would doubtless better to consider the species as made up of a number of more less distinct races, and that the common form in America is a different from the common form in Europe. 2880. 70. C. A. Impatientatum, L.v.S., Syn. Car. 442, Link pag. 57 in also Bethlehem. (442. 13. [Ecidium] Impatientis Sz. A. effuse, large, becoming pale, peridia in the center, s crenate, spores rather large, yellow-fuscous, simple. Frequent in May on the leaves of Impatiens maculata. Its the leaves and stains a broad yellowish spot, darker in the i ‘ Represented by part of a leaf, about 3 cm. long, and 2 cm. mounted, bearing a single large group of cia, and by an packet labelled inside “ AZcidium Impatientis Salem,” and c¢ “ ZEcidium Impatientat LvS Salem.” Link, at the place cited, indicated the possibility that this f might belong with the preceding one. It is, however, dif a although having much similarity in gross appearance. It is, in the xcial form of the American Puccinia Impatientis (Schw.) At (P. perminuta Arth.), having telia on Elymus, Agrostis and grasses, as proven by cultures. The name was changed by Sch nitz to Zcidium (Ceoma) impatientatum on page 309 of his I work. PAPERS GIVING RUSTS OF NORTH AMERICA. * 225 2881. 71. C. A. Berberidatum, Lk. 157, on Berberis canadensis, Carolina. This number is not starred, and it is probable that a reference to the record in the North Carolina list was omitted unintentionally. It is here added. (437. 8. [AEcidium] Berberidis. Rather rare on leaves of Berberis vulgaris, covering the mountains of Wilkes County.) Represented by a mounted specimen of a stout, ash-gray stem, 3.5 cm. long, having two fascicles of leaves, two full-grown leaves in one fascicle and three in the other, each.leaf 1.5 by 3 cm. or some- what less, bearing a number of small groups of young ecia; one group only appearing mature (see cut). There is also an empty aS a i tc 2 eso oe = Fic. 1. From a photograph of the mounted specimen in the Academy of Natural Sciences of Philadelphia, basis of Schweinitz’s No. 2881. Each specimen in the mounted set is treated essentially in the same manner. The writing was done by Michener. Engraved full size. packet labelled inside “ A£cidium Berberidis,’ and on the outside “7Ecidium Berberidat in Berb canad Salem.” The rust is the ecial stage of Puccinia pocultformis (Jacq.) 226 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Wettst. (P. graminis Pers.), whose telia are very common on ¢ and other grasses. The ecia have never been taken in An upon wild species of barberry, unless this record by Schweinitz such an instance. In the Carolina list they are said to occur « Berberis vulgaris “covering the mountains of Wilkes Co Evidently Schweinitz sometime after collecting his specimen som where in the vicinity of Salem learned that the native Berberis the mountains near Salem, N. C., is B. canadensis, and his colle _ tion was later so labelled and so recorded in his North American lis There is in the herbarium of the Academy of Natural of Philadelphia an ample and characteristic phanerogamic of B. canadensis from Salem, N. C., collected by Schweinitz another from Statesville, N. C., collected by Gray, Sargent, field and Canby, making it certain that B. canadensis did occur stated. But comparing the mounted cryptogamic specimen, must certainly have been the original collection, it is easy to see it does not agree well with the phanerogamic specimen by _ c nitz or the same species by others, as it has the ash-gray bark o vulgaris, instead of the dark reddish-brown bark of B. canae The evidence goes to show that although Schweinitz may observed the native barberry “‘ covering the mountains,” yet the was “rather rare,” and on Berberis vulgaris, as it has ge been found to occur during the years that have followed, in the Carolinas but throughout the eastern United States. T! is no reason to think that the rust will not as readily infect ; Berberis in its native state as it does the cultivated species, but to the present time there is no such authentic record. *26ne 92, CA, grossulariatum, Lk. 162. Very frequent on various § of Grossularia in the mountains of Pennsylvania. Represented by twenty leaves mounted loose in a packet, largest about 2 cm. across, showing a number of small grou Is zcia, and by an empty packet labelled on the outside “A grossulariat Mauchunk in Gros oxya,” with an evident emend written within “et Mauch Chunk Pensylva. in Rib oxyacanth Ly. Except one greenish fragment, the leaves are all of a brownish tint and similar in appearance. They may well be Grosst laria oxyacanthoides (L.) Mill. (Ribes oxyacanthoides L.). PAPERS GIVING RUSTS OF NORTH AMERICA. 227 _ The rust is the «cial stage of Puccinia Grossularie (Schum.) pe having telia on many species of Carex. 2 73. C. A. Hypericatum, L.v.S., Syn. Car. 451, Lk. 159, here and there, also near Philadelphia. (451. 22. [£cidium] Hyperici frondosi Sz. A. suborbicular (orange), peridia cylindric, elevated (white when dry), spores white. Frequent on leaves of Hypericum frondosum. Narrows of Yadkin, very beautiful, bright orange, making rather small but numerous spots sometimes almost devoid of the distinctive color. Peridia elevated as in A‘tc. Rhamni, to which somewhat related. Spores oblong, white, rather pellucid.) Represented by a dozen or so leaves, partly attached to short ‘stems, mounted loose in a packet, the leaves showing a few small, circular groups of white, cylindric zcia, and by an original packet _ containing a few leaves labelled inside “ A®cidium Hyperici frondosi Narrows of Yadkin,” and outside “ Aicidium Hypericatum LvS Hyp trond Narrows of Yadkin Carol.” The host agrees with a phanerogamic specimen, labelled by _ Schweinitz “Hypericum frondosum, Salem,” now in the collection _of the Philadelphia Academy, which is identified as H. prolificum L. The rust is the ecial stage of Uromyces Hyperici-frondosi (Schw.) -Arth., and is undoubtedly on Hypericum prolificum L. (H. frondo- sum Michx.). The combination £cidium (Ceoma) hypericatum _ Schw. was made on page 309 of the later work. 2884. 74. C. A. Violatum, Lk. 158, Syn. Car. 439, on leaves of various violets of Carolina and Pennsylvania, e. g., V. cucullata, obliqua, hastata, and the like. (439. 10. [Ecidium] Viole Conspect. fung. Niesk. p. 118. Occurs especially on Viola hastata, but also on other stemmed violets.) _ Represented by two specimens mounted, one of them being the ___end of a stem with two folded, cordate leaves and one young seed g capsule, having zcia on the blade, petioles, stipules, and stem, and ___ by a corresponding empty packet labelled “ ZEcidium Violatum V. 228 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S — This specimen has large ecia, and spores that correspond to eciospores of Puccinia Viole (Schum.) DC, be The other mounted specimen consists of one reniform leaf, broad, bearing three groups of xcia; and there is a corresp empty packet labelled “ A2cidium Violatum V. obliqua Beth.” This specimen shows smaller cia, and much smaller zcios than the other, and is doubtless the zcial stage of Uromyces tus (Schw.) Sheldon. The host is in all probability Viola prim folia L. he *2885. 75. C. A. pedatum, L.v.S., in some years very common on leaves 1 _ petioles of Viola pedata, Bethlehem. ; C. spots very small, much elevated and proportionally thick, p almost everywhere covered with rather large, somewhat subcylindric pseudoperidia. Spores pale. ox Represented by six leaves and one flower, mounted Coe packet, showing many ecia, and an original packet, containing very small leaves bearing a few small, irregular groups of ecia, is labelled “ Afcid Viol. pedate Lv Bethl.” ey The zcia and spores of this specimen, which are least on V pedata L., agree with those which were shown by cultures in to be the cial stage of Uromyces pedatatus (Schw.) Sheldon Andropogonis Tracy), having telia on species of Andropogon. name was changed to Acidium (Ceoma) pedatatum on page 30 the same work. *2886. 76. C. A. sagittatum, L.v.S., on leaves of Viola sagittata, Bethle Scarcely the same. C. spots purple, but yellowish on the lower surface. Pseudo slightly elevated, sparse, without order, on bullate spots, Spores concolorous. Represented by a short caudex with five attached leaf and three leaf blades, two of full size, 3 cm. long, one blade a1 petiole bearing indefinite groups of ecia, and by an empty labelled “‘ 7Ecid. Viole sagittat LvS Bethl.” : The necessity of. discriminating microscopic fungi chiefly by their gross appearance and the effect produced upon the host led Sch PAPERS GIVING RUSTS OF NORTH AMERICA. 229 -nitz to think this collection “scarcely the same” as the preceding one on Viola pedata, although a careful microscopic examination : shows that it has the same small spores and other characters which - go with the xcia of Uromyces pedatatus (Schw.) Sheldon. The mame was changed to £cidium (C@oma) sagittatum on page 309 of the same work. 2887. 77. C. A. luminatum, L.v.S., Syn. Car. [as A.] nitens, 458, also fre- ; quent in Pennsylvania on Rubus. The leaves, which with the whole plant are infested by this Aicidium, are degenerate (year after year.) (458. 29. [£cidium] nitens Sz. A. simple, elongate, peridia very large, yellow, brilliant, at length irregularly ruptured, spore mass orange. | Frequent on leaves, petioles and younger shoots of Rubus stri- gosus. Its perennial return so infests plants of the whole region that finally it entirely destroys them; summer. Resembles a Uredo, but it has a distinct peridium. Peridia finally confluent with each other.) Represented by five parts of leaves, each about 4 or 5 cm. long, in a mounted packet, and by many leaves and leaflets in the original _ packet, which is labelled inside “ A£cidium nitens in Rubo villoso Salem Bethl Neujork,” and in another place “ Czoma luminatum,” and on"the outside “ A2cidium luminatum LvS in Rub. villos Bethl _ & Salem.” All the leaves are covered with the rust and show the characteristic degeneration of the host. It was the custom generally followed by Schweinitz to preserve _ but the one original collection to represent each species. It is- quite evident from its appearance that the ample material of the present _ Species was all gathered at one time, and that it is all, or nearly all, _ from one plant, as it is very uniform. A part of the material has been seen by Dr. P. A. Rydberg, who monographed the genus Rubus for the “‘ North American Flora,” and he states that the host can not possibly be R. strigosus, but that it may be R. procumbens Muhl., or _ more likely its southern representative R. Enslenit Walt., both of which usually passed under the name of “R. villosus,’ a century ago. It will be noticed that Schweinitz labelled his collection R. _ villosus and did not change it afterward, although he added Beth- _ Iehem and New York for additional localities, and even changed the name of the rust to what he doubtless considered a better name, and — ARTHUR-BISBY—TRANSLATION OF SCHWEINTSS@ the turned the packet and placed on the outside his final r still with the host as R. villosus. It is impossible even to su why he used R. strigosus in the last printed account. The ha never been found on KR. strigosus in all the intervening years the use of that name by Schweinitz may certainly be —? as error. The rust itself is of special interest. Until very receiifie @ been identified with a similar rust of Europe, Gymnoconia stitialis (Schl.) Lagerh., a long cycle, autcecious form, as proven by cultures. The same long-cycle form also occurs in this country, also proven by cultures. Recently investigations by Kunkel ha’ shown that there also occurs in this country a short-cycle for orn whose telia are indistinguishable in appearance from the ecia i) long-cycle form, but differ in their mode of germination, and | t only the short-cycle form has so far been observed in the souther states, although both forms occur northward. The senior auth has recently (Bot. Gaz. 63:504. 1917) erected a new short-c genus with Schweinitz’s Salem collection as the type, so that i comes Kunkelia nitens (Schw.) Arth. The combination Acid (Ceoma) luminatum was made on page 309 of Schwere work, my inaccurate words in Syn. Car—‘ Spores bilocular,” i1 by a slip of the pen from the description of Puccinia Podop. an entirely different fungus—has wrongly placed this Aci the most remarkable of all, among the Puccinias. Ours u occurs with thick bullate spots, rendering the broad leay Podophyllum contorted and deformed—with a diameter o 4-6 inches. Pseudoperidia located in the center, slightly eley very densely crowded, rather large, and innumerable. The gin of the spot, however, always sterile. Spores are not bilo (435. 6. [A&cidium] Podophylli Sz. A. very large, orbicular, at length diffuse, golden yellow, er’ dense, spores somewhat elevated, bilocular. Usually it extensively and injuriously affects the leaves an stems of Podophyllum, attracting the eye by its beautiful color.) Represented by four pieces of leaves about 4 by 6 cm., mount loose in a packet, which are well covered with large groups of ecia, and by an original packet containing a number of large fragments of PAPERS GIVING RUSTS OF NORTH AMERICA. 231 leaves, bearing ecia, which is labelled “ Aécidium Podophyllat LvS Sal & Beth.” The rust is the xcial form of the long-cycle, autcecious species, _ Puccinia Podophylli Schw. (see no. 2939), on Podophyllum peltatum —L. The combination Zcidium (Ce@oma) podophyllatum was made on page 309 of the later work. *2889. 79. C. A. tenue, L.v.S., rather rare on leaves of Eupatorium agera- “ toides, Bethlehem. : C. spots yellowish, evanescent, very delicate. Pseudoperidia sparse, slightly elevated, but, what is peculiar, erumpent on both sur- faces, closed on the upper, open on the under. Spores pale. ‘ Represented by a mounted portion of leaf, cut 3.5 cm. square, bearing six or eight groups of zxcia, and by an empty packet, which is labelled inside “ AEcidium tenue Nobis In fol ignot Deetwiler,” and afterward “Eupat. agerat” substituted for “ignot,’ and is labelled outside “ A=cidium tenue in fol Eupat ageratoid Dettyler.” This is the ecial form of Puccinia tenuis (Schw.) Burrill, an _autcecious rust. The name is written Zcidium (C@oma) tenue on _ page 309 of the same work. —*2890. 80. C. A. Euphorbie hypericifoliz, L.v.S., frequent on leaves of E. hypericifolia, Salem and Bethlehem. It is not identical with C. Euphorbiatum Lk., nor does it make the leaves degenerate. C. spots small, deep purple on the upper surface, yellowish on the lower. Pseudoperidia aggregated, subconically elevated, and somewhat excavated. Spores orange. Although this number is starred and the earlier work is not di- rectly cited, yet the naming of Salem as a locality undoubtedly has _ reference to Syn. Car. 455, which in fact must be considered the basis of Schweinitz’s new name. (455. 26. [£cidium] Euphorbiz. Here and there on the leaves of Euphorbia hypericifolia, but does not make them degenerate.) Represented by a mounted fragment of a leaf, about I cm. square, well covered with ecia, and by an empty packet labelled “ Ecidium Euphorb. hypericif Salem.” 232 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S | The rust is the ecial stage of Uromyces proéminens (DC. and the host is Chamesyce Preslii (Guss.) Arth. (Euphorbia } Guss.), which passed under the name of E. hypericifolia in Sc nitz’s time. Link’s Ceoma Euphorbiatum is an entirely dif species, being the ecial stage of a hetercecious form. On page of Schweinitz’s later work the ‘name is changed to (Ceoma) Euphorbie hypericifolie. *2891. 81. C. A. Houstoniatum, L.v.S., rather rare, but where oc very copious on stem, leaves and peduncles of Ho cerulea, Bethlehem. = C. without distinct spots. Pseudoperidia elevated, pale, sub apex contracted, and somewhat excavated. Spores orange. 1 infected and somewhat degenerate plaets, nevertheless, fic Represented by three or more entire plaints mounted loose packet, all considerably drawn, but a few with flowers, and bi? original packet containing many rusted plants, which is~ “ ZEcidium Houstoniatum LvS Beth.” \ The rust is the ecial stage of Uromyces houstoniatus (Sc Sheldon, having telia on Sisyrinchium, as proven by. cultures. © combination Aicidium (Ceoma) houstoniatum is made on age of the same work. *2892, 82. C. A. Claytoniatum, L.v.S., on C. virginica from New Communicated by Dr. Torrey. C. almost simple and without spots, occupying the entire Pseudoperidia broad, sparse. Spores orange. ve Represented by a mounted stem, 5 cm. long, with one unop flower and two leaves, the leaves covered with ecia, and by | an inal packet containing one narrowly linear leaf, 6 cm. long, labelled “ AEcidium Claytoniat LvS Torrey.” A rather common rust, being the ecial stage of Puccinia toniata (Schw.) Peck. Schweinitz made the combination (Ceoma) claytoniatum on page 309 of the same work. *2803. 83. C. A. Pyrolatum, L.v.S., on the under side of the leaves of rotundifolia. Dr. Torrey. C. without spots. Pseudoperidia sparse, occupying the whole but not transforming it, pulvinate-elevated, pale, or orange PAPERS GIVING RUSTS OF NORTH AMERICA. 233 the spores. Finally these having fallen out Peziza-form cavities are left in the leaf. © @eoresented by half of a leaf, nearly 4.5 cm. broad, mounted, Wich is thickly covered with uredinia, and by an empty packet labelled “ ZEcidium Pyrolatum LvS in P. rotundifol Torr.” : The rust is the uredinial stage of Melampsoropsis Pyrole (DC.) _ Arth. (Chrysomyxa Pyrole Rostr.), but was naturally mistaken for an A&cidium by Schweinitz, as it possesses catenulate spores. The _ host may have been P. uliginosa Torr., rather than P. rotundifolia LL. The name Zcidium (C@oma) pyrolatum is used by Schweinitz on page 309 in the same work. *2804. 84. C. A. Myricatum, L.v.S., on leaves and especially on petioles of Myrica cerifera, communicated to me from New York by my friend Dr. Torrey. C. spots on strongly swollen petioles, dark purple, black where dry, and out of the spots project the dense pseudoperidia, rather large, widely open, brown, filled with yellowish spores. Represented by a mounted specimen, consisting of a terminal portion of stem, 2 cm. long, with four leaves attached, three being = somewhat over 4 cm. long and 18 mm. wide, and with an abundance of xcia ‘on the hyertrophied terminal bud, 2.5 cm. long, and by an original packet containing 3 cm. of stem with four leaves attached but without zcia, which is labelled on the inside “ ZEcidium Myrice on Myrica cerifera L,” and on the outside “ AZcidium Myricatum LvS in Myr. cerifera Torrey.” This is the ecial form of Gymnosporangium myricatum (Schw.) Fromme (G. Ellisii Farl.), as proven by cultures, the telia of which occur on Chamecyparis thyoides (L.) B.S. P. The name is changed to Zcidium (Ceoma) myricatum on page 309 of the same work. _ *2895. 85. C. A. Osmundatum, L.v.S., found on the fronds of Osmunda spectabilis and communicated by Torrey, but in drying so de- stroyed, that it is not possible correctly to describe it: the species nevertheless evidently distinct: spores ferruginous. Represented by a narrowly triangular, lateral part of frond, 2.5 em. long, blackish purple, mounted, and by an empty packet labelled _ “ Z&cid? Osmundatum in O. spectab Torrey.” Schweinitz used the 234 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S g name Azcidium (Ce@oma) osmundatum on page 309 of the work. ae The structure of this Ng is not evident, although th 3 abundance of globoid, brown spores present. The spores are in diameter, and echinulate or verruculose. They resemble spores, but Clinton in his monograph of the Ustilaginales’ North American Flora (7:24. 1906), where it is mentioned Ustilago Osmunde Peck, excludes the species from that order | suggests that it may be a Hyphomycete. The latest name is M: syrinx Osmunde Peck (N. Y. State Mus. ai IQII, page 43). *2806. 86. C. A. ylation, L.v.S., rather rare on leaves of Pyren) coro! Bethlehem. By no means identical with C. Roestelites, C. spots on upper surface, orbicular, red, on the border ochra center blackish. On the lower side there appear pseudop very densely crowded, subconcentric, only a little ele margin beautifully multifid-fimbriate; the parts straight, not all revolute, divergent, pale. Spores fuscous. Represented by one oblong leaf, 3.5 cm. long and 1.5 em. | broken across the middle, and mounted loose in packet, bearing - merous zcia on a somewhat hypertrophied spot, and by an en packet labelled on the inside “ Czeoma (Reestelia) coronariatum Salem in Pyr. coronar,” with “ A&cid” later substituted for F and on the outside “ Acidium Coronariatum LvS in Pyro con Salem.” The leaf is clearly that of Malus coronaria (L.) Mill, (Py coronaria L., P. angustifolia Ait.), the rust being ecia of ( sporangium Juniperi-virginiane Schw. To the mounted specime: attached another packet containing a little larger, more lz nceol leaf, with numerous ecia of the same sort, bearing an inscript by Dr. W. G. Farlow, saying it is from the Herb. Curtis, on Py angustifolia, Society Hill, N. C., no. 1226, and corresponds Schweinitz’s type of A. pyratum. The name Aicidium (Ce pyratum is given on page 309 of the same work. *2897. 87. C. A. sambuciatum, L.v.S., Syn. Car. 441, frequent on pe and leaves of Sambucus canadensis, also Bethlehem. A ft diagnosis follows. C. spots intumescent, often very large (i. e. 2 inches) on petic io PAPERS GIVING RUSTS OF NORTH AMERICA. 235 rather pale. Pseudoperidia large, dense, elevated, orange or pale, margin fuscous. Spores orange-fuscous, becoming de- colored. All much smaller on the leaves—pseudoperidia densely aggregated. (441. 12. [£cidium] Sambuci Sz. A. maculiform, large, thick, contorting the leaves, orange, becoming white, peridia minute, and spores simple, pale. Chiefly on the larger veins on the leaves, and on the petioles of Sambucus Canadensis. It distorts the leaves. Color orange- saffron; peridia sparse, spore-mass pale yellowish white.) Represented by parts of two compound leaves and bits of hyper- trophied rachis, mounted loose in a packet, showing numerous small groups of xcia, together with an original packet containing frag- ments of two leaves, also bearing small groups of ecia, labelled on the inside “ AXcidium Sambuci In Samb canad. Sal & Bethl,” and on the outside “ AEcidium Sambuciatum LvS Bethl.” This is the zxcial condition of Puccinia Sambuci (Schw.) Arth. (P. Bolleyana Sacc.), a common rust in the eastern United States, having telia on Carex. The asterisk before this number is a typo- graphical error. The name cidium (Caoma) sambuciatum is given on page 309 of the same work. 2898. 88 C. A. Urticatum, Lk. n. 169, Syn. Car. 436, very rare on Urtica. Salem, also at the same place on Cynoglossum amplexicaule. (436. 7. [Ecidium] Asperifolii, Rather rare on Cynoglossum amplexicaule. ) Represented by neither a specimen nor an original packet at _ Philadelphia or in the Michener collection at Washington, or in the Herb. Curtis at Harvard University. Cynoglossum virginicum L. _(C. amplexicaule Michx.) is not known to bear a rust. Neither is any rust known on Urtica so far south as North Carolina, although cia are common north of the 39th parallel of latitude. | The association of Urtica and Cynoglossum probably is carried over from European observations as given in the work by Albertini & Schweinitz (1. c., p. 117). It is probable that some appearance of the leaves misled Schweinitz into thinking that he had found in _ America the same rusts he had observed in Saxony. 236 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S” : y Subgen. Ra@STELIA or CERATITES, 2899. 80. C. R. Cylindrites, Lk. n. 172, Syn. Car. 432, under this n name included the following Czomata, perhaps to be separated species. , a. C. Crategi punctate, pseudoperidia divergent fibrous, swoll the middle—white. Pennsylvania. . 8. C. Crategi arborescentis, spots small, red, pseudoperidia not f of various forms, fuscous-red. Near Fayetteville, a y. C. Oxyacanthe, very large, very frequent near Philadelphia hedges. - 8. C. Mali, on leaves of Pyrus aii and coronaria, spots small effuse. Pseudoperidia minute. : (2. 3. [4Ecidium] Crategi var. Oxycanthe. A rare § leaves of various Crategi.) fe Represented in each of the four forms by specimens and o: inal packets from which it is possible to show that Schweinitz’s s mise was right, that they belonged to four distinct species. a. Represented by one leaf, 8 cm. long, of what is pro Crategus punctata Jacq., mounted, bearing six groups of ecia by two smaller but similar leaves, about 6 cm. long and 4.5 cm. b bro with no mature ecia, in the original packet, labelled inside “ “Ra (cornuta) oxyacanthe In Crat. pyrifol Bethl,’ with “ “corn crossed out, and “‘Czoma cylindrites”” written above, and ou labelled “ Cazeoma (Ceratites) Crategi punctate Bethl aff. pi at. The rust proves by microscopic examination to be the eecia Gymnosporangium globosum Farl. 8. Represented by a mounted leaf, 4.5 cm. long and about width, of what is probably Crategus viridis L. (C. arborescens bearing four groups of cia, and by half of a similar leaf with group of pycnia, in the original packet, labelled outside “ C (Ceratites) AEcidium Crategi arborescentes Fayetteville.” a similar but smaller leaf, with one group of ecia, is in the Mi collection at Washington, property of the U. S. Department of Ag culture. | This zcial rust is that of the very distinctive southern spe Gymnosporangium hyalinum (Cooke) Kern, whose telia are n¢ known. y. Represented by a large, 4.5 cm. broad and originally m PAPERS GIVING RUSTS OF NORTH AMERICA. 237 longer leaf, mounted loose in a packet bearing five large, circinating groups of zxcia, and by a small fragment of leaf about 3 cm. long, bearing cia, in the original packet, labelled inside “Reestelia oxy- acanthz a in Crat. oxyacant prope Philadelphia,” and above this written later “ Czoma cylindrites,” and labelled outside “2 Czoma _(Ceratites) cylindrites oxyacanthe in Hedgerows Philad. vulgatis- This zcial rust observed by Schweinitz to be very common, on what was doubtless the English hawthorn (Crategus Oxyacantha L.) and thought distinctive, was not again recognized until a trip by _ Dr. Frank D. Kern and the senior author to South Carolina in __ March, 1910, brought it to light. It belongs to Gymnosporangium = trachysorum Kern, having telia on Juniperus virginiana. ' _ 8. Represented by one large, 5 cm. broad, and originally 10 cm. long, strongly pubescent leaf of the cultivated apple, bearing numer- ous small groups of xcia, one half, 4 cm. long, being mounted, and the other half, 5 cm. long, in the original packet, which is labelled inside “‘ Reestelia cancellata In Pyro coronario Salem,” with all but the first word afterward crossed out, as if it were an error, and “8 penicillatum var Mali” substituted, and added below “var. in Malo Bethl,” ‘and still later there was written above “‘ Cezoma cylindrites,” while on the outside the packet was labelled “2 Czeoma (Ceratites) _cylindrites 8 penicillat in Pyr. Malo Beth.” | The rust proves to be the ecial stage of Gymnosporangium Juniperi-virginiane Schw. and on the common apple Malus Malus (L.) Britton (Pyrus Malus L.). The entry in the North Carolina list, no. 432, is not represented by a specimen, and is too indefinite to be associated with any certain _ species, unless the form £ be considered to cover it. 2900. 90. C. R. Roestelites, Lk. 173. cid. cancellatum, Syn. Car. 433 {error for 431]. In Bethlehem in an old orchard rejoicing in huge trees of Pyrus malus. In late autumn I have seen some of these trees, for 6-7 years, so covered by this fungus that the leaves appear red from a long distance. (431. 2. [cidium] cancellatum. Very rare, only once on pear leaves.) Represented by two sets of very unlike leaves, part of each being PROC. AMER. PHIL. SOC.. VOL. LVII. Q. JULY 17. 1918. 238 | ARTHUR-BISBY TRANSLATION OF SCHWEINITZ’S mounted. One of these consists of parts of two apple leaves, lengthwise, 5 or 6 cm. long, mounted, bearing many small groups zcia, and two similar pieces of leaves in the original packet, ona labelled “ Czoma ZEcid. Reestelites cancellat in Pyro malo arb maximas ad mortem zgens 1829 Bethl.” The other consists of ovate pear leaves (Pyrus communis L.), 6 cm. long, mounted, two similar, smaller leaves with another fragment in the orig packet, each leaf bearing one to three large groups of zcia, the pa being labelled “2 Czeoma (Ceratites) A=cidium Restelites can cell in Pyro Bethlehem.” i The ecia on the apple leaves belong to Gymnosporuiiaa J peri-virginiane Schw., and those on the pear leaves ie to G globosum. . 2901. gt. C. R. Fraxinites, L.v.S., Syn. Car. 430, Lk. 170, AEcidium fra Rather to be placed here; here and there; Rei on leaves. (430. 1. [A&cidium] Fraxini Sz. A. peridia elevated into a depressed chestnut-colored cone, length splitting into the broad lacinie. It makes round chestnut spots on the leaves, prominent beneath, flat above, surrounded — fuscous margin.) Represented by two lengthwise halves, 1.5 by 6 cm., of bro lanceolate leaflets, mounted, together bearing thirteen round grou, zecia on much swollen dark spots, but too young to show open per and by an empty packet, labelled inside “ Reestelia Fraxini In Salem,” with a later addition above “ Czeoma Reestelites Fraxinitu with “ Reestelites” afterward crossed out, and labelled on th : side ‘‘ AZcidium (Ceratites) Fraxinites LvS Salem & Beth.” — The rust is the ecial form of Puccinia fraxinata (Link) on species of Fraxinus, having its telia on the marsh grass, Spa [*]2902. 92. C. R. Botryapites, L.v.S. Very rarely observed on leaves Aronia botryapium, Bethlehem; but where it occurs, rather quent. C. entirely distinct—spots yellowish-buff, somewhat effuse. On under side the pseudoperidia appear central, aggregated as tt cles, globose, yellowish-green, at first obtusely conic and p closed, at length somewhat open and much fimbriated at opening, the divisions chestnut-brown, flexuous. Spores s dark. Pseudoperidia few, even at times single. PAPERS GIVING RUSTS OF NORTH AMERICA. 239 3 Represented by four leaves, one of them 4 by 6 cm., the others tr down to that size from larger leaves, mounted loose in a packet, bearing seven characteristic galls, and by an original packet with eight similar leaves, 4-7 cm. long, having bleached spots but no rust, which is labelled “®cidium (Ceratites) Botryapit LvS - Bethl 1830.” : _ The rust is the ecial stage of Pvacirpormapun botryapites (Schw.) Kern. At page 310 of the same work Schweinitz changed the name to Ceratites (Ceoma) botryapites. The asterisk was er- roneously omitted from this number. 5. Sybgen. PERIDERMIUM. 2903. 93. C. P. Pineum, Lk. 175, Syn. Car. 456. In Pennsylvania near ‘ Philadelphia and elsewhere, not rare. Specimens ample, a foot long, found by me on the trunk itself of Pinus inops, suggesting a resemblance to Gymnosporangium Juniperini. (456. 27. [£cidium] Pini. Rare with us, and only on young leaves.) _ Represented by two specimens. One of these consists of the sec- tion of a woody gall, 3 cm. across, mounted, with an empty packet, labelled “Czeoma Peridermium Pini in Ligno Philad.” A similar portion of a gall is in the Michener collection at Washington, prop- erty of the U. S. Department of Agriculture. __ The other consists of about a dozen slender leaves from a 2-leaved pine, none full length, now about 5.5 cm. long, mounted loose in a packet, bearing a few xcia, with an empty packet, labelled “ Czoma _ Peridermium Pini in acubus Salem.” _ Microscopic examination shows the woody form to be Perider- mium cerebrum Peck, the ecial stage of Cronartium Quercus (Brond.) Schrét., and the leaf form to be P. intermedium Arth. & sg 94. C. P. germinale, L.v.S., very rare on the fruits of roses. Com- = municated to me by Mr. Collins. C. pseudoperidia very long, cylindric, somewhat compressed, at length white, fimbriate, divisions cleft to the bottom, free. Spores effuse, pale. Pseudoperidia rising from little pits in the fruit, without any spot, usually three lines long. 240 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S _ Represented by a single oblong fruit, 1.5 cm. long, bearing long and colorless peridia, and by the original empty packet lab “Czoma Peridermium germinale LvS in germinib, Rose Collins Although the fruit has considerable resemblance to a mum1 rose hip, yet it is certainly the fruit of some species of Crate, and the fungus is the ecial stage of Gymnosporanginm ge n (Schw.) Kern (G. clavipes Cooke & Peck). The name is given _ Peridermium (Ceoma) germinale on page 312 of the same work. Genus 212. Puccinia Lk. and Dic#oma Fr. 2905. I. P. Graminis, Lk. n, 1, Syn. Car. 492. Very common also i in Pe sylvania on grasses. (492. 7. [Puccinia] Graminis. Frequent on the culms of gr. especially Andropogon.) pees Represented by the original packet containing a crumpled leaf. and some fragments of stem and sheaths, all apparently of wl Triticum vulgare Vill., bearing blackish, open telia of Puccinia po formis (Jacq.) Wettst. (P. graminis Pers.), together with six seven parts of conduplicate leaves, about 3 mm. wide, the pieces I ing from 6 to 15 cm. long, and heavily covered with dark bre blackish telial sori. The narrow leaves are undoubtedly some s of Carex, and the rust some species other than P. poculiformis, the identity of neither rust nor host has been definitely determi The packet is labelled “ Puccinia Graminis cerealis Germ. Sal. Be One of the pieces of sheath bears a small strip of gummed across the middle, showing that it had originally been attache sheet (see Shear, U. S. Dept. Agric. Bull., 380, p. 6, Jan. 15, 1¢ The writing on the packet appears to have been done all at one t It is, of course, impossible to say definitely if the material in packet is wholly, American, or partly obtained in Germany, as labelling might indicate, but from the appearance it may be infe that it represents two collections, both from this country. *2906. 2. P. striola, Lk. n. 2, on various Cyperacee and grasses. Beth Represented by the original packet containing a dozen or short pieces, 1-6 cm. long, of a Juncus, probably J? effusus, b PAPERS GIVING RUSTS OF NORTH AMERICA. 241 uredinia and telia, the spores being those of Uromyces Junci-effusi Syd. The packet is labelled “2 Puccinia Striola Beth,” the “2” in- - dicating that the original collection had been divided into numbered portions, of which no. 2 only had been retained. - 2007. 3. P. Arundinariz, L.v.S., Syn. Car. 487, Lk. p. 68 in a note. Very “ good species, also on Miegia (Arundinaria) cultivated in the Bartram Gardens, Philadelphia. (487. 2. [Puccinia] Arundinariz Sz. P. rather large, elevated, pulvinate (not surrounded by the epi- dermis), blackish-brown, spores oblong, bilocular, pedicel long. _ Rather rare on leaves of Arundinaria. Of the size of a mouse dropping, beautifully scattered over the leaves. Cells of the spores equal to each other, color under a lens yellow, pedicels longer than the spore, radiately divergent, white, pellucid.) _ Represented by an original packet containing a part of a leaf, 1 ; by 5 cm., which bears three telial sori in a row, two being empty of spores. The single sorus with spores is prominent, oblong, and dark rown or blackish. The. packet is labelled “ Puccinia Arundinarie LvS Salem.” __ The rust still bears the name given it by Schweinitz. Its ecial orm has not yet been discovered. *2908. 4. P. punctum, Lk. n. 3, on Carex and Scirpus, Bethlehem. Represented by two packets, one containing Carex and the other Scirpus, both rusted, together with a duplicate packet of the latter. One packet has a dozen or more, rather soft, crumpled leaves with a few stems, all heavily rusted, labelled “ Puccinia graminis var. -hortensis Beth,” and afterward graminis crossed out and “ Punc- tum” substituted. The rust proves to be the telial stage of Puc- cinia Grossularie (Schum.) Lagerh., and on some species of Carex. _. Another packet contains twenty-five or more pieces of leaves, 3-9 cm. long, of what appears to be Scirpus cyperinus (L.) Kunth, _ abundantly rusted, labelled on the inside “ Puccinia Caricicola LvS Beth,” with “Puccinia punctum Lk” added ‘ater, and on the out- _ side “Puccinia punctum Beth in Caricibus.” The rust is that of _ Puccinia angustata Peck, being the telial stage, only a few uredinio- Spores with their two superequatorial pores being found. 242 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S ~ A duplicate packet of the last contains two pieces of s rusted leaves, about 3.5 cm. long, and is labelled “5 Paceinia tum Lk caricicola LvS.” . *2909. 5. P. Scirpi, Lk. n. 4, on various Scirpi, Bethlehem, Represented by the original packet, containing very scanty s shi of leaf blades or sheaths, some of them 3-6 cm. long, and lL. d on the inside “ Puccinia Scirpi Beth 1826,” and on the jit 2 Puccinia Scirpi Beth.” The rust is clearly the telial stage of cinia angustata Peck, and the host is doubtless Scirpus cyperinu (L.) Kunth. It is entirely different from genuine P. ei: se *2910. 6. P. Sorghi, L.v.S., frequent on the seven of Sorghaae and cultivated. P. without spots. Sori broad, difform, variously lobed, at first ered by the epidermis, at length naked but surrounded at margin, and then the epidermis lacerate. Sori often also lobed from the center—2-4 lines long and broad. Larger occur on the nerves of the leaves. Spores blackish, large, sh pedicelled. . Represented by some twenty-five pieces, I-3 cm. wide and cm. long, of leaves of Indian corn, abundantly covered with t contained in the original packet, which is labelled “ Puccinia Sor LvS Lititz,” with a later addition of “& Zez.” ee The leaves in the original packet are all without otinadie of Zea Mays L., and the rust is the one common to that host. can only surmise why Schweinitz called the rust P. Sorghi, an it was on Sorghum, a genus which has never been known to | the rust. But it would seem from the labelling of the pack Schweinitz thought at first he had to do only with a Sorghum and afterward found it was certainly on Zea, so assumed that was on both kinds of hosts. am Because of the inappropriateness of the specific name, s taxonomists have adopted some other name, but most autho still use Schweinitz’s original name on the ground of priority. alternate stage has been found by cultures to occur on spe Oxalis. PAPERS GIVING RUSTS OF NORTH AMERICA. 243 *2011. 7. P. Andropogi, L.v.S., very frequent in autumn on leaves and culms also sheaths of various species of Andropogon, Bethle- F: oa obscure, sori densely aggregated, elevated, fuscous, obtuse, linear, short. Spores fuscous. Although not confluent, yet occupying almost the whole leaf. Represented by an original packet containing four or more stems and many leaves in pieces 7-10 cm. long, bearing an abundance of -telia, labelled “ Puccinia in Andropogi LvS.” The host is undoubt- edly Andropogon scoparius Michx., and the rust still bears Schwei- _ nitz’s name, although generally written P. Andropogonis. _ The two methods of writing the specific name indicate a differ- _ ence in the method of forming the genitive of this and similar Latin- ized Greek words, common among classical writers of the very early _ as well as more modern times. The longer form is now generally : am 8 P. emaculata, L.v.S., here and there on leaves of Panicum, espe- cially Panicum pubescens in fields, Bethlehem and Philadelphia. P. entirely without spots; at first the sori are all covered, rather few, sparse, erumpent; later often confluent, minute, short, nar- row, parallel, mostly acuminate at both ends. Spores very dark, rather small; immersed in water, brownish. Represented by an original packet containing five fragments of _ grass leaves, I-2 cm. wide by 2-10 cm. long, with a scanty showing of telia. The packet is labelled “ Puccinia emaculata LvS in Pan- ico pubes. Bart Gard.” The leaves are somewhat pubescent and _ considerably weathered. They can scarcely be the leaves of Pant- cum pubescens Lam., but rather are those of the more widely dif- fused P. capillare, judging from the soft pubescence, and from the general association of the rust. A portion of the Schweinitz col- lection has been seen by Prof. A. S. Hitchcock and by Mrs. Agnes Chase, the eminent agrostologists of Washington, D. C., who pro- nounce the host to be P. capillare. *2913. 9. P. Junci, L.v.S., on culms of J. effusus, Bethlehem, frequent. P. scarcely with any spots; sori irregular, erumpent, somewhat cov- ered by the epidermis, rather broad, applanate. Spores large, _ blackish brown. _ Represented by an original packet, containing three pieces, 5-7 244 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S cm. long, from a terete stem split lengthwise bearing a few sot It is labelled “ Puccinia Junci LvS in J. effuso Beth.” The teli spores are one-celled, and together with the few urediniospe seem to agree with later collections on the same host, now cz Uromyces Junci-effusi Syd. *2914. 10. P. Windsorie, L.v.S., on leaves of Windsoria (Poa) quin dentata, Bethlehem. . P. spots yellowish, at length evanescent. Sori long, linear, un prominent, erumpent from the epidermis, not confluent. Sp compact, from purple to dark fuscous, long pedicellate. Represented by an original packet containing ample mat i consisting of parts of nine leaves, 3-10 cm. long, and four stems, 7- cm. long, well covered with telia. It is saheltes “Puccinia W sorie LvS in culm & fol Poe quinquedent Beth. a The rust still bears the name given to it by Schweinitz, but | has been impossible to trace the origin of the name of the host. — such specific name is known under Poa or Windsoria. ‘Professor A. S. Hitchcock has suggested that it was a slip intended for ¢ quifida, a specific name used under Poa by Pursh, but never trans- ferred to Wéindsoria. Neither name is given in Muhlenberg “Catalogue,” but he does have Poa seslerioides Michx. es L.), which is clearly the host in — now called Tridens f (L.) Hitche. *2415. 11. P. Zizanize, L.v.S., on the fallen leaves of Zizania. Kaign’s Po near Philadelphia. 5 P. without spots, minute, at first covered, at length linearly erump the epidermis persistent about the margin of the sori; sori e gate, abbreviate, dark, held to the light somewhat Spores loose, usually scattered about, short pedicelled, de ‘not much smaller than in related species. Represented by an original packet containing two very st shreds of much weathered leaves 1-1.5 cm. long, bearing a few te sori. It is labelled “ Puccinia Zizanie LvS Kaines Pt.” The fragments remaining of this collection are so very sca that it seemed at first that no certain conclusion could be reached s to the identity of either host or fungus. The slightly reddish tint, PAPERS GIVING RUSTS OF NORTH AMERICA. 245 the character of the surface, the veining, and the rough edges show it these leaves could not have been those of Zizania. They do _ suggest Andropogon, however, and in spite of being weathered, they match well the leaves of A. scoparius and A. virginicus. Moreover, _ the teliospores, as well as a few urediniospores seen, agree fully with _Puccinia Andropogonis, n. 2911. While the two Andropogons named can not be told apart by their leaves, we probably have to do with A. virginicus which occurs on damp soil about Philadelphia. 2916. 12. P. Smilacis, L.v.S., Syn. Car. 494, also in Pennsylvania. i (494. 9. [Puccinia] Smilacis Sz. P. rather large, confluent, difform and stellate, dark fuscous, on Smilax rotundifolia occupying all of the somewhat dried leaves.) Represented by an empty packet labelled “Puccinia Smilacis LvS Salem.” _ There is no doubt that this number is based upon the telia of the common southern Smilax rust, still called Puccinia Smilacis Schw. 2917. 13. P. Polygonorum Lk. n. 6, Syn. Car. 488, on P. pennsylvanicum and P. virginicum, also Pennsylvania. (488. 3. [Puccinia] Polygoni Pensilvanici Sz. P. rather small, aggregated, somewhat elevated, chestnut brown, opaque, at first closed, seated on. pale spots, spores obovate-truncate. Frequent on Polygonum Pensylvanicum; rendering the plants sterile. Spores bilocular, pedicel short; cells almost broader than long, fuscous under a lens.) Represented by a packet containing two leaves, one about 3.5 by 5 cm., and the other somewhat smaller, bearing a few sori, which is labelled “ Puccinia Polygonorum P. virginice LvS Salem & Beth.” | The leaves are ovate-lanceolate, smooth with ciliate margins, and doubtless belong to Tovara virginiana (L.) Raf. (Polygonum virginianum L.). The other host named was also correctly deter- mined, without question. The rust is now given the earliest name for it, P. Polygoni-amphibii Pers. _ *2918. 14. P. concentrica, L.v.S., very frequent toward the end of autumn on half alive and dead leaves of P. coccineum, Bethlehem. P. spots very large, confluent, bright red on upper surface, paler on the lower. Sori very crowded, aggregately concentric, at first 246 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S i somewhat compact, black-shining, at length the spores loo minute, fuscous black. Possibly it is P. Polygoni-amp! Candolle—but very certainly specifically distinct fro ceding. Represented by a packet containing about ten fragments leaves, some nearly complete, showing an abundance of telia, in par circinating about single uredinia. The packet is labelled “ Puccis Polygoni coccinei concentricum omnino differt a P. Pore virg et aliis Bethl.” The leaves are broadly lanceolate, about 4 by 10 cm., and with those of a phanerogamic specimen at the Philadelphia Aca of Sciences, collected by Schweinitz, locality not given, and lz by him Polygonum coccineum, which is now determined to emersum (Michx.) Britt. The rust is Puccinia Polygoni-am as thought likely by Schweinitz, and differs from the eS di: species only as influenced by the host. #2919. 15. P. bullata, L.v.S., Syn. Car. 501, Lk. n. 8 In Pennsylvania, fou very large, two to even three inches, especially on stems Vernonia noveboracensis. (sor. 16. [Puccinia] bullata Sz. P. very large, oblong, pulvinate, chestnut brown, axvoendee epidermis, spores very dense, oval, bilocular, long pedicelled. Abnormal, erumpent from dried stem of various plants, e Ambrosia, Chenopodium. Very large, usually an inch long and lines thick, surrounded and often covered by the epidermis of | plant. The peduncles of the spores are five times as long, oval, short, the two cells equal.) Represented by three packets. The principal packet con four sections of stem, 3.5-5 cm. long, the largest being 8 mm. diameter, and se, labelled “ Puccinia bullata LvS Salem & Bethl Caulibus variis.” The two duplicate packets, one with three, t other two, similar fragments of stem, are labelled, the first “3 cinia bullata LvS,” and the second “ 5 Puccinia bullata LvS.” of the same original collection is in the Fries Herbarium at Upsa according to Lagerheim (1. c., p. 64), who renamed the speci longipes, because the specific name had been antedated by — (Obs., 1815). PAPERS GIVING RUSTS OF NORTH AMERICA. 247 All the fragments show very large sori, reaching 3 cm. long, characteristic of the rust on Vernonia when occurring on the stems. This is undoubtedly the same rust as the leaf form, recorded under no. 2926, as P. Vernonie, a name still generally applied to this rust. The leaf form has been grown by sowing spores from the large stem sori. The asterisk before this number is a typographical error. ‘aa 16. P. Pycnanthemi, L.v.S., rather related to P. Clinopodii, frequent on P. incanum, Bethlehem. P. spots purple, minute, persistent. Sori small, fuscous. Spores loose, long pedicelled. Represented by an empty packet, labelled on the inside “ Ce2oma i(redo) Labiatarum in Pycnanth glauci fol Beth,” with Uredo -erossed out and “ Puccinia” substituted, and on the outside “ Puc- cinia Pycnanthemi LvS in Pyc incano Beth.” _ The host can be accepted as correctly named, and the rust as identical with Puccinia Menthe Pers. #2921. 17. P. compositarum, Lk. n. 19, common, Bethlehem, especially on the stems and leaves of dead Cnicus or Cirsium (P. caulincola). Represented by an original packet containing six sections of weathered stems about 5 cm. long, the largest being 5 mm. in diam- eter, and all bearing telia. The packet is labelled “ Puccinia caulin- cola vere in caulib. Cnici altissimi,” with “compositarum” after- ward written in. __ The cobwebby hairs on these stems indicate that they are thistles, and there is every reason to believe that they belong to Cirsium altissimum (L.) Spreng. (Cnicus altissimus Willd.) as labelled by _Schweinitz. The rust agrees with Puccinia Cirsii Lasch. The ref- _ erence to “ P. caulincola” undoubtedly indicates the author’s opinion _ that his material might possibly be referred to the European Ceoma _ caulincola Nees, which was originally found on stems of Centaurea _ paniculata (Syst. Pilze, 16, 1816). By later authors the specific name was transferred to Puccinia and applied to other forms. 248 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S *2922. 18. P. maculosa, [L.v.S., not] Straus[s]. Bethlehem, here and there on leaves of Prenanthes or Hieracium. Entirely distinet fror the preceding by the broad, white spots, spores much paler. Represented only by an empty packet labelled “ Puccinia me losa LvS in fol. Hieracii.” There is a specimen, however, in the Michener Collection at Washington, consisting of a glabrous, pal a green leaf, a little more than 4 cm. long and 2 cm. wide, bea ing four groups of telia, labelled “ 2922-18—Syn. Car. Puccinia macu- losa Strau. in foliis Hieracii Beth. ex Herb. Schw.” There is also a tae representation in the Herb. Curtis at Harvard Univers “Prenanthis aut Hieracii,” a small portion of which, through Pr kindness of Dr. W. G. ae the writers have been able to examine. Both host and fungus from these two sources agree fectly with the material published as 1855, Ellis & Everhart, “ North American Fungi,” on Cynthia virginica from Illinois, 1882, A, Seymour, and as 3413, Rabenhorst-Winter, “ Fungi Europzi,” Krigia virginica (Cynthia virginica) from Missouri, 1885, C. Demetrio. A good description of the rust was given by Burrill in his “ Parasitic Fungi of Illinois,” p. 188. It is evident that Schwe nitz was very uncertain about the name of the host as he calls Hieracium on packets, and “ Prenanthes or Hieracium” in the lished account, and quite naturally so, if we consider it to be Krig or Cynthia virginica, now called Adopogon virginicus (L.) Kunt: for that plant has the aspect when growing that might well cz it to be considered under either genus. Even Muhlenberg mt have been uncertain about it, as his catalogue either does not m tion it, or merges it with some other species, although a comme plant of the flora. ee Strauss gave the name Uredo maculosa (Ann. Wett. Ges. 2:1¢ 1810) to a European rust on Prenanthes purpurea, apparently im cluding both uredinia and telia, with which no rust in America been identified. The rust found by Schweinitz is a short-cycle for not known in Europe. The specific name maculosa, under the genus Puccinia, is, therefore, to be credited to Schweinitz. - 2923. 19. P. Helianthorum, L.v.S., Syn. Car. 495, Lk. p. 74, clearly distin frequent on various Helianthi, and in Pennsylvania best develope on dead leaves. On cultivated H. tuberosus, it occupies the low surface of almost all leaves. PAPERS GIVING RUSTS OF NORTH AMERICA. 249 (495. 10. [Puccinia] Helianthi Sz. P. rather small, orbicular, aggregated, black, spores globoid- oval, bilocular, very long pedicelled. Common on many Helianthi—Spores fuscous yellow, pedicel white, pellucid.) _ Represented by an original packet containing twenty or more fragmentary leaves, I-4 cm. wide by 6-10 cm. long, and a leafy stem, 5 cm. long, bearing one mature flower head, the leaves richly supplied with telia. The packet is labelled “Puccinia Helian- _ thorum LvS 18 _ The leaves of this collection are lanceolate or ovate-lanceolate, -and probably came from the upper part of the plant. Examination of the leaves together with the flowering head makes it certain that the host is Helianthus tuberosus L., and the date, “1826,” shows that the collection was made in Pennsylvania, doubtless at Bethle- hem. The rust still generally goes by the name first given by Schweinitz, P. Helianthi, although his specific name for the «cial _ stage (no. 2871) has priority of place in the same publication and _ technically should replace it as P. Helianthi-mollis. Schweinitz evidently inserted “clearly distinct” under this entry, and similar expressions in the following and other entries to em- phasize his dissent from Link’s opinion (1. c.) that the species might _ be the same as the European P. Syngenesarum Link. 2924. 20. P. Heliopsidis, L.v.S., Syn. Car. 493, Lk. p. 74, and Pennsylvania— entirely distinct. (493. 8. [Puccinia] Heliopsidis Sz. P. rather irregular, aggregated, surrounded by the epidermis, chestnut brown, spores oval, elongate, long pedicelled, bilocular. Frequent on dried leaves of Heliopsis, also on Vernonia.—Cells of the spores equal, septum situated exactly in the middle of the spore.) Represented only by an empty packet, labelled “ Puccinia Heliop- sidis LvS.” The rust on Heliopsis is still known by the name given to it by Schweinitz. It has only been found on H. helianthoides (L.) weet. Although given as “ frequent,” yet it is represented by only _ five collections in the Arthur herbarium, all from the Mississippi 250 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S — region, three with ecia, one with uredinia, and only one r no ; telia. The species must be local, although widespread. 2925. 21. P. Verbesine, L.v.S., Syn. Car. 496, Lk. p. 74, not yet in Penn: vania—good species. (496. 11. [Puccinia] Verbesine Sz. P. punctiform, sparse, fuscous black, spores ovate, pedicel short. On flourishing leaves of Verbesina, Sigesbeckia (Richweed) Spores narrower at the apex than at the base, bilocular: cells Not surrounded by the epidermis. ) ; = Represented by an empty packet, which is labelled on ihe insic “Diczeoma Verbesine Salem,” and on the outside “ Puccinia, Ve besine LvS Salem.” | Schweinitz’s name still holds good for the Verbesians rust reel : region he explored. It most likely does not occur on Siegesbe 7 (richweed ), on which no rust has yet been found. *2926. 22. P. Vernoniz, L.v.S., very common on Vernonia, Bethlehem. P. without spots. By the rather pulvinate sori and by the beat rusty color of the spores it differs from P. Helianthi,’ It: ®) also occasionally on Helianthus. Represented by an empty packet, labelled “ Puccinia Ver 7 LvS in Heliant ferrugin.” a This is without doubt the leaf form of the common Ve rust, the stem form of which Schweinitz had already named bullata (no. 2919). The rust does not occur on Helianthus though occasionally the Helianthus rust simulates the one on nonia. 2927. 23. P. Xanthii, L.v.S., Syn. Car. 500, Lk. n. 23. Also frequent leaves of Xanthium in Pennsylvania. Beautiful and consp from a distance. Sori usually concentric and aggregated center of the spot. 2 (500. 15. [Puccinia] Xanthii Sz. P. spots delicate, orbicular, pale, beneath fuscous brown pale margin, spores oblong, bilocular, pedicellate. On lower surface of the leaves of Xanthium strumari sandy places. Beneath it shows at first pale vesicles resembling t cells of the leaf, these being broken and encircled by the epi the spores appear in a coherent fuscous pustule, yellow under a the pedicel longer than the spore.) PAPERS GIVING RUSTS OF NORTH AMERICA. 251 _ Represented by an original packet, containing part of one leaf . bout 3 by 6 cm., bearing many groups of telia, which is labelled -“Puccinia Xanthii LvS Sal & Beth.” A very common short-cycle rust still designated by Schweinitz’s *2928. 24. P. Helenii, L.v.S., rather rare, but prominent, on leaves of Helenium autumnale, Bethlehem. P. spots golden yellow, expanded, sori pulvinate, sparse, and close to each other, convex, at first brown, later pies chestnut. Spores rather large, compact. Represented by a packet containing the tip of a stem, about 2 em. long, with six small, sessile leaves attached, together with parts _of three maturer, lanceolate leaves, the largest 2.5 cm. broad and 7 em. or more long. The packet is labelled “ Puccinia Heleniit LvS Bethl.” An empty duplicate packet is labelled “2 Puccinia Helenii 5.” A careful study of this material leaves little doubt that the host is Aster salicifoliuns Lam., and that the rust is the common Puccinia Asteris Duby. The leaves of Helenium have a peculiar lower sur- _ face due to a sparse pubescence, quite unlike the smooth lower sur- _ face of the material in the packet, or of Aster salicifolius and of similar lanceolate-leaved species of the genus Aster. From an orig- _ inal specimen in the Fries Herbarium at Upsala, Lagerheim (Tromsé _ Mus. Aarsh., 17:60, 1894) has given detailed characters as a good _ Species, not remarking any error in the host. Even if the host had _ been Helenium, yet the rust would undoubtedly have proven to be the same species that occurs on Aster, judging by the description given and the relationship and characteristics of the hosts. *2929. 25. P. Silphii, L.v.S., sent from Carolina, on leaves of S. trifoliatum, | by my friend Denke. P. spots rather small, purple. Sori thick, pulvinate, confluent, aggre- gated, black. Spores compact, concolorous. Represented by four small fragments of leaf, the smallest one, 5 cm. long, bearing a group of telia. The packet is labelled “ Puc- cinia Silphii LvS in Sylph trifoliat Denke.” 252 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S” Both rust and host appear identical with those ne go under the same names at the present time. *2930. 26. P. Asteris, L.v.S., a handsome species, sreauene on leaves of paniculatus, Bethlehem, P. spots flattened, bullate, yellow, not widely expanded. Sori » dense, subconcentrally placed, beautifully fuscous. mini! loose, concolorous. Represented by a packet shabainhas a short stem with hie attached and also by ten much crumpled, similar, ovate-lane leaves, all with long, slender petioles, and all sparsely bearing tel The packet is labelled “Coma (Ur) Asterum LyS Bethl in A paniculat,” with the first two words crossed out and “ Puccinia substituted. | _ The leaves are doubtless Aster cordifolius L., and the rust short-cycle form first given the name Puccinia Asteris by Duby 1830, two years before the Schweinitz name was published. D less the early collection on Aster paniculatus, this being its common host, was entirely given away, leaving only a later col on A. cordifolius. *2931. 27. P. Kuhniz, L.v.S., common on the leaves of Kuhnia, EB P. without any spots. Sori amphigenous, pulvinate, Sonesta § gated, blackish brown. Spores rather large, loose, long celled. A Phragmidium? Represented by an original packet labelled on the inside “Ur Kuhniz in K. eupator Bethl & Salem,” with “Uredo” crosse. and “ Puccinia ” substituted, and on the outside “ Puccinia LvS Beth.” The packet contains a tiny fragment, 2 by 3% bearing a few large telial sori. The peculiar glands and hairs : the host unmistakable, and the amphigenous sori with their elli teliospores fully justify the record. The rust is not common eastward, Schweinitz’ s record being only one known to the writers east of Wisconsin and although in the middle west, especially between Illinois foothills of Colorado, itis not infrequent. Inthe Carolina list is mentioned as host for a rust (see no. 2844), and the earli on the packet reads ‘‘ Bethl & Salem,” but the packet was pro PAPERS GIVING RUSTS OF NORTH AMERICA. 253 not labelled until after Schweinitz became a resident of Pennsyl- -vania. We may safely assume that the packet with its fragment represents a collection made at Bethlehem, Pa. Unless this were true the asterisk before the number would have to be considered erroneous, and the omission of “Syn. Car. 478” unintentional. _ Furthermore, if Salem were the place where the collection was made, the record would have been Salem & Bethl., as may be seen under nos. 2832, 2846, 2875, 2888, 2917, 2919, 2927, etc. Some ob- servation at Salem may have been in mind, but with no specimen _ preserved. _ *2932. 28. P. investita, L.v.S., frequent, observed with AXcidium gnaphali- tatum on the tomentose leaves of Gnaphalium polycephalum, Bethlehem. Always hidden by the tomentum. P. without spots; sori minute, sparse, roundish, very black, scarcely showing at first through the tomentum, sometimes aggregated- confluent. Spores compact, very dark. Surface of the sori as if furrowed. Represented only by an empty packet labelled “ Puccinia inves- tita LvS in avers pag Gnaphalii polycephali cum AEcidio vulgari Beth.” There is no reason to doubt that this record applies to the rust still passing under Schweinitz’s name, P. investita, which pos- sesses ecia and telia, and is identical with no. 2873, and now better called P. gnaphaliata (Schw.) Arth. & Bisby. _ 2033. 29. P. Galii, L.v.S., Syn. Car. 499, Lk. p. 76, a rare species but suffi- ciently distinct—not a Sclerotium. (499. 14. [Puccinia] Galii Sz. (near Sclerotium). P. erumpent, globose-ovate, dark fuscous, spores clavate, biloc- ular, short pedicelled. On living leaves of Galium purpureum, but more perfect on dead ones, then a line long—Tubercle-like it pushes up the epidermis, which surrounds it. Spores a little darker in color than those of Puccinia graminis.) tea ie i Sper eats sat a pen gs Se ee ies i : Bee: Represented only by an empty packet, labelled on the inside 4 “Diczeoma Galii Salem,” and on the outside “ Puccinia Galii LvS Salem.” There is no specimen of this number at Philadelphia, or in the Herb. Curtis at Harvard University or in the Michener Col- lection at Washington. PROC. AMER. PHIL. SOC., VOL. LVII, R, JULY 17, 1918. 254 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ'S — It is particularly unfortunate that no specimen is available substantiate the record, as there is no other record of a Galiwm having been collected in North Carolina. The most southern lection in the Arthur herbarium is that of Puccinia punctata I on G. triflorum made by Mrs. Emily Arthur at Salt Sulphur § W. Va., in 1914, a locality on the opposite slope of the Alleg! Mountains to the northwest from Salem, N. Car. There is no ¢ record than Schweinitz’s of a rust on G. purpureum Walt., now ferred to G. pilosum Ait., although it would not be an unlikely ne for P. punctata. The bubartie. bie emergence of the sori and thei distinctly blackish color are sufficiently characteristic of the telia P. punctata Link to make it probable that Schweinitz had this sp in hand, although it must be a rare fungus in the Carolina flora i *2934. 30. P. Myrrhis, L.v.S., on leaves and stems of Myrrhis procumbe Bethlehem. ; P. without spots. Sori dense, minute, surrounded by the rupt epidermis, pulvinately applanate. Spores very loose, g brown. Ses Represented by a scanty specimen, consisting of numerous ver small fragments showing a few pale telia, in the original pa which is labelled “ Puccinia Myrrhis procumb LvS Beth.” — Both urediniospores and teliospores are present in the sp pecim r and are characteristic for the species now called Puccinia Pi le (Str.) Mart. The host is clearly as named by Schweinitz C. phyllum procumbens ae. Crantz (Myrrhis procumbens =P0 € *2935. 31. P. Bullaria, Lk. n. 32, on stems of Hyssopus veges Bethlehem. Represented only by an empty packet labelled “ Puccinia cola in Hyssop. nepet. Bet,”’ with second word crossed out and " laria” written in. There is a specimen in the Michener Collection at Washin exactly answering the requirements of this number. It is a piec smooth stem 4 cm. long, split lengthwise and originally 3 mn diameter. Protruding from a longitudinal fissure 2 cm. in length a fungus-like growth, brown and bullate, that may be the PAPERS GIVING RUSTS OF NORTH AMERICA. 255 stage of some ascomycete but is certainly not a rust. No spores e found. As the host is a labiate and not an umbellifer, Link’s mame could not in any case be applicable. wa6 32. P. anemones, Lk. n. 33, very rare on leaves of A. quinquefolia, but most distinct, Bethlehem. ___ Represented only by an empty packet, which is labelled inside ; “ Diczeoma aguienase Deetw,” with the later addition above of “ Puc- cinia anemones,” and on the outside “2 Puccinia Anemones Beth _ Deetwyler,” and also a word preceding the last one which is not _ wholly legible. _ There is practically no doubt that this number covers the rust : on the host as stated, now called Polythelis fusca (Pers.) Arth. 2037. 33. P. solida, L.v.S., Syn. Car. 486. [as P.] Anem. virginian, frequent on leaves of Anemone virginiana, Salem and Bethlehem. P. without spots. Sori sparse, rather large, so very compact that they appear solid, black. Spores at length somewhat loosened. Sori dispersed over the whole leaf, at first yellow and more or less impressed. (486. 1. [Puccinia] Anemones Virginian Sz. P. punctiform, sparse, chestnut brown, spores clavate, at- tenuate into a short pedicel, bilocular. Spores under the lens yellowish-white; they pass into the pedicel so that it is not possible to distinguish where they begin.) Represented by an empty packet, labelled ‘“‘ Puccinia solida LvS in Anem. Vir.” A widespread and well-known species, still bearing the earlier _ name here given. 2938. 34. P. circee, Lk. 43, Syn. Car. 491, common, and Bethlehem. : (491. 6. [Puccinia] Circee, frequent on leaves of Circza.) Represented by a packet containing parts of three leaves, the best preserved being about 3 by 5 cm., and labelled “2 Puccinia Circeze Germ & B & S,” with cancellation marks across “Germ.” Two of the leaves are faded and pressed smooth, the third is nat- ural green and crumpled by drying. The rust and host, undoubt- a edly C. Lutetiana, are common and widespread, although no other - collection of the rust is yet known so far south. 256 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S ~ 2939. 35. P. aculeata, L.v.S., Syn. Car. 489. P. podophylli, likewise cc mon on Podophyllum [in Pennsylvania]. Very distin’. on, Ay count of the aculeate spores. ae (489. 4. [Puccinia] Podophylli Sz. “ P. rather large, subconcentric, chestnut black on die spots, spores oblong, bilocular, aculeate. Here and there on leaves of Podophyllum.—Spores oval ender a lens yellowish, the points prominent, straight. Pedicel not cinta, very short.) Represented only by an empty packet, labelled “ Puccini seaeat LvS in Podoph Sal & B.” Owing to Schweinitz’s slip of the pen in calling the spores of cidium Podophylli (no. 2888) “bilocular,” Link transferred that form to the genus Puccinia, which necessitated a new name for the present form, so he made a descriptive name from a very distine- tive character (1. c., p. 79). Schweinitz adopts the name, but evi- dently considers himself responsible for the species, and, as in other such cases, does not cite Link’s work. Schweinitz’s earlier name ‘ still in use for this rust. 2940. 36. P. Lespedeze procumbentis, L.v.S., Syn Car. 497, Lk. p. 83, extra- ordinary species, and in Pennsylvania. ; (497. 12. [Puccinia] Lespedeze procumbentis Sz. P. rather small, subpunctiform, sparse, somewhat — ‘tesco erumpent, spores oblong, bilocular. Here and there on leaves of Lespedeza procumbens.—It lifts tl epidermis of the lower surface of the leaf into blisters, which tured are white, pellucid. Spores with septum situated oe the middle of the spore, and the pedicel (white, rather long) is tinct from the spore.) Represented only by an empty packet, labelled “ « Pace es pedeze procumbent LvS Salem.” : The senior author in his first publication on the subject of » (Amer. Nat., Jan., 1883, pp. 77-78) pointed out that doubth Schweinitz was led into the error of describing the spores as locular by looking at the dry spores under a magnification of abe seventy-five diameters. At any rate the microscopic details” Schweinitz’s description can be attested in this way. The gr thickened wall at the upper part of the teliospore, often equal half the spore’s length, under these conditions takes on the appe: PAPERS GIVING RUSTS OF NORTH AMERICA. 257 ance of an upper cell separated from the lower by a transverse tum. In reality the teliospores are one-celled, and the rust be- : longs under the genus Uromyces, as U. LAPEREEE LVF EMN OS sly M. A. Curt. : 2040 37. P. Lespedeze violacezx, L.v.S., Syn. Car. 498, Lk. p. 83, much more bs frequent on L. violacea than on L. palystachya, also in New Jersey. (498. 13. [Puccinia] Lespedeze polystachye Sz. P. rather small, punctiform, surrounded by the epidermis, black shining, spores oblong, attenuate at both ends, somewhat bilocular. Frequent on the lower surface of the leaves——Surrounded by the epidermis. Spores more elongate and attenuate into the pedicel, septum scarcely visible, it appears vaguely now near the apex of the spore, again lower. Color of the spores, under a lens, yellow.) Represented by neither specimen nor original packet. The rust is an abundant one, and is considered by all recent mycologists to be identical with the preceding, Uromyces Lespedeze-procumbentis (Schw.) Curt., and to be both on L. hirta (L.) Hornem. (L. poly- stachya Michx.) and L. violacea (L.) Pers. "The elaborate but elusive description of this species, when taken in connection with that of the preceding number, illustrates the im- perfect equipment possessed by Schweinitz and others of his time for the study of microfungi, and leaves us astonished at the large measure of success attained. The present number also illustrates the futility of long and cumbersome specific names for correctly designating a species. Before a decade had passed Schweinitz said that the rust which he specifically limited to Lespedeza polystachya was found by him “much more frequent on L. violacea.”’ How _ Much better it would have been to have designated this rust as P. __ affinis, or by some such simple appellation, and avoided bestowing a _ mame that would be burdensome to other mycologists. ; to. 38. P. Phaseoli trilobi, L.v.S., on leaves of P. trilobus sent from New York. Appears related to P. fabe. P. sori minute, hypophyllous, partly covered by the epidermis. Spores black, spots none. Represented by an empty packet, which is labelled “ Puccinia Phaseoli Newyk in Phaseolo trilobo.” There appears to be no 258 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZz’S _ doubt that this number belongs under Uromyces appen i (Pers.) Fries, and on Strophostyles helvola (L.) Britton ¢ lobus Michx.). *2043. 39. P. Fabe, Lk. n. 45, on leaves of V. faba, Nazareth. ‘Represented by neither specimen nor packet. Doubtless th rust was Uromyces Fabe (Pers.) DeBary, and on the host r at that time a plant more often cultivated in America than present. *2944. 40. P. Hyssopi, L.v.S., on leaves of H. scrophularizefolius, B occasionally. P. spots yellowish, effuse. Sori aggregated, compact, fied: what circinate and undulately confluent with each other, first blackish, small but occurring copiously upon sa le Spores fuscous, at length rather lax. Represented by a packet, containing a stem, 4 cm. long, wi opposite and petioled leaves attached, and by parts of three o leaves, the largest being 2.5 cm. wide, bearing many groups of 1 The packet is labelled “ Puccinia Hyssopi scrophul LvS tie, The host is now placed under Agastache, as A. scrophulari (Willd.) Kuntze, and the rust is identical with P._ (Schultz) Link. *2945. 41. P. Potentille, L.v.S., not Phragmidium, Lk., on mature of P. canadensis, on lower surface, Bethlehem. : P. sori minute. Spores fuscous, at length black, ec pedicelled. Spots almost disappearing. Represented by an empty packet, labelled “ Puccinia Pote with one other word, not deciphered. The rust is undoubtedly the one often called Phragmidium tentille-canadensis Diet. It was transferred to the genus Kueh by the senior author some time since, and again very recenth the genus Frommea, under which it is F. obtusa (Strauss) A #2946. 42. P. Ari triphylli, L.v.S., on lower surface of the leaves triphyllum, Bethlehem. P. spots pale, very broad, on the margins of the leaves. Sori PAPERS GIVING RUSTS OF NORTH AMERICA. 259 often confluent, at first covered by the epidermis, soon ruptured. Spores brown fuscous, loosely attached and Uredo-like but nevertheless a true Puccinia. Represented by a packet containing two large leaflets, 8 by 15 cm., bearing several loose groups of telia, and labelled “ Puccinia Ari triphylli Mauch Chunk.” _ This number is now called Uromyces Caladii (Schw.) Farl., and on Arisema triphyllum (L.) Schott (Arum triphyllum L.), other stages of the life cycle being listed under nos. 2839, 2860 and 1861. Genus 213. PHRAGMIDIUM. Ii is worthy of note that I have never met with a Phragmidium in America on the leaves of Rosa or Rubus, but the following very common species without doubt belongs here. 2047. 1. P. Hedysari, L.v.S., Syn. Car. 503, frequently occurs on leaves 3 of H. paniculatum and others, Bethlehem and Salem. P. sori minute but thickly scattered over the whole leaf, resting upon the epidermis. Spores long pedicelled, pedicel articulate, pellucid, remainder opaque, ovate, obtuse, not cylindric, ob- scurely septate, not constricted at the articulations, fuscous black. (503. 18. [Puccinia] Hedysari paniculati Sz. P. punctiform, sparse, fuscous, spores ovate-globose, fus- cous, pedicel very long, filiform, pellucid. Frequent on the under face of the leaves of Hedysarum paniculatum. I see no septum in the spore. Pedicel filiform, pellucid.) Represented only by an empty packet, labelled on the inside “Diceoma Hedysari paniculat Salem,” and on the outside “ Puc- cinia Hedysari panic Salem.” If the genus Uromyces had been in use at the time, Schweinitz _ undoubtedly would have placed this species under it, certainly at first, for he says he could see no septum. What his idea of the genus Phragmidium was, it is now difficult to say, but the senior author has explained in the paper referred to under no. 2940, that when the teliospores are seen dry under low magnification “the pedicels being delicate cylinders collapse and twist like a ribbon, and what appear to be three or four joints in each pedicel are very dis- 260 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S tinctly shown” A. c.), hence Schweinitz says “ pedicel artic The earlier specific name is still retained, the species being Ure Hedysari-paniculati (Schw.) Farl., and on Meibomia pani (L.) Kuntze (Hedysarum paniculatum L., Desmodium paniculatt DC.). age Note.—The genera numbered 214-246 include the rae Series I and all of Series I-IV. Under Series IV, ay d 1 the following species belong with the rusts. oF Genus 241. SEIRIDIUM. *3084. 1. S. marginatum, Lk. p. 126, n. 1. Our plant, very common on corymbosa growing in inundated places, agrees exactly Nees’s illustration and description. But it is not the same size; for usually it occurs on branches, living or half alive, y huge sori, very thick, two inches, encircling the branches, often many sori joined together. Spores so large that they clearly visible to the naked eye, or at least through a se very low power. : Represented by a mounted specimen, conatetiue of three stems, 4-6 cm. long and 6 mm. thick, well provided with 1: blackish sori. No original packet was to be found. The rust is cle the very distinctive Earlea speciosa (Fries) Arth. (Phragmie speciosum Fries), on Rosa carolina L. (R. corymbosa Ehrh pauciflora Muhl.). . *3085. 2. S. Similacis [typographical error for Smilacis], L.v.S., here ; there erumpent from the stems of Smilax caduca and o' species, Bethlehem. = S. sori very long, confluent, yet much smaller, and not so th in the preceding species]. Spores cylindric, dark fuss pedicels very long, contorted, white. Represented by neither specimen nor packet at Philadelphia, the Michener Collection at Washington, there are two stems, 5 cm. long by 5 mm. in diameter, and the other 4 cm. long and 3° in diameter, with many long weak prickles and well covered masses of telia. They are mounted and are labelled “Schw. | Seiridium obtusiusculum on rosa, Smilacis Beth. ex Herb. but without number. PAPERS GIVING RUSTS OF NORTH AMERICA. 261 _ The host is undoubtedly some species of Rosa, and may well be virginiana Mill., while the rust is undoubtedly Earlea speciosa (Fries) Arth. The appearance of this material corresponds to Schweinitz’s description. Genus 243. GyMNOSPORANGIUM. *3004. I. G. Juniperi, Lk. p- 127, n. 1. Not frequent, but very distinct from Podisoma Juniperi, found near Easton, Pennsylvania, on Juni- perus virginiana. Represented by a mounted specimen, consisting of a woody stem, Ir cm. long and 8 mm. in diameter, with a fusiform swelling from which the sori have dropped away. The stem was broken into two unequal parts before mounting. No original packet has been found. _ The rust is that of Gymnosporangium germinale (Schw.) Kern (G. clavipes C. & P.), of which the ecial form is. given under | Genus 244. PopisoMa. *3095. 1. P. Juniperi, Link, p. 127, found by me on a single Junip. Sabina in this region—copiously developed. _ Represented by a mounted specimen, consisting of a four- branched, woody stem, 5 cm. long, having a few subulate leaves each about 5 mm. long, and with a few, slender, corneous sori remaining, most of the telia having dropped out or been eaten by insects. No original packet has been found. The rust is that of Gymnosporangium clavarieforme (Jacq.) DC., and the host is most likely Juniperus communis L., being the common juniper, and not the red cedar as the name used by Schwei- nitz would seem to indicate. 3096. 2. P. macropus, L.v.S., Lk. p. 127 [error for 128], wrongly under Gymnosporangium [in earlier work]. In the parts of North Carolina best known to me a rather rare fungus. In Pennsyl- vania very common, particularly affecting Juniperus virginiana _ that has suffered by much pruning, and commonly known by the name “Cedar apple,” under which name it is offered in the market as a powerful, though imaginary, vermifuge remedy. Link expresses regret that I did not examine the structure of the underlying sporidochium. Now such things as were not dis- 262 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S cussed by me, upon this point, I gladly add here. In the place this very puzzling base ought by no means to be reg: as a sporidochium, if by this term it is intended to designate structure so called in Podisoma Juniperi. That body, whic gelatinous and composed of the interwoven stalks of the corresponds exactly with the tremellose ligules of our P. | ropus. On the other hand the basilar capitulum, the part question, is of a wholly different nature. Never, moreover, is wanting. In fact it always constitutes the first evidence for fungus; showing itself in the earliest stage on the ¢ branches of J. virginiana of the size of a rather large pin enlarging gradually, usually without altering the aff inches. Its texture when dry and old is fibrous-corky, as in F tulina but not succulent-fleshy, as if composed of fibers ra from the broadly obconic pedicel—otherwise presenting : time a somewhat woody condition. The immature capit on the other hand, may be easily cut like an apple, or even : Externally it has an epidermis-like cortex from lilac to fu purple in color, entirely juiceless like the skin of an apple. — the whole surface appear regular pits, polygonal or mo pentagonal, at first merely applanate, soon impressed and 1 nate; finally during wet weather, the cortex rupturing i center, the ligular gelatinous sporidochia an inch long are truded—bedecking all the trees during a rainy spring night | were with the richest crop of ripe oranges. If the wet continues for some days, the ligules in this condition begin dissolve. In the sunshine, however, the ligules are soon out—and they never again revive. The capitulum pe through the year. Old specimens are internally not unlike crescences of trees. Yet never can a capitulum be found wit out ligules, at least at first, nor ligules without a capitulum. is usual where trimmed juniper trees are forced artificially a pyramidal or other shape for this fungus to attack them credible abundance—but according to my observations c: made during ten years, such trees are not destroyed, ne they appear even to be harmed. There are therefore people, and not a few educated ones, who thoroughly believe fungus to be the inflorescence or genuine fruit of the ju I am thoroughly convinced by careful study that the base nothing to do with insect work. Yet it is not to be posi asserted that it is fungous. It seems to me to be a v normal growth, concerning which there is nothing more to but it should be further studied. Note.—The structure of the base of this fungus in its young before it protrudes the gelatinous ligule, accidentally omitted in its p PAPERS GIVING RUSTS OF NORTH AMERICA. 263 is as follows. The texture of the base at that time inside is like the sh of a ripe apple—if cut into slices with a knife—the color is greenish as in a green apple; oozy-cellular, apparently radiating from the stalk. ‘The green color soon changes to tawny orange—and then may be seen a few white branching fibers radiating from the stalk. As soon as the ligules are protruded on account of rainy weather, the base grows no more; but if the weather is not rainy the base enlarges day by day. The epidermis of the younger sporidochia, before their full maturity, has a somewhat filamentous- scaly texture, and the thickness of the skin of an apple. In their mature condition the ligules are covered with sporidia, just as in P. Juniperi—but _ the ligules are usually longer and not conic, often subflexuous and more attenuate toward the apex. The asterisk was probably omitted from this number by mis- take. Schweinitz evidently had many doubts about the true nature of this fungus and its generic position. In the North Carolina list he did not add “Sz.” to the name, nor did he supply a technical diagnosis, as in the case of his other new species. This may have been an accidental omission while in editorial hands, but is more likely an indication that Schweinitz hesitated regarding the best pro- (504. 1. [Gymnosporangium] Juniperi Virginiane. N. B. Wholly to be separated, I believe, from the genus Puc- cinia, and to constitute with Podisoma Juniperi, on the European Sabina, a new genus, even of this order? (that Podisoma Juniperi should be reunited with Gymnosporangium Juniperi, Nees himself affirms). In both the form and substance of the gelatinous ligule, loaded with spores, it agrees with the European fungus mentioned; but ours has a remarkable base, a thing never seen in the European. This base, as I have termed it, a somewhat corky-fleshy body, is quite like the flesh of Boletus hepaticus, even in color, and is borne on an obconic stalk, attached by its tip to the slender branches of our cedars (Juniperus Virginiana) at the very top of the trees ;— from this it changes into a hard (almost woody) capitulum, ex- panded, with incurved margins, marked with many pits, from which in wet weather are protruded the ligules, which when they have been dropped leave the pits empty. The color of the base is flesh- gray, of the fungus when fruiting and extruding the ligules strongly greenish golden, attracting the eye from a distance. Also, the capit- ulum is pendulous, and has a diameter of two to four inches. Spores covering the external surface of the ligules, linear- oblong, somewhat curved, two-celled, when again wet after drying yellowish, exactly like Nees’s illustration characterizing Gymno- sporangium.) 264 | ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S © Represented only by a mounted specimen, consisting of three galls, each about 1.5 cm. broad, one of which bears numerot jecting telia, 5 mm. long, the other two much eaten by insects. fungus is one of the best known American rusts, now ge listed under the name first given by Schweinitz. Why Link chan; the name, having no information except that supplied i in Schweinit Carolina list, and not having seen a specimen, is not Schweinitz accepts Link’s substitute name, but places the under the genus Podisoma for reasons which he states. Note.——The above account includes all numbers pertaining rusts in Schweinitz’s “Synopsis Fungorum in America Bot It also includes all numbers possibly relating to rusts, given Ecidium, Uredo, Puccinia and Gymnosporangium in his “$ sis Fungorum Caroline Superioris ”” except two. No. “460. [Uredo] confluens B., rare on softer eaves Veratrum album,” is represented by no specimen or original p: at Philadelphia, and the identity of the collection must be left u cided. No rust is known that would answer the requirement. No. “475. [Uredo] Bete, a and £, here and there on lea the garden beet and on Jpomea pandurane,” is represented specimen or original packet. The most probable suggestion to species of the Phycomycetous genus its to account doe number. = SCHWEINITZ’S UREDINALES IN SYSTEMATIC ARRANGE! (ED COLEOSPORIACE, CoLeosporiuM Ipoma@z (Schw.) Burr. (Uredo Ipomee Ceoma Ipomee Link). On Ipomea pandurata L., II, III, North Carolina, 2824. CoLEosPporIUM ELEPHANTopopis (Schw.) Thiim. (Uredo Ele podis Schw., Ceoma Elephantopodis Link). : On Elephantopus tomentosus L., II, North Carolina, 2825. (Corzosportum VERNONI# Berk. & Curt. On Vernonia noveboracensis (L.) Willd., North 2826.) PAPERS GIVING RUSTS OF NORTH AMERICA. 265 EosportuM SoLmacinis (Schw.) Thiim. (Uredo Solidaginis 3 Schw., Ceoma Solidaginis Schw.). On Solidago altissima L., 11, North Carolina, 2826. Solidago rugosa Mill., II, Pennsylvania, 2826. Solidago sempervirens L., 11, New York, 2826. _ Solidago serotina Ait., II, Pennsylvania, 2826. CoLEosPorIUM TEREBINTHINACE# (Schw.) Arth. (Uredo Terebin- thinacee Schw., Ceoma Terebinthinacee Schw.). On Silphium terebinthinaceum Jacq., 2827. OLEOSPORIUM HELIANTHI (Schw.) Arth. (Ceoma Helianthi ; Schw.). On Helianthus giganteus L., ii, 111, Pennsylvania, 2828. UrReDINACEZ (Melapsoracee). AMPSORA Mepus& Thiim. On Populus dilatata Ait. (P. italica Moench), II, sig 4 vania, 2855. _MELampsora BIGELow1 Thiim. On Salix nigra Marsh., II, Pennsylvania, 28 56. Puccrnrastrum Acrimon1£ (Schw.) Tranz. (Ceoma Agrimonie ~ Schw.). _ On Agrimonia parviflora Soland., II, North Carolina (Penn- sylvania), 2835. IASTRUM Myrtititr (Schum.) Arth. (P. minimum Arth., Uredo minima Schw., Ceoma Azalee Schw.). _ On Azalea nudiflora L., II, North Carolina, Pennsylvania, 2838. SHNEOLA Urepinis (LinK) Arth. (Phragmidium albidum . Lagerh.). On Rubus ideus L., I, Pennsylvania, 2833. 266 ARTHUR-BISBY TRANSLATION OF SCHWEINITZ’ MELAMPsoropsis Pyrota (DC.) Arth. (Ce@oma pyrolatum ZEcidium pyrolatum Schw., Chrysomyxa Pyrole On Pyrola uliginosa Torr. (P. rotundifolia Am. New York, 2803. Hyatopsora Aspipiotus (Peck) Magn. On Phegopteris Dryopteris (L.) Fee (Aspidims Muhl.), II, New York, 2836. CronaRTIUM Quercus (Brond.) Schrét. (Paviereee Peck). | On Pinus virgimicus Mill. (P. inops Sol.), I, - Penns) 2003. : PERIDERMIUM INTERMEDIUM Arth. & Kern. On Pinus sp., I, North Carolina, 2903. ZECIDIACEZ (Pucciniacee). RAVENELIA EPIPHYLLA (Schw.) Diet. (Spheria epiphylla On Cracca virginiana L. (Tephrosia virginiana Pers., Gi virginiana L.), III, North Carolina, 1474. TRANZSCHELIA PUNCTATA (Pers.) Arth. (C@oma Schw., Aicidium hepaticatum Schw., Puccinia Spinose Pers.). On Hepatica Hepatica (L.) Karst. (H. triloba Chaix, one Hepatica L.), I, Pennsylvania, 2878. PoLYTHELIs FuscCA (Pers.) Arth. (Puccinia Anemones Pers. On Anemone quinquefolia L., III, Pennsylvania, 2936. PotyTHELIs THALictrr (Chev.) Arth. (Ceoma Thai Puccinia Thalictri Chev.). “i On Thalictrum polygamum Muhl. (T. Cornuti Auet. New York, 2849. ?PHRAGMIDIUM IMITANS Arth. On Rubus ideus L., 1, Pennsylvania, 2854. PAPERS GIVING RUSTS OF NORTH AMERICA. 267 SPECIOSA (Fries) Arth. (Seiridium marginatum Schw. not Nees, S. Smilacis Schw., Phragmidium speciosum : Cooke). On Rosa carolina L. (R. corymbosa Ehrh., R. pauciflora Muhl.), I, North Carolina, 2832; III, Pennsylvania, 3084. - Rosa virginiana Mill., III, Pennsylvania, 3085. MMEA oBTUSA (Strauss) Arth. (Puccinia Potentille Schw., Phragmidium Potentille-canadensis Diet., Kuehneola ob- tusa Arth.). On Potentilla canadensis I1,, II,, North Carolina, Pennsyl- vania, 2834; III, Pennsylvania, 2045. NKELIA NITENS (Schw.) Arth. (Zcidium nitens Schw., A. lumi- ? natum Schw., Ceoma luminatum Link). On Rubus Enslenii Tratt., III, North Carolina, 2887. Rubus sp., III, Pennsylvania, 2887. YMNOSPORANGIUM MYRICATUM (Schw.) Fromme (G. Ellisti Farl., : Caeoma myricatum Schw., Zcidium myricatum Schw.). On M yrica cerifera L., 1, New York, 2894. ZYMNOSPORANGIUM JUNIPERI-VIRGINIANZ Schw. (G. macropus Link, Podisoma macropus Schw., Ceoma pyratum Schw., ZEcidium pyratum Schw.). On Malus coronaria (L.) Mill. (Pyrus coronaria L., P. angus- - tifolia Ait.), I, Pennsylvania, 2896. Malus Malus (L.) Britton (Pyrus Malus L.), I, Pennsyl- vania, 28998, 2900. : Juniperus virgimiana L., Ill, North Carolina, Pennsyl- vania, 3006. YMNOSPORANGIUM GLOBOSUM Farl. On Crategus punctata Jacq., I, Pennsylvania, 2899 a. Pyrus communis L., I, North Carolina or Pennsylvania, or both, 2900. 268 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’ GYMNOSPORANGIUM HYALINUM (Cooke) Kern (Reste |! Cooke). On Crategus viridis L. (C. arborescens Ell.), I, North lina, 28098. GYMNOSPORANGIUM TRACHYSORUM Kern. On Crategus Oxyacantha L., 1, Pennsylvania, 28097. GYMNOSPORANGIUM BOTRYAPITES (Schw.) Kern (Go vs Ellis, Ceoma botryapites Schw., Com bo Schw.). On Amelanchier canadensis (L.) Medic. Saale Bo Pers.), I, Pennsylvania, 2902. : GYMNOSPORANGIUM GERMINALE (Schw.) Kern (G. clavipes P., Ceoma germinale Schw., Periderminm ge Sew), eee On Crategus sp., I, Pennsylvania, 2904. eae Juniperus virginiana L., III, Pennsylvania, 3094. GYMNOSPORANGIUM CLAVARLEFORME (Jaeq.) DC. : ie On Juniperus communis L., III, Pennsylvania, 3095. — Uromyces Junci-EFFusi Syd. (Puccinia Junci Schw.). : On Juncus effusus L., Il, III, Pennsylvania, 2906, 2913. Uromyces CALapIt (Schw.) Farl. (4cidium Caladii Sch aroidatum Link, A. dracontionatum Schw., ee ) triphylli Schw. On Arisema triphyllum (L.) Schott (Arum ripen ’ III, Pennsylvania, 2946. ; Muricauda Dracontium (L.) Small (Arum Dracontt Arisema Dracontium Schott), I, North Carolina, 3 ‘sylvania, 2861. Peltandra virginica (L.) Kunth (Arum virginicum L. PAPERS GIVING RUSTS OF NORTH AMERICA. 269 North Carolina, Pennsylvania, 2839; I, North Carolina, 2860. : Uromyces HOUSTONIATUS (Schw.) Sheldon (Ce@oma houstoniatum = Schw., £cidium houstoniatum Schw.). On Houstonia cerulea L., 1, Pennsylvania, 2891. Unomyces Hyperici-rronpost (Schw.) Arth. (2cidium Hyperici- frondosi Schw., A. hypericatum Schw., Ceoma Hyperici Schw., C. hypericatum Link). On Hypericum prolificum L. (H. frondosum Michx.), I North Carolina, 2883. Hypericum sp., Il, North Carolina, 2842; I, Pennsylvania, 2883. Uromyces PEDATATUS (Schw.) Sheldon (U. Andropogonis Tracy, ZEcidium pedatatum Schw., A. sagittatum Schw., Ceoma pedatatum Schw., C. sagittatum Schw.). On Viola pedata L., I, Pennsylvania, 2885. Viola primulefolia L., I, Pennsylvania, 2884 p.p. Viola sagittata Ait., I, Pennsylvania, 2886. URoMYCES APPENDICULATUS (Pers.) Fries (Uredo appendiculata Pers., Puccinia Phaseoli-trilobi Schw.). On Phaseolus vulgaris L., Il, North Carolina, Pennsylvania, 2845. Strophostyles helvolva (L.) Britton (Phaseolus trilobus Michx.), III, New York, 2942. Uxomyces Fas (Pers.) DeBary (Uredo Vicie Pers., Ceoma leguminosarum Schlecht., Puccinia Fabe Link.) On Vicia Faba L., Il, 111, North Carolina, Pennsylvania, 2847, 2943. Uromyces LEspEDEZ#-PROCUMBENTIS (Schw.) M. A. Curt. (Puc- a cinia Lespedeze-procumbentis Schw., P. Lespedeze-poly- stachye Schw., P. Lespedezee-violacee Schw.). On Lespedeza hirta (L.) Hornem. (L. polystachya Michx.), III, North Carolina, 2941. PROC. AMER. PHIL. SOC., VOL. LVII, S, JULY I7, 1918. 270 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S _ Lespedeza procumbens Michx., II, North Carolina, is sylvania, 2940. 2 Lespedeza violacea (L.) pas III, North Caroling, Nev Jersey, 2041. Uromycgs HEpySARI-PANICULATI (Schw.) Farl. (Puccinia H sari-paniculati Schw., Phragmidium Hedysari Schw On Meibomia paniculata (L.) Kuntze (Hedysarum pa * tum L., Desmodium paniculatum DC.), ath Carolina, Pennsylvania, 2947. UroMYces pRoEMINENS (DC.) Pass. (Aicidium Euphorbiechag icifolie Schw., Ceoma Euphorbie-hypericifolie Schw On Chamesyce maculata (L.) Small (Euphorbia maculata L. I, North Carolina, 2846. Chamesyce Preslii (Guss.) Arth. (Euphorbia. Guss.), II, III, North Carolina, Pennsylvania, ; I, North Carolina, Pennsylvania, 2890. ey Wee Uromyces SpERMACcOcES (Schw.) M. A. Curt. (Ceoma Spe coces Schw., Puccinia Spermacoces Schw.). : On Diodia teres Walt. (Spermacoce diodina Michx.), i i, North Carolina, Pennsylvania, 2840. | PUCCINIA POCULIFORMIS (Jacq.) Wettst. (cidium Berbe Pers., Ceoma berberidatum Link, Puccinia grat ’ Pers.). 3 On Berberis vulgaris L., I, North Carolina, 2887. Triticum vulgare Vill., II, New York, 2817; III, sylvania, 2905. | PUCCINIA EPIPHYLLA (L.) Wettst. (P. Poarwm Niessl). On Poa pratensis L., II, North Carolina, 2878. Puccinia virGATa Ellis & Ev. (Ce@oma Andropogi Schw.). On Sorghastrum nutans (L.) Nash (Andropogon av Michx.), II, iii, Pennsylvania, 2820. PAPERS GIVING RUSTS OF NORTH AMERICA. 271 1a MajANTH# (Schum.) Arth. (4cidium Uvularie Schw., ZEcidium uvulariatum Schw., Ceoma convallariatum Link, C. uvulariatum Schw.). On Uvularia perfoliata L., 1, North Carolina, 2858. Vagnera racemosa (L.) Morong (Smilacina racemosa Desf.), I, Pennsylvania, 2857. ccIn1A ANpDROPOGONIS Schw. ( P. Zizanie Schw., £cidium Pen- : tastemonis Schw., A. pentstemoniatum Schw., Ceoma ; pentstemoniatum Schw.). On Andropogon scoparius Michx., III, Pennsylvania, 2917. | Andropogon virginicus L., III, Pennsylvania, 2915. Pentstemon australis Small, I, North Carolina, 2864. CCINIA FRAXINATA (Link) Arth. (cidium Fraxini Schw., Caoma fraxinatum Link, C. fraxinites Schw.). On Fraxinus sp., 1, North Carolina, Pennsylvania, 2901. PuccINIA ARUNDINARL® Schw. On Arundinaria sp., III, Carolina, 2907. ccrntA CLeMaTipis (DC.) Lagerh. (P. Agropyri Ellis & Ev., Zicidium Clematitis Schw., A. clematitatum Schw., Ceoma clematitatum Schw.). On Clematis virginiana L., I, North Carolina, Pennsylvania, 2874. CCINTA Eaton1# Arth. (4cidium Ranunculi Schw.). _ On Ranunculus abortivus L., I, North Carolina, Pennsylvania, 2875. CCINIA HIBIscIATA (Schw.) Kellerm. (P. Muhlenbergie Arth. & Holw., Zcidium hibisciatum Schw., Ceoma hibi- sciatum Schw.). On Hibiscus militaris Cav., I, Pennsylvania, 2877. 272 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S- Puccrni1a Impatientis (Schw.) Arth. (P. perminuta Arth., dium Impatientis Schw., A. impatientatum Schw., C. impatientatum Schw.). On Impatiens biflora Walt. (I. maculata Muhl. de I, Ne Carolina, Pennsylvania, 288o. Puccrinia Sorcui Schw. On Zea Mays L., ii, III, Pennsylvania, 2970. PuccrInia EMACULATA Schw. On Panicum capillare L., III, Pennsylvania, 29172. Puccrnia WINpsorL® Schw. On Tridens flavus (L.) Hitche. (Poa quingutae Pursh, seslerioides Michx., P. flava L.), III, — 2014. PuccINIA LYSIMACHIATA (Link) Kern (P. limose Magn., 2 ium Lysimachie Schw., Ceoma lysimachiatum Link) On Lysimachia quadrifolia L., 1, North Carolina, 2863. _ Lysimachia terrestris (L.) B. S. P. (L. racemosa Lam, stricta Ait.), I, North Carolina, Pennsylvania, 28¢ PUCCINIA HIERACIATA (Schw.) Arth. & Bisby (P. potructts : ZEcidium hieraciatum Schw., Ceoma hieraciatum Se On Hieracium paniculatum L., I, Pennsylvania, 2868. — Puccinra AsTERUM (Schw.) Kern (P. extensicola Plowr., ium Asterum Schw., A. asteratum Schw., A. tum Schw., A. Solidaginis Schw., Ceoma ast Link, C. erigeronatum Schw.). On Aster paniculatus Lam., I, North Carolina, 2870. Erigeron annuus Pers. (E. heterophyllus Muhl.), I, sylvania, 28690. Solidago sp., I, North Carolina, 2870. Puccinia Irwis (DC.) Wallr. (C@oma Iridis Schw.). On Iris versicolor L., II, Pensylvania, 2812 PAPERS GIVING RUSTS OF NORTH AMERICA. 273 cCINIA PoLyGoNI-AMPHIBII Pers. (P. polygonorum Link, P. Polygoni-pensilvanici Schw., P. concentrica Schw., # ZEcidium Geranii-maculati Schw.). On Geranium maculatum L., I, North Carolina, 2879. 3 Persicaria emersa (Michx.) Small (Polygonum coccineum Muhl.), I, III, Pennsylvania, 2978. Persicaria pennsylvanica (L.) Small (Polygonum pennsyl- vanicum L.), III, North Carolina, 2917. Tovara virginiana (L.) Raf. (Polygonum virginianum L.), III, Pennsylvania, 2917, 2918. Puccinta CLAYTONIATA (Schw.) Peck (cidium claytoniatum Schw., Ceoma claytoniatum Schw.).° On Claytonia virginica L., 1, New York, 2892. Puccinra ANEMONES-VIRGINIAN® Schw. (P. Solida Schw.). On Anemone virginiana L., III, North Carolina, Pennsylvania, 2937- Pucctnta Popopuyiii Schw. (P. aculeata Link, Zcidium Podo- phylli Schw., A. podophyllatum Schw.). On Podophyllum peltatum L., I, Ill, North Carolina, Penn- sylvania, 2888, 2939. a7 HeEucHER# (Schw.) Dietel (Ce@oma Heuchere Link, fe Uredo Heuchere Schw.). On Heuchera americana L., 111, North Carolina, 2843. Heuchera villosa Michx., III, North Carolina, 2843. Puccrnta Viotz (Schum.) DC. (Zcidium Viole Schum., Ceoma violatum Link). On Viola hastata Michx., I, North Carolina, 2884 p.p. -Puccinta Samsucr (Schw.) Arth. (P. Bolleyana Sacc., Zcidium Sambuci Schw., A. sambuciatum Schw., Ceoma sam- buciatum Schw.). On Sambucus canadensis L., I, North Carolina, Pennsylvania, 2897. 274 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Puccinia GRossULARL2 (Schum. ) Lagerh: (Ceoma grosst Link). ~ On Carex sp., III, Pennsylvania, 2908 p.p. ; Grossularia oxycanthoides (L.) Mill. (Ribes oxycan | L.), I, Pennsylvania, 2882. Puccinia Eveocuaripis Arth. (Ce@oma compositarum Lin ; On Eupatorium purpureum L., I, Pennsylvania, 2867 B. PUCCINIA ANGUSTATA Peck. On Scirpus cyperinus (L.) Kunth, III, Penmaylvaia, P-P., 2909. PUCCINIA CANALICULATA (Schw.) Lagerh. (Spheria ca. Schw.). On Cyperus sp., III, Pennsylvania, 1487. Puccin1a Smitacis Schw. (42cidium Smilacis Schw., A. sm tum Schw., Ceoma smilacinatum Link, beds Sn Schw.). A On Smilax rotundifolia L., 1, North Carolina, 2859; I 2822; III, same, bere. Smilax sp., II, Pennsylvania, 2822; III, same, 2916. Puccinia Circ# Pers. fae On Circea Lutetiana L., III, North Carolina, Penns 2831, 2938. PUCcCINIA Paces (Strauss) Mart. (P. Myrrhis Sch Osmorrhize C. & P., Ceoma Anemonis Schw., C. phylli Schw.). On Cherophyllum procumbens (L.) Crantz (Myrrhis ) bens Spreng.), II, III, Pennsylvania, 2934. a Osmorrhiza Claytoni (Michx.) Clarke (Myrrhis ( Michx.), II, III, Pennsylvania, 284r. ) Osmorrhiza sp. (not Anemone or Chelidonium), II. sylvania, 2829; II, III, New York, 2857. PAPERS GIVING RUSTS OF NORTH AMERICA. 275 CCINIA MENTH# Pers. (P. Pycnanthemi Schw., Ceoma labia- : tarum Link, Uredo Clinopodii Schw.). On Koellia incana (L.) Kuntze (Clinopodium incanum L., Pycnanthemum incanum Michx.), Il, North Carolina, Pennsylvania, 2823; III, Pennsylvania, 2920. Puccrnia verrucosa (Schultz) Link (P. Hyssopi Schw.). On Agastache scrophulariefolia (Willd.) Kuntze (Hyssopus scrophulariefolius Willd.) III, Pennsylvania, 2944. ‘Puccrnta MaAcutosa Schw. not Strauss (?Zcidium Dandelionis Schw.). On (?) Adopogon Dandelion (L.) Kuntze (Krigia Dandelion Nutt., Tragopogon Dandelion L., Cynthia Dandelion DC.), III, North Carolina, 2867 a. Adopogon virginicus (L.) Kuntze (Krigia virginica Willd., Cynthia virginica D. Don), III, Pennsylvania, 2922. Poccinta Xantun Schw. On Xanthium sp., III, North Carolina, Pennsylvania, 2927. -Puccinia VeRNoNIz Schw. (P. bullata Schw. not Link, P. longipes Lagerh.). On Vernonia noveboracensis (L.) Willd., III, Pennsylvania, 2919. Vernonia sp., III, North Carolina, 2919; III, Pennsyl- vania, 2926. -Puccinta Kunn1z Schw. On Kuhnia eupatorioides L., 111, Pennsylvania, 2931. -Pucernta TENUIS (Schw.) Burr. (Zcidium tenue Schw., Ceoma tenue Schw.). On Eupatorium ageratoides L.f., 1, Pennsylvania, 2889. Pucctnia Hettopsipis Schw. _ On Heliopsis sp., 111, North Carolina, Pennsylvania, 2924. 276 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Puccin1A HELIANTHI-MOLLIS (Schw.) Arth. & Bisby (P. H. thi Schw., P. Helianthorum Schw., A2cidium Helianthi- mollis Schw., A. helianthatum Schw., A. trachel Schw., Ceoma helianthatum Schw., C. trachelifoliatum Schw.). On Helianthus mollis Lam., I, North Carolina, Pennsylvania, 2871. Helianthus tracheliifolius Willd., I, Pennsylvania, 2872. Helianthus tuberosus L., III, Pennsylvania, 2923. . Helianthus sp., III, North Carolina, Pennsylvania, 2923. Puccin1a VERBESIN” Schw. (cidium Verbesine Schw.). — On Verbesina [occidentalis (L.) Walt.], I, North Carolina, 2870; III, same, 2925. PuccINIA GNAPHALIATA (Schw.) Arth. & Bisby (P. investita Schw Zcidium gnaphaliatum Schw., Ceoma quan u Schw.). ae On Gnaphalium obtusifolium L. (G. polycephalum Michx. I, Pennsylvania, 2873; III, same, 2932. Pucctnra Crrsit Lasch (P. compositarum Link, p.p.). : On Cirsium altissimum (L.) Spreng. (Cnicus altissimus Willd. III, Pennsylvania, 2921. PucciniA Asteris Duby (P. Asteris Schw., P. Helenii Schw.). On Aster cordifolius L., III, Pennsylvania, 2930. Aster paniculatus Lam., III, Pennsylvania, 2930. Aster salicifolius Lam. III, Pennsylvania, 2928. PucciniA SitpHit Schw. On Silphium trifoliatum L., III, North Carolina, 2929. 7Ecip1um ApocyNi Schw. (4. apocynatum Schw., Ceoma apocy tum Schw.). On Apocynum cannabinum L., I, North Carolina, 2865. PAPERS GIVING RUSTS OF NORTH AMERICA. 277 a. ZZcwium cimicirucatum Schw. (Ceoma cimicifugatum Schw.). ____ On Cimicifuga racemosa (L.) Nutt., I, Pennsylvania, 2876. ExctupED NAMES. __ The following names and numbers apply to forms that are not rusts, or if so are impossible of identification. Ceoma (Uredo) rimosum Link. * On Scirpus lacustris L. (S. acutus Muhl.), New York, 2819; no fungus present, probably mechanical injury. Ca@oma (Uredo) Campanularum Link (Uredo Campanule Schw.). On Specularia perfoliata (L.) A. DC. (Campanula perfoliata L., C. amplexicaulis Michx.), North Carolina, Pennsyl- vania, 2830; no specimen preserved, probably not a rust. Ceoma (Uredo) Teucrii Schw. On Teucrium canadense L. (T. virginicum L.), Pennsylvania, 2837; a Hyphomycetous fungus, Cercospora Teucrit (Schw.) Arth. & Bisby (C. racemosa E. & M.). Caoma (Uredo) apiculosum Link (Uredo flosculosorum Alb. & Schw.). On various hosts, North Carolina, Pennsylvania, 2844; no speci- men preserved, and the name so loosely applied as to have no value. @oma (Uredo) Lobelie-cardinalis Schw. On Lobelia cardinalis L., Pennsylvania, 2848 ; a Hyphomycetous fungus, usually called Cercospora effusa (B. & C.) Ellis & Ev. C@oma (Uredo) brunneum Schw. On an undetermined leguminous plant, Pennsylvania, 2850; some pathological condition, but no fungus present. 278 ARTHUR-BISBY—TRANSLATION OF SCHW EINITZ’S _ Ceoma (Zicidium) rubellatum Link ( Bicidinm Rumicis Schw. . On Rumex “and Grossularia,” North Carolina, Pennsyh at 2862; no specimen preserved, probably ater ag 1 perfecti, certainly not a rust. ; durane Schw., A. convolvulatum Schw.). On Ipomea pandurata L., North Carolina, Pennsylvania, 286: not a rust, but one of the Peronosporales, Aimar p mee-pandurane (Schw.) Swingle. Ceoma (Zicidium) osmundatum Schw. (A:cidium comand HU Schw.). On Osmunda spectabilis Willd., New York, 2805; not a1 a but a fungus of uncertain affinity, Mycosyrinx O Peck (Ustilago Osmunde Peck). Caeoma (Aicidium) urticatum Link (4icidium Astonia Sch On Cynoglossum virginicum L. (C. amplexicaule Muhl.) Urtica sp., North Carolina, 2898; no specimen pee \ very doubtful, but certainly not a rust. Puccinia Bullaria Schw. not Link. On Agastache nepetoides (L.) Kuntze (Lophanthus nepeto Benth., Hyssopus nepetoides L.), Pennsylvania, . no specimen in Philadelphia, but one in Washing not a rust, may be an ascomycete. = CHRONOLOGICAL ENUMERATION. After the serial numbers the corresponding numbers from Carolina list, when there are any, are given .in parentheses. Schweinitz name is followed by the name at present in use, or oth identification. An original specimen at the Philadelphia Acai of Sciences is indicated when in an autographic packet by an terisk *, when mounted by a dagger 7. “ “ PAPERS GIVING RUSTS OF NORTH AMERICA. "#41474 (130) Spheria epiphylla LvS. “ * rimosum Lk. “ Tridis Lv.S. 279 = Ravenelia epiphylla (Schw.) ; Diet. canaliculata L.v.S. = Puccinia canaliculata (Schw.) Lagerh. Ceoma (Uredo) Rubigo Lk. =Puccinia poculiformis (Jacq.) Wettst. linearis Lk. = Puccinia epiphylla (L.) Wettst. =no fungus, mechanical injury. Andropogi L.v.S. = Puccinia virgata Ell. & Ev. = Puccinia Iridis (DC.) Wallr. Smilacis Ly.S. = Puccinia Smilacis Schw. “ Labiatarum Lk. = Puccinia Menthe Pers. “ Ipomee Lv.S. “ = Coleosporium Ipomee (Schw.) Burr. Elephantopodis =Coleosporium Elephantopodis L.v.S. (Schw.) Thiim. Solidaginis L.v.S. = Coleosporium Solidaginis (Schw.) Thiim. Terebinthinacee =Coleosporium Terebinthina- Lv.S. cee (Schw.) Arth. Helianthi Ly.S. = Coleosporium Helianthi (Schw.) Arth. Anemonis L.v.S. = Puccinia Pimpinelle (Str.) Mart. Campanularum = uncertain, probably not a rust. Lk. Onagrarum Lk. = Puccinia Circee Pers. mintata Lk. =Earlea speciosa (Fries) Arth. ruborum Lk. = Kuehneola Uredinis (Link) _ Arth. PotentillarumLk. = Frommea obtusa (Str.) Arth. Agrimonie Lv.S. = Pucciniastrum Agrimonie (Schw.) Tranz. Filicum Lk. = Hyalopsora Aspidiotus (Peck) Magn. Teucrit Lv.S. =Cercospora Teucrii (Schw.) Arth. & Bisby, not a rust. Azalee Lv.S. = Pucciniastrum Myrtilli (Schum.) Arth. Ari virginici = Uromyces Caladii (Schw.) ; Lv.S. Farl. Spermacoces = Uromyces Spermacoces ines. (Schw.) M. A. Curt. Cherophylii = Puccinia Pimpinelle (Str.) Lv.S. Mart. Hyperict Lv.S. = Uromyces Hyperici-frondosi (Schw.) Arth. Heuchere Lv.S. = Puccinia Heuchere (Schw.) ; ; Diet. 280 2844 (478) Ceoma (Uredo) apiculosum Lk. = uncertain, name of no value. *+2845 (477,490) “ *72846 (459,474) “ 2847 (476) 72848 72849 +2850 +2851 2854 *42855 #12856 $2857 72858 (453) 728590 (452) 72860 (457) *+2861 2862 (433) 42863 (438) *2864 (449) *+2865 (448) *+2866 (454) *2867 (434) 72868 +2869 ‘pe (444,446) “ 2870 (445) “ a“ “ “ . ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S “ “ epiteum Lk. = Melampsora Bigelowti Thiim. (Zecidium) Convallaria- = Puccinia Majanthe (Schum.) tum Lk. Arth. & Holw. — . Uvulariatum =Puccinia Majanthe (Schum.) Lv. Arth. & Holw. Smilacinatum == Puccinia Smilacis Schw. Lvs. Aroidatum L.v.S. = Uromyces Caladii (Schw.) Farl. -Dracontionatum = Uromyces Caladii (Schw.) Las Farl. rubellatum Lk. = uncertain, not a rust. Lysimachiatum = Puccinia lysimachiata (Link) Lk. Kern. Pentstemonia- = Puccinia Andropogonis Schw tum L.v.S. Apocynatum = Zcidium apocynatum Schw. LD. Convolvulatum = Albugo Ipomee-pandurane — L.v.S. (Schw.) Swingle, not a rust. “ appendiculosum ==Uromyces appendiculatus Lk. (Pers.) Fries. ie punctuosum Lk. = Uromyces proéminens (DC) Pass. Leguminosarum = Uromyces Fabe (Pers.) De- Lk. Bary. Lobelie cardi- =Cercospora effusa (B. & C) nalis L.v.S. Ell. & Ev., not a rust. © Thalictri Lv.S. = Polythelis Thalictri (Chev.) Arth, brunneum L.v.S. = uncertain, not a fungus. Chelidonti L.v.S. = Puccinia Pimpinelle (Str.) Mart. : gyrosum Lk. = uncertain, may be PE ium imitans Arth. cylindricum Lk. = Melampsora Meduse Thim. a=uncertain, may be Puccinia — Compositarum maculosa Schw. Lk. |8=Puccinia Eleocharidis Arth. Hieraciatum = Puccinia hieraciata (Schw.) © Lv.S. Arth. & Bisby. Erigeronatum = Puccinia Asterum (Schw.) Ly. Kern. Asteratum L.v.S. = Puccinia Asterum (Schw.) Kern. Asteratum L.v.S. = Puccinia Verbesine Schw. _— . 2873 a « 2874 (447) “ 72875 (440) “ ox a $2870 (43) “ 72880 (442) “ 42881 (437) “ i“. «= ) *72883 (451) “ $2884 (430) “ #42885. - a - *72887 (458) “ 42888 (435) “ Bo ts “ ow - 6° eo 72893 - a $2894 - “ 72895 - . “ “ “ce PAPERS GIVING RUSTS OF NORTH AMERICA. 281 | $2871 (450) Ceoma (Zcidium) Helian- = Puccinia Helianthi-mollis thatum L.v.S. (Schw.) Arth. & Bisby. Trachelifoliatum = Puccinia Helianthi-mollis L.v.S. (Schw.) Arth. & Bisby. Gnaphaliatum =Puccinia gnaphaliata (Schw.) Lv.S. Arth. & Bisby. Clematitatum | —=Puccinia Clematidis (DC.) Lv.S. Lagerh. Ranunculaceatum = Puccinia Eatonie Arth. Lk. Cimicifugatum = Zcidium cimicifugatum Schw. Lv.S. Hibisciatum = Puccinia hibisciata (Schw.) LS. Kellerm. Hepaticatum =Tranzschelia punctata (Pers.) L.v.S. Arth, Geraniatum Lk. = Puccinia Polygoni-amphibiu Pers. Impatientatum = Puccinia Impatientis (Schw.) -Lv,5. Arth. Berberidatum =Puccinia poculiformis (Jacq.) Lk. Wettst. grossulariatum = Puccinia Grossularie (Schum.) Lk. Lagerh. Hypericatum = Uromyces Hyperici-frondosi LwS: (Schw.) Arth. Puccinia Viole (Schum.) DC. Violatum Lk. =Uromyces pedatatus (Schw.) Sheldon. pedatatum L.v.S.= Uromyces pedatatus (Schw.) Sheldon. sagittatum L.v.S. = Uromyces pedatatus (Schw.) Sheldon. luminatum L.v.S. = Kunkelia nitens (Schw.) Arth. Podophyllatum = Puccinia Podophylli Schw. Lvs tenue L.v.S. = Puccinia tenuis (Schw.) Burr. “ Euphorbie-hyper- = Uromyces proéminens (DC.) ictfolie L.v.S. Pass. Houstoniatum = Uromyces houstoniatus Eev.S. (Schw.) Sheldon. Claytoniatum = Puccinia claytoniata (Schw.) La: Peck. Pyrolatum L.v.S. = Melampsoropsis Pyrole (DC.) Arth. myricatum L.v.S. = Gymnosporangium myricatum (Schw.) Fromme. Osmundatum = Mycosyrinx Osmunde Peck, Lv.S. not a rust. 282 | ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S — Ceoma (Aicidium) Pyratum = Gymnosporangium Juni} , bigs Lv.S. virginiane Schw, — *+2807 (441) “ “~ sambuciatum = Puccinia Sambuci -_ Liw.S: Arth. ih 2808 (436) “ “ Urticatum Lk. = uncertain; not a rust. 4 *+2809 (432) “ (Restelia) Cylindrites = Gymnosporangium globorem Lk. a Farl. Soe. = “ hyalinum (Cooke) ern y= “ trachysorum Kern, &’= “ Juniperi-virginiane Sch * ‘Tenthericonoiaanae *+29000 (431) “ “ Restelites Lk. =} « pares pies ; f2901 (430) “ “ Fraxinites L.v.S. = Puccinia fraxinata (icky Arth. *+2002 - ¥ “ Botryapites = Gymnosporangium bot Lyv.S. (Schw.) Kern. ang Cronartium Quercus (Brond. 42903 (456) “ (Peridermium) Pineum Schrot. Lk. =) Peridermium intermedium ; Arth. & Kern, =. $2904 - “ “~ germinale L.v.S. = Gymnosporangium verminae (Schw.) Kern. : *20905 (492) Puccinia graminis Lk. = Puccinia poculiformis (Jace Wettst. in part. — Bee *2906 - “ striola Lk. = Uromyces Junci-effusi Syd. *2007 (487) “ Arundinarie L.v.S. = Puccinia Arundinarie Schw. Puccinia Grossularie *2908 - “ punctum Lk. = (Schum.) Lagerh. — Puccinia angustata Peck. *29009 - “ Scirpi Lk. = Puccinia’ angustata Peck. *2010 - “ Sorghi Lv.S. = Puccinia| Sorghi Schw. — *2011 — “ Andropogi Lwv.S. = Puccinia Andropogonis § *2012 - “ emaculata L.y.S. = Puccinia emaculata Schy *2013. - “ Junci Lv.S. = Uromyces Junci-effusi S *2014 - “ Windsorie Lw.S. = Puccinia Windsorie Sch *2915 - “ Zizanie L.v.S. = Puccinia Andropogonis 2016 - “ Smilacis L.v.S. = Puccinia Smilacis Schw. *2017. - “ Polygonorum Lk. = Puccinia Polygoni-amphibii Pers. *2918 - “ concentrica L.v.S. = Puccinia Polygoni-amphibii Pers. *2919 (501) “ bullata Lv.S. = Puccinia Vernonie Schw. 2920 - “ Pycnanthemi L.v.S. = Puccinia Menthe Pers. *2021 - “ compositarum Lk, = Puccinia Cirsii Lasch. 2922 - = morula [L.v.S. not] = Puccinia maculosa Schw. — Strauss ae 29023 (405) “ Helianthorum Lv.S. = Puccinia Helianthi-mollis (Schw.) Arth. & Bisby. © PAPERS GIVING RUSTS OF NORTH AMERICA. 283 2924 (493) Puccinia Heliopsidis Lv.S. = Puccinia Heliopsidis Schw. 2925 (496) Verbesine Lv.S. = Puccinia Verbesine Schw. - “ Vernonie Lv-.S. = Puccinia Vernonie Schw. (500) “ Xanthii L.v.S. = Puccinia Xanthit Schw. *2928 - “ Helentt Ly.S. = Puccinia Asteris Duby. *2929_ - “ Silphii Lv.S. = Puccinia Silphit Schw. #2930 - “ Asteris Lv.S. = Puccinia Asteris Duby. "2031 — “ Kuhnie Lvy.S. = Puccinia Kuhnie Schw. - “ investita L.v.S. = Puccinia gnaphaliata (Schw.) Arth. & Bisby. 29033 (4909) “ Galit LS. = Puccinia punctata Link. *2034 - “ Myrrhis Lv.S. = Puccinia Pimpinelle (Str.) Mart. i 7 “ Bullaria [L.v.S. not] Lk. = uncertain; not a rust. 2936 - “anemones Lk. = Polythelis fusca (Pers.) Arth. 2937 (486) “ solida Lwv.S.. = Puccinia Anemones-virginiane Schw. - *2938 (4901) “ Circee Lk. = Puccinia Circee Pers. 2039 (480) “ aculeata L.v.S. = Puccinia Podophylli Schw. 2940 (497) “ Lespedeze procumbentis = Uromyces Lespedeze-procum- ou - Lyv.S. bentis (Schw.) M. A. Curt. violacee L.v.S. = Uromyces Lespedeze-procum- bentis (Schw.) M. A. Curt. Phaseoki trilobi Lv.S. = Uromyces appendiculatus (Pers.) Fries. Fabe Lk. =Uromyces Fabe (Pers.) De- Bary. Hyssopi Lv.S. = Puccinia verrucosa (Schultz) Link. : Potentille Lwv.S. = Frommea obtusa (Strauss) Arth. Ari triphyllii Lw.S. = Uromyces Caladii (Schw.) Farl. 2047 (503) Phragmidium Hedysari L.v.S. = Uromyces Hedysari-paniculatt (Schw.) Farl. Sits — Seiridium marginatum [L.v.S. = Earlea speciosa (Fries) Arth. Z not] Lk. 3085 - “ S$(4) milacis L.v.S. = Earlea speciosa (Fries) Arth. +3004 -— Gymnosporangium Juniperi = Gymnosporangium germinale : Lk. (Schw.) Kern. 73095 - Podisoma Juniperi Lk. = Gymnosporangium clavarie- 73096 (504) “ forme (Jacq.) DC. macropus L.v.S. = Gymnosporangium Juniperi- virginiane Schw. 284 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Actinomeris sp. 2870 Adopogon Dandelion 2867 virginicus 2922 Agastache nepetoides 2935 scrophulariefolius 2044 Agrimonia Eupatoria 2835 parviflora 2835 Agrostis sp. 2880 Andropogon avenaceum 2820 Scoparius 2911, 29015 virginicus 2915 sp. 2885 Anemone Hepatica 2878 quinquefolia 2829, 2936 virginiana 2037 Apocynum cannabinum 2865 Arisema Dracontium 2861 triphyllum — Aronia Botryapium pore Arum dracontium 2861 triphyllum 2946 virginicum 2839, 2860 Arundinaria SP. 2007 Aspidium obtusum 2836 Aster cordifolius 2930 paniculatus 2870, 2930 salicifolius 2928 sp. 2826 Azalea nudiflora 2838 Berberis canadensis 2881 vulgaris 2881 Inpvex To Hosts Caladium sagittifolium 2839 Campanula amplexicaulis 2830 perfoliata 2830 _ Carex ee sp. 2882, 2897, 2905, 2908 Cherophyllum fagraty: procumbens 2034 ete sp. 2841 Bae es. Chamecyparis thyoides 2894 Chamesyce maculata 2846 Preslii 2846, 2890 Chelidonium sp. 2851 Cimicifuga racemosa 2876 Circea canadensis 2831, on Cirsium altissimum 2921 Claytonia virginica 2892 Clematis virginiana 2874 Clinopodium incanum 2823 Cnicus altissimus 2921 Convolvulus pandurata 2866 Cracca virginiana 1474 Crategus arborescens 2899 8 Oxyacantha 2899 Y punctata 2899 4 viridis 2899 8 Sp. 2904 Cynoglossum amplexicaule 2898 Cynthia Dandelion 2867 . virginica 2867, 2922 Cyperus sp. 1487 lesmodium : i m 2947 i teres 2840 hypericifolia 2846, 2890 maculata 2846 Preslii 2846, 2890 virginiana 1474 Galium 4 purpureum 2033 _ Geranium carolinianum 2879 maculatum 2879 Gnaphalium obtusifolium 2873 polycephalum 2873, 2032 Grossularia oxyacanthoides 2882 rotundifolia 2882 sp. 2862 Hedysarum paniculatum 2947 Helenium autumnale 2928 Helianthus giganteus 2828 mollis 2871 trachelifolius 2872 PAPERS GIVING RUSTS OF NORTH AMERICA. Helianthus tuberosus 2923 Sp. 2926 Heliopsis Sp. 2924 Hepfatica Hepatica 2878 triloba 2878 Heuchera americana 2843 villosa 2843 Hibiscus militaris 2877 Hieracium . maculatum 2868 paniculatum 2868 Sp. 2922 Houstonia cerulea 2891 Hypericum frondosum 2883 prolificum 2883 sp. 2842 Hyssopus nepetoides 2935 scrophulariefolius 2944 Impatiens biflora 2880 maculata 2880 Ipomea pandurana 2824, 2866 triloba 2824 Tris versicolor 2821 virginica 2821 Juncus effusus 2906, 2913 Juniperus communis 3005 Sabina 3095 virginiana 2899, 3004 Kellia incana 2823 Krigia Dandelion 2867 virginica 2867, 2922 PROC. AMER. PHIL. SOC., VOL. LVII, T, July 20, 1918. 286 ARTHUR-BISBY—TRANSLATION OF —a Kuhnia eupatorioides 2931 sp. 2844 Lespedeza hirta 2941 polystachya 2941 procumbens 2940 violacea 2941 Lobelia cardinalis 2848 Lophanthus nepetoides 2935 ‘Lysimachia quadrifolia 2863 racemosa 2863 stricta 2863 terrestris 2863 Malus coronaria 2806 Malus 2899 5, 2900 Meibomia paniculata 2947 Miegia ; SP. 2907 Muhlenbergia sp. 2877 Muricauda Dracontium 2861 Myrica cerifera 2894 Myrrhis Claytoni 2841 procumbens 2934 Osmorrhiza Claytoni 2841 sp. 2829, 2851 Osmunda spectabilis 2895 Oxalis sp. 2910 Panicum capillare 2912 pubescens 2912 Peleanaes virginica 2839, 2860 Pentstemon australis 2864 hirsutus 2864 Phalaris sp. 2857 Phaseolus trilobus 2942 vulgaris 2845 sp. 2844 Phegopteris Dryopteris 2836 Pinus inops 2903 Sp. 2903 Pisum sativum 2845 Poa flava 2914 pratensis 2818 quinquedentata 2914 quinquifida 2914 — seslerioides 2914 Podophyllum peltatum 2888, 2939 Polygonum coccineum 2918 emersum 2918 pennsylvanicum 2917 © virginicum 2917, ath Populus dilatata 2855 Potentilla canadensis 2834, 2045 Prenanthes sp. 2867, 2922 Pycnanthemum incanum 2920 Pyrola rotundifolia 2893 uliginosa 2893 Pyrus angustifolia 2896 coronaria 2896, 2899 6 communis 2900 Malus 2899 5, 2900 oxyacanthoides 2882 _ rotundifolium 2882 carolina 2832, 3084 corymbosa 3084 _ pauciflora 2832, 3084 - virginiana 3085 Enslenii 2887 tdeus 2833, 2854 strigosus 2833, 2887 villosus 2887 Rumex sp. 2862 terebinthinaceum 2827 trifoliatum 2929 _ racemosa 2857 Smilax -caduca 3085 _ rotundifolia 2822, 2859, 2916 Solidago rugosa 2826 sempervirens 2826 serotina 2826 sp. 2870 Sorghastrum 5 nutans 2820 Sorghum sp. 2910 PAPERS GIVING RUSTS OF NORTH AMERICA. Spartina Sp. 2901 Specularia perfoliata 2830 Spermacoce diodina 2840 Strophostyles helvola 2942 Tephrosia virginiana 1474 Teucrium virginicum 2837 Thalictrum Cornuti 2849 polygamum 2849 Tovara virginiana 2917 Tragopogon Dandelion 2867 Tridens flavus 2914 Triticum vulgare 2817, 2905 Urtica sp. 2898 Uvularia perfoliata 2858 Vagnera racemosa 2857 V erbesina occidentalis 2870 Sp. 2925 Vernonia noveboracensis 2826, 2919 sp. 2926 Vicia Faba 2847, 2043 Viola cucullata 2884 hastata 2884 obliqua 2884 pedata 2885 primulefolia 2884 sagittata 2886 287 288 ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Windsoria (Poa) quinquedentata 2914 Xanthium Sp. 2927 InpEXx To Funct. Zicidium Zicidium apocynatum 2865 Apocyni 2865 Aroidatum 2860 Asperifolii 2808 asteratum 2870 Asterum 2870 Berberidis 2881 Caladii 2860 cancellatum 2900 cimicifugatum 2876 claytoniatum 2892 clematitatum 2874 Clematitis 2874 convolvulatum 2866 Crategi oxycanthe 2899 Dandelionis 2867 dracontionatum 2861 erigeronatum 2869 Euphorbie 2890 Euphorbie-hypericifolie 2890 Fraxini 2901 Geranii-maculati 2879 gnaphalitatum 2873 helianthatum 2871 Helianthi-mollis 2871 hepaticatum 2878 hieraciatum 2868 hibisciatum 2877 houstoniatum 2891 hypericatum 2883 Hyperici-frondosi 2883 impatientatum 2880 Impatientis 2880 Ipomee-pandurane 2866 luminatum 2887 Lysimachie 2863 myricatum 2894 nitens 2886 osmundatum 2895 pedatatum 2885 Pentastemonis 2864 Zizania pe ie Albugo Ceoma (A2cidium) Mays 2910 Sp. 2015 pentstemoniatum 2864 — Hs Pini 2903 ge oe podophyllatum 2888 a ie Podophylli 2888, 2939 Prenanthis 2867 pyratum 2806 pyrolatum 2893 quadrifidum 2878 Ranunculi 2875 Rumicis 2862 sagittatum 2886 Sambuci 2897 sambuciatum 2897 smilacinatum 2859 Smilacis 2859 Solidaginis 2870 tenue 2889 trachelifoliatum 2872 Uvularie 2858 uvulariatum 2858 Verbesine 2870 Viole 2884 Ipomee-pandurane 2866 e apocynatum 2865 aroidatum 2860 asteratum 2870 berberidatum 2881 cimicifugatum 2876 claytoniatum 2892 clematitatum 2874 compositarum a Prenanthis compositarum B Eupatoria convallariatum 2857 convolvulatum 2866 dracontionatum 2861 erigeronatum 2869 Euphorbie-hypericifolie 2800 geraniatum 2879 : gnaphaliatum 2873 _ grossulariatum 2882 helianthatum 2871 hepaticatum 2878 hibisciatum 2877 _ hieraciatum 2868 houstoniatum 2891 hypericatum 2883 Hypericorum 2842 impatientatum 2880 luminatum 2887 lysimachiatum 2863 myricatum 2894 osmundatum 2895 pedatatum 2885 pentstemoniatum 2864. podophyllatum 2888 pyratum 2806 pyrolatum 2893 ranunculaceatum 2875 rubellatum 2862 Sagittatum 2886 sambuciatum 2897 smilacinatum 2859 tenue 2889 trachelifoliatum 2872 urticatum 28098 uvulariatum 2858 violatum 2884 Ceoma (Peridermium) germinale 2904 Pineum 2903 Ceoma (Restelia) botryapites 2902 cylindrites 2899 ‘ 8 Crategi-arborescentis 2899 a Crategi-punctate 2899 3 Mali 2890 ¥ Oxyacanthe 2899 fraxinites 2901 restelites 2900 Caeoma (Uredo) Andropogi 2820 Anemonis 2829 Agrimonie 2835 apiculosum 2844 appendiculosum 2845 Ari-virginici 2839 Azalee 2838 PAPERS GIVING RUSTS OF NORTH AMERICA. 289 Caoma (Uredo) brunneum 2850 Campanularum 2830 Cherophylli 2841 . Chelodonit 2851 cylindricum 2855 Elephantopodis 2825 epiteum 2856 Filicum 2836 gyrosum 2854 Helianthi 2828 Heuchere 2843 Hyperici 2842 Ipomee 2824 Tridis 2821 labiatarum 2823 leguminosarum 2847 linearis 2818 Lobelie-cardinalis 2848 mintata 2832 onagrarum 2831 Potentillarum 2434 punctuosum 2846 rimosum 2819 rubigo 2817 - ruborum 2833 Smilacis 2822 Solidaginis 2826 . Spermacoces 2840 Terebinthinacee 2827 Teucrit 2837 Thalictri 2849 Ceratites (Ceoma) botryapites 2902 fraxinites 2901 Cercospora effusa 2848 racemosa 2837 Teucrii 2837 Chrysomyxa Pyrole 2893 Coleosporium Elephantopodis 2825 Helianthi 2828 Ipomee 2824 Solidaginis 2826 Terebinthinacee 2827 Vernonia 2826 290 | ARTHUR-BISBY—TRANSLATION OF SCHWEINITZ’S Cronartium Quercus 2903 Earlea Speciosa 2832, 3084, 3085 Frommea obtusa 2945 Gymnoconia interstitialis 2887 Gymnosporangium botryapites 2902 clavarieforme 3095 clavipes 2904, 3004 Ellisii 2894 germinale 2904, 3004 globosum. 2899, 2900 hyalinum 2899 Juniperi 3004 Juniperi-virginiane 2896, 2899, 2900, 3096 myricatum 2894 trachysorum 2899 Hyalopsora Aspidiotus 2836 Kuehneola obtusa 2834 Uredinis 2833 Kunkelia nitens 2887 Melampsora Bigelowiti 2856 Meduse 2855 Melampsoropsis Pyrole 2893 Mycosyrinx Osmunde 2895 Peridermium cerebrum 2903 germinale 2904 intermedium 2903 Phragmidium albidum 2833 Hedysari 2947 Phragmidium Podisoma Polythelis Puccinia imitans 2854 “a Potentilla-conolles ik speciosum 2832, wie Juniperi 3005 macropus 3006 Thalictri 2849 aculeata 2939 Agropyri 2874 Andropogi 2911 fi Andropogonis 2864, side Anemones 2936 NN het a: Anemopnes-virginiane 2037, angustata 2909 Spe, Ari-triphylli 2946 Arundinarie 2907 — Asteris 2928, 2930 Avicularie BB Fabe aay Bolleyana 2897 aye bullaria 2935 bullata 2919, 2926 canaliculata 1487 Caricis-Asteris 2870 Caricis-Erigerontis ph Caricis-S olidaginis 2870 — Circe@ 2831, 2938 =e Cirsii 2921 ‘ any ane claytoniata 2892 compositarum 2921 concentrica 2918 Eatoni@a 2875 Eleocharidis 2867 emaculata 2912 epiphylla 2818 extensicola 2869, 2870 Fabe 2943 fraxinata 2901 Galii 2933 gnaphaliata 2873, 2932 graminis 2817, 2881, 2905 Grossularie 2882 Hedysari-paniculati 2947 Helenii 2928 Helianthi 2871, 2023, Helianthi-mollis 2871, 2872 Helianthorum 2923 | Heliopsidis 2924 _hibisciata 2877 _Heuchere 2844 _ Hyssopi 2944 ___Impatientis 2880 investita 2873, 2932 Tridis 2821 Junci 2913 Krigie 2867 Kuhnie 2931 Lespedeze-polystachy@ 2941 Lespedeze-procumbentis 2940 Lespedeza-violacee 2941 limose 2863 lysimachiata 2863 maculosa 2867, 2922 Majanthe 2857 Menthe 2823, 2920 Muhlenbergie 2877 Myrrhis 2934 _ Osmorrhize 2829, 2841, ii patrueclis 2868 perminuta 2880 Phaseoli-trilobi 2942 Pimpinelle 2829, 2841, 2851, 2934 _ Poarum 2818 poculiformis 2817, 2881, 2905 Podophylli 2888, 2939 Polygoni-amphibiti 2879, 2917,2918 Polygoni-pensilvanici 2917 Polygonorum 2917 Potentille 2945 _ Pruni-spinose 2878 punctata 2933 punctum 2908 Pycnanthemi 2920 Sambuci 2897 __ Scirpi 2909 _ Silphit 2929 Smilacis 2822, 2859, 2916 solida 2937 _ Sorghi 2910 _Spermacoces 2840 striola 2906 tenuis 2889 _Thalictri 2849 _ Verbesine 2870, 2925 PAPERS GIVING RUSTS OF NORTH AMERICA. Puccinia Vernonie 2919, 2926 verrucosa 2044 Viole 2884 virgata 2820 Windsorie 2914 Xanthii 2927 Zizanie@ 2915 Pucciniastrum Agrimonie 2835 Circee 2831 minimum 2838 Myrtilli 2838 Ravenelia epiphylla 1474 Seiridium marginatum 3084 Smilacis 3085 Spheria canaliculata 1487 epiphylla 1474 Tranzschelia punctata 2878 Uredo Alchemille B 2834 appendiculosa 2845 Caladti 2839 Campanule 2830 Circee 2831 Clinopodtt 2823 Elephantopodis 2825 Euphorbie 2846 - flosculosorum 2844 Heuchere 2843 Ipomee 2824 linearis 2818 minima 2838 Rose 2835 scutellata 2846 Smilacis 2822 Solidaginis 2826 Terebinthinacee 2827 Vicie 2847 Uromyces Andropogonis 2885 291 _ Uromyces appendiculatus 2845, 2042 Caladii 2839, 2860, 2861, 2946 Fabe 2847, 2043 Hedysari-paniculati 2047 houstoniatus Ror Hyperici-frondosi 2842, 2883 Junci-effusi 2906, 2913 Purpue University, LAFAYETTE, INDIANA, 2N> > Y OF SOME ANT LARV, WITH A CONSIDERA- TION OF THE ORIGIN AND MEANING OF THE SOCIAL HABIT AMONG INSECTS. By WILLIAM MORTON WHEELER. . (Read April 19, 1918.) _ The care lavished by ants on their brood is a matter of such fre- -and easy observation that it has always excited wonder. and nent. When a colony is disturbed at the height of the breeding son the brood is at once seized and carried to a place of safety, 1 when more closely observed in artificial nests the behavior of . workers is seen to be very largely a constant round of-the four rent activities of feeding, licking, transporting and defending the oung. Swammerdam says in that wonderful volume, the “Biblia . lature ” (1737-1738) : “ Incredible oropy7 et cura Formic educant nmamque dant operam, ne vel tantillum quod ciate eorum Ver- ¥iculorum educationem atque nutritionem omittant ”— with in- dible affection and care the ants bring up their vermicules and mit not the least thing appertaining to their education and nurture.” le intentionally uses the word oropyy, from the verb orépyev, to ess his belief that love or affection impels the ant to care for — r young, since there is no Latin equivalent for a term which to Greek meant the affection of the members of a human family - one another as distinguished from other forms of “tender ling.”? And observers since Swammerdam seem uniformly to ve aerend with him, though more modern writers often use such ms as “care of the brood” as more suited to the present colorless d noncommittal stage of natural history. Now it must be conceded that Swammerdam’s statement calls 1 Contributions from the Entomological Laboratory of the Bussey Insti- ion, Harvard University. No. 147. Plato in the “Laws,” 754B, uses the verb in its typical Greek sense he Says: wats orépye te al orépyera brd ray yernodvtwv. _ PROC. AMER. PHIL. SOC., VOL. LVII. U, AUGUST 8, I918. 293 294 WHEELER—ANT LARV. attention to a real problem, but one that belongs to the psycho ; i and not to the biologist. I am quite willing to admit that there 1 y be in ants some feeble analogue of the parental feelings of man and the highest animals, but as a biologist I am bound to seek an possible to find some ethological or physiological basis for the ant’ behavior toward her brood. Like other students of insects I hav no doubt, often taken too much for granted and have unquestionably ) committed the eighth deadly sin, called by the orthodox behaviorists “anthropomorphism,” not once but many times. By _ of pa penance I offer the following paper. < I confess that I took the Swammerdamian conception for gran d till recently, while studying a collection of ants made in the Belgian Congo for the American Museum of Natural History by my fri Mr. H. O. Lang, came upon some facts which seem to throw a fi of light on the true meaning of the relations of ants to their brood. These relations now appear to me so simple and unequivocal tha find difficulty in understanding how they could have remained long unperceived, especially as a host of other facts had been sistently pointing in the same direction. Our blindness seems have been due to regarding the adult ants as the only active facto in the brood relationship. We supposed that the larve, probab because they are such sluggish, legless maggots, were merely tk inert and passive objects of the feeding, licking, transportation protection. One result of this assumption has been a general negl ; of the study of larval ants. Even their morphology has tr c little attention. There are a few valuable papers by Berlese (1901 Karawaiew (1896) and Pérez (1902) on the metamorphosis of ant a single paper by Emery (1899) devoted to the external char of the larve in a selected series of species and a number of sc descriptions and figures, published mainly for taxonomic purpo by myself and others. I regret that in the past I failed to study the larval ants n closely and more continuously, especially as the meaning of som the unpublished records in my notebooks of 1899 and 1900 is to me only now after the lapse of nearly twenty years. W took up my work at the University of Texas in the fall of 899 8 a morphologist accustomed to well-furnished northern and Europeai WHEELER—ANT LARV2. 295 embryological and anatomical laboratories and libraries, I found so little apparatus for the work in which I had been trained, that I fell to a peculiar listlessness and was for some weeks unable to con- trate my attention on any subject that seemed worthy of investi- gation. One day, while I sat on the bank of Barton Creek, near ustin, in the very spot where, as I later learned, MacCook had worked on the famous agricultural ant (Pogonomyrmex molefa- ciens), I happened to see a file of cutting ants (Atta texana), each with its piece of leaf poised in its mandibles. I vividly remember the thrill of delightful fascination with which I watched the red- brown creatures trudging along under their green loads, and it seemed to me that I had at last found a group of organisms that would repay no end of study. At that time there was no active myrmecologist in the country. MacCook had completed his work and Pergande was no longer deeply interested in the ants. Prof. Emery, however, and later Prof. Forel extended helping hands to me and forthwith sent me their numerous and important publica- tions, and several of my students, notably C. T. Brues, A. L. Me- lander, C. G. Hartman and W. A. Long, never wearied of accom- panying me on long excursions into the dry, sunny woods and canyons about Austin. For a time I was greatly interested in the habits of three large ants of the primitive subfamily Ponerine, Odontomachus clarus, Pachycondyla montezume and Lobopelta elongata, which are com- ‘mon in Central Texas. I was able-to show that their peculiar tuber- culate larve are not fed with regurgitated food, like the larve of more specialized ants, but with pieces of insects (1900). Concern- ing the feeding of the Odontomachus larva I published the follow- ing remark (p. 24) : These larve are placed by the ants on their broad backs, and their heads and necks are folded over onto the concave ventral surface, which serves as’ a table or trough on which the food is placed by the workers. An unpublished note, the significance of which I did not appreciate at the time, refers to Pachycondyla and was recorded while I was studying the behavior of its extraordinary Phorid commensal, Meto- pina pachycondyle (1901). It runs as follows: 296 WHEELER—ANT LARVZ. As soon as the fragments of insects are placed on the larva’s ti or ventral surface, the latter is sometimes inundated with a copious, col liquid, which is at once eagerly lapped up by the attendant nurse. { should now describe this behavior in the following words as the fragments are placed on its ventral surface, the larva | charges from its salivary glands a supply of secretion which is som times very abundant. This secretion, by means of a strong lytic ferment which it contains, digests the food ai t and thus enables the larva to swallow and assimilate it, at same time serves in part as an agreeable draught for th The strong mandibles of the Ponerine larve are used for cc ing the insect food and thus preparing it for the action of the s The larval feeding habits of our small northern species of Pe Stigmatomma are essentially the same as those of the Texan as I showed in‘a special paper (1900a). _ Within recent years I have examined the larve of a m different Ponerinz and have found them all to possess w: oped mandibles. All, in my opinion, except, perhaps, during he! very youngest stages, are fed with fragments of insects, s directly by their nurses. In some species, the insect prey is ably given to the larva without previous dismemberment. — describe and figure the young of three genera, Myrmecia, Me ponera and Bothroponera, as they differ considerably from one other and from all previously described Ponerine genera and serve therefore to illustrate the diversity of larval structure y the subfamily. e Fig. 1 is from a photograph of the adult larva of Myrmecia guinea, one of the larger Australian bulldog ants, the most primi of existing Formicide. It is milk-white, has the form of a table marrow, with all the segments distinct, except those a extreme posterior end of the body, the anterior segments are slender and curved and the head is very small. The body is ra uniformly clothed with short, rapidly tapering, bristle-like Under a higher magnification the head (Fig. 2) is seen to ha projecting bilobed clypeus (c), broad, heavily chitinized, tridentate mandibles (mm) and well-developed maxille (#) ar labium (J), the former with two pairs of strongly chitinized 4 WHEELER—ANT LARVE. 297 shaped sensille (s’), the latter terminating in a broadly elliptical chitinous plate, with a single pair of knob-shaped sensille (s”) and the opening of the salivary duct (d) near the middle of its anterior border. The upper surface of the short, rounded cranium bears a Fic. 1. Adult larve of Myrmecia sanguinea Fabr. pair of minute antennal rudiments (#). When I collected this larva in New South Wales I was unable to learn anything of its feeding habits. Indeed, he who would make such observations would have to don a suit of armor specially designed to ward off the stings of this powerful and ferocious ant or be able to keep it in a large artificial nest. As I was continually travelling about I was unable to resort to the latter alternative. It is, however, not improbable that the Myrmecia larva is fed on whole insects, since the small head and very long mobile neck are very much as in certain solitary wasp larve (e. g., Sphecius), which gnaw a small hole in their prey and oS ge 298 WHEELER—ANT LARVE, then reach into its body cavity and devour its soft parts. . mandibles of the Myrmecia larva certainly show that it feeds insect food. ) ne The second larva (Fig. 3) is that of the “ Matabele ant,” Me ponera fatens, of which Mr. Lang secured many specimens in Belgian Congo. Arnold (1914) and others have shown that a | Fic. 2. Head of Myrmecia sanguinea larva. A, dorsal; B, ventral : lateral view; m, antenna; c, clypeus; m, mandible; +, maxilla; s, maxillz sensilla; 1, labium; s”, labial sensilla; d, opening of salivary duct. Nas ant preys on termites, the bodies of which it carries home agglu nated in the form of pellets (Alluaud in Santschi, 1914). It is 1 restless and changes its nesting site frequently, so that it is oblig to carry its young about a great deal and for considerable distan WHEELER—ANT LARV2. 299 The larva is cylindrical, covered with a very tough, opaque, grayish, “hairless skin and furnished with long, falcate mandibles. The pupa is enclosed in a very tough, black cocoon. These peculiarities are evidently adaptations to exposure to the air and light, to the exigen- cies of frequent and protracted transportation and to feeding on the bodies of termites brought into the nest by the workers. Mr. ‘Lang actually observed the exposure of the black cocoons to the sunlight, a peculiarity of behavior which I had also observed in cer- tain Australian Ponerine of the genera Diacamma and Rhytido- _ponera (1915). The third larva (Fig. 4), that of Bothroponera sublevis, one of four species of the genus, which I collected in Australia, has a very broad elliptical body, with a short, stout neck, strongly folded over onto the ventral surface, which is somewhat concave. The integu- — ment is also hairless and of a peculiar opaque, gray color. The sides of the three thoracic segments and first abdominal segment are furnished with fleshy tubercles and the mouthparts are very _ highly developed. It is placed on its back by the nurses and fed with fragments of insects deposited on its trough-like ventral sur- _ face as in our North American Ponerine.* The feeding of the larve with pieces of insect food is not, how- ever, confined to the Ponerinz. Miss Fielde and I have shown that one of the commonest Myrmicine ants of the North Eastern States, Aphenogaster fulva, has the same habit. During late June, at the height of the breeding season, it is hardly possible to remove the ____ Stone covering a nest of this ant without finding one or more larve lying on their backs or sides in the act of feeding on the legs of __ flies or fragments of other small insects. Janet has observed simi- 3 Mayr described Bothroponeraas a genus, but Emery, Forel and Santschi have been treating it as a subgenus of Pachycondyla. I return to Mayr’s conception, because the adult, at least, of the Australian species of Bothro- ponera exhibits several peculiarities of behavior, such as the extrusion when captured of a mass of frothy bubbles from the tip of the gaster, and because of the structure of the larva, which is very different from that of Pachy- condyla as will be seen by comparing Fig. 4 with my previously published figures of P. montezsume. The larve of Diacamma, Leptogenys and Odontomachus bear a greater resemblance to those of Pachycondyla. Bothroponera is, moreover, confined to tropical Africa, Asia and Australia, whereas Pachycondyla is neotropical. 300 WHEELER—ANT LARV2. lar behavior in Tetramorium ce@spitum and in some Camp ants of the genus Lasius. Hungry larve of Aphenogaster w attack and devour smaller larve of their own species that lie reach of their sharp mandibles. Still the very young larve of Aphenogaster and pesily Fic. 3. A, nearly adult larva of Megaponera fetens Fabr.; B, head of dorsal view. of the Ponerine are fed with liquid food regurgitated on mouths by the workers. Miss Fielde thus describes the eis Aphenogaster (1901) : | The feeding of the larva, which is bent nearly double in the egg, W regurgitated food begins as soon as it straightens itself and protruc mouth. When the larve begin to appear in the egg-packet, the workers the packet and hold it free and still, while one of their number holds a t lucent white globule of regurgitated food to the larval mouth projecting WHEELER—ANT LARV2. 301 the surface of the egg-packet. I have repeatedly seen the workers thus feed- _ ing the very young larve, a single globule of regurgitated food serving for __ a meal of which four or five larve successively partook. __ Undoubtedly the majority of Myrmicine, Dolichoderine and Cam- " ponotine, the three most highly specialized subfamilies of ants, feed the brood throughout its larval stages with regurgitated liquids. Concerning larval feeding in the Doryline nothing is known. I come now to a consideration of some of the ant larve collected __ by Mr. Lang in the Belgian Congo. Four of these, all belonging to - the subfamily Myrmicine, are of unusual interest. One of the "species is a new Pedalgus which I shall describe elsewhere as P. _ termitolestes sp. nov., the third of the genus to come to light, as _ only one Indian and one other West African species were previously known. The workers of termitolestes are minute brownish yellow q ants which live in the masonry of large termite hills and undoubtedly 4 prey on their inhabitants. Their habits therefore resemble those of ¢ a the well-known thief-ants, Solenopsis molesta of North America and Ss. fugax of Europe.- The larva (Fig. 5) has a singular shape, _ being almost spherical, with a short neck, small head and minute, | bidenticulate mandibles. The delicate integument is studded with i very short, stiff hairs, each of which has two recurved branches. : a ‘The larve, which are held together in compact masses by the inter- _ locking of these hooked hairs, are fed with liquid food by regurgi- 3 tation as is evident from the contents of their large spherical : 7: stomachs and the very feeble development of their mouthparts. t £ _ Altheugh, like other Myrmicine, they do not spin cocoons but form naked pupz, they nevertheless possess huge salivary glands. Even _ in the very young larva (Fig. 5A) the salivary receptacle on each _- side is full of a clear liquid secreted by the large cells of the two branches of the gland. In the nearly full-grown female larva (Fig. _ 5B) the glands are very voluminous and have their lumen and that ___ Of the receptacle full of secretion shown as dark, compact masses in _ the figure, which was, of course, drawn from a specimen hardened and dehydrated in alcohol. As such an amount of saliva would _ hardly be necessary for digestive purposes and as it is not used in _ the form of silk by the full-grown larva, it probably serves as a store of food for the nurses. The Pedalgus larve, therefore, would etc wee etna] sc inpepittian ame 2 802 WHEELER—ANT LARVZ. seem to resemble the repletes of honey ants (Myrmecocystus, 1 tomyrmex, etc.), except that the food for the workers is metabo and stored as saliva by the larva, instead of merely being ingu: > ibe wes Fe Fic. 4. Adult larva of Bothroponera sublevis Mayr. A, ventral He view; C, head, dorsal view; D, head, in profile. tated and stored in the ingluvies, or crop by a certain num workers. From the fact that other Myrmicine ants, although spin no cocoons, often have well-developed salivary glands, w infer that these organs have much the same function as in Pea To prove this, however, additional observations are neces though other reasons for believing it to be the case, will app ear the sequel. In all the larval stages of the Dolichoderinz and in immature larve of Camponotine the salivary glands are fp put to a similar use. In the species of Ecophylla and Polyrhi that employ their young larve for spinning the silken portion WHEELER—ANT LARVZ. 303 the nest inhabited by the whole colony, we must suppose that the - spinning habit, which in other Camponotine ants is manifested only _ atthe end of larval life, has become secondarily precocious, but even 4a in such larve the saliva in the stages immediately after hatching 4 may, perhaps, still serve as an agreeable draught for the nurses. __ The three remaining larve which I wish to describe belong to _ species formerly included in the genus Sima but now for taxonomic _ Fic. 5. A, very young; B, nearly adult larva of Pedalgus termitolestes sp. nov.; lateral view to show the development of the salivary glands. _ reasons referred to Tetraponera and Pachysima. These ants live _ in hollow twigs like the species of the closely related neotropical _ genus Pseudomyrma. A large collection of Tetraponera tessmanni, . made by Mr. Lang, included larve and pupe in all stages of de- 804 WHEELER—ANT LARV2. velopment. The adult larva differs little from the youn shown in Fig. 6. It is long, cylindrical and hypocephalic, the head on the ventral side instead of being terminal. i various larve described above, it has a pair of swollen’ Fic. 6. Adult larva of Tetraponera tessmanni Stitz. — head, and a large protuberance, evidently representing fused appendages, on the ventral side of the first abdominal The dorsal surface is furnished with long, hook-shaped means of which the larva is evidently suspended from the the nest in the same manner as some of our American ant (Pheidole, Leptothorax, etc.) which have similar dorsal WHEELER—ANT LARVZ. 305 NEE PTR LY SMAI RTD NGS ape larve of Tetraponera (natalensis, allaborans, etc.) are not e those of T. tessmanni. The meaning of the thoracic and abdominal appendages becomes clear when we examine the larve of Pachysima ethiops and lati- ons. Four distinct stages, probably separated by moults, or ec- dyses, may be recognized in e@thiops. The first stage larva, just _ after hatching, is represented in Fig. 7 as it appears in ventral and “SS ‘Fic. 7. First larva stage (“trophidium”) of Pachysima ethiops F. Smith. A, ventral; B, lateral view. ‘lateral view. The body is curved, convex dorsally and concave ventrally, and terminates behind in a cylindrical projection, with the anus shifted to the ventral surface near its base. The creature is _ strongly hypocephalic like the larval Tetraponera and Pseudoponera. The head is surrounded by a cluster of prominent, tubercle-like ap- -pendages. On the prothorax, which is large and forms a great hood over the head, there are three pairs of these appendages, an anterior : truncate pair, a median pointed pair and a large posterior pair, 306 WHEELER—ANT LARVZ. swollen and rounded and embracing the sides of the head. — evidently correspond to the single prothoracic pair of the ponera tessmanni larva. The mesothoracic segment has a pa smaller appendages nearer the mid-ventral line. Between arises a very peculiar organ with a swollen, pear-shaped base pro longed into a slender, apparently erectile, tentacle-like process 1 ni extends up in front of the head and terminates in a small im The first abdominal segment bears a pair of large swollen a dages lying at the base of the mesothoracic pair and united y large and very prominent mid-ventral tubercle. This tubercle its lateral appendages are represented in the T. tessmanni lary the others, with the exception of the third prothoraciec pair, a sent. Sections and stained, cleared preparations of the He 2. show that the various tubercles contain portions of the fatb least in the bases of their cavities, and next to the hypo dense, granular substance, evidently a coagulated liquid proc by the underlying adipocytes, or trophocytes. The same ic fills the unpaired tentacle, except its pear-shaped base, whic tains fat cells. Around the bases of the tubercles are mus arranged that their contraction must increase the pressure on and granular liquid and in all probability cause the latter to through the hypodermis and delicate chitinous cuticle onto face. The whole arrangement of the tubercles, in fact, co n system of exudate organs, or exudatoria, as I shall call adapted to secrete substances that can be licked up by the ants they are feeding and caring for the larve. In this stage the : dibles are small, soft, blunt and unchitinized so that the larva be fed with regurgitated liquid food. The labium has a pe pair of fleshy appendages, shown just beneath the mandibles The body is naked, except for a few sparse, pointed bristles dorsal surface and the median pair of prothoracic appendages. nothing like this larval stage is known among ants or inde the Hymenoptera, I propose to call it the “trophidium.” The second stage larva is shown in Fig. 84. The various | toria are smaller in proportion to the remainder of the body still much like those of the trophidium. The body is more el the mandibles are more pointed and distinctly falcate, but eve WHEELER—ANT LARV2. 307 this stage they are unchitinized and therefore nonfunctional. The coarse hairs are visible on the dorsal surface but a more uniform _ investment of small hairs has made its appearance. They are blunt - or even clavate, especially on the prothoracic segment. In this and the trophidium stage I am unable to find any salivary glands in cleared preparations though rudiments may, perhaps, be present. Fic. 8. A, second, B, third and C, fourth (adult) larval stages of Pachysima @ethiops F. Smith, The third stage larva (Fig. 8B) is larger and very regularly ellip- tical. The exudatoria can all be recognized, except the unpaired tentacle. It is, however, present in some of the younger individuals but in a greatly reduced and vestigial condition and at the bottom of the deep depression which now appears as a definite pocket just back of the mouth and under the midventral swelling of the first ab- dominal segment. In many larve I found in this pocket a small, 308 WHEELER—ANT LARVA. rounded, dark-colored pellet, which puzzled me at first. In se however, it was at once seen to consist of the triturated and pacted bodies and parts of small insects. It is, in fact, a food-p placed by the worker ants in the pocket just behind the 1 mouth. In this stage, therefore, the larva is fed on solid food the strongly chitinized, acute and bidentate mandibles corrob this statement. Slender salivary glands may also be detected, cating that the substance of the food-pellet is subjected to ext testinal digestion. The longer hairs on the dorsal integumen’ almost completely disappeared. The first pair of appendages on prothorax has disappeared and the second pair is obsolescent. — In the fourth, or adult stage (Fig. 8C) the larva is more | and cylindrical and much more hypocephalic, the prothorax fe a great protuberance in front of the head. The exudatoria are s recognizable, with the exception of the first and second prothora pairs, which have disappeared entirely. The labial appen are reduced. A food pellet was found in the postcephalic in several of the larve of this stage but is not represented figure. The coarse hairs have disappeared from the integ which is now uniformly covered with very short, delicate hai the structure of the posterior end of the body is very differ t | that of the preceding stages. _ We owe the only account of the ethiops larva in the litera Emery (1912). He describes what corresponds to my fourth larva very briefly and figures its anterior end with some | exudatoria, but erroneously interprets the large prothoracic Da “ ébauches de pattes,” or rudiments of the anterior pair of i legs.* | The larve of Pachysima latifrons are quite as extraor: as those of @thiops and also pass through four stages. The phidium, or first stage, shown in Fig. 9, is very hypocephalic, the pt 4In the same paper Emery created the subgenus Pachysima for the commodation of what was formerly called Sima ethiops and for species described as latifrons, because they have the frontal carine worker and female much more widely separated than in the numerous of Tetraponera (Sima auctorum). I have raised Pachysima to generic because the larve of the two species are so very different from th Tetraponera. WHEELER—ANT LARV2. 309 thoracic segment being greatly enlarged and projecting anteriorly. Both preparations stained in toto and sections show that the portion of the fat-body in this segment is sometimes heavily charged with urate crystals, so that it undoubtedly functions as a ‘storage kidney | till the Malpighian vessels are sufficiently developed to excrete. The _ first and second pairs of prothoracic appendages of the ethiops - larva are absent, but the third pair is very large and embraces the _ sides of the head. The meso- and metathoracic segments each bear a pair of slender pointed appendages, the first abdominal segment a ‘Fic: 9g. First larval stage (“trophidium”) of Pachysima latifrons Emery. A, ventral; B, lateral view. huge leg-like pair, which are swollen and fusiform at the base and run out into a long slender process which forms an obtuse angle with the basal portion. The sternal region between these appen- dages is protuberant and its cuticular covering, like that of the four pairs of appendages is minutely prickly, unlike the smooth cuticle of the remainder of the body. Sections show that both the appen- _ dages and the sternal swelling are exudate organs, though the pro- _ thoracic and abdominal pairs are much more important than the _ others. The prothoracic appendages are filled with blood and very PROC. AMER. PHIL. SOC., VOL. LVII, V, AUGUST 8, I918. 310 WHEELER—ANT LARVA, little fat tissue, but their hypodermis is much thickened and of crowded cells arranged in peculiar clusters. In section t dominal appendages appear as in Fig. 10, The fusiform 1 filled with large, clear trophocytes, or fat-cells, some of which in middle of the swelling may contain urate crystals, like those in prothoracic storage kidney, but the slender, tubular distal po contains a granular liquid, which can only be regarded as an ext Fic. 10. Longitudinal section through exudatorium of first abdominal eg of trophidium of Pachysima latifrons Emery. . derived from the trophocytes in the basal enlargement. — This date is evidently filtered through the thin cuticula covering 1 pendage by pressure, for there is a rather elaborate system muscles, as in the ethiops larva, surrounding the bases of the appen dages and capable of subjecting their contents to pressure. . head is small and has soft, blunt, rudimentary and unchitinized 1 dibles, and the labium bears a pair of long, palp-like append: which project forward in the deep depression between the head the swollen sternal portion of the first abdominal segment. 1 are probably also exudatoria and seem roughly to correspond to unpaired tentacle of the ethiops larva. The structure of the mo parts shows that the larva in this stage is fed with liquid food gurgitated by the workers. The convex dorsal surface is beset sparse, curved bristles of uniform thickness, with blunt tips. a: segmentation of the body is indistinct and its posterior end cu forward and terminates in a large tubercle with the anal orifice anterior to its base. Fig. 9B, drawn from a stained and cle * re: Ey ; * WHEELER—ANT LARV#. 311 preparation, shows the nervous system and alimentary canal. The Malphigian vessels have only just begun to develop at the blind end of the proctenteron where it abuts on the posterior end of the large, elliptical mesenteron, or stomach, but no salivary glands can be de- tected. In the second stage larva (Fig. 114) the body is more elongate and cylindrical and the four pairs of appendages can still be recog- Fic. 11. A, second, B, third and C, fourth (adult) larval stages of Pachysima latifrons Emery. nized though considerably smaller in proportion to the remainder of the body. The mandibles are becoming chitinized. Many of the long hairs on the dorsal surface are still present but a general covering of short, sparse hairs has made its appearance. The third stage larva (Fig. 11B) is larger and still more elongate 312 WHEELER—ANT LARV 2. and cylindrical and shows a further regressive development of | exudatoria. Those on the meso- and metathoracic segments disappeared and the abdominal pair has short broad bases with t distal portions attenuated to slender points. The labial append have also disappeared. The mandibles are well chitinized and | larva is now fed with pellets of crushed insects, like the et larva in the corresponding stage. These pellets were found in in several of the alcoholic specimens as represented in Fig. 11 The pellet lies in the deep pocket between the head and the s protuberance of the first abdominal segment and is therefore easy reach of the mandibles and labium of the larva. Cleared arations show that the salivary glands have made their appez though they are small and slender. The anterior end of the fourth stage, or adult larva is sham Fig. 11C. The exudatoria of the prothoracic segment now a merely as a pair of welts or folds embracing the sides of the | and continuous with the more dorsal portions of their segn _ which is relatively smaller and less projecting than in the pre : stages. The exudatoria of the first abdominal segment are ae tinct but their distal portions are reduced to mere points, s absent in larve just before pupation, and the sternal swelli much less prominent. In this stage the larva resembles that Tetraponera throughout its various stages. In the third and stages of the latifrons larva, as in the corresponding stag ethiops, the salivary glands probably furnish secretions which a useful both in the extraintestinal digestion of the food pellet < substances that can be imbibed by the workers. The fact that two species of Pachysima the exudatoria decline pari passu wi development of the salivary glands certainly suggests that both of organs have to some extent a common function. In fort Pedalgus and probably many other Myrmicine, in which - velopment of the salivary glands is more precocious, the exuc are not developed. I believe, therefore, that we must interpret the exudatoria as’ primitive glands, but they differ so much from the ordinary hypo mal glands of insects that it will be necessary to consider them closely before proceeding further. They are, as we have seen, WHEELER—ANT LARV#. 313 _ divertictila like the embryonic legs, consisting of hypodermis and its overlying cuticula and containing a portion of the fat-body sep- arated from the hypodermis by a granular liquid. Now the fat- body of insects may be regarded as a diffuse ductless gland, the cells (trophocytes) of which take certain substances from the blood in _ which they lie, store them in the cytoplasm as fat-globules or proteid granules and later return them to the blood in a more finely divided, if not chemically modified form. The exudate which accumulates in the distal ends of the exudatoria is therefore merely blood charged with nutrient substances from the fat-cells, and either filters gradu- ally through the-hypodermis and overlying cuticle or is forced through them by muscular pressure. At first sight it would seem that the cuticle must be impervious to such a liquid, but a considera- tion of the more recent work on the minute structure of chitin by Holmgren (1901, 1902), Biedermann (1902, 1903), Kapzov (1911), Casper (1913) and others shows that there is nothing to prevent the _ passage of a thin fatty liquid, even if it were not under pressure _ and even if the cuticle were much thicker than it is in the ant larva. _ The cuticle is a colloid, either of a reticular structure, as Kapzov be- _ lieves, or formed of horizontal layers of very fine fibrille crossing one another at an angle of 60° as most investigators, including _ Biedermann and Casper, maintain. Between the fibrille are regu- larly distributed and extremely fine openings or “pore canals,” through which a liquid might readily pass as if the cuticle were a 5 The question arises as to whether the larval exudatoria of Pachysima are the homologues of the true appendages on the thoracic and first abdominal segments of embryo insects. In other words, do the exudatoria represent true legs or are they new formations? The trophidium of P. latifrons seems _to point to the former alternative. The large leg-like exudatoria on the first abdominal segment are certainly very suggestive of the embryonic “ pleuro- podia” to which I devoted a paper many years ago (1890). On the other hand, the four pairs of trophidium appendages in latifrons seem to be homologous with the four pairs of papille in the larva of Bothroponera (Fig. 4), and the latter are almost certainly merely remnants of a consider- able number of similar papillae which are scattered over the whole surface is of the larval Pachycondyla, Diacamma and Ponera. Furthermore, two of _ the pairs of exudatoria on the prothorax of the Pachysima ethiops tro- phidium and the unpaired tentacle-like exudatorium just behind the head’ cannot be brought into the homology. It would seem, therefore, that the _ exudatoria must be regarded as ccenogenetic, or new formations peculiar to _ the young larve of certain Old World genera of Pseudomyrmini. 314 WHEELER—ANT LARVZE. filter. Even in the case of the hypodermal glands fatty liquid known to pass through the thin chitinous cuticle with which secreting surface of the cytoplasm is always covered, even Revie ends of the ducts are intracellular. v3 In this connection attention may be called to a very similar ¢ dation of blood plasma charged with certain substances (e. g., ¢ tharidin) through the hypodermis and cuticle in many Mel Cantharid, Lampyrid, Coccinellid and Chrysomelid beetles. It long been known that when these insects are roughly handled t discharge from the articulations of their legs a white, yellow greenish, bad-smelling liquid, which Magretti (1881, 1882), I (1895), De Bono (1889) and Berlese (1909) have shown to blood plasma. It accumulates in pockets at the articulations a passing through the integument and leaving the blood cells (ame cytes) behind, and is clearly an exudate though it is repugnator E instead of having an alluring or nutrient function like the plasma of Pachysima. omy It is unnecessary, however, to seek confirmation of my in pretation of the circumoral appendages of the Pachysima larva b merely pointing to the conditions in the Meloids, Coccinellids, Wasmann, Holmgren and Tragardh have published valuable studi of exudate organs much more like those of the ant larve. To W: mann (1903) belongs the credit of having first made an exte investigation of the trichome glands and exudatoria of num myrmecophiles and termitophiles. Many of these structures more or less modified hypodermal glands with tenuous ducts | ing at the base of hairs (trichomes) which either diffuse the s 3 tion so that it can evaporate quickly or spread it out so that it ¢ readily licked up by the ants, but in such termitophiles as the S linid beetle Xenogaster inflata, the fat-body in certain parts o abdomen forms “ blood tissue,” which becomes the “ exudate passing through a layer of hypodermis at the base of p (“exudate buds”).® The latter seem to consist of cuticular stance perforated by delicate canals that conduct the exudate | surface. Wasman says: | The exudate of these buds seems therefore to be a component blood fluid, which is as it were filtered through the hypodermal layer. 6 The trichome glands may be compared with similar structures in oth WHEELER—ANT LARVZ. 315 _ Some of the details in his account and figures are far from clear, _but there cam be no doubt about his meaning. Equally interesting is his description of the larve of certain symphilic myrmecophiles _ (Lomechusa, Atemeles and Xenodusa) concerning which he writes: The cuticula of the whole body, excepting the head, is membranous and ; whitish. Outer exudate organs (i. e., trichomes) are lacking. The exudate tissue is exclusively the fat-body. He believes, in other words, that in these larve the voluminous fat- E body functions as a huge exudatorium which pours a fatty exudate : onto the surface of the body. This at once suggests that in many ant larve the general fat-body may have the same function, so that there would be in these insects three possible sources of liquid sub- stances agreeable to the worker ants. the salivary glands, the exuda- developed in any given species, but at any rate there is just as much ‘reason for supposing that the general fat-body may function as an exudate organ in the ant-larva as in the larve of the Lomechusine e myrmecophiles. Kriiger (1910) and Jordan (1913) have cleared up some of the obscurities in Wasmann’s paper, especially in regard _ to the trichome glands of hypodermal origin, but in my opinion have _ not invalidated his general conclusions in regard to the role of the a fat-body and blood in exudation.” animals. Many nonmyrmecophilous insects have similar glands that serve to _ diffuse sexually attractive secretions. The question arises as to whether many of the hair-tufts in mammals may not have an analogous function. Anthropologists seem not to have explained the retention of hairs in the axillary and pubic regions of man. It is evident that the hairs in the arm- pits serve rapidly to diffuse and evaporate the secretions of the sudorific glands. _ The pits full of trichomes on the thorax of many symphilic Paussid beetles are strangely suggestive in this connection. The function of the public hairs is not so clear, but perhaps certain bats which have peculiar tufts about the genitalia (see, e. g., the figures of the Congolese Hipposideros langi Allen in _ Bull. Amer. Mus. Nat. Hist., 37, 1917, pp. 436, 437) may indicate that in the remote past the pubic hairs had a sexual function in the ancestors of man. 7 For a critique of Jordan’s work and for further discussion of the struc- ture and development of Lomechusa and Atemeles the reader is referred to _Wasmann’s recent monograph: “ Neue Beitrage zur Biologie von Lomechusa und Atemeles” (1915). toria sensu stricto and the fat-bo¢y. They would not all be equally —— 316 WHEELER—ANT LARVZE. Wasmann has shown in a number of papers that the tr of termites, the symphiles, are physogastric, i. ¢., have the enormously distended with fatty tissue. This condition striking in certain Staphylinids (Xenogaster, Corotoca, Sp Termitomimus, etc.) and Diptera (Timeparthenus, Terr Termitoxenia, Thaumatoxena, etc.). Tragardh (1907) has s! sections of the beetle Termitomimus, which lives: in me side I quote the greater part of it: The relation of the fat-body to the hypodermis and the cuticle ferent in different parts of the body. 1. The hypodermis is exceedingly thin, sometimes scarce dl and pressed close to the cuticle by the underlying fat-body. The pr no distinct endostracum and is penetrated by an immense number tremely fine pores, arranged in transverse rows. This is the case ventral, lateral and posterior part of the pseudoabdomen, i. ¢., exactly the cuticle is of a bright reddish-yellow color (“symphilous col mann) and where the termites may most easily get access to it. — 2. The hypodermis is thick and withdrawn from the cuticle y thicker, with well-developed epiostracum and, endostracum, ‘eaten 2 wide space, which is filled with liquid. . . . The fat-body is contiguous hypodermis. The space between the cuticle and the hypodermis is less filled with a cyanophilous tissue of a spongy appearance which si exhibits a very distinct radial structure, sometimes is concenheiial fied and contains numerous granules which are also to be found in chogenic cells. This is evidently a fluid, which has either passed thr hypodermis and is a derivate from the fat-body or it is a secretion pr by the hypodermis and is coagulated by the method of fixation. ... The above stated facts concerning the relation of the fat-body hypodermis and the cuticle differ in some essential respects from what mann has found in the termitophilous physogastrous insects studied b In Spirachtha, Termitoxenia, the larve of Orthogonius and Glyptus, gaster and other Aleocharini the hypertrophied fat-body is always surr by large tracts of blood-tissue, consequently the exudation is deriv rectly from the blood-tissue and only indirectly from the fat-body. exudation is no fluid but evaporates through the membranous cutee has no pores. To support the theory of the exudation being only an attractive od not offering the termites any source of subsistence Wasmann points fact that the symphili as a rule only occur in small numbers in the nests. These statements, however true they may be with regard to the mentioned genera, do not apply at all to Termitomimus. In this genus WHEELER—ANT LARVZ. 317 “contrary, in the part of the abdomen which is easiest accessible to the termites, iz., the ventral, lateral and posterior side of the pseudoabdomen, 1. The fat-body is not surrounded by the nied -tnewe but contiguous ie extremely thin hypodermis and 2. The cuticle is penetrated by an immense number a pores (and the endostracum is not distinctly discernible). _ 3. Furthermore Termitomimus does occur in great numbers in the nests £ the termites. _ These facts seem to me to suggest that in Termitomimus the fat itself or a derivate of the fat-body may possibly be secreted as a fluid through -mumerous pores of the cuticle and not merely evaporate and that Ter- ‘omimus thus offers to the termites a source of subsistence. The com- | paratively very large extension of the area of the cuticle which exhibits this” structure also argues in favor of this theory. In another paper (19072) Tragardh describes a peculiar Tineid caterpillar with exudatoria even more like those of the Pachysima larva. He found it in the tree nests of Rhinotermes in Zululand. The relations between the caterpillar, which feeds on the woody sub- stance of the nest, and the termites are evidently friendly. When disturbed, the larve were seen to make their way to other parts ff the nest, coming along one after the other, with regular intervals, like in ‘procession, each larva being escorted by a few soldiers and workers. ich of the seven anterior abdominal segments of the caterpillar on its sides a pair of long, tapering appendages, which ragardh regards as exudatoria and each appendage contains a lobe the fat-body, surrounded by blood. The imperforate hypodermis ; separated from the thin cuticle, the space between being filled with exudate. In this case he believes that the exudate must evap- orate on the surface of the body, since he says: _As the larva emits a strong odor, and the termites were scarcely seen touching the appendages, the exudation is very likely an alluring odor. He compares the organs with the various osmateria described by ® Packard i in the caterpillars of Megalopyge and Hemileucide. __._ Certain organs in the larve of two groups of Hymenoptera may also be interpreted as exudatoria. In 1907 I called attention to peculiar blister-like organs on the sides of pseudonymphs of certain Eucharine parasites of ants, notably in Orasema. These structures are shown in Figs. 19 and 21, PI. 2 of the paper referred to and in Fig. 251 F, G, p. 415 of my ant book (1910). In the pupa of the 318 WHEELER—ANT LARVZ. same insect the abdomen has similar organs in the form of tr; verse welts. Reichensperger (1913, Pl. 6, Fig. 12) describes figures the very same organs in an Abyssinian Eucharine, ase fraudulenta, which lives with Pheidole megacephala, and sg) that they may be exudate organs. Forel (1890) had prey: mentioned similar structures (“asperités et boussoufflures ”) « pupa of the large Eucharis myrmecie taken from the cocoon of Australian bulldog ant, Myrmecia forficata. On recently reé ing my preparations I find that the organs of Orasema viridis be interpreted as exudatoria. They are knob-shaped, with v hypodermis and cuticle and are filled with blood but contain no tissue, although the fat-body in the abdomen and thorax is voluminous. In life the knobs are colorless and glistening. the pseudonymphs and pupz are assiduously licked by th Pheidole instabilis, so that the knobs of the former and the y the latter probably produce substances agreeable to the ants. The other group of Hymenoptera comprises the singular African bees of the genus Allodape. In 1902 Brauns show they make very primitive nests, consisting of a single cavity, 12 cm. long, in the stems of various Liliaceous plants, but un other solitary bees, feed their larvee from day to day with i brei” (honey-soaked pollen?). In the warmer portions | Colony and German Southwest Africa Allodape breeds throu the year. The single cavity of the nest contains eggs, larvae stages, pupze and freshly emerged bees intermingled. The la are unique among bees in possessing peculiar tubercles on the 7 of the fifth to tenth segments. Friese (1914) publishes photogr, of some rather shrivelled half-grown larve and describe: bercles as “bladderlike evaginations of the outer skin.” | seems to regard them as legs (pseudopods) and says that t used to hold the food, but it seems probable that they are 8 While this paper was in the hands of the printer Dr. R. J. New South Wales sent me the larve, pupe and an adult male of a h described Eucharine, which he found attacking the brood of the r ant (Myrmecia gulosa). Prof. C. T. Brues believes that the par belong to the genus Psilogaster and will describe it in the near future. larve and pupe are covered with exudatoria like those of Orasema bu prominently developed. WHEELER—ANT LARV2. 319 exudate organs. If this proves to be true, the resemblance of Allo- dape to Pachysima, which also rears its brood in all stages in hollow ‘stems and feeds the older larve with food-pellets, would be very striking. Allodape is also of considerable interest im connection with Roubaud’s observations on the wasp Synagris which will be consid- ered in the sequel. : More important in their bearing on the exudate organs of Pachysima are Holmgren’s observations on the termites. He de- ‘votes the twelfth chapter of his volume (1909) on the anatomy of these insects to the exudate tissue. Termites are really themselves physogastric like their guests, and Holmgren shows that all the castes, but especially the queens, have extensive exudate tissues, consisting of the peripheral layers of the abdominal fat-body. In these layers the trophocytes do not contain fat-globules but nu- ‘merous minute granules which are discharged into the blood and thus convert it into the exudate that passes through numerous pores or lacunz in the chitinous cuticle to the surface. There it is licked up by other members of the colony. He finds that the development of the exudate tissue differs considerably not only in the different castes but also in their various developmental stages and that the intensity of the licking and feeding of the individuals of a termite * colony is directly proportional to the amount of their exudate tissue. Those with the largest mass of exudate tissue are the best fed and the most licked. In other words, the care bestowed by the workers on the various members of the colony is not an immediate expression of an altruistic philoprogenitive ‘‘mstinct (Brutpflegeinstinct), but depends essentially on egoistic motives, 4. e., exudate hunger. _ To this point I willingly follow Holmgren, but both he and Was- _ mann have used their respective observations as a basis for what seem to me to be rather dubious speculations, a consideration of _ which will have to be deferred till the more general part of my dis- cussion is reached. Escherich (1911) gives a more vivid, not to say more spectacular _ account of the exudate hunger of termites. So eager are the F. workers of the Ceylonese Termes redemanni for the exudate of their huge physogastric queen that they actually tear little strips out of her cuticle in order to get at the liquid more readily! _Escherich found that old queens sometimes have their white ab- 320 WHEELER—ANT LARVZE. domens dotted with the small brown scars of the wounds 1 flicted by their progeny. Here the feeding behavior of the and offspring is the reverse of that in incipient ant colonies, the queens are fed with regurgitated food by the workers a the latter with exudates, but this is, in all probability, also the in established ant colonies when the workers have matured and queen no longer feeds the brood. The facts collated in the foregoing paragraphs relate to the date organs, but we had previously seen that the salivary glan larval ants probably subserve a similar function in the life o colony in addition to digesting proteid foods extraintestinally producing silk at the time of pupation. The question arise whether there is any evidence that in other groups of social the salivary glands of the larva produce substances which are sumed by the worker nurses. Fortunately there are some very tinent observations at hand in the French literature which i is in splendidly original works on the habits and taxonomy of The observations to which I refer relate to the social wasps. Buysson (1903) observed that the larve of Vespa “ secret the mouth an abundant liquid. When they are touched th is seen to trickle out. The queen, the workers and the mal very eager for this secretion. They know how to excite the spring in such a way as to make them furnish the beverage. Janet (1903) was able to prove that the secretion is a pra the salivary, or spinning glands and that it flows from an op the base of the labium. “This product,” he says, “is often i by the imagines, especially by the just emerged workers and males, which in order to obtain it, gently bite the head of the The most illuminating study of this matter, however, is | in a fine paper by Roubaud on the wasps of Africa (1916, account of the primitive wasps of the genus Belonogaster pre striking picture of one of the earliest stages in the social li wasps, as will be seen from the following quotation : In the species of Belonogaster as well as in those of the genera and Polistes we have been able to observe this proceeding in detail. larve, from birth, secrete from a projection of the hypopharynx, on ferior surface of the buccal funnel, an abundant salivary liquid, whick the slightest touch spreads over the mouth in a drop. All the adult 1 WHEELER—ANT LARVZ. | 321 males as well as females, are extremely eager for this salivary secretion, the taste of which is slightly sugary. It is easy to observe, especially in Belonogaster, the insistent demand for this larval product and the tactics employed to provoke its secretion. As soon as a nurse wasp has distributed her food pellet among the va- rious larve, she advances with rapidly vibrating wings to the opening of each cell containing a larva in order to imbibe the salivary drop that flows abun- dantly from its mouth. The method employed to elicit the secretion is very easily observed. The wing vibrations of the nurse serve as a signal to the larva, which, in order to receive the food, protrudes its head from the orifice of the cell. This simple movement is often accompanied by an immediate w of saliva. But if the secretion does not appear the wasp seizes the larva’s head in her mandibles, draws it towards her and then suddenly jams it back into the cell, into which she then thrusts her head. These movements, wolving as they do a stimulation of the borders of the mouth of the larva, compel it to secrete its salivary liquid. ‘One may see the females pass back and forth three or four times in front of a lot of larve to which they have giveri nutriment, in order to imbibe the secretion. The insistence with which they perform this operation is such that there is a flagrant disproportion between the quantity of nourishment dis- tributed among the larve by the females and that of the salivary liquid which they receive in return. There is therefore a real exploitation of the larve by the nurses. . _. The salivary secretion may even be demanded from the larva without a compensatory gift of nourishment, both by the females that have just hatched and by the males during their sojourn in the nest. The latter employ the same tactics as the females in compelling the larve to yield their secretion. They demand it especially after they have malaxated an alimentary pellet for themselves, so that there is then no reciprocal exchange of nutritive material. It is easy to provoke the buccal secretion of the larvz artificially. Merely touching the borders of the mouth will bring it about. The forward move- ; _ ment of the larve at the cell entrance, causing them to protrude their mouths 3 _ to receive the food pellet, is also easily induced by vibrations of the air in the neighborhood of the nest. It is only necessary to whistle loudly or emit 3 shrill sounds near a nest of Belonogaster to see all the larve protrude their heads to the orifice of the cells. Now it is precisely the vibrations of the air _ ¢reated by the rapid agitation of the bodies of the wasps and repeated beating f their wings that call forth these movements, either at the moment when food is brought or for the purpose of obtaining the buccal secretion which is ‘so eagerly solicited. Roubaud summarizes the general bearing of his observations in the following paragraph: __ The reciprocal exchange of nutriment between the adult females and the Tarver, the direct exploitation of the larval secretion without alimentary com- Pensation by the males and just emerged females are trophobiotic phenomena ‘the elucidation of which is of great importance to an pe eending of the 322 : ._ WHEELER—ANT LARVZE. origin of the social tendencies in the Vespide, : as we shall show in poe The retention of the young females in the nest, the associations isolated females, and the coéperative rearing of a great number of all rationally explained, in our opinion, by the attachment of the v larval secretion. The name e@cotrophobiosis (from olkos, family, given to this peculiar family symbiosis which is characteciebs by re exchanges of nutriment between larve and parents, and is the raison d the colonies of the social wasps. The associations of the higher Vespids in our opinion, as its first cause the trophic exploitation of the larvae adults. This is, however, merely a particular case of the trophol which the social insects, particularly the ants that cultivate aohida coccids, furnish so many examples. * It does not seem to me that the term “ cecotrophobiosis ” i chosen. Apart from its length, it implies, as Roubaud states, lationship between adult and larval members of the same co family, comparable with that existing between ants on the one and Aphids, Coccids, Membracids and Lycenid larve on the This relationship, however, is, so far as nutrition is concerned, sided since the ants exploit the aphids, etc., and may def even transport them, but do not feed them. Moreover, 7 Belonogaster the feeding of adults and larve is reciprocal, and latter could not be reared if they were actually exploited — an extent as to interfere with their growth. As the relatior clearly codperative or mutualistic, I suggest the term trof (from rpody, nourishment and é\Adrrev, to exchange) as less ward and more appropriate than “ cecotrophobiosis.” That the feeding of the young by the mother wasp without | pensation is more primitive than the condition in Belonogas shown by Roubaud’s beautiful observations (1908, 1 1916) on three species of Synagris in the Belgian Congo sicheliana and cornuta). These wasps represent important in the transition from the solitary to the social forms, since make earthen cells like other Eumenids, lay eggs in them a vide the young with paralyzed caterpillars of Hesperid butte favorable seasons, when sentaaeety are bundant, the behavior the latter sealed up (“approvisionnement massif accéléré’ when the season is less unfavorable and food scarcer, the was WHEELER—ANT LARVZ. 323 ing but paralyzed caterpillars (‘‘éducation surveillée indirecte”’) d eventually to a stage essentially like that of the social wasps in reached this last stage. The mother insect, nk malaxating the caterpillar, herself imbibes its juices. : The internal liquids having partly disappeared during this process of malaxation, the prey is no longer, as it was in the beginning, soft and juicy and full of nutriment for the larva. It is possible, in fact, to observe that the caterpillar patée provided by the Synagris cornuta is a coarse paste which has partly lost its liquid constituents. There is no exaggeration in stating that such food would induce in larve thus nourished an increase of the ‘salivary secretion in order to compensate for the absence of the liquid in the prey and facilitate its digestion. t is here that the further development to the condition seen in lonogaster and other social wasps sets in. The mother wasp finds the saliva of the larva agreeable and a trophallactic relation- pe i is established. As Roubaud says, ‘the nursing instinct having evolved in the manner here described in the Eumenids, the wasps acquire contact with the buccal secretion of the larva, become acquainted with it and seek to provoke it. Thence naturally follows tendency to increase the number of larve to be reared simultaneously in order at the same time to satisfy the urgency of oviposition and to profit by _ the greater abundance of the secretion of the larve. Although considerable evidence thus points to trophallaxis as the _ Source of the social habit in wasps, ants and termites, it must be ad- ; mitted that the phenomenon has not been observed in the social bees. That the latter may have passed through a phylogenetic stage like that of Synagris seems to be indicated by the solitary bees of _ the genus Allodape to which I have already referred (p. 318). _ Brauns’ observations, though meager, show nevertheless that Allo- dape has reached Roubaud’s fourth stage, that of direct feeding of the larve from day to day, and if I am right in supposing that the peculiar appendages of the larve are exudate organs, there would be grounds for assuming that trophallaxis occurs in this case. On 324 WHEELER—ANT LARVZ. the other hand, it has often been suggested (e. g., by von B Reepen) that the three social subfamilies, the stingless bees ponine), bumble-bees (Bombine) and honey bees (Apine developed from the solitary bees by another and more direct for the Meliponinz, though living in populous societies, still b up their brood in essentially the same way as the solitary bees, : by sealing up the eggs in cells provisioned with honey-soaked 4 ) The Bombinz, however, keep opening the cells from time to and giving the larve a little food at a time, and in the hone cells are left open till pupation and the larve fed more contin Numerous facts indicate that the Bombinz are the most p the Apinz the most specialized of existing social bees, and Meliponine, though closely resembling the solitary bees in the of the young, are nevertheless in other respects very hi cialized (vestigial sting, elaborate nest architecture, ete.) therefore not improbable that these bees, after passing throt stage more like that of the Bombine, have reverted secondar a more ancient method of caring for their brood. At any ri Meliponine have been so little studied, as compared with the binze and Apinz, that they can be left out of the present discu Sladen (1912) has given us a good account of the queen bee feeding the larve, but he says nothing about the salivary of the latter. These are very large, as we know from the Bordas (1894), but their development is perhaps fully acco for by the complete cocoon spun by the mature larva. E honey bee, which has been so thoroughly studied, I find no e * that the adult workers feed on larval secretions. In bott however, it is impossible under natural conditions ac observe the behavior of the larva while it is being fed. might, perhaps, be done if the bees could be induced to r young in glass tubes made to resemble the cells.° But even should be found, on further investigation, that there is no tion of reciprocal feeding between the larval and adult and Apinz, we might still contend that these very highly sp ®Dr. E. F. Phillips informs me that it would be possible to o behavior of honey-bee larve and their nurses in cells built against and formed by the glass wall of an observation hive. WHEELER—ANT LARV#. 325 | insects had in their evolution passed far beyond the stages repre- sented by the termites, ants and social wasps. There can, indeed, little doubt that the bees are descended from wasp-like ancestors nd that they must therefore have passed from an animal to a vege- ible diet. If the change of diet took place after the social habit bad been established, as is possible and as is so clearly shown to be the case in the harvesting and fungus-growing ants, the loss of a Tes ort to the larval secretions by the adult social bees could be readily explained as due to the abandonment of a scarce animal food, pro- ‘cured with considerable difficulty, for nectar and pollen, which are undant and easily obtained. _ Another objection that may be urged against the view that ophallaxis is so fundamental as I contend, is the behavior of the ants towards their inert pupe, which though transported and de- ded as assiduously as the larve, yield neither liquid exudates nor stions. This does not seem to me to be a serious objection, be- use the pupz evidently have an attractive odor and may therefore => said to produce volatile exudates like certain myrmecophiles. oth the larve and pupe, moreover, evidently represent so much tential or stored nutriment available for the adult ants when the od-supply in the environment of the colony runs very low or entirely. Infanticide and cannibalism then set in with the sult that the devouring of the young of all stages may keep the It personnel of the colony alive till the trophic conditions of the nvironment improve. Certain predatory tropical species (Dory- “tine, Cerapachyini) regularly raid the colonies of other ants and carry home and devour their brood. In northern Eurasia and North merica Formica sanguinea makes similar raids on colonies of ‘ormica fusca but permits a certain number of the pupz to hatch 9 nd become “slaves.” The latter, however, represent only a small ortion of the pupz secured during the course of the summer. Was- nn believes that the fusca pupz are plundered for the sake of g reared. This I doubt, but if true, we should havé to account it by supposing that to the sanguinea workers the odor of the ca pupe is, if anything, even more attractive than that of their If we confine our attention largely to the ants, I believe it can PROC. AMER. PHIL. SOC., VOL. LVII, W, AUGUST 9, I918. 326 WHEELER—ANT LARV#. be shown that trophallaxis, originally developed as a casio ; relation between the mother insect and her larval brood, — panded with the growth of the colony like an ever-widening till it involves, first, all the adults as well as the brood and th fore the entire colony; second, a great number of species of insects that have managed to get a foothold in the nest as 8 i. e., other species of ants (social parasitism) ; fourth, ali ‘ gers, predators or parasites Comune) third, alien social ns that live outside the nest and are “milked” by the ants C biosis), and, fifth, certain plants which are visited or som partly inhabited by the ants (phytophily). In other eee hel have drawn their living environment, so far as this was into a trophic relationship, which, though imperfect or one-si the cases of trophobiosis and photophily, has nevertheless ; the peculiarities of trophallaxis. A brief sketch of each five expansions, indicated as annular areas in the accompz diagram (Fig. 12), may not be out of place. eo 1. There isa very close resemblance between the behavior of ants towards one another and their behavior towards their } The adults feed one another with regurgitated food or ev secretions as is the case with Crematogaster (Physocrema) i an Indomalayan species, the workers of which have great glands in the back of the thorax. Many ants transport and the transported ant assumes a quiescent, larval or pupal This is best seen in certain Ponerine, e. g., in the species pelta, which carry their males under the body as if t larve or pupz. On such occasions the males keep their antenne in the pupal position. Moreover, when the fe of the colony is cut off ants often devour other ants of ti as if they were larve or pup. The largest workers (sol eliminated first, either because they represent more stor because their continued life in the colony constitutes a gre: on the food resources, or for both reasons. Some yea corded an instance of this behavior in an Arizona ant, wh Pheidole militicida, because it regularly kills and eats al headed soldiers in the colony during the winter when the food § is very limited. In artificial nests of Camponotus, which WHEELER—ANT LARV#. 327 _ Fic. 12 Diagram to illustrate the expansions of the trophallactic and trophic relationships within and outside the ant colony. The confines of ie nest are indicated by the double line. lations with their hosts. Among ants especially these relations are _ so intimate that the symphiles may be regarded as integral members of the colony. The adult Lomeschusine beetles, e. g., are not only fed and licked, but their young are treated as if they were ant larve, ‘owing to the presence of trichome glands (“external exudate or- -gans” of Wasmann) in the former and fatty, or internal exudatoria in the latter. 328 WHEELER—ANT LARVZE. 3. The various parasitic ants, of which a number of opecion come to light within recent years and have been described by of its subspecies in' three of my former papers (1901, 1903, 4. The relations of ants to plant-lice and other Homopte: they may be defended by the ants. The Homoptera are not | probably for the simple reason that their mouthparts are so liarly specialized for piercing plant-tissues and sucking their jt and the Lepidopteron larve have, as a rule, no occasion to ab their leaf diet. There are, however, several cases in which caterpillars and Homoptera have entered into more intimate ciation with the ants. Many of the root aphids and coccids their eggs are collected and kept by the ants in their nests, at lez during certain seasons of the year. Two of the caterpillars have acquired closer relations with the ants are so instructiv: illustrating one of the ways in which the myrmecophilous habit been developed, that they merit more detailed description. F, P. Dodd (1912) found that the first-stage larva of a sn gray Queensland moth, Cyclotorna monocentra, is ectoparasitic a Jassid Homopteron which feeds on certain trees and is att and “ milked” by an ant of the genus Jridomyrmex. The ant ries the parasite but not the Jassid into its nest. There the forme spins a temporary cocoon and later emerges from it as a peculie flat, bright red (symphilic color), second stage larva, with long tails. In this stage it subsists “solely on the ant grut sucking out their juices,” but as in the case of Lomechusa in. nests of the European Formica sanguinea, the ant is partiall compensed for the loss of its brood. Dodd says: | Reference has been made to the caterpillars raising their terminal s ments, even the small ones from the cocoons doing so, This was quite cient to warrant investigation. Consequently at various times I have p them with ants and grubs under glass, in order that they could be se advantage and without risk of disturbance. When the anal parts are WHEELER—ANT LARV. 329 truded, an ant generally soon becomes aware of the fact and will be seen to - these great attention. I soon noticed that a liquid, often perfectly trans- parent (it looks so on the blue-green ground, probably was pale bluish), is emitted, and that it is greedily drunk up by the ants. Over and over again, with and without a lens, I have seen this issue, and the ants speedily absorb it. Some ants, perhaps hungry or more enterprising than others, would take a supply from a second caterpillar. If an ant is not satisfied with the quantity given out, she deliberately seizes the protruding parts and gives them a gentle nip, the mandibles can plainly be seen to press upon the juicy flesh; if the hint is not immediately acted upon a more vigorous squeeze is given, and the tails may be gripped and pressed. This is very comical, the t’s meaning is unmistakable and the caterpillar so thoroughly understands too, for a second hint never fails. This liquid, though frequently quite _ Clear, is often mixed with yellowish matter, and at times some jelly-like sub- stance is extruded; the latter the ants do not care about, for after the mois- ture is licked up this is in their way, and if they have not been imprisoned o long, will seize and tug at it until it comes off, and carry it to a spot set apart for waste matter, such as their own pellets and pupal skins, etc., are the nest, travels to the nearest tree in company with the foraging $, spins its cocoon in a crevice of the bark and pupates. In about enty days the moth emerges.*® _ The second case is the caterpillar of Lycena orion, which has ‘been recently studied by Chapman (1916, 1916a) and Frohawk (1916) in England. The butterfly lays its eggs on thyme and other plants. On these the larva feeds, and is often attended by ants as it possesses a honey-gland like many other larval Lyceenids. When t has reached the third, or last moult it crawls down to the ground d on encountering a foraging worker of Myrmica levinodis or cabrinodis hunches up the anterior segments of its body in a singu- manner. Frohawk interprets this behavior as a “ signal” which duces the ant to seize the caterpillar and carry it into the nest. _10When I was in Queensland Dodd generously gave me a fine series of the stages of this extraordinary insect, together with specimens of its st ant. The latter, which I had previously found regularly nesting in the superficial portions of large, flat termitaria at Koah and Townsville, is not, as Dodd states in- his paper, Iridomyrmex purpureus Smith (= detectus Smith), but J. sanguineus Forel. It is smaller and paler than detectus, but every bit as fierce and aggressive. 330 WHEELER—ANT LARVZE. The individual ant which first finds the larva is always the one to it off. Although during its attendance several other ants may find the and stay by it a short time, and even milk it, they soon leave it to its o attendant, who apparently informs them that their services are not {!] Whether the ant signals to the larva for it to prepare itself for t or the larva gives the signal that it is ready to be taken, seems doubtful from what we have seen both Capt. Purefoy and I are inclined to think the larva gives the signal. No. 3 larva alluded to hunched itself both second and third time while the ant was about an inch away pes fe an opposite direction, and at the fourth hunching up the ant was sti over the larva ready for the signal, and when this was given it was seized and carried. Chapman observed that after the caterpillar was taken into” nest it fed on the Myrmica larve. During this period of its life. was not seen to yield the secretion of its honey-gland but treated by the ants as what Wasmann would call an | indiffe tolerated guest, or synoekete. 5. The fifth expansion of trophallaxis, namely the acquisition trophic relations with the myrmecophytes, or plants pos extra-floral nectaries or food-bodies, is also imperfect like ordir trophobiosis, since the ants merely obtain nutriment from the and possibly afford them some protection. The nectar and | plant-foods are for the purposes of the ants merely so many sweet secretions of the Lycenid caterpillars which feed « on foliage. ‘. As the foregoing study of trophallaxis has an ianpOriaale De: on Wasmann’s and Holmgren’s interpretation of symphily it w advisable to consider their views in greater detail. Wasmann elaborated his ideas in regard to the origin and meaning of s in several papers, but as an article published in 1910 embod mature and apparently final contentions, his earlier publi need not be drawn into the discussion. Having found that lar symphiles live only with particular host ants and termites, | cludes, first, that the latter have during their phylogeny 4 particular symphilic instincts as differentiations or modifice their original nursing and adoptive instincts, and second, true ant and termite guests have been developed by these syn instincts through a process called “amical selection,” w WHEELER—ANT LARVZ. 331 likens to the conscious artificial selection employed by man in perfecting the numerous, often bizarre varieties among his domes- ticated animals and plants. Escherich (1898, 1902, 1911), Schim- ‘mer (1909, 1910) and I (1910) have never accepted this view, and I am still unable to see that Wasmann has successfully disposed of ‘our arguments. The whole matter comes down to the answers to two questions: Do ants and termites possess special symphilic in- ‘stincts? and: Is the assumption of amical selection necessary to ac- count for the facts? In my opinion both questions are to be an- swered in the negative. It is unnecessary to consider all the various symphiles which Wasmann has so long and so carefully studied. A brief account ot _ Lomechusa strumosa, his chief battle-horse and according to his own statement one of the most typical of symphiles, will suffice. This is admittedly a predatory parasite in the colonies of Formica san- guinea. Its larve devour the ant larve and the adult beetles are fed and licked by the ants. The fat tissue of the larva probably supplies the ants with an agreeable exudate and the adults certainly furnish an agreeable secretion from their abdominal trichome glands. When the larve, which are evidently treated as if they were ant larve, mature, they are buried in the soil, just as the ant-larve are buried, in order that they may pupate. The pupe are also un- earthed like the ant pupe, after they have spun their cocoons, but this treatment is fatal to the parasites and only those that have been forgotten and left in the soil are able to develop into beetles. Often the greater part of the ant brood is destroyed by the Lome- chusa larve, but in some colonies, by a process which Wasmann _ has never adequately explained, many of the larve develop into _ pseudogynes, or forms intermediate between workers and females. These pathological individuals are unable to perform the functions of either of the castes which they imperfectly represent. This is in _ its essential outlines the history of Lomechusa. Now Wasmann be- lieves that Formica sanguinea has acquired during its phylogeny a special symphilic instinct which impels it to foster Lomechusa to the detriment of the colonies and therefore to the detriment of the species, and regards the case as furnishing a splendid argument against natural selection and an incontestible proof of the existence PARES Ss Wid Re oe ra hee 832 WHEELER—ANT LARVZ. of amical selection. The same reasoning is, of course, Pheidole, the physogastric Staphylinids which live with termites, etc. The bizarre structures of these symphiles, s deformities of some breeds of domestic animals and are sv to have arisen and to have been perfected in an analogous m The analogy,.as conceived by Wasmann, is indeed so close a is hard to see why the term amical selection should have been int duced for what would seem to be after all only another case Darwin’s artificial selection though performed by ants instead nien. ; ae The argument looks plausible till we examine it more critically) When we ask how'the particular symphilic instinct to foster | chusa became established i. e., hereditary, in sanguinea, we see t Wasmann has taken a great deal for granted. Of course, we real know nothing about the phylogeny of sanguinea in its relation Lomechusa. The sanguinea queen and her fertile female offspri in colonies that are old enough to be infested by the beetle, pa particular attention to the parasite and could therefore acquire an instinct as Wasmann postulates only by inspiration. — workers, which do look after the beetles, rarely reproduce and. ably never reproduce in infested colonies and would therefor be in a position to transmit even if they acquired such an ins And as the sanguinea brood is either largely devoured .or conv: into infertile pseudogynes, so that the whole colony tends to die o we have anything but a favorable environment for engendering transmitting an instinct so specialized as to be concerned with a ticular symphile. Furthermore, Lomechusa is a very sporadic site. It may be abundant in certain regions, as in certain parts. Holland, where Wasmann has worked and at St. Moritz, in» Upper Engadin of Switzerland where I once found it and its in considerable numbers, but there are many regions in which ° sanguinea colonies are entirely free from the pest and hence flourishing condition and one most favorable to the survival o species. Wasmann has not shown that Lomechusa introduced WHEELER—ANT LARV. 333 the colonies of such regions is treated with any less consideration than in young, previously uninfested colonies in regions where the parasite is common. As Lomechusa is very rare in England, the ex- periment could be readily performed by shipping a lot of the beetles from the continent to my friend Donisthorpe, with the request that he introduce them to the British sanguinea. I am willing to wager that even if they came from Germany they would be hospitably licked and fed by the ants of Albion. Wasmann might, however, contend that Lomechusa was once a universal sanguinea parasite or, at any rate, much more abundant and more uniformly distrib- uted than at present, but if this had been the case how could san- guinea have survived, if the ravages of the parasite are as great as he asserts, especially when we consider that sanguinea is itself a parasite on another ant, Formica fusca, and is therefore dependent on a host? The perusal of Wasmann’s papers leaves me with the impres- sion that he is bent on showing that symphily is something biolog- ically unique and that for every peculiarity in ant behavior we are bound to postulate a specific instinct. If three of my maiden aunts are fond of pets and prefer cats, parrots and monkeys, respectively, am not greatly enlightened when the family physician takes me aside and informs me sententiously that my aunt Eliza undoubtedly has an zlurophilous, my aunt Mary a psittacophilous and my aunt Jane a pithecophilous instinct, and that the possession of these in- stincts satisfactorily explains their behavior. It is only too appar- t that the physician has merely called the stimuli that severally affect my aunts by Greek names plus a suffix indicating “ fondness,” sumed their existence as entities in my aunts’ minds and naively __ drawn them forth as “explanations.” It is high time that such ‘seem to me to constitute the most formidable argument against the existence of special symphilic instincts, for in the first place, if in the social insects the relations between parent and offspring or be- 334 WHEELER—ANT LARV2ZE, is trophallactic, it is clearly essentially the same as the re between host and symphile. It becomes unnecessary, therefore, assume that in the ants and termites the primitive nursing insti which is a mutual feeding, has been specialized or modified ¢ the phylogeny in adaptation to particular symphiles. nee genetic modifications, well within the limits of the plastic, or “ ir ligent”” behavior of the ants, as responses to the specific organ’ tion of the symphiles, seem amply sufficient to account -~ phenomena. nf In the second place, trophallaxis is, of course, traceable: mutualistic hunger, or “exudate hunger” as Holmgren calls it, therefore to an appetite, in the sense in which this term is em by English psychologists. In view of the fact that psycholog have universally regarded the appetites as very primitive and f damental it is rather strange that they have received so little « tion from the animal behaviorist. Very recently, however, D (1917) and Craig (1918) have emphasized their importance in nection with instinct in two valuable contributions. Drever the appetites as very simple or primitive instincts or “as 1 senting an earlier stage of conscious life, which in the human and the higher animals, is overlaid by the stage to which the dev ment of the specific ‘instinct’ tendencies belong.” He enumer: the hunger, thirst and sex cites the yelhe for sleep or for exercise or activity, “nausea,” “primitive disgust” James’ “ instinct of personal isolation.” Cea contribution is ticularly interesting because he reaches his conclusions from a st of birds (doves) and deals with the matter more thoroughly. cording to him the appetites and aversions are constituents instincts. “Each instinct involves an element of appetite, or sion,.or both.” Perhaps his view is not essentially different Drever’s, since the most typical appetites, those of hunger and represent the basic reactions of organisms, and what are or called “ instincts,” 7. e., the chain-reflexes, or more elaborate r ized behavior of animals, are evidently later and superposed ties that, so to speak, adopt the general movement or pattern pression characteristic of the appetites. Craig, in fact, resol behavior of animals into cycles which run their course according WHEELER—ANT LARV#. 335 _ the appetite or aversion schema. He evidently regards sexual be- havior as the most typical expression of appetite. I should regard hunger as being certainly from a biological point of view the more _ primitive.** : __If we regard symphily and trophallaxis as expressions of essen- tially the same instinct with pronounced appetitive constituent or _ pattern, we can readily understand how Wasmann was led astray by the behavior of sanguinea towards Lomechusa, for the appetites are notoriously prone to perversion. In fact, Escherich’s compari- son of the appetite of sanguinea for the secretions of the beetle with alcoholism is not altogether inept. I should prefer to compare the ant’s behavior with that of.a cat presented with a sprig of catnip or of a leopard presented with a ball of paper sprinkled with oil of bergamot. If the secretions of the larval and adult Lomechusa _have an analogous influence on their hosts, as is very probable, the apparently anomalous behavior of the latter would be readily un- derstood. It would certainly be no more surprising than that my hypothetical maiden aunts prefer to have their bed-linen scented _ with lavender or that some of my bachelor friends prefer Havana cigars and cannot be persuaded to smoke the “domestic” variety. If the objection be raised that I overlook the fact that the rela- _ tion of sanguinea to Lomechusa is one of host to parasite, whereas that between the queen ant and her brood is one of parent to off- ; 11] find myself therefore in closer agreement with Jung than with Freud. The former’s term “libido” seems to be practically synonymous with “ appe- tite” in its general sense, as e. g., in the following very suggestive passage (1916, p. 149): “ We see the libido in the stage of childhood almost wholly occupied in the instinct of nutrition, which takes care of the upbuilding of the body. With the development of the body there are successively opened new spheres of application for the libido. The last sphere of application, and surpassing all the others in its functional significance, is sexuality, which seems at first almost bound up with the function of nutrition. (Com- pare with this the influence of procreation on the conditions of nutrition in lower animals and plants.) In the territory of sexuality, the libido wins that formation, the enormous importance of which has justified us in the use of the term libido in general. Here the libido appears very properly as an impulse to procreation and almost in the form of an undifferentiated sexual primal libido, as an energy of growth, which clearly forces the individual towards division, budding, etc. (The clearest distinction between the two. forms of libido is to be found among those animals in whom the stage of nutrition is separated from the sexual stage by a chrysalis stage.) ” 336 WHEELER—ANT LARV2ZE. spring, I would reply that from a general biological point of the resemblances between the two cases are still fundamental ; suggestive. This has been shown by Giard in one of his inter papers (1905, 1911). He says: Bae Comparative ethology permits us to go further and shows us in clearest manner that the relations between the parent organism and its eny are in principle absolutely the same as those which exist betw parasitized animal and its parasite and that after a period of unstable librium, in which one or the other of the two organisms in contact fin itself injured to the profit of its associate, there is a tendency to establish definitive status of mutual equilibrium in which the two partners find in the association an advantage in the struggle against the ensemble of coi causes of destruction, both cosmic and bionomic. Es, A partial attainment of the equilibrium mentioned by Giard, in the nursing relation of ants and that of sangwinea to Lome is brought about by mutual feeding. In neither case is the exc of food between the two parties quantitatively equal, but the dates as stimuli, in all probability make up in quality or intensity what they lack in quantity. | This brings us back to Wasmann’s amical selection whan sti remains to be considered. It has often been remarked that symphiles are strangely like our domestic animals in that they in a socia! environment where they are protected from enemies abundantly fed. In the case of the domestic animals Darwin I ago showed that such an environment favors the production te) traordinary variations, and Pearson (1897) and Trotter (19: agree that when organisms unite to form larger biological units s as the Metazoan body, herds, colonies and societies, the indivi though necessarily limited in their evolution along particular nevertheless in other respects escape from the stabilizing influence natural selection and exhibit unusual freedom of development ar specialization. Both the domestic animals and the symphiles really become integral members of the insect societies in which are permitted to live, show this freedom in the development usual structural and color characters, as we see in albinos, p breeds of fowl, pigeons and dogs, Paussids and Clavigerids monstrous antennz, ant-chalcids like Kapala and Isomeralia, huge thoracic spines, etc. Similar phenomena are common in m: WHEELER—ANT LARV. © 337 3 ecto- and entoparasites which are intimately associated with their sts (e. g., Sacculina, many Copepods, Isopods, tapeworms, etc.). le origin of these strange characters is evidently spontaneous, or mutational and dependent on the favorable conditions under which they arise. In the case of the domestic animals we know that the unusual characters are being continuously and rapidly perfected and established by man’s selective activity. It does not follow, however, that the analogous developments of symphiles are the outcome of a similar activity on the part of the ants and termites. The resem- blance of the aberrant characters of symphiles to “hypertelic” structures in many other insects’ has been noticed by Dahl. That _ the phenomena in both cases are due to the same cause, 1. e., the re- _ laxation or suppression of natural selection, is much more probable ' than Wasmann’s contention that the ants take the same interest in _~ breeding Paussids and Clavigerids with extraordinary antenne that e do in breeding lop-eared rabbits and fan-tail pigeons. Nor is there any evidence that even the biologically useful characters of the symphiles, namely their trichome glands and exudate tissues, are engendered or perfected by amical selection. The truly amazing cases of convergent or parallel development of these structures in symphiles belonging to the most diverse genera is, in all probability, attributable to the adaptive activities of the symphiles themselves, just as we attribute the convergent development of hooks, suckers, hermaphroditism, blindness, etc., in entoparasitic worms or aptery in ectoparasitic insects, such as lice, fleas, Polyctenids, Nycteribids, €tc., to the parasites themselves and not to specifically selective efforts on the part of the host organisms. Holmgren accepts Wasmann’s amical selection and carries it a step further in his contention that it accounts for the development f the various castes in the termite colony. He says (1909, p. 200) : _ If now the above described connection between feeding and exudate secretion holds good, so that the quantity of exudate secretion determines the kind of feeding, it would seem to be self-evident that the exudate secre- tion is intimately connected with the development of castes, for Grassi and Sandias have shown that feeding is probably to be regarded as a factor in caste development. And if, therefore, the exudate secretion is the cause of feeding we must regard it as the cause of the differentiation of the various 338 WHEELER—ANT LARVZE. That this opinion is no longer tenable, at least in the form in it is stated, is shown by the observations of Bugnion (1912), ¥ has proved that the soldier and worker castes of Eutermes are de mined in the egg, and the observations of Miss Thompson (1 who has been able to distinguish the sexual from the sterile ¢ of Leucotermes flavipes at the time of hatching. Holmgren’s could be accepted only on the assumption that the effects of fe had been carried back during the long phylogeny of the termites the embryonic stages. Incidentally it may be said that his 01 statement in regard to the development of the complemental neotenic males and females in the termite colony refer, sate | development of castes, but to the ontogenetic growth of the ¢ tissues, a process which is exhibited in the most extraord manner during the imaginal life of the true queens of many sp In conclusion it may be interesting to note in connection wit development of the social habit of insects from a trophallactic tion between parent and progeny, that the social or gregarious stinct in man has also been regarded by some authors as an tite. Drever (1917) cites the early British philosopher Hute! (“ Nature and Conduct of the Passions,” Sect. 4, 1728) as fying the gregarious instinct among the appetites, and refe: McDougal’s interesting comments on gregariousness (1910), (p. 184) : ; a There is in the instinct itself something which suggests such [as McDougal’s], something which might even lead the psychologist tain that it belongs to the “ Appetite” group in our system of cla an opinion to which Galton’s description [of the wild ox of D which cannot endure even a momentary severance from the herd] » some support. There is indeed something primordial about the wh ence involved in the operation of the gregarious instinct. _ The fact that higher gregarious and social animals are sz long as they are with their fellows but become uneasy when is certainly very suggestive of the appetitive type of behavio 12 As Trotter says (1916). “In interpreting into mental terms quences of gregariousness, we may conveniently begin with the sim conscious individual will feel an unanalyzable primary sense of the actual presence of his fellows, and a similar sense of discomfo absence. It will be obvious truth to him that it is not good for the be alone. Loneliness will be a real terror, insurmountable by reason, ve & thie Saw WHEELER—ANT LARV#. 339 this connection some of Le Dantec’s recent writings are of consid- erable interest. In a footnote (p. 288) at the end of “ Les Influences _Ancestrales” (1917) he asks: Does maternal love, which has assumed such great moral significance in the human species, originate among the females of the Mammalia as the desire (souci) to rid themselves of their milk? A similar tendency to show that the social relation in man has an egoistic instead of an altruistic foundation is even more forcibly dis- played in his startling not to say shocking, volume entitled “L’Egoisme Seule Base de Toute Société” (1916) .7% BIBLIOGRAPHY. : 1914. Arnold, G. Nest-Changing Migrations in Two Species of Ants. Proc. Rhodes. Sc. Assoc., 13, 1914, pp. 25-32, I pl. tgo1. Berlese, A. Osservazioni su Fenomeni che avvengono durante la Ninfosi degli Insetti Metabolici. Rivist. de Patol. Veget., 8, 1901, pp. 1-155, 6 pls., 42 text-figs.; 10, 1901, pp. 1-120, 8 pls., 5 text-figs. 13 Since the foregoing paragraphs were written I have found two quota- tions from Charles Bonnet’s “Contemplation de la Nature,” which are the more remarkable because they were published in 1764. On p. 213 he says: “In order the better to insure the well-being of their progeny, would not Nature have engaged the affection of the mothers in such a manner that the young would become for them a source of agreeable sensations and material usefulness? Certain facts seem to confirm this conjecture. ... The mammz have been constructed with such art that the sucking and pressure exerted by the young excite the nerves which impart to these organs a delicate disturb- ance or soft commotion accompanied by a feeling of pleasure. This pleasure sustains the natural affection of the mother, if indeed it be not one of its principal causes. The same may be said of the action of licking, which is reciprocal. Finally, mothers are sometimes incommoded by the abundance of their milk; the young relieve them by sucking.” The second quotation (p. 272) is even more astonishing in its bearing on the conditions in the social insects: “ The neuters [of bees] have no sex and do not reproduce. How can we suppose that they have the same affection for the offspring of their queen as the mothers of other animals? They behave nevertheless in the same manner under the same circumstances. If, therefore, Nature has known how to insure the attachment of mothers by the agreeable sensations derived from their offspring or by the services they render, it would certainly seem that she must employ much the same means in the case of the worker bees and that she has placed in the young a secret source of delectable sensations which attaches them to the workers and induces them to disgorge into the cells the kind of porridge with which the young are nourished.” These quotations are from a work entitled “La Psychologie Animale de Charles Bonnet” published in 1909 in Geneva by Ed. Claparéde who cites them and the work of Giard in support of his views on the reciprocal relations of the mother to her offspring. is oe WHEELER—ANT LARV. 1909. Berlese, A. Gli Insetti, 1, 1909, p. 535, figs. 618, 619. 1902, Biedermann, W. Ueber die Struktur des Chitins bei Ir Crustaceen. Anat, Anzeig., 21, 1902, pp. 485-490. 1903. Biedermann; W. Geformte Secrete. Zeitschr. f. allgem. 1903, Pp. 395-481, 4 pls. 1894. Bordas, L. Appareil Glandulaire des Hyménoptéres. Ann, Zodl. (7), 19, 1804, pp. 1-362, 11 pls. : 1902, Brauns, H. Eucondylops n. g. Apidarum. Zeitschr. rai A 2, 1902, pp. 379-380. 1912, Bugnion, E. Observation sur les Termites. Diffé Castes. Compt. Rend. Soc. Biol., Paris 1, 1912, pp. 1903, 1905. du Buysson, R. Monographie des Guépes ou Vespa. A? Ent. France, 72, 1903, p. 260; ibid., 74, 1905, p. 485. ae 1913. Casper, A. Die Kérperdecke und die Driisen von Dytiscus 1 L., ein Beitrag zum feineren Bau des Insektenkérpers. Zei wiss. Zoél., 107, 1913, pp. 387-508, 44 text-figs. bes 1916. Chapman, T. A. What the Larva of Lyczna orion does Last Instar. Trans. Ent. Soc. Lond., 1916, pp. 291-297, 9 PD 1916a. Chapman, T. A. Cbservations Completing an Outline of tl History of Lycena orion L. Trans. Ent. Soc. Lond., ais. pp. 312, 3 pls. 1918. Craig, W. Appetites and Aversions as - Constituents of Biol. Bull., 34, 1918, pp. 91-107. 188. De Bono, F. P. Sull’ umore segregato della Timarcha | pith Scheffer. Natural. Sicil., 8, 1880. 1912. Dodd, F. P. Some Remarkable Ant-friend Lepidoptera of land. Trans. Ent. Soc. Lond., 10, 1912, pp. 577-590, 1 pl. — : 1917. Drever, J. Instinct in Man, a Contribution to the Psy Education. Cambridge University Press, 1917. 1898. Escherich, K. Zur Anatomie und Biologie von Paussus turciet - Zoél. Jahrb. Abt. Syst., 12, 1898, pp. 27-70, 1 pl., 11 text-figs. 1902. Escherich, K. Biologische Studien iiber algerische Myrm zugleich mit allgemeinen Bemerkungen tiber die Entwi Bedeutung der Symphilie. Biol. Centralbl., 22, 1902, pp. 4 figs. : : 4 1911. Escherich, K. Termitenleben auf Ceylon. Jena, Gustay Fi 1899. Emery, C. Intorno alle Larve di Alcune Formiche. Mem. Sc. Ist. Bologna (5), 8, 1809, pp. 3-10, 2 pls. 1912, Emery, C. Etudes sur les Myrmicine. Ann. Soc. Ent. Belg., 56, 1 PP. 94-105, 5 figs. 1901. Fielde, A. M. A Study of an Ant. Proc. Acad. Nat. Sc. P. IQOI, pp. 425-449. eee 1890. Forel, A. Un Parasite de la Myrmecia forficata Fabr. Compt. Re Soc. Ent. Belg., 1890, p WHEELER—ANT LARVZ. as Friese, H. Bienenlarven mit Pseudopodien und neue Arten der Gat- tung Allodape. Deutsch. Ent. Zeitschr., 1914, pp. 144-146, 1 fig. Frohawk, F. W. Further Observations on the Last Stage of the Larva of Lycena orion. Trans. Ent. Soc. Lond., 1916, pp. 313-316, 2 pls. tg11. Giard, A. Les Origines de !Amour Material. Bull. Inst. Psych. Internat., 5, 1905, p. 1; Revue des Idées, 1905, p. 249; Republished in Giard’s “ CEvres Diverses,” I, I9II, pp. 207-235. Holmgren, N. Ueber das Verhalten des Chitins und Epithels zu den unterliegenden Gewebearten bei Insekten. Anat. Anzeig., 20, 1901, pp. 480-488. Holmgren, N. Ueber die morphologische Bedeutung des Chitins bei den Insekten. Anat. Anzeig., 21, 1902, pp. 373-378, 5 figs. I a 1909. Holmgren, N. Termitenstudig¢n 1. Anatomische Untersuchungen. K. Svensk. Vetensk. Handl., 44, 1900, pp. 1-215, 3 pls., 75 text-figs. Janet, C. Observations sur les Guépes. Paris, Carré et Naud, 1903. Jordan, H. C. Zur Morphologie und Biologie der myrmecophilen Gattungen Lomechusa und Atemeles und einiger verwandter Formen. ea Zeitschr. f. wiss. Zodl., 107, 1913, pp. 346-386, 20 text-figs. ; a 1916. Jung, C. G. Psychology of the Unconscious. Transl. by Beatrice M. a Hinkle. New York, Moffat, Yard & Co., 1916. = 1911. Kapzov, S.. Untersuchungen iiber den feineren Bau der Cuticula bei Insecten. Zettschr. f. wiss. Zoél., 98, 1911, pp. 297-337; 3 pls., 3 text- figs. : Karawaiew, W. Die Nachembryonale Entwicklung von Lasius flavus. a Zeitschr. f. wiss. Zodl., 64, 1898, pp. 385-478, 4 pls., 15 text-figs. + 1gto. Kriiger, E. Beitrage zur Anatomie und Biologie des Claviger testa- r ceus Preyssl. Zeitschr. f. wiss. Zoél., 95, 1910, pp. 327-381, 2 pls., .: 33 text-figs. ; e 1916. Le Dantec, F. L’Egoisme Seule Base de Toute Société. Paris, Ern. Ks Flammarion, 1916. 1917. Le Dantec, F. Les Influences Ancestrales. Paris, Ern. Flammarion, ae 1917. Lutz, K. G. Das Blut der Coccinelliden. Zodél. Anzeig., 18, 1895, pp. 3 244-255. . : a 1910. McDougall, W. An Introduction to Social Psychology. 2d Ed., Bos- ; we ton, J. W. Luce & Co., 1910. z 1881. Magretti, P. Del Prodotto di Secrezione Particolare di Alcuni . 3 Meloidi; Esame Microscopico. Bull. Scient. Pavia, 3, 1881, p. 58. 1882, Magretti, P. Ricerche Microscopiche sopra i Liquidi di Secrezione e di Circolazione nelle Larve di Alcuni Imenotteri Tentredinei. Boll. Scient. Pavia, 4, 1882, p. 58. : Pearson, K. The Chances of Death and Other Studies in Evolution. 2 vols. London, New York, E. Arnold, 1897. PROC, AMER, PHIL. SOC., VOL. LVI, x, AUGUST 17, I918. 342 - WHEELER—ANT LARVA. 1902. Pérez, C. Contribution a I’Etude des Métamorphoses. Ball Scient, 37, 1902, pp. 195-427, 3 pls. 1913. Reichensperger, A. Zur Kenntnis von Myrmecophilen aus Abessinien, Zoél. Jahrb, Abt. f. Syst., 35, 1913, pp. 185-218, 2 pls., 15 text-figs. 1908. Roubaud, E. Gradation et perfectionnement de l’instinct chez les Guépes Solitaires d’Afrique, du genre Synagris. Compt. Rend. aes Sc., 147, 1908, pp. 659-607. 7 1910. Roubaud, E. Recherches sur la Biologie des Synagris. Evolution de l’Instinct chez les Guépes Solitaires. Ann. Soc. Ent. Baoseed 70, IQIO, pp. I-21, 4 pls. : 1910a. Roubaud, E. Apercus biologiques sur les Guépes Sociales d’Afrique des Genres Icaria et Belonogaster. Compt. Rend. Acad. Sc., 1910, . pp. § 1916. Roubaud, E. Recherches Biologiques sur les Guépes Solitaires Sociales d’Afrique. Ann. Sc. Nat. Zoél. (10), 1, 1916, pp. 1-160, 34 figs. 1914. Santschi, F. Formicide in “ Vyas de Ch. Alluaud and R. Jeanna en Afrique Orientale (1911-1912),” Ins. Hymen., 2, 1914, pp. 43-148, 2 pls., 30 text-figs. 1909. Schimmer, F. Beitrag zu einer Monographie der Gryllodeengattung Myrmecophila Latr. Zeitschr. f. wiss. Zodl., 93, 1909, PP. 409-534, 3 pls., 29 text-figs. ‘1910. Schimmer, F. Ueber die Wasmannische Hypothese tee “ Duldungs- instinktes” der Ameisen gegenuber synodker Myrmekophilen. Zo Anzeig., 36, 1910, pp. 81-95. 1912. Sladen, F. W. L. The Humble-Bee, its Life-History and how to Domesticate it. London, Macmillan & Co., 1912. 1737, 1738. Swammerdam, J. Biblia Nature, sive Historia Tae 2 vols. Leydz, Severin, etc. 1737, 1738. 1917. Thompson, Caroline B. Origin of the Castes of the Common Termite, Leucotermes flavipes Kol. Journ. Morph., 30, 1917, pp. 83-136, 8 p 5 text-figs. 1907. Tragardh,I. Description of Tesnithedamed New Genus of Termitop ilous Physogastric Aleocharini, with Notes on its Anatomy. Zoi Studier., Upsala, 1907, pp.172-190, 1 pl., 10 text-figs. 1907a. Tragardh, I. Notes on a Termitophilous Tineid Larva. Ark. 3 Z06 3; 1907, pp. t=7) I pl. 1916. Trotter, W. Instincts of the Herd in Peace and War. New Y. or! Macmillan Company, 1916. 1903. Wasmann, E. Zur naheren Kenntnis des echten Gastverhaltni (Symphilie) bei den Ameisen- und Termitengasten. Biol. Cen 23, 1903, PP. 63-72, 195-207, 232-248, 261-276, 298-310, 25 text-figs. 1910. Wasmann, E. Ueber das Wesen und den Ursprung der Sympk Biol. Centralbl., 30, 1910, pp. 97-102, 129-138, 161-181. WHEELER—ANT LARV2. 343 1915. Wasmann, E. Neue Beitrage zur Biologie von Lomechusa und Atemeles mit kritischen Bemerkungen iiber das echte Gastver- haltnis. Zeitschr. f. wiss. Zodl., 114, 1915, pp. 233-402, 2 pls., 1 chart and 2 text-figs. 1890. Wheeler, W. M. On the Appendages of the First Abdominal Segment of Embryo Insects. Trans. Wis. Acad. Sc. Arts. Lett., 8, 1890, pp. : 87-140, 3 pls. 1900. Wheeler, W. M. A Study of Some Texan Ponerine. Biol. Bull., 2, _- 1900, pp. I-31, Io figs. _ 19002. Wheeler, W. M. The Habits of Ponera and Stigmatomma. Biol. - Bull., 2, 1900, pp. 43-60, 8 figs. ' Igor. Wheeler, W. M. An Extraordinary Ant-Guest. Amer. Natural., 35, IQ0I, pp. 1007-1016, 2 figs. ' Igo1a. Wheeler, W. M. Compound and Mixed Nests of Ants. Amer. : Natural., 35, 1901, pp. 431-448, 513-539, 701-724, 791-818, 20 figs. ' 1903. Wheeler, W. M. Ethological Observations on an American Ant Ss (Leptothorax emersoni Wheeler). Arch. Psych. Neurol., 2, 1903, pp. _ I-31, I fig. ' -1907. Wheeler, W. M. Notes on a New Guest Ant, Leptothorax glacialis, Res and the Varieties of Myrmica brevinodis. Bull. Wis. Nat. Hist. Soc., y 5, 1907, pp. 70-83. _ 1907. Wheeler, W. M. The Polymorphism of Ants, with an Account of a Some Singular Abnormalities Due to Parasitism. Bull. Amer. Mus. a Nat. Hist., 23, 1907, pp. 1-92, 6 pls. _ 1910. Wheeler, W. M. Ants, Their Structure, Development and Behavior. Es New York, Columbia University Press, 1910. 3915. Wheeler, W. M. On the Presence and Absence of Cocoons Among Ants, the Nest-Spinning Habits of the Larve and the Significance of the Black Cocoons Among Certain Australian Species. Ann. Ent, Soc. Amer., 8, 1915, Pp. 323-342, 5 figs. ag DEVELOPMENT OF MAGNETIC SUSCEPTIBILITY I MANGANESE STEEL BY PROLONGED HEAT TREATMENT. nee By CHARLES F. BRUSH. (Read April 19, 1978.) During the past three years I have had the honor of presenting several papers on “ Spontaneous generation of heat in recently har ened steel,”* showing that all the specimens treated spontanec generated heat in easily measurable quantity after hardening quenching in water at various temperatures above the critical perature of decalescence. Carbon tool steel, “ high-speed ” tu : chromium steel, and several specimens of nickel-chromium ste were tested. In all cases the generation of heat was most f nounced immediately after quenching and diminished rapi though it continued observable a week or more. Generation of h was always conditional on true hardening, by quenching at a te perature above the critical point, except in the case of one of nickel-chromium specitnens. Here moderate, but unequivocal, ¢ eration of heat followed quenching at a temperature just below critical point, reached by falling slowly from a higher temper at through full recalescence. While true hardening could not occurred, yet there was a well-marked “ stiffening” or inci hardening of the metal as shown by subsequent hardness When, however, the same steel, after annealing, was slowly to the same temperature (but not above) and quenched, no g tion of heat followed. In this connection another interesting nomenon developed as follows: When the same lot of nickel- mium steel, after annealing, was quenched at several success higher (rising) temperatures, but all below the critical tempera 1 Proc. Am. Phil. Soc., Vol. LIV., No. 217, May-July, 1915. Phys. Re view, N. S., Vol. IX., No. 3, March, 1917. Proc. Royal Soc., Series A 93, No. A, 649, Apl. 2, 1917. Joint paper with Sir Robert Hadfield. Am. Phil. Soc., Vol. LIV., No. 4, 1917. 344 BRUSH—MAGNETIC SUSCEPTIBILITY. 345 each quenching was followed by a small but distinct absorption of heat. __ The phenomena above outlined appear to be new, and have _ aroused the interest of some eminent metallurgists, among others Sir Robert A. Hadfield, who kindly furnished all the specimens of nickel-chromium steel employed, and to whom I am greatly indebted _ for joining in some of the later work, and incidentally confirming the most important of my findings with different apparatus of his own design. q The foregoing outline of former work is introduced here because q it underlies, and is closely associated with, the subject matter of the _ present paper. _ Sir Robert Hadfield long ago suggested that interesting results might follow similar experiments with manganese alloy-steel (with which his name is so intimately connected). As is well known, this remarkable steel is exceedingly tough, and _ difficult to work with machine tools though not hard; its softest and _ toughest condition is brought about by water-quenching at a high temperature, after which it is almost completely non-magnetic; it has no critical temperature, and hence cannot be hardened in the or- ‘dinary sense; when heated a long time at a moderate temperature it becomes magnetic, loses much of its tensile strength, and all its toughness, and becomes brittle and considerably harder. __. For the purposes of the following experiments Sir Robert Had- field sent me 19 numbered bars of his manganese steel, each 6 inches long and ¥% inch in diameter (round), all cut from the same long __ bar and ground to size after treatment. Following is his specification : Analysis —C, 1.18 per cent.; Si, .14 per cent.; Mn, 12.29 per Bars 1 to 6, As forged. Non-magnetic. _ Bars 7 to 12, Toughened by water-quenching at 995° C. pies) 3 netic. P Bars 13 to 18, Toughened as above, then reheated to 500° and kept at that temperature 63 hours. Magnetic. © Bar 19, Treated like 13 to 18, then reheated to 995° and water- quenched. Non-magnetic. 346 BRUSH—DEVELOPMENT OF MAGNETIC SUSCEPTIBILITY I carefully tested one bar of each description for sclerose hardness, with following results: Bar No. 1 (as forged). Hardness 37.3. Bar No. 7 (toughened). Hardness 28.5. Bar No. 15 (toughened, reheated). Hardness 51.6. Bar No. 19 (toughened, reheated, retoughened). Hardness Each of the above hardness numbers (and those to follow) the mean of at least ten consistent measurements made on the cz fully ground, horizontal end surface of the bar, a fresh spot t used for each measurement. I subsequently heated bar No. 13 to 1074° and cooled Coil | in the furnace. Its hardness, which presumably had been about 51. like its companion No. 15, was then 28.8, and it was oe seeming to show that quenching at high temperature, and anne: from a still higher temperature, gives the same hardness and 1 no! magnetic condition whatever previous treatment may have be The hardness of bar No. 19 seems to contradict this conclusion, respect of hardness, but it was quenched at a very considera lower temperature. In the following experiments ten of the 6-inch bars of mangar steel were used, so as to approximately equal in weight the t 5-inch bars of other steels employed in former experiments. First quenching: Bars 1 to 5 and 7 to 11 (10 in all) were heat in an electric muffle furnace to 1013° C. and quenched in wate: This treatment was followed by no appreciable generation or abso: tion of heat when tested in the calorimeter employed in former | periments and described in the earlier papers referred to. : Hardness was now: Bar No. 1, 30; Bar No. 7, 28.3, she that the first lot “as forged” and the second lot “toughened” w brought to substantially the same “ toughened ” condition. 4 Second quenching: The ten bars were again heated to 1 allowed to cool in the furnace to 800° and quenched. Again followed no appreciable generation or absorption of heat. Hardness was now: Bar No. 1, 27.6; Bar No. 7, 26.3. Third quenching: The bars were heated to 818°, allowed cool in the furnace to 607°, and quenched. There was no s1 quent generation or absorption of heat. IN MANGANESE STEEL BY HEAT TREATMENT. 347 Hardness was: Bar No. 1, 26.3; Bar No. 7, 26.6. The ten bars were next heated slowly to 645° and allowed to cool slowly in the furnace to room temperature. Hardness was: Bar No. 1, 26.5; Bar No. 7, 25.9. All the bars were now very moderately magnetic, though in their softest condition. The foregoing quenching temperatures were falling ones. The _ following quenching temperature was a rising one, from the an- ____ nealed condition last described. _ Fourth quenching: The bars were heated slowly to 615° and quenched. i Hardness was now: Bar No. 1, 38; Bar No. 7, 30.3. Notwithstanding this considerable increase of hardness, there followed no appreciable generation or absorption of heat. The bars remained very moderately magnetic. The results of the foregoing experiments make it highly prob- _ able that no spontaneous generation or absorption of heat can be _ brought about by quenching this manganese steel at any tempera- ture, rising or falling, while in its normal, useful non-magnetic con- _ dition. But it was thought worth while to make further experiments with the steel in its magnetic condition and, incidentally, to study the development of this magnetic condition during the prolonged moderate heating necessary to bring it about. The latter study has proved so interesting that I have pursued it to considerable length and made it the titular subject of this paper. The apparatus employed in the following study consists, in part, of a vertical cylindrical electric furnace heated by small spirals of “nichrome” wire carrying alternating current regulated by a theostat. The heating spirals are so disposed as not to produce any magnetic field inside or outside the furnace. Instead of the usual sheet-iron casing, this furnace is cased with sheet brass slotted longi- tudinally to prevent induction currents in it when the external mag- netizing solenoid is excited. The furnace is surrounded by a solenoid 16 inches inside, and 20 inches outside diameter, and 4 inches long (high), consisting of 860 turns of No. 12 insulated copper wire wound in two equal coils adapted to be placed in series or parallel relation by means of a suitable switch. The axes of the 848 BRUSH—DEVELOPMENT OF. MAGNETIC SUSCEPTIBI furnace and solenoid are coincident. The solenoid is excited b: current from a 65-volt storage battery, controlled by a rheostat, and the circuit is closed and opened by a switch which breaks sir taneously at three points in series, so as to avoid the destructive which would occur at a single break. An ammeter and switch are ineluded in the line. A single turn of asbestos-insulated platinum wire is located | the furnace, and the ends of this loop are connected by a twi cable with a ballistic mirror-galvanometer of 600 ohms’ res ceile When the solenoid circuit is closed, a brief electric current induced in the platinum loop in the furnace and causes a minimw swing of the galvanometer scale easily read with considerable : cision. : When a bundle of ordinary steel or iron bars is placed within ¥ platinum loop the galvanometer deflection is, of course, many time greater, and is fairly proportional to their magnetic susceptibilit after deducting the minimum deflection due to the platinum alone, and when the excitation of the solenoid is not too small too great. In the following experiments with the manganese s 9 amperes was found to ba suitable exciting current with the soleno’ coils in series. Small variations of exciting current were reduce to this value in computing results. Residual magnetism was me: ured by the usual method of reversals, and allowed for. The above described apparatus was originally designed constructed for a rough study of the magnetic properties of meta and alloys at temperatures up to and above their melting points, has proved very useful. A high-temperature furnace with sla brass casing is included in the general outfit. Preparatory to the following study of magnetic suscep the manganese steel brought about by prolonged heating, ten ha inch round bars, 6” long, of Swedish charcoal iron were pla within the platinum loop in the furnace, and the galvanometer d flection was repeatedly observed when the solenoid was excited various amounts of current. Nine amperes was found to give cor veniently large deflection, which was closely proportional to. current through a wide range about this value. This condition also found approximately true when the manganese steel bars, mac IN MANGANESE STEEL BY HEAT TREATMENT. 349 “magnetic by heating, were subsequently substituted for the charcoal P. In the following experiments galvanometer deflection, less that _ amount due to the platinum loop alone, is taken as the measure of magnetic susceptibility, and the susceptibility of the Swedish iron is used as a standard and assigned a value of 100. All other values are reduced to and expressed in terms of this standard. As a preparatory measure, the ten bars of manganese steel were brought to their softest and toughest condition by quenching at 1000°. Hardness was: Bar No. 1, 26.7; Bar No. 7, 25.8. _All the bars were quite free from any trace of magnetism. The bars were next placed within the platinum loop in the elec- tric furnace and heated 170 hours to a temperature fluctuating be- - tween 505° and 525°. The growth of magnetic susceptibility is _ plotted in the curve shown in Fig. 1. There is no doubt that the Swedish Chart aS a 100 ae : CA - off 3 iy fos 7 20 40 Bo _ 100 120 140 160 - Hours of Heating 505° 525°C. Fic. I. curve would have been smoother if the temperature had remained constant. It was intended to use about 515° temperature, and it was held near that value by the rheostat during the first few hours _ Subsequent fluctuations were due to variations of voltage in the 850 BRUSH—DEVELOPMENT OF MAGNETIC SUSCEPTIBILI’ | alternating heating current. The higher temperatures usually curred in the latter part of the night, and were always accomp by more than average rise of susceptibility. But the large dep sion in the central part of the curve is thought to be due to some ¢ scure cause, and not to temperature variation. . The entire absence of growth of susceptibility during tie last or more hours prompted the belief that the steel had reached a stable condition at the temperature of treatment, and led to the discontinu- ance of this experiment. Permanent magnetism, which had considerable while susceptibility was rising, fell off very — cn ing the last two or three days. Fifth quenching: At the end of 170 hours the bars we quenched, after which they exhibited moderate, but Pd i and equivocal generation of heat. ; Hardness was: Bar No. 1, 48.1; Bar No. 7, 47.2. This great increase of hardness (from 26.7 and 25.8 in the nealed condition) brought about by the long heating, doubtless counts for the spontaneous generation of heat observed. & During the long heating the bars acquired a rather thick coating of black oxide which peeled off almost completely in quen leaving clean metal surface. The oxide was strongly magnetic ; b its weight was so small, compared with the total weight of the b: that it could not have affected, materially, the foregoing magnet observations. . The bars were again placed in the furnace and heatel to a high temperature than before, fluctuating between 590° and 598°, for t t first 90 hours (from 170 to 260 hours, reckoned from beginning treatment). 3 The results of this procedure are plotted in the first curve. Fig. 2. It is seen that magnetic susceptibility started at a very siderably higher value than it had at the end of the previous tr ment. The reason of this increase during the intervening few d without heating, is not clear. It may have occurred at the mon of quenching; or, more likely, during the period of spontaneous | generation which followed the quenching. S The curve shows a very regular, but steadily diminishing, gro IN MANGANESE STEEL BY HEAT TREATMENT. 351 of susceptibility at this higher temperature, until it reaches nearly double the value it had at the end of the previous treatment. Fig. 2 shows that at the 260-hour point of total treatment the temperature was quickly raised about 25°, 4. ¢., to 619°, and con- tinued at that point about ten hours. This moderate rise in tem- perature brought about a sudden and very considerable fall in sus- ceptibility, approaching stability at the end of the ten hours. | Bee aia ee SO le ee a see et 591 597° |-----4-----7 noose ‘619+ % (6 Quenched 220 280 300 320 240 260 Hours of Heating Fic. 2. When the temperature was next quickly lowered to its former value and then continued to the end of the experiment, 175 hours (340 hours total), there was at first a sudden rise of susceptibility, followed by steady growth as before. All these changes are clearly shown in Fig. 2. Great sensitiveness to temperature change is indicated at about 600°. Sixth quenching: At the end of 344 hours, total, of treatment, if any, spontaneous generation of heat. Hardness was: Bar No. 1, 37.4; Bar No. 7, 36.2. This shows a very considerable softening since the last qu ing, notwithstanding the large increase of magnetic suscef The softening may account for the absence of heat — the quenching. The magnetic susceptibility of the cold quenched bars: was ¢ the same (slightly lower) as before quenching. —T 590; SSS TT 2 “ 5 st IZ aL ‘ we : i ae ' he " 5 ‘ ERIE 2 : : 2) \ 1 5 Lee a. a. S 360 380 400 Hours of Heating Fic. 3. The ten bars were again: heated, slowly this time, to 590° held nearly at that temperature until the 381st hour of total ment, as shown in Fig. 3. Susceptibility rose slightly, reaching highest value, 68.5. As this is comparable with the susceptibility « ordinary steel, the manganese had apparently almost completely its influence. IN MANGANESE STEEL BY HEAT TREATMENT. 353 At this stage it was thought that decalescence might possibly be brought about by cautiously raising the temperature, and the effect of doing so is shown in the great and nearly vertical drop in the sus- ceptibilty curve. The stations in this part of the curve represent observations at half-hour intervals, indicating two hours for the total drop, with the temperature steadily rising to the maximum of 692°. It seemed clear that decalescence was not taking place, because loss of susceptibility was far too slow in time, and the maximum temperature reached was not sufficiently high. Probably the man- ganese was simply resuming its sway. The temperature was next rapidly lowered to 605°—590°, bringing on a rapid recovery of magnetic susceptibility, amounting to 30 points in 21 hours as shown. Again the temperature was raised, but much more rapidly than before, resulting in a much steeper drop in the curve, the observa- tion stations shown representing only five-minute intervals. Seventh quenching: At the end of the curve shown in Fig. 3 the steel bars were quenched at 687°. Subsequently there was no trace of generation or absorption of heat. Hence it is virtually certain there had been no decalescence. Hardness was: Bar No. 1, 42; Bar No. 7, 41.8. Sir Robert Hadfield long ago assured me that the study of manganese steel is full of surprises for the investigator. I have experienced some of them, and hesitate at present to draw definite conclusions from the results of the experiments just described. The manner in which the manganese operates in completely obliterating the magnetic quality of seven times its weight of iron is, so far as I am aware, not yet known; and the instability of the alloy or, prob- ably, mixture of alloys, in which the carbon present may play an important part, at about 600° temperature as herein shown, is most remarkable and promises a fertile field for future investigation. I am contemplating the study of a similar PETLO-RADEAS alloy free from carbon. CLEVELAND, O., April, 1918. PRELIMINARY NOTES ON SOME NEW SPECIES 7 OF AGARICS., Sd By GEO. F. ATKINSON. (Read April 19, 1918.) Amanita brunnescens nov. sp. Gregaria vel dispergens, 8-15 cm. alta: pileo convexo dein expanso, fuligineo-brunneo, 4-9 cm. lato, subumbonato, in centro obscuriore, ad marginem pallidiore et fre- quenter brunneo virgato vel maculato, e calyptra volve floccoso squamoso, squamis albis vel pallido-ochraceis, interdum rgir pilei albo, rare pileo albo, matura margine pilei interdum striatulo lamellis leniter attigentibus, albis: sporis quaternis, albis, globosis, 8-10 : stipite albo, abrupte bulboso, floccoso-squamuloso, 4-10 mm. crasso: annulo amplo, membraneo, superiore, albo vel pallide albo demum murino: bulbo stipitis 2-3 cm. crasso, margine bulbi limbo volve curto, acuto predito. Stipite, bulbo, et interdum carne p tactu brunnescens. This is one of the most common species of Amanita in the eastern United States, and I have observed it for many years. It is usually interpreted as a dark brown form of Amanita phalloides. For number of years I have regarded it as a distinct species. Typi Amanita phalloides which I have collected in France has been com- pared and confirms this opinion. The spores of typical Am. phal- loides are subglobose to oboval or subellipsoid, and bruises do not change to a reddish brown color. The American plant being a dif- ferent species may account for different results, obtained in chemical analyses, from those reported for the typical Amanita phalloides Europe. ‘ No. 24277, type, C. U. Herb., ground, woods, west of Cayt Lake, near Ithaca, N. Y. July 30, 1917. Leva B. Walker collectoi Amanita pachysperma nov. sp. Gregaria vel dispergens, 4-5 ¢ alta: pileo campanulato dein convexo, umbonato, 1.5-2 em. la cinereo, e calyptra volve dense et minute squamuloso, margine ad umbonem striato, carne tenue: lamellis albis, ellipsoideo-ventricosis, basidiis abrupte clavatis, 35-45 X 12-15 »: sporis binis, subellipsoid: obovalibus, 11-15 X 8-10p, sed 208, 18X12, 128, 11 10 X 7, uniguttulatis: stipite albo, leniter bulboso, floccoso-squar uloso, 3-4 mm. crasso:‘annulo membraneo, albo, superiore et pet sistente : limbo volve libero, persistente. 354 ga i aS MS ATKINSON—NEW SPECIES OF AGARICS. 355 In shape and color resembles Amanita cinerea Bres. (Fung. Trid. I, 7, pl. 1, 1881), but spores larger, basidia two-spored, and stria- tions on pileus more extended. No. 3711, type, C. U. Herb., in sandy ground in woods, Blowing Rock, Watauga Co., N.C. Aug. 19-Sept. 19, 1899. G. F. Atkinson collector. Hypholoma comatum nov. sp. Gregarium vel dispergens, 3-5 em. altum: pileo ovali dein campanulato, albo, demum nigro-griseo et leniter striatulato, 0.5-1.5 cm. lato, adolescente radiatis hyphis dense predito, dein sericeo-fibrilloso, margine appendiculato: lamellis stipite adnatis, ellipticis, cinereis dein cinereo-nigrescentibus : cystidiis solum in acie lamellarum, numerosis, clavato-ventricosis vel subcylin- dricis, membrana tenui, 40-55 X 10-12: sporis quaternis, anguste subellipsoideis, purpureo-brunneis vel purpureo-nigrescentibus, levi- bus, 12-14 X 6-7 p: stipite equali, albo, bulbilloso, recto vel flexuoso, fibrilloso-squamoso, I-I.5 mm. crasso. No. 24195, type, C. U. Herb., on ground on humus and buried twigs, lawn by East Ave., Campus, Cornell University, Ithaca, N. Y. July 16, 1917. G. F. Atkinson collector. Hypholoma confertissimum nov. sp. Dense cespitosum, 8-12 cm. altum et 10-20 cm. latum: pileo ovali dein convexo, I-3.5 cm. lato, hygrophano, pallide argillaceo, demum pallide ochroleuco vel pallide albo, in centro argillaceo et ruguloso, velo universali albo- floccoso, demum pileo fibrilloso-squamoso et appendiculato et pallide cinereo-brunneo maculato vel virgato: lamellis stipite adnato-adnexis, demum liberatis, ellipticis, confertissimis, cinereis dein purpureo- brunneis, acie alba: cystidiis in superficie et acie lamellarum sub- fusoideis-ventricosis, 35-50 X II-14: sporis quaternis, purpureo- brunneis, levibus, 5-7 X 2.5-3: stipite equali, recto vel flexuoso, albo, cavo, albo-floccoso-squamoso, fragili, 3-5 mm. crasso, mem- brana mycelii oriente. No. 25063, type, C. U. Herb., tufts of 50-100 individuals from dead roots or buried wood, in forest of mixed hard wood, Hoop Pole Ridge, west of Oakland, Md. Sept. 24, 1917. G. F. Atkinson col- lector. This species is related to Hypoloma aggregatum Peck, but differs in its smaller spores, different cystidia, yellow mat of my- celium, etc. Lactarius nigroviolascens nov. sp. Dispergens, 5-6 cm. altus: pileo 3-5 cm. alto, convexo, incurvato, demum expanso et in centrum depresso, interdum umbonato, atro-brunneo, demum fulvo-olivaceo et in centro obscuriore, pruinoso et ruguloso: margine pilei demum 356 ATKINSON—NEW SPECIES OF AGARICS. _ striato et crenato: lamellis stipite adnatis in lineis decurrentibus, sub- distantibus, albis dein ochraceis et obscurioribus: sporis quaternis, subglobosis: echinulatis, 8-10: stipite squali, pileo concolor minute et densissime tomentuloso: lacte carneque dulci, albo, con- tactu aéris nigro-violascente. No. 24257, type, C. U. Herb., ground, woods, Coy Glen near Ithaca, N. Y. July 25, 1917. Leva B. Walker collector. Related to Lactarius fuliginosus. Lactarius villosozonatus nov. sp. Dispergens wa subcespitosu 3-5 cm. altus: pileo 8-12 cm. lato, ochraceo vel ochroleuco, villoso, zonato, convexo et depresso, demum infundibuliforme: pilei fortissime incurvato, demum expanso et prominenter crenato: 4 lamellis stipite adnexis et sinuatis, angustis, subdistantibus : bi i fs subglobosis vel leniter elongatis, echinulatis, 8-10, vel X 10-12: stipite glabro, 1.5-3 cm. crasso: lacte ex hyalino brun- nescente, tarde acri. No. 20215, type, C. U. Herb., ground woods, Port Jeetereen, te L, N. Y. Aug. 26-Sept. 2, 1904. G. F. Atkinson collector. Lepiota rhombospora nov. sp. Gregaria vel solitaria, 3-4 cm. alta: pileo convexo, 8-15 mm. lato, 1 cm. alto, adolescente ochraceo- fulvo, demum ochroleuco granuloso et squamuloso: squamulis ellis” rotundatis vel obovatis vel pyriformibus preditis: lamellis sti adnatis vel adnexis, sinuatis, ventricosis, bal pen cystidiis solum — in acie lamellarum, lanceoideis, 24-32 X 6-8: sporis quaternis, a lateralibus, albis, a fronte rhomboideis et utrimque acutiusculis, a latere inequilateralibus, levibus, 4-5 X 3-44: stipite equali, pileo concolore, granuloso et squamuloso. Nos. 24323 and 24324, type, C. U. Herb., in edge of mixed woods, on leaf mold, by Stewart’s Camp, Seventh Lake, Adirondack Mts, _ Aug. 16 and 21, 1917. F. C. Stewart collector. Pholiota retiphylla nov. sp. Gregaria, 2-3 cm. alta: pileo lateritio vel vinaceo-cinnamomeo, I-2.5 cm. lato, convexo dein ex- panso, sericeo-tomentosulo, margine incurvato, carne vinaceo tincta lamellis stipite adnatis, leniter emarginatis, angustatis, confertissimis, vinaceo-rufis, superficie lamellarum venosa et reticulata: cystidiis nullis: sporis quaternis, subovalibus, levibus, rufo-luteis, 5.5 X 4-4.5 m: stipite interdum bulboso, demum equali, fibroso-striato, cavo, carne vinaceo-tincta 4-8 mm. crasso: annulo mebranaceo, tenui, vinaceo. No. 18540, type, C. U. Herb., on very rotten moss-covered log, woods, Malloryville moor eee McLean and Cortland, N. Re Aug. 18, 1904. H. S. Jackson and H. H. Whetzel collectors. THE GENUS GALERULA IN NORTH AMERICA. By GEO. F. ATKINSON. (Read April 19, 1918.) a Galerula’ is a genus of yellow-spored Agaricacee including _ small plants or those of medium size, but slender in form, and fragile. The species have no claim to rank of economic importance, — while their ecological role as saprophytes is not large, owing to the _ comparatively small number of individuals. Many species are usu- ally associated with mosses on logs or ground in the woods or swamps. A number of species occur on dung heaps or in recently -manured grass lands. The larger number of species are some shade of yellow, or tawny, or ochraceous. In taxonomic works the genus is usually divided into sections according to external characters and ecological relations. By this method the species are not grouped according to their real affinities, and in a few cases forms not closely related are assembled under a single specific name. __A high degree of internal structural differentiation has taken place in the evolution of the species. In the present study this vantage point has been employed to group the species into sections _ more nearly in accord with their true relationships. ss In some respects the genus, as usually recognized, occupies the same position in the yellow-spored Agarics that Mycena does in the white-spored group. The pileus is usually campanulate; the stem Se s Ass debs ees Ey i wines aN pte See ea Esko oe 1Galera Blume, Bijdr., 415, 1825, is employed for a genus of orchids.. Galera, by Fries in Syst. Myc., 2: 264, 1821, as a tribe of Agaricus, was raised to generic rank by Quélet in 1872 (“ Champ. Jura et Vosges,” 135). There- __ fore while Galera Fries antedates Galera Blume by four years, it was used : Pa as a subgenus, or tribe, and cannot take precedence over Galera Blume, in accordance with rule 49 of the International Rules for Botanical Nomencla- ture. Galerula was employed by Karsten (Bidr. Finl. Nat. Folk., 32, 442) in . % _ 1879 as a genus for several species which he separated from Galera. Galerula ___ Karsten is employed here in the broader sense of the genus with practically __ the same limits as used by Murrill in 1917 (“N. Am. FL,” 10, 161, 1917). _-—s« PROC. AMER. PHIL. SOC., VOL. LVII, ¥, AUGUST 19, 1918. 357 358 ATKINSON—GALERULA IN NORTH AMERICA, has a cartilaginous rind; in the young stage the margin of the is straight, lying parallel with the stem, i. ¢., the margin r incurved; a distinct veil is usually absent, or if present it usually of sufficient tenacity to form a distinct annulus on the This feature of the straight pileus margin is often difficult to mine, since, in many cases, the number of individuals of a collected is often too few, and they are in a too advanced “— development to determine the relation of the pileus margin to stem. Nevertheless, to one possessing some familiarity with genus, the external form or “habit” of the plants, taken in junction with their color, serves in a large number of cases reasonably sure provisional means of differentiation from the r genera. i In some well-recognized species of the genus, bowers margin of the pileus is incurved in the young stages and sei comes straight. A notable example is Galerula angusticeps. thermore, there appear to be structural characters of great imp tance which indicate that certain species with a convex pileus, o1 few with the margin incurved when young, are more closely : to the Galerula type than to other types, for example, certain sp which, on the basis of the “ habit” principle would fall in the Naucoria, Pluteolus, or even Hebeloma; while a few species plac in Galerula because of the “habit” formula, are excluded when 1 morphological, or structural, principle is employed as the basi determining relationships. On the basis of the morphological principle the species can be ranged in two groups. First, those in which the pileus is hon geneous, and second, those in which the pileus is corticated. © This pr ciple of grouping the species was, in fact, employed a quarter o century ago by Fayod? who carried the principle still furth recognizing two generic concepts. He recognized Conocybe f species with a corticated pileus, and Galera for those we ah geneous pileus. : In his concept of the genus Conocybe, the cortex of the was merely indicated as “ pseudoparenchymatous.” This de 2 Fayod V., “Prodrome d’une histoire naturelles des Agaricinés,” Sci. Nat. Bot., VIL. 9: 181-411, 1889. : ATKINSON—GALERULA IN NORTH AMERICA. 359 of the cortex does not appear to be sufficiently clear, and does not separate the true species of this section of Galerula from species of _Naucoria and Hebeloma having a more or less pseudoparenchymatous cortex. According to the principle followed in the present arrange- ment, the cortex of the pileus in this section of the genus is more highly differentiated than the simple pseudoparenchymatous cortex. There is an outer layer clavate to pyriform cells, in the young stage forming a more or less definite palisade layer. In age these cells in certain species swell to a large size, as well as certain cells beneath this layer, so that the cortex takes on a pseudoparenchymatous aspect, but close examination shows the large pyriform cells of the _ surface, and it can be seen that the pseudoparenchyma is of a dif- ferent origin and structure from that which I have termed simple pseudoparenchyma in certain species of Naucoria and Hebeloma. Still further differentiation is manifest in the structure of the lamellz. All species recognized here as belonging to Galerula are- provided with specialized cells in the hymenium, usually termed _cystidia. These vary in form and in their distribution on the Jamellz. They will not be discussed in detail at the presemt time- In the synopsis of the species presented below their form and ar- rangement is clearly indicated. The species with a corticated pileus _ are regarded as more highly specialized, those of the last section having reached the highest stage of specialization with corticated pileus and lecythiform or stopper-shaped cystidia.’ This specializa- tion is further indicated by the large number of species with two- spored basidia. There appear to be rather clear evidences of progression in de- velopment from the simpler forms of the species with a homogeneous pileus, and diverging in two lines, one line culminating in the corti-— cated species, the other line retaining the homogeneous structure of the pileus. Galerula angusticeps proves to be an interesting species in this connection. Some individuals have the pileus homogeneous in structure, while others show a rudimentary cortex of the Galerula type. It is therefore placed in both. of the principal groups in the a synopsis. After a critical study of the species in allied genera has been made, it may be possible to make some suggestions concerning the relationships within the group. 360 ATKINSON—GALERULA IN i AMERICA. This structural study of the genus Galeria | in North based on an examination of material collected by myself duri years in this country and in Europe. The latter material was lected principally in Sweden and France. The determinations confirmed or made by Dr. Robert Fries, son of Elias Fries for species from Sweden, and by E. Boudier for the species irc France. Through the courtesy of the State Botanist, Ke A the State Museum at Albany. Thewaie the courtesy of Dr. Britton, director of the New York Botanical Garden, and Dr. W. A Murrill, I have examined also nearly all of the types in the He barium of the N. Y. Bot. Gard., and some European species. _ There are 58 species from North America in the Bhs synopsis. eee SYNOPSIS OF THE SPECIES. PILEUS HOMOGENEOUS. Cystidia flask-shaped-lanceoloid-fusoid, Cystidia on sides and edges of lamelle. i | , Galerula besseyi (Pk.) Murr. N. Am. FI. 10, 163, 1917. _ Galera besseyi Peck, N. Y. State Mus. Bull. 131, 35, pl. 5, figs. 1: 1909. Nebraska. Galerula hypnorum (Fr. emend. Pat.) Atkinson. Agaricus (Galera) hypnorwm Fr. ex Schrank. Syst. Myee a 2 1821. Galera hypnorum Quélet Champ. Jura et Vosges, 1: 137, Emend Patouillard, Tab. analyt. Fung. 1: 103, fig. 230, 1 Galerula hypni Murr. pr. pte. N. Am. Fl. 10: 163, 1917. Galerula reflexra Murr. N. Am. FI. 10: 169, 1917. Oregon, Mexico, Europe. Galerula muricellospora nov. sp. Gregaria vel solitaria: cm. alta: pileo campanulato, interdum convexo, hygrophano, s' ATKINSON—GALERULA IN NORTH AMERICA. 361 fulvo vel ochraceo, demum pallidiore, glabro, homogeneo, 3-6 mm. lato, 3-5 mm. alto: lamellis stipite adnexis vel anguste adnatis, sub- ventricosis, subdistantibus, pallide ochraceis: cystidiis in superficie et in acie lamellarum, sublanceoloideis vel subfusoideis, ad basem subventricosis, 40-90 X 10-15 p: sporis binis, ferrugineis, ovatis vel late subfusoideis, a latere inequilateralibus, demum minute echinu- latis, 10-14 X 6-9: stipite zquali, recto vel flexuoso, ochraceo- - brunneo, sursum pallide luteo et pruinoso, fistuloso, I-I.5 mm. crasso. No. 7837, type, C. U. Herb., among living mosses, Coy Glen, near Ithaca, N. Y. Oct. 12,1901. J. M. Van Hook collector. New York, Colorado. Galerula paludicola nov. sp. Gregaria vel dispergens, 5-13 cm. alta: pileo campanulato, dein convexo-expanso et umbonato, I-2.5 em. lato, hygrophano, -ochraceo, striatulo, demum _ ochroleuco, homogeneo: lamellis stipite adnatis, leniter emarginatis, pallide- ochroleucis vel pallide-brunneis: cystidiis in superficie lamellarum subfusoideis, 45-50 X 10-12: in acie lamellarum cystidiis simi- laribus: sporis quaternis, subellipsoideis, 10-14 X 6-7: stipite zequali, interdum bulbilloso, leniter floccoso-squamoso, luteo vel rubescente-luteo, deorsum obscuriore, 2-3 mm. crasso. 23576, type, C. U. Herb., among sphagnum, Malloryville Moor between McLean and Cortland, N. Y. Oct. 17, 1913. G. F. Atkin- son collector. New York. Cystidia only on edges of lamelle. IT. Galerula cerina (Bres.) Atkinson nov. sp. Galera cerina Bresadola n. sp. “in Herbario non publicaris.” Gregaria, 1.5-4 cm. alta: pileo campanulato vel convexo, obtuso, interdum subumbonato, glabro, striato, 2-8 mm. lato, 2-5 mm. alto, hygrophano, ochraceo-fulvo vel pallide ochraceo, demum ochroleuco, homogeneo : lamellis stipite adnatis, subventricosis, pileo concoloribus : cystidiis solum in acie lamellarum, numerosis, subcylindricis, ad basem subventricosis, interdum subcapitatis, rectis vel flexuosis, 30-60 X 8-12: sporis quaternis, ferrugineis vel ochraceo-fulvis, ovatis vel subnavicularibus, ad basem latioribus, levibus, 8-14 XK 6— 8»: demum membrana spore rugulosa vel saccata: stipite equali, subbulboso, pileo concolore vel decorsum obscuriore, sursum pruinoso, I mm. crasso. 362 ATKINSON—GALERULA IN NORTH AMERICA. No. 25019, type, C. U. Herb., on humus near sphagnum in 2.5 miles south of Oakland, Md. Sept. 16, 1917. G. F. / collector. ce Maryland, North Carolina, New York. New York. Galerula heterocystis nov. sp. | Galerula hypni Murr. pr. pte. N. Am. Fl, 10, 163, 1917. 3 Forma et color Galerule hypnorum: cystidiis in acie lz n ; confertissimis, in superficie nathis, tibiiformibus vel clavate natis, 25-40 X 10-14: sporis quaternis, ovatis, a latere i eralibus, 12-15 X 6-7 p. Type “ Fungi of Jamaica, No. 435, Cinchona, 4,500-$,200 0 te. 3 N. Y. Bot. Gard. Herb. 2 Galerula inculta (Pk.) Murr. N. Am. FI. 10, pe 1917, Galera inculta Peck, N. Y. State Mus. Ann. Rep. 4t, 169, -, New York. Galerula lignicola Murr. N. Am. FI. ro, 165, ge New York. Galerula minuta (Quél.) n. comb. Galera minuta Quélet, Champ. Jura, Vosges, 3, 438, pl. 5 fg. 1875. New York. Galerula parvula Murr. N. Am. FI. 10, 162, 1917. Tennessee. Galerula sphagnicola nov. sp. Gregaria vel solitaria, 5-8 alta: pileo campanulato dein expanso, umbonato, 1.5-2.5 cm. | ATKINSON—GALERULA IN NORTH AMERICA. 363 __ hygrophano, cinnamomeo-brunneo, striato, demum pallidiore: con- texto homogeneo: lamellis stipite adnexis, in lineis decurrentibus, distantibus, cinnamomeo-brunneis, cystidiis solum in acie lamellarum, __ numerosis, crassis, cylindicis et deorsum ventricosis, interdum flex- uosis, 30-45 X 9-I2m: sporis quaternis, ovalibus vel subnavicular- ibus, ad basem crassioribus, 8-11 X6-8 »: membrana spore interdum rugulosa vel subinflata: stipite zquali, pileo concolore sed interdum pallidiore, demum pallide-cinnamomeo, glabro, sursum pruinoso, cavo, 2—4 mm. crasso. No. 18587, type, C. U. Herb., among sphagnum in very wet places, Junius, N. Y.. Sept. 15, 1904. H. H. Whetzel and H. S. Jackson collectors. New York, Alabama. Galerula sphagnorum (Fr. ex Pers.) Murr. N. Am. Fl. 10: 167, 1917, emend Atkinson. Agaricus hypnorum sphagnorum Pers. Syn. Fung. 386, 1801. Name only. Type not determined. Agaricus hypnorum sphagnorum Fr. Syst. Myc. 1, 267, 1821. Name only. Agaricus sphagnorum Lasch, Linnaea, 3: 417, 1828. Galera sphagnorum Sacc. Syll. Fung. 5: 869, 1887. © Galera hypnorum var. umbonata Peck, N. Y. State Mus. Bull. 25, 655, 1899. Maryland, New York, Europe. Galerula subhypnorum nov. sp. Galerula hypni Murr. pr. pte. N. Am. Fl. 10, 163, 1917. Gregaria vel dispergens, 2-6 cm. alata: pileo campanulato-con- vexo, obtuso vel umbonato, ad marginem striato, hygrophano, fulvo vel ochraceo-fulvo, demum pallidore, homogeneo, 4-10 mm. lato: lamellis stipite adnatis, subdistantibus, ochraceo-fulvis: cystidiis solum in acie lamellarum, numerosis, lanceoloideis vel subcylindricis et ad basem subventricosis, in apice interdum crassioribus, 40-70 X 6-1I-: sporis quaternis, ochraceis, a fronte subellipsoideis vel subfusoideis, a latere leniter inequilateralibus, levibus, 10-14 XK 6-8 p: stipite pileo concolore sed pallidiore, ad basem interdum obscuriore, cavo, I mm. crasso. This species differs from G. hypnorum in the absence of cystidia 364 ATKINSON—GALERULA IN NORTH AMERICA, on the sides of the gills, in their different form, and in the I colored, thinner-walled spores. New York, Maine, Washington, Europe. Cystidia tibiiform, only on edges of lamelle. ITI. Galerula bryophila (Pk.) nov. comb. Galera bryophila Peck, N. Y. State Mus. Ann. Rept. 54, ; Galerula hypni Murr. pr. pte. N. Am. Fl. 10: 163, 1917, New York, Colorado, Mexico. Galerula lasiosperma nov. sp. Gregaria vel solitaria, 6-9 ¢ alta: pileo ovali-campanulato, dein convexo et expanso, obtus hygrophano et ad marginem striatulo, 1 cm. lato, 4-5 mm. alto, ca taneo, demum fulvo vel ochraceo-fulvo, homogeneo: lamellis adnatis, ventricosis, subdistantibus, ochraceis : cystidiis solum lamellarum, tibiiformibus, ad basem subventricosis, sersum te et capitatis, 30-45 X 6-8: sporis quaternis, ferrugineis, ¢ vel subreniformibus vel fabiformibus, leniter echinulatis X 5-7»: stipite zquali, recto vel flexuoso, pileo concolore s¢ lidiore, sursum pruinoso, ad basem albo-myceleoideo, cavo, 1-1 mm. crasso. a No. 25033, type, C. U. Herb., on sphagnum in open field, ‘e Farm, Cranesville moor, Western Maryland. Sept. 18, 1917. G. Atkinson collector. : Maryland. Galerula lenticeps (Pk.) nov. comb. Agaricus lenticeps Peck, N. Y. State Mus. Ann. Rept. 31, 34, ‘187 Naucoria lenticeps Sacc. Syll. Fung. 5, 838, 1887. New York. . Galerula pistillicystis nov. sp. Gregaria, 2-3 cm. alias = hemisphaerico-campanulato, 4-10 mm. lato, hygrophano, luteo, dein pallide ochraceo, homogeneo: lamellis stipite adnatis ventricosis, ferrugineis : cystidiis solum in acie lamellarum, numero pistilliformibus, ad basem ventricosis, sursum teretibus et capit 25-35 X 6-8: sporis quaternis, ferrugineis, subellipsoideis, levi 7-10 X 3.5-5 pm: stipite equali, pileo concolore sed pallidiore, g surstum pruinoso, cavo, I mm. crasso. No. 24072, type, C. U. Herb., on a rotten log among mosses ATKINSON—GALERULA IN NORTH AMERICA. 365 swamp bordering on Labrador Lake, near Apulia, Onondago Co., _ N.Y. June 12, 1917. G. F. Atkinson collector. New York. Galerula rufipes (Pk.) Murr. N. Am. FI. 10, 164, 1917. Galera rufipes Peck, N. Y. State Mus. Ann. Rept. 42, 116 (20), 1889. New York. Galerula semilanceata (Pk.) n. comb. Galera semilanceata Peck, Torr. Bot. Club Bull. 23, 415, 1896. Washington. Galerula stylifera nov. sp. Gregaria, 3-5 cm. alta: pileo ovali _ dein campanulato, matura expanso et subumbonato, I-2 cm. lato, _hygrophano, ochroleuco-fulvo vel pallide-ochraceo, demum ochro- leuco, glabro, homogeneo: lamellis stipite adnatis, emarginatis, ven- tricosis, ochraceis: cystidiis solum in acie lamellarum numerosis, ‘styliformibus vel tibiiformibus, ad basem subventricosis, apice capi- __tato, 30-45 X 5-8: sporis quaternis, ochraceis, subellipsoideis, 6-8 XX 3.5-4.5: stipite equali, ochraceo-fulvo vel castaneo, subvilloso, & sursum pruinoso, 2 mm. crasso. ~ No. 24399, type, C. U. Herb., on very rotten wood mold and de- caying leaves of Pinus strobus, McGowan’s Woods, near Ithaca, N. Y. Oct. 10, 1917. H.E. Stork collector. New York. Galerula subannulata nov. nom. Naucoria lateritia Murr. N. Am. FI. 10, 172, 1917. Maryland, New York. Galerula tibiicystis nov. sp. Gregaria, 6-9 cm. alta: pileo cam- panulato dein convexo et interdum umbonato, 1-2 cm. lato, hygro- phano, fulvo, glabro, demum pallido-ochraceo, homogeneo: lamellis stipite adnatis uncinatis, ventricosis, subdistantibus: cystidiis solum in acie lamellarum, numerosis, tibiiformibus, ad basem subventricosis, apice capitato, 30-45 X 7-9: sporis quaternis, ovatis vel subellip- soideis, a latere inequilateralibus, ferrugineis, 8-12.5 X 5-7.5p: stipite zequali, pileo concolore sed pallidiore, sursum pruinoso, 2-2.5 mm. crasso. No. 25080, type, C. U. Herb., on sphagnum in a spruce moor 366 ATKINSON—GALERULA IN NORTH AMERICA. near Miller’s Run, 3-4 miles north of gr Md. Sept. 25, G. F. Atkinson collector. Maryland, New York, Mabiarhulette, Sweden (Upsala). Cystidia lecythiform or stopper-shaped, only on edge of IV. | | Galerula angusticeps (Pk.) Murr. N. Am. FI. 10, 168, 1917 Galera angusticeps Peck, Torr. Bot. Club Bull. 24, 143, 1897. Conocybe angusticeps Murr. Mycologia, 4, 248, 1912. California. Galerula teneroides (Pk.) Murr. N. Am. FI. 10, 166, 1917. Agaricus teneroides Peck, N. Y. State Mus. Rept. 29, 39, 1878, Galera teneroides Sacc. Syll. Fung. 5, 861, 1887. a : This species is placed here provisionally. The portion s th type material examined was in such poor condition that the stru 1 of the pileus could not be determined. | New York. Piteus CorTIcATED. Cystidia large, fusoid or clavate, or sublanceoloid, not sp cia Cystidia on sides and edges of lamelle, emerging. __ V. Galerula kellermani (Peck) Murr. N. Am. FI. 10, 165, 1917. Galera kellermani Peck, Jour. Myc. 12, 148, pl. 89, 1906. ay Ohio. Galerula cervinialba (Murr.) nov. comb. Prunulus cervinialbus Murr. N. Am. F1. 9: 326, 1916. Cystidia subfusoid : basidia large, sterile, therefore the pure lamellz. New York. ATKINSON—GALERULA IN | NORTH AMERICA. 367 Cystidia only on edge. VI. Galerula -cyanopus nov. sp. Gregaria vel solitaria, 2-3 cm. alta: pileo ovali, campanulato-convexo, dein expanso, obtuso, ad mar- ginem striatulato, fragili, brunneo-ferrugineo, 5-10 mm. lato: cortice pilei cellis pyriformibus przedito: lamellis stipite adnexis, angustatis, ochraceis: cystidiis in acie lamellarum numerosis, ad basem ventri- cosis, in apice cylindricis vel mucronatis, 30-40 X IO-I7: sporis quaternis, ellipsoideis, apice minute truncato, 8-10 X 5-6»: stipite I-I.5 mm. crasso, leniter bulboso, albo ad basem cyaneo, sursum pruinoso, deorsum leniter velutino, fragili. Ground among grass, No. 23302, type, C. U. Herb., Ithaca, N. Y. New York. Galerula filipes nov. sp. Gregaria, 4-6 cm. alta: pileo ovali dein campanulato, obtuso, ochraceo, 3-6 mm. lato: cortice pilei cellis obo- valibus vel pyriformibus predito: lamellis stipite late adnatis, dente decurrentibus, ochraceis vel ochraceo-fulvis: cystidiis solum in acie lamellarum, lanceoloideis vel anguste ovatis, ad basem ventricosis, 30-50 X 10-18: sporis quaternis, ferrugineis vel fulvo-ochraceis, ellipsoideis, 7-8 X 4-5 u: apice spore truncato: stipite zquali, bul- billoso, sursum pruinoso, I mm. crasso. The type material consists of specimens collected by Peck in a grass plot at North Elba, Essex Co., Adirondack Mts., N. Y., now _ in the N. Y. State Mus., Albany, as Galera capillaripes (see N. Y. State Mus. Bull. 94: 32, 1905). - This species resembles in form and color Galerula capillaripes, but the basidia are four-spored, the spores are much smaller and the cystidia are different. New York. Galerula mirabilis nov. sp. Gregaria vel dispergens, 4-6 cm. alta: pileo campanulato-convexo, .dein expanso et umbonato, 1.5-2 em. lato, glabro, striatulo vel rugoso, vinaceo-cinnamomeo, in cen- trum castaneo, cortice cellis obovalibus vel pyriformibus przdito: lamellis stipite adnexis, subdistantibus, subventricosis, fulvo-ochra- ceis, acie lamellarum alba: cystidiis solum in acie lamellarum, nu- merosis, abrupte clavatis, 40-70 X 12-18: sporis binis, a fronte navicularibus vel cymbiformibus, a latere inequilateralibus, demum minute tuberculatis, fulvis, 12-25 X 7-11 p: stipite equali, recto vel flexuoso, cavo, glabro, pileo concolore sed pallidiore, sursum pruinoso. No. 15117, type, C. U. Herb., in mixed woods in a swamp near McLean, N. Y. June 17, 1903. H. H. Whetzel collector. New York. 868 ATKINSON—GALERULA IN NORTH AMERICA. — Galerula plicatella (Pk.) Murr. N. Am. FI. 10: aL 191 Agaricus coprinoides Peck, Buffalo Soc. Nat. Sci. Bull. I, 52, 187, Agaricus plicatellus Peck, N. Y. State Mus. Rept. 29, 66, 1878. ‘ Galera coprinoides Sacc. Syll. Fung. 5, 867, 1887. Galera plicatella Earle, Torreya 3, 136, 1903. New York. Galerula sulcatipes (Pk.) Murr, N. Am. Fl. 10, 166, 191 Agaricus sulcatipes Peck, N. Y. State Mus, Ann. Rept. 35, 132, ‘Galera sulcatipes Sacc. Syll. Fung. 5, 866, 1887. New York. . This species is very closely related to Galerula tortipes fro: which it differs only in its smaller size. cat * Galerula tortipes (Mont.) Murr. N. Am. FI. 10, ee 1937 Agaricus tortipes Mont. Syll. Crypt. 119, 1856. Galera tortipes Sacc. Syll. Fung. 5, 867, 1887. - Ohio. Galerula viscosa (Clem.) nov. comb. Galera viscosa, Clements, Cryp. Form, Coloradensum, Ne 1906, Colorado. Cystidia specialized, lecythiform or stopper-shaped, on edge of lamellae. Cystidia on sides imbedded, clavate or clavate mucronate on edge lecythiform or stopper-shaped. VIL. Galerula ayeyets nov. sp. Gregaria, 6-10 cm. alta: p ovali dein campanulato, 1.5-2.5 cm. lato, ochraceo, ochraceo-ft vel ferrugineo, demum pallidiore, sparsim villoso: cortice pilei pyriformibus vel ovalibus predito: lamellis stipite adnexis, cystidiis in acie lamellarum lecythiformibus, 15-22 X 6-8 p, in s ficie lamellarum clavatis, non projicientibus: sporis binis, a fre late ellipsoideis, 12-22 x S12: stipite zequali, pileo concolore | pallidiore, striato, sursum pruinoso, 2-4 mm. crasso. ATKINSON—GALERULA IN NORTH AMERICA. 369 a No. 127, Pacific Slope Fungi, distributed as “ Galera tenera” by q _C. F. Baker, type in C. U. Herb. “A common little mushroom on | decayed horse manure in old pastures. Foothills near Stanford University, Santa Clara County, California, November 30, I90I, Coll. C. F. B.” | | a This species is related to Galerula macrospora, but differs in the striate stem, imbedded cystidia on the sides of the lamelle, etc. California, Europe (Trento, May, 1901, Bresadola). Cystidia only on edge. VII. Galerula angusticeps (Pk.) nov. comb. See above under section IV. Pileus homogeneous. Some speci- mens have a rudimentary cortex. Galerula antipus (Lasch) emend Atkinson. Agaricus antipus Lasch in Linnaea, 3, 415 (no. 401), 1828. Galera antipus Quél. Champ. Jura et Vosges, 1, 136, 1872. Galera antipus Gillet, Champ. France, 553, with figure, 1878. _ Galera antipoda Sacc. Syll. Fung. 5, 863, 1887. Spores in front view limoniform to subangular, in side view ellipsoid. North Carolina. Galerula capillaripes (Pk.) Murr. N. Am. FI. 10, 163, 1917. _ Galera capillaripes Peck, Torr. Bot. Club Bull. 26, 166, 1899. ° iy Ohio, New York. Galerula crispa (Longyear) Murr. pt. pte. N. Am. FI. 10, 167, 1917. Galera crispa Longyear, Bot. Gaz. 28, 272, 1899. Michigan, New York. 370 ATKINSON—GALERULA IN NORTH AMERICA. — Galerula crocospora (Berk. & Curt.) Murr. N. Am. FI. 10, 168, 1917. Agaricus crocosporus B. & C. Ann. Mag. Nat. Hist. I, 12, 4 Galera crocospora Sacc. Syll. Fung. 5, 866, 1887. . ‘South Carolina. This es probably belongs here but the tainty. Galerula curta nov. sp. Gregaria, 4-6 cm. alta: pileo campar lato dein subexpanso, 2-4 cm. lato, I-2 cm. alto, obtuso, an r striato, ochraceo vel ochraceo-fulvo demum ochroleuco, : cellis obovalibus vel pyriformibus predito: lamellis stipite a ellipsoideis, ochraceis: cystidiis solum in acie lamellarum, num lecythiformibus, 15-25 X 7-10: sporis binis, obovalibus ra oF ellipsoideis, ochraceis, 10-18 X 8-11 »: stipite quali, glabro, tria- tulo, sursum pruinoso, cavo, 3-5 mm. crasso. Nos. 3209 and 3210, type, C. U. Herb., in grass in curbing a of Buffalo St., Ithaca, N.Y. July 20,1899. G. F. Atkinsom| CO - New York. . Galerula distantifolia Murr. N. Am. Fl. 10, 169, 1917. Mexico. Galerula flava (Pk.) Murr. N. Am. FI. 10, 166, 1917. Galera flava Peck, N. Y. State Mus. Ann. Rept. 45, 79, (19) ; New York. Galerula flexipes (Karst.) nov. comb. Galera fleripes Karsten, Myc. Fenn. 3, 371. New York. Galerula fragilis (Pk.) Murr. N. Am. Fl. 10, 164, 1917. Galera fragilis Peck, Torr. Bot. Club Bull. 24, 144, 1897. Kansas. Galerula “ lateritia.” A medium-sized Galerula very closely re- lated to Galerula crispa, which may be only a variety. It is usually) regarded in this country as equal to Agaricus lateritius Fr.; 1 am n ATKINSON—GALERULA IN NORTH AMERICA. 371 at present convinced of the correctness of this interpretation. It is _ common in lawns and grassy places in rainy weather, May to July, and also occurs on dung. Galerula macrospora nov. sp. Galera tenera obscurior Peck, N. Y. State Mus. Rept. 50, 130, 1899. Gregaria, 6-15 cm. alta: pileo campanulato, I-2.5 cm. lato, 6-10 mm. alto, ochraceo vel ferrugineo, non striato, demum ochroleuco, cortice pilei cellis pyriformibus vel obovatis predito: lamellis stipite adnexis, ellipsoideis, ochraceis: cystidiis solum in acie lamellarum, numerosis, lecythiformibus, 15-25 X 6-Q: sporis binis, late ellip- soideis, 12-25 X 8-15: stipite zquali, recto vel flexuoso, sursum pruinoso, pileo concolore, cavo, 1.5-2.5 mm. crasso. No. 15759, type, C. U. Herb., on ground among mixed grasses and moss (Hylocomium squarrosum), on the edge of a coniferous wood, near Stockholm, Sweden. Aug, 24-28, 1903. G. F. Atkinson collector. North America (New York), Europe (Sweden and France). Galerula mexicana Murr. N. Am. FI. 10, 169, 1917. Mexico. Galerula neoantipus nov. sp. Gregaria, 3-7 cm. alta: radix 2-5 cm.: pileo campanulato-convexo, 1.5-2.5 cm. lato: cortice pilei pseudoparenchymato: lamellis cum in Galerula antipus: cystidiis solum in acie lamellarum, numerosis, lecythiformibus, 15-30 X 6-9 p: sporis quaternis, ellipsoideis, 12-17 K 6-9 n. On newly seeded lawn, Middlebury, Vt., Aug. 1896, E. A. Burt collector, in E. & E. N. Am. Fungi, second edition, No. 3510, type, C. U. Herb. The specimens in Cooke’s Ill. Brit. Fungi, No. 463, “spores ellipsoid, 16 X 8,” probably belong here. Vermont, (?England). Galerula ovalis (Fr.) Karsten, Bidr. Finl. Folk 32, 443, 1879. Agaricus (Galera) ovalis Fr. Monogr. Hymen. Suec. 1, 389, 1857. Galera ovalis (Fr.) Gillet, Champ. France 3, 554, 1876. New York, North Carolina, Europe. 372 ATKINSON—GALERULA IN NORTH AMERICA. Galerula pilosella (Fr. ex Pers.) Atkinson emend. Agaricus (Galera) tener var. pilosellus Fr. Syst. Myc. 1, 266, ia Agaricus pilosellus Pers. Synop. Fung. 387, 1808. ¥ Gregaria vel solitaria, 3-6 cm. alta: pileo campanulato, ad mar. ginem interdum expanso, 1-2.5 cm. lato, 1 cm. alto, hygrophano, ochraceo vel cinnamomeo-brunneo, striatulato, demum pallide ochraceo vel fulvo-ochraceo, leniter villoso: cortice pilei cellis pyri formibius et rare cystidiis praedito : lamellis stipite adnexis, angustatis, ellipsoideis, numerosis, luteo-ochraceis vel ochraceo-cinnamoméis : cystidiis in acie lamellarum lecythioideis, 15-24 X 6—Qm: sporis quaternis, anguste ellipsoideis, apice minute truncato, Bi stipite zquali, subtiliter villoso. No. 20851, C. U. Herb., type of the emended species, on ground in a spruce forest north of Pontarlier, Jura Mts., France, Aug. 21, 1905, G. F. Atkinson collector. No. 24314, C. U. Herb., near Seventh Lake, Adirondack Mts., N. Y. Aug. 15,1917. G. F. A. = F. C, Stewart collectors. New York, Maryland, Europe. Galerula plumbeitincta nov. sp. Gregaria, 3-5 cm. alta: pileo” convexo, dein campanulato, I-1.5 cm. lato, adolescente lubrico, sub-— striato, plumbeitincto, cortice cellis clavato-pyriformibus predito: lamellis stipite late adnatis, subdistantibus, ventricosis, ochraceo-— cinnamomeis: cystidiis solum in acie lamellarum, numerosis, ampul- leformibus, frequenter subcapitatis, 25-35 X 12-15: sporis qua- — ternis, ochraceo-cinnamomeis, late ellipsoideis, 12-15 X7-IOp: stipite albo dein plumbeitincto, sursum pruinoso, deorsum substriato, | cavo, 2-3 mm. crasso. : On dung hills, Cascade Glen, Ann Arbor, Mich. June 9, i C. H. Kauffmann collector, no. 565, type, C. U. Herb. . Michigan. Galerula procera nov. sp. Gregaria, 10-12 cm. alta: pileo cam- panulato, obtuso, ad marginem demum leniter expanso, 3-4.5 cm. lato, 2-2.5 cm. alto, ochraceo-fulvo, dein ochraceo vel pallide-ochro-- leuco, glabro, non striato: cortice pilei cellis obovalibus vel pyri-— formibus preedito : lamellis stipite adnexis, ellipsoideis, ochraceis: cystidiis solum in acie lamellarum, numerosis, lecythiformibus, 18-25 X 6: sporis quaternis, ochraceis, a fronte late ellipsoideis, apice minute truncata, 12-20 X 8-IIp: stipite «quali, ad basem leniter crasso, recto vel flexuoso, glabro, striato, sursum pruinoso, pileo con-— colore sed pallidiore, 3-4 mm. crasso. . ATKINSON—GALERULA IN NORTH AMERICA. 373 No. ggi10, type, C. U. Herb., on ground, in humus, among leaves in moist woods, Buttermilk Gorge, near Ithaca, N. Y. July 23, 1902. C. H. Kauffmann collector. New York. Galerula pulchra (Clem.) Murr. N. Am. FI. 10, 166, 1917. alero pulchra Clements, Bot. Surv. Nebr. 4: 22, 1896. _ Nebraska. Galerula spartea (Fr.) - nov. comb. i E Galera spartea Quél. Champ. vane Mier £ ee ye: = @&«New York. Galerula tenera (Fr. ex Schaeff.) Murr. | N. Am. FI. 10, 166, 1917. 7 * Agaricus (Galera) tener Fr. Syst. Myc. 1, 265, 1821. Galera tenera Quélet, Champ. Jura et Vosges, 136, 1872. _ Throughout greater part of N. Am., Europe. Galerula tenerella (Atkinson) Murr. N. Am. FI. 10, 164, 1917. Galera tenerella Atkinson, Ann. Myc. 7, 369, 1909. New York. Galerula tenuissima (Weinm.) nov. comb. _ Agaricus tenuissimus Weinm. F1. Ross, 219, 1836. Galera tenuissima Quélet, Assoc. Fr. Av. Sci. 1884, 280, pl. 12, fig. 8, 1885. New York. Species NEEDING FuRTHER STUDY. GALERULA GLABRA Murr. N. Am. FI. 10: 163, 1917, has not been examined. GALERA GRISEA Earle, No. 43 Inform. An. Estac. Centr. Agron. _ Cuba 1: 237, 1906, needs further examination of fresh material. PROC. AMER. PHIL. SOC., VOL. LVII, Z, AUG. 19, 1918. 374 ATKINSO GALERA sIMULANs Earle, Inform. An. 1 : 1: 236, 1906, is probably identical with Ga Species EXCLUDED. — GALERA ALBA Peck, Torr. Bot. Club Bull. 24: 14; Agaricus (GALERA) CALLISTUS sas Buff. I: 52, 1873. é Acaricus (GALERA) EXPANSUS Peck, Buff, 4 I: 52, 1873. GALERA RETICULATA Peck, N. Y. State Mus. 1901. _ GALERA STRIATULA Clements, Bot. Surv. Nebr, GALERA VERSICOLOR Peck, Torr. Bot. Club Bull, CorNELL UNIVERSITY, Irnaca, N. Y., atts May, 1918. ORGANIZATION, REPRODUCTION AND INHERITANCE IN PEDIASTRUM. (Pirates V. anp VI.) By R. A. HARPER. (Read April 19, 1918.) I have discussed in previous papers (’08, ’12) the problems of _ organization as seen in strictly ccenobic plants in which the col— ony shows little or practically no differentiation in either the struc-. _ ture or the functions of its cells. In Pediastrum we have a type in which at least the incipient steps in differentiation’can be recognized- The margins of the flat plate-shaped cell colonies are in some species quite entire, in others more or less lobed or toothed, and _ present the problems of the development and inheritance of specific _ and differentiated form in plants at a relatively critical stage. We { have here the first beginnings of cell, and, we may say, tissue differ- entiation. In such ccenobes as most species of Spirogyra, Hydro- dictyon, Gonium and others, though the colonies have definite and probably adaptive structure, the cells are all alike in form and func- tion, but in certain species of Pediastrum the lobed peripheral cells differ markedly from the interior cells of the colony. In other species the lobing is almost equally developed in all the cells. The genus thus presents us with the processes of differentiation in vary- ing degrees of expression in what are plainly rather closely related species. I have also discussed elsewhere (716) the interrelations of the cells in the eight- and sixteen-celled colonies of Pediastrum Borya- num as giving the basis for a definite conception of plant types and the comparison of this species with the other members of the group brings out still more clearly the idea of biological form types as I have discussed it there. The group is also well adapted to illus- ‘ trate the relations of heredity and environment in morphogenetic 375 376 HARPER—ORGANIZATION, REPRODUCTION processes and to give evidence as to the possible mode of origin interrelations of different types. The shape of the cell is the main basis of group distinctions the genus. As Braun (’55) first pointed out, the number of « in the colony has been shown not to be diagnostic of species larger aggregates. Fig. 21 shows two daughter colonies, one 1 sixteen and one with thirty-two cells, both of which came from same thirty-two-celled mother colony. The shape of the cells « given species varies in minor details of proportion about a norm the species. The form of the cells of different species are of 3 me fundamentally diverging types. Fic. 1. Pediastrum integrum. Cells almost without spines or lob The colony shows only fourteen cells but it is possible, as is common for species, that it is partially two layered. about 225. 7 Fic. 2. P. simplex. Cells with one spine. Configuration te XX about I50. tae Fic. 3. P. clathratum. Cells with two very long and slender per and two shorter basal lobes, large intercellular spaces. Configuration s+ I Colonies bilaterally symmetrical about the axis m,n. about 350. ee PIG, (2s Fic. 2. It is a question of first importance morphogenetically whether these cell forms are hereditary or whether they are newly 2 under the stimulus of the intercellular environment in each gen I tion. It is plain that the form is not inherited directly as such since the two-spined cells do not divide in such a way as to produce once two equivalent two-spined daughter cells. 2 The main distinctions in the group are between forms who cells have no spinous processes or very rudimentary ones, and ho whose:cells have more or less developed spines. These distinetic AND INHERITANCE IN PEDIASTRUM. 377 _ have been regarded as of subgeneric rank and are the basis of the _ integrum (Fig. 1), Monactinia or simplex (Fig. 2), the Diactinia _ or two-spined (Fig. 3), the Triactinia or three-spined and the _ Tetractinia or four-spined series. A tendency to the formation of a four-spined type of cell is shown in the species P. biradiatum Meyen, and the peripheral cells of P. tricornutum Borge are de- scribed as having three spines. Braun (55) recognized the diag- nostic value of cell form in making four sections of the genus, the Monactinia, P. Simplex, Anomopedium, P. integrum (Nag.), the Diactinia, P. Boryanum, etc., and the Tetractinia, P. Ehrenbergit. A further class of differences which has been given specific rank _ is found in the degree of similarity in form between all the cells of the colony. In P. Boryanum (Fig. 4) the interior cells are quite different in form from the peripheral series, while in P. integrum (Fig. 1), on the one hand, and P. clathratum (Fig. 3) on the other, the cells are very much alike throughout the colony. It seems natural to assume that such forms as P. integrum (Fig. “a 1) with its oval cells, sometimes with two papillz, but with no _ striking form differentiations, represent the more primitive species, _ though perhaps P. integrum itself is only an environmental form as _ Iam suggesting in another paper, and that evolution has progressed toward the simplex type with one spine.on the one hand, and the _two-spined type on the other. The tendency to the development of spines more strongly on the peripheral cells probably came first (P. Boryanum, Fig. 4) and later the development of the strongly four- lobed form in all the cells of the colony (P. clathratum, Fig. 3). Positive evidence that the evolution of the group has followed this course is, however, lacking. The origin of the tendency to pro-— duce spines is also not obvious. We shall find a direct relation between the cell form and the intercellular relations in the colony in comparing the colonies of P. Boryanum (Fig. 4) and P. asperum (Fig. 5) and we have evidence for the assumption that the result of a direct environmental influence has been transformed into an inherited cell character in such cases but no support for such an hypothesis is found in the three-spined cells of P. tricornutum Borge or the four-spined cells of P. biradiatum Meyen. - As noted, the number of cells in a colony has been shown te 378 HARPER—ORGANIZATION, REPRODUCTION like P. Boryanum may have eight, sixteen, thirty-two or sixty-fe cells. Still certain species tend toward higher and others tow lower cell numbers. P. Ehrenbergii commonly occurs with four, eight, or sixteen cells. The rule that the number of cells in nized. Individuals with fifteen cells instead of sixteen cells or Fic. 4. Fic. 5. Fic. 4. Pediastrum Boryanum. Peripheral cells with two spines very slender for the species. Interior cells with merely a reéntering angle to indi- cate the spines. No intercellular spaces. Configuration 1+5-+10. Colony bilaterally symmetrical about the axis m,n. > about 300. : Fic. 5. P. asperum. Cells with two fairly long peripheral spines and two short, blunt basal lobes. Intercellular spaces well developed, the curves which — bound them suggesting the origin of the cell lobes by catenoidal deformation. 2 Configuration 1-+-5-++-10. Colony bilaterally symmetrical about the axis — m,n. Possible polarites of the cells suggested by the plus and minus signs. about 600. thirty-one instead of thirty-two cells occur, but as Nitardy (14) has emphasized, they are great rarities and may be properly re- garded as abnormal. The figure of P. integrum ee shows only fourteen cells (Fig. 1). The relative development of the spine is directly correlated with : that of the intercellular spaces and we may consider first the organi-_ zation of a form with well-developed intercellular spaces. Perhaps AND INHERITANCE IN PEDIASTRUM. 379 _ the most abundant species as I have found them, next to P. Borya- num Kg., is P. asperum (Figs. 5 and 6). P. asperum has been regarded by many as a form of P. Boryanum. Whether or not it is to be considered a good species, the forms as I have found them reproduce the type so far as cell form, involving as it does the char- acteristic intercellular spaces, is concerned with a high degree of constancy. I have never seen a colony with the asperum cell form produced from a Boryanum parent nor a colony without intercellu- lar spaces from an asperum parent. I have been able to study and photograph P. asperum in all stages of the vegetative growth and reproduction of the colonies and it may well serve to illustrate the two-spined forms with well- developed intercellular spaces and cells similar throughout the col- ony in contrast with P. Boryanum with intercellular spaces small or wanting and interior cells only slightly or even not at all two- spined, as I have already described it (’16). ARRANGEMENT OF THE CELLS AND INTERCELLULAR SPACES IN THE SIXTEEN-CELLED COLONIES OF P. ASPERUM. The spatial interrelations of the cells are essentially the same in the typical sixteen-celled colony of P. asperum as in the typical sixteen-celled colony of P. Boryanum, as I have described it else- | where (16), and as it has long been known in the literature. But in P. asperum the cells are all very much alike in form, the interior cells having only slightly shorter lobes than those on the periphery of the colony. We may consider first the sixteen-celled colony, which is perhaps the most common, though 8-, 32-, 64- and 128- celled colonies are found. I shall leave the colonies with 32, 64, 128, etc., cells without discussion at this time, since certain further elements in morphogenesis implied in the larger number of units are involved which I shall take up in a later paper. A common ar- rangement in the thirty-two-celled colony is 1+6+ 10+ 15, as noted by Braun (55). The presence of six cells in series II. of the thirty-two-celled colony as compared with the five in series II. of the sixteen-celled colony introduces quite fundamentally differ- ent relations between the oblong four-lobed form of the cells and the natural surface tension factors in such groups. 880 HARPER—ORGANIZATION, REPRODUCTION As in P. Boryanum, the general arrangement of the cells in most common form of sixteen-celled colony of P. asperum is in the center with five around this and two cells in the third or o series, the 1-+- 5 + 10 group, as Nageli and Braun called it. Na (’49, p. 93) quite fully described these conditions for the veiled: by comparing the areas of the concentric circles, assuming the r: diameters of the cells are the same, showed geometrically that emi cell numbers are what he calls the “ most natural,” that is, the cells so arranged come the nearest to filling the spaces in the concentric circles as well as being the ones most commonly found. I have described this arrangement (’16) as coming the nearest to that of a least-surface configuration for such a group of rounded cells lying in one plane. As I shall further discuss in another connec- tion, Nageli is, however, probably mistaken, at least for P. Borya- num, as Braun’s figures (’55) show, in saying that for eight cells the arrangement 2-++-6 is much less common than the “more nat- ural” arrangement I + 7. I shall follow the same method of numbering the cells bake in the case of P. Boryanum (Fig. 25, and 16, Fig. 1b), making t central cell No. 1 and proceeding outward as shown in the diagram (Fig. 7). The colony is bilaterally symmetrical as in P. Boryanum the axis bisecting cells 1, 4, 7 and 12 and passing through the face of contact of cells 2 and 6. For describing the structure of the colony more conveniently we may here, as in the case of P. Boryanum, call the ends of ing axis of the colony its m and n poe ; respectively. . The central cell is in contact with five cells. The cells of. the 3 second series are each in contact with six cells and the cells of the outer series are alternately in contact with three and four cells ‘ The cell walls meet in threes on the principle of least surfaces, ex- cept where cells 2 and 6 are in contact as a pair right and left of the axis of the colony. In P. Boryanum this grouping of inter- sections is universal in the more regular sixteen-celled colonies the paired contact of cells 2 and 6 is one of the most obvious differ- ences between the two species in their intercellular relations. — If the tips of the lobes of a cell of series II. be connected in serial order, g', g®, d*, d?, by straight lines, we have a trapezoid wi AND INHERITANCE IN PEDIASTRUM. 381 _ its longest side away from the center of the colony. In the case of the central cell we have a trapezoid with its longest side connecting the apices of the basal lobes. As in P. Boryanum, five of the outer cells stand radially outward from the five cells of series II. and five stand radially opposite the surfaces of contact of the cells of "series II. Cells eight and sixteen, nine and fifteen, ten and fourteen, eleven and thirteen are paired to the right and left of the axis of the col- Fic. 6. Fic. 7. Fic. 6. Pediastrum asperum. Very typical colony organization as de- scribed for Fig. 5. This colony served as the model for the type diagram for the species given in Fig. 7. < about 325. Fic. 7. Type diagram for P. asperum. Angles of intersection of all the walls made 120°. Intercellular spaces and peripheral spines sketched free- hand from the colony shown in Fig. 6. Numbering of cells and lettering of angles as in diagram of P. Boryanum, fig. 35. ony. The adaptive and symmetrical adjustment by which, with the products of cellular bipartition to be grouped, we get five instead of six cells in series II., thus leaving ten for series III., which can be so spaced as to place one cell opposite each cell of series II. and one cell opposite the surface of contact of each pair of cells in series 382 HARPER—ORGANIZATION, REPRODUCTION II., is just as conspicuous here as in the case of P. Boryanum. — delicacy of the pressure and contact responses which prevents th placing of six, the normal least-surface number, around the central cell, thus leaving only nine for the peripheral series, and resu in quite asymmetric contact relations between series II. and affords good evidence of the efficiency of cellular interactions i production of morphogenetic adjustments. : The four-lobed form of the cells and the spacing of the pape of series II. about the central cell and the ten cells of series about the five cells of series II., leave free intercellular areas an id the symmetrical distribution of Hea areas in such fashion as to maintain most perfectly the rigidity of the colony and the equality of the cells has led to the development of one of the very o differences between P. asperum and P. Boryanum. This is haps the most conspicuous difference between the two species. © cells instead of being in contact on the entire extent of their adjacent surfaces show a series of intercellular openings which perfo the plate-shaped colony like the holes in a sieve. An intercellt space is formed between each pair of contact walls of the sixt cells. They are of such form and are so placed that the inte: cells of the colony have the four-lobed form of the peripheral c« The cells of the whole colony are thus, as noted above, much | differentiated in form than those of the colony of P. Boryan The inner cells differ from the outer cells, broadly speaking, on in that their peripheral lobes are blunted and shortened where they meet the inner lobes of the outer series. P. asperum confor much more nearly to the definition of a coenobe as a colony of sells quite similar in form and function. One might conclude that we have here a simpler type in wh cell differentiation has not yet gone so far as in P. Boryanum. am inclined to the view, however, that in reality P. asperum is more specialized type and expresses more fully the form-dete ing tendencies which have led to the development of the spines lobes on the peripheral cells of the colony of P. Boryanum. As we shall find from a study of the method of reproducti in the colony, the cells all inherit alike the tendency to the fo lobed form. This is obvious even in the interior cells of the col AND INHERITANCE IN PEDIASTRUM. 383 of P. Boryanum. They all show a reéntering angle on their periph- eral sides. In P. asperum this tendency to the four-lobed form has come more fully to expression, overcoming to quite a degree the adhesion of the cell surfaces so that each cell may assume the form to which its inherited tendencies predispose it. The possible origin of this tendency in the intercellular relations of such groups of cells, I shall discuss later. In form these intercellular spaces vary from youth to the adult reproductive condition, as will be noted later in discussing reproduction. They also show some degree of variation in differ- ent individuals of apparently the same stage of development. In their curved outlines they express very fully the tensions existing in the protoplasm of the cells and the contact relations of the cells in the colony. In colonies with atypical cell arrangement, they may become highly varied in different regions, expressing fully the dif- ferent and irregular tensions set up in such abnormal or subnormal groupings (Figs. 26-28). In number, form and position these intercellular spaces form two series, four between the central cell and series II. and ten be- tween series II. and III., the outer series. Of the four intercellular spaces of the inner series two, a and c (text-fig. 7) lie on the axis of the colony, and the other two, b and e, are symmetrically placed to the right and left. These four spaces are of three forms and sizes, a, the largest, is broadly triangular to shield-shaped (Figs. 5, 6) ; c, the second in size, is roughly an asymmetrical ellipsoid with its greater convexity toward the central cell, and b and e are still smaller and more or less symmetrical ellipsoids with pointed poles, somewhat lemon-shaped. The outer series consists of five larger, ovoid to shield-shaped spaces, i, k, 0, t and q, and alternating with them five smaller spaces, f, h, 7,1 and p. One of the larger spaces, g, and one of the smaller, 7, are bisected by the axis of the colony. The remaining four of the larger group may be regarded as forming two pairs, a pair, i and k, symmetrically placed to the right and left of the axis, nearer the m pole of the colony, and nearer together than o and f, the similarly placed pair toward the m pole. The remaining four 884 HARPER—ORGANIZATION, REPRODUCTION smaller spaces also form two pairs, f, p and h, l, placed as : larger pairs but with the nearer pair toward the n pole of the ec In their relation to the cells between which they occur the inte cellular spaces form two classes, those formed at the points w three cells come together and where three walls intersect and t formed by retraction of the plane contact walls of two adj: _ cells; in other words, the intercellular spaces which are bounded three cells and those which are bounded by two cells. There six of the first class, five in the outer series, and one, a, in the inne series, and there are eight of the second class, five in the outer § r and three in the inner series. a These spaces, as noted, get their outlines and positions frou 1 tendencies of the cells to assume their hereditary four-lobed fo: and not merely as an expression of surface tension and rounding up such as results in the triangular intercellular spaces in many loc parenchymatous tissues and in the colony of Gonium. The fit and definite form of each cell in which it differs from its neigh arises during growth and in irregular colonies these differences m ! be very marked (Figs. 25-27). They may consist in inequality t the length of the lobes with greater or less blunting of their » curving of the lobes, or deformation of the whole trapezoidal out- line of the cell till it becomes rhomboidal or some other form. growth of the cell will result in the protoplasm as flowage and ter sions in the direction of the growing lobes and such tensions exertec at four points of the viscous mass will tend to produce a simple « noidal deformation of the whole such as arises in all semi-fl viscous bodies under tension of any sort. Retraction would nat-— urally occur on the surface of contact midway in the regions of ten. sion simulating a tendency of the whole mass to break up into f droplets. The curves bounding the intercellular spaces in the ures (see Fig. 5 especially) are the obvious expression of tens exerted on the cell mass in the direction of the four lobes. degree to which this tendency to catenoidal deformation will co to expression will depend on the viscosity of the protoplasm, th adhesion of the cells to each other, etc. It is in these parti ale that the species differ and the same species may under varying ditions or at different stages of development differ in such charac- AND INHERITANCE IN PEDIASTRUM. 385 teristics. There is no question that P. Boryanum may at times, and in the same fashion, develop intercellular spaces, though I have _ mever seen them so symmetrically and strongly developed in this - species as in P. asperum. The production of the general four-lobed outline is, as evidence given below shows, a result of the inherited form of each cell rather than its pressure relations in the colony. There is further to be considered a certain suggestion of a want of correlation between the form of the cells and their pressure rela- tions in the colonies of Pediastrum. Regarding the young cells as equal and rigid globular bodies in one plane, the central cell in a sixteen-celled group would be in contact with and pressed by five cells symmetrically placed about it. Each of the cells of the second series would be in contact with four cells which are not equally spaced about it. The cells of the third series in such a figure would be alternately in contact with one and two cells, correspond- ing to the contacts of the cells of series III. in the actual colony which are alternately in contact with three and four cells (Figs. 6,7). If now the globular cell bodies yield to pressure and flatten upon each other, filling the empty spaces in the group, they will (except the central cell) tend to become oblong and four-cornered like the mature cells of Pediastrum. On the other hand, in the commonest form of the thirty-two-celled colonies the central cell is im contact with the six cells of the second series, and each of the six assumed globular bodies would be in contact with five cells not simi- larly placed. In the third series of ten, two such globular cells would be in contact with four each, six would be in contact with three each, and two would be in contact with two each. In the outer series fourteen of the fifteen globular cells would be each in contact with one cell, while one would be in contact with two cells, and yet all these cells in both the sixteen- and thirty-two-celled colo- nies grow into the characteristic four-lobed form with the resulting contact relations found in the ‘mature colony in which in the sixteen-celied colony, for example, the cells of the second series are each in contact with and pressing against six cells and the cells of the outer series are alternately in contact with one and two cells each. The new specific contacts and pressures arising in growth are due to the inherited four-lobed form of the cells, which none 386 HARPER—ORGANIZATION, REPRODUCTION the less, in my opinion, may have arisen in evolution from the pres- sure and contact relations of the young cells in the common sixteen- celled colony, regarding them merely as surface tension globules. We may conceive, as is described below, that transitorily the young - cells at once flatten upon each other and thus give the figure of P. Boryanum with no intercellular spaces. The spaces then start to form with the first growth of the young colony in such a fashion as to make the resulting four-lobed cells as nearly isodiametric as" is possible with the numbers involved and the inherited tendency _ to adhere together in the most compact figure possible (groups of three). It is quite conceivable, then, for the sixteen-celled colony, — that out of the incompatibilities resulting from the laws of biparti- tion and surface tension, both operating in the case of an organism > | whose cells tend to adhere in colonies, the four-lobed form of the cells has been developed from cell forms such as we find in P. integrum. It is easy to recognize certain physical stimuli which have been present; first, surface tension tending to keep each cell — isodiametric ;second, perhaps during growth, functional hypertrophy | : as we have observed its action in the case of Hydrodictyon; third, — adhesion tending to keep the contact surfaces plane; fourth, cate- — noidal deformation, due to growth tension in the direction of the lobes. Each cell is originally like its fellows—a globular or ovoid bit — of jelly. Adhesion during the writhing motions of the final stages of the swarming period leads to its being flattened upon its neigh- bor. This may be favored by low turgor at this period, but we have little evidence on this point. In P. Boryanum the cells, as a rule, adhere permanently over their entire original contact surfaces and thus maintain their primary contact relations. In P. asperum, however, the surfaces of the cells tend to separate. The degree and location of this separation may be determined by the tendency of the cells to catenoidal deformation as they grow rapidly in the axes of greatest resistance as developed by their four-lobed form. — The cells of the whole colony, with the exception of the central cell, : are longer in their tangential axes than in their radial axes, since five cells fill the space which six should occupy in the second series - of a least-surface group and ten fill the spaces of twelve in the outer series. Surface tension tends in each cell to equalize these two | AND INHERITANCE IN PEDIASTRUM. 387 axes. Taking cell 4 as an example, the pressures upon it are from the cells 1, 3, 11, 12, 13 and 5 and in tending to round up while developing its pressure relations with these cells it forms the wedge- shaped surfaces of contact inclined to each other at the character- istic angles of 120°. The axes of major growth of the cells after the early stages are just what they would be expected to be on the principle of functional hypertrophy, though as shown below, the cells are capable of reaching their typical form even when their con- tact with other cells has been reduced to a minimum (Figs. 29-33). The strongly four-lobed cells are in their form adapted to the expression of a typical surface tension configuration for the whole sixteen-celled colony, but this form does not apparently depend for its expression in ontogeny on the contact relations to which it is so perfectly adapted. Whether in phylogeny, as noted, it was not a result of the cellular interactions existing in the colony is, of course, quite another question. The fluctuating variations in the form of the cells and of the intercellular spaces are the direct ex- pression of the effects of functional response to environmental influ- ences operating in ontogeny. But the cell is able independently to develop its fundamentally four-lobed form. : It may be that we have here a case of an adaptive form char- acter which arose in direct response to and as it were determined by environmental factors of cellular interaction which has become so fixed that it is now transmitted in cell division and comes to typical expression without the need of the stimuli of adhesion and pressure relations which originally called it into being. The considerations here developed apply, of course, especially to the sixteen-celled colony. The case of the thirty-two-celled colony will be considered, as noted below, in another connection. It is during growth that the intercellular spaces are developed. As the cells of the colony reach their mature size the intercellular Spaces become again relatively smaller. The cells now increase their volume by expanding in the direction of least resistance, that is, toward the intercellular spaces. As they prepare for reproduc- tion by swarmspore formation this swelling becomes very marked (Fig. 8). The ovoid spaces become triangular, as does the large 388 HARPER—ORGANIZATION, REPRODUCTION space. The spaces between two cells become flattened to thin convex lense-formed figures, This change of form is accompanied by increased density of the cell content, which also becomes more deeply green in color. It appears that the cells are accumulating reserve storage products to be used in the reproductive period. — The distinction between this sort of growth and that at the earliest - period, when the cells were developing their long lobes and large intercellular spaces, is obvious and parallels the distinction between the periods of growth and of maturing with the accumulation of — _ reproductive reserves in adult many-celled plants and animals. REPRODUCTION IN P. ASPERUM. Al. Braun (’51). reported De Bary’s discovery of the gametes of Pediastrum in 1851 and Askenasy (’88) has described their con- jugation and shown the similarity of the life history to that of Hydrodictyon. Askenasy (’88) shows fully that the final forma- : tion of the new colony is essentially the same process when it comes ~ from a zygospore through the polyeder stage and when it is formed — directly by asexual reproduction from a cell of a parent colony. — The method of asexual reproduction in Pediastrum has been correctly understood since the work of Meyen (’28), who saw the escape of the swarmspores, their free motion and later their com- bination to form the young colony and Nageli (’49) correctly con- cluded that since the number of cells in a colony is regularly a multiple of two the cells must have arisen by bipartition. To be sure, Conn (’o8) rather casually describes the swarmspores as con- tinuing to divide after they have come to rest. Whether this state- ment of Conn’s is based on his own observations or is merely an — a priori guess at what seems to him probable is not clear. Nageli (’49) in 1849 had already convinced himself to the contrary. There ~ can be no question but that the process of cell nul is com- pleted before the swarming period. I have photographed more or less ssidieaa fully a large number of colonies with mother cells at various stages of division. There can be no question that here we have a process of successive biparti- tion of a multinucleated sporeplasm. Smith (’16) has figured from — AND INHERITANCE IN PEDIASTRUM. 389 sections stages showing the cleavage of numerous segments of the sporeplasm and my photographs show the first division, four-cell stage, etc. Smith’s figures show clearly that the cells divide by furrows and that regular karyokinetic figures appear in nuclear _ division. The swarmspore is uninucleated and the sporeplasm be- - comes multinucleated before cell division begins. When the colony is to reproduce itself the cells become extremely _ plump and to some extent lose their deeply four-lobed form (Fig. 4 8). The spinous projections tend to be drawn in perhaps, though : they do not entirely disappear in any of the species. This is well _ shown in figures of P. asperum, 8, 9, and 10, and at a later stage when cleavage is well along in Figs. 11, 12.and 14. These forms would hardly be recognized as belonging to P. asperum if it were not possible to find all conceivable intermediate forms between these _ very turgid types and the immature colonies with slenderly lobed - cells, such as are shown in Fig. 6. In such species as P. clathratum _ (Fig. 3) with its very deep sinuses and slender-branched cells it is plain that the four-pointed cell form maintains itself strongly even against the effects of increase of turgor. As Askenasy (’88): noted, the nuclei are readily recognizable at this stage. Askenasy mentions, but without figures, that the cell division proceeds by a series of successive bipartitions. Figs. 10, a and b, show the mother cell dividing into two. Cells c, d, e in the same figure show at least the beginnings of the four-cell stages. It is very difficult owing to the density of the cell contents at these stages to bring out the cleavage clearly in photographs, though the furrows are easily visible on focusing. The divisions apparently occur by constriction furrows from the plasma membrane inward ____as described by Smith (’16), as noted, and by Timberlake (’o2) for _ ‘Hydrodictyon. There is no indication of the formation of cell _ plates, so far as the appearance of the living material is concerned. That we really have constriction furrows forming here and not merely lines of simultaneous cleavage as claimed by Swingle (’97) for Stypocaulon and by various authors for a number of fungi (Baum, ’oo, Kusano, ’o9, Barrett, 12, Griggs, ’12) is suggested by PROC. AMER. PHIL. SOC., VOL. LVII, aa, AUG. 19, 1918. 890 HARPER—ORGANIZATION, REPRODUCTION the rounding up at the edges of the clefts resulting in open fu over the surface before the cell is completely cut through. In these plump-celled forms which are about to reproduce the intercellular spaces are much reduced in area and their form is somewhat changed (Figs. 9-14). The cells may, however, reach — maturity and divide without becoming so plump as those shown the cases just noted. In the stage shown in Fig. 13, cleavage complete and yet the cells are much more deeply lobed than those in Fig. 11, where sleavage has only reached the eight- to sixteee celled stages. The study of the living material is very convincing as to the existence here of division by centripetal furrowing, with rectangular intersection of the cleavage planes. The first cleavage plane is regularly in the short axis of the cell (Fig. 10). The successi planes of division cut each other approximately at right angles through the four- and eight-celled stages. In the sixteen-celled and thirty-two-celled stages there are modifications due to the irregular outline of the mother cells. In general, however, the whole series of divisions tends to illustrate the principle of rectangular inters section. These processes of division as a rule take place during the niet as Smith has noted (’16), and swarming occurs at daylight, or a little later in cool weather. This is also the case in Hydrodictyon (’08). By sealing up with hot paraffine an ordinary watér mount containing well-grown colonies of Pediastrum this rhythm may be readily disturbed. It is probable that the accumulation of waste products is an important factor in such cases. At any rate in prep- arations sealed up on the previous evening colonies in various stages of division may be found the next day. The division goes on much more slowly under these conditions or may be inhibited all together. I have kept cultures so sealed for long periods without visible de- terioration of the colonies. The young colonies may continue to grow for some time, but after the first day or two no cell division occurs. The colonies may ‘be kept alive for six months if they are kept sealed and not allowed to dry out. The appearance of the mother cell after division is practically AND INHERITANCE IN PEDIASTRUM. 391 complete is shown in Figs. 12, 13 and 14. Not all the division lines can be brought out in the photograph. The mass is too thick and it is plain that the swarmspores are in at least two layers. In these reproductive stages, when the cells have become very turgid, and especially during the stages of cell division, the mother colonies tend ‘to lose their flatness and become curved and bent in various ways so that it is extremely difficult to get any large number of cells in focus at once. Colonies with cells in which division is complete when freshly mounted at daybreak are liable at any moment to show the beginning of the swarming period. The cells of a colony rarely swarm all at once. As a rule only part of the cells divide in any one night and in the same way there is a succession in the initiation of active movement in the mother cells, suggesting that internal as well as external factors may be concerned in bringing about the active swarming period. When once the swarm cells begin to move, however, the succeeding stages are normally run through with great speed and with no halting. The daughter cells first seem to round up against each other and lose the rectangular outlines which have been maintained by their mutual pressure. Slight twitchings and - quiverings can be seen in the mass whose significance is hard to grasp but almost immediately the swarmers begin to glide upon one another and show writhing, struggling movements. The mother cell now bursts by the familiar crescent-shaped slit on its upper or under surface and its contents slide out in the form of a sack or vesicle containing the writhing mass of young swarmspores. This sack, Al. Braun (’51) rightly observed, is the inner layer of the wall of the mother cell. It is elastic and expands to over twice the diameter of the mother cell as soon as it is free. In thus expanding, how- ever, it still maintains the four-cornered outline of the mother cell, as my photographs (Figs. 18-20) show. The two peripheral spinous projections of the mother cell are still recognizable on it as two symmetrically placed papilla. In the increased space provided by the expansion of the vesicle the movements of the swarmspores become much more active. The writhings become zigzag dashings to and fro. The swarmspores shoot about in all directions through the mass, which has become much looser, and into the open space 392 HARPER—ORGANIZATION, REPRODUCTION around it. This is a perfectly free swimming movement like that | at the corresponding stage in Hydrodictyon. Oltmanns (’o4, p. 194) remarks that the zodspores ‘probably remain connected threads of protoplasm but apparently has only Klebs’ (’90) mis contention as to the corresponding stages in Hydrodictyon as the basis for the statement. A single spore can frequently be followed clear through the mass or halfway round its periphery. This most active swarming period continues for three to four minutes when conditions are favorable. It is followed by a rather sudden slowing down and now the outlines of the future colony appear. The sudden appearance of order out of the chaos of swarming bodies is most striking. The circular outline of the plate appears first and the peripheral cells seem to slow down in their movements, while those in the interior are still quite active. The free-swarming period thus passes over into a second and much longer period of writhing and struggling in which the cells do not. move far from their places, but push this way and that between and over each other, crowding and turning around and over without getting completely out of connection with their neighbors as they did in the earlier free-swimming stage. Coincidently with the slow- ing down of the movements of the swarmspores they begin to take’ on the four-lobed form of the adult cells. This change is very conspicuous. The oval swarmspores seem as if they were about to divide into two (Figs. 15-20). Each cell, as the figures show, — almost seems to be made up of two pear-shaped halves, the narrowed eo ends of the halves being the future spines. This sudden assumy tion by the swarmspores of the four-lobed form is a very striking and conspicuous fact at this stage and is accompanied by rapid growth and mutual pressure between the cells. Walls are formed and the cell contours become more clear cut and definite. All these changes begin with the slowing down of the movements of the swarmspores. The process suggests very strongly the effort of the cells to get into very specific relations with each other and as close together as possible, thus forming the compact plate-shaped colony. As the movement dies away, the behavior of individual cells can be followed with more exactness. A change of position of one cell AND INHERITANCE IN PEDIASTRUM. 393 apparently may lead to a change of position by all its neighbors and these mutual adjustments and readjustments continue till it seems that a sort of equilibrium is reached. Smith has figured the changes in position of the cells at successive intervals (’16). As noted, the cells in the peripheral series apparently get their definitive positions first, while movement is still quite active in the interior of the young colony. I have not been able to determine just when cilia appear in _ the swarmspores or ‘when and how they disappear, but doubtless the duration of the active movements of the cells is an index of the limits of the ciliated stage. _ To give some hint of the size relations and general appearance ___ of the mother cells and the daughter colonies as they are first formed. _ I have had reproduced in Figs. 15, 16 and 17 photographs of three stages in the development of a mother cell into a young colony. The difficulties in photographing such stages are very great, as rather long exposures are required and it is not easy to find any consider- able number of cells in the same focal plane. Fig. 15 shows the last two cells of a mother colony in which cleavage is complete and swarming is about to begin. All the other cells are already empty and the walls of some of them show faintly. Two young colonies partly out of focus lie nearby. The irregular grouping of the Swarmspores in the mother cell shows nothing of the organization of the future colony. There is no evidence of mosaic or any other _ type of predetermined form inheritance here. In Fig. 16, from a photograph taken a few minutes later, swarming has begun in one of the two mother cells. The vesicle has escaped and lies partly __ beneath the remaining mother cell. The swarmspores are in active 4 motion and appear only as a gray cloud. It would, of course, re- quire a very short exposure to catch these moving bodies in sharp _ focus. This figure is less highly magnified than the preceding one . and, in all, seven young colonies and part of an eighth are shown _ lying near, all of them having come earlier from the same parent colony to which the cell just swarming belongs. The figure is printed more deeply, so that the contents of the remaining mother cell appear black, and parts of the outlines of the adjacent empty cells appear more clearly. The attempt was made to focus on the 394 HARPER—ORGANIZATION, REPRODUCTION active mass of swarmspores but the plane of the picture is a too high and the faintness of the swarming group is partly « this fact. Fig. 17 is from a plate exposed four or five min later. The young colony lies partly beneath the second mother cell but its outlines are in evidence and it is seen to be made up three concentric rings of cells like the other young colonies 1 about. Six of these sister colonies and parts of two more are sho in this figure. The plane of the picture is a little below that. Fig. 16 and other parts of the young colonies are in focus. I! no positive proof of continued ciliary activity in colonies as old these, but none the less as one watches them they are seen to ; from side to side and even shift their position slightly. Such nc ments may well be due to slight currents in the water. As noted, these photographs cannot be regarded as success but they give a notion of the relative sizes of the mother cell. the young colony at the time of its birth and the general relat under which the morphogenetic processes take place. With the c plete cessation of the movements of its component cells it is ob y at once that the organization of the future colony has been achie With the establishment of the peripheral series the young spin projections are in general all pointing radially outward (Figs. 20). In the same way in the inner series, generally speaking, two-lobed cells have the long axes of their lobes in the radii of colony viewed as a whole, just as in the full-grown colony. — same exceptions and variations from these rules of orientation be found in such young colonies as are present in the older The relative number of cells in the outer and inner series, ‘tl contacts and the shape of the intercellular spaces are all fixed as th remain throughout subsequent growth and development. Ra rapid growth continues for an hour or so, and with the increase size of the individual cells their mutual pressure and the deve ment of the intercellular spaces make the specific oT the colony more conspicuous. Summarizing, we may distinguish roughly five phases vegetative reproduction of Pediastrum. a 1. Nuclear division which goes on during the vegetative gro of the cells and by which they become multinucleated. AND INHERITANCE IN PEDIASTRUM. 395 2. Cell division by repeated bipartitions in general according to _ the principle of rectangular intersections. 3. Escape of the vesicle contemporaneous with quiverings and then slow writhings of the swarmspores for a very brief period, passing at once into 4. The free-swimming stage, lasting several minutes, in which the swarmspores dart about in entire freedom from each other. _ 5. Slow writhing stage, in which the swarmspores gradually _ perfect the spatial interrelations found in the adult colony. The description of reproduction just given may be regarded as representing the process near its optimum as to speed and efficiency. The resulting young colonies (Figs. 15-17) are fairly regular in the arrangement of their cells and rounded in outline. Such con- ditions are as a rule only achieved in the case of colonies favorably situated and brought under observation only a short time before swarming occurs. In the case of colonies sealed under a cover glass -as described above for from six to twenty-four hours before repro- duction occurs there are some marked modifications in the process which lead to characteristic changes in the shape of the young col- ony. One of the commonest deviations from type in the colonies as found in nature is seen in the tendency to be oblong or oval instead of circular in outline. The explanation of this modification in form can be discovered at once by observing the reproduction of colonies which are unfavorably placed. As noted, divison and swarming may sometimes be completely inhibited if the colonies are mounted under a cover glass and sealed before they are full grown. If, how- ever, they are about ready to swarm and, for example, are mounted some time in the night before the morning on which they would nat- urally swarm, the process may take place later in the day. Swarm- ing may thus be delayed for several hours and in such cases it is at once seen that it is less vigorous. The free-swarming period is shorter or may disappear entirely, the swarmspores only writhing ‘and twisting about without really getting out of contact with each other. The whole active stage is shortened. This may go so far in certain media that the swarm- spores scarcely move at all (’88). The fact is to be at once observed in all such cases of a less vigorous and shortened swarming period 396 HARPER—ORGANIZATION, REPRODUCTION that the resulting colonies are never circular in outline but al somewhat elongated in one axis. The spinous projections are constantly radial in their position and the intercellular cor and intercellular spaces are far from what I have described aboy as typical. Photographs of such irregular colonies still inclosed the mother vesicle are shown in Figs. 18-20. It is difficult, as n to bring out the vesicle in a photograph. I have traced over dilute India ink the outlines of the vesicle in Figs. 18 and 20 bring out more sharply the points involved. Fig. 19 is left printed, but the faint outline of the vesicles and their papille be made out. In Fig. 20 an edge view of a young colony in vesicle is shown. The figures show clearly enough that the long axis of these irregular oblong and oval colonies lies in the long axis of the mother cell. The vesicle, although gelatinous and swollen, is apparently elastic and always maintains the outline of the mother — cell even to the retention as noted of two papille representing spinous projections. The conclusion is obvious that these unfay ably situated colonies with reduced activity in swarming are unable ‘ to achieve the typical compact circular plate form and the outline of the young colony is influenced by the oblong shape of the | closing vesicle. The achievement of all the nice adjustments n sary in making a typical least-surface figure requires more « than the cells of these weakened colonies are capable of and as result they conform more or less to the outlines of the vesicle. Sometimes the result is a flattening of the outline of 1 colony along one or both edges. Again the result is a more elli soidal form. In cases of extreme weakness the colonies may be rather angular (Fig. 23), conforming quite completely to the outlin of the mother cell and, as Askenasy (’88) noted, sometimes no escaping from it. In these extreme cases the colonies are practicall always irregular in all their dimensions, with the cells more or le piled upon each other, so that the colony is more than one lay thick. The spirious projections also tend to disappear under the conditions. The effect of environment in modifying and disturbing the morphogenetic processes is thus most clearly shown, and can class a whole series of such divergences from type as strictly epigenetic and environmental in their origin. The shape and struc- AND INHERITANCE IN PEDIASTRUM. 397 ture of any given colony of Pediastrum conforms to its type in so far as the environment permits. The multiplicity of divergences 2 from the type form for the species is an index of the varying de- grees of favorableness in the surroundings of the parent colony and mother cells. The relations of unstable equilibrium between .the units of a group of sixteen or thirty-two, as compared with a group of nineteen or thirty-seven, give unusual opportunity for the play of such environmental influences. FLUCTUATING VARIATIONS IN THE INTERCELLULAR RELATIONS IN THE COLONIES OF P. ASTERUM AND P. BoryANUM. The evidence from the above account of asexual reproduction is clear that the colony of Pediastrum is formed by the interaction of a group of free-swarming zodspores without the possibility of any predetermination of its form as such in the arrangement of the parts of the mother cell. A cell can apparently fill any place in the group which forms the young daughter colony. A colony under favorable conditions may attain the rounded outline of a typical least-surface configuration for such a group of cells or under less favorable con- ditions it may conform more or less wholly to the outline of the mother cell even to the extent of remaining two-layered. I have also described above the general arrangement of the cells and inter- cellular spaces in a typical adult colony of P: asperum, and we may now turn to the question as to the kind and degree of variability which the colonies as they are found in nature exhibit. In the continuous disk of the typical sixteen-celled colony of P. Boryanum I have shown (’16) that the angles of intersection of the cells walls vary considerably in different ‘parts of the colony, the correspondingly placed angles right and left of the axis of the col- ony tending to be equal. The angles of contact in any fairly typical sixteen-celled colony of P. asperum (Fig. 6) are, so far as I am able to determine, quite constantly 120°. The difficulties of measure- ment here are much greater than in P. Boryanum, since the presence of intercellular spaces reduces the surfaces of the cells in contact and the length of the lines by which the angles are measured. Small variations are no doubt quite regularly present, but they are within 398 HARPER—ORGANIZATION, REPRODUCTION the limits of error by any method of measurement I have bi to use. If the ordinary semicircular protractor is used and placed for each reading the results show fluctuations of from 1° between the angles. If, however, the circular protractor with marking the angles of 120° is carefully placed so that the angles can, as it were, be simultaneously read, it is at once ap pare how closely the angles approximate 120°, and that the deviations most cases are so slight as to be practically indistinguishable appearances due to inequalities in the thickness and density of cell walls, middle lamellz, intercellular substances, etc., as the photographs. TABLE I. ANGLEs OF INTERSECTION OF THE CELL WALLS IN A Seuecrep: Cotony or P. asperum. Angles. eit Points. ae a. 4, es Wee gsi, we oe 120° 120° 120° Fens Fe aT 122 119 119 tt PO re eT CE 120 120 120 OF ce Saas viata # GE 120 120 120 Dons sheds vue ee 120 120 120 Woah Bx nee a nas 123 120 117 Gee ultec ses ete nes 120 120 120 Cinen sues a a homas 120 120 120, Wo iran vw eae 120 120 120° Gan pe igs 122 120 1x6: Gos Sa e Sei eee I2I 119 120° QW ORES 119 120 1900 Ge So RRs oe 122 118 120 Pigs eae 120 120 120 Totals j.355.2 eee 1689° 1676° 1675° BVO ea ee 120° 119° I19° Higher magnifications making the wall lines bounding the at longer do not essentially help the situation. I have enlarged to diameter of about 8 cm. the photograph of P. asperum (Fig. 6) an carefully and repeatedly measured the angles at each point of con- tact for the whole colony. The results are given in Table I. order in which the angles were read is indicated by the le a, b, c, at the points g' and d. I have not been able in these meas- urements on P. asperum to distinguish between the angles adjacent AND INHERITANCE IN PEDIASTRUM. 399 to the large and small intercellular spaces, that is, the regions of contact of three and two cells respectively about the points d, d', d?, etc., as I did in the case of P. Boryanum. Such differences if they are present in P. asperum are within the limits of error with the methods of measurement I have so far been able to employ. The deviations from 120° range from 117° to 123°. The method is not, however, exact and the successive series of measurements from which the averages in the table are derived have little value further than to show plainly enough that the fluctuations here are in most cases relatively slight, and that they appear to range rather equally about 120° for all the cells in the colony. I have some slight evidence that the inner angle a of the three at the points g', g?, etc., and d, d’, d?, etc., averages a little larger than the other two angles, b and c, but the evidence is by no means ade- quate on this point and I reserve the data till some further method of testing the questions involved has been devised. _ It is obvious enough that the fluctuations in the values of these angles of inter- section are much less in P. asperum than in P. Boryanum and this fact is evidently correlated directly with the formation of the large intercellular spaces in P. asperum, which permit a curving of the lobes and a general distortion of the whole cell body which results in a more perfect equilibrium between the surface forces at the points of contact of the cells. The relative degree of conformity to type in a series of the sixteen-celled colonies of P. asperum can be better studied by com- paring the angles subtended by each cell about the center of the col- ony. These angles can be measured with more accuracy, and, while they show considerable individual variations, the averages of the rather small numbers obtained show considerable constancy. We may take first a series of fourteen colonies whose cells are arranged in the most common order, namely, one in the center with five ‘around this in the so-called second series and ten in the third or outer series. It is plain from the table that the bases of the cells of both the second and third series tend to be equal and regarded as arcs of a tircle about the center of the colony r (Fig. 7) subtend approxi- mately equal angles. The cells of series II. occupy on the average 400 HARPER—ORGANIZATION, REPRODUCTIO! TABLE II. Dimensions oF CeLLs IN A SERIES oF FourTEEN SIXTEEN-CELLED Ca P. asperum As INDICATED BY THEIR Bases MEAsuRED As ARCS OF Circte Apout THE CENTER OF THE COLONY. — Cells 2, 3 4, 5 and 6. Colony. gig? Est. tet. S04... 02: ae 67° 68° bc ABA ss 70 76 68 16055 eo ae 76 73 74 $05 ea ae 75 71 74 IO. Ss 72 72 76 co Gear 71 73 yA Sut. <4, . 72 14 65 G0l. 355556 72 75 70 1704-3 kG 72 65 73 ob BRO e eae 67 75 73 Th Ga 76 13 71 WOES oi 08 clas 72 q1 78 $86, 66 78 74 Dy ee rer 5 74 76 64 Totals ....45. 1003° 1002° 1020° AM ae 71° ¥r° 72° Cells 7, 8, 9, 10, 12, 12, 13, 14, 15 and 16, d-di di-d2, | dds, | d-d4. | dt-ds, | ds-d’. | d-di. | di-di. pe eee 34° |} at® | 37° | 38°.) 3s) oe ae ee ABA. 6s es 34 37 36 38 36 34 36 35 169 : 36 36 36 39 35 36 34 35 10S..+..+5| 35 35 39 37 35 36 34 36 1105-5 ie 36 37 36 37 35 35 38 32 38%. ...5.- 34 36 35 37 37 36 36 37 By Chee 40 36 36 40 35 36 33 35 O0% fc the apical angles. In the cells of series II. (except cell 4) we shall have two basal, two lateral and two apical angles and in cell 4 and in all the cells of series III. three basal angles, for example, in cell 12 p*d*r*, p°d*r°, and ae and in cell 4 the corresponding three angles. In the central cell the unpaired angle igi® on the axis of the colony is the largest of the five in both cases. It is 115° for the average of the series and 111° for the single colony. The apical pair of angles bounding the tips of the lobes 7*gi* 90° and i*g*i> 90° are the smallest of the five. Their values are nearly equal in the averages from the series, as they should be for the bilateral symmetry of the colony. In the single colony, No. 55, they are markedly unequal, 81° and 93°, though their average, 87°, is only three degrees different from the average of the series 90°. The average values of the lateral angles ig*t' 107° and ig*i® 105° are intermediate between those of the basal and apical angles of the cell in the averages of the series as they also are in the single colony. In the latter again they differ by 11° though their average, 108°, differs by only 2° from that of the series, 106°. On the whole, the differences between the typical shape of the central cell as de- termined from the selected individual and as determined by the average of a selected series are within the range of errors in meas- urement by the methods used for angular dimensions in photographs of such objects. 408 HARPER—ORGANIZATION, REPRODUCTION 7 TABLE V. AVERAGE VALUES oF Eacu oF THE INCLUDED ANGLES OF THE Coss AGES OF THE Ricut AND Lerr Parrep ANGLES IN A SERIES OF CELLED CoLonies oF P. Boryanum CoMPARED WITH THE Col _ ANGLES IN THE SELecTED Type DrAcram. (’16 pp. 98, 99. ORME 1. Includes Angles of the Central Cell. : Average. OE os os aa IIs 115° i GOO cts ciueeg)) 20T t OO 6 ia + cca 106° 3 PR ee OOF RS cians wees 90° 90° 2. Basal Angles of Cells 2-6. GOR fcces Cases) «390".. (PO. a eee 121° SO SS SP ree SA SS eae eee 126° PO opens 130° | PES eta a eee 126° § CPM | 8 (OR Ca ee 134° POs ieee sap Sgt c 1 I. teat Se aan 134° “Totals. ie Hy SE 6 7 ly ad 639° 641° BW et ls 127° 128° 3. Lateral Angles of Cells 2-6. R05. Fees ee ok EOS ERO ees chee Meee 110° eh erect EEA FAS. so Or 110° Wey Ee ta ci 5 be Oe RAMOS oo vows 115° 112° Fe et gan Pare BEND PR. Co. ee Oe tx2” MAG sess xs 113 | a ee re 116° 114° Totelass ences 587° AV Sete sere 111° 4. Apical Angles of Cells 2-6. ODP. csv ccccecs}. SEG Fates isan] 208.5 tra> ODD 6 oo s essed eae OOP iss sins TI0.5 III.2 OPP. 5 oe eel ee o'd"p",......%.| 108.7 105.8 o8d3p3, ......---| 100.0 Ofdtpt.... 2.2...) 106.0 103 ofd4p4. .........| 105.0 POP, ive s'ce'se| IIT.O 108 Totals ....vc « sce ot cee 544.7° 540.0° 5 Sees Mra Mee ce fe 108.9° | 108° 5. Mid-basal Angles of Cells 7, 9, 11, 13 and 15.: WOON estes tse 0 136° ; ct | Ps ge at Oe OO: 136 POO cee ene 138 137 POP CE os 120 O80 cee rane 137° 133 AND INHERITANCE IN PEDIASTRUM. 409 TABLE V.—Continued. 6. Basal Angles of Cells 7-16. 4 OL eee eee ) eee ey go 122° a] en ease | 122 Fong sl 122 pease t 128 5 1 al eager S 123 a Rae | 128 Ds a a } 125 a eas | 128 Pe a ae 123 walls... -....: | 626° 615° MM aha Lis crn o i 125° 123° EE ees ; PE OM alae oy ) 128 RN ae | 125 a i : 127 Me oS: fo Sgag = hee eo. k 127 ee i: 331 ra gc I Oe ; 128 Ee P2326 a poof te fe a aya pe 126 tele... | 634° | 636° Mg ie als ys 3 2) 820" 737° The basal angles of the cells of series II. igk, i®*gk, ig*k*, 1°g*k*, etc., also constitute five pairs of correspondingly placed angles right and left of the axis of the colony mn. The average values of each for the series of seven colonies are arranged in corresponding pairs in the table, section 2. In this series the average values of the right and left pairs are seen to be progressively smaller as we pass in the direction from pole m towards pole m in the colony. * The values range from 121° in the first pair about the point g to 134° in the pairs about g? and g*. The equivalence of the adjacent basal angles of cells 2-3, 126°, and 3-4, 134°, is to be expected as a characteristic of the type configuration though it is more or less accidental doubtless that it should appear on the basis of so few measurements. The pro- . gressive increase in size of the basal angles of these cells is correlated with the reduction in value of the corresponding included angles of cell 1 proceeding from basal to lateral and apical angles. As shown in the table, column 4, the values of the corresponding angles in the selected type diagram from colony 55 agree fairly well with these averages from the seven colonies. The first pair are smaller in the average from the series. The next three pairs are the same, and the last pair, the basal angles of cell 4, are smaller in the average type than in the selected type. . 410 HARPER—ORGANIZATION, REPRODUCTION We have two sets of points of intersection between the w the cells of series II. and series III., due to the fact that there a ten cells in series III. to five cells in series II. We may con: first the included angles of cells 2-6 about the five points of inter section marked ee'e*, etc. The values of these angles for the s colonies and their averages are given in Table IV. Point e is the axis. The remaining four points, e'e*, are placed symmetrical right and left of the axis mm as is indicated by their position in t table. The average values of the lateral angles of cells 2, 3, and 6 about these points e, e*, etc., are arranged in five corresp right and left pairs in Table V., section 3. es, " There is no adequate evidence of any regularly progressive a change in value in the averages of these five pairs as compared the progressive change in value of the basal angles of these This is doubtless owing to the increased distance of these < from the five-sided central cell with its unequal included gas to its inherited form tendencies. There is a manifest tendency the angles of intersection of the cell walls to become equal in 2 mony with the general symmetry of a least-surface The values of these angles from the averages of the series a fairly well with the values of the corresponding angles in the sele type figure. In the case of the lateral angles of cell 4, an arbitrary value, 120°, was assigned in the diagram. The actual average value of these angles in colony 55 was 112°, as shown in the table (’16, p. 98), and this is only two degrees from the average value of = angles in the series of seven colonies. The values of the right and left pairs of the apical nate of - cells 2-6 are given in Table V., section 4. There is considerable fluctuation in the values of these angles but the average of the series, 108°, agrees closely with the value obtained for the s: angles in the selected colony 55. In this case again an arbitrary value (100°) was given to the apical angles of cell 4 in the type © diagram and the average for the series agrees with the measure- ments from the selected colony, 109°, and not with the arbitrarily assigned value. i There is no clear evidence from the series that there is any pro- gressive change in value of these apical angles odp, o'd’p* of the AND INHERITANCE IN PEDIASTRUM. 411 _ cells of series II. as we pass from pole n to pole m of the colony. _ The average values of these angles, comparing all seven colonies, _ have an extreme range of 9° from 103° to 112°, with, as noted, an average of the averages for all the colonies of 108°. In the single _ colony, No. 55, the range of variation was 17° from 97° to 114°. _ Colony No. 87 showed a range of variation of 30° from go° to 120° in the values of the angles odp, o*d'p’, etc. The average values of the five mid-basal angles of cells 7, 9, 11, 13 and I5 are given in Table V., section 5. We have here a case in which the average values of the angles in the series of seven colonies is the same as the value (133°) arbitrarily assigned in the selected type diagram for the mid-basal angles of cells 11 and 13, and differs by seven degrees from the average value as measured (140°) of these two angles in colony 55, from which the selected type diagram was derived. This is the reverse of the result as we have found it in the other two cases noted above and a fourth case noted below in which an arbitrary value was given to angles in the selected type diagram. In the three other cases the average value of the angles in colony 55 as measured is nearer to the average value of the corresponding angles in the series of seven colonies than to the arbitrarily assigned value. As I have already pointed out (716), the angles about the point e* in colony 55 are obviously to the eye the most unsymmetrical and aberrant in the whole colony and the possibility of making consistent arbitrary corrections for these asymmetries and their correlated effects on the other angles in the region is not very great. The values of the unpaired angle oeo® and the pair at e* and e* agree closely with those obtained from the selected colony No. 55 and given in the selected type diagram. The values of these mid-basal angles of the peripheral cells are larger than those of the other angles of the groups about the points of intersection e, e*, e?, etc., and this we may regard as a direct cor- relative of the fact that the tangential diameters of the cells of series II. and III. are regularly greater than their radial diameters. lf this tangential elongation of the cells were directly determined by the fact that there are but five cells in series II. and ten cells in series III., instead of six and twelve, the normal numbers for a least-surface configuration in which all three angles would be 120° 412 HARPER—ORGANIZATION, REPRODUCTION _ each, we might be able to give a fixed value to this relation k the inequality of the diameters and the inequality of t about e and to get some notion of the viscosity of, the p: the adhesion of the cells to each other, etc. As noted, howe is obvious that the inequality of the diameter may exist degree at least in approximately free cells, as is shown ¢ 29-33: o In the basal included angles of cells 7, 8, 9, 10, etc., of s we have two series consisting of five pairs each : lee equal. The average for the different sets is: For the Series. Type Diagram. 125° 127° 123° 123° 126° 130° 127° 131° Toh... s., Se 128° PGS s alvees 125° 635° 127° In the selected type diagram derived from the selected col No. 55, the average for these angles is 127°. The colony N ) is, however, especially irregular in the region of cell four, and ar arbitrary value, 131°, was assigned to the two angles otd*r* a o°d®’y®. If the measured value for these two angles, 123°, is : instead of the arbitrary value, the average of these angles for the single selected colony, No. 55, becomes 126°, only 1° different from that for the series. On the whole, the values for the included angles of the cells agree fairly well when derived by averaging the correspon AND INHERITANCE IN PEDIASTRUM. 413 pairs in a series of colonies of similar cell arrangement with those _ derived by averaging the values of the corresponding pairs of angles _ ina colony selected for its obvious symmetry. ___ I have emphasized the difficulty of measuring the angles in pho- _ tographs of such small objects with the lack of sharpness in the cell walls, especially at points of intersections, etc. Only approxima- tions can be achieved with the technique I have used. The fact _that the colonies are not absolutely flat figures so that the plane of the photograph cuts the colony at different levels and the further fact that the walls are not absolutely vertical to the plane of the colony present difficulties inherent in the nature of the material. The average values of the angles measured from the center of the colony which the corresponding cells in the series of seven colo- nies subtend are given in Table VI. The exact determination of the center of the colony is not easy in view of the irregularities of its boundaries. The point taken as the center in each colony was midway between the point of intersection of the major axes of the colony and the point of intersection of the major axes of the central cell. The values obtained from averaging the series like those for the single selected colony No. 55 and those of the corresponding angles in P. asperum (Table II.) are fairly close to 36°, the angle for a strictly surface tension configuration. There can be little doubt that these angles tend to be equal as would be expected were surface tension and adhesion alone operative. TABLE VI. Dimensions oF CELLS 7-16 IN A SERIES oF SEVEN CoLontes oF P. Boryanum As INDICATED BY THEIR BASES MEASURED AS ARCS OF A CIRCLE ABOUT THE CENTER OF THE COLONY. Colony. 7: 8. 9- Io. Ir. 12. 13. 14. 15. 16. .« So eae 38° 30° 40° 35° a7" 30° 38° 36° 35° 41° I3ZI.....-. 37 35 35 36 36 36 36 36 37 36 | 34 37 37 ee — —_ — or 38 34 = phe Ber 36 38 C2 36 35 28 40 33 Bee 35 36 34 36 36 38 | 36 36 35 38 i | Totals. ...| 217° | 210° | 220° | 208° | 261° 250° / 259° | 238° 256° | 258° ; ’ ; eee... mer) as? | aa ee ae" Fak | an” | 34°] 36° |. 208 414 HARPER—ORGAN IZATION, REPRODUCTION How much significance can be attached to the degree of < ment obtained by the two methods used in obtaining the norms all these fluctuating elements can only be determined by more tensive statistical studies both of especially symmetrical single nies and extensive series of colonies chosen only on the basis their having a similar arrangement of their cells. The tendency equality shown in the values of the corresponding angles both the intersection of the walls and for the spaces occupied by the entire cells certainly suggests that the structure of the colony t : to be a least-surface configuration with all angles of intersection the cell walls equal to 120° and all the cells subtending equal ang about the center of the colony, this tendency being in every c limited, however, by the inherited form of the cells and the a dents of environment. - = The results of my measurements of the homologous shell: the colonies of Pediastrum are in agreement with Rhumbler’s (’o2) results obtained by measuring the homologous marginal angles of various Foraminifera and confirm still further the conception the semi-liquid nature of plant and animal protoplasm. Rhumbler uses these results primarily as evidence on this point. The fact that only homologous angles tend to be equal leads him at once, however, to emphasize the heterogeneity in structure of the proto- plasm resulting in an anomogenous consistency and anomogenous tensions in different regions of the cell. The cell of Pediastrum with its inherited four-lobed form operating always with surf tension in determining the value of any given cell angle is also notably anomogenous system, as compared with a simple fluid di let. In the morphogenesis of the colony it is obvious that th anomogeneity is quite as important a factor as is the principle surface tension. It is in the indisputable evidence from bot Rhumbler’s material and my own of the interplay of capillarity protoplasmic anomogeneity, and especially of the principle of binary fission that we get a basis for interpreting the complexity of th ; form elements with which we are confronted even in such simple organisms as the Foraminifera and ccenobic alge. The least-sui face configuration comes to expression in Pediastrum in so far 2 is consistent with the inherited cell form and consistency and wi é 5 or ‘. nT Pek ie: ‘ Ps Vid, res oes AND INHERITANCE IN PEDIASTRUM. 415 cell numbers produced by bipartition. The endless variations in form found on comparing a series of colonies are the expression of the unstable equilibrium, arising especially from the simultaneous operation of the law of bipartition and the physical principle of least surfaces. In the series of colonies chosen I have included as noted only individuals with the common plan of 1-+5-+ 10, since it is im- possible to compare the angles fairly in individuals with different geometric plans. This is obvious in the case of the superficially Fic. 25. Pediastrum Boryanum. Colony with cell number 6 so displaced that the side which should be adjacent to cell 5 is in contact with cell 1, the side that should be in contact with cell 1 is in contact with cell 2, etc. The remainder of the cells are normally placed. > about 600. rather regular colony numbered 67 (Fig. 25). In this colony all the cells are regularly and symmetrically placed except No. 6, in series Il. This cell has swung around until its major axes, both radial and tangential, are displaced about 45°. Its reéntering angle nor- mally under the middle of cell 16 is now under cell 15. The en- tire cell has been, broadly speaking, rotated through one sixth of its circumference so that each of its sides has been displaced by one in its relations with the sides of the neighboring cells. The side normally adjacent to cell 5 is now adjacent to cell 1. The side normally adjacent to cell 1 is now adjacent to cell 2 and much reduced in length. The side normally adjacent to cell 2 is now adjacent to cell 7, etc. Cells 15 and 7 have slipped in toward the center of the colony so that its peripheral outline has been flattened 416 HARPER—ORGANIZATION, REPRODUCTION at these points. The central cell is more e nearly i i c In such a colony it is plain that the included angles of cell ; those of all its immediate neighbors at least, are under quite ferent pressure relations than they would be with the normal arrangement. Their fluctuations will be of a different order than they were normally placed. The writhing, struggling motions c the final stage of swarming as described above could only sulted in the normal relations of equilibrium in case they had sufficient violence to bring the cell out of its present abn 0 tions into the normal so that the properly matched sides it neighbors would be in contact. The cell has reached 4 condition of equilibrium in its relations with the adjacent cells, but this equilibriun achieved anything like equivalence in value or position f responding sides and angles. The palpable asymmetry 6 and its neighbors is most convincing proof that the form « cell is not influenced merely by its pressure relations in the but also by its inherited tendency to assume the characteristic lobed outline. The displacement of cell 6 is brought out clearly by comparing the right and left pair of cells 3 and 5 the right and left pair 2 and 6. In spite of fluctuating vau cell 3 could be superimposed on cell 5 by merely rotating it axis of the colony mn through an angle of 180°. Whereas to s impose cell 2 on cell 6 there would be necessary a further re of No. 6 about its center and in the plane of the colony thro about 45° so as to bring the corresponding sides and angles of two cells together. ay Colony 67 illustrates asymmetry originating in the displa : of a single cell to the extent of bringing unmatched sides togethe This we may call anomogenous asymmetry as contrasted with asymmetries involving merely fluctuation in the values of sides angles without the passing of a critical point which alters the fu mental arrangement of the cells by bringing about abnormal jw positions of sides and angles. All grades and degrees of this anomogenous displacemeli cells can be found in nature. Two general types of irregularity be distinguished. The first is an irregularity which does not AND INHERITANCE IN PEDIASTRUM. 417 marily affect the outline of the colony or the arrangement of the cells in concentric series about a center, the displacement ‘being largely in the position of the major axes of the individual cells. This is illustrated as noted in colony 67 (Fig. 25). The second results in an abnormal form for the colony as a whole and the loss of the concentric arrangement of the cells (Figs. 26 and 28). The two types are more or less combined, of course, in the majority of cases. Fic. 26. Pediastrum asperum. Colony irregular, crescent-shaped form. X about 425. : Fic. 27. P.asperum. Colony fairly regular in outline but interior cells __- very irregularly placed. about 275. a Fic. 28. P. asperum. Irregular colony, somewhat triangular in outline. X about 425. The most extreme case of the first type which I have observed is seen in the thirty-two-celled colony (Fig. 27). Here almost all the interior cells are anomogenous in their position and interrela- tions with their neighbor cells, though there is the common general arrangement for a thirty-two-celled colony; one in the center, six cells in series II., ten cells in series III., and fifteen cells in series IV. All the cells of the outer series are normally placed with long spines outward and the general outline of the colony ig circular, but the interior cells are in absolute confusion, as compared with the typical arrangement. Only three cells of series III. have their long lobes radially outward and their long axes tangential and these three are not in normal contact relations with the cells of series II. This colony emphasizes the existence of the stage described under the head of reproduction, where it was noted that the cells of the pe- ripheral series seem to get their definitive positions first while the 418 HARPER—ORGANIZATION, REPRODUCTION interior cells are still actively swarming. The appearance here if the development of the colony had proceeded normally until peripheral cells had gotten their places and then for some re activity was checked before the interrelations of the cells in interior had been worked out. It would be possible to express this confusion in degrees of displacement of the major axis of ea cell, but the situation is sufficiently clear from a comparison of th colony with that shown in Fig. 4 without giving it such a ‘mathe- matical expression. I shall include data on this point in a study from a wider comparative viewpoint involving the effect of increasing numbers of cells on the orgnnlee colonies. a An extreme of irregularity of the second type while is ex pressed in the general contour of the colony is shown, above, in Fig. 26. Here the colony is crescentic in outlin type arrangement of cells for a sixteen-celled colony has disappeared and we can no longer recognize a central cell with concentric series about it and yet the inherited form of the cells quite perfectly developed in all cases. The modifications are as are obviously due to the special pressure relations under w each cell finds itself. It is possible, of course, in many case identify all the angles of any particular cell and to combine ther in homologous groups for the colony and a whole series of colonies as I have done for the series of more regular individuals. — ‘But an, these irregular colonies the major axes of the colony are f quite unrecognizable and any particular included angle of a cell will be so obviously misplaced and with such unusual pressure relat 7 ) that the comparison of its values in different colonies decors very complex problem. I have brought together in Table VII. the values of all ee cluded angles in a series of both regular and irregular sixteen-celled specimens of P. Boryanum. The values are given for groups di fering by three degrees. The data could, of course, be rep graphically in a curve, but the main point illustrated is brought quite well from a glance at the figures. They form a series culmi- nating in 120° and ranging somewhat similarly above and below this number. The total number of angles with greater value than AND INHERITANCE IN PEDIASTRUM. 419 TABLE VII. ; FREQUENCIES OF THE ANGLES OF VALUES BETWEEN THE EXTREME 86° AND 176° In A SERIES oF THIRTEEN Cotontes oF P. Boryanum TAKEN BY CHANCE. THE VALUES ARE GROUPED BY THREES. 86° | 8°} 92° | 95° | 98° ' ror? 104° | 107° 110° 113° | 116° | 119° to to to to to | to | to to | to to to to 88°. | gx°. | 94°. | 97°. 100°.) 103°. 106°,| 109°.) r12°.| 135°. 118°.) 121°. No. of angles......... ae ee a|2|5|3|% 26 | 33 | 42 44 | 121 : i | | 322°} 125° | 128° | x3x° | 134° | 137° | 140° | 143° , 146° i to to I to | 150°.) 155°.) 176°. 124°, =e 130°.| 133°.| 136°.) 139°.| 142°.| 145°./ 148°.) No. of angles......... 98 | 6r | 39 | 15 6lal4iziad: nah 120° is 252. The total number which have a value less than 120° is 189. The number recorded for 120° is perhaps unduly large. _ Since, as I have noted, it is quite impossible to judge within one or two degrees with any certainty there is perhaps a tendency to assign too frequently an angle of 120° rather than 119° or 121°. On the other hand, it is not impossible that the angle 120° being the point of equilibrium in the surface tension group, there may be some especial fixity in the configuration when it is once achieved, so that when it arises by chance as a result of the more protracted slow _ writhing movements of the final stage in forming the colony, it may be maintained in a certain number of points of intersection at the expense even of greater inequalities at other adjacent points. When once attained exactly it may be more persistent than any other angu- lar value. But whether or not the number of angles with a value of 120° is correctly determined there is no question as to the general tendency in these angles to fluctuate about 120° as a center or modal _ point. On the other hand, it would be quite inadmissible in the light of the results obtained by comparing the corresponding angles of a series of colonies with similar arrangement (Table V.) or a single colony selected for its symmetry (16, p. 98) to conclude that the type colony of P. Boryanum should have all its included angles equal to 120°. With sufficiently large numbers of cases the sec- ondary modal points representing the special values of the included angles of the central cell, etc., should emerge. The average values for the corresponding sides of the cells in 420 HARPER -ORGANIRSTION, Seen eatame TABLE VIII. LENGTH IN Mac, ov 3mm Ceuz Bounnantes 2 a SmUBS Cotontes or P, Boryanum. Sides of Central Cells. Colony. i. A, : 73, fhe iy Ree Maye a SE ee 8 1 4.5 5 bee a P aas ees 9 7 4:5 5.5 Pes CoN wi GA ee Ir 9 7 6.5 SO ids pe 7.5 9 4.5 45 B2i saek cameo’ yaa 8.5 9.5 4.5 «4S Beak ced oe eae 9.5 9.25 6. 5 TOMO s Su idiicees cis 53.0 50.7 31.0 31 OT ree en eee 8.8 8.4 5.1 5.1 0. Radial Walls of the Cells of Series I]. B® A a. . 1 EGRESS GC aU a et ie al vi3 7 7 PD r ah bw ck Croley Wee ew ae Sa 7 7 7 TOs HR eae ae eosin idles aes 10 8 9.5 BYR ype eran rc 7 SY A RA 9.5 7.5 8 S55 eS Poe Wale ke bee oe ae Si: 6.5 5 Be ga ek vied oe aaeeca Daas Soe an 8 9.0 6.5 TOMB i vescia ads coe eres 49.0 45.0 43.0 Bee aie io aes eae has 8.1+ 5+! 7+! Basal Walls of Cells 7, 9, 11, 13 and 15. 0 ol, 0, | Pee a o | of 88. cao 7 6.5 6.5 5.5 6.5 6 7 rk § RR a Ia 5.5 5 6 5 4.5 5.5 ye Bee es 7.5 9 7 7 7 6 7 I ae 5.5 6 6 6 5-5 5.5 : Be C7 pe 5 ays 6 6.5 5.5 5.5 5 Be icsiee 7.5 6 6.5 7 7 4 8 Totals... | 37.0 | 39.5 | 37.0 | 38.0 | 36.5 | 31.5 39.5 Aver, | 62) 68) 62 ead 6 5.2 | 6.5 Radial Walls of Series III. r. A, Pe ag oh ee ee ee ed Bae oy nC Ee 6.5 7 7 7 6.5 7 7.5 CIR se hes 7.5 75 a5 8 7 8 7 PN 8 8 8.5 9 8.5 8 S00 sinc 7 8 bs 8 9 9 8 eT eh cs 8.5 7 7 7.5 8 8.5 7 55--+++++| 95 | TO 9.5 8 8.5 9 8.5 Totals. ...| 47.0 | 47-5 | 46.5 | 47-5 | 47-5 | 49.5 | 38.0 Bh Sic 7.8 | 7.9 7.9 7.9 7.9 8.2 7.6 AND INHERITANCE IN PEDIASTRUM. 421 the series of seven colonies of P. Boryanum are given in Table VIII., and agree well with those obtained for the selected colony (16, p. 98). I have not included measurements of the basal walls of cells 8, 10, 12, 14 and 16, as their tendency to equality is suffi- ciently obvious. The corresponding radial walls between the cells of series II. tend to be equal as do also the radial walls between the cells of series III. The indication from the measurements of the selected colony that the walls i and 7° of the central cell should be regarded as typically a little longer than the walls 7* and # is not confirmed by the measurement of the series, though there is a difference of .2 mm. in the average. Measurements of the dimensions of the central cell in a larger series of colonies would be of interest. Its inherited oblong and two-lobed form, and the pentagonal outline in the colony formed by the bases of the five cells of series II. afford an interesting case of disharmony in morphogenetic factors. INHERITANCE OF CELL Form. I have pointed out that the four-lobed form of the cells appears’ immediately in the young colonies and have referred to it through- out as inherited rather than as the direct and epigenetic expression of the pressure and other interrelations of the cells in the growing colony. I have pointed out the adaptation of this four-lobed cell form to the exigencies of group formation, when the number of units is strictly limited by the principle of bipartition. The prin- ciple of least surfaces here requires that five cells instead of six are to be placed about one in the center and ten cells in the third series. This arrangement involves just such a tangential elongation of the individual cells as we find has actually occurred and favors the maximum of compactness in the arrangement 1-++5-+10. I have suggested the possibility that the environmental complex may have led in successive generations to the development of this four-lobed form and its fixation as an hereditary character of the cells. Evi- _ dence that the typical form of the cells can be achieved independ- ently of their being in normal contact and pressure reactions in the colony is rather easy to obtain. Many colonies are found in nature PROC. AMER. PHIL. SOC., VOL. LVII, CC, AUG. 20, 1918. 422 HARPER—ORGANIZATION, REPRODUCTION in which by some accident certain cells are only attaches at point to the remainder of the group. : In Fig. 29 we have an eight-celled colony of P. ashe which one cell, a, is attached by only one of its basal lobes” yet is quite symmetrically developed. The second base lobe rounded rather than sharply wedge-shaped and this difference may be taken as the measure of the influence of the epigen pressure and contact relations as compard with heredity in termining the form of the cell. The asymmetric positions of two central cells in this colony also have produced character effects on their form and it is plain that the relatively free cell much more nearly achieved its full development than have 1 Fics. 29 AND 30. P. asperum. Irregular, eight-celled colonies, one in each case attached by one lobe only, but quite typical in outline. 550. Fhe cells with their asymmetric contact relations. In Fig. 30 we have a colony with only seven cells visible, one cell, a, attached by on y one basal lobe, another by a basal and a peripheral lobe, b. It is plain here that in neither case has the free basal lobe tended in degree to assume the more tapering form of a peripheral lobe, 1 have the long and short axes of the cells been reversed. In this colony the eighth cell may have been present in its early life and possibly may have been connected with some of the cells now partly free. The form of cell b, however, has certainly been achieved under the same contact conditions in which it appears in the figure. The rounded end of its free basal lobe as well as that of cell a shows that the wedge form is an environmental effect. . Figs. 31 and 32 (less highly magnified) also show cells which have attained the normal form while attached by only one basal AND INHERITANCE IN PEDIASTRUM. 423 lobe. These are probably thirty-two-celled colonies, though they are too irregular to permit accurate counting. The colony shown in Fig. 31 is quite young, while that shown in Fig. 32 is well on toward maturity. The cells attached by only one lobe have gone through their whole development in apparently normal fashion. FE nts —— arse ep MTOR ON 34) Fics. 31 AND 32. Pediastrum asperum. Young colony showing as in Figs. 29 and 30 cells which are without their normal contact and pressure relations in the colony and have still developed the characteristic form for the species. Fig. 31 X about 500; Fig. 32 X about 175. Fic. 33. P.asperum. Like the last two figures, but with a cell attached in reversed position by one of its peripheral lobes and still showing the typical form. X about 225. Fic. 34. P. asperum. Colony with one of its peripheral cells reversed but showing one typical long spine directed toward the center of the colony. The other is blunted by contact with an adjacent cell. > about 425. Most interesting is the case shown in Fig. 29, where we have a cell attached only by one of its peripheral lobes and to a peripheral lobe of a peripheral cell of the colony. Here the basal lobes are both peripherally placed and yet have retained their blunter form. 424 HARPER—ORGANIZATION, REPRODUCTION The peripheral lobes are basally placed and one of them functions for attachment and yet they retain their more tapering form. It is over and over illustrated in abnormal colonies that the polarity shown in the difference between the basal and peripheral lobes is a matter of cell organization and not of colony organization. A very characteristic case is shown in Fig. 34. Here in an other- wise quite regular colony one of the ten peripheral cells has been reversed in position and thrust partly back into the second series. The one free peripheral lobe is quite normally developed though pointing toward the central cell of the colony. The diagonally oppo- site basal lobe which is free to grow radially outward has shown no tendency to do so. The second true peripheral lobe has had its natural growth tendencies quite inhibited by the limitations of space in which it finds itself. The tendency to functional hypertrophy if operative here is not equal to the production of a normal peripheral lobe under such conditions. The whole grouping in such a case gives a very clear picture of the exact part played by heredity and environment respectively in morphogenetic processes. Such examples as are shown in Figs. 25-30 can be multiplied almost without limit and it is clear that however much the contact and pressure relations in the group may have influenced the evolu- — tionary processes by which such oval cells as those of Gonium be- E come the four-lobed oblong cells of Pediastrum, at present these — cells are able to attain their characteristic forms, diagnostic for the species, when almost entirely free from their normal environmental relations with the other cells of the colony. j Nitardy figures several marked cases (’14, Taf. VL., p. 2) in, which the single spinous outgrowth and general triangular form of the cells of P. simplex are shown to be an hereditary growth-hab of the cells rather than a response to their pressure relations 0 orientation in the colony. In the figure referred to a peripheral cell is shown with its poles reversed and the spine projecting toward the center of the colony and an intercellular space quite as in a Fig. 30 described above. 2% It would be natural, perhaps, to expect that ae four-lobed form should be strictly epigenetic and achieved anew by each generati AND INHERITANCE IN PEDIASTRUM. 425 under the influence of the stresses and pressures developed in the _ growing colony. We have noted, however, that the cell form char- acteristic for the species is visible at once as the swarmspores cease moving and is only sharpened and made more definite with the further growth of the cells. On the other hand, it is equally obvious Fic. 35. Type diagram from a selected individual, colony No. 55. Re- Produced from 16, Fig. 1b. that abnormalities of form—shortening, elongation, change of direc- tion of the lobes, etc—as well as the regular blunting of the peripheral lobes of interiorly placed cells are all direct environ- ' mental epigenetic results dependent on the interrelations of con- tact and pressure between the cells in the forming and growing stages of the colony. 426 HARPER—ORGANIZATION, REPRODUCTION GENERAL DISCUSSION. Cell Form.—There are plainly two sets of form-determfinnig! in- fluences operating on the cells of Pediastrum. First, the cell heredity which, given free play, as in the case of cells largely out ; contact with other cells of the colony, develops what we may call: the typical cell form. And, second, the environmental pressure and contact relations which exist between the cells as ordinarily placed in the colony. This may result in the extreme difference which we_ find between the internal and peripheral cells of P. Boryanum or the very slight differences which we find between these same cells in P, clathratum (Fig. 3). The familiar antithesis between heredity and environment in the determination of adult form and sivectare | is fully in evidence in Pediastrum, but under such relatively simple conditions as to permit of an attempt at analysis. It is os here, as has been held in general by students of heredity, that inheritance through cell division may perpetuate the type form structure, while many at least of the fluctuating variations in 1 type are directly traceable to environmental conditions of interac- tion between the cells and favorable or unfavorable outside anaes B tions at the time the colony is formed and during its growth. . The evidence is clear, as I have shown from the cases of acci- ; dentally misplaced cells, that all the various cell forms found in the genus are transmitted from one colony to the next by inherita in some fashion of other. I shall discuss the method of transmis- sion below. ‘The cell form also obviously determines the character of the colony as a whole. The form and character of the colon . as a whole may be said also in turn to influence the form of the cells, " but the modifications so produced are of the nature of environmental limitations on the complete development of the cells, as, for example, the shortening of the lobes on the interior cells as compared with ° marginal cells of the colony. The position of the cell in the colo influences its form only in minor, though perfectly obvious 2 definite degrees, but the structural and organic characters of colonies which are the basis for their classification into subgen and species are the direct expression of the inherited characters the cells. AND INHERITANCE IN PEDIASTRUM. 427 The prevailing oblong, four-lobed form of the Pediastrum cell is probably adaptive for the general metabolism of the cells and is also the form which permits the closest possible approximation to a least-surface configuration in a colony composed of units arising by binary fission. It is the form of cell which would be expected to arise under the pressure relations existing in such plates of cells held together by adhesion and yet as noted it can and does arise in cells almost entirely free from such contact and pressure relations. We have here evidence that a cell form which may well have arisen first simply as a response to environmental stimuli has become fixed in heredity until now the series of growth processes by which it develops can go on quite independently of the stimulative conditions which originally called them forth. In Pediastrum we do not have the extreme differentiation of germ plasm and soma which under the conditions in the higher metaphytes makes such direct interrela- tions of environment and heredity so difficult to conceive. It is sufficiently obvious that the oblong four-lobed form is directly transmitted through vegetative reproduction by cell division and there is no reason to question that the same is true in sexual reproduction by the fusion of gametes. In asexual reproduction the mother cell divides by successive bipartitions to produce a swarm of oval ciliated swarmspores which at first show no trace of their adult form. I have noted, however, how promptly, almost instan- taneously, the four-lobed form appears as the swarmers come to rest in the contact and pressure relations of the colony and with the very first growth expansion, so that almost as soon as it is formed the young colony has all the essential structural characteristics of the adult. ; We may consider briefly at this point the difficult question as to the method of representation and transmission in heredity of the characters of differentiated tissue cells and the characters of tissues, organs and entire organisms considered as wholes. Inheritance of Cell Form.—tThe inheritance of cell form cannot be said to be direct in the sense’ that the inheritance of green color may be direct. The division of a green cell giv:s at once two green cells. Greenness is inherited.as such by division of the chloroplast 428 HARPER—ORGANIZATION, REPRODUCTION and is a metidentical character in Detto’s sense. The cole) as eck may be thought of, at least, as present throughout the whole process, of producing the daughter cells from the mother cell. The cylin- drical form of a cell of Spirogyra may also be inherited directly as such in this same fashion by the division of a Ram tac: pag colt into two shorter cylindrical daughter cells. In the case of the cell of Pediastrum the lobed or spinous fond disappears in the successive bipartitions of the mother cell and we have an oval ciliated swarmspore essentially similar in form to those of other more or less distantly related green alge. Repro- duction by cell division here has involved the return to what is generally assumed to be a more primitive type of cell both as to its” form and its motility. The adult form typical of the species only reappears as a result of ontogenetic development by which the primitive cell form becomes differentiated into the more specialized adult. In all filamentous and ccenobic alge which reproduce by swarmspores we have this advance beyond the conditions in Spiro- gyra, and related types in which the germ cell differs only in size from the adult. The reproductive cycle in Pediastrum, for example, parallels that of the higher plants in its essential stages. A mother cell forms undifferentiated germ cells which become specialized dur- ing ontogeny in their form and structure and by their combination produce the many-celled colony which shows also a more or less highly specialized and adaptive organization. In the higher meta phytes it is not so directly obvious as in Pediastrum that the form characteristics of the many-celled plant body are the direct expres- sion of the form, polarities, adhesiveness, and other characteristi of the individual cells. The inheritance of cell form and of form of the colony are indirect as compared with cell color. To be sure, in the latter case swarmspores may be relatively or entirely free from the green color which then reappears in ontogeny but the transfer of the capacity to form green pigment is assumed to involve the division of plastids which thus carry on the pigment-forming bases, chromogens, just as the nucleus, chromosomes, etc., are per- petuated directly by division. In the vegetative reproduction « these simple alge we do not need to say that the capacity to fo: chlorophyll is represented by an hereditary factor in the germ plasm, AND INHERITANCE IN PEDIASTRUM. 429 ’ for we have the visible organ or plastid of the cytoplasm to provide for such transfer. The evidence as to the behavior of the plastids in zygospore formation in Spirogyra indicates that the same is true in sexual reproduction by cell fusion. In the case of the lobed cell form, however, every visible trace of the adult character as such __ seems to be lacking in the germ cell, and it seems natural to postu- late a gene or factor which without being the character itself may as a granule or in some other form represent the adult form when it his disappeared. As a matter of fact, however, we have no plastid for form determination and to assume a granule like a plastid in the chromosome which transmits the determiners of the adult cell form meets with obvious difficulties in this case. The sudden appearance of the lobed, spinous cell form in reproduction as I have described it does not suggest the working out of influences emanating from ele- ments in the chromosomal organization of the nucleus, but rather the direct expression of the organization of the cell as a whole when it begins to grow. This organization shows the most direct rela- tions of adaptation to and interdependence with the pressures and _ contacts established in such a group of cells and may well have been achieved as a response to such environmental surroundings, but it is quite independent of them and-comes to full expression, as noted, in cells which are for accidental reasons quite free from the other members of the colony. It seems to me to be most obviously the expression of the anomogenous organization of the protoplasmic mass involving localized growing points on its surface, specific polar differentiations, etc. This general organization of the cell may well be transmitted indirectly through cell division involving as such transmission would only a sort of regeneration by each daughter cell of the general symmetry relations between the parts of the mother cell. We do not need, then, as it seems to me, to imagine any spatially differentiated organization of a special germ plasm to account for the inheritance of cell form in Pediastrum. It would be possible, but probably premature, to attempt to express in diagrammatic form for Pediastrum the organization of the cell as a whole which is im- plied in its behavior in forming the colonies. Much further cyto- logical work such as has been done by Smith on nuclear and cell di- 430 HARPER—ORGANIZATION, REPRODUCTION vision, the centrosome, blepharoplast, plastids, pyrenoids, etc., Tetradesmus (’13), Scenedesmus (’14), Characium (’16), Pedi- astrum (’16), etc., is needed before we shall be able to correlate the evidence for cell polarities, adhesions, growing points, surface tension, etc., with the data from the chemical study of colloids. B it is obvious that it is only on the basis of such studies that we « hope to lay the foundations for a proper theory of the heredit transmission of characters and the morphogenetic processes by: wl a mother cell is transformed into a mass of tees swarm- spores and these in turn into the adult colony of Pediastrum. —_— Inheritance of the Characters of the Colony—The cbarinees about 550. Fics. 12 AND 13. Cleavage complete and the daughter cells more or rounded up. The whole mass conforms to the outlines of the four-lob mother cell and there is little evidence of the rectangular intersection af cleavage planes. Fig. 12 X about 700; Fig. 13 X about 1,125. Fic. 14. Less highly magnified view of a thirty-two-celled colony i in. of whose cells cleavage is complete. about 500. Fic. 15. Two young thirty-two-celled daughter colonies and two ce AND INHERITANCE IN PEDIASTRUM. 439 of the same mother colony just ready to swarm. Empty cells of the same mother colony are also shown. The cells of the two daughter colonies already four-lobed and looking almost as if they were dividing in their short axes. X about 1,050. ; Fic. 16. The same group a few minutes later less highly magnified and _ showing seven daughter colonies and part of an eighth all more or less out of _ focus. The swarmspores have just escaped from the right-hand mother cell of the two shown in Fig. 15 and appear as a rounded cloud partly beneath the left-hand mother cell. They are swarming vigorously and are also a little out of focus. > about 500. Fic. 17. The same group a few minutes later than the stage shown in _ Fig. 16. The swarmspores have come to rest in the newly formed colony and the rounded outline of the colony and its concentric series of cells can be _ made out though it is somewhat out of focus in this figure also. \< about 500. Priate VI. Fics 18, 19 AND 20. Young daughter colonies still enclosed in the vesicles _ in which they escape from the mother cell. The vesicles are very transparent and hard to bring out in the photographs. In Figs. 18 and 20 I have traced their outlines over with dilute India ink; in Fig. 19 they are left as they ap- peared in the print and while they are very faint they can be seen and it is clear that in shape they still maintain the outlines of the mother cell even to _ the slight projections representing the larger pair of spines. The walls of the mother cells, also shown, enable us to compare the size of mother cell, vesicle and young colony. The tendency of the young colony to conform _ more or less to the oblong shape of the vesicle is obvious, but the edge view of a colony in Fig. 29 shows that in these cases at least this tendency has not prevented the formation of the plate of a single layer of cells though the space relations in the vesicle would favor the formation of two or three layered groups such as are shown in Fig. 23. X about 1,050. fe Fic. 21. Two young daughter colonies, one with sixteen, the other with ~ _ thirty-two cells, both being the offspring of a thirty-two-celled mother colony. The young colonies are of the same age, but the cells of the sixteen-celled colony are proportionally larger. >< about 500. Fic. 22. A mother colony and six young daughter colonies, all of which __ show more or less the influence of the oblong mother cell vesicle on their shape. X about 150. Fic. 23. Ten extremely young daughter colonies, all from the same mother colony. Reproduction occurred in this case after the mother colony had been sealed up as described for about eighteen hours and the free swim- ming movements of the swarmspores were almost entirely suppressed. The _ colonies are two or more layers thick and-the interrelations of the cells are entirely abnormal and yet the cells themselves have taken on the four-lobed form typical of the species. >< about 150. Fic. 24. Four young colonies from the same mother colony, illustrating still further, as do the preceding figures 15 to 23, the extreme range of fluctu- ating variation in cell arrangement which can be found in the offspring of a single mother colony, while in all the type of cell form remains quite con- stant. X about 400. ’ LIGHTING IN ITS RELATION TO THE EYE. By C. E. FERREE anp G. RAND. (Read April 13, 1917.) I. INTRODUCTION. The work of which this paper is a brief outline was done the auspices of the American Medical Association’s subcommitte the hygiene of the eye, of which Dr. William Cambell Posey Philadelphia, is chairman, and has been in progress for six The object of the work has been to compare the effect of d lighting conditions on the eye, and to find the factors in a li situation which cause the eye to lose in efficiency and to xf discomfort. In all 52 different lighting situations have been i gated, selected with special reference to the problem in hand. A a number of miscellaneous experiments have been conducted pe ‘taining to the hygienic employment of the eye. - Confronting the problem of the effect of different lightin ditions on the eye, it is obvious that the first step towards sys ; work is to obtain some means of estimating effect. The promin effects of bad lighting systems are loss of efficiency, temporary. progressive, and eye discomfort. Three classes of effect, howe description of tests designed especially for this work has previo appeared in print. Some of these tests have been designed termine the eye’s aggregate loss in functional power, others 1 in the analysis of this effect. Time can be taken here only f briefest mention of the principles on which they are based. one with which the greater part of the work has been done is : for determining the power of the eye to sustain clear seeing. two principles are involved in this test. One is that visual or clearness of seeing may be measured by the smallest visual 440 LIGHTING IN ITS RELATION TO THE EYE. 441 which the eye is able to discriminate; the other, a principle equally old, is that a loss of efficiency in a machine, apparatus, or a living organ or organism will show out more plainly when a prolonged rather than a momentary performance is required. These principles in their simplest terms have been combined into a test of the com- parative ability of the eye to maintain its power of clear seeing or aggregate functional activity under different conditions of lighting and under different kinds and conditions of use. Such a test for clear seeing was needed because the conventional acuity test had not been found to be sufficiently sensitive to fatigue conditions to warrant adoption in our work. It, we scarcely need to point out, was designed to test the dioptric condition of the eye and may be used with more or less success as a test of how far a given lighting condition is con- ducive to clear seeing with a maximum of momentary effort; but it has not the essentials of a fatigue test, nor of its converse, the ease with which clearness of seeing is maintained, which are the features needed primarily for the selection of lighting conditions for the greater part of the work that we are ordinarily called upon to do. Almost, if not quite, as good results, for example, may be gotten with it after work as before when there is every other reason to believe that the eye has suffered considerable depression in func- tional power. The reason for this is obvious. Although greatly fatigued, the eye can under the spur of the test be whipped up to give almost if not quite as good results as the non-fatigued organ when only a momentary effort is required. If fatigued, however, it can not be expected to maintain this extra effort for a period of time. The demonstration of this fact led early in our work to the introduction of a time element into the test. The principle involved is not a new one. It is merely the application of a very old and well- known one to the work of testing for ocular fatigue. If, for ex- ample, a sensitive test is wanted for the detection of fatigue in a muscle, as good results can not be expected if the test requires only a momentary effort on the part of the muscle as would be attained if the endurance of the muscle were taken into account. For our purpose, therefore, the old acuity test subjected to certain features _ of standardization for the sake of greater reproducibility has been made into an endurance test in which the fatigue or loss of func- age a iy ne - peat Neer nied 442 FERREE AND RAND—LIGHTING tional efficiency of the eye is measured by its power to sustain seeing for a period of time. In operation the test may be dese briefly as follows: The power of the eye to sustain a certain stan¢ of acuity for three minutes is measured before and after a 3-h period of reading from uniform type and paper under the ligh conditions to be tested. That is, by means of a visual acuity object, with the proper auxiliary apparatus for its control and ob vation, and a kymograph and chronograph, records are made the time the eye can be held up to this standard of pai: the time it drops below. The ratio of these quantities to each o or to the total time for which the record is made, is taken as the measure of the ability of the eye to sustain its power of clear s ‘in before and after work under the lighting conditions to be tested, Thus far. the analytical tests have been confined to the reti and the extrinsic muscles of the eye. There are four ways in wi the retina might be expected to show a depression of functi power: in a lowering of sensitivity to colored and white light; in - increase in the rate of exhaustion to light stimulation and a co sponding decrease in rate of recovery; and in an increase in the lag or time required to give its full response to light stimulation. We have already made tests for the first three of these features for effect of different lighting conditions and work is under way for testing of the fourth feature. In the work on the extrinsic mus we have again found it advisable for the sake of sensitivity in de ing small effects to use an endurance test instead of one requi only a momentary performance. That is, we have supplemented conventional abduction and adduction tests by a determination of tl power to sustain the codrdination of action on the part of the muscles needed for binocular seeing—measured by the power maintain under strain the accurate combination of binocular imz of a simple test-object before and after a period of work under lighting conditions to be tested. The eyes are put under strain combine their images to give the needed sensitivity to the When this is done even when the muscles are fresh, if the object looked at or fixated for an interval of time, it will be seen alternz as one or as two. The proportion or ratio of the time seen as to the time seen as two or to the total time of the observation IN ITS RELATION TO THE EYE. 4438 be regulated by the amount of initial strain under which the eyes are put to combine their images. The regulation of this ratio is empirical and of importance; for, as is the case with the test for loss of efficiency for clear seeing, the sensitivity of the test depends _to a considerable extent upon the initial value that is given to this ratio. The eyes may be put under strain to combine their images by interposing between them and the object viewed weak prisms and by adjusting them and regulating the distance of the object from the eye so that with the maximum of effort to see it as one, it is seen alternately as one or as two in the proportion desired. We have also tested the tendency of different conditions of lighting to produce ocular discomfort, and have explored the field of vision for the purpose of determining the liability to discomfort from the exposure of the eye to surface brilliancies of different orders of magnitude. This tendency was measured by the time re- quired for just noticeable discomfort to be set up, in the former case both with the eye at work and at rest under the lighting conditions in question, and in the latter with the eye systematically exposed to a given area and brilliancy of surface at different points in the visual field, by means of a large perimeter constructed especially for the purpose. The following aspects of lighting sustain an important relation to the eye: the evenness of illumination, the diffuseness of light, the angle at which the light falls on the object viewed, the evenness of surface brightness, the intensity of light, and its composition or color value. For convenience of treatment in this paper we have grouped the first four of these under the heading distribution factors. The work throughout has been conducted primarily for the purpose of finding out the comparative importance of these factors to the com- fortable and efficient use of the eye rather than to test the merits of various types and varieties of lighting. On the other hand, however, the investigations have not been abstract in character. That is, all the variations obtained were gotten in actual lighting situations by employing so far as possible lighting installations in common use. In order that a correlation might be had between lighting conditions and effect on the eye, the following specifications of illumination effects and conditions was made in each case. 444 FERREE AND RAND—LIGHTING | 1. A determination was made of the average illumination of test room under each of the installations of lighting used, and of the distribution of light in the room. The room was laid out in 3 ft. squares and measurements were made of the horizontal, vertical and 45° components of illumination at 66 of the intersections of the sides of these squares, and at the point of work. In all cases in which the variation of intensity was not the special point of inves- tigation, the illumination for each installation was made as peices equal as possible at the point of work. ives 2. A determination was made in candlepower per square inch of the brightness of prominent objects in the room, such as the test surface and reading page; the ceiling spots above the reflectors for the indirect installations; the reflectors and the ceiling spots abo the reflectors for the semi-indirect installations ; the reflectors, open- ings of reflectors and the lamps in so far as they were visible for the direct installations; the specular reflections from surfaces, etc and the surfaces of lowest brightness to get the range. 3. Since the angle of presentation is an important feature in ‘the effect on the eye, a determination was made also of the angle of ele- _ vation of some of the more important surfaces such as the reflector, opening of the reflector, etc., above the plane of the obserayes bi: when held in the working position. 4. Photographs were taken of the room from Care ssidion under each system of illumination. In the selection and use of observers for the work the following . are some of the precautions that were taken: Care was exercised in the first place to choose only those who had shown already a satis- factory degree of precision in other work in physiological optics and whose clinic record showed no uncorrected defects of consequence. All were under 30 years of age. Before being allowed to take part in the actual work of testing each observer was trained to a satis- factory degree of precision in the 3-minute record under a given light- ing condition and in the 3-hour test under several of the conditions to be tested. In the actual work of testing the results were com- piled from several observations and the precision was checked up by the size of the mean error. No results were accepted as signifi- cant unless the variation produced by changing the conditions to be IN ITS RELATION TO THE EYE. 445 tested was largely in excess of the mean variation or mean error for each condition tested. This, the accepted conventional check on the influence of variable extraneous factors was carefully ap- plied at each step in the work. In attempting to make any presentation of results for a problem so complicated as the one under investigation, in the space allotted, we have had to choose between giving the details for some particular piece of work and trying to draw some general conclusions from the work as a whole, supplemented by an incomplete statement of data,* so far as the tests have been applied up to the present time. We have chosen the latter alternative, although caution and our own preference are on the side of the former. As already stated, the work has been in progress for six years. 1 For a detailed statement of the data obtained in these experiments the reader is referred to the Transactions of the Illuminating Engineering So- ciety, 1913, VIII., pp. 40-60; 1915, X., pp. 407-447; 448-501; 1097-1138; 1916, XL, pp. 1111-1137; 1917, XIL., pp. 464-487. In these references will be found data on the following points for the lighting conditions tested: (a) The horizontal, 45° and vertical components of illumination at the 66 stations in the test room, the mean deviation of these values from the average illumination and the percentage mean deviation in some of the more important cases. (b) Measurements in candlepower per square inch of the brightness of prominent objects in the room, such as the test surface, the reading page, the ceiling spots above the reflectors for the indirect installations; the reflectors and the ceiling spots above the reflectors for the semi-indirect installations; the specular reflections from surfaces; etc.; and the surfaces of lowest brightness to get the range. (c) Ratios be- tween surfaces of the first, second, third, etc., order of brilliancy and surfaces of the lowest order of brilliancy, and between surfaces of the first, second and third order of brilliancy and the brightness at the point of work, to show the gradations in surface brightness. Again in some of the more important cases the mean deviation of the brightness values of the different surfaces from the average brightness of all the surfaces measured, and the percentage mean deviation have been given. (d) The angle of elevation of some of the more important surfaces such as the reflector, opening of the reflector, etc., above the plane of the observer’s eye when in the working position. (e¢) Photographs for each system of illumination representing to the eye the de- tails of the test room, the location and type of lighting units, the position of the test station, the apparatus with which the tests were made, the illumina- tion effects (distribution of light and surface brightness), etc. And (f) tables giving a detailed numerical statement of the results of the test includ- ing among other items a comparison of the average error of each set of determinations with the change of result produced by changing the lighting conditions tested, as a check on the significance of the results. 446 FERREE AND RAND—LIGHTING We have avoided, therefore, as far as possible, making any | m parison of results in different series or years; but wherever this h been done, the comparisons are based on the results of the m observer with sufficient check experiments to show that the e of observation is safely within the variation in result upon w . the conclusion is based. The following are some of the results 1 at have been obtained. ay 1. Of the lighting factors that influence the welfare of the ye those we have grouped under the heading distribution are appa ently fundamental. Thus far in the work they seem to be the m important we have yet to deal with in our search for the condi i that give us the minimum loss of efficiency and the maximum com- fort in seeing. If, for example, the light is well distributed in the field of vision and diffuse and there are no extremes of surface brightness, our tests indicate that the eye, so far as the problem 0 lighting is concerned, is practically independent of intensity of light. That is, when the proper distribution effects are obtained, intensities high enough to give the maximum discrimination of detail may employed without causing appreciable fatigue or ‘discomfort te . eye. «The work on composition or color value of light is ‘still progress. While, therefore, we are not in a position to con fully on this point, our belief based on the work which has done is that the color differences that are ordinarily present in a tificial light are not nearly so important as are, for example, h differences in the precautions that are being used to exclude h brilliancies from the field of view. The defects with regard to value are, however, as a practical problem harder to remedy. _ 2. For the type of control of distribution factors given b semi-indirect reflectors of low and medium density and the direct flectors which present, as many of them do, excessive brilliancies to opening, surface of reflector, or wholly or partially exp sources, our results show that often too much light is used in or nary work for the comfort and welfare of the eye. That is, y these reflectors, means have not yet been found to produce thi amount of light without introducing ae brilliancies into field of view. 3. The angle at which the light falls on the object viewed is IN ITS RELATION TO THE EYE. 447 important factor especially if the light is not well diffused and the surface of the object viewed is not sufficiently mat in character; but not so important, for example, as a certain evenness or gradation of surface brightness in the field of view. High brilliancies in the field of view seem in fact to be the most important cause of the eye’s discomfort and loss in power to sustain clear seeing in lighting sys- tems as we have them at the present time. In lighting from exposed ‘sources it is not infrequent to find the brightest surface from one million to two and one half million times as brilliant as the darkest; and from three hundred thousand to six hundred thousand times as brilliant as the reading page. These extremes of brightness are, our tests show, very fatiguing to the eye, especially when the high brilliancies occur in certain zones or regions of the field of view. | 4. Of the commercial systems of artificial lighting tested thus far, unmodified, the best results have been obtained for the indirect systems, and the semi-indirect systems with reflectors having a high density. By means of these reflectors the light is well distributed ; ‘in the field of view and extremes of surface brilliancy are kept within the limits which the eyes are prepared to stand. A great deal of loss in power to sustain clear seeing has been found to result from the use of semi-indirect reflectors of low and medium density and from the use of direct reflectors of shallow and medium depth. With regard to the degree of density that is most favorable to the eye, the direct reflector seems, however, to present a special case. With translucent reflectors of medium depth, our best results have been gotten so far with reflectors of medium density. This, how- ever, is not in contradiction to our principle that extremes of bright- ness are fatiguing to the eye. For if the physical efficiency of the reflector is not to be lowered by increasing its density, its opening must become brighter in some proportion to the increase of density ; i. e., in a totally opaque reflector all, and in the denser reflectors nearly all of the light sent to the working plane must come from the opening. Moreover, in case of the denser reflectors, the ceiling and the reflectors are relatively dark, while standing out in sharp con- trast to them is the bright opening of the reflector. In the reflectors of medium density, however, the reflector need not have such a high brilliancy and there is little contrast between it and its surroundings. 448 FERREE AND RAND—LIGHTING When installed on or near the ceiling in rooms of moderate I € igh the best results seem to be obtained when the opening, the surfa of the reflector and the ceiling have as nearly as possible equal bril liancy. It seems probable that the effect on the eye of the den: reflectors can be very much improved by increasing the depth of reflector and by other devices that will lower the brilliancy of the opening. In fact the best results we have as yet gotten from any type of reflector have been from a direct opaque reflector of the de bowl type, modified so as greatly to reduce the brightness of the opening, giving a field of view with the lowest maximum of bril- liancy of any we have as yet been able to obtain in an actual light- ing Situation. This reflector, 10% in. in diameter and 11% in. deep, was lined to a depth of 3.7 in. with a mat surface having a reflec- tion coefficient of about 4 per cent. Moreover, a result almost good as any we have obtained by indirect lighting was gotten giving this band or lining a reflection coefficient of about 38.5 per cent. In the former case the brightness of the opening taken from_ the position of the observer’s eye was 0.0129 cp. per sq. in., a reduc- tion of 99.8 per cent. in the maximum brilliancy of the opening; an in the latter, 0.1815 cp. per sq. in., a reduction of 96 per cent. f the former the illumination of the room was reduced on the aver 25 per cent.; and in the latter, 12.4 per cent. Poor results are giv & by shallow direct reflectors of all densities unless they are installed so high above the working plane as to be almost if not entirely ae moved from the field of view. 5. We have frequently been asked to fix an upper limit oe br : _ness which the eye can stand without any considerable loss in pe ov to sustain clear seeing through a period of work. At present thi: can be done at best only very approximately; moreover, the v assigned can not be made independent of the grouping of conditic in which this brightness occurs. For example, a lighting ins tion which has its highest brightness well within the field of viev demands a smaller maximum than one in which these brightnes: are carried outside the zone of most harmful effects on the That is, higher brightnesses can be tolerated for the totally ind: reflectors, or for direct reflectors installed on the ceiling, than semi-indirect reflectors in case of which the highest brightnesses, IN ITS RELATION TO THE EYE. 449 namely, the brightnesses of the reflectors, are in rooms of moderate height dropped well into the field of view. It is obvious also that the effect will depend on the number and size of the bright surfaces in the field of view as well as on the angle of presentation to the eye. For rooms of the size of the one in which we worked, an ap- _ proximation of a maximum brightness may be gotten from the fol- lowing data based on the testing of 52 lighting situations. For the. indirect installations the eye fell off 8.6 per cent. in power to meet the standard imposed by the test as the result of 3 hours of con- tinuous reading with the maximum brightness in the field of view of 0.138 cp. per sq. in. For the direct installation the loss was 6.6 per cent. for a brightness of 0.0129 cp. per sq. in.; 8 per cent. for a brightness of 0.1815 cp. per sq. in.; and 32.9 per cent. for a bright- ness of 0.66 cp. per sq. in. For the semi-indirect installations the loss was 15 per cent. from a brightness of 0.264 cp. per sq. in.; 48” per cent. for a brightness of 0.361 cp. per sq. in.; and 60 per cent. for a brightness of 0.614 cp. per sq. in. We would not feel inclined to recommend a maximum brightness greater than 0.15-0.2 cp. per sq. in. with the grouping of distribution factors ordinarily found in the lighting of rooms. In contrast with this, the brightness of the gas flame and oil lamp is from 3-8 ep. per sq. in.; the Welsbach mantle from 20-50 cp. per sq. in.; the carbon filament from 375- 480; the filament of the vacuum tungsten lamp from 875-1,000; the filament of the gas-filled tungsten lamp 10,271-16,433; and the open arc lamp from 10,000-50,000. 6. A marked characteristic of the effects produced by the dense - and completely opaque direct reflectors was the low illumination of the ceiling and upper part of the room, and the high and in some cases almost glaring illumination of the floor and objects in the _ working plane. So far as the effects on the eye of the kind regis- tered by our tests are concerned, however, these irregularities of illumination and of low surface brightness extraneous to the lamp and reflector seem to be of comparatively little consequence, so long as the higher brilliancies of lamp and reflector are themselves properly taken care of. With the direct reflectors, translucent and opaque, we have had quite wide variations in the distribution of illumination ranging from the well-illuminated ceilings and the com- 450 FERREE AND RAND—LIGHTING paratively evenly illuminated walls and working plane for the re- flectors of medium density to the dark ceilings and upper part of the room and highly illumined lower half for the opaque refl And with the opaque reflectors turned towards the ceiling, the trans- lucent reflectors turned both up and down, and with reflectors of both the focusing and distributing types, we have had the greate amount of light first in the upper half of the room, then in the lower half, and within limits lanes of light have been produced ; a still it has been possible to get in all of these cases comparatively good effects on the eye so long as no excessive brilliancies were in- troduced in the field of view. Again, however, we do not wish to say that this is the only factor that makes for the welfare of the eye. We wish only to call attention to its very great importance. 7. The problem of installing is not the same for the semi-in- direct as for the totally indirect reflector. In the latter case the height should be adjusted so as to give as nearly as possible a even distribution of surface brightness on the ceiling and even- ness of illumination on the working plane. In the case of the semi- indirect reflectors, especially those of low and medium densities and in rooms of medium height, if the distance from the ceiling is made great enough to produce these effects, the bright reflectors are dropped too low in the field of view for the highest comfort and efficiency of the eye. Apparently the denser they are, the more nearly they should be installed as are the indirect reflectors; and the — less dense they are the more nearly they should be installed as are — the direct reflectors of similar density, so far as eye effects of the kind revealed by our tests are concerned. In this connection it. may be pointed out that in current practice direct reflectors f general illumination are usually installed on the ceiling or as near to it as is possible, especially in rooms of low and medium height. However, while this may be a good general rule for the installation — of direct reflectors of low and medium density and of shallow and | medium depth, the question of most favorable height for the dense and completely opaque reflectors is, we believe, still open to investi- gation. 8. In the work of providing general illumination the most diffi cult feature presented in the problem of protecting the eye is en IN ITS RELATION TO THE EYE. 451 countered in the lighting of rooms of low and medium height. The difficulty decreases with increase of the height of the ceiling. In rooms whose ceilings are very high in proportion to other dimen- sions of the room, it seems safe to say that comparatively good results could be gotten with almost any reflector of modern design ; for it is much easier in such rooms to get the bright sources of light, primary and secondary, out of the zone of most harmful in- fluence on the eye. g. The loss of efficiency sustained by the eye in an unfavorable lighting situation seems to be muscular, not retinal. The retina has been found to lose little if any more in efficiency under one than under another of the lighting systems employed. 10. The observation of motion pictures for two or more hours causes the eye to lose heavily in efficiency. The loss decreases rather regularly with increase of distance from the projection screen. It seems little if any greater, however, than the loss caused by an equal period of steady reading under much of the artificial lighting now in actual use. In making these tests care was taken . to choose a projection apparatus which gave a picture compara- tively steady and free from flicker. 11. In all the conditions tested a rather close correlation is found to obtain between the tendency of a given lighting condition to cause loss of visual efficiency and to produce ocular discomfort. The tendency to produce ocular discomfort, as already stated, was estimated by the time required for just noticeable discomfort to be set up with the eye both working and at rest under the conditions to be tested. The results of this work were also carefully checked up by the determination of the mean error of the observation. II. Some oF THE CONDITIONS TESTED (COMMERCIAL TYPES OF LIGHTING). The tests throughout the work were conducted in a room 30.5 ft. long, 22.2 ft. wide and 9.5 ft. high. In Fig. 1 this room is shown drawn to scale: north, south, east and west elevations, and plan of room. In the plan of room are shown by a cross and the appro- priate numeral the 66 stations at which the illumination measure- 452 FERREE AND RAND—LIGHTING ments were made; also the positions of the outlets: A, B, C, D, E F, G and H for the lighting fixtures. In the drawing east eleva tion, one of the positions at which the tests were taken is repre- Fic. 1. Plan of test room. sented, namely, the one with six reflectors in the field of view. The walls and ceilings of this room are of rough plaster painted mat white. The floor is of medium dark tiling. 1. Direct, Semi-indirect and Indirect Systems of Lighting. — In our choice of the first set of conditions to be tested it was our purpose to make a selection that would give a wide variation in the distribution factors. Three types of lighting were chosen. One may be called an indirect system; one a direct system; and one a 3 semi-indirect system. The direct reflectors were not of the most — modern make, although they may be said to have given effects ver similar to much of the lighting in actual use at the present time. They were of porcelain ware 16 inches in diameter and only slightly ; IN ITS RELATION TO THE EYE. 453 concaved. When placed above the lamps employed, they served merely to direct the light to the working plane. No protection from the brilliancy of the light source was afforded to the eye. For the semi-indirect system inverted alba reflectors 11 inches in diameter _ POSITION A POSITION B ing units 4n field of view | 4 lighting units in field of|view 7 Po ALi V4 Al f POSITIDE C POSITIDE D 2 lighting units in field of view No lighting units in field of view ee ee —— - = Sent= ~<] tadivect - Cuart I. (Direct, Semi-indirect and Indirect Systems of Lighting.) Showing the power of the eye to sustain clear seeing under direct, semi- indirect and indirect systems of lighting, and the-effect of varying the ob- server’s position in the room or the number of bright sources, primary and secondary, in the field of vision. Power to sustain clear seeing before and after work is represented on the ordinate and hours of work on the abscissa PROC. AMER. PHIL. SOC., VOL. LVII, EF, SEPT. 14, 1918. 454 FERREE AND RAND—LIGHTING were employed. These reflectors are of modern design and : repre- sent very well glassware of medium density. In case of the indirect system corrugated mirror reflectors were used inclosed in brass bowls. These reflectors are also of modern design and give effects which may be taken to represent very well those obtained in good indirect lighting. The tests were taken at four positions in the room, one with six, one with four, one with two, and one with none of the lighting units in the field of view. The last three of these positions are marked with a cross in Fig. 1, east elevation. A graphic representation of the results of the tests for the four po- sitions is given in Chart 1. Because of the amount of space that it would require, a tabular statement of results from which these and subsequent charts were constructed will not be given in this paper. In the second series of experiments we undertook to determine the most favorable intensities of illumination for the three types of installations we had used in the first series; and in addition the effect of varying the intensity of the illumination with the particu- lar grouping of distribution factors represented in each case. ‘ tests were made in the same room, with the same fixtures, and in general with the same conditions of installation and methods of working as were described in the account of the experiments of the first series. To secure the various degrees of intensity of light needed, lamps of different wattages were employed. In order keep the distribution factors as nearly constant as possible for a given type of system, the lamps used in making the tests for that type of system were all of one wattage, i. e., were all 15’s, 25’s, 40’s, 60’s, or 100’s. For the indirect and semi-indirect systems 25, 40, 60, and 100-watt lamps were employed. Our fixtures for the direct system were so installed that either one or two lamps could be used in each fixture, totalling respectively 8 and 16. In order to get a wider range of intensities both numbers of lamps were used, 7. ¢@., one series of tests was made with 8 lamps, and another with 16, Also four intensities of light were employed in each case. Thes : intensities were secured in the 8-lamp system by using lamps to- talling 120, 365, 400 and 800 watts. In case of the semi-indirect and indirect reflectors socket extenders had to be used with the and 40-watt lamps. That is, without the extenders these lamps, IN ITS RELATION TO THE EYE. 455 on account of their smaller size, came so low in the reflectors as to change the distribution effects given by the reflector. For example, without the socket extenders with these shorter lamps, the spot of light on the ceiling, for the indirect system especially, was made smaller and correspondingly more brilliant. It was considered to be a point of interest in relation to the general problem to determine whether this comparatively small change in illumination effects would cause any difference in the eye’s ability to hold its power to e sustain clear seeing. The results of the tests for the different in- a tensities of light for the three systems of lighting are shown in 4 Chart II. Space need not be taken here to represent the compara- _ tive effects with and without socket extenders (see Trans. IIl. Eng. Soc., 1915, X., pp. 473-476). In this connection it will be suffi- cient for our purpose here to state that quite an appreciable dif- ference in result was obtained especially in case of the 25-watt lamps. These experiments constitute but one feature of a series conducted to show the effects of faulty installation. 2. Semt-indirect Reflectors Differing in Density. In the work under the first and second sets of conditions the influence of differences in the distribution factors, more especially surface brightness, was clearly revealed by the use of wide varia- tions in illumination effects. In the third set of conditions much smaller variations were employed. Such differences in effects were included as could be obtained by employing semi-indirect reflectors alone ranging from medium to dense. Six sets of reflectors were used, similar in size and shape and differing only in density. These reflectors were furnished by the Holophane Works of the General Electric Co. (now Ivanhoe-Regent Works) with special reference to the needs and purpose of the investigation. They are all of the bowl type and 8 inches in diameter. Reflector I. is a pressed Sudan toned brown; Reflector IJ. a blown white glass toned brown (an experimental product); Reflector III. a pressed Sudan; Reflector IV. a pressed Druid; Reflector V. a blown Veluria; and Reflector _ VI. a blown white glass (also an experimental product). Reflectors I., IIL., IV. and V. are commercial products, II. and VI. are in- 456 + IvI ozb Lor ogg x ogo og'I Lor ooh 2 og'o ogi Lor Sof g bSo Ez1 Lor ‘obz vy *yequoz *7eO “SOA “SHEA “HoH “he, “sa[pues 300.7 *(sdurwy or) ang smayshg Suny3ry. OZ ZOI og 6f'r Sgo LOI Sgo Sto gr't 6ro zo go “oS® = *yequoz = *[¥9 “HoH “RPA “so]purD y0J Lor ogb qd Lor oz€ 9 Lol ooz g Lol ozl V “SOA “SHEAL (sdurey g) saxyq] : wewshg ZapysyT OSE ofr zs LOI Solo ore gor 6to pA Lg'o =6(6f'0)=— EE" *oSh «= *yeyuoz = *[eo “HOWY -TH2A *sa]pued 1007 Lol oog q Lor og’ oO Lor ozf gq Lol oot ¥ "SOA "SHEA *yooulpuy : wayshg BunyZryq - ‘Spr og’s cyl ogo vGo of€ zgo «=f gS'0 ozz bgtvo zlr Sto og! *‘jequ0z =*[B9 “HoH “Hie, *sa[purr) 1004 Lor ool Lol 00g Lor ogt OIL oz Lor ozt OII 002 Lol 002 "SOA “SEAL *PeIpUl-tuleg : wayshg Zunysry “NOAHKM IN ITS RELATION TO THE EYE. 457 serted in the series to give gradations in density. These reflectors were installed 30 inches from the ceiling in accord with the prin- ciples of indirect lighting. Clear tungsten lamps were used as light sources with each installation. These reflectors are numbered in order of their density from greatest to least, that is, Reflector I. is oyv eae NGE —T a a8 Iv y : 20 ] E E oO 04 os 08 Cuart III. (Semi-indirect Reflectors Differing in Density.) Showing the tendency of the six types of semi-indirect reflectors to cause loss of power to sustain clear seeing. In A, power to sustain clear seeing before and after work is represented on the ordinate and hours of work on the abscissa. In B, percentage of drop in power to sustain clear seeing after work for the different reflectors is plotted along the ordinate and brightness of reflector in candlepower per square inch along the abscissa. Foot-candles, Candle Power'per : Volts. 2, come Matera aes Vertical. Horizontal. =o ig es chp ak ras III cieer % 1.14 2.7 0.264 Be ieee Sc Fe IIo i Oy 1.13 2.6 0.361 | SE Sep Seen 107.5 ey RAR PS wk 2.6 0.392 MMe tra os nets 105.5 3.8 / 1.15 2.5 0.614 &. eee oC wecs abies a 105.5 3-7 1.15 2.6 0.848 a. «ESE Rapes os ee 107.5 42 |. 1.16 2.7 0.920 Es 5 By multiplying the above values by 486.8 they may be converted into Ss millilamberts, a term frequently used by engineers to specify small brightness _-—s« Quantities. 458 FERREE AND RAND—LIGHTING of the greatest and Reflector VI. is of the least density. In connection it is scarcely needful to mention that the arena i density of the reflector, the lower is the brilliancy of the s which it presents to the eye. The results of this series of ¢ ox) ments are represented in Chart III. iz 80 80 _ uf 40 ae I I 0 6 iW 16 aT 80 : 7 Peek. Vie Ww mm I 40 i I on 80 40 $00 600 7000 Py, re AMG Cuart IV. (Semi-indirect Reflectors Differing in Density.) Showi the tendency of the six types of semi-indirect reflector to cause loss of po .to sustain clear seeing. In Curve A, percentage drop in power to clear seeing after work for the different reflectors is plotted against ratio o: average brightness to brightness at point of work; in B, against ratio of light est surface to brightness at point of work; in C, against average brightness in D, against ratio of lightest surface to average brightness; in E, agai ratio of lightest surface to darkest surface; and in F, against ratio of avera brightness to darkest surface. IN ITS RELATION TO THE EYE. 459 In the tables referred to on a previous page (footnote I, p. 445), we have shown for the sake of completeness of representation the gradation of surface brightness in three ways: (1) Brightness meas- urements of prominent surfaces have been made. (2) Ratios have been given between surfaces of the first, second, third, etc., order of — brilliancy, and surfaces of the lowest order. of brilliancy; and be- tween surfaces of the first, second and third order of brilliancy and the brightness at the point of work. And (3) the mean variation from the average and the percentage of mean variation have been shown. In the consideration of these specifications a number of single items might be selected as of possible significance in relation - to the effect on the eye. Among these may be mentioned the order of magnitude of the highest brilliancies ; the average brilliancy ; the ratio of the highest to the lowest order of brilliancy; the ratio of the highest order of brilliancy to the brilliancy at the point of work (brightness of test-object and reading page); etc. In order to see which of these correlate most closely with the results of the test, curves have been constructed in which some of these features are plotted against the results of the test. These curves are given in Charts III. and IV. In Chart III., B, percentage loss of visual effi- ciency is plotted against the highest order of brilliancy, namely the brightness of the reflector. In Chart IV. are grouped the remainder of the curves. 3. Translucent Direct Reflectors Differing in Density. In the fourth series of experiments it was decided to use the same reflectors as were used in the third with one omission (the blown Veluria) because of its close similarity to another in the series, and to instal them in accord with the principles of direct lighting. In this series was included also a set of reflectors of prismatic glass- ware, differing somewhat from the others in size and design.. They will be designated by the numerals L., II., III., IV., V. and VI., num- bered for convenience of treatment in the tables in the order of their effect on the eye from best to worst. Reflector I. is the pressed Druid; Reflector II. the blown glass toned brown (experimental) ; Reflector III. the blown white glass (experimental) ; Reflector IV. 460 FERREE AND RAND—LIGHTING “the pressed Sudan; Reflector V. the pressed Sudan toned ete and Reflector VI. the prismatic. The size and type of the first fir of these reflectors have already been given. Reflector VI. is of the extensive type, 834 in. in diameter and 534 in. deep. Reflectors I. Cuart V. (Translucent Direct Reflectors Differing t in Density.) — ‘Shows ing the tendency of the six types of translucent direct reflectors to cause loss of power to sustain clear seeing. In A, power to sustain clear seeing bef re and after work is represented on the ordinate and hours of work on the work for the different reflectors is plotted along the ordinate and highest — brightness of reflector in candlepower per square inch along the abscissa. “ Foot-candles Fe Candle Veuuier. Reflector. | Volts. Moka td tntoak 48°: Square Inch. Typed cess 110 5.0 1.47 3.30 0.66 — Type Hy bose 113 4.84 1.43 3-35 _ 0.924 — Type tit soi ssase ce 110 5.0 1.44 3-40 1.23 Type lV .cssier tes 110.5 5.0 1.44 3-40 1.364 Type V ain esous III 4.80 1.44 3.30 1.87 Type Vissseen ie 107 5.10 1.47 3.40 2.05 IV., V. and VI. are commercial products, but IT. and III. are expe mental products inserted in the series to give gradations in density The blown Veluria used in the former series of experiments was omitted from this series because the illumination effects obtained anc IN ITS RELATION TO THE EYE. 461 the effects on the eye, as determined in preliminary experiments, differed so little from those gotten from the blown white glass as to be considered as of little significance for the present work, These re- flectors were all used with 2% in. form “H” holders, and were in- stalled on the ceiling pendant in accord with the principles of direct lighting. Full-frosted tungsten lamps were used as light sources with each installation. The results of this series of experiments are represented in Chart V.? : In the tables* referred to in footnote 1, p. 445, we have again 2In considering these results it should be borne in mind that these re- flectors have been used to produce certain variations in illumination effects and that the work has not been conducted as a specific test of reflectors. For example, in order to secure in all cases approximately equal illumination at the test object, the lamps had to be operated at slightly higher voltages for some reflectors than for others. This produced for the different reflectors slightly different relative brightness values for outer surface and opening than would have been obtained had the lamps all been operated at the same voltage. Also clear and bowl-frosted lamps are more commonly used with these reflectors than full-frosted lamps. One effect of using clear or bowl- frosted lamps with them in this work would have been to have increased the brightness of both opening and outer surface of the reflectors and to have given, there is good reason to believe, a correspondingly uniformly poorer result for the eye. It is never entirely safe to predict results under conditions differing even slightly from what have been used, but from data at hand there is no reason to think that the change would have produced any significant difference in the relative rating of these reflectors. The full-frosted lamps were used for two reasons: (a) to test the whole group of reflectors under conditions as favorable as possible for the eye. This is admittedly only one point of view; the results might have had a more direct practical bearing had clear or bowl-frosted lamps been used. And (b), which is the chief reason, for the sake of making the work as far as possible comparable with the pre- vious work, we desired to make the illumination of the test object as nearly equal as could be for the different reflectors, translucent and opaque, and equal to that used in the former work. This was best accomplished by the selection of lamps made. 3 The following points might perhaps be cited in connection with the brightness specifications given in these tables. In case of the translucent re- flectors, installed pendant, two important items of surface brightness should be taken into account, the brightness of the opening and the brightness of the outer surface of the reflector. If a dense reflector is chosen, for example, the brightness of the opening tends to become excessively high; also its ap- parent or physiologic brightness is increased by induction from its dark sur- roundings which effect does not register on the photometer. If, on the other hand, the reflector chosen transmits too much light the brightness of the outer surface of the reflector becomes too high for the comfort and welfare of the 462 FERREE AND RAND—LIGHTING shown the gradation of surface brightness in the manner described in the preceding section. And in order to ascertain which of the — brightness specifications—order of magnitude of highest brilliancy, — average brilliancy, ratio of highest to lowest order of brilliancy, ratio of highest order of brilliancy to average brilliancy, ratio of average to lowest brilliancy, ratio of highest-order of brilliancy to brightness at point of work (brightness of test object and reading page), etc.—correlate most closely with the results for the tendency | to cause loss of power to sustain clear seeing, curves have been con- structed in which a number of these features were plotted against — the results of the tests. These curves are given in Charts V. and VI. In Chart V., B, per cent. loss in power to’ sustain clear seeing is plotted against the highest order of brilliancy that varies by any . considerable amount from installation to installation, namely, the — brightness of the reflector—outer surface and opening. In Chart VI. are grouped the remainder of the curves. o, Three points may be noted perhaps with reference to these ; charts: (1) The prismatic reflectors, which differ in design from the a rest of the series, more or less conspicuously fall out of the curve in every case but two. The effect of difference in design on the smooth- ness of the curve comes out especially in the results with the opaque reflectors (shown in the report of the next series of tests), in which — case there are marked differences in both size and design. All of the curves plotted on the above bases are very irregular in case of these reflectors with the exception of separate curves for three which are — similar in design. For a statement of the probable reasons for this inequality see this paper, pp. 468-9. (2) The greater regularity of — the curves is rather strikingly marked in which the highest order of brilliancy that varies by considerable amounts or the ratios in which eye. For the translucent reflectors used in these tests, the best results have been obtained with the reflectors of medium density. The reverse of this was true, it will be remembered, when the same reflectors were installed in- verted. The highest brightnesses when these reflectors are installed pendant — are the filament spots on the lamps. Only very small areas of these spots are visible, however, and their brightness and the brightness of the lamp differ so little from installation to installation as to be, in all probability, of - relatively little consequence in a comparative study of effects on the eye. The significant variables are thus the brightnesses of the outer surface and opening of the reflector. IN ITS RELATION TO THE EYE. 463 this quantity appears, is plotted against the results of the test. This, it will be remembered, was true also of the work of the preceding I< Wee E bbe F | .- isa 68700 600 300 i000 Cuart VI. (Translucent Direct Reflectors Differing in Density.) Show- ing the tendency of the six types of translucent direct reflectors to cause loss of power to sustain clear seeing. In Curve A, percentage loss in power to sustain clear seeing after work is plotted against ratio of highest brightness of reflector to brightness at point of work; in B, against ratio of highest brightness of reflector to average brightness; in C, against ratio of highest brightness of reflector to darkest surface in field of view; in D, against aver- age brightness; in E, against ratio of average brightness to brightness at point of work; and in F, against ratio of average brightness to darkest sur- face in the field of view. 464 FERREE AND RAND—LIGHTING experiments. (3) The range of brightness for this set of experi- 5 ments is quite a little higher in the scale than for that of the former : experiments with the same reflectors. This fact should be borne in mind, for example, in comparing the shape of the curves for the two sets of experiments in which highest order of brilliance is” plotted against the results of the tests. In case of the present ex- | periments, for example, the curve begins at a point, 0.66 candle- = power per square inch, which is well above the knee of the former curve. That is, the two curves are in general quite similar in shape when they are compared for the same range of brightness values. : The former curve has, however, the greater regularity; but in con- nection with this fact it should be borne in mind that a variation of — the brightness factor in separation can be more nearly accomplished : with these reflectors when they are installed inverted than ine they are installed pendant. ) ee ir 4. Opaque Direct Reflectors Differing in Dimensions, ; - i | Lining and Design. . In this series of experiments the testing of pendant reflectors was continued. Seven totally opaque reflectors differing in lining, — dimensions and design were used. In connection with the work of | this series of reflectors, two points of perhaps more than usual in- — terest may be noted: (a) By means of a modification of one of, the reflectors used, Reflector IV., made to reduce the brillianey of the 3 opening, a field of view was given having the lowest maximum of brilliancy of any that we have as yet been able to obtain in an actual — lighting situation; and (b) we were able to test more effectively than — in any lighting situation previously used, the importance of evenness — of surface brightness compared with evenness of illumination as av factor influencing the ability of the eye to maintain its power of E clear and comfortable seeing. The opaque reflectors represent a more promiscuous selection than the pendant translucent reflectors previously used. That is, in case of the translucent reflectors, all but one, Reflector VI., were of — the same size and design, and the variations in the illumination — effects were obtained by varying the density of the reflector alone; IN ITS RELATION TO THE EYE. ' 465 while in case of the opaque reflectors the significant variations in the lighting effects were produced by changing the size (more especially the depth) of the reflector, the lining and the design. Of these Re- flector IV. was of blown silvered glass with a system of spiral corrugations on its reflecting surface for the diffusion of light. This reflector was of the deep bowl type, 10% in. in diameter and 11% in. deep. Reflector V. was also of blown silvered glass and had a vertical system of finer corrugations than had Reflector IV. This reflector was 93 in. in diameter and 8 in. deep and was more dis- tributing in type than was Reflector IV. Both of these reflectors were used with 3% in. form “A” holders. Reflector VI. was a steel, aluminum-finished reflector of the intensive type, 814 in. in diameter and 7% in. deep. This reflector was provided with a clip ring which was attached directly to the socket. Reflector VII. was a porcelain-enameled steel reflector of the shallow dome or dis- tributing type, 15 in. in diameter and 6% in. deep. This reflector was used with a 2% in. form “O” holder. Early in the work with these reflectors it was found that if good results for the eye were to be obtained with dense or com- pletely opaque reflectors, some way must be gotten either of shield- ing the eye from the opening of the reflector or of reducing its bril- liancy which increases as the density of the reflector is increased. Obviously something can be accomplished in this direction by using reflectors of the deep bowl type, giving specular rather than diffuse reflection, if the angle of presentation to the eye is not too great. Reflector IV., for example, is of this type. The opening of this reflector was of low brilliancy when viewed at an angle of 13° 19’, the angle of presentation to the eye for the two reflectors farthest from the observer in the present series of experiments. High up in the reflector, however, was a small but brilliant image of the lamp, a part of which was visible at an angle with the eye of 20° 17’, the angle made by the two reflectors at Outlet B (Fig. 1), and still more at an angle of 40°, the angle made by the two reflectors nearest the observer. It was thought advisable to find out how much this re- flector could be improved in its effect on the eye by reducing the amount of reflection for a certain distance above the lower edge of the reflector. This was accomplished for the purpose of these ex- 466 FERREE AND RAND—LIGHTING periments by lining the reflector to a depth of 9.5 cm. (3.7 in.) with a mat surface of low reflection coefficient. This lining formed all _ of the surface that was visible in the openings of the four reflectors - at Outlets A and B, and cut out the image of the lamp in the two reflectors at Outlet B. In the two reflectors nearest the observer, however, some of the bright lining and image were still visible, but the angle of presentation was here great enough that comparatively little effect on the comfort of the eye and its power to sustain clear seeing was had or was to be expected. Two sets of lining were used, one a very dark gray (reflection coefficient of about 4 per cent.) ; the other a lighter gray (reflection coefficient of about 38.5 per cent.). Reflector IV. provided with the first of these linings is designated in the charts as Reflector I., and with the second, as” Reflector II. Still another modification was made of this reflector to lessen the effect of the opening on the eye. The apparent or physiologic brightness of this opening, as was the case with the other opaque reflectors, was enhanced by induction from the dark green coating or backing on the outer surface of the reflector. This effect is quite noticeable on inspection where a comparison with a reflector presenting less or no induction is afforded, but does not register in the photometer because the surroundings are not included in the photometric field. In case of Reflectors I. and II. this induction was lessened a great deal by covering the outside of the reflector with a closely fitting cap of mat white paper. Because of the favorable results obtained with these modified reflectors, a similar modification of Reflector V, was made by the manufacturer for the purpose of reducing the brilliancy of its open- ing. In this case the band was made permanent by sand-blasting the corrugated glass surface of the reflector. The coefficient of re- flection of the surface thus prepared was approximately 52 per cent. The band was made 5 cm. in width. While considerable improve- ment in the effect on the eye was produced by this modification, not nearly so good results were gotten as in the other case because (@) the coefficient of reflection was not sufficiently reduced by the sand Ee blasting; and (b) Reflector V. was not deep enough to give the best a results with this type of modification. The tip of the lamp, for 3 example,.was visible to the observer in case of four out of the six oS IN ITS RELATION TO THE EYE. 467 reflectors in the field of view. This reflector is designated in the charts as Reflector III. All of the reflectors of this series were in- stalled on the ceiling pendant in accord with the principles of direct lighting. It was our wish to conduct this investigation, as has been the case in all of our work on the distribution factors, with the color value and the intensity of light as nearly equal as possible at the test object. Clear tungsten lamps were used with Reflectors I., II. and IV.; full-frosted lamps with Reflectors VI. and VII.; and bowl- frosted lamps with Reflectors III. and V. Clear lamps were used in the cases mentioned because in the first place in reflectors of this type the lamps were not visible to the observer at the point of work; and secondly, although the illumination given was high in the aver- age, the distribution was such as to give a low illumination at the point of work. That is, the tendency of these reflectors, installed at the height used in our test room, was to give lanes of light directly beneath the two rows of reflectors, shading off to a correspondingly low value on either side. Full frosted lamps were used with Re- flectors VI. and VII., because with Reflector VI. a part and with Reflector VII. all of the filament would otherwise have been visible to the observer ; also the value of the illumination at the test object would have been much too high as compared with the other reflectors in the series and higher than the values used in previous work. In case of Reflectors II. and V. both of the above objects, namely, the better protection of the eye from the filament and the performance of the tests with the illumination values as nearly as possible equal to those obtained with the other reflectors and in previous work, was best accomplished by the use of bowl frosted lamps. The results of this series of tests are represented in Chart VIL. Again in order to ascertain which of the brightness specifications— order of magnitude of highest brilliancy ; average brilliancy ; ratio of highest to lowest order of brilliancy; ratio of highest order of bril- liancy to average brilliancy; ratio of average to lowest brilliancy; ratio of highest order of brilliancy to brightness at point of work (brightness of test object and reading page) ; etc.—correlate most closely with the results for the tendency to cause loss of power to sustain clear seeing, charts were constructed in which a number of 468 FERREE AND RAND—LIGHTING pared with the corresponding charts given for the preven expe ments, these charts show great irregularity unless separate — ; fe RY Bic, SS Nee | | Cuarr VII. (Opaque Direct Reflectors Differing in Lining, Dimensions and Design.) Showing the tendency of the seven types of cones direct sustain clear seeing before and after work is represented on the ordinate and hours of work on the abscissa. Foot-candles, Reflector. Volts. Vertical. Horizontal. Typed. ess IIL 2.25 0.72 Type:khigs A . * a. an y fe Lu TES ib ie aie ° =< .° 1 | y = 7F oO ° & 1 rr ? s ! a S S Fic. 1. Chart showing distribution of regions on which small nebule were counted. the fact that the present program deals primarily with the regions containing spirals, it is necessary to investigate further any possible effect which a concentration of regions near the north galactic pole might have upon the resulting estimate. This objection was urged by Fath (loc. cit.) in explanation of the difference between his esti- mate of 162,000 and the 500,000 of Perrine.. He found, on plotting Perrine’s regions, that 33 per cent. of these were within 45° of the north galactic pole, while less than 20 per cent. of the Kapteyn areas were within this distance. In the present program I find that 117 regions, or 26 per cent. of the regions which I have used, are situ- ated within 45° of the north galactic pole, in an area amounting to but 14.6 per cent. of the sky, and these 117 regions contain 2,997 spirals, or about 48 per cent. of the whole. To avoid the effect of this concentration in the north polar galac- tic area, it will be advisable to subdivide the material available. We 516 CURTIS—NUMBER OF SPIRAL NEBULZ, may divide the celestial sphere into four areas, two of which will be the zones of 45° radius about the two galactic poles; a third area will comprise the two zones, each 15° in width, extending from — 30° to —45° and from + 30° to + 45° galactic latitude, and the fourth will be the zone 60° wide extending from — 30° to + 30° galactic latitude. The results are indicated in the following short table: EE Galactic Latitude. Rogices. | Dey. | Speak, | Sq eg oP 457 80-400? oa os as eae II7 88.50 2,097 34 205,000 =~ 467 to 90%! oo. 5s ck cots 43 32.25 918 28 - 169,000 © 90" £0 FA is eee ee 62 46.50 1,117 24 204,000 = 30° 10 + 908 eae 217 162.75 1,179 7 144,000 Fotele. ica bccs eee 439 6,211 722,000 ; It will be seen from the above that there is a concentration o the smaller nebulz in the vicinity of the north galactic pole similar to that which obtains among the larger, visually discovered, spirals, and that the density about the south galactic pole is somewhat less marked. The data given in the third line of the table are of special interest, showing that the small spirals persist to a distance of at least 60° from the galactic poles, with only a slight diminution the degree of density which obtains in the polar areas. As this revised estimate, 722,000, is equal to that of Fath pl that of Perrine, with several thousand to spare, a discussion of the possible reasons for the discrepancy becomes imperative. The has lowing points may be considered: A. It may be urged: that my count has not been sufficient con- servative, and that I have possibly included many spurious objects. The detection of the faintest and’ smallest nebulz is very largely a matter of experience, and all who have worked with photographic plates soon learn, by hard necessity, to recognize the average fla at a glance. A very large proportion of the objects are unmistak- ably nebule. As to the faintest nebulz, it is astonishing how faint and small are the nebule which two “clean” duplicate plates will reproduce. For a large proportion of my regions no duplicate plates exist, and I have been necessarily guided by the experience deriv CURTIS—NUMBER OF SPIRAL NEBULZ. 517 from regions taken in duplicate. I am unwilling at present to admit that as many as five per cent. of the nebule counted by me are spurious. Even if twenty per cent. were spurious, we should still have to account for over half a million nebule. B. The theory may be advanced that the small spirals occur in greatest profusion in the regions contiguous to the larger members of the class, which might explain why fewer nebulz were found by Fath. -His plates were taken at the centers of the Kapteyn areas where the larger nebulz would be included only by chance, whereas, from the purpose of the Crossley nébular program, nearly all the _ plates have some N. G. C. object central. This point is difficult to prove or disprove without a special investigation comprising many plates taken at random in the galactic north polar region. It is certain that the small spirals frequently show a gregarious tendency ; sometimes one half of a plate will record many small spirals while the other half records very few. The greatest number of nebule found on a single plate was 304 (checked by a duplicate plate) in the region of 12" 55™, + 28° 30’ (shown in Fig. 2) ; the region about N. G. C. 4826, less than 7° distant from this, shows but 2. While the small nebulz are evidently quite irregular in their distribution, it would seem that the large number of regions included in this discussion is sufficient to afford a true representation of their average frequency. C. Sharp focus and perfect images are essential for the detec- tion of the smallest and faintest spirals. On plates where large num- bers of small nebulz are found, the majority are, as a rule, detected in the area 20’ in radius about the optical axis as center, comprising only 0.35 of a square degree. Ata distance of 30’ from the optical axis the parabolic images are very poor, and only the brighter of the small nebulz can be detected in these regions, the faintest nebule being obliterated by the blurring and spreading of the image. Dr. Fath used very large plates, 612 X 8% inches in size, in his work on the number of the small nebule, with the 60-inch reflector, and it appears that he used almost this full area, inasmuch as he states that the used area of his plates was 1.88 square degrees. As the full area of the Crossley plate is 0.9 square degrees (and the outer portions of this are so poor because of the parabolic distortions that 518 CURTIS—NUMBER OF SPIRAL NEBULZ. the effective area used in my counts is believed to be less than 0.75 square degrees) this would mean, if the two reflectors were of the same focal length, that the images on one half of the angular area of the large plates used by Fath were worse than those in the re- jected edge strips of the Crossley plates. But the greater focal -m Fic. 2.. Region of many small nebulz at 12" 55" + 28° 30’. 249 small nebulz in an area 38’ X 39’. ratio of the 60-inch would increase this disadvantage, even allowing for certain advantages which might partially counterbalance this due to the greater linear scale of the plates. The following com- CURTIS—NUMBER OF SPIRAL NEBULZ. 519 parisons of the aberrations of the image of an infinitely distant object in the focal plane of a single, perfect, parabolic mirror, as calculated from Schwarzschild’s formule* for the Crossley and the 60-inch reflectors, will illustrate this point. . Tue Cross_tey ReFiector; Focat Ratio=—1: 5.8. Distortion by Field Distortion by Dist, from Optical Axis. Curvature. Coma. le ada tivaes ce hci akpceccesb ss at 2” 8 10”.1 mio ncorner -Of Plate). 650.2505 cic se eee 13”.7 7 THE 60-INCH REFLECTOR; Foca, RatTio—=1I: 5.0. Dk eee ee 23 12. .5 me tcormer Of plate) ......-..2..s0.<- ek 27”.5 From my own experience in counting these minute objects on the x Crossley plates it would appear to me that the actual effective area _ used by Fath must have been very much less than the 1.88 square _ degrees assumed by him in his calculations.® It is my opinion also that the Lumiére Sigma plates which he used are not the best for the end in view. These plates are of very great speed, and are in- valuable for some purposes. I have long since ceased to use them for nebular work, however, believing that the slightly slower, but beautifully “clean” Seed 27 and Seed 23 plates really show the faintest details better. With the smaller grain and clear background existing in the Seed plates, very small and faint nebulz “stand out” much more plainly than on the more rapid Sigma plates. _ D. It is not impossible that a considerable proportion of the thirty or so plates which Fath took within 45° of the north galactic pole happened to strike regions of comparatively few small nebule. Had he chanced to include four such regions as the following: *Untersuchungen zur geometrischen Optik,” Abh. Kén. Ges. d. Wiss. su Géttingen, Math.-Phys. Kl., N. F., 4, 1, 2, and 3, 1905. 5 Since the completion of the manuscript of this paper Dr. Fath has pub- lished a note on “ The Probable Number of Nebule” in A. J., 728, March 12, 1918, in answer to a letter in which Dr. Perrine had called his attention to the large angular area and the increased parabolic distortions in the outer parts of the plates as a factor in the smaller estimate made by Dr. Fath. Dr. Fath finds, by taking the counts found from the areas 40’ square in the center of the Mt. Wilson plates, that there is a marked increase in the number of nebulz found, amounting to about 60 per cent., and increasing his earlier esti- mate to 262,000. 520 CURTIS—NUMBER OF SPIRAL NEBULA, ® 9" + 55° 34 sehen ee eeeepeeeeeeessserteseserees es it ee + 17° ye rieteeeeeaeeenereseetereeceasasennes 12" 1™+ 6° 1 ; 12 55" + 28° 30! In conclusion, I see no reason, at present existing the estimate made in this paper, that at least 700,000 are within reach of large reflecting telescopes. Beca that the faintest and smallest members of the — ible rn well exceed one million. THE NAYADES (FRESHWATER MUSSELS) OF THE UPPER TENNESSEE DRAINAGE. WITH NOTES ON SYNONYMY AND DISTRIBUTION. ¥ By A. E. ORTMANN, Pu.D., Sc.D. (Read April 19, 1978.) The present enumeration of the mussels of the upper Tennessee is the result of the writer’s work carried on in this region since 1912, under the auspices of the Carnegie Museum of Pittsburgh, Dr. W. J. Holland, director. It is intended, herein, to give a complete synopsis of this fauna, using the modern system, and the accepted rules of nomenclature, together with a full synonymy of the various forms, as far as it has been firmly established. Much stress has been laid upon the facts of the geographical distribution, because it has become evident that not all of the species are uniformly distributed in this area: The latter includes all of the upper Tennessee drainage from Chattanooga, Tenn., upward, comprising largely eastern Tennessee, _ a small section of northern Georgia, and parts of North Carolina (in the high mountains), and southwestern Virginia. It appears that the Walden Gorge of the Tennessee River, below Chattanooga, forms some kind of a barrier to Nayad distribution, at least for cer- tain species ; at any rate, it forms a natural division within the Ten- nessee system. Of course, not all parts of this drainage have been investigated by myself; but collections have been made in all of the more important streams; and, together with the records obtained from other sources, it is believed that this fauna is now rather com- pletely known. The region in question is known as one of the chief centers of Nayad development, and may be called the most prolific section of the world in this particular group. Many of the species described by the older authors (Conrad, Lea, and others) originally came from this region. But no synopsis of the whole fauna has been pub- lished, except the attempt made by Lewis (1871 and 1872). PROc. AMER, PHIL. SOC., VOL. LVII. II, SEPT, 30, I918. 521 522 ORTMANN—NAYADES OF In this connection it should be noted that Lewis’s “ Holst , River” actually is the Tennessee River below Knoxville (chiet the region of Concord). Indeed, even at present, the natives in Cor cord call the river there “ Holston”; but the maps of the U. S. Ge logical Survey give the name Tealesses to the river below the jun tion of the Holston (proper) and French Broad. In Knoanite } river is called Tennessee. Sa et ‘ Aside from Lewis’s paper, only a few are at hand ie tribute to the fauna of this region. One of them has been pub a1 by Pilsbry and Rhoads (1896). This is, however, by no me: synopsis, containing hardly half of the forms which are found But on account of the good locality-records it is very valuable fact, it is, up to the present time, the most accurate publicati this respect. And further, two preliminary papers have been p lished, based upon my own collections; one by myself (Ortma 1913b), the other by Goodrich (1913). These, however, treat only of the headwaters-region of Powell, Clinch, and Holston in Virginia, In addition to the material collected by myself, I have examined the upper Tennessee shells in the collection of Mr. B. Walker in De- troit, and I want to express to Mr. Walker my best thanks for the privilege of examining his shells, and the delightful days I spent his home in April, 1916. Mr. Walker has a great number of shells _ from this region obtained from older collections, which in part: are” cotypes, topotypes, or other authentic material. But the greatest treasure in his collection are the Nayades collected by Professor Dr. C. C. Adams in 1899 to 1901, in the course of his work on Jo, be- cause Professor Adams always was very careful in recording. s: localities. eee A large number of the “species” described by Lea ons re from very insufficient material), and of those listed by Lewis, synonyms. Additional species have been subsequently described by various authors; but also these are mostly synonyms. There i: rule, observed in many cases, and indicated first by Wilson & Clark, ’14, that one and the same shell assumes different shapes in the lars rivers and in small streams and headwaters, a rule the existence which will be shown elsewhere; and it is easily understood why various local races have been regarded as good species, as long 523 UPPER TENNESSEE DRAINAGE. 4 poop? : . 4 i : ne on) , is H Sn, i. ee ae ae 3 ye o ( ee 5 } ak ne C , sy, ge } ; a Pd f ; { oe wnenn ans eee s adaw econo ee aie o i ‘yet Hy Py tabensd wince ba ¢ cS . nAowad, & / =~ A WOQONVAL 4 Sa ‘ sy ! eQ f \ H -* 1 - } : of of / enor r™ H <=" i i preety , Meeag agua” ik 7 1 4 b ‘ r a7 VD of OM mi \ | snl” a ‘ PESTHIAGIINS Am J9VNIVUG JISSINNIL YaddNn 524 ~ ORTMANN—NAYADES OF the intergrades were not known. But the discovery of the latter- and this was one of the problems to which I directed my attention- has necessitated the cancellation of a great number of these | m1 species. Nevertheless, the fauna still remains remarkably ich, which surely in a large part is due to the comparatively old age of this river system, to the diversity of its character, and also to certain changes of the — which have taken place in the geological past. i, In the following pages, the correct names of the vaste | are given, conforming to the systematic arrangement published | the writer (Ortmann, 1910, 1912b), and conforming with the rules of priority. It should be remarked, that of practically all Tennessee- forms the anatomy has been investigated, but has not yet been pu lished of all of them: the description of the rest will appear in time. The two great papers of Simpson (1900 and 1914) are taken as a basis, and the quotations are from the last paper, so that it can be easily seen, where changes in nomenclature have been introduce Also the names used by Lewis, those “used by Pilsbry and Rhoads, y myself and Goodrich are given, in order to facilitate comparison with our list. The synonyms are all quoted unless they have been recognized and accepted as such by Simpson; but other references have been largely omitted, for the reason that they ee found in Simpson’s paper. as No full descriptions of the forms are given, but ae chief characters are briefly indicated. The extralimital distribution has not been given in detail, Here and there it has been referred to, but only in especially interesting cases. In this respect, much work remains to be done, and in many North American Nayades the exact boundaries of the distribution have not yet been exactly located. mt The material, upon which the present paper is founded, has deposited in the Carnegie Museum of Pittsburgh, and comes chic from the collections made by myself; a very small part has come from other sources. The Carnegie Museum is in possession of the old collections of Hartman, Holland, and Juny, and a more recent collection has been bought from H. H. Smith. In addition, th museum is indebted to Messrs. Frierson and Walker for occasion UPPER TENNESSEE DRAINAGE. 525 exchanges of rare forms. As has been mentioned above, also Walk- er’s collection has been examined, and incidentally some of the ma- terial of the U. S. National Museum, in Washington, chiefly some - of Lea’s types, has been studied. It is believed that the Carnegie Museum possesses now the best collection representing the Upper Tennessee fauna, and with regard to the illustration of the distribution of the various forms, it has no equal, not to speak of the fine collection of soft parts. In view of the gradual, slow but steady, deterioration of the fauna in conse- quence of stream-pollution, there is great danger that the fauna will largely become destroyed, and that it will be impossible, in the future, to duplicate this collection. At the present time, conditions are fair, in some parts splendid; but there are already polluted streams, in which the fauna is gone. Such are: the Powell River, for a certain distance below Big Stone Gap, Va. (wood extracting plant); the North Fork Holston for a distance below Saltville, Va. (salt and plaster of Paris industries) ; French Broad River at Asheville, N. Car. (pollution comes—as I have been informed—from Davidson River, farther up) ; Big Pigeon River, from Canton, N. Car., all the way down (woodpulp and paper mill) ; Tellico River below Tellico Plains, Tenn. (old wood pulp and extracting mill). The building of dams (for water power, etc., for instance in Nolichucky River near Greenville, Tenn.) also has a deteriorating effect upon mussel life, and all this surely will increase in the future. BIBLIOGRAPHY. Aside from the well-known older papers of Lamarck, Say, Rafinesque, Conrad, Call, and others, the following have been principally used in the preparation of this paper. Frierson, L: S. 1911. Remarks on Unto varicosus, cicatricosus and Unio compertus new species. Nautilus, 25, 1911, pp. 51-54. 1914a. Remarks on Classification of the Unionide. Nautilus, 28, 1914, pp. 6-8. 1914b. Observations on the Genus Symphynota. Nautilus, 28, 1914, p. 40. Goodrich, C. ; 1913. Spring Collecting in Northwestern Virginia. Nautilus, 27, 1913, pp. 81, 82; 91-95. Hinkley, A. A. 1906. Some Shells of Mississippi and Alabama. Unionide. Nautilus, 20, 1906, pp. 52-55. 526 ORTMANN—NAYADES OF Ne Lea, I. 1832-1874. Observations on the Conus Unio. Vol. 1-13, 1832-1874. Lewis, J. 1871. On the Shells of the Holston River. Americ. Journ. 1871, pp. 216-219. ie 1872. Shells of Tennessee, No. 2, Proc. Acad. Philadelphia, 187: 108-114. Ortmann, A, E. 1910. A New System of the Unionide. Nautilus, 23, 1910, pp. 114- 1g11a. The Anatomical Structure of Certain Exotic Najades. — 24, IOII, pp. 103-108; 114-120; 127-131. 1911b. A Monograph of the Najades of Pennsylvania. Mem. Carne Mus., 4, 1911, pp. 279-347. 19I2a. Cumberlandio, a New Genus of Najades. Nautilegs, %, ‘QI: 13-14. 1912b. Notes upon the Families and Genera of the Najades. negie Mus., 8, 1912, pp. 222-365. ea 1913d-I015. Studies 1 in Najades. Nautilus, 27, 1913, pp. ene pp. 28-34; PP. 41-47; pp. 65-69; 1915, pp. 106-108; pp. 29, I915, pp. 63-67. : 1913b. The Alleghenian Divide and Its Influence upon the Fre: Fauna. Proc. Amer. Philos. Soc., 52, 1913, pp. 310, 311 1916. The Anatomy of Lemiox rimosus (Raf.). Nautilus, 30, 1916, pf 39-41. 1917. A New Type of the Nayad-Genus Fusconaia, Group of ft nesiana Lea. Nautilus, 31, 1917, pp. 58-64. Pilsbry, H. A., and Rhoads, S. N. 1806. Céntriiptions to the Zodlogy of Tennessee, No. 4. Proc. Acad. Philadelphia, 48, 1896, pp. 487-506. Simpson, C. T. 1900. Synopsis of the Naiades or Pearly Freshwater Migieiea U. S. Nat. Mus., 22, 1900, pp. 501-1044. 1914. A Descriptive Cotalnase of the Naiades or Pearly Fresh 1 Mussels. Bryant Walker, 1914. cart Utterback, W. J. nee 1916. The Naiades of Missouri. Amer. Midland Naturalist, . pp. I-200. Vanatta, E. G. we: 1915. Rafinesque’s Types of Unio. Proc. Acad. Philadelphia, 191: 549-559. ae Walker, B ee 1g10a. Description of a New Species of Truncilla, Nautilus, 24, PP. 42-44. : : 1910b. Notes on Truncilla, with a Key to the Species. Nautilus, 24, pp. 75-81. 1911. Notes on the Distribution of Margaritana monodonta Say. tilus, 25, I911, pp. 57-58. 1916. The Rafinesque-Poulson Unios. Nautilus, 30, 1916, pp. 43-47. UPPER TENNESSEE DRAINAGE. 527 Wilson, C. B., and Clark, H. W. 1914. The Mussels of the Cumberland River and its Tributaries. Bur. of Fisher., No. 781, 1914, pp. I-63. Family : MARGARITANID& Ortmann (19114, p. 129). Genus: CUMBERLANDIA Ortmann (19124¢, p. I3). I. CUMBERLANDIA MONODONTA (Say), 1829. Unio monodonta Say, ’29.—Unio monodontus Lewis, ’71—Marga- ritana monodonta Ortmann, ’12), p. 233 (anatomy ).—Marga- ritana monodonta Simpson, ’14, p. 521. Widely distributed in the larger and medium rivers: Tennessee, Clinch, and Holston. According to Walker (’I1), it goes up, in the Clinch, to Union Co., Tenn., but I found it also at Clinch River Sta- tion, Claiborne Co. In the Holston, I have traced it up to Austin Mill, Hawkins Co., Tenn. Walker gives it also from Little River, but this can be only in the lower part of it. This is generally regarded as a rare species; but in the lower Clinch, in Anderson Co., and in the lower Holston, in Knox Co., Tenn., it is locally abundant. Type locality: Falls of the Ohio. Family: UNJIONIDZ Ortmann (19114, p. 129). Subfamily : UNIONINZE Ortmann (1910, p. 116). Genus: FusconarA Simpson (1900). Ortmann (1912), p. 240). 2. FUSCONAIA PILARIS (Lea), 1840. Unio pilaris Lea, ’40——Unio globatus Lea, ’71-—Unio ebenus and pilaris Lewis, ’71 and ’72.—Unio pilaris Pilsbry & Rhoads, ’96.— Quadrula andrewsi Marsh, ’02—Quadrula beauchampi Marsh, ’02.—Quadrula pilaris, andrewsi, beauchampi, globata Simpson, "14, pp. 893-899. : This is the upper Tennessee representative of F. subrotunda Lea of the Ohio drainage, and it may be merely a dwarfed, globular form 528 ORTMANN—NAYADES OF of the latter. I have seen the normal subrotunda from the Tet in northern Alabama, but I cannot tell whether the two each other. The typical form of F. pllabie « passes into the follow races in an upstream direction, and it is hard to draw a line | them. In order to preserve the oldest names given, I have arb concluded to call by this name those more swollen forms where the : transverse diameter of the shell is 55 per cent. of the length and over. This typical F. pilaris is restricted to the large rivers: Tenness lower part of Little Tennessee, lower French Broad, lower Hol: and lower Clinch. In the two latter rivers it is rare, and interg with the next form, going up, in the Holston, to baht. 2/3 Co. in the Clinch, to Anderson Co., Tenn. is Type locality: French Broad River, Tenn., and Holston Ri Tenn. (topotypes examined). a 3. FUSCONAIA PILARIS LESUEURIANA (Lea), eae ; Unio lesueurianus Lea, ’40.—Quadrula flexuosa Simpson, o Quadrula flexuosa Simpson, ’14, p. 887. £ According to Simpson, ’14, p. 894, lesueuriana is syno: pilarts. Like pilaris, but less swollen; the diameter of the shel is 45 per cent. to 54 per cent. of the length. = Of Q. flexuosa, I have seen, in the Walker collection, two spe mens from Holston River, Knoxville (Mrs. Andrews), and Ot specimen from Holston River, Hamblen Co. (opposite Grainger (Marsh). They had been compared with the type of flerwosa, thus labeled by Marsh. Two of these, with the diameter of 53. 54 per cent., are lesueuriana, while one (Knoxville), with dia of 56 per cent., is pilaris, Simpson’s measurements of flexuosa the diameter of 51 per cent., and thus it should be made a syn of lesueuriana, This is the form of the medium rivers; it turns up in single viduals in the region-of Knoxville, and becomes the prevailing ft in the lower Clinch and in the whole Holston proper. In the Ci it has been traced up, in single individuals, to Scott Co., Va., and UPPER TENNESSEE DRAINAGE. 529 also goes into the lower Powell, up to Claiborne Co., Tenn. From the Holston, it goes in the lower parts of the North and South Fork: but at and near its upper limit, it always intergrades with the next form (bursa-pastoris). : Type locality: Caney Fork River, Tenn. (and Holston River, Tenn.). (Thus it seems to exist also in the Cumberland drainage, but this requires further investigation. Wilson and Clark, ’14, do not mention it.) 4. FUSCONAIA PILARIS BURSA-PASTORIS (Wright), 1896. Unio bursa-pastoris Wright, ’96—Unio kirtlandianus Pilsbry & Rhoads, ’96.—Fusconaia bursa-pastoris Ortmann, ’I3a, p. 90 (anatomy), and °13b, p. 311—Fusconaia bursa-pastoris Good- rich, ’13, p. 93 —Quadrula bursa-pastoris Simpson, ’14, p. 890. This is the compressed headwaters-form of F. pilaris, connected with it through the intermediate form /esweuriana. It has the same relation to F. pilaris, as has F. kirtlandiana (Lea) of the upper Ohio drainage to F. subrotunda; it is practically the same thing, except _ that it does not reach the size of it, and does not develop the “ wing” seen on the posterior upper margin of kirtlandiana. It is perfectly clear that the form mentioned by Pilsbry & Rhoads from Watauga River, Johnson City, Tenn., is this, and, indeed, I have found it very close to this locality. The diameter of this form is less than 45 per cent. of the length, and it appears as if the lack in this direction is compensated by a greater circumference of the shell, bursa-pastoris often reaching a size (length and height) not seen in either pilaris or lesueuriana. The metropolis of this form is in Clinch River; a few specimens have been found in the lower Clinch, in Knox and Anderson Cos. (associated with leswewriana and even pilaris); farther up, from Claiborne Co., Tenn., into Virginia, it becomes more and more abundant, and north of the state line, where Jesweuriana disappears (only found at Clinchport), it is developed as a pure race, going up _into the headwaters as far as Cedar Bluff, Tazewell Co., Va. In Powell River, similar conditions prevail; in the lower Powell it is found associated and intergrading with Jesueuriana, but north of the 530 ORTMANN—NAYADES OF state line, i in Virginia, it is exclusively paceean going up to Bigs In the Holston, this form is not so well developed, pers no ce abundant, but is present nevertheless, and also has the same rela to lesueuriana, first being associated with it, but then being excl present. The latter is the case in North Fork Holston, in We t down as Grainger Co., so that in this region (as is the case i Clinch) occasionally all three forms of the species may be - together. a Type locality: Powell River, Va. (topotypes exaniiedine hee Note: The color of the soft parts and eggs in the pilaris ¢ varies greatly.. The soft parts may be of the orange type, eggs (and placente) may be red: this is chiefly the case in and upper Clinch rivers. Elsewhere, in Holston, French Bro Tennessee, pale soft parts, with whitish eggs prevail. © 5. FuscoNAIA CUNEOLUS (Lea), 1840. Unio cuneolus Lea, ’40.—Unio cuneatus Reeve, Conch. Icon. Unio, ’64, pl. 16, £. 73—Unio cuneolus Lewis, ’71—Unio olus Pilsbry & Rhoads, ’96.—Pleurobema cuneolus Simpso P. 743- This is not a Pleurobema, but a Fusconaia, cadet the plas group of F. flava (Raf.) (= rubiginosa Lea) in the upper Tent region. It is a very good species. Bar The original and typical cuneolus represents a compressed waters-form, which, farther down, changes into a more s' form. I have, arbitrarily, decided to call specimens with the diz less than 50 per cent. of the length by this name. This form is found in Powell, Clinch, and Holston rivers, in upper parts. In the Powell, from Union Co., Tenn., up to O Lee Co., Va. (also in Puckell Creek, Lee Co.) ; in the Clinch Anderson Co., Tenn., up to Clinchport, Scott Co., Va.;in the Holsto UPPER TENNESSEE DRAINAGE. 531 from Grainger Co., Tenn., up the North Fork at Mendota, Washing- ton Co., Va., and Big Moccasin Creek, Scott Co., Va. It has not been found in South Fork Holston, Watauga, or any of the eastern tributaries of the Tennessee, although the var. appressa goes to the lower Nolichucky and into the Little River drainage. In addition, it is in some of the tributaries of the lower Clinch: in Poplar Creek, Roane Co. (intergrading here with cuneolus ap- pressa) and in Emory River, at Harriman, Roane Co., Tenn. Type locality: Holston River, Tenn. (topotypes examined). 6. FUSCONAIA CUNEOLUS APPRESSA (Lea), 1871. Unio appressus Lea, ’71—Unio tuscumbiensis Lea, *71.—Unio flavidus Lea, ’71——Pleurobema tuscumbiensis Simpson, ’14, p. 748. There has been great confusion about this form, for the reason that the type-set in the U. S. Mus. (examined by myself) contains three specimens, of which only one, the figured type, is this, while the other two belong to Le-ringtonia dolabelloides conradi (Van.). But several authors seem to have taken the latter as the types, and thus, for instance, Pleurobema appressum Simpson (’14, p. 747) is not this, but Lexingtonia dolabelloides conradi. The figured type of Lea is a more swollen form of F. cuneolus, and resembles the latter in every respect except obesity. I have drawn the line between the two at the diameter of 50 per cent. of the length, and specimens with this diameter, and over, I call F. cuneolus appressa. This variety belongs to the larger rivers. I have not seen it from the Powell, and in the Clinch it turns up for the first time below the mouth of the Powell, at Offutt, Anderson Co., Tenn., but it goes into a smaller tributary of the lower Clinch, Poplar Creek, Roane Co., associated here with typical cuneolus. In Emory River, only cune- olus has been found. In the Holston, appressa turns up first at Austin Mill, Hawkins Co., Tenn., associated and intergrading with typical cuneolus. I also found F. cuneolus appressa in the Noli- chucky, at its mouth, and then it seems to be the prevailing form in the Tennessee at and below Knoxville. 532 ORTMANN—NAYADES OF Remarkably enough, I have found a single specimen of th in a very small creek tributary to Little River—Pistol Cree! ford, Blount Co., Tenn.—it has the diameter of 50 per cent.,. stands just on the line dividing the two varieties. This spe rather stunted in growth, and may be only an individual abne Type locality: Tuscumbia, Ala. (Tennessee ieee and H River, Tenn.) (type examined). a Note: This type of shell is also present in the Teanesses di in North Alabama, where again the appressa-form inhabits tl rivers, while in the smaller streams typical cuneolus is found, — 7. Fusconara cor (Conrad), 1834. Unio cor Conrad, ’34.—Unio edgarianus Lea, ’ 40—Unio Lea, ’71.—Unio edgarianus Lewis, ’71.—Unio andersonensis *72,—Unio edgarianus Pilsbry & Rhoads, bas. e edgarianum Simpson, ’14, p. 741. According to Frierson (Naut., 29, ’16, p. 102 ff.) the wee cor Conrad is the same as shells which have been called edg tuscumbiensis, andersonensis. Frierson has sent for inspectior specimen, which had been compared with the type of cor, and proved to be edgarianus. This is also supported by Conrad’s ¢ tion, which says that cor has rays, some of them broad. __ This species is closely allied in shell and anatomy to F. cune but is distinguished by very smooth and shining epidermis, ¢ beautiful color: upon a brownish or yellowish background are dark green to blackish rays, while cuneolus has greenish or ish-olive epidermis, with finer, greenish rays. The original cor is a much swollen form, and belongs to the rivers; but in the smaller streams, it again passes into a mor pressed form (F. cor analoga). I have drawn the line betwe two at the diameter of 50 per cent. of the length, so that spe with this or a greater diameter fall under cor. : I have found this form only in Clinch River, at Edgemoor, derson Co., Tenn., and, according to Pilsbry & Rhoads, it is Clinch in Roane Co. In the Walker collection, there are spe belonging here from Needham Ford, Union Co., Tenn. (assa UPPER TENNESSEE DRAINAGE. 533 with var. analoga). In the same collection are typical cor also from Poplar Creek, Roane Co., a rather small stream, where we should expect the headwaters-form. (It happens, sometimes, that a small tributary of a large river has, at or near its mouth, the large-river form.) Lewis reports this form from the Tennessee below Knoxville. Type locality: Elk River, Ala. (and Flint River, Ala.) (topotypes examined, loaned by Frierson). Note: Conrad’s Flint River is not the one in Madison Co., but the one in Morgan Co., Ala. 8. FUSCONAIA COR ANALOGA nov. var. Unio edgarianus Reeve, Conch. Icon. 16. Unio. 1864, pl. 15, f. 65.— Fusconaia appressa or edgariana Goodrich, ’13, p. 93. This flat form of cor (or edgarianus) has never been separated from the swollen typical form, probably because there never was any doubt about its affinity, on account of the peculiar color character of the shell. However, it differs from typical cor as strongly as does F. cuneolus from F. cuneolus appressa, and thus deserves a varietal name. The shell figured by Reeve as U. edgarianus surely is this, _ since the compressed shape is mentioned. Specimens with the diam- eter less than 50 per cent. of the length fall under this variety. This is the form of the headwaters and small streams. It is in the Powell, and goes up to Lee Co., Va. It is in the Clinch, from Needhams Ford, Union Co., Tenn. (Walker coll.) up to Cleveland, Russell Co., Va., and becomes quite abundant in the Virginian part of this river. In addition, it is in the North Fork Holston from Hawkins Co., Tenn., up to Holston, Washington Co., Va. _ It has never been found in any other part of the Holston drain- age, or any other river or creek in East Tennessee, and its restric- tion to North Fork Holston, Clinch, and Powell, and its continuation ‘ downward, as typical cor, in the Tennessee proper alone, is quite remarkable. Type locality: Clinch River, Speers Ferry, Scott Co., Va. (types in Carn. Mus. Cat. no. 61.6326). Note: The group of F. cor is also represented in the Tennessee 534 . ORTMANN—NAYADES OF drainage of North Alabama, and, also here, cor is in the Ten proper and the lower parts of the larger tributaries (Elk, both and Paint Rock Rivers), while in the upper parts, at least of Rock River, it passes into the analoga-form, Group OF FUSCONAIA BARNESIANA (Lea). Forms of this group have been listed by Lewis (’71) as L nesianus, pudicus, and tumescens. In Simpson’s papers (’o 14), they stand under Pleurobema, and their proper positio mutual relation have not at all been understood. All these shells belong to Fusconaia, but form a separate up within this genus, distinguished from the other species by very low beak cavities, and the peculiar color of the eggs and placent which is darker or lighter purplish-black, or some similar shade (s Ortmann, ’17). se Also here we have the phenomenon that flat and comp: forms are found in the headwaters, swollen forms in the 1 rivers, with the intergrades between them in the rivers of size. The transition is complete and very gradual, so that it tremely difficult to draw separating lines. The division into 1 also in general nailing development of beaks, and color mar I shall distinguish three forms: barnesiana bigbyensis, with diameter less than 40 per cent. of the length; barnesiana, with diameter from 40 to 49 per cent.; and barnesiana tumescens, the diameter of 50 per cent. and over. : g. FUSCONAIA BARNESIANA (Lea), 1838. Unio barnesianus Lea, ’38.—Unio meredithi Lea, ’58.—Unio p Lea, ’60.—Unio lyoni Lea, ’65——Unio barnesianus and pua Lewis, ’71—Unio tellicoensis Lea, ’72—Unio lenticularis ’°72,—Pleurobema barnesianum, pudicum, lenticulare, mer UPPER TENNESSEE DRAINAGE. 535 Simpson, ’14, pp. 754, 755, 790, 791.—Fusconaia barnesiana Ort- mann, ’17, p. 59. Simpson, ’14, p. 754, makes tellicoense a synonym of barnesianum, and (p. 755) lyont a synonym of pudicum. U. pudicus is practically the same as barnesianus, but with dis- tinct rays; meredithi is a pudicus with only few rays; U. lyoni is large and has rays; it also has somewhat elevated beaks, inclining thus toward the var. tumescens; tellicoensis and lenticularis are mod- erately large, with indistinct rays, practically identical with bar- nesianus. This form, which we must regard as the typical species, unfor- tunately represents the intermediate condition between the flat and swollen extremes. It is represented by shells from the Powell, Clinch, and Holston, going up here well toward the headwaters: in Powell to Big Stone Gap, Wise Co., Va.; in the Clinch to Richland, Tazewell Co., Va.; in the Holston to the North Fork at Holston, Washington Co., Va., and to the South Fork at Bluff City, Sullivan Co., Tenn. In the downstream direction, it can be traced to the Tennessee River below Knoxville. However, it is most abundant in lower Powell, in the middle Clinch, and in the Holston near the Forks. In addition, it turns up in many tributaries: Cove Creek, Campbell Co., Tenn.; Coal Creek, Anderson Co., Tenn.; Big Flat Creek, Knox Co.; in Little Pigeon River, Boyd Creek, Pistol Creek, Tellico River, Cane Creek (McMinn Co.), Hiwassee River, etc. It is a common form, of rather universal distribution under the proper conditions. Toward the headwaters, it passes generally into the var. bigbyensis, but there are some small creeks where this is not the case (at any rate, where the bigbensis-form has not been found). In the larger rivers it gradually passes into the form tumescens, and is often found associated with it. Type locality: Cumberland River, Tenn. (not found by Wilson and Clark, ’14, although they mention “Fleurobema crudum” == barnesiana tumescens; the latter, however, surely has been mis- identified). 536 ORTMANN—NAYADES OF 10, FUSCONAIA BARNESIANA BIGBYENSIS (Lea), 1841. Unio bigbyensis Lea, ’41.—Unio estabrookianus Lea, ’45.—U) sinans Lea, 68.—Unio fascinans Pilsbry & Rhoads, ’96.—P. bema fassinans rhomboidea Simpson, ’00,—Fusconaia est iana Goodrich, ’13, p. 93.—Pleurobema bigbyense, P. fas and var. rhomboideum, P. estabrookianum Simpson, ’14, pp 75€ 797, 798, 803.—Fusconaia barnesiana bighyensis — "17, P. 59. cats This is the form of the headwaters and small tributaries of : Tennessee system, with a diameter of less than 40 per cent. of length of the shell. U. bigbyensis and P. fassinans rhomboideum represent normal specimens, the former with distinct, the latter with — indistinct or missing rays. U. estabrookianus is founded upon v large, somewhat distorted specimens; and the only type (examined in Washington) of U. fassinans is an exceptionally a serie large specimen without rays. ie Also in this form we have the phenomenon that the co m: headwaters shell, so to speak, gains in circumference what it has lost in diameter. The var. bigbyensis grows much larger than forms farther downstream. It is very variable in size, shape, and sometimes it is hard to distinguish it from Pleurobema out argenteum. Generally, darker (brownish) color of epidermis, fine rays (when present), forming no blotches, and a more cer position of the beaks distinguish F. barnesiana tiger course, the anatomy is entirely different. i This race is very generally distributed over the whole | t Tennessee drainage, but disappears in the larger rivers. It oft in very small streams of the headwaters (North Fork Powell, other small streams in the Powell drainage; in the Clinch at Te well, Va.; Big Moccasin Creek, all three forks of the Holston, also Laurel Creek, Watauga River, and Little Pigeon River; CO n Creek at Knoxville; Sale Creek in Rhea Co.; Little River; Abr Creek in Blount Co.; Tellico River; Spring Creek, Polk Cans Cane Creek, McMinn Co.). . As has been stated, even at the uppermost localities, in very sm: creeks, F. barnesiana in its typical form may be present, and may UPPER TENNESSEE DRAINAGE. 537 associated with bigbyensis. Farther down, barnesiana soon begins to prevail. The form bigbyensis has been traced down in the Powell to Combs, Claiborne Co., Tenn.; in the Clinch to Clinchport, Scott Co., Va.; but a single specimen has been found at Solway, Knox Co., Tenn., apparently an exceptional case; in the Holston it has not been found below the point where the North and South Fork unite. Type locality: Big Bigby Creek, Maury Co., Tenn. (trib. to Duck River and lower Tennessee). II, FUSCONAIA BARNESIANA TUMESCENS (Lea), 1845. Unio tumescens Lea, ’45.—Unio crudus Lea, ’71.—Unio radiosus Lea, ’71.—Unio tumescens Lewis, ’72——Unio twmescens Pilsbry & Rhoads, ’96.—Pleurobema twmescens and P. crudum Simpson, "14, p. 751, 753.—Fusconaia barnesiana tumescens Ortmann, ’17, Pp. 59. (Simpson makes radiosus a synonym of tumescens.) This is the swollen form of the large rivers, with a diameter of 50 per cent. of the length, and over. It also generally has higher ‘beaks, making the outline of the shell more nearly triangular; but higher beaks are also sometimes observed in the typical F. barnesiana (lyoni-type). U. twmescens represents the extreme in obesity, and radiosus (type examined!) is very close to it. U. crudus is the much eroded form of the French Broad River (topotypes at hand). The two former have more or less developed rays, the latter is rayless. F, barnesiana tumescens has its metropolis in the Tennessee River at and below Knoxville. But it is also found some distance in the rivers above this point. From the Clinch it is known from the lower part; the uppermost locality is at Edgemoor, Anderson Co. It also has been reported (by Pilsbry and Rhoads) from Emory River at Harriman, Roane Co. In the Holston proper I have observed it as far up as Noeton, Grainger Co. It is also in the lower French Broad, at Boyd Creek, Sevier Co., in the Little Tennessee in Monroe Co., and in Hiwassee River, at Austral, Polk Co. (intergrading, at these places, with typical F. barnesiana). Remarkably enough, a small tributary of Little River, Pistol Creek, Rockford, Blount Co., Tenn., contains shells, most of which, PROC. AMER. PHIL. SOC., VOL. LVII, JJ, SEPT, 30, 1918. ars eee cat Kodo ics meascaations nae iat con arial alae neaas oh me) on ee ae BEC At he. tie ene it a eee oe, Ss Gee tae - eat eo ae a ee ae ce as lec ee Weebeiess ¥ 538 ; ORT MAN eo OF measurements > barnesiana, Type locality: “Alexandria, La.” This is surely ne Specimens from the Tennessee River in northern Alabama been recognized as this form by Call, Pilsbry and Rhoads, Bae ice fe River and Elk River (Estill Springs). In Shoals Creek, La dale Co., Ala., a very peculiar mixture is found, containing all types side by side, intergrading completely, but with Oi : barnesiana prevailing. : Genus: AMBLEMA Rafinesque, 1820. Crenodonta (Schlueter, ’36), Ortmann, 1912), p. 245.—Ambler Frierson, 19144, p. 7. 12. AMBLEMA PLICATA COSTATA (Rafinesque), 1820, Amblema costata Rafinesque, ’20.—Unio undulatus Lewae ee Unio undulatus Pilsbry & Rhoads, ’96—Crenodonta und; Ortmann, ’12), p. 246 (anatomy ).—Crenodonta undulata rich, 13, p. 93——Quadrula undulata Simpson, ’14, p. Amblema costata Frierson, ’14a, p.7.—Quadrula costata V; 15, p. 556.—Amblema (plicata) costata, Utterback, ’16, p. As to nomenclature, see Frierson and Vanatta (/. c.). Th tity of Raflnesque’s species has been recognized already by Conrz UPPER TENNESSEE DRAINAGE. 539 1834 and 1836, and the name of costata has been used by Kuester and Reeve. Moreover, the true U. undulatus Barnes, ’23, is actually U. heros Say, ’29. U. plicatus Say, 1817, is the Lake Erie form (=hippopeus Lea, ’45), and costata Raf. of the Ohio drainage, apparently is the ancestral form to this. But according to the laws of priority, plicatus, although being a local race, has to stand as the main species. Generally distributed in the upper Tennessee drainage, in Powell, Clinch, lower Emory, Holston, French Broad, Nolichucky rivers, lower Little River, and Tennessee proper. Goes up to Shawanee, Claiborne Co., Tenn., in the Powell; to Cleveland, Russell Co., Va., in the Clinch; to Hilton, Scott Co., Va., in North Fork Holston; and Pactolus, Sullivan Co., Tenn., in South Fork Holston; to the mouth of the Nolichucky at Chunn’s Shoals, Hamblen Co., Tenn. It also occasionally enters some rather small streams, for instance: Poplar Creek, Roane Co., and Boyd Creek, at Boyd Creek, Sevier Co., Tenn. : Type locality: Ohio River (Raf.) (according to Vanatta, the type is from small creeks in Kentucky). Genus: QuapDRULA Rafinesque, 1820. Ortmann, 1912), p. 250. 13. QUADRULA PUSTULOSA (Lea), 1831. Obliquaria bullata Refinesque, ’20.—Unio pustulosus Lea, ’31.— Unio pernodosus and U. pustulosus Lewis, ’71—Unio pustulosus and sphericus Pilsbry & Rhoads, ’96—Quadrula pustulosa Ort- mann, ’12b, p. 251 (anatomy).—Quadrula pustulosa Simpson, "14, p. 848. According to Vanatta (’15, pp. 556, 557), Obliquaria retusa Raf., ’20, “probably” is U. pustulosus Lea, ’31, and O. bullata Raf., ’20, is U. pernodosus Lea, ’45: but the latter is not different from pustu- losus. The identity of O. bullata and U. pustulosus has been asserted by Conrad in 1834, and is evident from Rafinesque’s description. Also Kuester and Reeve have used the name of bullatus. However, as Vanatta points out, the specific name bullata is preoccupied by 540 ORTMANN—NAYADES OF Obliquaria flexuosa bullata Raf., ’20 (Internat. Rules of Nomencl., Art. 11), and.since the identity of O. retusa is not: certain, the name given by Lea becomes available. ae I cannot see anything but an individual variation in what Lea ie called U. pernodosus. This is given from “ North Carolina,” acc ing to Lea probably from the tributaries of the Tennessee (i mountains). But its occurrence in these parts has never been con- firmed. U. sphericus (?), reported by Pilsbry & Rhoads from 1ez Chattanooga, surely is this. This species prefers the larger rivers, and is not rare is chiefly so in the Tennessee at and below Knoxville. I traced the Clinch to Offutt, Anderson Co., Tenn., but a single specimen is in the Walker collection also from the Powell, at Bryant SI oals, Claiborne Co., Tenn. In the Holston, it goes up to McBee Ford, near Hodges, Jefferson Co., and in the French Broad, it reaches the lower part of the Notichueky at Chunn’s Shoals, Hamblen es z Type locality: Ohio. ae 14. QUADRULA VERRUCOSA (Rafinesque), 1820. Obliquaria verrucosa Rafinesque, ’20.—Unio conjugans W Naut., 13, ’99, p. 89.~Quadrula tuberculata Ortmann, 712 254 (anatomy).—Tritogonia tuberculata and T. conjugans ‘Si son, ’14, pp. 318, 322.—-Tritogonia verrucosa Vanatta, ’15, { —Quadrula verrucosa Utterback, ’16, p. 62. 3 This form has been previously reported, as U. conjugans, yn} from Hiwassee River, and the type (examined in Washington) is indeed a remarkable shell, an apparently stunted, shortened male this species (individual abnormity). That this species is actuall present in the lower part of the Hiwassee, in Meigs Co., Tenn. shown by three fine specimens in the Walker collection (Adams It is very remarkable that I did not find a trace of this strik species in any other part of the upper Tennessee drainage. | Type locality: Ohio River. Note: Below the Walden Gorge, in Sequatchie River, Tenn, the Tennessee drainage in northern Alabama, this species is mo: abundant, both in the Tennessee River and its tributaries (Pz UPPER TENNESSEE DRAINAGE. 541 Rock River, Flint River in Madison Co., Elk River, Shoals Creek, Bear Creek). I5. QUADRULA METANERVA (Rafinesque), 1820. Obliquaria metanerva Rafinesque, ’20.—Unio metanerva Lewis, ’71. —Quadrula metanerva Ortmann, ’12b, p. 255 (anatomy).— Quadrula metanerva Simpson, ’14, p. 834. In the Tennessee at and below Knoxville. Above Knoxville, I found only a single specimen in the Holston, at Mascot, Knox Co. Rare. Type locality: Kentucky River (according to Vanatta, ’15, the types are from Ohio River). 16. QUADRULA INTERMEDIA (Conrad), 1836. Unio intermedius Conrad, ’36.—Unio tuberosus Lea, ’40—Unio sparsus Lea, ’41-——Unio intermedius, sparsus, tuberosus Lewis, *71.—Quadrula sparsa Ortmann, ’12b (anatomy )—Quadrula in- .termedia Goodrich, ’13, p. 93—Quadrula tuberosa, tuberosa sparsa, intermedia Simpson, ’14, pp. 836, 837. According to Simpson, Q. tuberosa is more swollen than Q. in- termedia, and I have seen rather swollen specimens in the Walker collection from the Cumberland River and the Tennessee in North Alabama. However, the original figure of U. tuberosus Lea is not much swollen. My specimens from the upper Tennessee drainage are all more or less compressed. We might have here again a case where in larger rivers a more swollen form turns up. But even if . this should be correct, we should only regard these forms as races ‘3 of the same species. Reported from Nolichucky, Holston, Clinch, and Tennessee ‘ rivers, and said to be abundant. According to the material collected and seen by myself, it is decidedly a rare species. I know it from the _ . Holston at Church Hill, Hawkins Co., Tenn.; from the South Fork ¢ Holston at Pactolus and Bluff City, Sullivan Co., Tenn.; ffom North ae Fork Holston at Mendota, Washington Co., Va.> and from the Clinch River at Clinchport, Scott Co., Va. (Walker coll.), and Cleveland, Russell Co., Va. 542 ORTMANN—NAYADES OF It is remarkable that I did not find it in the lower parts of thes rivers, although it has been reported from the Knoxville regia ni from the Tennessee in northern Alabama. 2 Type locality: Nolichucky River, Tenn. 17. QUADRULA CYLINDRICA (Say), 1817. Unio cylindricus Say,’17—Unio cylindricus Lewis,’71 —Unio dricus Pilsbry & Rhoads, ’96.—Quadrula cylindrica Ortman 12b, p. 256 (anatomy ).—Quadrula cylindrica Simpson, 4, { 832. . Reported from the Tennessee below Knoxville and ‘eon the Holston River. It is rather frequent in the larger rivers, anc in the Holston, up to Hawkins Co., Tenn. In the region of the fork it begins to incline toward the next form; however, in Big Mocassin Creek, Mocassin Gap, Scott Co., Va., a tributary of the North I found a rather normal cylindrica. In the Clinch, it goes to Clinch. port, Scott Co., Va. It is also in the Powell, up to Claiborne Co Tenn. Type locality: Wabash River. 18. QUADRULA CYLINDRICA STRIGILLATA (Wright), 1898 Unio cylindricus strigillatus Wright, Nautilus, 12, ’98, p. 6— rula cylindrica strigillata Ortmann, ’13), p. 311 Quatre oe " drica strigillata Goodrich, ’13, p. 93. Simpson (’14, p. 834) does not admit this as a separate | form but it is a very good and well-marked local race, belonging to headwaters, being compressed, devoid of large tubercles, but thi covered with small ones. Its metropolis is in the upper Clinch River. Intergrades betws this and the main fotm are found in Scott Co., Va., and thence ther up, it becomes a pure race, and goes up to Cedar Bluff in T: well Co., Va. The same conditions prevail in Powell River, v it intergrades with the normal type in Claiborne Co., Tenn., and goes up to Pennington Gap in Lee Co. It also exists in North Holston, from Hawkins Co., Tenn., to Mendoto, Washington Va., and in the South Fork Holston at Pactolus, Sullivan Co., T n UPPER TENNESSEE DRAINAGE. 543 But the specimens from the Holston drainage are not so well and typically developed as those from the upper Clinch. Type locality: “Clinch River, Lee Co., Va.” A case of inexcus- able carelessness, for the Clinch never touches that county. Genus: RotunpDarIA Rafinesque (1820). Ortmann, 1912), p. 257. 19. ROTUNDARIA TUBERCULATA (Rafinesque), 1820. Obliquaria tuberculata Rafiesque, ’20.—Unio verrucosus Lewis, *71.—Unio verrucosus Pilsbry & Rhoads, ’96.—Rotundaria tuber- culata Ortmann, ’12b, p. 258 (anatomy)—Quadrula (Rotun- daria) tuberculata Simpson, ’14, p. 903. Abundant in the larger rivers, going rather far up toward the headwaters. The extreme points are: Clinch River, Clinchport, Scott Co., Va.; North Fork Holston River, Mendota, Washington Co., Va.; South Fork Holston River, Pactolus, Sullivan Co., Tenn. ; Nolichucky River, Chunn’s Shoals, Hamblen Co., Tenn. I have it also from Boyd Creek at Boyd Creek, Sevier Co., Tenn., a very small tributary of French Broad. In the other eastern tributaries of the Tennessee (Little River, Little Tennessee, and Hiwassee) it has not been found, and also no records are at hand from Powell River. Type locality: Ohio River. Genus: PLETHOBASUS Simpson (1900). Ortmann, 19120, p. 259. 20, PLETHOBASUS COOPERIANUS (Lea), 1834. Unio cooperianus Lea, ’34—Unio cooperianus Lewis, ’71.—Unio cooperianus Pilsbry & Rhoads, ’96.—Plethobasus cooperianus Ortmann, 12), p. 261 (anatomy).—Quadrula cooperiana Simp- son, ’14, p. 852. In the Tennessee River at and below Knoxville, down to Chatta- nooga (Pilsbry & Rhoads). Also in the lower Clinch River; I have 544 ORTMANN—NAYADES OF it from Edgemoor, Anderson Co., Tenn., and Pilsbry and R it from Patton’s Ferry, Roane Co., Tenn. I also have fo French Broad River, at Boyd Creek, Sevier Co., Tenn. from “ Holston River” probably refer to the Tennessee, at any it must be a rare shell above Knoxville. ‘ Type locality: Ohio River. 21. PLeTHOBASUS CyPHyYUs (Rafinesque), 1820 "15, p- 556 In the larger rivers, Tennessee, Clinch, Powell, Ho French Broad, going up, in the Powell, to Bryant Shoals, Cl. Co., Tenn. ; in the Clinch to PB Scott Co., Va.; in Tenn. Type locality: Falls of the Ohio. ce 3 (middle and inves ais .—Pleurobema compername! Sir 14, p. 809. : Specimens in the Walker collection from “ Holston River, Kr ville,’ are authentically labeled varicosus Lewis, and comp. 1 Frierson, and they agree very well with specimens collected br self in French Broad River, at Boyd Creek, Sevier Co., Tenn. specimens (I have five) are larger than Frierson’s figures, most of them are more drawn out posteriorly. They were foun sociated with one specimen of the true P. cyphyus, and resemble latter very much, except that the-radial row of knobs on the is poorly developed, almost obliterated, and that the soft par not of orange, but of pale color. (For the rest, the anatomy of and female is like that of P. cyphyus.) I consider this as a variety of P. cyphyus, since in some specimens, the row of knobs is more distinct, and since UPPER TENNESSEE DRAINAGE. 545 cyphyus often has these knobs nearly obliterated. Whether the soft parts are always pale remains to be seen (the specimen of typical cyphyus from the same locality had orange soft parts): but several other shells have, in the French Broad, the tendency to develop pale soft parts, or pale nacre, while they are tinted elsewhere. This seems to be a rare shell, which has been reported by Lewis from the Tennessee below Knoxville, by Frierson from the “ Clinch and Holston”; Walker has specimens from “ Holston, Knoxville” (probably Tennessee), and I found it in the lower French Broad. It might be that this is a local race, restricted in its distribution, and possibly with its center in the lower French Broad: but more mate- rial with exact localities should be secured. Type locality: Clinch and Holston rivers. Genus: LEXINGTONIA Ortmann (1914). Ortmann, 1914, p. 28. 23. LEXINGTONIA DOLABELLOIDES (Lea), 1840. Unio dolabelloides Lea, ’40——Unto thorntoni Lea, ’57-——Unio moore- sianus Lea, ’57.—Unio recurvatus Lea, ’71-—Unio circumactus Lea, ’71—Unmio subglobatus Lea, ’71—Unio dolabelloides and mooresianus Lewis, ’71.—Pleurobema dolabelloides Simpson, ’14, P. 752. U. thorntoni, mooresianus, recurvatus, circumactus, and subglo- batus have been recognized by Simpson as synonyms of this. Also here we have a case where a swollen form (dolabelloides) is found in the larger riyers, and a compressed one (conradi) in the smaller streams, with the intergrades existing between them. I have drawn the line between the two at the diameter of 50 per cent. of the length, so that forms with the diameter 50 per cent. or over are dolabelloides, and those below 50 per cent. are conradi. L. doladelloides in its typical development is a swollen form, gen- erally also with more elevated beaks. It is known from the Ten- nessee River below Knoxville (Lewis), down to Rathburn, Ham- ilton Co. (Walker collection), and from the lower Clinch, up to Agee, Campbell Co. (where it intergrades with conradi) ; and it is in French Broad River, at Boyd Creek, Sevier Co. 546 ORTMANN—NAYADES OF It is remarkable that this form has not been found in the proper, from Knoxville up to the forks, while the beat rt is both in the North and South Fork Holston. Type locality: “ Holston River, Tennessee,” which stands eae ently for Tennessee River. 24. LEXINGTONIA DOLABELLOIDES CONRADI (Vanatta), 1915. Unio maculatus Conrad, ’35.—Pleurobema maculatum Goodric p. 94.—Pleurobema maculatwm Simpson, ’14, p. 737—Pleur bema appressum Simpson, ’14, p. 747. —Pleurobema Vanatta, ’15, p. 559. A form, largely misunderstood. Specimens ot this hake 0 been called U. appressus Lea, because two specimens are in the U Nat. Mus. with the figured type of appressus, but are different from the latter. Also Simpson’s P. appresswm (’14, p. 747) undoubt dy is this, since he quotes Sowerby’s figure of U. argenteus (Co icon. 16. Unio. ’66, pl. 37, f. 204), which is a fine ra f this form. 4 This variety often resembles very much Pleurobema cee (Conr.) (chiefly such forms which have been called clinche Lea). However, it may be recognized by the subtriangular out more forwardly inclined beaks, and the more distinct truncati the posterior slope. The most important character, however, is in the soft parts: in the present form, these are generally orange, with the outer gills marsupial, filled, when charged, red, subcylindrical placente. These are the characters of the -Lexingtonia. The main species has the same characters, at ! the Clinch; in the French Broad, however, I have found spe with pale soft parts. But the same tendency has been noticed other species from the French Broad. An abundant form in the headwaters of Powell, Clinch, No and South Fork Holston. In the Powell, it goes up to Big Gap, Wise Co., Va., and is also in Puckell Creek at Pennington G Lee Co., Va. In the Clinch, it is found up to Cedar Bluff, Taz Co., Va., and down to Clinton, Anderson Co., Tenn., interg in the lower part with typical dolabelloides. It is everywhere UPPER TENNESSEE DRAINAGE. 547 North Fork Holston up to Saltville, Smyth Co., Va., and in the South Fork Holston at Fish Dam (Walker coll.), Emmett, and Bluff City, Sullivan Co., Tenn. However, it does not go down the Holston proper (see above). Type locality: (of U. maculatus Conr.) Elk and Flint rivers, northern Alabama (Conrad’s Flint River is in Morgan Co., Ala., and is different from the Flint River in Madison Co., Ala.). Note: This group of forms is also abundant in the Tennessee drainage in northern Alabama. The typical dolabelloides is found in the Tennessee proper, and also in lower Paint Rock River, Flint River (Madison Co.), and Limestone Creek; while the conradi-type is found in smaller streams, for instance, in the headwaters of Paint Rock River, the headwaters of Flint River (Madison Co.), Flint River (Morgan Co., according to Conrad), and in Elk River (Conrad). . Genus: PLEUROBEMA Rafinesque (1820). Ortmann, 19120, p. 261. 25. PLEUROBEMA OBLIQUUM (Lamarck), 1819. Unio obliqua Lamarck, ’19—Unio obliquus Pilsbry & Rhoads, ’96.— Pleurobema obliquum Ortmann, ’12), p. 264 (anatomy ).—Quad- rula obliqua Simpson, ’14, p. 881. This consists of a group of forms very variable in shape, which has been divided into a number of “species.” In the upper Ten- nessee region several of the latter are found, but they all intergrade with each other, and there is very little indication of their separation into geographical or ecological races. Mostly, the various forms are found associated, so that they are hardly more than individual variations. However, in deference to the nomenclature hitherto accepted, and in view of the fact that in other regions some of these variations _ become local varieties, I have kept these forms apart. The typical P. obliquum is rather upright, with a distinct radial furrow, and is.subtriangular in outline. The nacre is generally white. This form is quite abundant in the larger rivers, Tennessee, 548 | ORTMANN—NAYADES OF Clinch, Holston, and French Broad. It goes up, in the Union Co., Tenn.; in the Holston, to Grainger Co., Tenn. Tennessee proper it is known down to Chattanooga. It a strange that it is absent in Lewis’s list. In specimens from the upper section of Holston and Clinch, radial furrow is quite shallow. Type locality: Ohio River. 26. PLEUROBEMA OBLIQUUM CORDATUM (Rafinesque), 1820. Obovaria cordata Rafinesque, ’20——Unio plenus Lea, *4c plenus Lewis, ’71—Quadrula plena Simpson, ’14, p. 886 rula cordata Vanatta, *15, p. 558.—Pleurobema i Utterback, ’16, p. 77. I accept the nomenclatural change introduced by Vasil Rafinesque’s description and figure can very well be referred to form. ye Upright, more rounded, and more elevated than the norm: fi radial furrow less developed. This is the most poorly marked form of the group, and i with the main form all over its range, representing merely an vidual variation. In the upper Tennessee region it is rath Tyee locality: Ohio River. 27. PLEUROBEMA OBLIQUUM CATILLUS (Conrad), 1836. Unio catillus Conrad, ’36.—Unio solidus Lea, °38.—Quadrula so . Simpson, ’14, p. 885.—Pleurobema obliquum catillus Uttert ck. ’16, p. 79. | Subtriangular, rather swollen, with the radial furrow oblit or absent. Nacre white or reddish. Individual variation of the main form, all over its range, b rare in the upper Tennessee; there are mighty few specimens show the characters of this form well developed. In other regions (upper Ohio, and west of the Mississip form assumes frequently the character of a local race, in fa of the Mississippi, this, and forms like P. obliquum rubrum, pr UPPER TENNESSEE DRAINAGE. 549 _ while the form coccineum is scarce, and the typical obliquum is absent. Type locality: Scioto River, Ohio. 28. PLEUROBEMA OBLIQUUM COCCINEUM (Conrad), 1836. Unio coccineus Conrad, ’36.—Pleurobema coccineum Ortmann, *12b, p. 263 (anatomy)—Pleurobema sp. ? Goodrich, 713, p. 94.— Quadrula coccinea Simpson, ’14, p. 883. A compressed form, typically merely a compressed catillus, with the radial furrow absent. Such forms have been reported hitherto only once from the upper Tennessee by Call (from “Holston River”). I have found only a few of them, corresponding entirely to the coccineum of the upper Ohio drainage in Pennsylvania; in the Clinch at Solway, Knox Co., Tenn., and in the Holston at Hodges, Jefferson Co., and at Noeton, Grainger Co., Tenn. They stand very close to the catillus forms of this region, representing merely an individual variation of it. Type locality: Mahoning River, near Pittsburgh (= Mahoning River, Lawrence Co., Pa.). | In addition to the above form, there is, in the upper Clinch, a very peculiar form of this group, not found elsewhere, which may be described as a compressed obliquum, with traces of the radial furrow still present. I have seen the soft parts of this (including — a gravid female), so that there is no doubt about the affinities of this shell. This form requires further study, and might deserve a varietal name, for it is found, in the Clinch, at, and a good deal above, the upper limit of P. obliquum. In the Walker collection there are several such specimens from Needham’s Ford, Union Co., Tenn., and I have it from Clinchport, Scott Co., Va., and from Cleveland, mer Xe» Va. This form may be more abundant in the poorly known portion of the Clinch from Claiborne and Grainger Cos., through Hancock Co., Tenn., to the Virginia state line. 550 ORTMANN—NAYADES OF 29. PLEUROBEMA OBLIQUUM RUBRUM (Rafinesque), 1820. Obliquaria rubra Rafinesque, ’20.—Unio pyramidatus Lea, Unio mytiloides Lewis, ’71.—Unio pyramidatus Pilsbry & Rhoa ’96.—Pleurobema pyramidatum Ortmann, ’12b, p. 264 (anatomy —Quadrula pyramidata Simpson, ’14, p. elie r Vanatta, ’15, p. 557. ay peat to Vanatta, the type of Obliquaria rubra Raf. adobe (which is EES this nomenclatural change shor be accepted. | Shell oblique, with high, forwardly inclined beaks. Radial row more or less developed. Nacre generally red. ni Although, in its typical phase, quite distinct from normal i quum, intergrades do exist, and also intergrades toward cor atun and catillus have been observed. - Also this form generally accompanies P. obliquum, but b Clinch and Holston it ascends the rivers a little farther. In Clinch, it goes to Oakman, Grainger Co., Tenn., and in the Holst to Austin Mill; Hawkins Co., Tenn. It is sbundiaae and well de oped in this region, and of all the forms of the ae group, has the best claim to be regarded as a local race. c In the upper Ohio, this form is quite rare, and often not West of the Mississippi are peculiar forms of it, often with thes furrow quite obliterated. Type locality: Kentucky River. GROUP OF PLEUROBEMA OVIFORME. Also here we meet with a group in which the rule holds ¢ that flat forms of the headwaters are represented, fate stream, by more swollen forms. Generally speaking, the oviforme group represents, in the Tennessee, the P. clava of the Ohio drainage, but it is much variable, and has developed, in the headwaters, a very peculiar pressed type, which does not find a parallel in the upper Ohio s All these forms have the characteristic Pleurobema anatomy, UPPER TENNESSEE DRAINAGE. 551 resemble very much in this the P. clava, although some peculiar variations are observed in the color of certain parts. As it happens, the oldest name (oviforme) has been used for an _ intermediate form, and thus, in order to retain this in its original sense, it seems advisable to distinguish three types within this species: the headwaters form (argenteum) has a diameter of less than 40 per cent. of the length; the form of the medium sized rivers (oviforme) has a diameter from 40 to 49 per cent.; and the big tiver form (holstonense) has 50 per cent and over. Also here we see the phenomenon that the headwaters form gains in circumference what it has lost in obesity. The argenteum type shows this possibly to the greatest extent; shells of this form reach in length and height dimensions entirely unknown in the more swol- len forms. 30. PLEUROBEMA OVIFORME (Conrad), 1834. Unio oviformis Conrad, ’34—Unio ravenelianus Lea, ’34.—Unio patulus Conrad, ’38 (not patulus Lea, ’29)—Unio lesleyi Lea, -*60——Unio ornatus Lea, ’61—Unio clinchensis Lea, ’67—Unio conasaugaensis Lea, ’72—Unio clinchensis, lesleyi, patulus Lewis, *71.—Pleurobema oviforme Goodrich, 713, p. 94——Pleurobema clinchense, lesleyi, oviforme, ornatum, conasaugaense, raveneli- anum Simpson, ’14, pp. 743-800. This resembles much the upper Ohio form of P. clava, but is less cuneate, with the beaks less anterior. It is extremely variable in shape, higher or more elongate, and the color pattern is hardly ever alike in any two individuals. In the Walker collection are topotypes of U. ravenelianus Lea (from Asheville): six specimens have the diameter of oviforme (40-49 per cent., while one has the diameter of 38 per cent. and the latter would thus fall under argenteum. But it has the typical shape of oviforme. Since the type of ravenelianus has 43 per cent., accord- ing to Simpson, we should place this here, and these shells are, indeed, nothing but oviforme without rays, of a general dull color (pale brownish, not yellowish). Such specimens are found else- where, and several sets in the Walker collection, from Poplar Creek, 552 ORTMANN—NAYADES OF Roan Co., from the Little Tennessee, Monroe Co., and Cane Cr oviforme ravenelianum). The figure of U. conasaugaensis Lea is an witinnd diameter, but is an argenteum according to shape and size. topotypes and authentic material of conasaugaensis, examined ino Walker collection, fall partly under oviforme and a se Grgentew, this i isa real be hard to place. i Little Taniensee cad in many of their tribetarien passing i s lower parts of the larger rivers into oviforme holstonense, and i the upper parts into oviforme argenteum. It is to be noted that ovifo goes well into the headwaters, and apparently it does not pass argenteum in some instances. This is chiefly the case in Broad River, where it goes into the North Carolina mountains ( ville) without assuming the characteristic features of argenteum. Type locality: Tennessee. 31. PLEUROBEMA OVIFORME ARGENTEUM (Lea), 1841. Unio argenteus Lea, ’41—Unio striatissimus Anthony, Am. Conch., 1, ’65, p. 155.—Unio planior Lea, ’68.—Unio brevis ’72.—Unio argenteus Lewis, ’72—Unio swordianus W Naut., 11, ’97, p. 4—Pleurobema fassinans Ortmann, 71 310 (not fassinans of Lea)—Pleurobema argenteum Go 13, p. 94.—Pleurobema swordianum, argenteum, breve, ple Simpson, ’14, pp. 757-802 —Pleurobema [ane Ortmann, p. 31 (per erroreum). The soft parts of this form have been described by me under erroneous name of P. fassinans. UPPER TENNESSEE DRAINAGE. 553 Simpson (p. 804) makes U. striatissimus Anthony a synonym of P. estabrookianum (which actually is Fusconaia barnesiana big- byensis). Specimens from Blount Co., Tenn., received from the Alabama Museum of Nat. Hist. as striatissimus, and similar ones with the same label in the Walker collection, agree fully with speci- mens collected by myself in Little River in Blount Co., and are this form. : Walker has four specimens labeled P. swordianum (Wright) from the Wright collection, which thus are authentic specimens. They are all typical P. oviforme argenteum. In addition, he has three others from the Sword collection (original lot), of which two are this form, while the third is Fusconata pilaris bursa-pastoris. The type of swordianum has, according to Simpson, the diameter of 40 per cent., and thus would stand under oviforme. -This, how- ever, seems to be an extreme specimen. This is the compressed form of oviforme, peculiar to the head- waters and other small streams. It also generally attains a larger size than the typical oviforme, and is more rhomboidal in outline. Lea’s only type of U. argenteus (examined by myself in Washing- ton) is not a normal specimen; it is tapering behind, which is a character of ovtforme. U. planior represents the normal shape of this shell, rhomboidal, while U. brevis is.exactly the same thing, only slightly shorter. The color markings are generally less bright than in oviforme, and very often they are obscure or missing, chiefly in old shells. _ This variety is found in Powell River from Big Stone Gap, Wise Co., Va. (where it alone is present), downward (associated with oviforme) ; in the Clinch, from Tazewell Co., Va., down to Kyle Ford, Hancock Co., Tenn. (also associated with oviforme and inter- grading with it). In the Holston drainage it is pure in Big Mocassin Creek, and in the North Fork at Saltville, Smyth Co., and Holston, Washington Co., Va. It is also pure in the Middle Fork at Chil- howie, Smyth Co., in the South Fork at Barron, Washington Co., Va., and in Watauga River-at Watauga, Carter Co., Tenn. Farther down, it passes into, and is associated with, oviforme, but has not been found in the Holston proper. It is in Little Pigeon River, at Sevierville, Sevier Co., Tenn., but not very well developed here, the PROC. AMER. PHIL. SOC., VOL. LVII, KK, OCT. I, 1918. 554 ORTMANN—NAYADES OF majority of the specimens belonging to oviforme. In Little Ri at Walland and Melrose, Blount Co., Tenn., it is well dev 0 ed, pure, and not accompanied by oviforme. The Little River form 1a: been called striatisswmus, and it has one peculiarity in - soft parts; they are of the orange type, and the placentz are red. . forme argenteum is also in the tributaries of Hiwassee Ri ; Conasauga Creek with oviforme), and in South Chickamauga C: at Ringgold, Catoosa Co,, Ga. The latter specimens have, in the average, a greater Meaokied: resembling that of oviforme, but they are stunted in growth (truncated behind). A few of normal shat are clearly argenteum. In the larger rivers, this form is missing. _ Type locality: Holston River, Tenn. (not recently aise in f ol- ston proper). ess 32. PLEUROBEMA OVIFORME HOLSTONENSE (Lea), 1840. | Unio holstonensis Lea, ’40.—Unio mundus Lea, ’57—Unio tes- serule Lea, ’61.—Unio pattinoides Lea, ’71—Unio acuens Lea, ’71.—Unio lawi Lea,’71.—Unio holstonensis and tesserule Lewi: *71.—Unio bellulus Lea, ’72.—Unio acuens and lawi Pilsb: Rhoads, ’96.—Pluerobema holstonense, acuens, and tess Simpson, ’14, pp. 739, 749, 749. U. mundus, lawt, pattinoides, and bellulus have already been r ognized by Simpson as synonyms of P. holstonense. . This is the swollen, larger river form, of oviforme. U. hahetome with regard obesity. : This form has been reported from the Tennessee below Kuo ville, and down to the mussel shoals in northern Alabama; and ther from the lower Clinch, the lower Emory River, and the chucky. According to the material examined by the writer, it up, in the Clinch, to Edgemoor, Anderson Co., Tenn.; in is ston, to Mascot, Knox Co., Tenn.; it is in French Broad at Creek, Sevier Co., Tenn.; in Little Tennessee River, Mourest Ce Tenn. (Walker coll.); and in the Hiwassee at Austral, Polk Tenn. At all these points, near its upper limit, it is associated intergrades with oviforme. UPPER TENNESSEE DRAINAGE. 555, Type locality: Holston River, Tenn. (Simpson says: Tuscumbia, Ala.) (topotypes examined). Note: The group of Plewrobema oviforme is also quite abundant in the Tennessee drainage in North Alabama. Also here the hol- stonense-form is in the Tennessee proper (also in lower part of Paint Rock River and Limestone Creek) ; in the tributaries, oviforme is the prevailing form, and in some of the headwaters, the argenteum- form is fully as well developed as in the upper Clinch and Holston, for instance: in Paint Rock River at Princeton, Jackson Co., Ala. ; Dry Creek, Holly Tree, Jackson Co., Ala. (tributary to Paint Rock) ; in Hurricane Creek, Gurley, Madison Co., Ala. (tributary to Flint River) ; in Elk River, Estill Springs, Franklin Co., Tenn., and its tributary: Boiling River, Cowan, Franklin Co., Tenn. It should be remarked that in this region also the true Pleuro- bema clava (Lam.) turns up; this species is distinguished by much more anteriorly situated, pointed beaks, swollen anterior part of the shell, and cuneiformly compressed posterior part. The Carnegie Museum has this species from the old Smith collection, labeled Tus- cumbia, Ala., and I have seen it also in the Walker coll. from Florence, Ala. In addition, in the latter collection, are specimens. from Sequatchie River, at Jasper, Marion Co., Tenn. (Wetherby coll.). Genus: Extiptio Rafinesque (1820). Ortmann, 1912), p. 265. 33. ELLIPTIO NIGER (Rafinesque), 1820. Unio nigra Rafinesque, ’20.—Unio crassidens Lewis, ’71.—Unio cras- sidens Pilsbry & Rhoads, ’96.—Elliptio crassidens Ortmann. ’12), p. 266 (anatomy ).—Unio crassidens Simpson, ’14, p. 606.—Unio crassidens Vanatta, °15, p. 555.—Elliptio nigra, Utterback, ’16, p. 88. -The identity of U. migra Raf. is now firmly established (Say, Conrad, Kuester, Sowerby, and Vanatta), and it is the species com- monly called crassidens. However, the type of crassidens Lamarck, 1819, is not this, but is the trapezoides Lea (Frierson, ’14a, p. 7). _ Thus Rafinesque’s name should be used for the present species. 556 ORTMANN—NAYADES OF Everywhere in the larger rivers: Tennessee, Holston, Frenc Broad, Clinch, and Powell; goes up, in the Powell, to J Lee Co., Va.; in the Clinch, to Clinchport, Scott Co., Va. n Holston to the South Fork at Pactolus, Sullivan Co., Tenn.; in French Broad, it goes to the Nolichucky at its mouth( Ch Shoals, Hamblen Co., Tenn.). It is also in Emory River Harriman Junction, Roan Co., Tenn., and in Hiwassee Riv Kincannon Ferry, Meigs Co., Tenn. In the Tennessee below ville, and down to Chattanooga, it is extremely abundant. 3 Type locality: Ohio River. 34. Extrprio piLatatus (Rafinesque), 1820. Unio dilatatus Rafinesque, ’20.—Unio gibbosus Lewis, 7 —Un gibbosus Pilsbry & Rhoads, ’96.—Elliptio gibbosus ’'12b, p. 271 (anatomy ).—Elliptio dilatatus Ortmann, ’13b, p —Elliptio gibbosus Goodrich, ’13, p. 94.—Unio gibbosus Sit "14, p. 597.—Unio dilatatus Vanatta, ’15, p. 555 —Ellip Utterback, ’16, p. go. a Common, in large rivers as well as in small creeks, pos most widely distributed species in the upper Tennessee reg that it is heardly required to name special localities. However should be mentioned that it is one of the gee which ; more frequently hits unica. Type locality: Ohio River (the type is bie Kentucky F cording to Vanatta). Genus: LAsTENA Rafinesque (1820). Ortmann, ’12b, p. 297, and ’I5, p. 106. 35. LASTENA LATA (Rafinesque), 1820. Anodonta lata Rafinesque, ’20.—Margaritana dehiscens Lewis Lastena lata Goodrich, ’13, p. 94.—Lastena lata Ortmann, 106 (anatomy ).—Lastena lata Vanatta, ’15, p. 554. UPPER TENNESSEE DRAINAGE. - 557 Reported from Tennessee River, below Knoxville (Lewis), and : : the Holston River (Call). I have never found it in the Holston, but only in the Clinch, at Edgemoor and Clinton, Anderson Co., Tenn.; at Oakman, Grainger Co., Tenn. In the Walker collection it is from - Clinch River at Clinchport, Scott Co., Va.; and I collected it still farther up at St. Paul, Wise Co., and Cleveland, Russel Co., Va. It ’ is undoubtedly a rare shell. Type locality: Kentucky River. Subfamily: ANODONTINZ Ortmann. Ortmann, 1910, p. I17. Genus: LAsMIGONA Rafinesque (1831). Symphynota Lea (1829), Ortmann, ’120, p. 280 —Lasmigona Frier- son, *14}, p. 40. 36. LASMIGONIA (SULCULARIA) BADIA (Rafinesque), 1831. Alasmodon badium Rafinesque, ’31—Margaritana holstonia Lewis, *71.—Symphynota holston(ia) Goodrich, 13, p. 94.—Alasmidonta holstonia Simpson, ’14, p. 502—Symphynota (Sulcularia) badia Frierson, ’14a, p. 7——Symphynota (Alasminota) holstonia Ort- . mann, ’I4, p. 43 (anatomy). The subgenus Alasminota Ortmann, ’14, p. 42, is synonym to Sulcularia Rafinesque, ’31 ; see Frierson, 1. c. A characteristic small stream species, abundant locally, and re- ported by Lewis from small streams in Monroe Co., Tenn., _ by Call from tributaries of the Holston in East Tennessee. The largest streams where I have seen it are the upper Holston proper at Church Hill, Hawkins Co., Tenn., and the Hiwassee, at Austral, Polk Co., Tenn. In both cases the specimens came from a small slough. In the headwater streams, it is in upper Powell, in Va., in the uppermost Clinch in Tazewell Co., Va.; in Little Mocassin Creek, Scott Co., Va.; South Fork Holston at Bluff City, Sullivan - Co., Tenn.; Watauga River, Carter Co., Tenn. Other small streams are the following: Cove Creek, Campbell Co., Tenn. (to Clinch) ; Bull Run, Knox Co., Tenn. (to Clinch) ; 558 ORTMANN—NAYADES OF Big Creek, Hawkins Co., Tenn. (to Holston) ; Long Creek, Cocke Co., Tenn. (to French Broad); Little Pigeon River, Sevier ¢ Oxy Tenn. (to French Broad) ; First Creek and Third Creek, Knox Co., Tenn. (to Tennessee); Piney River, Rhea Co., Tenn. (to Ten- nessee) ; Conasauga Creek, Monroe Co., Tenn. (to Hiwassee ) South Chickamauga Creek, Catoosa Co., Ga. (to Tennessee), Thus this species has practically a anivertal cutie in nour region, but it strictly avoids larger streams. as Type locality: Small streams of the Knobs, Kentucky (ead waters region of Cumberland and Kentucky Rivers). 37. LASMIGONA (LASMIGONA) cosTATA (Rafinesque), 1820, _ Alasmidonta costata Rafinesque, ’20——Margaritana rugosa Lewis, ’71.—Alasmodonta rugosa Pilsbry & Rhoads, ’98.—Symphynota costata Ortmann, ’12b, p. 283 (anatomy).—Symphynota costata * Ortmann, ’13), p. 311.—Symphynota costata Goodrich, ’13, P. 4 —Symphynota (Lasmigona) costata Simpson, ’14, p. 488. — A common species, most abundant in small and motel fee streams, rarer in the larger rivers, but not absent there (Tennessee, lower Clinch and Holston). Goes up, in the Powell, to Olinger, Co., Va.; in the Clinch, to Cedar Bluff, Tazewell Co., Vas Ei in to Tolisnae City, Wathivictos Go. Tenn. It is aioe in ‘South Chic - mauga Creek, Ringgold, Catoosa Co., Ga. =: Type locality: Kentucky River. Genus: ANoponta Lamarck (1799). Ortmann, ’12), p. 286. 38. ANODONTA GRANDIS GIGANTEA (Lea), 1834. Anodonta gigantea Lea, ’34.—Anodonta grandis (incl. gigani Ortmann, 712), p. 292 (anatomy).—Anodonta grandis gigas Simpson, ’14, p. 420. : No Anodonta has ever been reported from the upper Tenne UPPER TENNESSEE DRAINAGE. 559 region. However, in the collection of B. Walker, there are two large specimens of a form of Anodonta grandis Say, collected by Mr. M. D. Barber, of Knoxville, in a “small mud pond near French Broad River, 8 miles above Knoxville.” The largest has a length of 168 mm., the other of 127 mm. In the latter, the beak sculpture is vis- ible, and corresponds to that of A. grandis; it has white nacre, and much resembles specimens of the var. gigantea Lea, as found in western Pennsylvania. The other (larger) is a little distorted and tapering behind, and has pale purple nacre, resembling in these char- acters some of the southern forms of the species. The occurrence of this form in this isolated locality is quite re- markable, but supports the view that Anodontas may possess excep- tional means of dispersal. According to more detailed information obtained by Mr. Walker in 1916 from Mr. Barber, the pond is - “plainly natural, some 4 or 5 rods long by 2 rods wide, with soft deep mud. There was a small stream of water running to the French Broad River, just a short distance. This was Io or II years ago.” “Two years ago, I took a nephew and found the same pond, but although we waded the pond in every direction, up to our knees in mud, we could not find even a fragment of gigantea. A man liv- ing near there said he had seen large shells in another pond near, but I have not examined it.” These ponds, a short distance from the river, probably are on the flood plain of French Broad; the writer has not been able to locate them on the U. S. Geol. Surv. maps. Type locality: Port Gibson, Claiborne Co., Miss. Genus: ANODONTOIDEs Simpson (1898). Ortmann, 12), p. 294. 39. ANODONTOIDES FERUSSACIANUS (Lea), 1834. Anodonta ferussaciana Lea, ’34——Anodontoides ferussacianus Ort- .mann, ’12b, p. 294 (anatomy)—Anodontoides ferussacianus Simpson, ’14, p. 467. Anodonta oblita (?) Lewis, 71, may possibly stand for this species. Lewis doubtfully reports this from the Tennessee below Knoxville, but this probably is a mistake. Originally, A. oblita Lea - 560 ORTMANN—NAYADES OF and Anodonta denigrata Lea, which are this species, have been de scribed from Campbell Co., Tenn.: this is surely in the Cumberla drainage, since Wilson & Clark report it (’14) from Clear Fork, at Jellico, Campbell Co., Tenn., and other places in the upper Cumber- land region in Kentucky (I found it myself in Cumberland River at Orby, Bell Co., Ky.). ce In the Walker collection are two specimens from Powell ee ‘Lee Co., Va., collected by G. F. Sword. There is no mistake about — them, and thus this species must be listed with the upper Tennessee i shells, although known only from a single and somewhat indefinite, locality, and remarkable for its absence all over the rest mt this region. Type locality: Ohio River, Cincinnati, Ohio. Genus: ALASMIDONTA Say (1818). Ortmann, ’12), p. 294. 40. ALASMIDONTA (PRESSODONTA) MINOR (Lea), 1845. Margaritana minor Lea, ’45——Margaritana minor Lewis,. "72.— Alasmidonta (Pressodonta) minor Ortmann, ’12b, p. 295.—'14, p. 46 (anatomy).—Alasmidonta minor Ortmann, ’13), p. 311.— Alasmidonta iminor Goodrich, ’13, p. 94.—Alasmidonta (Presso- donta) minor Simpson, ’14, p. 498. A characteristic small creek species, locally abundant. Iti is joan all over the region, but strictly avoids the medium-sized and larger rivers. I have never seen it in the Clinch South of the Va.-Tenn. state line, and never in the Holston proper. I know it from the f . lowing stations. South Fork Powell River, Big Stone Gap, Wise Co., Va.; ; North Fork Clinch River, Tazewell, Tazewell Co., Va. (Walker coll.) 5 Clinch River, Cedar Bluff, and Richland, Tazewell Co., Va.; Clinch River, Cleveland, Russell Co., Va.; Clinch River, St. Paul, Wise | Va.; Clinch River, Speers Ferry, Scott Co., Va.; Brush Fork (tt tary to Poplar Creek and Clinch), Marlow, Anderson Co., Te Big Mocassin Creek, Mocassin Gap, Scott Co., Va.; North Fork ston River, Saltville, Smyth Co., Va.; Middle Fork Holston Ri UPPER TENNESSEE DRAINAGE. 561 Chilhowie, Smyth Co., Va.; South Fork Holston River, Barron, ' Washington Co., Va.; Big Creek (tributary to Holston), Rogerville, = Hawkins Co., Tenn. (Walker coll.) ; Little Pigeon River, Sevier- . ville, Sevier Co., Tenn. ; Boyd Creek, Boyd Creek, Sevier Co., Tenn. ; Little River, Melrose, Blount Co., Tenn.; Pistol Creek, Rockford, * Blount Co., Tenn.; Conasauga Creek, Monroe Co., Tenn. (Walker coll.) ; South Chickamauga Creek, Ringgold, Catoosa Co., Ga. é Type locality: Tennessee (and “ North Carolina,” but no exact locality given). 41. ALASMIDONTA (DECURAMBIS) MARGINATA (Say), 1819. _ Alasmodonia marginata Say, ’19.—Margarttana marginata Lewis, *71.—Alasmodonta marginata Pilsbry & Rhoads, ’96.—Alasmt- donta marginata Ortmann, ’12), p. 297 (anatomy ).—Alasmidonta marginata Ortmann, ’13), p. 311—Alasmidonta marginata Good- rich, ’13, p. 94—Alasmidonta (Rugtfera) marginata Simpson, "14, Pp. 504. As to the subgenus Decurambts Raf., ’31, see Frierson, ’14a, p. 7. Generally distributed over the whole upper Tennessee region, but _ ‘apparently more abundant toward the headwaters. Goes, in the Powell, to Olinger, Lee Co., Va.; in the Clinch, to Richland, Taze- well Co., Va.; in the Forks of the Holston, to Saltville and Chithowie, Smyth Co., Va., and is also in Big Mocassin Creek, Scott Co., Va. Pilsbry & Rhoads report it from Watauga River at Johnson City, Washington Co., Tenn. It is also in Ocoee River, at Ducktown, Polk Co., Tenn. (Walker coll.). | Type locality: Scioto River, Chillicothe, Ohio. 42. ALASMIDONTA (DEcURAMBIS) RAVENELIANA (Lea), 1834. Margaritana raveneliana Lea, ’34.—Alasmidonta (Rugifera) ravene- liana Simpson, ’14, p. 507. Differs from A. marginata chiefly in the absence of rugosities on the posterior slope. However, in very young specimens, traces of them are sometimes seen, so that this characteristic feature of the subgenus Decurambis (Rugtfera) is still indicated. This form un- doubtedly is an offshoot of the marginata-stock, separated from the 562 ORTMANN—NAYADES OF rest in the high mountains of North Carolina, and developed into a good species. The type locality is French Broad and Swananoa rivers, ville, Buncombe Co., N. Car. I have not been able to find this there, for the French Broad is polluted in this region (lumber : tries on Davidson River). But I have rediscovered it in Big P River, at Canton, acdabee Co., N. Car., where it is not — a proper places. Specimens from “North and South Fork of the Cur River,” referred to this species, do not belong here, but “a a form A. marginata, with well-developed rugosities upon the — slope (atropurpurea Raf., ’31). Genus: Pectas Simpson (1900). Ortmann, ’14, pp. 45 and 65 (as subgenus). On account of the very peculiar glochidia I consider. it . to retain Pegias as a genus, closely allied to Alasmidonta. 43. PEGIAS FABULA (Lea), 1836. Margarita (Margaritana) fabula Lea, ’36 —Alasmidont Ortmann, ’130, p. 311.—Alasmidonta (Pegias) fabula "14, p. 65 (anatomy) .—Pegias fabula Simpson, ’14, p. 473. A rare species in the upper Tennessee drainage, apparen’ y ferring smaller streams. Possibly it has been often overlook« account of its small size and the peculiarity of being genet a eroded. I know it from the following localities. oe Wallen Creek, Lee Co., Va. (Walker coll.) ; Powell Rive den, Lee Co., Va.; Big Mocassin Creek, Mocassin Gap, § Va.; North Fork Holston River, Saltville, Smyth Co., Va.; - Fork Holston River, Holston, Washington Co. (Walker col Mendota, Washington Co., Va.; South Fork Holston tolus, Sullivan Co., Tenn. Type locality: Cumberland River, Tenn. Genus: StropHitus Rafinesque (1820). Ortmann, 712), p. 299. UPPER TENNESSEE DRAINAGE. 563 44. STROPHITUS EDENTULUsS (Say), 1829. Alasmodonta edentula Say, ’29.—Anodonta edentula Lewis, ’72— Alasmodonta edentula Pilsbry & Rhoads ,’96.—Strophitus eden- tulus Ortmann, ’12b, p. 299 (anatomy).—Strophitus edentulus Ortmann, °133, p. 311.—-Strophitus edentulus Goodrich, °13, p. 94.—Strophitus edentulus Simpson, 14, p. 345. Walker thinks that the upper Tennessee-form might be distin- guished from the normal edentulus as var. shaefferianus (Lea). He believes that the latter is more compressed and more projecting an- teriorly, and has more frequently reddish nacre. In the average, this appears to be correct, yet there are many specimens in my rich mate- rial which do not exhibit these characters, and cannot be distin- guished from specimens of the normal edentulus, as found, for in- stance, in western Pennsylvania. Thus I do not think it advisable to separate the two forms. This species is abundant, both in larger rivers and smaller streams: in the Tennessee, in the lower French Board, the Holston and its forks and tributaries, and all over the Clinch and Powell drainages. It goes up, in the Powell, to Big Stone Gap, Wise Co., Va.; in the Clinch to Cedar Bluff, Tazewell Co., Va.; in the Forks of the Holston, to Saltville, Smyth Co., Va., and Bluff City, Sullivan Co., Tenn. ; It has not been found by myself in the eastern tributaries of the Tennessee south of the French Broad, but possibly this is accidental. Type locality: Wabash River. : Subfamily: LAMPSILINZE Ortmann, 1910. Ortmann, ’I0, p. 118. Genus: ELLipsartA Rafinesque (1820). Ptychobranchus Simpson, ’0o.—Ortmann, 12b, p. 305.—Ellipsaria Frierson, ’144, p. 7. 45. ELLIPSARIA FASCIOLARIS (Rafinesque), 1820. Obliquaria fasciolaris Rafinesque, ’20——Unio phaseolus Lewis, ’71. —Unio phaseolus Pisbry & Rhoads, ’96.—Ptychobranchus 564 ORTMANN—NAYADES OF phaseolus Ortmann, ’12b, p. 306 (anatomy) —Piychobraielna phaseolus Goodrich, ’13, p. 94.—Ptychobranchus phaseolus Simp- son, ‘14, p. 333—~Ellipsaria fasciolaris Frierson, ’14a, P. Lees Ptychobranchus fasciolaris Vanatta, ’15, p. 554. Widely and uniformly distributed over the upper ‘Teinenege region, but nowhere in great numbers. In the Tennessee, French — Broad, Holston, Clinch, lower Emory, and Powell. It goes up, in in the Powell, to Pennington Gap, Lee Co., Va.; in the Clinch, to Cleve- land, Russell Co., Va.; in North Fork Holston, to Mendota, Wash- ‘i ington Co., Va.; in South Fork Holston, to Emmett, Sullivan Coy Tenn. It is one of the species which has been reported from French _ Broad River, at Asheville, Buncombe Co., N. Car. It has not ‘yet been found in Little River, Little Tennessee, and Hiwassee, but a is in South Chickamauga Creek, at Ringgold, Catoosa Co.,Ga. Type locality: “ Ohio, Wabash, Kentucky rivers.” (The type is from Kentucky River, according to Vanatta.) ’ 46, ELLIPSARIA SUBTENTA (Say), 1825. Unio subtentus Say, ’25—Unio subtentus Lewis, ’71.—Unio sub- tentus Pilsbry & Rhoads, ’96.—Ptychobranchus subtentus Ort- mann, ’12b, p. 308 (anatomy).—Ptychobranchus subtentus Ort- mann, ’13), p. 311.—Ptychobranchus subtentus Goodrich, "13, I 94.—Ptychobranchus subtentum Simpson, 14, p. 339. Known from Tennessee, Powell, Clinch, Holston, and Nolichu rivers, but more abundant toward the headwaters, and rather 1 in the big rivers. Goes up to Big Stone Gap, Wise Co., Va.; Cedar Bluff, Tazewell Co., Va.; and to Smyth Co., Va. (in N and Middle Fork Holston). Also in Big Mocassin Creek, in Co., Va. Thus it ascends, in the small streams, father than fasciolaris. In the headwaters it is locally quite abundant. . Type locality: North Fork Holston River (Say says in § Carolina, but this is in Virginia) (topotypes examined). Genus: Ostiquarta Rafinesque (1820). Ortmann, 7120, p. 309. UPPER TENNESSEE DRAINAGE. 565 47. OBLIQUARIA REFLEXA Rafinesque (1820). Obliquaria reflera Rafinesque, ’20.—Unio cornutus Lewis, ’71.— Unio cornutus Pilsbry & Rhoads, ’96.—Obliquaria reflexa Ort- mann, ’12), p. 310 (anatomy) —Obdliquaria reflexa Simpson, ’14, P. 330. : | Only in the larger rivers. Tennessee below Knoxville (Lewis), and at Rathburn, Hamilton Co., Tenn. (Walker coll.). Lower Clinch, Patton’s Ferry, Roan Co., Tenn. (Pilsbry & Rhoads). I found it in the Clinch at Solway, Knox Co.; Edgemoor and Clinton, Anderson Co., Tenn. In the Walker coll. is a specimen from below Agee, Campbell Co., Tenn. Type locality: Kentucky River and Letart Falls (according to Vanatta, the types are from the latter place, below Parkersburg, W. Va.). Genus: CyproGENIA Agassiz (1852). Ortmann, 12), p. 312. 48. CYPROGENIA STEGARIA (Rafinesque), 1820. Obovaria stegaria Rafinesque, ’20.—Unio irroratus Lewis, ’71.— Unio irroratus Pilsbry & Rhoads, ’96.—Cyprogenia irrorata Ort- mann, ’12b, p. 312 (anatomy).—Cyprogenia irrorata Simpson, "14, p. 320.—Cyprogemia stegaria Vanatta, 15, p. 554. Known from the Tennessee below Knoxville (Lewis), and the Holston, at Boyd Island, Knox Co. (Pilsbry and Rhoads). It is also in the Tennessee at Rathburn, Hamilton Co., Tenn. (Walker collec- tion). I traced it up, in the Holston, to Turley Mill, Grainger Co., Tenn. ; and in the lower Clinch it is pat abundant, going up to Oak- man, Grainger Co., Tenn. Type locality: Ohio River. Genus: Dromus Simpson (1900). Ortmann, ’12b, p. 314. 566 ORTMANN=NATADES OF 49. DromMus DROMAS (Lea), 1834. Unio dromas Lea, ’34.—Unio dromas Lewis, ’71.—Unio dromas Pilsbry & Rhoads, ’96.—Dromus dromas Ortmann, ’12b, Pp. 315° (anatomy ).—Dromus dromas Simpson, ’14, p. 341. The typical form is rather swollen, and has a large knob or hump on each valve. This knob, however, is very variable. ae This form is found in the Tennessee proper. It has ees: re- ported by Lewis from below Knoxville, and by Pilsbry and Rhoads from Chattanooga, Hamilton Co. It is in the Walker collection from Chattanooga and from Rathburn, Hamilton Co. Pilsbry and Rhoads: report it also from the Holston, at Boyd Island, near Knoxville, : found it only in the Tennessee, three miles below Knoxville. | ad There are occasional specimens, which might be called by name, in the lower Clinch and Holston, but in this region it is gen erally replaced by the next form, with which it intergrades. __ Type locality: Harpeth River, Tenn. (and Cumberland River, Nashville, Tenn.). 50. DRoMUS DROMAS CAPERATUS (Lea), 1845. Unio caperatus Lea, ’45—Unio caperatus Lewis, ’71 Dn : caperatus Simpson, ’14, p. 343. This form has lost the “ hump,” and it is more comphanas the normal form. It begins in the Tennessee at Knoxville (Lewis) ! and ascends both the Clinch and Holston, where it intergrades, i lower parts, with typical D. dromas. In the Holston, it goes t ) Holston Station, Grainger Co., Tenn., and in the Clinch, it goes” Clinch River Station, Claiborne Co., enn: It enters also the Pow and goes up to Shawanee, Claiborne Co., Tenn. | : It is quite abundant in the Holston in Knox and Jefferson although it is frequently associated here with the typical form, intergrades with it. Its metropolis, however, is in the Clinch Anderson Co. upward, and in the Powell, where it is a pure Type locality: Clinch River (topotypes examined). Note: The tendency to develop here a more compressed for the smaller rivers agrees entirely with the similar phenomenon served in species of Fusconaia, etc. UPPER TENNESSEE DRAINAGE. 567 2 Genus: Oxovaria Rafinesque (1820). Ortmann, 128, p. 320. 51. OBovartA (OBOVARIA) RETUSA (Lamarck), 1819. __ Unio retusa Lamarck, ’19—Obovaria retusa Ortmann, *12b, p. 321 = (anatomy ).—Obovaria (Obovaria) retusa Simpson, 714, p. 290. Reported by Call from the Holston River in east Tennessee, but missing in Lewis's list. There are specimens in the Carnegie Mu- seum (from the Smith collection) labeled: Tennessee River, Knox Co., Tenn.; and others in the Walker collection labeled: Holston > River, Knox Co., and Holston River, Knoxville, Tenn. In all these cases, apparently, the Tennessee River at and below Knoxville is meant. I found this species only once: a young specimen in Clinch River, at Clinton, Anderson Co., Tenn. In the Walker collection are also specimens from the Tennessee at Bridgeport, Jackson Co., Ala., and from Florence, Lauderdale Co., Ala. (also reported by Hinkley). Thus this species seems to belong to the upper Tennessee fauna, going up to Knoxville and into the lower Clinch; but it apparently is very rare. Some of my specimens of O. subrotunda from the Holston (Mascot) have slighly incurved beaks and purple nacre, and resemble O. retusa to a dergee. Already Wilson and Clark (’14) have indi- _ cated a similar approaching of the two species in Cumberland River. Type locality: ? (Nova Scotia per errorem). ; 52. OBOVARIA (OBOVARIA) SUBROTUNDA (Rafinesque), 1820. Obliquaria subrotunda Rafinesque, ’20—Unio circulus Lewis, ’71.— Unio circulus Pilsbry & Rhoads, ’96——Obovaria circulus Ort- mann, ’12b (anatomy )—Obovaria (Obovaria) circulus Simpson, *14, p. 291.—Obovaria subrotunda Vanatta, ’15, p. 552. The identity of Obliquaria subrotunda and Obovaria striata Rafinesque with U. circulus: Lea has been recognized by Conrad in 1834, who selected subrotunda as name, which thus must be used. s Apparently rare in the upper Tennessee region. Reported from the Tennessee below Knoxville (Lewis), and at Knoxville (Call, 568 ORTMANN—NAYADES OF and Pilsbry & Rhoads). I know it sy fond the Holston River rv. Mascot, Knox Co., where I found three specimens with the diameter of 60, 61 and 62 per cent. of the length. These must be placed with typical subrotunda (diameter 60 per cent. and over). A fourth specimen, found associated with these has the diameter of 55 per : cent., and should be called O. subrotunda levigata (which see). _ ie In this region, there is evidently no tendency of this form to 0 : into the small streams of the headwaters, although my spe ie from the Holston indicate an inclination toward the small. stream form levigata. : Type locality: Ohio River (type from Kentucky River, scond- ing to Vanatta). 53. OBovaria (OBOVARIA) —— ‘LEVIGATA (Rafinesque), Unio levigata Rafinesque, ’20.—Obovaria (Obovaria) lens mas ’14, p. 293.—Obovaria.levigata Vanatta, ’15, p. 552. Already Conrad (1834) has seen that levigata Raf. is the s same aA, Ss lens Lea. e This is the small stream form of the main species, distinguished by greater compression of the shell (diameter less than 60 per cent of the length). It is quite abundant in the tributaries of the Ten- nessee below the Walden Gorge (Sequatchie River, Tenn., Flint River and Hurricane Creek, Madison Co., Ala., Elk River and Bear Creek). It intergrades here with the main species, as it does. in a He upper Ohio region. From above the Walden Gorge, I have it from South Chicka- mauga Creek, Ringgold, Catoosa Co., Ga. From the headwaters region, above Knoxville, I have just two specimens from the Hol- ston, one from Mascot, another from Holston Station, Grainger Co. in which the diameter falls under 60 per cent. (to 55 per cent. in first, to 57 per cent. in the other). I have never seen a trace of form in the small streams of this region. : Type locality: Kentucky River. UPPER TENNESSEE DRAINAGE. 569 Genus: NEPHRONAIAS Fischer & Crosse (1893).7 Ortmann, ’12b, p. 324. 54. NEPHRONAIAS LIGAMENTINA GIBBA (Simpson), 1900. Unio ligamentinus Lewis, ’71.—Unio ligamentinus Pilsbry & Rhoads, ’96.—Nephronaias ligamentina (incl. gibba) Ortmann, 712, (anatomy ).—Nephronaias ligamentina gibba Goodrich, ’13, p 94.—Lampsilis ligamentina gibba Simpson, ’14, p. 82. The main species is not represented in the upper Tennessee region, although occasional specimens turn up, which are hard-to dis- tinguish from it. The var. gibba is extremely abundant in all the larger rivers: Powell, Clinch, Holston, French Broad, Tennessee, but in the upstream direction it disappears before it reaches the head- waters region. In the Powell, it has been observed up to Shawanee, Claiborne Co., Tenn.; in the Clinch, up to St. Paul, Wise Co., Va.; in the North Fork Holston, up to Holston Bridge, Scott Co., Va.; in the South Fork, up to Pactolus, Sullivan Co., Tenn. From the French Broad it also enters the lower part of the ———— at Chunn’s Shoals, Hamblen Co., Tenn. Type locality: “ Ohio River and southward.” 55. NEPHRONAIAS PECTOROSA (Conrad), 1834. Unio pectorosus Conrad, May, ’34——Unio perdix Lea, August, ’34. —Unio biangularis Lea, ’40—Unio btangulatus Lea, ’43—Unio biangulatus Lewis, ’71—Unio biangularis Pilsbry & Rhoads, ’96. —Nephronaias perdix Ortmann, *12b, p. 326 (anatomy).— Nephronatas perdix Ortmann, ’13b, p. 311.—Nephronaias perdix Goodrich, *13, p. 94.—Lampsilis biangularis and perdix Simpson, "14, pp. 59 and 88. The date of publication of U. perdix Lea is incorrectly given by Simpson as 1827. There is no question that biangularis and perdix are the same species, and it is quite astonishing that this identity has not been recognized by Simpson. Recently, Frierson has suggested to me that vittatis Rafinesque, 1831, from Greene River, Ky., might 1 Possibly the generic name should be changed to Actinonaias; compare Frierson, Nautilus, 31, 1917, p. 48. PROC. AMER. PHIL. SOC., VOL. LVII, LL, OCT. 1, 1918. 570 ORT MANE ane OF be this species. However, I hesitate to accept this, till it has been shown that this species actually exists in Greene River. Agena e N. pectorosa is found in the Tennessee below and at Kooxville (Lewis, Pilsbry & Rhoads), but it seems to be rare there. Farther up, it goes into Clinch, Powell, Holston, French Broad, and lower Nolichucky, and it is also in Little River. It ascends toward the headwaters farther than does N. ligamentina gibba, and goes, in the Powell, to Pennington Gap, Lee Co., Va.; in the Clinch to Cleve- land, Russell Co., Va.; in North Fork Holston, to Saltville, Smyth Co., Va. ; ; in South Fork Holston, to Barron, Washington Co., Va.; and in the Watauga, to Watauga, Carter Co., Tenn. Just in the region, where N. ligamentina gibba begins to disappear (north of the Va.-Tenn. state line), N. pectorosa is most abundant and in its best development. Type locality: Elk River, North Alabama. Genus: AMyGpALonatAs Fischer & Crosse (1893). — Ortmann, 12), p. 327. 56. AMYGDALONAIAS TRUNCATA (Rafinesque), 1820. 5 Truncilla truncata Rafinesque, ’20——Unio elegans Lewis, ve Amygdalonaias elegans Ortmann, ’12b, p. 328 (anatomy) Plagiola (Amygdalonaias) elegans Simpson, ’14, p. 307. Plagiola elegans Vanatta, ’15, p. 553.—Amygdalonaias trunce Utterback, ’16, p. 148. ea, Vanatta does not accept Rafinesque’s specific name on account fe Unio truncata Spengler, 1793. However, these do not conflic (Walker, ’16). Not abundant, and missing in the headwaters. Tennessee be Knoxville (Lewis) ; Clinch River, Solway, Knox Co.; Edge and Clinton, Anderson Co.; Black Fox Ford, Union Co.; ¢ River Station, Claiborne Co.; Oakman, Grainger Co., Tenn. ston River, McBee Ford, near Hodges, Jefferson Co., Tenn. Type locality: Ohio River (type from Falls of the Ohio, ac ing to Vanatta). UPPER TENNESSEE DRAINAGE. 571 Genus: PLAGrIoLa Rafinesque (1820). Ortmann, ’120, p. 329. 57. PLAGIOLA LINEOLATA (Rafinesque), 1820. ? Obliquaria lineolata Rafinesque, ’20—Unio securis Lewis, ’71.— Unio securis Pilsbry & Rhoads, ’96.—Plagiola securis Ortmann, ’12b, p. 329 (anatomy).—Plagiola (Plagiola) securis Simpson, 14, p. 304.—Plagiola lineolata Vanatta, ’15, p. 553——Plagtola securis Walker, ’16, p. 45. Of the names of Rafinesque (depressa, lineolata, and ellipsaria) given to this species, lineolata has been selected by Conrad in 1834. (Walker, 1. c.) This name has been used also by Say, Agassiz, and Call. : Only in the larger rivers. Tennessee at and below Knoxville, also at Rathburn, Hamilton Co., Tenn. (Walker collection). In the Clinch at Patton’s Ferry, Roane Co. (Pilsbry & Rhoads), and at Offutt, Anderson Co., Tenn. Rare. Type locality: Falls of the Ohio. Genus: PARAPTERA Ortmann (I9QII). Ortmann, 12), p. 330. If the first species (leptodon) should actually belong here, the generic name must be dropped in favor of Leptodea Rafinesque, ’20 (see Frierson, ’14a, p. 6). Utterback (’16, p. 151) uses Lasmonos Raf. 58. PARAPTERA LEPTODON (Rafinesque), 1820. ? Unio leptodon Rafinesque, ’20—Unio tenuissimus Lewis, ’71.— Lampsilis (Proptera) leptodon Simpson, ’14, p. 188.—Lampsilis leptodon Vanatta, 15, p. 551—Lasmosos leptodon Utterback, "16, p. 156. A rare shell. Lewis gives it from the Tennessee below Knoxville. In the Walker collection it is from the Clinch, at Needham Ford, Union Co., Tenn., and I found it in the Clinch at Edgemoor, Ander- son Co., Tenn. I found it also in the Holston at Holston Station and Noeton, Grainger Co., Tenn. 572 ORTMANN—NAYADES OF Type locality: Lower Ohio River (Vanatta says that the from Kentucky River). 59. PARAPTERA FRAGILIS (Rafinesque), 1820. Unio fragilis Rafinesque, ’20.—Unio gracilis Lewis, 9 gracilis Pilsbry & Rhoads, ’96.—Paraptera gracilis Ortmann, ’12b, p. 331 (anatomy).—Lampsilis (Proptera) gracilis Simpson, ’14, p. 181 oe (?) fragilis Frierson, ’14a, Pp. 7- La fragilis Vanatta, ’15, p. 552.—Lasmonos fragilis Utterba p. 152. i In the larger rivers, not rare. Tennessee Rives, below ville (Lewis) and Holston River, Boyd Island, near (Pilsbry & Rhoads). In the lower Nolichucky at Chunn’s. Sho Hamblen Co.; up the Holston to Holston Station, Grainger Co.; n the Clinch up to Clinch River Station, Claiborne Co. ; in the Poy ell River, up to Combs, Claiborne Co., Tenn. Type locality: Ohio River (the type is, according to vi from creeks in Kentucky). si Genus: PRoprera Rafinesque (1819). Ortmann, ’12), p. 332. 60. PROPTERA ALATA (Say), 1817. : Unio alatus Say, ’17—Unio alatus Lewis, ’71—Unio alatus Is & Rhoads, ’96.—Proptera alata Ortmann, ’12b, p. 333 (anatomy) ——— (Proptera) ies —. 4, p. 162. the Clinch to Clinchport, Scott on Va.; up the Powell, to Comb Claiborne Co., Tenn.; it reaches the North Fork Holston t mouth, at Rotherweas: Hawkins Co., Tenn. It is also i in Broad River, and in the lower part of the secs at C Shoals, Hamblen Co., Tenn. Type locality: ? Genus: ToxoLasMA Rafinesque (1831). Corunculina (error typ.) Simpson, ’98.—Carunculina Ortmann, P. 337; 714, p P. 68.—To.rolasma Frierson, ’144, p. 7. : UPPER TENNESSEE DRAINAGE. 573 61. ToxXOLASMA LIivipUM (Rafinesque), 1831. Unio lividus Rafinesque, ’31—Unio moestus Lea, ’41—Unio cylin- drellus Lea, ’68.—Unio glans Lewis, ’71—Unio glans Pilsbry & Rhoads, ’96—Lampsilis (Carunculina) cylindrella and moesta Simpson, ’14, pp. 155, 156—To-xrolasma livida Frierson, ’14a, P- 7- The upper Tennessee form is not the real U. glans of Lea (’43), as already hinted at by Pilsbry & Rhoads. The latter is more swol- len, and has more inflated beaks, and possibly, it is the big river and lowland form of To,rolasma lividum. What Lea has described as U. moestus (from French Broad River, Tenn.) undoubtedly is this: I have specimens from Little Pigeon River (tributary to French Broad), which are fully identical with moestus. U. cylindrellus Lea (Duck River, Tenn.) is in shape absolutely identical with T. lividum; however, it differs by paler color of epidermis and nacre. Such pale specimens occasionally are found in the upper Tennessee drainage as individual variations. Yet it might be, that elsewhere (Tennessee drainage in northern Alabama, and Alabama drainage in Alabama and Georgia) this pale form becomes a distinct race. U. pullus Conrad, ’38, from the original locality, Wateree River, S. Car., may be a different species. But it is rather sure that the specimens from “ Warm Springs, N. Car.” (= Hot Springs, Madi- son Co., N. Car., on the French Broad) are actually T. lividum. There is great variation in the color of the nacre: it may be en- tirely dark purple, or may have (generally) a whitish margin along the edge of the shell; or it may (very rarely) have a pale, yellowish, color. The epidermis is mostly blackish or blackish brown, but it may become pale brown or greenish brown. This species seems to have a wide range over the upper Ten- nessee drainage, but, on the other hand, it seems to be rather local. I know it from the following localities: South Fork Powell River, Big Stone Gap, Wise Co., Va. (Walker coll.) ; Powell River, Jones- ville, Lee Co., Va. (Walker coll.) ; Shawanee, Claiborne Co., Tenn. (Walker coll.); Green’s Ford, Union Co., Tenn. (Walker coll.) ; Clinch- River, Speer’s Ferry, Scott Co., Va.; Emory River, Harri- 574 ORTMANN—NAYADES OF man, Roane Co., Tenn.; North Fork Holston River, at Saltville, — Smyth Co. (O. A. Peterson), and at Holston, Washington Co., Va. (Walker coll.) ; Holston Bridge, Scott Co., Va. (Walker coll.) ; ., Holston River, Rogersville (= Austin Mill), Hawkins Co., Tenn. (Walker coll.) ; Little Pigeon River, Sevierville, Sevier Co., Tenn. ; Pistol Creek, Rockford, Blount Co., Tenn. In addition, it is in Lewis’s list from the Tennessee Selon Knoxville. s Type locality: Rockcastle River, Ky. Its actual presence in eee "ee y castle River, at Livingston, Rockcastle Co., Ky., has recently been confirmed by Williamson (as U. glans, see: Ohio Naturalist, 5, 1905, — p. 311). Wilson & Clark, ’14, do not give it from this place, | but cae, have seen specimens in the Walker coll. from this locality. — Note: This form is also present in the tributaries of the Ten- nessee in North Alabama, as has been indicated long ago by Conrad. It also seems to be represented in the Alabama drainage, but from — this region it generally goes under the name of corvunculus Lea, ’68. I have a number of specimens of the latter before me, which I cannot distinguish from the upper Tennessee form. Genus: Lemrox Rafinesque (1881). Frierson, ’14a, p. 7——Ortmann, ’16, p. 39. 62. Lemr1ox rtmosus (Rafinesque), 1831. Unio rimosus Rafinesque, ’31.—Unio celatus Lewis, ’71.—Microm celata Goodrich, ’13, p. 94.—Micromya celata Simpson, ’14, p. 34.—Lemiox rimosus Frierson, ’14a, p. 7——Lemiox rimosus s Ort- mann, 16, p. 39 (anatomy). This species has been reported from the Tennessee below Kno ville (Lewis), and from Powell, Clinch, and Holston rivers (Call It seems to have a wide distribution, but it is found nowhere great numbers, and is thus a rare shell.” In the Powell, it goes up t C Jonesville, Lee Co., Va.; in the Clinch, to St. Paul, Wise Co., Vi in the North Fork Holston, to Holston, Washington Co., Va. found here and there in the Holston proper and the lower but I have never seen it in the eastern tributaries of the Hols Tennessee system. UPPER TENNESSEE DRAINAGE. 575 Type locality: Cumberland River (not reported from the Cum- berland drainage by Wilson & Clark, ’14). Genus: Mepionipus Simpson (1900). Ortmann, ’12b, p. 334; "15, p. 143. 63. MEDIONIDUS PLATEOLUS (Rafinesque), 1831. Unio plateolus Rafinesque, ’31—Unio conradicus Lewis, ’71—Unio conradianus Pilsbry & Rhoads, ’96.—Medionidus conradicus Ort- mann, ’120, p. 335, and *15, p. 142 (anatomy )—Medionidus con- radicus Ortmann, ’13), p. 311—Medionidus conradicus Good- rich, *13, p. 94—Medtonidus conradicus Simpson, ’14, p. 247. The identity of U. plateolus and conradicus Lea (’34) has been suggested to me by Frierson, and I think, this is correct. Very abundant in the headwaters and in small streams generally, but quite rare in the larger rivers. It has been found practically all over our region, and often goes up into the smallest streams which contain at all Nayades. It is hardly worth while to record single localities, and it suffices to say that it is in the Powell, Clinch, and Holston, and their tributaries; in the French Broad, it goes up to Asheville, N. Car.; it is in Little River, and in South Chickamauga Creek, Ga. From the Tennessee proper, below Knoxville, it has been reported by Lewis. Type locality: Falls of the Cumberland River. (Its existence at this place, below the falls, has been confirmed by Wilson & Clark, + 14.) Genus: Euryni Rafinesque (1820). Ortmann, 712), p. 336. 64. EuryNniA (MicroMYA) FABALIs (Lea), 1831. Unio fabalis Lea, ’31—Unio fabalis Lewis, ’71—Eurynia (Mi- cromya) fabalis Ortmann, 12), p. 339 (anatomy )—Eurynia fabalis Goodrich, ’13, p. 94——Micromya fabalis Simpson, ’14, P- 33- Rather rare, but possibly in part overlooked. Lewis gives it from Tennessee River, below Knoxville, and the Walker collection 576 ORTMANN—NAYADES OF has it from Tennessee River, Little River Shoals, Knox Co., Tenn. I found it chiefly toward the headwaters, in Powell, Clinch, and : ol ston. In the Powell, it goes up to Combs, Claiborne Co., Tenn. ; : in the Clinch, to Cleveland, Russell Co., Va.; in the North Fork ‘Hol- ston, to Hilton, Scott Co., Va.; in the South Fork, to Pactolus, Sul- livan Co., Tenn. It seems to be absent from the eastern tributaries of the system. Type locality: Ohio River. 65. Eurynta (MicromyA) TRABALIS (Conrad), 1834. Unio trabalis Conrad, ’34.—Eurynia (Micromya) trabalis Ortmann, Extremely rare. I found a single specimen in Clinch River, at Speer’s Ferry, Scott Co., Va.; another one in South Chickamauga Creek, Ringgold, Catoosa Co., Ga.; and three specimens in Hiwassee River, at Austral, Polk Co., Tenn. At the first locality, it baie ciated with the next species. The only difference of E. trabalis and perpurpurea is, that a former has white, the latter purple nacre. The specimens from the Hiwassee are exceptionally large, and do not taper so much behind as usual, sas This species seems to have its metropolis in the Cunntiegl system (Wilson & Clark, ’14). Call cites it from Clinch and Powell Rivers in Tennessee, but Lewis does not mention it from the “ ‘Hol- ston” (== Tennessee), and I have never seen a trace of it in ( e whole Holston drainage. According to Hinkley, it is rare in the mussel shoals of the Tennessee in Alabama; the Carnegie Museur has one specimen from Paint Rock River, at Paint Rock, Jack Co., Ala. It may be that my localities in Chickamauga Creek < Fliwaeeee River are connected with the range in northern Alabama. Type locality: ? 66. Eurynta (MIcRoMYA) PERPURPUREA (Lea), 1861. Unio perpurpurea Lea, ’61.—Eurynia perpurpurea Ortmann, 30, p. 311.—Eurynia purpurpurea Goodrich, ’13, p. 94.—Lampsilis perpurpurea Simpson, ’14, p. 133.—Eurynia (Mires ¢ purpurea Ortmann, ’15, p. 63 (anatomy). *12b, p. 340 (anatomy ).—Lampsilis trabalis Simpson, ’14, p. I UPPER TENNESSEE DRAINAGE. 577 This may be only a variety with purple nacre of E. trabalis, but I have not yet seen any intergrades. It is characteristic for the Clinch River, where it is not rare in Virginia (from Speer’s Ferry, Scott Co., up to Cedar Bluff, Taze- well Co.). In addition, it is in the Powell at Olinger, Lee Co., Va. (Walker coll.) ; in the North Fork Holston, from Rotherwood, Hawkins Co., Tenn., to Mendota, Washington Co., Va.; and in Emory River, Harriman, Roane Co., Tenn. _ In the Tennessee and Holston proper, and their eastern mountain tributaries, no trace of this species has ever been found. Type locality: Tennessee. 67. EURYNIA NEBULOSA (Conrad), 1834. Unio nebulosus Conrad,’34.—Unio cumberlandianus Lea,’36.—Unio ‘cumberlandicus Lea, ’38—Unio notatus Lea, ’38.—Unio glaber Lea, ’38.—Unio creperus Lea, ’38—Unio muehlfeldianus Lea, °38.—Unio simus Lea, ’38.—Unio obscurus Lea, ’38.—Unio zeig- lerianus Lea, ’38—Unio amenus Lea, ’40-——Unio fatuus Lea, *40.—Unio dactylus Lea, ’40-——Unio tener Lea, ’40.—Untio regu- laris Lea, ’41.—Unio puniceus Haldeman, ’42—Unio radians Lea, *57——Unio jonesi Lea, ’59.—Unio discrepans Lea, ’60.— Unio scitulus Lea, ’60.—Unio lingueformis Lea, ’60.—Unio per- pictus Lea, ’60—Unio planicostatus Lea, ’60—Unio sparus Lea, 68.—Unio dispansus Lea, ’71.—Unio glaber, iris, cumberlandi- anus, jonesi, sparus Lewis, ’71, and ’72—Unio muhlfeldtianus Pilsbry & Rhoads, ’96—Eurynia nebulosa, dispansa, and plani- costata Ortmann, ’13), p. 311—Eurynia nebulosa Goodrich, ’13, p. ’94—Eurynia (Micromya) nebulosa Ortmann, ’15, p. 64 (anatomy). Simpson, 714, has recognized the identity of a great number of these nominal “ species,” but not of all of them. He gives (pp. 119 and 120) : cumberlandianus (Cumberland R., Tenn.) ; notatus (Cum- berland and Holston R., Tenn.) ; glaber (Holston R., Tenn.) ; radians (Othcalooga Cr., Gordon Co., Ga.) ; jonest (Euharlee Cr., Ga.) ; dis- crepans (North Alabama) ; scttulus (Tuscumbia, Ala.) ; lingueformis (Columbia, Ga., and French Broad R., Tenn.) ; perpictus (Bull R. 578 ORTMANN—NAYADES OF and Holston R.) ; sparus (Swamp Cr., Whitfield Co., Ga.), and I am fully convinced that all these actually are synonyms of nebulosus, — But I believe that there are many others, and on account of the ex- _ traordinary size of the list of synonyms, I think it advisable to make a few remarks as to these. ‘s U. creperus Lea, ’38 (Tennessee), made a synonym of L. i“ (Lea) by Simpson (p. 115), is. founded upon an old half shell, hardly recognizable, but it may be this. In the Walker collection is _ a specimen from Clinch R., Va. (Wright), labeled creperus, which is distinctly E, nebulosa. It is a male, has purple nacre, and agencies | rather broad, only slightly interrupted rays. . U. dispansus Lea, ’71 (east Tennessee), has been put by Sinpede (p. 106) in the synonymy of L. vanuxemensis; I think it ‘fee 3 here. ‘ U. puniceus Haldeman, ’42 (Simpson, p. 104) (Holston R, ; Washington Co., Va.). I have found forms corresponding to the description in the same region (topotypes), and they simply are E nebulosa with a peculiar reddish-orange nacre. U. obscura Lea, ’38 (Nashville, Tenn.), and U. seiglerian Les, : ’°38 (Cumberland R., Tenn.), made synonyms by Simpson (p. 117) — are also this, with rather fine, uninterrupted rays, and purple nacre. U. fatuus Lea, ’40 (Holston R., Tenn.), and U. dactylus Lea, ’40 (Caney Fork R., Tenn.), made synonyms by Simpson (pp. 116, 117) are rather elongated and unusually swollen forms of £. cee with the rays less developed, and not interrupted. _U. planicostatus Lea, ’60 (Tuscumbia, Ala.) (Simpson, p. 11 3 A strongly compressed male, with rays distinct, rather wide, and little interrupted. Such specimens are frequent, chiefly in the Clinch, U. muehlfeldianus Lea, ’38 (Cumberland R., Tenn.) (Simpson, — p. 121). According to Simpson, only a single specimen is known, which is undoubtedly this, probably a female. Pilsbry & Rhoa give this also from Watauga R., near Johnson City, Washington Tenn.: near this place (Watauga, Carter Co.), I have found nebulosa. U. amenus Lea, ’40 (Holston R., Tenn.) (Simpson, P. 122) typical female of E. nebulosa. U. tener Lea, ’40 (Big Pigeon R., Tenn.) and U. regularis UPPER TENNESSEE DRAINAGE. 579 *41 (French Broad R., Tenn.), made synonyms by Simpson (pp. 122, 123). There are two specimens (¢ and Q) in the Walker collection, labeled tener (from French Broad, Asheville, N. Car.). They are rather thin-shelled, have rays, which are fine, and subcontinuous in © the male, but somewhat spotted in the female. They undoubtedly are a form of E. nebulosa. U. simus Lea, ’38 (Cumberland R., Tenn.) (Simpson, p. 123). A male, with strongly developed rays: specimens of this type occur frequently, and are practically identical with U. notatus, admitted by Simpson as a synonym of nebulosus. This tremendous synonym indicates that we have to deal here with a very variable species. The variation concerns first of all the color pattern of the epidermis (rays). In the second line it is shown in the color of the nacre, the shape of the shell, and, of course, some- times the females have been made “ species” for the reason of their different shape. It is not excluded that additional synonyms may be discovered. E. nebulosa belongs in the affinity of E. iris (Lea), ’30, and has practically the same anatomy. Indeed, some of its forms are hardly distinguishable from E£. iris, and it may be that the latter is only the western and northwestern representative of it (that of the Ohio drainage). Thus it is also explained why iris has been recorded for the upper Tennessee River (Lewis). According to my observations, there is only one species of the iris-group in the upper Tennessee region. It is generally an elon- gated-elliptical shell, more or less pointed behind in the male, slightly dilated and rounded posteriorly in the female, of a yellowish, green- ish, or brownish color, covered more or less with rays, which nor- mally are well developed, and very often broken up into spots. These rays may be fine or wide (in the female, they are often very broad and distinct on the dilated part of the shell), may cover all of the shell, or only part of it, and may be indistinct or nearly miss- ing. The interior of the shell may be white, or of all shades of salmon, orange, pink, or purple. Although I have tried hard, I have been unable to siete this group of forms into species, and it is also impossible for me to dis- tinguish local races. It is true that sometimes specimens from a cer- 580 ORTMANN—NAYADES OF tain locality, chiefly from some small creeks, show uniform and pe- culiar characters, but this holds good only for short distances. a the longer rivers (Powell, Clinch, Holston) the variation is consid- erable and irregular, apparently without any recognizable rules. _ The distribution of E. nebulosa extends over the whole of the | upper Tennessee region, but the species decidedly favors the head- waters and small streams, and often goes up to the uppermost limit ot of Nayad-distribution in this region. It is not necessary to give a list of the localities: it is practically everywhere in the Powell, Clinch, Holston, French Broad, Little River, Hiwassee. (It apparently is . only accidental that it has not been recorded from the Little Ten- — nessee.) It is also in Chickamauga Creek, and it deserves special — mention that it goes up, in French Broad, to Asheville, N. Car. — In the larger rivers, this species is rare, yet it is present. In addition, E. nebulosa has a wide range not only in the Cumber- land drainage, but also in that of the Tennessee in North Alabama, — and it also has invaded the headwaters of the Coosa~Alabama system — in northern Georgia and Alabama. It is very singular that Wilson & Clark (’14) do not mention it in their paper on the Cumberland shells, although a good number of the synonyms have their ek locality in this system. ; Type locality: Black Warrior River, Ala. 68. EurynrA (MicroMyA) VANUXEMENSIS (Lea), 1838. Unio vanuxemensis Lea, ’38—Unio nitens Lea, ’40—Unio wm- brosus Lea, ’57 (==umbrans Lea, ’57).—Unio tenebricus Lea, | 57.—Unio pybasi Lea, ’58.—Unio fabaceus Lea, 61 —Unio copei Lea, ’68.—Unio pybasi Lewis, ,’71.—Unio pybast and caliginosus Pilsbry & Rhoads, ’96.—-Eurynia (Micromya) vanuxremensis Ortmann, ’12b, p. 342.—’I5, p. 65 (anatomy).—Nephronaias — copet and Eurynia vanuxemensis Ortmann, ’130, p. 311.—Eurynia vanuxemensis Goodrich, 13, p. 95.—Lampsilis vanuxemensis Simpson, ’14, p. 105. : Unio dispansus Lea, ’71, placed by Simpson here, belongs to E nebulosa (see above). Probably there are other synonyms. This shell is shorter and higher than £. nebulosa, has generally eS UPPER TENNESSEE DRAINAGE. 581 dark epidermis with indistinct or no rays, and a deep copper-colored or purplish nacre; rarely the latter is lighter, and even when whitish, it has at least some purple or red. The female is characterized by a very strong dilatation of the postbasal margin, and very often, chiefly in old shells, it has a strong “constriction” behind this dilatation. The distribution of this species is very similar to that of E. nebu- losa, preferring also the small streams, and very often the two species are found associated. Also here it is unnecessary to give a list of localities, and it suffices to state that it is found practically over the whole upper Tennessee region. It should be remarked, however, that I did not find it in the headwaters of the Clinch (above Speer’s Ferry, Va.), but it may have been overlooked here. While locally abundant in smaller streams, it becomes rare in the larger rivers. Also here, local races cannot be distinguished, except that the shell attains, in certain streams, a much greater size than in others. This is the case, for instance, in the Middle Fork Holston in Smyth Co., Va., where exceptionally large specimens (U. copei Lea) are found, while the North Fork contains a small race. Specimens of larger rivers are also generally rather large and well developed, and often lack, in the 9, the posterior constriction. Such specimens have a resemblance to U. lienosus Conrad, being more drawn out at the posterior end. U. lienosus is a southern form, not found in the Tennessee drainage. There is no doubt that U. caliginosus Lea (=lienosus), recorded by Pilsbry & Rhoads from the lower Clinch, is founded upon such specimens. Also this species has a wide distribution in the Cumberland drain- age, the Tennessee drainage in Alabama, and the Coosa-Alabama system, and it is very likely that also the form of the Cumberland, questionably referred to lienosa by Wilson & Clark (’14), is this. Type locality: Cumberland River, Tenn. 69. Eurynia (Eurynia) recta (Lamarck), 1819. Unio rectus Lamarck, ’19——Unio rectus Lewis, ’71—Unio rectus Pilsbry & Rhoads, ’96.—Eurynia (Eurynia) recta Ortmann, ’12b, P. 344 (anatomy ).—Eurynia recta Goodrich, 713, p. 95 ——Lamp- silis recta Simpson, ’14, p. 95. Abundant in the larger rivers: Tennessee in Knox Co., lower 582 ORTMANN—NAYADES OF French Broad and lower Nolichucky (Hamblen Co.) ; all the way up the Holston to the North Fork at Rotherwood, Hawkins Co., Tenn. Also in the Clinch up to St. Paul, Wise Co., Va., and the — Powell up to Combs, Claiborne Co., Tenn. Type locality: Lake Erie (topotypes examined). : Note: The Lake Erie form differs somewhat from that of the central basin. If it should be found to be desirable to express this in nomenclature, E. recta should be reserved for the lake form, and E. recta latissima (Rafinesque), ’20, should be used for the other. Genus: LAMPSILIs Rafinesque (1820). Ortmann, ’12b, p. 345. 70. LAMPSILIS VIRESCENS (Lea), 1858. Unio virescens Lea, ’58—Lampsilis virescens Simpson, ’14, p. 93. I am not quite satisfied as to the proper position of this species _ within the genus Lampsilis. According to external appearance, it seems to be related to L. Iuteola (Lam.), although it also has some — features, which resemble those of L. teres (Raf.) (==anodontoides Lea). In either case, however, it would be a Lampsilis. The type locality is Tennessee River, Tuscumbia, Colbert Co., Ala., and it has been reported, by Call, from Spring Creek at Tus- cumbia. The Carnegie Museum has it from tributaries of the Ten- nessee in northern Alabama (Paint Rock River, and Bear Creeks: It has never been found anywhere else. oo But I have found a few specimens (all males) in Emory River, at Harriman, Roane Co., Tenn., and in the Walker collection are others from Coal Creek, Anderson Co., Tenn. These two streams are not far apart, and flow from Walden Ridge to the Clinch. Emory River is the only western tributary which completely cuts through Walden Ridge and drains, in its headwaters, a section of the Cumberland Plateau. 71. LAMPSILIS OVATA (Say), 1817. Unio ovatus Say, ’17——Unio ovatus Lewis, ’71—Unio ovatus Pils- bry & Rhoads, ’96.—Lampsilis ovata Ortmann, ’12), p. 350° (anatomy ).—Lampsilis ovata Simpson, 14, p. 48. UPPER TENNESSEE DRAINAGE. 583 Distinguished by the distinct and sharp posterior ridge, depressed (truncated) posterior slope, and the peculiar, wedge-shaped anterior part of the shell. But these characters gradually pass into those of the variety ventricosa. The typical L. ovata is restricted to the larger rivers, and quite abundant there. It is in the Tennessee, the Little Tennessee, Hol- ston, Clinch, and Powell. In the Powell, it goes up to Shawanee, Claiborne Co., Tenn. ; in the Clinch, to Clinchport, Scott Co., Va. It is the prevailing form in the Holston proper, but does not go into the Forks of the Holston. All along its range, and chiefly above Knox- ville, it is accompanied by the var. ventricosa, and intergrades with it. But at the points just named, it disappears, and leaves the field to ventricosa. Type locality: Ohio River. 72. LAMPSILIS OVATA VENTRICOSA (Barnes), 1823. Unio ventricosa Barnes, ’23.—Lampsilis ventricosa Ortmann, ’12), p. 351 (anatomy )—Lampsilis ovata ventricosa Ortmann, ’130, p. 311—Lampsilis ovata ventricosa Goodrich, *13, p. 95—Lampsilis ventricosa Simpson, 714, p. 38. According to Vanatta (’15, p. 551), the type of Lampsilis car- dium Rafinesque, ’20, is this, and also Conrad (’34) says so. How- ever, this conflicts with Rafinesque’s description, from which it is evident that L. cardium is the female of L. ovata. We have here a case where the “type” does not agree with the original description, and it should be borne in mind that the co-called “types” of Rafin- esque, in the Philadelphia Academy, are not types in the strict sense, but merely “ authentic specimens ” of somewhat doubtful value. In this variety, the posterior ridge becomes indistinct, the pos- terior slope is not excavated, and the anterior part of the shell is not remarkably compressed. Also, the shell is generally less convex. But there are all stages of transition. The upper Tennessee form of ventricosa very rarely has the distinct rays of the corresponding form of the upper Ohio region. L. ovata ventricosa is found associated with the normal L. ovata in the larger rivers, but is less frequent there. It goes, however, 584 ORTMANN—NAYADES OF beyond the upper limit of L. ovata in the headwaters, where it is — found in its best development and as a pure race. In the Powell, it goes to Olinger, Lee Co., Va.; in the Clinch, to Cedar Bluff, Taze- well Co., Va.; in North Fork Holston, to Saltville, Smyth Co., Va. — (also in Big Mocassin Creek) ; and in the South Fork, to Emmett, — Sullivan Co., Tenn. It is in Nolichucky River, at Chunn’s Shoals, — Hamblen Co., Tenn. (inclining here toward ovata) ; in Little Pigeon River, Sevierville, Sevier Co., Tenn. ; in Little River, Melrose, Blount — Co., Tenn.; and in Emory River, Harriman, Roane Co., Tenn. I . have seen it also in Chickamauga Creek, Ringgold, Catoosa Co., Ga. — Type locality: Wisconsin River, and Mississippi River, — du — Chien, Wis. . ; 73. LAMPSILIS FASCIOLA Rafinesque (1820). Lampsilis fasciola Rafinesque, ’20.—Unio multiradiatus and perra- — diatus Lewis, ’71—Unio multiradiatus Pilsbry & Rhoads, ’96.— Lampsilis multiradiata Ortmann, ’12), p. 352 (anatomy )—Lamp- silis multiradiata Ortmann, ’13b, p. 311.—Lampsilis multiradiata — Goodrich, ’13, p. 95——Lampsilis multiradiata Simpson, ’14, p. 55. Also in this case, I do not follow Vanatta’s (’15, p. 551) de- — termination of the “type” of fasciola, but rely on Rafinesque’s de. scription, which indicates a shell of the cardiwm-ovata type, with 3 unequal, flexuous rays, which fits multiradiata Lea, but not luteola Lamarck. Moreover, Conrad, in 1836, held the same opinion, also — pointing out the undulated rays as the main character of this species. — Practically everywhere in the larger rivers as well as in smaller streams, but apparently more abundant toward the headwaters. In the Tennessee, South Chickamauga, Hiwassee, Little Tennessee, Little River. In the French Broad drainage, going up here to Ashe- ville, N. Car., and, in Big Pigeon River, to Canton, N. Car. In the Holston, up to the South Fork at Barron, Washington Co., Middle — Fork at Chilhowie, Smyth Co., and North Fork at Saltville, Smyth 7 Co., Va. Also in Big Mocassin Creek, and Watauga River. In the Clinch, it goes up to Cedar Bluff, Tazewell Co., Va.; in the Powell, to the North Fork at Big Stone Gap, Wise Co., Va. Also in the — Emory at Harriman, Roane Co., Tenn. a Type locality: Kentucky River. UPPER T ENNESSEE DRAINAGE. 585 74. LAMPSILIS ORBICULATA (Hildreth), 1828. Unio orbiculatus Hildreth, ’28—Lampsilis orbiculata Ortmann, ’120, Pp. 343 (anatomy )—Lampsilis orbiculata Simpson, 14, p. 76. Not reported previously from the upper Tennessee region, but in the Walker collection are two specimens labeled “ Holston R., Tenn.,” which come from the Lewis collection (and thus probably are from the Tennessee). I have found myself two specimens in the Clinch, one at Offutt, Anderson Co., the other at Solway, Knox Co., Tenn. This species is also in the Tennessee in North Alabama, at the mussel shoals near Florence (reported by Hinkley, and represented in Walker coll. and Carnegie Mus.). Type locality: Muskingum River, Marietta, Ohio. Genus: TRUNCILLA Rafinesque (1820). Ortmann, ’12b, p. 354. The subgenera distinguished by Simpson require revision, also with regard to nomenclature. I disregard them for the present, re- marking only that the soft parts furnish good criteria for their defi- nition. 75. TRUNCILLA TRIQUETRA Rafinesque, 1820. Truncilla triqueter Rafinesque, ’20—Unio triangularis Lewis, ’71.— Unio triangularis Pilsbry & Rhoads, ’96.—Truncilla triquetra Ortmann, ’12), p. 355 (anatomy ).—Truncilla triquetra Simpson, "14, p. 5. Rather frequent both in larger and smaller rivers, but nowhere in great numbers. Tennessee below Knoxville (Lewis) and at Knoxville (Pilsbry & Rhoads) ; Little River in Knox Co. (Walker coll.) ; lower Nolichucky; Holston River up to the South Fork at Pactolus, Sullivan Co., Tenn., and the North Fork at Mendota, Washington Co., Va.; in the Clinch up to Clinchport, Scott Co., Va.; in the Powell, up to Shawanee, Claiborne Co., Tenn. Type locality: Falls of the Ohio. PROC, AMER. PHIL. SOC., VOL. LVII, MM, OCT, I, 1918, 586 ORTMANN—NAYADES OF 76. TRUNCILLA ARCZFORMIS (Lea), 1831. Unio arceformis Lea, ’31—Unio arceformis Lewis, ’71—Unio sistas Pilsbry & Rhoads, ’96.—Truncilla arceformis si a son, "14, p. 12. In the larger and medium rivers: Tennessee, Knox Co., Tenn.; . French Broad at Boyd Creek, Sevier Co. In the Holston locally abundant: Boyd Island, near Knoxville (Pilsbry & Rhoads) ; Mc- Millan, Knox Co.; Mascot, Knox Co.; Gant Island near Straw- berry Plains, Jefferson Co. (Walker coll.); McBee Ford, near : Hodges, Jefferson Co.; Turley Mill, Noeton, and Holston Station, Grainger Co.; Austin Mill, Hawkins Co., Tenn. In the Clinch at Clinch River Station, Claiborne Co.; and Oakham, Gramere Co., Tenn. — : Type locality: “ Tennessee River.” Lea (Tr. Amer. Philos. Soe. 3 1834, p. 86) corrects this, saying that, according to Troost, this species is not in the Tennessee, but only in the Cumberland River (meaning, of course, the Tennessee of northern Alabama). 3 77. TRUNCILLA INTERRUPTA (Rafinesque), 1820. Obliquaria interrupta Rafinesque, ’20.—Unio brevidens Lea, ’31 (not ’24, as given by Simpson).—Unio brevidens Lewis, ’71—Trun-_ cilla brevidens Simpson, ’14, p. 7. The identity of Rafinesque’s species with that of Lea has been recognized by Conrad, accepted by Kuester and Reeve, and con- firmed by Vanatta (715, p. 550). But Vanatta does not use the nam of interrupta, for reasons which do not hold good, as shown by Walker (’16, p. 45). The only objection to interrupta is that it given, originally, from Kentucky and Ohio rivers, while it seems be absent at least from the Ohio. But we must remember that Lea’ brevidens also was originally given from Ohio, corrected subse: quently, ’34, to Cumberland River. In the larger and medium rivers: Tennessee (Lewis), Ho Clinch, and Powell. In the Powell, up to Rose Hill, Lee Co., in the Clinch, up to Clinchport, Scott Co., Va.; in the Holston to the North Fork at Hilton, Scott Co., Va. Type locality: “ Kentucky and Ohio Rivers.” (Probably i rect; type from Ohio River, according to Vanatta.) UPPER TENNESSEE DRAINAGE. . 587 78. TRUNCILLA LENIOR (Lea), 1843. Unio lenior Lea, ’43.—Truncilla lenior Simpson, ’14, p. II. A rare species. I have found it in the Clinch, Speer’s Ferry, Scott Co., Va.; in North Fork Holston, Rotherwood, Hawkins Co., Tenn. ; in South Fork Holston, Pactolus, Sullivan Co., Tenn.; and in the uppermost Holston proper, Church Hill, Hawkins Co., Tenn. There are specimens in the Walker coll. (from Mrs. Andrews), labeled: Holston River, Knox Co. (probably Tennessee River), but Lewis has not recorded it from these parts. For the rest, it is miss- ing in this region, but it turns up again in the Tennessee drainage in North Alabama: Simpson gives it from Paint Rock River, Wood- ville, Jackson Co., Ala., and the Carnegie Museum possesses quite 2 number of specimens from this river at Paint Rock, Holly Tree, and Trenton, Jackson Co., Ala. The male of this species may be easily recognized by the fine denticles on the margin of the posterior end; for the rest, it looks like a pale-colored Eurynia nebulosa. Type locality: Stones River, Tenn. (Cumberland drainage; but _ ‘missing in the list of Cumberland shells published by Wilson & Clark, ’14). 79. TRUNCILLA HAYSIANA (Lea), 1833. Unio haysianus Lea, ’33—Unto haysianus Lewis, ’71—Unio haysi- anus Pilsbry & Rhoads, ’96—Truncilla haysiana Ortmann, ’12), p. 357 (anatomy) —Truncilla haysiana Ortmann, 713), p. 311.— Truncilla haysiana Goodrich, *13, p. 95.—Truncilla haysiana Simpson, ’14, p. 16. Widely distributed, but always only in small numbers at a given locality. Tennessee in Knox Co. (Lewis) and at Knoxville (Pilsbry & Rhoads) ; Little Tennessee, Coytee, Loudon Co., Tenn. (Walker coll.) ; in the Powell, up to Pennington Gap, Lee Co., Va.; in the Clinch, up to Raven, Tazewell Co., Va.; in the Holston, it goes in the North Fork to Hilton, Scott Co., Va., and in the South Fork to Pactolus, Sullivan Co., Tenn. Type locality: Cumberland River. 588 - ORTMANN—NAYADES OF 80. TRUNCILLA STEWARDSONI (Lea), 1852. Unio stewardsoni Lea, ’52.—Unio stewardsoni Lewis, — 4 cilla stewardsoni Simpson, ’14, p. 21. I think that what Walker (‘10a, pl. 3, f. 4) has figured as ‘the ‘ 3 of T. lewisi is actually an old ¢ of stewardsoni. S A rare species. The Carnegie Museum has specimens from the Tennessee at Knoxville, and Lewis reports it from this region; and — there are also specimens from Clinch River, Clinton, Anderson Co., _ in the Carn. Mus, I found it myself in the Holston, at McMillan and Mascot, Knox Co., and at Holston Station, Grainger Co., Tenn. Type locality: “ Chattanooga River, Tenn.” There is no such river in Tennessee. Generally, Lea’s “ Chattanooga River” is the Chattooga River in northern Georgia (tributary to Coosa); but in the present instance this cannot be, since this species is not found in the Coosa drainage. . 81. TRUNCILLA LEWIst Walker (1910). Unio foliatus Lewis, ’71.—Truncilla lewisi Simpson, ’14, p. 20. As stated above, Walker’s figure of the ¢ (’10a, pl. 3, f. 4) prob- ably belongs to T. stewardsoni. I have found a single small male specimen (soft parts examined !) : in the Holston, at Holston Station, Grainger Co., Tenn., and another, somewhat larger one, in the Powell, at Combs, Claiborne Co., Tenn. These differ from the male of T. stewardsoni by a wider radial fur- row, with the two ridges confining it, more divergent. My speci- mens are not full grown, and even if they should be young females - (as Walker suggested after examination of the one from the ston), they would present to us the shape of the male, as all young females do in the genus Truncilla, exactly as the soft pare of yo females resemble those of the males. The Carnegie Museum has (from the Hartman coll.) two fe- males, labeled: Tennessee River, Knox Co., Tenn. Lewis gives # form (as U. foliatus) from the Tennessee at Little River Sh below Knoxville; and Walker reports it from Clinch and Hols rivers in Knox Co. (also from Cumberland River, Burnside, Pulask Co., Ky., but not found here by Wilson & Clark, 14). UPPER TENNESSEE DRAINAGE. 589 Thus this is undoubtedly one of the rarest species of Truncilla, and its distribution should be studied more closely. Type locality: Holston River, Tenn. (topotype examined). 82. TRUNCILLA PROPINQUA (Lea), 1857. Unio propinquus Lea, ’57—Unio propinquus Lewis, ’71—Unio propinquus Pilsbry & Rhoads, ’96.—Truncilla propinqua Simp- son, "14, p. 27. 4 Reported, by Lewis, from Tennessee River, Knox Co., and from 4q the Clinch by Call. Pilsbry & Rhoads give it from the Tennessee at Knoxville, the Holston at Boyd Island, near Knoxville, and from the q Clinch at Patton’s Ferry, Roane Co. The Carnegie Museum pos- a sesses it from the Tennessee, Knox Co. (Smith coll.), and I found it a myself in the Clinch at Edgemoor and Clinton, Anderson Co., Tenn. 4g Type locality: Florence and Tuscumbia, Ala. (topotypes ex- - amined). 83. TRUNCILLA TORULOSA (Rafinesque), 1820. Amblema torulosa and gibbosa Rafinesque, ’20.—Unio perplexus Lewis, ’71.—Truncilla perplexa Simpson, ’14, p. 24——Truncilla torulosa Vanatta, *15, p. 550. The identity of torulosa and gibbosa with perplexus has been in- dicated by Conrad (’34), and he selected the name of torulosa. Al- though he later (’36) uses gibbosus (and so do Agassiz and Reeve), the first selection has to stand. The typical T. torulosa has a radial row of prominent knobs across the middle of the shell. But these knobs vary greatly, and in the upstream direction, they have the tendency to become reduced, finally disappearing, thus passing into the condition seen in the next form. nae Restricted to the larger rivers. It turns up in the Tennessee at and below Knoxville (Lewis), and continues down the river, but is hardly found above Knoxville. The Carnegie Museum has three young specimens from near Knoxville (Hartman coll.), which have distinct knobs. In addition, there is a specimen from Chattanooga, Hamilton Co., Tenn. (Juny coll.), which is typical in all respects, 590 ORTMANN—NAYADES OF except that it has salmon-colored nacre; to my knowledge, this color of the nacre is extremely rare. . : Farther down the Tennessee, at the mussel shoals in Alabama, — this species is abundant (Conrad, Hinkley, and Carn. Mus.). : Type locality: Ohio River and Kentucky River (the type is from Kentucky River, according to Vanatta). 84. TRUNCILLA TORULOSA GUBERNACULUM (Reeve), 1865. Unio gubernaculum Reeve, Conch. icon. 16. Unio. ’65, pl. 28, f. 146. Reeve’s figure undoubtedly is this form. Simpson (’14, p. 26) makes this a synonym of the var. rangiana (Lea), and it surely is the parallel form to rangiana of the upper Ohio drainage. It differs, however, distinctly by the dark green color of the pose expan- sion of the female shell. From the typical torulosa, this variety differs by the poorly de- veloped or wanting knobs, and by the rather more compressed shell. This is the headwaters form of torulosa, and begins to take its place in the region of Knoxville. I have it from the Nolichucky, Chunn’s Shoals, Hamblen Co.; here, as also in the lower Holston, faint knobs may yet be present. Farther up, the shell is entirely smooth. In the Holston, it goes up to the South Fork at Pactolus, Sullivan Co., Tenn., and to the North Fork at Holston Bridge, Scott Co., Va. In the Clinch, it goes to Dungannon, Scott Co., Va. Itis also in the Powell, up to Shawanee, Claiborne Co., Tenn. Locally, it may be quite abundant. Type locality: ? 85. TRUNCILLA TURGIDULA (Lea), 1858. Unio turgidulus Lea, ’58 (male).—Unio deviatus Reeve, 64 (fe- male).—Truncilla deviata Walker, ’10b, pp. 77, 78, 81 a glo ; deviata Simpson, ’14, p. 31. Unio turgidulus has been regarded as the male of the female U. florentinus, but it belongs to the female deviatus. In Walker’s key — (10b), this has been indicated by the grouping, but it has not been — expressly mentinoned. : T. turgidula stands nearest to T. biemarginata (Lea), a species — UPPER TENNESSEE DRAINAGE. 591 known from the Tennessee in North Alabama, but which has not yet been recorded from the upper Tennessee. T. turgidula agrees with the latter in the biangulate posterior ridge, but the biangulation is much less pronounced, and the depression or furrow in front of it is less developed. In the female (deviatus), the biangulation is also present, but indistinct, and the furrow is obliterated, being filled by the expansion of the shell. The female resembles, to a degree, that of T. florentina, but has the shell, as Simpson states, more elongated, and has fuller and higher beaks (the latter charac- ters hold also good for the male). This species has been recorded from Cumberland and Tennessee rivers (not recently found by Wilson & Clark, ’14, in Cumberland), and from Duck River, Tenn. (Call), but only one definite locality is known (Florence, Lauderdale Co., Ala.). Hinkley reports it from Shoals Creek, Lauderdale Co., Ala. The Carnegie Museum has it from Bear Creek, in Franklin Co., Ala. I have found it in the upper Tennessee drainage, where it is not rare in the Holston proper from Knox Co. up to Austin Mill, Hawkins Co., Tenn. I found it also in Emory River, Harriman, Roane Co., Tenn. In the Walker coll. it is represented from the Holston, Rogersville, which is prac- tically the same locality as Austin Mill. Type locality: Cumberland River and Florence, Ala. 86. TRUNCILLA FLORENTINA (Lea), 1857. Unio florentinus Lea, ’57.—Truncilla florentina Simpson, ’14, p. 30 (excl. turgidulus). This species has not a biangulate posterior ridge, but this ridge is rounded, and the radial depression in front of it is hardly de- veloped, indicated only by a flattening of the shell. In the female, the posterior expansion of the shell may become very large, and is generally of the color of the rest of the shell, or lighter, but not uni- formly dark green, as in T. capseformis.. By the latter character, by the more strongly developed and more numerous denticulations of the margin of the expansion, and by the greater convexity of the shell, T. florentina is distinguished from capseformis. The male of T. florentina is shorter, higher, and more swollen than that of T. capseformis. 592 ORTMANN—NAYADES OF Known from the Cumberland and Tennessee Rivers, from the latter, however, reported hitherto only from North Alabama. I have found it in the Holston, from Knox Co., up to Holston Station, — Grainger Co., Tenn., but not in great numbers. Type locality: Florence, Ala. (and Cumberland River) (topotypes examined). 87. TRUNCELLA WALKER! Wilson & Clark (1914). Truncilla walkeri Wilson & Clark, ’14, p. 46, pl. 1, f. 1. This is practically a large, compressed T. florentina. It agrees with it in every respect, except that it attains a larger size, and is as compressed as T. capseformis. It may be only the headwaters form of T. florentina. From T. capseformis it is distinguished by larger size, and by the absence of any dark green tints (except rays) upon the posterior expansion of the female. Also the denticulations on the margin of the expansion are stronger and more numerous. Wilson & Clark do not mention these denticulations: but his specimens seem to have been more or less mutilated in this region. My specimens, belong- ing to the type set, show only traces of them, while the fine material I collected myself in the headwaters of the Holston shows them well developed. : The males of T. walkeri and capseformis are very similar: that of walkeri is possibly somewhat larger and of a lighter color, yellow- ish brown, with light green rays, while that of capseformis is green- ish olive, with dark green rays. But these differences are rather un- certain. However, I was never put to the task of separating them, since I have never found the two species associated. T. walkeri is very local in the upper Tennessee region. I found | it only in the South Fork Holston at Emmett, Sullivan Co., Tenn., and at Barron, Washington Co:, Va. (Walker has it from Barron) ; and further, I found it in Middle Fork Holston, at Chilhowie, Smyth Co., Va. At the latter place it was not rare. In addition, the Carnegie Museum has it from Flirt River and © Hurricane Creek at Gurley and Maysville, Madison Co., Ala. There are also specimens at hand from the type locality. UPPER TENNESSEE DRAINAGE. 593 Type locality: East Fork Stones River, Walterhill, Rutherford Co., Tenn. (cotypes examined). 88. TRUNCILLA CAPS#FORMIS (Lea), 1834. Unio capseformis Lea, ’34—Unio capseformis Lewis, ’71.—Trun- cilla florentina and capseformis Ortmann, ’12b, p. 359 (anatomy). —Truncilla capseformis Ortmann, ’13b, p. 311.—Truncilla cap- seformis Goodrich, ’13, p. 95.—Truncilla capseformis Simpson, "14, p. 29. The specimen of “ florentina,’ of which I have described the anatomy, was actually a capseformis. Distinguished from the two preceding species by the combina- tion of the following characters: shell rather compressed, with the beaks not much elevated; expansion of the female uniformly dark green, with only small, few, and remote denticles on the margin. There are differences in the soft parts of T. turgidula, florentina, walkeri, and capseformis, which will be discussed elsewhere. T. capseformis is very abundant in the upper Tennessee drain- age, all over the region. It is in the Tennessee below Knoxville, in Powell, Clinch, Holston, Nolichucky, Little Pigeon, and goes up, in the French Broad, to Asheville, N. Car. (Walker coll.). It is also in Little Tennessee, at Coytee, Loudon Co., Tenn. (Walker coll.). In the Powell, it has been traced up to Shawanee, Claiborne Co., Tenn.; in the Clinch, to Cedar Bluff, Tazewell Co., Va.; in the North Fork Holston, to Mendota, Washington Ca., Va. (also in Big Mocassin Creek) ; in the South Fork Holston, it is at Pactolus, Sullivan Co., Tenn., but not farther up. Type locality: Cumberland River. The above enumeration contains nearly all nominal species ever recorded from the upper Tennessee region. However, there are yet two additional ones, which have not been mentioned: Unto apacus Haldeman (1824)—Holston River, Tenn. Pleurobema abacus Simpson, ’14, p. 810. A spurious species, which never has been positively recognized. Specimens under this name in the Lea collection (U. S. Nat. Mus.) 594 ORTMANN—NAYADES OF are, as Simpson states, much like “ Plewrobema appressus,” that is to say, Lexingtonia dolabelloides conradi (Van.). I have examined : these in Washington. Also specimens in the Walker collection, la- 3 beled abacus, from Flint River, Gurley, Madison Co., Ala., are this. MARGARITANA QUADRATA Lea (1861)—Eastern Tennessee. Symphynota quadrata Simpson, ’14, p. 487. The type is lost. This species never has been recognized. < have the suspicion, from description and figure, that it is identical with Alasmidonta minor (Lea), 1845. LIST OF LOCALITIES, AND OF THE NAYAD-FORMS z FOUND AT THEM. The following list is submitted for two reasons: first, to give an idea of the richness of the material upon which this paper is founded ; second, to facilitate, for subsequent collectors, the search for cer-, tain forms. There will be a time, not far distant, when the fauna of many of the localities will have deteriorated or disappeared in consequence of stream pollution, and thus it is important to know all the localities where a given form has been found. The exact location of all the collecting stations is given on the accompanying map, sO that also a change of local geographic nomenclature will not inter- - fere. (See page 523.) The localities have been arranged according to river systems, be- ginning in the northwest (Powell), and proceeding downstream and eastward. The smaller streams not belonging to the headwaters, have been placed together at the end of each system. The Ten- nessee proper stands at the end of the list. ) Forms found intergrading at one locality are connected by. braces. In several instances of well-known and easily recognized ; species, I have not taken home specimens at certain localities, but only seen them (mostly dead shells). These are marked “seen.” The record of this fact always was made in the field with actual specimens before me, and is absolutely reliable. UPPER TENNESSEE DRAINAGE. 595 PoWELL DRAINAGE. North Fork Powell River, Big Stone Gap, Wise Co., Va. Walker coll. (C. C. Adams, Sept. 4, ’99). _ 1. Fusconaia pilaris bursa-pastoris 4. Eurynia nebulosa 2. F. barnesiana bigbyensis 5. Lampsilis fasciola 3. Medionidus plateolus South Fork Powell River, Big Stone Gap, Wise Co., Va. *— Walker coll. (Adams, Sept. 4, 99); {== Carnegie Mus. (Ortmann, May 15 and Sept. 6, ’13). Fi. Fusconaia pilaris bursa-pastoris +7. Alasmidonita minor 72. F. barnesiana 78. Strophitus edentulus *43. F. barnesiana bigbyensis t9. Ellipsaria subtenta 74. Lexingtonia dolabelloides con- *10. Toxolasma lividum radi *711. Medionidus plateolus *75. Pleurobema oviforme argen- *{12. Eurynia nebulosa teum 713. E. vanuxemensis 76. Lasmigona badia Powell River, Olinger, Lee Co., Va. Walker coll. (Adams, Sept. 5, ’99). 1. Fusconaia pilaris bursa-pastoris 7. Lasmigona costata 2. F. cuneolus 8. Alasmidonta marginata 3. F. barnesiana bigbyensis ; 9. Ellipsaria subtenta 4. Lexingtonia dolabelloides con- 10. Medionidus plateolus radi 11. Eurynia perpurpurea 5. Pleurobema oviforme 12. E. nebulosa 6. Elliptio dilatatus 13. Lampsilis ovata ventricosa Powell River, Dryden, Lee Co., Va. * — Walker coll. (Adams, Sept. 3, ’99) ; -==Carn. Mus. (Ort- mann, Sept. 7, 713). fi. Fusconaia pilaris bursa-pastoris 79. Alasmidonta marginata f2. F. barnesiana fio. Pegias fabula #43. F. barnesiana bigbyensis fii. Strophitus edentulus *44. Lexingtonia dolabelloides con- *}12. Ellipsaria subtenta radi *713. Medionidus plateolus 75. Pleurobema oviforme *414. Eurynia nebulosa *76. Elliptio dilatatus #715. E. vanuxemensis +7. Lasmigona badia 716. Lampsilis ovata ventricosa *78. L. costata 4717. L. fasciola 596 ORTMANN—NAYADES OF Cane Creek, Pennington Gap, Lee Co., Va. Walker coll, 1. Fusconaia pilaris bursa-pastoris 2. Pleurobema oviforme Puckell Creek, Pennington Gap, Lee Co., Va. Walker coll. 1. Fusconaia cuneolus 3. Pleurobema oviforme — 2. Lexingtonia dolabelloides con- radi Wallen Creek, Lee Co., Va. (near Jonesville)... * == Walker coll.; }==Carn. Mus. (from G. H. Clapp). *1, Fusconaia barnesiana bigbyensis *4. Medionidus plateolus +2. Elliptio dilatatus +5. Eurynia nebulosa *3. Pegias fabula Powell River, Lytton Mill, Pennington Gap, Lee Co., Va. Walker coll. (Adams, Sept. 1, ’99). 1. Fusconaia pilaris bursa-pastoris 8. Lasmigona costata 2. F. cor analoga 9. Ellipsaria fasciolaris — 3. F. barnesiana bigbyensis 10. Nephronaias pectorosa 4. Quadrula cylindrica strigillata 11. Eurynia nebulosa 5. Lexingtonia dolabelloides con- 12. E. vanuxemensis radi 13. Lampsilis fasciola 6. Pleurobema oviforme 14. Truncilla haysiana 7. Elliptio dilatatus Powell River, Jonesville, Lee Co., Va. * = Walker coll.; j==Carn. Mus. (Hartman coll.). *1. Fusconaia pilaris bursa-pastoris *4. Ellipsaria subtenta *2. Pleurobema oviforme ‘ *5. Toxolasma lividum *3, Elliptio niger *+6. Lemiox rimosus Powell River, Rose Hill, Lee Co., Va. : Walker coll. (Adams, Aug. 5, ’or). 1. Fusconaia pilaris bursa-pastoris 8. Nephronaias pectorosa 2. F. cor analoga 9. Lemiox rimosus 3. F. barnesiana bigbyensis - 10. Medionidus plateolus 4. Pleurobema oviforme 11. Eurynia nebulosa 5. Elliptio dilatatus 12. Lampsilis fasciola 6. Alasmidonta marginata 13. Truncilla interrupta 7. Ellipsaria fasciolaris UPPER TENNESSEE DRAINAGE. 597 Powell River, Shawanee, Claiborne Co., Tenn. Walker coll. (Adams, Aug. 31, ’99). 1. Fusconaia pilaris a | 2. F. pilaris lesueurian 3. F. cuneolus : 4. F. cor analoga 5. F. barnesiana bigbyensis 6. Amblema plicata costata 7. Pleurobema oviforme 8. P. oviforme psceirellal 9. Elliptio niger 10. E. dilatatus 11. Ellipsaria subtenta 12. Dromus dromas caperatus 13. Nephroaias ligamentina gibba 14. Nephronaias pectorosa 15. Toxolasma lividum 16. Medionidus plateolus 17. Eurynia nebulosa 18. Lampsilis ovata 19. L. ovata ventricosa 20. L. fasciola 21. Truncilla triquetra 22. T. interrupta 23. T. haysiana 24. T. torulosa gubernaculum 25. T. capseformis Powell River, Bryant Shoals, Claiborne Co., Tenn. Walker coll. (Adams, Aug. 30, ’99). 1. Fusconaia pilaris bursa-pastoris 2. F. cor analoga 3. Amblema plicata costata 4. Quadrula pustulosa 5. Quadrula cylindrica strigillata 6. Plethobasus cyphyus 7. Lexingtonia dolabelloides con- radi 8. Pleurobema oviforme 9. P. oviforme ee 10. Elliptio dilatatus 11. Lasmigona costata 12. Alasmidonta marginata 13. Ellipsaria fasciolaris 14. E. subtenta 15. Dromus dromas caperatus 16. Nephronaias ligamentina gibba 17. N. pectorosa 18. Lemiox rimosus 19. Medionidus plateolus 20. Eurynia nebulosa 21. Lampsilis fasciola 22. Truncilla interrupta 23. T. capseformis Powell River, Combs, Claiborne Co., Tenn. Carn. Mus. (Ortmann, Sept. 12, ’13 and Sept. 13, ’15). 1. Fusconaia pilaris bursa-pastoris 2. F. pilaris lesueuriana 3. F. cuneolus 4. F. cor analoga 5. F. barnesiana bigbyensis 6. Amblema plicata costata 7. Quadrula cylindrica 8. Plethobasus cyphyus 9. Lexingtonia dolabelloides con- radi 10. Pleurobema oviforme 11. P. oviforme gales SE 12. Elliptio dilatatus 13. Lasmigona costata 14. Alasmidonta marginata 15. Strophitus edentulus 16. Ellipsaria fasciolaris 17. E. subtenta 18. Dromus droma caperatus 20. Nephronaias pectorosa 19. Nephronaias ligamentina gibba 21. Paraptera fragilis 22. Proptera alata 23. Lemiox rimosus 598 ORTMANN—NAYADES OF 24. Medionidus plateolus 31. L. fasciola 25. Eurynia fabalis 32. Truncilla triquetra 26. E. nebulosa 33. T. interrupta 27. E. vanuxemensis 34. T. haysiana 28. E. recta 35. T. lewisi 29. Lampsilis ovata 36. T. torulosa oubernaculum ‘ 30. L. ovata baie Le Pao iF Copsey erent Powell River, Green’s Ford, Union Co., Tenn. Walker coll. (Adams, Aug. 23, ’99). (Adams says: Campbell Co., Walker: Claiborne Co.; the lcalty is, where the three counties meet, and the ford is, according to Ieica,s -—sheet Maynardville—rather in Union Co.) 1. Fusconaia pilaris bursa-pastoris 10. Strophitus edentulus 2. F. pilaris lesueuriana 11. Ellipsaria fasciolaris 3. F. cuneolus 12. E. subtenta 4. F. barnesiana 13. Toxolasma lividum 5. Amblema plicata costata 14. Medionidus plateolus 6. Quadrula cylindrica 15. Eurynia fabalis 7. Elliptio niger 16. E. nebulosa 8. E. dilatatus 17. Truncilla interrupta 9. Lasmigona costata 18. T. capseformis Powell River, Powell River P. O., Campbell Co., Tenn. Walker coll. (Adams, Aug. 23, ’99). 1. Amblema plicata costata 5. Eurynia fabalis 2. Elliptio dilatatus 6. Truncilla triquetra 3. Lasmigona costata 7. T. capseformis 4. Ellipsaria subtenta CLINCH DRAINAGE. North Fork Clinch River, Tazewell, Tazewell Co., Va. Walker coll. (Adams). 1. Alasmidonta minor 2. Eurynia nebulosa Clinch River, Tazewell, Tazewell Co., Va. Carn. Mus. (Ortmann, Sept. 19, ’12). 1. Fusconaia barnesiana bigbyensis 2. Lasmigona badia UPPER TENNESSEE DRAINAGE. 599 Clinch River, Cedar Bluff, Tazewell Co., Va. Carn, Mus. (Ortmann, Sept. 20, ’12, and May 11, 13). 1. Fusconaia pilaris bursa-pastoris 2. F. barnesiana bigbyensis 3. Quadrula cylindrica strigillata 4. Lexingtonia dolabelloides con- radi 5. Pleurobema oviforme 6. P. oviforme acai 7. Elliptio dilatatus 8. Lasmigona badia 9. L. costata 10. Alasmidonta minor 11. Strophitus edentulus 12. Ellipsaria subtenta 13. Medionidus plateolus 14. Eurynia perpurpurea 15. E. nebulosa 16. Lampsilis ovata ventricosa 17. L. fasciola 18. Truncilla capseformis Clinch River, Richland, Tazewell Co., Va. Carn. Mus. (Ortmann, Sept. 20, *12). 1. Fusconaia pilaris bursa-pastoris 2. F. barnesiana 3. F. barnesiana bisbronsist 4. Quadrula cylindrica strigillata 5. Lexingtonia dolabelloides con- radi 6. Pleurobema oviforme 7. P. oviforme cade 8. Elliptio dilatus 9. Lasmigona badia 10. L. costata 11. Alasmidonta minor 12. A. marginata 13. Strophitus edentulus 14. Ellipsaria subtenta 15. Medionidus plateolus 16. Eurynia perpurpurea 17. E. nebulosa 18. Lampsilis ovata ventricosa 19. L. fasciola Clinch River, Raven, Tazewell Co., Va. Carn. Mus. (Ortmann, Sept. 21, ’12). 1. Fusconaia pilaris bursa-pastoris 2. F. barnesiana bigbyensis 3. Quadrula cylindrica strigillata 4. Lexingtonia dolabelloides con- radi 5. Pleurobema oviforme 6. P. oviforme argenteum 7. Elliptio dilatatus 8. Lasmigona costata 9. Strophitus edentulus 10. Ellipsaria subtenta 11. Medionidus plateolus 12. Eurynia perpurpurea 13. E. nebulosa 14. Lampsilis ovata ventricosa 15. L. fasciola 16. Truncilla haysiana 17. T. capseformis Clinch River, Cleveland, Russell Co., Va. *— Walker coll. (Adams, Aug., 99); {==Carn. Mus. (Ort- mann, May 13, 13). t1. Fusconaia pilaris bursa-pastoris *+2. F. cor analoga +3. Amblema plicata costata *+4. Quadrula intermedia 600 ORTMANN—NAYADES OF +5. Q. cylindrica strigillata 715. Strophitus edentulus *+6. Lexingtonia dolabelloides con- *+16, Ellipsaria fasciolaris radi *+17. E. subtenta +7. Pleurobemaobliquumcoccineum *+18. Nephronaias pectorosa 78. P. oviforme *+19. Medionidus plateolus to. P. oviforme argenteum *+20. Eurynia fabalis : *+10. Elliptio dilatatus *+21. E. perpurpurea *t+11. Lastena lata j22. E. nebulosa *+12. Lasmigona costata +23. Lampsilis ovata ventricosa +13. Alasmidonta minor *+24. L. fasciola +14. Alasmidonta marginata *+25. Truncilla capseformis Clinch River, Fink, Russell Co., Va. Carn. Mus. (Ortmann, May 12, 713). 1. Fusconaia pilaris bursa-pastoris 9. Alasmidonta marginata 2. F. cor analoga 10. Ellipsaria fasciolaris 3. F. barnesiana bigbyensis 11. Nephronaias pectorosa ' 4. Amblema plicata costata 12. Medionidus plateolus 5. Quadrula cylindrica strigillata 13. Eurynia nebulosa 6. Pleurobema oviforme argenteum 14. Lampsilis ovata ventricosa 7. Elliptio dilatatus 15. L. fasciola ' 8. Lasmigona costata 16. Truncilla capseformis Clinch River, St. Paul, Wise Co., Va. * == Walker coll. (Adams, Aug. 8, ’99); +—=Catn. Mus. (Ort- mann, May 14, 13). ue t1. Fusconaia pilaris bursa-pastoris *+13. Ellipsaria fasciolaris +2. F. cor analoga *+14. Ellipsaria subtenta : +3. F. barnesiana bigbyensis +15. Nephronaias ligamentina my “i *+4. Amblema plicata costata *+16. N. pectorosa ‘+5. Lexingtonia dolabelloides con- +17. Lemiox rimosus radi *+18. Medionidus plateolus +6. Pleurobema oviforme argenteum t19. Eurynia fabalis *+7. Elliptio dilatatus j20. E. perpurpurea | +8. Lastena lata ‘ *+21. E. nebulosa *+9. Lasmigona costata f22. E. recta t10. Alasmidonta minor 423. Lampsilis ovata ventricosa *+11. A. marginata *+24. L. fasciola t12. Strophitus edentulus *+25. Truncilla capseformis Clinch River, Dungannon, Scott Co., Va. Walker coll. (Adams, Aug. 11, ’99). 1. Amblema plicata costata 2. Truncillatorulosagubernaculum UPPER TENNESSEE DRAINAGE. 601 Clinch River, Clinchport, Scott Co., Va. *— Walker coll. (Adams, Aug., 99); ~==Carn. Mus. (Ort- mann, Sept. 8, ’13). *+1. Fusconaia pilaris bursa-pastoris) , *718. Alasmidonta marginata *+2. F. pilaris lesueuriana *+19. Ellipsaria fasctolaris 73. F. cuneolus *+4. F. cor analoga +5. F. barnesiana bigbyensis *+6. Amblema plicata costata *7. Quadrula intermedia 78. QO. cylindrica *9. Rotundaria tuberculata tio. Plethobasus cyphyus fui. Lexingtonia dolabelloides con- radi 712. Pleurobemaobliquum coccineum 713. P. oviforme *714. Elliptio niger *715. E. dilatatus *16. Lastena lata *+17. Lasmigona costata 720. E. subtenta *;21. Nephronaias ligamentina gibba *i22. N. pectorosa *+23. Proptera alata *+24. Medionidus plateolus 725. Eurynia perpurpurea 726. E. nebulosa 727. E. recta 728. Lampsilis ovata *+29. L. ovata Sec out *730. L. fasciola 731. Truncilla triquetra *+32. T. interrupta *+33. T. torulosa gubernaculum *734. T. capseformis Clinch River, Speer’s Ferry, Scott Co., Va. Carn. Mus. (Ortmann, July 8, ’13). 1. Fusconaia pilaris bursa-pastoris 2. F. cuneolus 3. F. cor analoga 4- Amblema plicata costata 5. Quadrula cylindrica 6. Rotundaria tuberculata 7. Pleurobema oviforme argenteum 8. Elliptio dilatatus 9. Lasmigona costata 10. Alasmidonta minor 11. A. marginata 12. Ellipsaria fasciolaris _ 13. Nephronaias pectorosa. (seen) 14. Toxolasma lividum 15. Medionidus plateolus 16. Eurynia fabalis 17. E. trabalis 18. E. perpurpurea 19. E. nebulosa 20. E. vanuxemensis 21. Lampsilis ovata ventricosa 22. Truncilla triquetra 23. T. interrupta 24. T. lenior 25. T. capseformis Clinch River, Church Ford, Scott Co., Va. Walker coll. 1. Fusconaia cor analoga PROC. AMER. PHIL. SOC., VOL. LVII, U, OCT. I, 1918. ns ; He had made intensive studies of various colonies of each of several species and endeavored 21“ New Cycads and Conifers from the Trias of Pennsylvania,” ibid. IQII, pp. 17-21. 22“ The Formation of Ripple Marks, Tracks and Trails,” ibid., 1911 pp 536-547. | 23 “ Variation in Some Jamaican Species of Pleurodonte,” Proc. Acad. Nat. Sci. Phila., 1911, pp. 117-164. Re a ENS TIRES AMOS PEASLEE BROWN. riir to correlate the variations which they exhibited with differences of environment, and to show the effect of isolation in their evolution. To illustrate graphically the extent of their variation he devised some ingenious plottings and curves, based upon actual measurements of each individual shell, for by adopting measurements as his basis of comparison he hoped to eliminate as far as possible the personal equation. This whole investigation illustrated the constant trend of his mind toward mathematical methods. Another paper along very similar lines dealt with variation in two species of Lucidella.** Still another Jamaican study was responsible for a paper on the method of locomotion in certain land snails**—a distinctly original piece of work. He had found that certain species possessed a com- paratively very rapid rate of progression and a careful study dem- onstrated that they had an entirely different method of locomotion from that of the majority of snails. The foot, he found, touched the surface upon which they walked only along its edges, while the © wave motions which traversed it were in the opposite direction to that usually prevailing in these mollusks. Furthermore the shell was carefully balanced on the operculum and swayed from side to side as the animal advanced. The rapidity of the wave motions and the exact rate of progress were worked out in much detail. During the period just described two mineralogical publications were issued in which Brown’s name appears as joint author. One of these, in the preparation of which he was associated with Dr. Persifor Frazer, consisted of a series of tables for the determina- tion of minerals by physical properties,?* while the other in which Dr. Frederick Ehrenfelt was his associate was a report on the min- erals of Pennsylvania?’ published by the Topographical and Geo- logical Survey of the State in 1913. In the summer of 1913 Brown made another trip to the tropics, touching at Georgetown, British Guiana, and spending some time 24“ Variation in Two Species of Lucidella from Jamaica,” ibid., 1913, », 3-21. d c 2 “The Method of Progression of Some Land Operculates from Ja- maica,” The Nautilus, XXIV., No. 8, December, 1910, pp. 85-90. 26“ Tables for the Determination of Minerals by Physical Properties,” Philadelphia, J. B. Lippincott & Co., 1910, pp. i—xiii, I., 1-125. 27“ Minerals of Pennsylvania,” Topog. and Geolog. Survey of Penna., Report No. 9, pp. 1-160, 1913. xiv OBITUARY NOTICES. on the island of Antigua. Here he made a representative collection of fossils and on his return published a comprehensive account of the geology of the island,’* reviewing the literature of the subject and describing his own collection, while in the following year in con- junction with Dr. Pilsbry he published a short paper on the fresh- water mollusca of the Antiguan Oligocene.*® In these contributions eleven new forms are described. His last publication,®° also in con- junction with Dr. Pilsbry, dealt with collections of Oligocene fos- sils from Cartagena, Colombia and Haiti, collected by Mr. Lloyd B. Smith. Twenty-one new species and subspecies were here described. While Dr. Brown’s trips to the tropics were distinctly beneficial and he was able to return to his classroom, nevertheless his recoy- ery was only partial. In 1911 and 1912, moreover, he was twice operated upon for gallstones, an additional strain on his weakened constitution, and during the past two or three years he seemed to have little ambition to engage in any serious scientific research. In spare moments, however, in the seclusion of his home, he —— his microscopical studies, now as ever his chief diversion. Another nervous breakdown at the close of 1916 was followed in the spring by a partial paralysis which compelled him to definitely resign his professorship at the University of Pennsylvania, where he had now been engaged in teaching for over twenty-seven years. In the autumn his condition was further complicated by the develop- ment of a severe carbuncle, which his weakened system was power- less to combat, and he passed away on October 9, 1917, at Atlantic City, N. J., where he had been taken in the hope that the change of air and surroundings might prove beneficial. | Dr. Brown had never married, and after the death of his parents he had continued to reside with his brothers and sisters at their home in Germantown. In his prime he was a strikingly handsome man, tall, broad-shouldered and dark-haired. His mind was always 28 “ Notes on the Geology of Antigua,”.Proc. Acad. Nat. Sci. Phila., 1913, pp. 584-616. 29“ Fresh Water Mollusks of the Oligocene of Antigua,” ibid., 1914, pp. 200-213. 80“ Oligocene Fossils from the Neighborhood of Cartagena, Colombia, with Notes on Haitian Species,” ibid., 1917, pp. 32-41. AMOS PEASLEE BROWN. xv active and alert. His eyes would kindle as his interest was aroused in conversation or congenial occupation, and among his intimates his fine sense of humor was constantly in evidence. His powers of observation were keen and his deductions remarkably accurate. He had well-defined opinions on scientific topics and, while not hesi- tating to express them, he was loath to force them upon others or to engage in argument or controversy, and in assuming without protest whatever tasks were allotted to him he often bore far more than his share of the burdens of life. His quiet unassuming manner attracted those with whom he came in contact, while he possessed none of the qualities that make enemies. In his death science loses an investigator and teacher of excep- tional ability and this Society an officer noted for his devotion and loyalty. : A long line of students will, in years to come, recall with pleasure their association with Amos Brown at the University—his kindli- ness ; his fairness; and his earnestness of purpose. A smaller group of scientific associates will cherish recollections of his reverence for the sciences which he helped to advance and his faithfulness to the trusts that were placed upon him. A still smaller group, who were privileged to know the real man in the intimacy of close companion- ship, will mourn the loss of that which has been described as the easiest thing to speak of, but the hardest to find—a true friend. WITMER STONE. eS) oar fi pas s MINUTES. ili MINUTES. Stated Meeting January 4, 1918. Curator C. L. Dootittie, C.E., Sc.D., LL.D., in the Chair. The decease was announced of Joseph P. Remington, Ph.D.,- on January 1, 1918, in his 71st year. Dr. Witmer Stone read an Obituary Notice of Prof. Amos P. Brown (see page i). Mr. Edgar L. Hewett read a paper on “Our Cultural Heritage from Ancient America,” which was discussed by Dr. Jastrow, Prof. L. W. Miller, Mr. E. S. Balch and Mr. Hewett. The judges of the Annual Election held this day between me hours of two and five in the afternoon reported that the following named members were elected according to the Laws, Regulations and Ordinances of the Society, to be the Officers for the ensuing year: President. William B. Scott. Vice-Presidents. Albert A. Michelson, George Ellery Hale, Joseph G. Rosengarten. ; Secretaries. I. Minis Hays, Arthur W. Goodspeed, Harry F. Keller, Bradley Moore Davis. Curators. Charles L. Doolittle; William P. Wilson, Leslie W. Miller. . iv MINUTES. Treasurer. Henry La Barre Jayne. Councillors. (To serve for three years.) Bertram B. Boltwood, Ernest W. Brown, Francis B. Gummere, Herbert S.: Jennings. The Finance Committee and the Treasurer presented their an- nual reports. Stated Meeting, February 1, 1918. WituraM B. Scort, Sc.D., LL.D., President, in the Chair. Dr. Alonzo E. Taylor, a newly elected member, subscribed the Laws and was admitted into the Society. The Society commemorated the Centenary of the Death of Caspar Wistar, M.D., and the following papers on Dr. Wistar were read: “As a Scientist and Philosopher,” by Dr. I. Minis Hays. “As a Human Anatomist,” by Dr. George A. Piersol. ““As a Comparative Anatomist,” by Prof. William B. Scott. Stated Meeting, March 1, 1018. Wi1am B. Scott, Sec.D., LL.D., President, in the Chair. Mr. Pierre S. duPont, a newly elected member, subscribed the Laws and was admitted into the Society. The decease was announced of Samuel G. Dixon, M.D., on Feb- ruary 26, 1918, zt. 67. 7 Prof. Vernon Kellogg read a paper entitled “ Behind the Ger- man Lines in Belgium and France.” Mr. Benjamin Smith Lyman presented notes on “ The Soul,” “The Word of God,” “Classical Studies,” and “Of.” (See page 627.) Stated Meeting, April 5, roré. WituiaMm B. Scort, Sc.D., LL.D., President, in the Chair. The decease was announced of John Fulton, at Johnstown, Pa., on January 20, 1916, zt. 89. i _ . * » ee ee ee eS ™ jigted tact MINUTES. Vv Mr. Eli K. Price read a paper on “ The Park System of Phila- delphia,” which was discussed by Prof. Miller, Dr. Keen, Prof. Scott and Mr. Bryant. Mr. Price, on behalf of the Committee on the Henry M. Phillips Prize, presented the following report: “The Committee on the Henry M. Phillips Prize have the honor to report that the Judges appointed by the Society to pass on the essays that might be submitted on ‘The Relation of the Initiative, Referendum and Recall,’ in competition for the Prize, have found to .their regret that no essay on the chosen subject has been received of such originality and importance as to justify an award of the Prize.” “The Committee therefore recommend the adoption of the fol- ‘lowing resolutions: “Resolved, That no award of the Phillips Prize be made in pursuance of the competition invited for the year 1918. “ Resolved, That the thanks of the Society be extended to the Hon. Charles Matteson, of Providence, Prof. Charles E. Merriam, ‘of Chicago, Dr. Westel W. Willoughby, of Washington, Prof. Henry J. Ford, of Princeton, and the Hon. Hampton L. Carson, of Philadelphia, for their valuable services as Judges of the competi- tion.” The Report was accepted and the resolutions appended thereto were adopted. Stated General Meeting, April 18, 19 and 20, 1918. Thursday Afternoon, April 18, 1918. Opening Session, 2:30 o’clock. WiuuiaM B. Scott, Sc.D., LL.D., President, in the Chair. The following papers were read: “Efforts of Food Control under Queen Elizabeth,” by Edward P. Cheyney, A.M., LL.D., Professor of European History, University of Pennsylvania. vi MINUTES, “Control of Commerce in War Time,” by William E. Lingel- bach, Professor of Modern European History, University of Pennsylvania. “The Influence of Russian Political Parties on Domestic and International Questions,” by Alexander Petrunkevitch, Ph.D., Professor of Zodlogy, Yale University. “Problems of War Finance,” by Thomas S. Adams, Ph.D., Professor of Political Economy, Yale University. : “Control of Railroads of the United States,” by Emory R. Johnson, Sc.D., Professor of Transportation and Commerce, University of Pennsylvania. “The Sanitation of Camps,” by Col. Frederick F. Russell, Med- ical Corps, U. S. A. Friday, April 19. Morning Session, 10:30 o’clock. J. G. RosEnGARTEN, LL.D., Vice-President, in the Chair. Dr. Frank Dawson Adams, a recently elected member, sub- scribed the Laws and was admitted into the Society. The following papers were read: “The Art of George Catlin,” by Edwin Swift Balch, A.B., of Philadelphia (see p. 144), which was discussed by Dr. Hol- land. “Surgical Shock,” by William T. Porter, M.D., LL.D., Pro- fessor of Comparative Physiology, Harvard University, which was discussed by Dr. Keen. “The Relations of French and American Thought in the 18th and 19th Centuries,” by Albert Schinz, A.M., Ph.D., Pro- fessor of French Literature, Smith College, Northampton, Mass. “Type-writer Keyboards; An Inquiry for Some Rational Ones,” by Charles R. Lanman, Ph.D., LL.D., Professor of Sanskrit, Harvard University. “ Changing of the Sex-Ratio of the Rat,” by Helen D. King, Associate Professor of Embryology, Wistar Institute, Phila- delphia, which was discussed by Prof. Jennings. MINUTES. vii “History of the Study of Greek Vase Paintings,” by Stephen B. Luce, Curator of Greek Antiquities, Museum of the Uni- versity of Pennsylvania. (See page 649.) “The Naiades of the Upper Tennessee Drainage,” by Arnold E. Ortmann, Ph.D., Sc.D., Professor of Physical Geography, University of Pittsburgh. (See p. 521.) “ A New Type of Insect Larva,” by William Morton Wheeler, Ph.D., Sc.D., Professor of Economic Entomology, Bussey Institution, Harvard University. (See p. 293.) “A Critical Survey of the Sense of Hearing in Fishes,” by George H. Parker, Sc.D., Professor of Zoology, Harvard University. (See p. 69.) “The Perfecting Principle,” by L. H. Bailey, LL.D., Late Pro- fessor of Horticulture, Cornell University. “Medicinal Plants—Present and Future Supplies,” by Henry Kraemer, Ph.D., Head of the Dept. of Pharmacology, Uni- versity of Michigan. “Parasitism among the Red Alge,” by William A. Setchell, Ph.D., Professor of Botany, University of California. (See p. 155.) Afternoon Session, 2 o’clock. ALBERT A. MicHEtson, Ph.D., Sc.D., LL.D., F.R:S., Vice-President in the Chair. The following papers were read: I. “ Preliminary Notes of Some New Species of Agarics” (see Pp. 354), and . IT. “ The Genus Galerula in North America,” by George F. At- kinson, Ph.D., Professor of Botany, Cornell University. (See p. 357-) . “Temperature, Imbibition and Growth,” by D. T. MacDougal, Ph.D., LL.D., Director of the Department of Botanical Re- search, Carnegie Institution of Washington. “Variation in Blueberry Hybrids,” by Frederick V. Coville, Curator of the U. S. National Herbarium, Department of Vili MINUTES. Agriculture, Washington, D. C., which was discussed by Pro- fessors Harshberger, Webster, Goodspeed and L. H. Bailey. “Organization, Reproduction and Heredity in Pediastrum,” by Robert A. Harper, Ph.D., Professor of Botany, Columbia University. (See p. 375.) I. “ Dependence of the Earth’s Magnetic State on Solar Condi- tions 1888-1916,” and II. “ The Potentials of Certain Magnetized Bodies,” by Louis A. Bauer, Ph.D., Sc.D., Director of the Department of Terres- - trial Magnetism, Carnegie Institution of Washington, which were discussed by Professors Webster, and Michelson. “ Development of Magnetic Susceptibility in Manganese Steel by Prolonged Heat Treatment,’ by Charles Francis Brush, Ph.D., Sc.D., LL.D., of Cleveland (see p. 344), which was discussed by Dr. Bauer. “ Accelerometers,” by N. W. Akimoff, of Philadelphia, which — was discussed by Dr. Webster. “Luminescence of Radium Salts,” by D. H. Kabakjian, As- — sistant Professor of Physics, University of Pennsylvania, and E. Karrer, of Philadelphia. Friday evening, 8:30 o’clock. Lieut.-Col. Robert Andrews Millikan, Ph.D., Sc.D., spoke on “ Science in Relation to the War.” Saturday, April 20. Executive Session, 9:30 o’clock. Witu1aM B. Scort, Sce.D., LL.D., President, in the Chair. On the recommendation of. the Officers and Council nominations for foreign membership were suspended by unanimous vote until the number of foreign members is reduced to seventy-five. Prof. Ernest W. Brown, on behalf of the Nominating Commit- tee, recommended that the term of service of the curators, like that of Councillors, be limited to three consecutive years. MINUTES. ix On motion this recommendation was unanimously adopted. Pending nominations for membership were read and the Society proceeded to an election. The tellers reported that the following nominees had been elected to membership: Residents of the United States. Henry Andrews Bumstead, A.B., Ph.D., New Haven. Philip Powell Calvert, Ph.D., Philadelphia. Clarence Griffin Child, Ph.D., L.H.D., Philadelphia. William T. Councilman, A.M., M.D., LL.D., Boston. Victor George Heiser, M.D., New York. Herbert C. Hoover, B.A., LL.D., Washington. Ale5 Hrdlicka, M.D., Washington. Gilbert Newton Lewis, A.M., Ph.D., Berkeley. Theodore Lyman, Ph.D., Cambridge. J. Percy Moore, Media, Pa. Louis Valentine Pirsson, M.A., New Haven. George Harrison Shull, B.S., Ph.D., Princeton. Joseph Swain, B.L., M.S., LL.D., Swarthmore. William Roscoe ore. A.M., LL.D., Litt.D., L.B.H., Cam- bridge. Samuel Wendell Williston, A.M., M.D., Ph.D., Sc.D., Chicago. Foreign Residents. Joseph Jacques Cesaire Joffre, Paris. Paul Painlevé, Paris. Raymond Poincaré, Paris. ERE a BRT remaee Morning Session, 10 o’clock. WiuturaM B. Scott, Sc.D., LL.D., President, in the Chair. Mr. William Roscoe Thayer, a newly elected member, sub- scribed the Laws and was admitted into the Society. The following papers were read: “ Motions in the Stellar Systems Struve 1836 and Struve 208,” by Eric Doolittle, Professor of ee coony, University of Pennsylvania. “The Number of the Spiral Nebulz,” by H. D. Curtis, As- . 3 MINUTES. tronomer, Lick Observatory, Mt. Hamilton, Cal. (see p. 513), which was discussed by Prof. E. W. Brown. “Ttaly in the Triple Alliance,” by William Roscoe Thayer, Litt.D., L.H.D., LL.D., Cambridge, Mass. “Ballistic Experiments by a New (?) Method,” by Arthur Gordon Webster, Sc.D., LL.D., Professor of Physics, Clark University, Worcester, and Mildred Allen. “Some Considerations on the Ballistics of a Gun of Seventy- five Miles Range,” by Arthur Gordon Webster, Sc.D., LL.D., Professor of Physics, Clark University, Worcester, Mass. “The Relation of Deposits of Iron and Coal to the Great War,” by William H. Hobbs, Ph.D., Sc.D., Professor of Geology, University of Michigan. “The Peculiar Geographical Features of Northwestern France and their Bearing on the War,” by William Morris Davis, Sc.D., Ph.D., Professor Emeritus of Geology, Harvard Uni- versity. * “ Rig-Veda Repetitions,” by Maurice Bloomfield, Ph.D., LL.D., Professor of Sanskrit and Comparative Philology, Johns Hopkins University. Afternoon Session, 2 o’clock. Wiu1aM B. Scott, Sc.D., LL.D., President, in the Chair. The following papers were read: “The Babylonian Origin of the Jewish Method of Slaughter,” by Paul Haupt, Ph.D., LL.D., Professor of Semitic Lan- guages, Johns Hopkins University. “ Soldiers’ and Sailors’ Insurance,” by Samuel McCune Lind- say, Ph.D., LL.D., Professor of Social Legislation, Colum- bia University, New York. (See page 632.) Symposium on Food Problems in Relation to the War— “ Physiological Effects of Prolonged Reduced Diet on Twenty- five Men,” by Francis G. Benedict, Ph.D., Sc.D., Director of the Nutrition Laboratory of the Carnegie Institution of Washington. (See p. 479.) “Food Conservation from the Standpoint of the Chemistry of Nutrition,” by Henry C. Sherman, Ph.D., Professor of Food MINUTES. xi Chemistry, Columbia University, New York. (See p. 491.) “Some Economic Aspects of the American Food Supply,” by J. Russell Smith, Ph.D., Professor of Industry, Wharton School of Finance and Commerce, University of Pennsyl- vania. (See p. 501.) “Food Control and Conservation in the United States Army,” by John R. Murlin, Major, Sanitary Corps, U. S. A. Stated Meeting, May 3, 1918. Wituiam B. Scott, Sc.D., LL.D., President, in the Chair. Dr. Philip P. Calvert, a newly elected member, subscribed the Laws and was admitted into the Society. Acknowledgments of election were received from Philip Powell Calvert, Ph.D. William T. Councilman, A.M., M.D., LL.D. Victor George Heiser, M.D. Herbert C. Hoover, B.A., LL.D. Ales Hrdlitka, M.D., Washington. J. Percy Moore. Louis Valentine Pirsson, M.A. George Harrison Shull, B.S., Ph.D. Joseph Swain, B.L., M.S., LL.D. William Roscoe Thayer, A.M., LL.D., Litt.D., L-H.D. Samuel Wendell Williston, A.M., M.D., Ph.D., Sc.D. The following papers were read: a ee Scan wees te dna’ os ees r ee Acre! ti "2 i tien — i dai ee SS ca ji i ae RT ey ee rah li ty er - ri ay a Se ey r 4 2 ae ee Piet ed Pee ne ee Ce eS et Na Oe ae “Tnorganic Evolution from the Astronomic and Atomic As- pects,” by Eric Doolittle, C.E., and Arthur W. Goodspeed, Ph.D., which were discussed by Professor Snyder, Presi- dent Scott, Mr. Willcox, and Dr. Keen. Stated Meeting, November 1, 1918. WiuuiaM B. Scort, D.Sc., LL.D., President, in the Chair. Professor J. Percy Moore and President Joseph Swain, newly- - elected members, subscribed the Laws and were admitted into the Society. xii MINUTES. Letters accepting membership were received from His Excellency Raymond Poincaré, Le Maréchal Joseph Joffre and Prof. H. A. Bumstead and a declination from Prof. Clarence G. Child. A communication from the Rector of the University of Lund stating that the University would celebrate on the 27th of Septem- ber of this year, the 250th Anniversary of its founding was read. The Secretaries were instructed to send to the University the con- gratulations and good wishes of the Society on the occasion. The decease was announced of Prof. Guido Cora, at Piedmont, on October 10, 1917, zt. 66. Grove K. Gilbert, A.B., A.M., LL.D., at Jackson, Mich., on May 1, 1918, et. 75. Frank Miles Day, B.S., M.A., at Philadelphia, on June 15, 1918, et. 57. Rt. Rev. John J. Keane, at Dubuque, Iowa, on June 22, 1918, et. 79. James Douglas, B.A., LL.D., at New York, on June 25, 1918, et. 81. Stephen Farnum Peckham, A.M., on July 11, 1918, zt. 79. William H. Greene, M.D., at Wenonah, N. J., on August 8, 1918, zt. 65. Maxime Bocher, A.B., Ph.D., at Cambridge, Mass., on Sep- tember 12, 1918, et. 51. Mr. Joseph Willcox, at Philadelphia, on October 1, 1918, zt. 89. _ Dr. Isaac Norris, at Florence, Italy, on October 22, 1918, zt. 84. Prof. John A. Miller read a paper on “The Total Solar Eclipse of June 8, 1918,” which was discussed by Prof. Snyder. Special Meeting, November 21, 1918. WitL1AM B. Scott, D.Sc., LL.D., President in the Chair. Dr. ETIENNE BuRNET, of the French Educational Mission to the United States, read a paper on “ Pasteur as a Representative of the French Scientific Spirit.” MINUTES. xili Stated Meeting, December 6, 1918. Witt1aM B. Scort, D.Sc., LL.D., President in the Chair. Letters accepting election to membership were read from Gilbert Newton Lewis, A.M., Ph.D. Theodore Lyman, Ph.D. The decease was announced of, George Francis Atkinson, Ph.D., at Tacoma, on Nevember cake 1918, zt. 65. Charles Richard Van Hise, M.S., LL.D., at Milwaukee, on November rgth, 1918, zt. 61. Samuel A. Green, M.D., at Boston, on December 5th, 1918, zt. 88. Prof. Franklin Edgerton read a paper on “ India’s Place in the Modern World,” which was discussed by Dr. Keen, Prof. R. G. Kent, Mr. Bryant, President Scott and Prof. Edgerton. The President delivered his Annual Address. INDEX. A /.bbott, archzological significance of an ancient dune, 49 Adams, problems of war finance, vi Agarics, preliminary notes on some new species of, 354 Akimoff, accelerometers, viii Algz, parasitism among the red, 155 American food supply, economic as- pects of the sor Ant larve, study of some, 293 Archeological significance of an an- cient dune, 49 Art of George Catlin, 144 Arthur & Bisby, annotated transla- tion of Schweinitz’s two papers giving the rusts of North America, 173 Atkinson, genus Galerula in North America, 357 Atkinson, preliminary notes on some new species of agarics, 354 Atkinson, twin hybrids from crosses Ginothera lamarckiana and fran- ciscana with GZ. pycnocarpa, in the F, and F., 130 B Bailey, the perfecting principle, vii Balch, art of George Catlin, 144 Bauer, dependence of the earth’s magnetic state on solar conditions, Vill Bauer, potentials of certain magnet- ized bodies, viii Benedict, physiological effects of a prolonged reduction in diet on twenty-five men, 479 Biochemical studies of the pitcher liquor of Nepenthes, 112 Bisby, Arthur &, annotated transla- tion of Schweinitz’s two papers giving the rusts of North Amer- ica, 173 Bloomfield, Rig-Veda repetitions, x Brown, Amos P., obituary notice of, i Brush, development of magnetic sus- ceptibility in manganese steel by prolonged heat treatment, 344 Burnet, Pasteur as a representative of the French scientific spirit, xii Cc Catlin, George, art of, 144 Cheyney, efforts of food control under Queen Elizabeth, v Classical education, Coville, variation in blueberry hy- brids, vii Curtis, on the number of spiral neb- © ule, 513 Davis, peculiar geographical features of northwestern France and their bearing on the war, x Diet, physiological effects of a pro- longed reduction in, on twenty-five men, 479 Doolittle, motions in the steilar sys- tems, Struve 1836 and 208, ix Doolittle, inorganic evolution from the astronomic and atomic aspects, xi Dune, archeological significance of an ancient, 49 E Edgerton, Franklin, India’s place in the modern world, xiii Election of officers, i Eye, lighting in its relation to the, 440 F Ferree and Rand, lighting in its re- lation to the eye, 440 Fishes, critical survey of the sense of hearing in, 69 Food conservation from the stand- point of the chemistry of nutri- tion, 491 —— problems in relation to the war, symposium on, 479 —— supply, economic aspects of the American, 501 Fuels, interrelation of fossil, I G Galerula in North America, genus, 357 Goodspeed, inorganic evolution from the astronomic and atomic aspects, xi Greek Vase-painting, brief history of the study of, 649 xiv INDEX. xv H Harper, organization, reproduction and inheritance in Pediastrum, 375 Haupt, Babylonian origin of the Jewish method of slaughter, x Hearing in fishes, critical survey of the sense of, 69 Heiser, American sanitation in the Philippines and its influence on the Orient, 60 Hepburn, biochemical studies of the pitcher liquor of Nepenthes, 112 Hewett, our cultural heritage from ancient America, iii Hobbs, relations of deposits of iron and coal to the Great War, x Hybrids, twins, from crosses (ine: thera lamarckiana and franciscana with CZ. pycnocarpa, in the F, and F,, 130 ‘ Insurance, soldiers’ and sailors’, 632 J Johnson, control of railroads of the United States, vi K Kabakjian, luminescence of radium salts, viii Kellogg, Vernon, behind the German lines in Belgium and France, iv King, changing of the sex-relation of the rat, vi Kraemer, medicinal plants—present and future supplies, vii L Lanman, typewriter keyboards, vi Lighting in its relation to the eye, 440 Lindsay, soldiers’ and sailors’ insur- ance, 632 Lingelbach, control of commerce in war time, vi Luce, brief history of the study of Greek vase-painting, 649 Lyman, brief notes on “ Soul,” “ The Word of God,” “Classical Educa- tion,“ “ Of,” 627 M MacDougal, temperature, inhibition and growth, vii Magnetic susceptibility in manganese “steel by prolonged heat treatment, development of, 344 Millikan, science in relation to the war, Viii Members admitted: Adams, Frank D., vi Calvert, Philip P., xi Moore, J. Percy, xi du Pont, Pierre S., iv Swain, Joseph, xi Taylor, Alonzo E., iv Thayer, William R., ix Members deceased: Atkinson, George Francis, xiii Bocher, Maxime, Ripe Cora, Guido, xii Day, Frank Miles, xii Dixon, Samuel G., iv Douglas, James, xii Fulton, John, iv Gilbert, Grove K., xii Green, Samuel A., xiii Greene, William H., xii Keane, Rt. Rev. John Ve Ki Norris, Isaac, xii - Peckham, Stephen Farnum, xii Remington, J. P., iti van Hise, Charles Richard, xiii Willcox, Joseph, xii Members elected, xi Membership, acknowledgment of election to: Bumstead, H. A., xii Calvert, Philip P., xi Councilman, William T., xi Heiser, George Victor, xi Hoover, Herbert C., xi Hrdlitka, Ales, xi Joffre, Le Maréchal Joseph, xii Lewis, Gilbert N., xiii Lyman, Theodore, xiii . Moore, J. Percy, xi Poincaré, His Excellency Ray- mond, xii Pirsson, Louis Valentine, xi Shull, George H., xi Swain, Joseph, xi Thayer, William Roscoe, xi Williston, Samuel Wendell, xi Membership declined, xii Minutes, iii Merlin, food control and conserva- tion in the army, xi N Nayades (fresh-water mussels) of the Upper Tennessee drainage, with notes on synonymy and dis- tribution, 521 Nebulz, on the number of spiral, 513 xvi Nepenthes, biochemical studies of the pitcher liquor of, 112 Newbold, Syriac dialogue “ Soc- rates,” 99 * Nutrition, food conservation from the standpoint of the chemistry of, 491 0 Obituary notices: Amos P. Brown, i CEnothera lamarckiana and francis- cana with C2. pycnocarpa, in the F, and F., twin hybrids from crosses of, 130 ‘ Of,” 630 Officers, annual election of, i Ortmann, Nayades (fresh-water mussels) of the Upper Tennessee drainage. With notes on synon- ymy and distribution, 521 ? Parasitism among the red alge, 155 Parker, critical survey of the sense of hearing in fishes, 60 Pediastrum, organization, reproduc- tion and inheritance in, 375 Petrunkovitch, influence of Russian political parties on domestic and international questions, vi Porter, surgical shock, vi Philippines, sanitation in the, 60 Phillips Prize, report of committee on the, v. Physiological effects of a prolonged... reduction in diet on twenty-five men, 479 Pitcher liquor of Nepenthes, bio- chemical studies of, 112 Price, Eli K., the park ve of Philadelphia, v R Rand, Ferree and, lighting in its re- lation to the eye, 440 Russell, sanitation of camps, vi Rusts of North America, annotated translation of Schweinitz’s two papers giving the, 173 Ss Sailors’ insurance, soldiers’ and, 632 Sanitation in the Philippines, Ameri- INDEX, can, and its influence on the Orient, 60 Schinz, relations of French and American thought in the eighteenth and nineteenth centuries, vi Schweinitz’s two papers giving the rusts of North America, annotated — translation of, 173 Setchell, parasitism among the red alge, 155 Sherman, food conservation from the standpoint of the chemistry of nutrition, 491 Smith, J. R., economic aspects of the American food supply, 501 “ Socrates,” Syriac dialogue, 99 Soldiers’ and sailors’ insurance, 632 “ Soul,” 627 Steel, development of magnetic sus- ceptibility in manganese, by pro- longed heat treatment, 344 Stevenson, interrelations of the fossil fuels, I Stone, obituary notice of ‘Amos P. ' Brown, 7, i Symposium on food problems in re- lation to the war, 479 Syriac dialogue “ Socrates,” 99 T Thayer, Italy in the Triple Alliance, x U Upper Tennessee drainage, the Na- yades of the, 521 v Vase-painting, brief history of the study of Greek, Ww Webster, ballistic experiments by a new method, x Webster, on the ballistics of a gun of seventy-five miles’ range, x Wheeler, study of some ant larve with a consideration of the phe. and meaning of the social habit among insects, 293 Wistar centenary, iv “Word of God,” 628 sate elie + ‘adA] UIM} DUDIYIADULD) JO UOLVB19U03 ‘UMOYS [JOM S4ajoBsIeYD Jnq oanjyeur A][NY puosas ‘pupryrapuny “PX wdavaougkd “sp *(VSI=) & ‘OY jou oayaso4 ‘juerd poayepeq B fadAy UIMy Ddavv0WIN “1 “D1 ‘pdavr0urtd "sp X VUunryIavUey] VAIYJOUTT) SSO1I JO VONRIIUIT 4S1VI — ALVA | 34V1d "HAT (008 ‘HdOSOTIHd ‘WY SONIDSS00"d ‘SOARD] OPJOSOA [WIA] ‘UOTPBIIUDS Puodas dy} JO Spliq -AY ULM} aYT YIM ‘odapr0uItd "AH X vunryrapuny sp) ‘SOABA] O}JOSO4 [VOIdA] ‘UOIFeAAUAT PUODaS ay} JO Splaq ‘bP OTT -XY ULM} AY) YUM ‘DuplyraDUtAD] “ZT X DdaprouINd “FP “EC “OT ‘Tl avid WV dA 4 ihe jAd I] 34VId j “HAT (90S “HdOSOTIH, ‘WY SONIGAIO0Ud "adA} UIM} DUDISIIUDAT "9 “DIST ad} UIMy DdavI0NING “S$ “DIY ‘UMOYS [JOM SsoJOVIeYOS ING ‘QanzZeU 9JINH yOU sSayjoesos pue szuRld pazejaq ‘UOTes9UaT Jsay UT SUIM} JO Sazjaso1 ‘DUDISIIUDAL “sy K vdavr0UINd DAgyjoUT) TI] avg | aLvId "IIA"1 (00 “HdOSO1IHd ‘WY SONIGAZO0UNd ‘adA} UIM} DUDISIUDAT “(O1'ZST =) Q “DIY adAy uM} Ddanpr0umh “(UE LSI=) Z ‘og ‘UOIJBIDUDT puodas ayy ut UIMy Ddap20NINd Jo Buryyds spunosvunaf “3 X vdavaouakd “sp ‘Al WV 1g SS'L9T OVLST } \ : LS es 7, i Y - a Al 341d ‘ITA ‘008 'HdOSOTIHd ‘WY SONIGZI00""” — >. PROCEEDINGS AM. PHILOSOPH. Soc. LVI. PLATE V PLATE VI - PROCEEDINGS Am. PHILOSOPH. Soc. LVII. "Sly tae sae Q American Philosophical eG: Society, Philadelphia Ps Proceedings PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY | pe ze he ty ike ‘ freer er fe anloeate aie if Bp tees ath ¥ ioe pst mk ee yeas at Ata oie ae Aye Vie ceadoete ay aOdy ia + 4 ‘ eat rhe bets Tahegeis pause l ay Ay IAT VEE malt Of nee® tee! wit) ads ae Hans oi os Raa 1s feat 464s eee vee, ae = Pat 35 + ® fas CC ae of a ee Pia gS 7 Habeas x z Se PVs Sy Gare reeuene ecosentes He eet ae Ee ee Sretenery eT heeteessia a er Pat ie Ce tae ua scpy et By Tt Sait eds ye As Sees ak #4 Pare: ae 5é ae abe Sve re hey oe A in eho Esty nites a ne mente: + ape set pey, Cath ae oete Ware aa Se siemreet Bes te hs Ra Remo h ct me oh eee ays toate anata te to ~ eo Sacre, ia any eset me tay a nar voeare rs : ine Seek + pare ~S 2 i te seeenracae sane eA Seem ELS f gasps N rye RR aigis 37 SIRS a atte ate Se % etaa: ibe ose % ay AS. anit SER VR SERS SNA Se se iA ac RP Seo ts . od he mempre ere At eolrpeneia” Ks. a Aas nt mre * cee Vy Thee ? Leatnate ay. has er se Ss anaes te pte eeeg er