THE OHIO JOURNAL OF SCIENCE (CONTINUATION OF THE OHIO NATURALIST) Official Organ of the OHIO ACADEMY OF SCIENCE and of the OHIO STATE UNIVERSITY SCIENTIFIC SOCIETY VOLUME XXI— 1920-21 OHIO STATE UNIVERSITY COLUMBUS \<1T2^Q) AUTHORS INDEX. bownocker, j. a 249 Braun, Annette F • 206 Drake, Carl J 201 Fink, Bruce 211 FoERSTi, Aug. F 33 Kennedy, Clarence H 19, 83 Knight, Harry H 107 KosTiR, W. J 267 Kraatz, Walter C 137 Mitra, Swanna K 89 Oberholser, Harry C 217 Rice, E. L 1 Schaffner, J. H 128, 185 Taylor, Mary A 240 .Tiffany, Lewis H 1 13, 272 Weiss, Harry B 104 Williams, Stephen R 123 Hoi the second series, vertical striae dominate; in the two lateral quadrants, horizontal striae dominate. In all cases the dominating strise are crossed transversely by finer stride. Nothing is known of the column, although the facet for its attach- ment is present. Nor is anything definitely known regarding the siimmit of the theca in case of any of the specimens here under investigation. The central part of the figure ace ompanying the original description consists of one of the hexagonal plates belonging to the second series. The nodes at the four extreme comers have weathered away and exposed some of the structure of the interior, but there is no evidence of pores connecting with this interior. The four exterior ridges connecting the central umbonal part of this plate with the nodes at its four extrem- ities are well shown, but the fifth ridge, extending directly downward to the basal angle of the thecal plate, is poorly indicated. From this basal angle, in a complete specimen, the ridge continues downward along the sutures between two of the basal plates, narrowing toward the base, but this part of the specimen is not preserved in the figured type, and the figure here is misleading. The group of horizontal ridges dominating the plates of the second series at mid-height is not ade- quately demarcated from the striae above and below this horizontally ridged zone. 6. Lycocystites nodosus (Hall). (Plate I, Figs. 11 A, B, C, D.) Echinocystites nodosus Hall, 20th Rep. New York State Cab. Nat. Hist., 1868, p. 316, PL 12, Figs. 10, 11.) Lysocystites nodosus Miller, N. Am. Geol. Pal., 1889, p. 259 (Echinocystites preoccupied) . Echinocystites nodosus was based on casts of the interior of the theca of a species described by Hall from the Racine dolomite at Racine, Wisconsin; later the generic name was changed to Lysocystites, Echino- cystites being preoccupied. The type, numbered 2024, is preserved in the American Museum of Natural History in New York City. This type does not present clear evidence of the direction of the sutures between the plates, and therefore has been found of no service in unrav- elling the plate diagram of the species, nor in determining which is the top or bottom of the specimen. The clew to its structure was discovered by Arthur W. Slocom, who noticed that a specimen evidently identical specifically with Lysocystites nodosus fomied the cast of the interior of a second specimen which was a cast of the exterior of a species having the same style of ornamentation as Aethocystites sculptns Miller, from the top of the Laurel limestone, at St. Paul, Indiana. Since the two spec- imens still were attached to each other, there was no possibility of error in the conclusion that they formed parts of the same specimen. With this clew as a guide it was found possible to orient several of the casts of Lysocystites nodosus belonging to Walker Museum at Chicago Uni- versity- The best of these (Plate I, Figs. 11 A, B, C) is numbered 21815 and is from the type locality, Racine, Wisconsin. This has the following structure : 46 The Ohio Journal of Science [Vol. XXI, No. 2, Casts of the interior of the theca. — At that end of the cast which corresponds to Figure 11 on Plate 12 accompanying Hall's original description of Lysocystites nodosus, three short narrow radiating ridges occur, and the outHnes of three basal plates may be detected. Two of these plates are truncated above and the third is more rhomboid in outline and is angular at the top, as in the basal series of plates in Aetho- cystites. Along the upper margin of this -basal series of plates are five more or less cuneate nodes, equidistant from each other, corresponding in position to the lower part of the sutures between the plates belonging to the second series. These cuneate nodes on the cast correspond in position to the lower series of strong angular nodes on the exterior of the theca. In addition to the five cuneate nodes there are three narrow vertical ridges, usually boldly defined, located at the upper end of the sutures between the three basal plates, and extending thence upward into the median part of those three plates of the second series which are angular at the base. All of these structures on the cast of the interior of Lysocystites correspond to salient features on the exterior of Aetho- cystites. Orientation of the specimen is facilitated by the fact that that one of the three short radiating ridges which is directed toward one of the nodes belonging to the lower series follows the median line of the narrow rhomboid basal plate. Passing from the node at the top of this plate two nodes toward the left, and thence vertically upward, the anal aperture will be found at the top of the third series of plates, slightly toward the right of this vertically directed line. In Hall's Figure 11, the base of the theca has been interpreted as the summit. The specimen is so oriented as to place the rhomboid basal plate at the top of the figure. The narrow vertical ridge at the lower end of the figure was determined incorrectly as the ovarian aperture; no aperture of any kind being present here. The two other corresponding vertical ridges, at the upper end of the sutures on the right and left of the rhomboid plate are not clearly defined in the cast figured by Hall, and are not indicated in his figure. Figure 10 represents the type in an inverted position, with the supposed ovarian aperture occupying the iniddle of the upper part of the figure. Returning to the Walker Museum specimen described above, a second series of cuneate nodes is located immediately above the first series, at the top of the second series of plates, but with the pointed end directed in the opposite direction. They locate the upper series of nodes visible on casts of the exterior of Aethocystites. The truncated ends of both the lower and upper series of cuneate nodes are separated more or less by a transverse groove from the rest of the nodes. If any pores connecting the plates are present they might be searched for here. The specimens at hand present no evidence on this subject. The upper or third series of plates can be diagrammed readily so as to conform to the system worked out for Aethocystites scidptus, although the suture lines are indicated only faintly. After studying specimen number 21815, just described from the Walker Museum collections, other specimens of casts of the interior of Lysocystites nodosus become readily intelligible. In the Walker Museum Dec, 1920] Cystids and Blastoids 47 of Chicago Universit}', for instance, there are five Racine specimens of Lysocystites nodosus, numbered 18943, all casts of the interior, in dif- ferent states of preservation. Using the entire series, practically all the details already described in the case of specimen number 21815 can be made out. They differ from the latter chiefly in the appearance of the two series of nodes (Plate I, Fig. 11 D). These are not cuneate in outline and are not crossed at what corresponds to their truncated ends by a transverse groove. Moreover, these specimens are smaller in size than the one numbered 21815, but in all other features they so closely resemble the latter as evidently to be congeneric. Since there is no question of specimen number 21815 being congeneric with Aethocystites, the specimens numbered 1S943 belong here also. The latter, however, are identical in character with the type of Lysocystites nodosus, so that the latter also evidently is congeneric with Aethocystites. Exterior surface. — In the preceding lines it was mentioned that the cast of the interior of Lysocystites nodosus had been found by Arthur W. Slocom forming the interior of a specimen of Aethocystites. At the time of my visit to Chicago, this specimen was not accessible, but another specimen of Aethocystites, presenting the cast of the exterior, found in the Cedarville colomite at Wilmington, Ohio, forms number 2193 of the Walker Museum collection at Chicago University, and evidently has some bearing on the subject since the Cedarville dolomite carries a typical Racine fauna. This cast of the exterior shows the very charac- teristic surface markings of Aethocystites, including the nodes at the four angles of the plates of the second series. Only two plates of the second series are preserved, one of them truncated at the base, the other pointed. The original height of this Wilmington specimen is esti- mated at 19 mm.; the height of the Racine specimen numbered 21815, the first one described here, is about the same, but the specimen appears more rotund. Locality and Horizon. — In the type area, Lysocystites nodosus is known definitely only from Racine, Wisconsin. Remarks. — It is possible that there are two species of Lysocystites in the Racine dolomite of Wisconsin. In that case, the term Lysocystites nodosus must be restricted to forms like the type, in which the nodes at the four corners of the plates of the second series are not conspicuously cuneate in outline, and do not have their truncated ends similarly separated from the remainder of the nodes by transverse grooves. Specimen 21815 possibly belongs to a second species, and the cast of the exterior of Aethocystites, numbered 2193, from Wilmington, Ohio, may not belong to either of these species. For the present, however, all are regarded as belonging to a single species, the term Lysocystites nodosus covering all the forms here discussed. 48 The Ohio Journal of Science [Vol. XXI, No. 2, Use of Term Lysocystites. — Since the generic name Lysocystites was proposed in 1889, five years earlier than Aethocystites, the latter is abandoned. Relationship. — Lysocystites was referred by Bather (The Echinoderma, 1900, p. 70) to the CryptocrinidcB, the sole family of the Aporita. The relationship is even closer than hitherto suspected. In Cryptocrinus, from the Ordovician of Russia, there are three basal plates formed by the fusion of two pairs of the original five, the unfused plate being in the right anterior interradius. Alternating with the five original basal plates is a second series of five plates, all free from each other and hexa- gonal in outline, followed in turn, in alternating order, by a third series of five plates, subpentagonal in outline. The anus lies between two plates of the third series, separated from the nearest plate of the second series by a small supplementary plate. In all of these features Cryptocrinus closely resembles Lysocystites. The latter differs chiefly in the more elevated position of the anal aperture, the latter being located at the top of the third series of plates rather than between two of the latter. None of the plates of the fourth series, nor any of the tegminal plates at the extreme summit of the theca, are known in case of Lysocystites. Holocystites Hall. Holocystites Hall, 20th Rep. New York State Cab. Nat. Hist., 1868. p. 311, 380. Genotype: Holocystites cylindriciis Hall. Megacystites Hall, Addenda of preceding report, p. 380. Substitute for Holocystites. Trematocystis Jaekel, Stammesgeschichte der Pelmatozoen, 1899, p. 414. Genotype: Holocystites subglobosus (Miller), PI. 4, Fig. 2. Holocystites of Hall. — The genus Holocystites was founded by Hall on six species of cystids which agreed in having the plates arranged in more or less alternating transverse or more or less alternating vertical rows, the plates of the more primitive species showing a marked tendency toward a predominance of hexagonal outlines. The oral aperture was terminal and the anal aperture was eccentric but only a short distance away from, the oral one. The arms evidently were not recumbent and there was no trace of their former location. The first three species described in the accompanying text agreed in being distinctly elongated in a vertical direction, the first species of this series being called Holocystites cylindricus on this account. In reality, three forms are figured under this name, of which figure 4 on plate 12 is regarded here as the type. The second described species, Holocystites alternatus, differs from the first by the intercalation of numerous sup- plementary plates between the transverse rows of more primitive plates. Dec, 1920] Cystids and Blastoids 49 The third species, Holocysiites abnormis, differs from both the preceding in having the intercalated plates inserted in a more or less transverse row beneath the fourth series of plates, counting downward from the top of the theca. Of these three species, Holocystites cyUndricus is regarded as the genotype. The other three species, Holocystites winchclli, H. ovatus, and H. scutellatus, are oval or approximately spherical in form, and to these must be added H. sphcericus described by Winchell and Marcy from the Racine of the Chicago area, and H. jolietensis described by Miller from the Niagaran of Joliet, Illinois. Holocystites of Miller. — The 40 so-called species of Holocystites described by Miller from the Osgood limestone of Indiana, one species being cited as thovigh from the top of the Laurel limestone in the Waldron area, all agree in having the plates pierced by diplopores, the pores of each pair being connected just beneath the surface of the plates by peculiar channels frequently resembling the Greek letter Omega. Both the oral and the anal aperture have polygonal margins, the polygonal outline of the oral aperture being accentuated by the food-grooves leading from each angle of this aperture to the proximal end of a facet, each facet evidently serving originally for the support of a brachiole, since the food-groove indents the margin of each facet, although no brachioles ever have been found. At least 10 of these species had 5 brachioles, 2 more species probably having the same number. At least 16 species had 4 brachioles, 5 additional species probably having the same number of brachioles. Holocystites amplus group. — Of the remaining 7 species described by Miller, 5 may belong to the group typified by Holocystites amplus. In this group, 3 food-grooves, narrowing distally, lead from angles of the oral aperture to large facets for the support of brachioles. The nearest relative to Holocystites amplus appears to be Holocystites adipatus, which preserves one of the long, rapidly attenuating food-grooves; Holo- cystites tumidus resembles H. adipatus in shape, and H. ventricosiis may belong here also. Holocystites gyriniis agrees apparently in having three main food-grooves, but the latter do not attenuate strongly distally as in H. amplus, and both the left and left anterior food-grooves appear to bifurcate on reaching the facets, as though two arms were supported on a single protuberance in each case. The facet at the end of the right- hand ray is not preserved. Unfortunately, the upper end of the theca has been crushed from front to rear, producing the tantalizing conviction that some extremely interesting structure is here obscured. In the species with 3 main food-grooves, the length of these grooves is so much greater than in other species that there is a possibility of their belonging to a distinct genus. It is not certain, however, that Holocystites gyrinus belongs to the same group as Holocystites amplus, although the limitation of the food-groove system in both cases to 3 main strongly divergent rays suggests such a relationship. 50 The Ohio Journal of Science [Vol. XXI, No. 2, Number of Arms. The absence of arms on all known Osgood forms referred to Holocystites suggests that the latter were stiff and readily broken off, as in the case of the arms of the very different genus Comarocystites. The very distinct food-grooves are sufficient evidence of arms on the protuberances at their ends, especially in view of the fact that the food-grooves indent the margins of these protuberances. There is no evidence of generic differences between species having 5 arms and those having 4 arms. Among species with 4 rays there are both smooth forms with the quadrangular margins surrounding the central oral aperture distinctly delim- ited, and coarsely papillate forms with the corresponding quad- rangular margins rendered more indistinct by the presence of the papillae. Moreover, there are variations in the length of the theca and in the arrangement of the plates which can not be brought into co-ordination with any oral structure. ANAL APERTURE OF HOLOCYSTITES. One distinct advance in our knowledge of the anal aperture has been made. This consists in the fact that the outline of the anal aperture in all cases is polygonal, usually either pentagonal or hexagonal, and covered with a pyramid consisting of as many triangular plates as there are sides to the aperture. Many of the specimens show the facets for the attachment of the plates forming the pyramid, but the pyramid itself has been observed in only one case, namely, in an undescribed species of Holo- cystites, of the elongate type, as in //. alter?iatus, which is num- bered 109G5 in Walker Museum of Chicago University, is labeled as coming from Jefferson County, Indiana, probably is from the Osgood formation, and retains distinctly 2 of the 5 plates belonging to its anal pyramid. These plates are triangular, almost equilateral in form, and are still in position. This specimen had 4 arms. A study of the material accessible at Chicago University led to the conclusion that no classification could be based on differences in the number of plates forming the anal pyramid. On the contrary, this number appeared to vary from 5 to 6 in the same species in several cases. A study of the form of the theca and of the arrangement of the plates of the theca also failed to show any generic differences but led to other significant observations. Dec, 1920] Cystids and Blastoids 51 HOLOCYSTITES CYLINDRICUS GROUP. For instance, there is a group of Osgood species, character- ized by their elongate form and by the more or less definite arrangement of their plates, which closely resemble the group of elongate Racine species: Holocystites cylindriciis, H. alter- natus and //. ahnormis. Corresponding to H. cylindricus is H. canneus, with 5 arms, although a few plates are irregularly inserted. Corresponding to H. alternatus are H. hacculus and H. perlongits, each with 5 arms. H. plenus with 5 arms probably also belongs here. Of the less elongate, more ovate species hav- ing supplementary plates inserted very much as in H. alternatus there are //. splendens with 5 arms, and H. faberi, H. parvulus, and H. spangleri with 4 arms. H. ahnormis appears to be repre- sented in its oblique form by H. colletti. After studying these more elongate Osgood forms it seems impossible to avoid the conclusion that we have here a series of species congeneric with the genotype of Holocystites. Unfortunately, the Racine species are known at present chiefly from casts of the interior of the theca, presenting little evidence of the structure of the exterior. The granulose surface of many of these Racine specimens is formed by the fillings of the pores traversing the thecal plates. While these pores appear to be arranged in pairs, no entirely conclusive evidence has been found. No trace of the Omega-like connections between the pores has been noticed. The outlines of the oral and anal apertures are not preserved with sufficient distinctness in the specimens at hand. Finally, insurmountable obstacles result from the entire absence of any knowledge of the food-grooves and of the supports for the arms. This is due to the fact that collectors preserve only the casts of the interior of the theca but make no prolonged search for impressions of the exteriors. The latter usually occur only in large rocks, incon- venient to carry home, and in a fragmentary condition are not readily recognized as of value. HOLOCYSTITES WYKOFFI GROUP. In contrast to the elongate Osgood species of Holocystites is a group of globose forms, usually supported by a broad, flat base, the flattened area including several of the thecal plates. On one side of the specimen the plates have an elongate hexagonal form and are arranged in more or less transverse rows, while on the 52 The Ohio Journal of Science [Vol. XXI, No. 2, opposite side of the same specimen this arrangement in rows may not be so evident. This arrangement is found in H. wykojffi and H. sphceroidalis, with 5 arms, to which H. madisonensis probably is closely related, with H. suhrotundatiis less certain. This tendency toward the arrangement of the plates in trans- verse rows is shown also by numerous ovate forms with coarsely papillate surfaces, having 4 arms, and including H. afinis, H. benedicti, H. ornatissimns, H. papulosus, and H. siibovatus. Species of this type can not be differentiated generically from the smoother ovate forms, with 4 arms, in which there is no very obvious arrangement of the plates in transverse rows, as in H. commodus, H. glohosus, H. gorbyi, H. hammelli, H. indianen- sis, H. parvus, H. rotundatus, H. scitulus, and H. subglobosus. The studies here outlined have led to the conclusion that the importance attributed to the types of species has given rise to certain disadvantages as well as advantages. Although numer- ous specimens of Holocystites were collected formerly in the Osgood formation of southeastern Indiana, only the types were highly valued and have gone into the important collections. The other specimens became objects of trade and sale and have become practically lost to science. Hence, from the meager rep- resentatives of any one species it frequently is impossible to determine the range of variation in form, arrangement of plates, outline of oral or anal aperture, or the surface ornamentation. While the writer is convinced that the species described by Miller number only about one-fourth as many as indicated by the list of names, the material for more exact discrimination is lacking. THE TERMS HOLOCYSTITES, MEGACYSTIS, AND TREMATOCYSTIS. The generic name Trematocystis, proposed by Jaekel rests on the Osgood species Holocystites subglobosus. This is one of the smoother species with 4 arms, with a sharply angulate quad- rangular border around the oral aperture, and with the plates more or less arranged in transverse rows, but not conspicuously elongated in a vertical direction along the lower of these rows. As far as may be judged from our present knowledge of the various Osgood forms, there is a considerable probability that this term eventually must be discarded for either Holocystites or Megacystis. Dec, 1920] Cystids and Blastoids 53 STUDIES BY BATHER. In last year's numbers of the Geological magazine Bather has added greatly to our knowledge of the Osgood forms of Holo- cystites. In these numbers he has elaborated with his usual acumen our knowledge of this genus, and has accompanied the same with numerous drawings which illuminate every phase of its structure. 7. Holocystites alternatus Hall. (Plate IV, Figs. 1, 2, 3, 4, 5, 6.) Holocystites alternatus Hall, 20th Rep. New York State Cab. Nat. Hist., 1868, p. 312, PI. 12, Fig. 9; PI. 12a, Fig. 6. Holocystites alternatus Foerste, Ohio Jour. Sci., 17, 1917, p. 233, PI. II, Fig. 4. In' the type of Holocystites alternatus (Plate IV, Fig. G) there are two sets of plates: a primary set, conspicuously larger in size, arranged in transverse rows with S plates in each row, and a distinctly smaller set inserted in transverse rows between the transverse rows of primary plates. Three of the rows of primary plates are lettered A, B and C; a fourth row, unlettered, occurs almost at the very base of the specimen. Between rows A and B, and between B and C, secondary plates were inserted at all points where three sutures met, resulting in octagonal outlines for the primary plates and pentagonal outlines for the secondary intercalated plates, the unpaired angle of the latter being directed alternately upward and downward. Between C and the lowest row of primary plates there was inserted first a secondary series arranged in transverse order, and above and below the latter a tertiary series was added, resulting in a very elongate theca. A somewhat similar arrangement is noted in a specimen from the same locality, Racine, Wisconsin, numbered 839, and preserved in the Public Museum of Milwaukee (Plate IV, Fig. 1). The arrangement of the intercalated plates immediately below some of the plates of series C is very similar, but the interpretation of the lower plates requires the intercalation of three transverse rows of secondary plates, all at about the same time. Three specimens are figured from the Cedarville dolomite at Cedar- ville, Ohio (Plate IV, Figs. 2, 3, 4). All of these are noteworthy for the simplicity of the plate system between rows C and D, there being only a single transverse series of secondary plates, with a few scattered, incon- spicuous, tertiary ones. In a specimen from Wilmington, Ohio, the intercalation of secondary and tertiary plates has been carried to an extreme, even the primary plates of row B being distinctly separated. This Wilmington specimen is remarkable also for the excellent preservation of the lines of gro\\i:h on the thecal plates. The most con- spicuous evidence of growth along the margin of the larger plates is formed by a flattened border, from a millimeter to a millimeter and a half in width, above which the central part of each plate rises mod- 54 The Ohio Journal of Science [Vol. XXI, No. 2, erately, but rather abruptly. From these zones it is evident that when some of the larger plates were smaller, some of the still smaller plates might have been absent altogether. In still earlier stages of growth only those plates here called primary probably were present. If this view be correct, then such species as Holocystites cylindricus and Holocystites abnormis might be regarded as retaining their primitive characteristics even in old age, and young specimens of species which in older age resemble Holocystites alternatus might be expected to resemble Holo- cystites cylindricus in their earlier stages of growth. 8. Holocystites greenvillensis Foerste. Holocystites greenvillensis Foerste, Ohio Jour. Sci. 17, 1917, p. 203, PI. 9, Figs. 3 A, B, C; Plate 10, Fig. 8. This species, described from the Cedarville dolomite four and a half miles east of Greenville, Ohio, is an excellent example of a small form of Holocystites preserving its primitive characteristics, namely, about five transverse rows of plates, each row consisting of about 8 plates, the successive rows alternating with each other. 9. Allocystites hammelli Miller. (Plate I, Figs. 13 A, B.) Allocystites hammelli Miller, N. Amer. Geol. Pal., 1889, p. p. 222, Fig. 242. Basal part, for a height of 5 mm., rapidly expanding from a width of 2.5 mm. to 6 mm.; bottom broken off; in place of a column there may have been merely an attachment area as in Holocystites. This basal part is followed by a circlet of 7 plates, the line of separation between two of these being indistinct. The next circlet contains 8 plates of larger size; in addition to this there are 3 pairs of accessory plates, vertically arranged, separating adjacent plates from each other, and to this is added another accessory area too poorly defined to be deciphered. The following or third circlet also consists of 8 plates of larger size; in addition there are accessory plates at 7 of the 8 intervening sutures; at 5 of these sutures the accessory plates are known definitely to be arranged in pairs, one plate directly over the other; at the other two sutures the same structure may exist but it is not decipherable. The following or fourth circlet also consists of 8 plates of larger size; in addition there appear to be pairs of accessory plates at 3 of the intervening sutures, and single accessory plates at the lower ends of apparently 4 of the remaining sutures. The anal aperture rests on the upper margin of two of these plates of the fourth circlet and is enclosed, laterally at least, by plates of the fifth circlet. The latter circlet consists of at least 7 plates, the two in contact with the anal aperture being of exceptional width, the remaining 5 being tall rather than wide. If there is a sixth circlet of plates, its presence can not be detected in the type specimen. Surmounting the entire specimen, at least in its present condition, is a protuberance 2 mm. in height, constricted at the base, with a pentagonal outline at the top, with a width of 5.6 mm. between the right anterior and left anterior angles, and a diameter of 4 mm. from Dec, 1920] Cystids and Blastoids 55 the anterior angle- to the posterior margin. From the five angles ridges extend toward the center of the rather flat top of the protuberance. These 5 ridges are not sharply defined and their actual structure remains obscure. There is no evidence of the presence of food-grooves. As far as may be determined from the single specimen at hand, the oral aperture may have been covered by a pyramid of 5 plates, somewhat as in SphcBronis, globidus Angelin, or as in Glyptosphara leuchtenbergi Volborth, from the Ordovician of the Baltic areas of Russia. Between the oral and the anal apertures there is a transverse ridge, interpreted as locating the madreporite. No gonopore can be detected. The margin of the anal aperture protrudes slightly. Its outline is shghtly elliptical, rather than circular, the diameters being 3 and 4 mm. The inner margin appears to be rhomboid with the major axis in a lateral direction, suggesting an anal pyramid of 4 plates, but the evidence is not clear. Surface granulose, with the granules varying from slightly more to sHghtly less than a millimeter apart. Under a lens, very minute gran- ules, not visible to the unaided eye, appear in great numbers. Where the surface of the plates has been removed in cleaning the top or fifth circlet of plates, there appear to be slight color changes suggesting the presence of the Omega-like ornamentation of the worn surfaces of the plates of Holocystites. Here, again, the evidence is not clear. Locality and Horizon. — From the Osgood formation on Riker's Ridge, about 4.5 miles northeast of Madison, Indiana. Type numbered 6006 in Walker Museum of Chicago University. Remarks. — Were it not for the oral protuberance, this spec- imen would be regarded readily as a typical species of Holo- cystites. It differs in plate system from such a form as Holo- cystites greenvillensis chiefly in the presence of numerous acces- sory plates. It is regarded as belonging to the same family. However, the anomalous oral protuberance is suflticient to estab- lish it as a distinct genus. fc>^ 10. Gomphocystites indianensis Miller. (Plate I, Figs. 5 A, B.) Gomphocystites indianensis Miller, N. Amer. Geol. Pal., 1889, p. 249, Fig. 319. The upper, more globular part of the theca evidently has been compressed, but its outline, as seen from above, may have been distinctly elliptical even before compression. This is suggested by the relatively straight direction of the anterior and right anterior food-grooves along part of their paths, as seen in the figure accompanying the original description of the species. It is suggested also by the parallel direction of the straight parts of these food-grooves and by the angular curvature of the right anterior food-groove at mid-length. It will require addi- tional specimens to determine how much of the elliptical form of the 56 The Ohio Journal of Science [Vol. XXI, No. 2, theca is due to compression. The height of the globular part of the theca is about 20 mm. At the base it appears to have narrowed strongly to a short stipe, probably not over 15 mm. in length, although this part of the theca is not preserved. The oral aperture is at the base of a deep, laterally elongated, tri- angular pit, the anterior food-groove departing from the anterior angle of this pit, the right anterior and right posterior food-grooves departing from the angle on its right, while the left anterior and left posterior food- grooves separate from each other at a point about a millimeter from the left angle of the pit. The most characteristic feature of the species consists in the relatively great width and depth of the food-grooves and their almost V-shaped cross-section. Along the proximal half of the length of the food-grooves their width equals about 1.5 mm. Along the sides of the food-grooves there are depressions or grooves which may locate the sutures between the bordering plates. If they represent branches of the food-grooves, then it should be observed that they lead to no facets for brachioles. Parts of the surface are so well preserved that the absence of any evidence of the presence of brachioles should be emphasized. The food-grooves complete more than half of a circuit, the distal end reaching the lower side of the globular part of the theca. The distal end of theleft ant erior food groove curves upward toward the anterior food- groove and becomes parallel to the latter again for a short distance, as indicated in Fig. 5B on Plate I, and also in the figure accompanying Miller's description. The anal opening is circular and 3 mm. in diameter. No trace of the anal pyramid or of the small plates covering the food grooves remains. The surface is strongly and irregularly granulose, most of the granules being about one millimeter, or a little less, distant from each other. Where the surface is worn, there are diplopores, the individuals of each pair being less than half a millimeter from each other. Where the surface is not worn, there is no trace of these diplopores. Locality and Horizon. — From the Osgood formation in Jefferson County, Indiana. Specimen numbered 6019 in the Walker Museum of Chicago University. 11. Gomphocystites sp. (Plate III, 2 A, B.) Theca depressed globular, narrowed strongly at the base to a short stipe probably a little over 5 mm. in length. Along part of the globular portion of the theca the sides appear flattened, but this may not be a constant feature. There is no tendency toward straightening of any part of any of the food-grooves as in the case of Gomphocystites indianensis. The distal ends of these food-grooves reach the lower part of the plat- tened sides of the theca. The distal end of the left anterior food-groove probably curves upward toward the anterior food-groove, as in Gompho- cystites indianensis. Along part of the anterior food-groove, several Dec, 1920] Cystids and Blastoids 57 short grooves branch off from the main food-groove and terminate abruptly near the middle of thecal plates, but without any evidence of facets for the support of brachioles. Compared with Gomphocystites indianensis, the surface is papillose rather than granulose, the papillae being about 1.5 mm. or more distant from each other. The width of the food-groo^'es is about a millimeter or slightly less, and their cross-section is shallow U-shaped, rather than deeply V-shaped. Locality and Horizon. — From the base of the Louisville limestone, immediately over the Waldron shale, at the railroad cut two miles east of Anchorage, Kentucky. Remarks. — While this Louisville specimen is regarded as belonging to a new species, the state of preservation of the specimen figured makes the latter undesirable as a type. 12. Gomphocystites bownockeri Sp. nov. (Plate III, Figs. 1 A, B; Plate I, Figs. 6 A, B; 9 A, B; Plate II, Figs. 9 A, B.) Ohio specimens. — Thecas obliquely compressed, originally probably depressed globose, rapidly narrowed at the base to a short stipe, probably not over 10 or 15 mm. in length. Oral aperture elongated transversely, about 4 mm. wide and 3 mm. long from front to rear along its median line. Thecal plates varying in size ; those along the outer margin of the food-grooves tending to be smaller, more or less elongated at right angles to the food-grooves, the intermediate sutures having the same direction, the resulting appearance being that of oblong bodies packed together laterally. Usually only that row of plates which is in imme- diate contact wnth the food-groove presents the laterally compressed appearance in a striking manner, but a second row frequently offers a similar appearance along a part of its length and in a less obvious manner. In the spaces between those rows of plates which are adjacent to the food-grooves there are other plates varying considerably in size and some of these plates are much larger than any forming the rows. Larger plates occur also along the lower part of the theca, including the stipe. The anal apertni-e is 4 mm. in diameter, and 3 mm. distant from the nearest part of the oral aperture. The food-grooves extend at least below mid-height of the globular part of the theca, and probably reach its lower part. The food-grooves are faintly impressed on the type specimen, the latter being regarded as a cast of the interior of the theca. The plates show marginal depressed areas, with central raised parts, interpreted as evidences of marginal growth. The marginal depressed bands of the larger plates are about 1 mm. in width, and those of the smaller plates are narrower, suggesting the later intercalation of the smaller plates. Locality and Horizon.— From the Cedarville dolomite at Cedarville, Ohio. Type, numbered 8736A, in the Museum at Ohio State University. A second specimen, 8736B, 40 mm. in width, shows similar features. 58 The Ohio Journal of Science [Vol. XXI, No. 2, Racine specimens. — Gomphocystites bownockeri occurs also in the Racine dolomite of the Chicago area. Several specimens, numbered 22944, are preserved at Chicago Univesrity. The one marked 22944A shows the row of laterally compressed plates, along the lower side of one of the food-grooves, very well. The thickness of the thecal plates is almost 2 mm. The middle layers of most of the plates have weathered away, leaving the fillings of some of the diplopores exposed in the form of columns. The exterior layers of a sufficient number of these plates has been preserved to indicate that the food-grooves are relatively narrow, only a millimeter or less in width, but rather sharply impressed; on casts of the interior of the theca they are only faintly indicated. At one point, several short lateral branches of the food-grooves seem to be present. In a specimen numbered 4523 at the U. S. National Museum, short lateral grooves branch off diagonally from the lower side of the main food-grooves and terminate at small facets, evidently for the support of brachioles. The combined length of the branches with the terminal facets is 2 mm. Along the proximal part of the main food-grooves- 5 facets occur in a length of 10 mm. No clearly defined lateral grooves and facets were noticed along the upper side of these main food-grooves. Surface ornamentation. — Two specimens of Gomphocystites from the Chicago area, numbered 22943 at Chicago University, present impres- sions of parts of the exterior surface. One of these consists of the impression of the top of a specimen, and shows food-grooves, a relatively coarsely papillate surface, and numerous diplopores, the pores of each pair averaging about half a millimeter or less apart. The second spec- imen retains the impression of the attenuate basal stipe, 15 mm. long, and also the impression of part of the lateral wall of the globose part of the theca. The tips of the food-grooves evidently reach the lower slopes of the globose part of the theca. The striking feature of this specimen is its strongly papillate ornamentation, the papillae on the upper surface of the theca frequently being 1.5 to 2 mm. apart, while those on the stipe may be one millimeter or even less distant from each other. Food-groove system. — The food-grooves are deeply incised into the outer surface of the theca. Three primary branches leave the oral aperture. Of these, the anterior branch remains undivided; the right branch divides slightly over one millimeter from the center of the cral aperture; the left branch divides at a distance of about 2 mm., the resulting total being 5 branches. The width of the food-grooves is cne millimeter or less. Variation in the form of the theca. — Two of the specimens from the Chicago area, numbered 22944 at Chicago University, namely B and C, are nearly globular rather than depressed globular in form, but present the same rows of laterally compressed small plates along the lower side of the food-grooves as in Gomphocystites bownockeri. Specimen 22945 (Plate I, Figs. 9 A, B) from the same area, differs merely in its more elongate form. In all of these specimens the tips cf the feed-grooves, extend practically to the bas^ cf the more globular part of the theca. Dec, 1920] Cyst ids and Blastoids 59 One of the specimens found at Cedarville, Ohio (Plate II, Fij^^s. 9 A, B) also presents a globose form, but the food-grooves apparently extend only half way down the globose part of the theca. It consists of a rather poorly preserved cast of the interior of the theca. Remarks. — Hall described only two species of Gomplio- cystites from the Racine dolomite, Gomphocystites glans and G. claims, both from Racine, Wisconsin. The type of the genus is Gomphocystites tetiax, from Lockport, New York. All of these species described by Hall have long stipes, while all described in the present paper have short stipes. Narrawayella Gen. nov. Cyclocystoides. — The genus Cyclocystoides was based by Billings and Salter (Geol. Surv. Canada, Dec. 3, 1858, p. 86) on Cyclocystoides halli Billings. The conspicuous part of this species, as far as known, consists of a ring of relatively large plates, surrounded by a peripheral margin of small imbricating plates. The proximal half of the large plates is strongly elevated above the distal half; it is evenly convex, and covered with low granules. The distal half of each plate usually shows two spoon-shaped depressions. The outer edge of the proximal half of the large plates is under-cut, and this under-cutting extends backward into proximal half of the plate as a funnel- shaped pit, but it is not definitely known whether this pit continues in the form of a pore entirely through the plate. In American strata, 6 species having this type of structure are known. These are: Cyclocystoides anteceptus Hall, from the Black River of the Escanaba River; Cyclocystoides halli Billings, from the Curdsville member of the Trenton in Canada; Cyclo- cystoides salteri Hall, from the Trenton near Saratoga Springs, New York; Cyclostoides bellulus Miller and Dyer, from the Fairmount at Cincinnati, Ohio; Cyclocystoides magntis Miller and Dyer, from the Fairmount at Morrow, Ohio; and Cyclo- cystoides huronensis Billings, from the Richmond on Rabbit Island, in Lake Huron. None of the other seven described American species referred to this genus are known to have the structure found in Cyclo- cystoides halli. There is no evidence for regarding them as true species of Cyclocystoides. In fact, they differ not only from that genus but also among each other. Narrawayella. — One of these distinct groups is typified by Cyclocystoides cincinnatiensis Miller and Faber, from the Corry- '60 The Ohio Journal of Science [Vol. XXI, No. 2, ville member of the Maysville formation at Cincinnati, Ohio. In this species the plates are coarsely pitted. Cyclocystoides nitidus Faber, from the Corryville member, near Transit, Ohio, also is coarsely pitted and may be the same species, differing only in having 24 instead of 30 plates in the ring of large plates. Cyclocystoides munduliis Miller and Dyer, probably from the Corryville at Morrow, Ohio, with 32 plates, probably belongs to the same group. Raymond describes and figures a similar specimen of the same type (Bull. Victoria Memorial Museum, 1, 1913, p. 28, Fig. 3, PI. 3, Fig. 4) from the Prasopora zone of the Trenton in the Axe Factory quarry, at Hull, Quebec. This specimen is in the Narraway collection and is described as having the large plates covered with small pits between which are rounded inosculating ridges. For this group of pitted species, typified by Cyclocystoides cincinnatiensis, the term Narrawayella is proposed, in recognition of the great service to paleontology rendered by Mr. J. E. Narraway during his life as a collector of fossils in the rich area surrounding Ottawa, in Canada. For the species described and figured by Raymond, the name Narraway- ella raymofidi is offered. In this group of species the outline of the large plates is cuneate rather than quadrangular, and there is no evidence of spoon-shaped ornamentation on the distal halves of the plates. Cyclocystoides minus Miller and Dyer and Cyclocystoides parvus Miller and Dyer, both from Morrow, Ohio, the first with 19 plates, the second with 26 plates in the main ring, are not sufficiently understood but may belong to the same group as the preceding. I do not know on what authority these species are cited from the Richmond. Agelacrinus arm. — A remarkable specimen figured by Miller and Faber (Jour. Cincinnati Soc. Nat. Hist., 15, 1892, p. 85, PI. 1, Figs. 13-15), from the hilltop at Cincinnati, Ohio, as probably a fragment of Cyclocystoides magnus, consists of a fragment of either Agelacrinus cincinnatiensis or Agelacrinus pileus, exposing both the upper and lower surface of one of the rays, and some of the adjacent plates. The floor plates of this ray overlap each other distally, and along the margins of the floor-plates are seen the basal extensions of the lateral covering plates, as in ray 3 of Figure 5A, Plate 1, accompanying my paper on Agelacrinidce, in the Bull. Denison Univ., 17, 1914, pp. :399-487. See also volume 18, 1916, pp. 340, 341. Dec, 1920] . Cystids and Blastoids 61 Savagella Gen. nov. Recently Savage described a remarkable form from the Orchard Creek shale, near Thebes, Illinois, under the name Cyclocystoides ornatiis. The large plates forming the conspicuous ring are similar to those of typical Cyclocystoides in being quad- rangular in shape, but here the similarity ends. Along their upper surface the plates are radially grooved, their inner face is vertical, and their lower surface is convex, their radial cross- section being subtriangular. There is no structure comparable with the spoon-like ornamentation of typical Cyclocystoides, moreover the steep inner face of the ring-plates must have been correlated with an altogether different structure of the theca interior to the ring. Therefore the new generic term Savagella is proposed, with this species as the genotype. Cyclocystoides illinoisensis. — A fourth generic type is repre- sented by Cyclocystoides illinoisensis Miller and Gurley, from the same locality and horizon as Cyclocystoides ornatits. The ring plates of this species are flattened, without conspicuous orna- mentation, and their radial cross-section has a flattened elliptical form. For this fourth generic type no new name is provided at the present time although it is beheved that it will prove distinct when better understood. These four genera — Cyclocystoides, Narrawayella, Savagella, and the unnamed genus having Cyclocystoides illinoisensis as a type — are included in the family Cyclocystoididce, proposed by S. A. Miller (Jour. Cincinnati Soc. Nat. Hist., 5, 1882, p. 223), with Cyclocystoides as the typical genus around which the others are grouped. In proposing these new names it is fully realized that the structure of these peculiar organisms is not fully understood. It is believed, however, that the first step to their understanding is to note that the species hitherto grouped under the single term Cyclocystoides differ greatly in structi^re and probably represent several distinct but closely related genera. 13. Savagella ornatus Savage, (Plate I, Fig. 18.) Cyclocystoides ornatus Savage, Trans. Illinois Acad. vSci. 10, 1017, p. 265, PI. 2, Fig. 1. Type: Disk 18 by 20 mm. in diameter. Submarginal ring, consisting of 20 plates. The individual plates are about 1 mm. long in a radial 62 The Ohio Journal of Science [Vol. XXI, No, 2, direction, and 2.5 mm. wide in a direction parallel to the circumference of the theca. Their radial cross-section is subtriangular. Along their inner faces the}^ are abruptly vertical. Their upper faces curve down- ward with even convexity as far as mid-height on the outer margin of the plates; these upper faces are crossed radially by low ribs, usually five, sometimes four, on each plate. The lower faces also are moderately convex. The narrow intervals between the plates are occupied by a darker substance which, originally, may have been flexible, permitting the submarginal ring to be flexible. The submarginal ring of plates is bordered exteriorly by a marginal band of imbricating plates. Those in contact with the ring usually equal or slightly exceed 1 mm. in width, sometimes attaining a width of 1.5 mm.; they appear to be short but wide, with broadly convex free margins. The other plates are successively narrower, those at the margin of the theca usually being about half, or slightly more than half, of a millimeter in width; they are longer than wide. The basal part of the specimen, within the submarginal ring, appears to be formed by numerous plates from less than one millimeter to slightly over one and a half millimeters in diameter. These are irregularly convex, so as to produce moderate depressions at numerous points. Their margins are too poorly defined to determine whether there is any definite system in their arrangement. Locality and Horizon. — From the Orchard Creek shale, near Thebes, Illinois. Specimen figured and described by Miller and Gurley as one of their types of Cyclocystoides illinoiscnsis (Fig. 28 on Plate 5, Bull. Illinois State Mus. Nat. Hist., 1895; also Plate I, Fig. 18 of the present paper). This specimen was described as having 13 submarginal plates, forming not much, if any, more than one-half of a circle. From this the inference is drawn that the complete specimen had 24 to 30 plates. As a matter of fact fourteen plates and half of a fifteenth are present, there are suffi- ciently distinct impressions to indicate the former presence of three addi- tional plates, and their total number could not have exceeded 20, although apparently there is room only for 18 or 19. The radial cross- section of the plates is subtriangular, with the inner face abruptly vertical as in Cyclocystoides ornatus. The radially directed ribs on the upper faces of the latter are absent, but this upper surface is so badly worn in the Miller and Gurley specimen, here described, that the failure of these ribs to appear has no diagnostic value. There are four or five series of small plates in the marginal band surrounding the submarginal ring. In the opinion of the present writer, this specimen, the second one of those figured and described by Miller and Gurley, should be referred to Cyclocystoides ornatus Savage, and the name Cyclocystoides illinoisensis should be restricted to forms resembling Figure 27 accompanying their original description. Miller and Gurley describe the plates of the marginal band as elon- gated nodes. Dec, 1920] Cystids and Blastoids 63 Remarks. — The characteristic features of Cydocystoides ornatiis are the subtriangular radial cross section of the sub- marginal plates, with the inner face abruptly vertical, and the upper face crossed radially by 4 to 5 low ribs; the individual plates are much wider tangentially than long radially, and usually are distinctly separated from each other laterally, often from half to three-quarters of a millimeter. 14. Cydocystoides (?) illinoisensis Miller and Gurley. (Plate I, Figs. 17 A, B.) Cydocystoides illinoisensis Miller and Gurlev, Bull. Illinois State Mu.s. Nat. Hist.,' 6, 1895, p. 61, PI. 5, Fig. 27. The species Cydocystoides illinoisensis was founded on two fragments of the submarginal ring, found in the Orchard Creek shale, on Orchard Creek, near Thebes, in Alexander County, Illinois. These specimens differ in character. In the specimen described first, forming Figure 27 on Plate 5 accompanying the original description, (Plate I, Fig. 17A of present paper), 9 plates of the submarginal ring are present. These are nearly square in outline, neither the tangential nor the radial diameter differing far from 2 mm. The exposed surface is much flattened in a direction parallel to the disk, and the radial cross-section is very depressed elliptical, with a vertical diameter of 1 mm. The plates are in close contact laterally. Exterior to the submarginal ring are ntrmerous small naarginal plates arranged in short diagonal rows of 3 or 4 plates. Of these, those nearest the submarginal ring are nearly 1.5 mm. in width, while those nearest the free margin are much smaller. In their description of this specimen, Miller and Gurley state that the nine large plates present appear not to form more than a third of a circle. In my own opinion, hqwever, they form nearly half of the submarginal ring, the latter being somewhat elliptical in form. At least in a more complete specimen, (Plate I, Fig. 17 B) collected at the same locality, in the same shale, and preserved in the Museimi of Illinois State University, exactly 20 plates are indicated either by plates actually present or by the impression left by those that are missing. Locality and Horizon. — From the Orchard Creek shale, on Orchard Creek, near Thebes, in Alexander County, Illinois. The type, numbered 6051 A, is preserved in the Walker Museum, at Chicago University. The specimen here figured belongs to the Museum of the University of Illinois. Remarks. — The second specimen figured and described by Miller and Gurley under Cydocystoides illinoisensis is regarded as belonging to Cydocystoides ornatus, Savage, described from the same locality and horizon. 64 The Ohio Journal of Science [Vol. XXI, No. 2, 15. Troostocrinus sanctipaulensfs Sp. nov. (Plate I, Fig. 16.) Closely similar to Troostocrinus reinwardti (Troost), from the Beech River division of the Brownsport limestone of Western Tennessee. Most of the differences are slight. The sinus in the upper part of the radials tends to be more narrow; the median part of the lower half of the radials, beneath the sinus, tends to be more angular, the intermediate part, along the sutures, being more or less concave; and the lower end of the radial sinus is slightly lower, being in direct contact with the most extended part of the median fold immediately beneath. Cross- sections of the theca agree in being pentagonal along the lower half of the radials and triangular along the basals, the angles occupying the median parts of each of the three basals. In the specimen figured, the basal part of the theca appears less attenuate but another specimen from the same locality shows greater attenuation. The most conspicuous dift'erences are to be noted in the ambulacra. While the number of side-plates in the same length appears to be about the same, namely 16 in a length of 5 mm., the individual side plates appear to be more convex, the median line separating the side-plates is much more con- spicuously grooved, and this groove zig-zags less from side to side. Locality and Horizon. — From the top of the Laurel lime- stone at St. Paul, Indiana. Four specimens numbered 22909, preserved in Walker Museum, at Chicago University; only one is figured; another presents the details of the oral end of the theca. Remarks. — Whether the differences noted above are suffi- cient to warrant the erection of a new species is an open question. Additional specimens are necessary to determine how constant the differences noted are. Students of the crinoidea are aware of the frequency with which species occurring at St. Paul find their nearest relatives in the Waldron, Brownsport, and Racine, many of them showing Gotlandian affinities. From this point of view, the occurrence of the blastoid Troostocrinus, hitherto known only from higher strata, in the upper part of the Laurel formation at St. Paul is entirely normal. IG. Troostocrinus reinwardti-minimus Var. nov. (Plate I, Fig. 14.) Closely related to Troostocrinus reinwardti, from the middle or Troostocrinus zone of the Beech river division of the Brownsport forma- tion in western Tennessee. It differs in being much smaller, and more slender; compared with the total length of the radials, the triangular parts between the radial sinuses are relatively shorter; moreover, the Dec, 1920] Cystids and Blastoids 65 tips of these triangular parts tend to be less convergent. Along the sutures between the radials the theca is sufficiently concave to give a distinctly pentagonal cross-section to the upper half of the theca, while the basal part has a triangular cross-section, as is usual in this genus. Locality and Horizon. — From the Bainbridge phase of the Niagaran, six miles west of St. Marys, in St. Genevieve County, Missouri. Collected by Doctor Herrick E. Wilson, and numbered 14791 in the collections of Walker Museum at Chicago University. Remarks. — The following new species have been described from the Niagaran locality at St. Marys, all by Prof. R. R. Rowley: Cordylocrinus ? dubius, Cyathocrinus ovalis, Lecano- crinus hemisphericiis, Pisocrinus glahelliis, Pisocrinus granulosus, Scenidium ? nodocostatum, and Stribalocystites missouriensis. In addition to these. Prof. Rowley identified two species as Pisocrinus glohosus Ringueberg and Pisocrinus gorbyi Miller. The fauna is regarded as equivalent to some part of the Browns- port formation of western Tennessee. Troostocrinus ? dubius and Melocrinus wittenhergensis were described by Rowley from a Helderbergian locality near Witten- berg, Missouri. The Troostocrinus should be re-examined to verify its generic reference. Of the species listed, Scenidium nodocostatum belongs to the same group as the species described originally by Hall and Whitfield, from the Louisville limestone of Kentucky, as Orthis nisis. This is not a Scenidium. 17. Troostocrinus subcylindricus (Hall and Whitfield). (Plate III, Figs. 3 A, B, C.) Penlremites subcylindrica Hall and Whitfield, Geol. Surv. Ohio, Pal. 2, 1875, p. 129, PI. 6, Fig. 13'. As in all other Eublastoidea, 5 fork-shaped radials are supported by 3 basals. The bases of the right posterior and left anterior radials rest on the truncated tops of two of the basals, the top of the third basal, occupying the right anterior interradius, being acutely angular. Theo- retically, each of the two truncated basals was formed by the lateral coalescence of two basals. Only the posterior deltoid can be detected readily, the other four being restricted to the extreme tip of the acute interradial areas. The anal aperture opens through the oral extremity of the posterior deltoid. In Troostocrinus subcylindricus, the surface of the radials rises on approaching the radial sinuses, the rise increasing toward the lower end of the sinuses. Immediatelv beneath the lower end of the sinuses the 66 The Ohio Journal of Science [Vol. XXI, No. 2, median part of each radial curves conspicuous outward for a distance of about one millimeter. Near mid-height of the radials the cross-section of the theca is pentagonal but along the basals this cross-section changes to triangular, as in other species of Troostocriniis. In the type (Plate III, Fig. 3A), the distance between the base of the radials and the lower end of the ambulacra is 17 mm., and from the latter to the acute tip of the areas between the sinuses the distance is about 11 mm. The lateral diameter of the theca just beneath the projecting lower ends of the sinuses is 15 mm. The basals diverge rather strongly, as in figure 3B on Plate III. The divergence of the lower part of the radials is much less, giving rise to the specific name subcylindrica. In another specimen (Plate III, Fig. 3C), recently found at Cedarville, Ohio, the theca is much more elongate. Locality and Horizon. — From the Cedarville dolomite. The type, numbered 3306, accompanied by the basal part of another specimen, is in the Museum of Ohio State University; both specimens were obtained at Yellow Springs, Ohio. A third specimen, here figured, was obtained in the quarry at Cedarville, Ohio. Remarks. — Troostocriniis suhcylindricus is characterized by the strong outward curvature of the median part of the radials immediately beneath the lower end of the radial sinuses. This is shown conspicuously on lateral view. In Troostocriniis reinwardti, from the Brownsport formation of Western Tennessee, there is no corresponding outward cur- vature of the median part of the radials. On the contrary, on lateral view the median parts of the radials curve slightly inward rather than conspicuously outward just before reaching the lower end of the radial sinuses. The length of this convex curvature is only about one millimeter. Moreover, the lower third of the theca usually is much more slender, the lower half of the radials diverging more strongly. In the Brownsport species, the side-plates number about 16 in a length of 5 mm. The surface of the ambulacra, compared with that of other species, is distinctly flattened, the groove between the series of side-plates being not consicuous, nor straight, but zigzaging rather strongly between the alternating side-plates. IS. Troostocrinus sp. (Plate I, Fig. 15.) Only a single specimen of Troostocriniis is known from the Chicago area. This specimen is numbered 221)07 in Walker Museum at Chicago University, and is from Bridegport, Illinois. It evidently is from the Dec, 1920] Cystids and Blastoids 67 Racine horizon. Considering^ the close resemblance between the Racine faunas of Illinois and Wisconsin and the Cedarville fauna of Ohio, this Bridgeport specimen might be expected to show close affinity to Trooslocrinus subcylindricus (Hall and Whitfield), from the Cedarville dolomite. Compared with the latter it presents the following differences: The radial sinuses are more narrow, varying from 1 mm. to slightly over 1.5 mm. in width. The triangular areas between these sinuses are relatively shorter, have a somewhat wider apical angle, are only slightly concave, and are more strongly and more abruptly inflected toward one another. Near the lower end of the radial sinuses the cross-section of the theca is pentagonal, the lower half of the radials being angular along the median line, but their curvature along this line is not outward on approaching the lower end of the sinuses, as in Troostocrinns subcylindri- cus. All parts of the theca are relatively shorter than in the latter species, the result being a blunter top, and a more rapidly attenuating base to the theca. Compared with Trooslocrinus sanctipaulcnsis, the theca is similar to the figured specimen, in the shorter form and resultant outline, in the narrowness of the radial sinuses, and in the absence of any outward curvature of the median part of the radials just beneath the lower end of the radial sinuses. It differs in the triangular areas between these sinuses being shorter and less abruptly curved inward. Additional specimens will be needed to discriminate this form if it be distinct from those already described. Ifl. Crinocystites chrysalis Hall. (Plate I, Figs. 2 A, B; 3 A, B.) Crinocystites chrysalis Hall, 20th Rep. New York State Cab. Nat. Hist., 1868, p. 318, Pl.'l2a, Figs. 10, 11. The type of Crinocystiles chrysalis (Plate I, Figs. 2 A, B), from the Racine dolomite at Racine, Wisconsin, is preserved in the American Museum of Natural History, in New York City. As figured by Hall, this type consists of a clavate, moderately curved body; the sutures between the plates are represented, correctly, but there are no arm sup- ports, the specimen having been incorrectly interpreted in this respect. Both of the figures presented by Hall are inverted from their natural position. Arthur W. Slocom, the curator of the Walker Museum at Chicago University, called my attention to the fact that Crinocystites chrysalis was merely the cast of the interior of some other body whose exterior . aspect was very different. This was shown by specimen 22914 in Walker Museum, which came from the same horizon and locality as the type of Crinocystites chrysalis. Prof. Stuart Weller at once recognized the similarity of this specimen to Eucalyptocrinus proboscidialis Miller (Plate I, Fig. 4) from the Cedarville dolomite, at Pontiac, Ohio (Jour. Cincinnati Soc. Nat. Hist., 5, 1882, p. 224, PI. 9, Fig. 2). Formerly Pontiac was the seat of a lime industry. It was a railroad station, five miles south of Sidney, in Shelby County. Here the Racine phase of the Cedarville dolomite is exposed. 68 The Ohio Journal of Science [Vol. XXI, No. 2, A comparative study of the type of Crinocystites chrysalis, of the Chicago University specimen of that species, and of Eucalyptocrinus proboscidian s demonstrates that Crinocystites chrysalis is a cast of the interior of the anal tube of a species of Eucalyptocrinus, using that term in the broad sense in which it is applied at present. The following is a description of the Chicago University specimen. Chicago University specimen. — The specimen retains distinct impres- sions of the exterior surface of the upper part of the 10 wing-like processes that surround the tegmen, and that serve as partitions between the vertical compartments sheltering the arms. No trace of these arms remains. The upper edge of these processes projects horizontally out- ward, their total extension across the entire width of the specimen being IS mm., while 7 mm. farther down the attenuated upper part of the teg- men has a width of only 5 or 6 mm. The upper surface of the wing-like processes forms a platform, above which rises a further extension of the tegmen in the form of an anal tube. At its base this anal tube has a width of 7.5 mm., decreasing to about 3.5 mm. in a length of 14 mm., above which it is not preserved. The plates forming the anal tube are elevated towards their centers in a strongly nodose manner. They are arranged in about ten vertical series, the lower plate of each series being directly above the top of one of the wing-like processes already described. Four or five plates occur in each vertical series, as far as preserved, the plates in adjoining rows alternating with each other. Counted in a transverse direction, the plates at the base are arranged in circles of five plates each, the plates of successive circles alternating. It is not known whether the number of plates in a circle continues to be five as far as the top of the anal tube. The cast of the interior of the anal tube is continued downward into the cast of the interior of that part of the tegmen which is included between the top of the wing-like processes. The upper part of this cast of the interior, for a length of 14 mm., retains traces of the plates forming the anal tube; the lower part, G mm. in length, shows a vertically elongated flattened area beneath each of the vertical rows of the anal tube, somewhat as in Eucalyptocrinus proboscidialis. In the latter, however, that part of the tegmen which is included between the upper part of the wing-like processes is constricted strongly beneath while in Crinocystites chrysalis the corresponding part is constricted only suf- ficiently to give the entire cast included under that name by Hall an inverted clavate appearance, with the maximum expansion on a level with the top of the wing-like processes. The Chicago University sjjecimen agrees with the type of Crino- cystites chrysalis in having the anal tube moderately curved lengthwise. That part of the cast of the exterior of the anal tube of the Chicago University specimen which is best preserved belongs to the concavely curved side of this tube, but the amount of this curvature is slight. That part of the cast of the interior of the anal tube which is best pre- served belongs to convexly curved side of the tube. This curvature is more distinct, and this is the side of the tube here figured. Dec, 102(1] Cystids and Blastoids 69 Locality and Horizon. — From the Racine dolomite at Racine, Wisconsin. The Chicago University specimen is num- bered 22914, and consists of the two parts described in the preceding lines. The type of the species, numbered 2023, is preserved in the American Museum of Natural History, and consists of the cast of the interior of the anal tube and of the top of the constricted part of the tegmen. Remarks. — While it is very probable that the Chicago University specimen here described belongs to the same species as the type of Crinocystites chrysalis, this is not absolutely certain. The anal tube of the latter is more curved lengthwise, is wider at the base, is more strongly clavate, and has a different arrangement of plates toward the top of the anal tube. Eiicalyptocrinus prohoscidialis Miller (Plate I, Fig. 4 of present paper), is a closely related species. The form most closely resembling Eiicalyptocrinus prohoscidi- alis is Eucalyptocrinus egani Miller (Jour. Cincinnati Soc. Nat. Hist., 3, 1880, PI. 4, Figs. 1-lc), from the Racine dolornite at the Bridgeport locality, in Chicago, Illinois. From the figures accompanying the original description of this species it is evident that an anal tube extended above the platform formed by wing-like processes, but only the five lower plates of this tube are indicated on the cast of its interior, and it is not known definitely how much longer the anal tube was. The original specimen used for figure 1 c, accompanying the original descrip- tion of this species, is in the museum of the Cincinnati Society of Natural History. An anal tube rising above the platform of wing-like processes occurs also in a specimen from Racine, Wisconsin, which closely resembles Eucalyptocrinus nodulosus Weller, a cast of which is preserved in the Springer collection in the U. S. National Museum. All of these species with anal tubes projecting above the platform of wing-like processes differ from Eucalyptocrinus rosaceus (Goldfuss), the type of the genus, from the Devonian of the Eifel, the anal opening of the latter consisting of a small aperture between four plates at the center of the platform. Among European Calyptocrinids, the American species here discussed resemble most the form originally described by Phillips as Ily panthocrinites decorus. The latter also has an anal tube distinctly rising above the platform of wing-like processes. 70 The Ohio Journal of Science [Vol. XXI, No. 2, In most American species of Eucalyptocrinus, the anal opening is among a series of small plates forming the central part of the flat platform of wing-like processes. Only in Eiicalyptocrinus lindahli Wachsmuth and Springer do the wing- like processes extend strongly outward horizontally as in Eucalyptocrinus rosaceus, the genotype. At present there is no disposition on the part of specialists to subdivide the genus Eiicalyptocrinus , so that Crinocystites chrysalis may be regarded provisionally as the anal tube of some species of Eucalyptocrinus . Eventually, however, it may be found desirable to segregate those species in which the anal tube rises conspicuously above the platform of wing-like processes. In that case it may be necessary to determine whether the American species here discussed are as closely related to Hy panthocrinites decorus as the general appearance of the latter suggests. Eucalyptocrinus proboscidialis Miller. (Figure 4.) Eucalyptocrinus proboscidialis Miller, Jour. Cincinnati Soc. Nat. Hist., 5, 1882, p. 224, PI. 9, Fig. 2. Calyx obconical, risinf^ from a small, flattened base, consisting of the basals alone. Radials almost as tall as wide; first costals fully as tall as wide; second costals with equilateral pentagonal outlines. First dis- tichals about equal in size to the second costals, those of the same ray in contact with each other laterally, so that the interdistichal does not truncate the second costal but has an angular basal margin. Second distichals much smaller; the palmars which support the arms are not differentiated clearly; the interdistichal is narrower but slightly longer than the second distichals. The lower interbrachial is almost twice as tall as wide; it is surmounted by a pair of interbrachials the upper parts of which project distinctly above the general margin of the calyx. The width of the calyx at its upper margin is about four-fifths of its vertical height. The lower part of the tegmen, for a vertical height of 5 mm., is only slightly narrower than the top of the calyx; above this point it contracts, at first rapidly and then more gradually, reaching its narrowest dimen- sions about 24 mm. above the top of the calyx. Farther up it widens again, at first gradually and then more rapidly, to a level 37 mm. above the top of the calyx, somewhat as in a very much elongated hour-glass. To this height extend the compartments sheltering the arms. Above this extends an anal tube about 53 mm. in length. The lower part of the tegmen, for a vertical height of 11 mm., con- sists of a circlet of 10 oblong plates separated toward the base by a series of smaller plates, also 10 in number. Each of the 10 larger oblong Dec, 1920] Cystids and Blast oids 71 ■ «','' plates supports one of the vertical radiating win.^^-like partitions sep- arating the compartments sheltering the arms. Alternating with the tops of the larger oblong plates is another series of plates, narrowing rapidly upward and forming the lower third of the narrowly contracted part of the tegmen. At its narrowest part, the diameter of this con- tracted portion of the tegmen is scarcely 3 mm. ; the individual plates of this portion can not be differentiated in the type specimen. The upper third of the much elongated "hour-glass" por- tion of the tegmen consists of plates narrowing rapidly downward and producing a structure similar to an elongate funnel. The number of plates forming the circlet here can not be determined; it seems to be 8 but may be 10. The top of the funnel like portion of the teg- men widens into a narrow platform, appar- ently 18 mm. in width, forming the top of the compartments sheltering the arms. About halfway between the top of the calyx and the platform at the top of the funnel-like portion of the tegmen, the body of the crinoid enlarges to a diameter of 25 mm. The anal tube rising above the platform of the tegmen is about 15 m^m. wide at its base, and tapers gradually to a width of 4 m_m. or less. It is composed' of hexagonal plates which are almost equilateral at the base but become narrower and smaller farther upward. At a distance of 15 mm. from the base of the calyx the column consists of columnals 3 mm. in height, intercalated between which are columnals only 1 mm_. in height. Toward the base of the. calyx, both sets of columnals diminish in height. The general diameter of the column is about 5 mm., but short vertical wing-like processes extend outward from the larger columnals and apparently also from the smaller columnals. In case of the larger col- umnals these wing-like processes slightly exceed 1 m_m_. in length. Apparently there are 5 of them within the circimiference of each colvmmal. The plates of the anal tube are very thick, those near the lower part of the tube varying from 2 to 3 mm. in thickness. The surface of these plates is coarsely and irregularly nodose, especially centrally. From this it is assumed that the plates of the calyx also probably were f^:: Figure 4. Eucalyptocrinns probosci- dialis Miller. Type speci- men. An outline of a col- umnal is added in the lower more or less protuberant centrally, and the comer of the figure. 72 The Ohio Journal of Science [Vol. XXI, No. 2, general surface may have been more or less coarsely papillate, but no direct evidence of this is at hand. Locality and Horizon. — From the Cedarville dolomite at Pontiac, six miles northeast of Piqua, Ohio. The type is num- bered 13867 in the Museum of Ohio State University. Remarks. — In most species of Eucalyptocrinus the interdis- tichal truncates the top of the second costal. In Eucalyptocrinus proboscidialis this is prevented by the lateral contact of those first distichals which belong to the same ray. Only two other American species of Eucalyptocrinus possessing this character- istic are known. One of these is Eucalyptocrinus obconicus Hall, described from the Racine dolomite at Racine, Wisconsin. In the type of this species the vertical height of the calyx is 19 mm., and its diameter at the top is almost seven-tenths of its height. The base of the calyx is obtusely rounded, instead of concave, and the basals may be seen on lateral view of the calyx, though only of short length. The second of the American species mentioned above was described by Slocom, also under the name Eucalyptocrinus obconicus (Field Columbian Mus., 2, Geol. vSeries, 1908, p. 301, PI. 86, Figs. 1, 2), though he suspected that it might be distinct. The type of this second species was found in the Racine lime- stone of the spoil heaps along the Chicago Drainage Canal near Lemont, Illinois. It differs from typical Eucalyptocrinus obconicus in being twice as tall and wide; the base of the calyx tapers to an acute angle ; the basals are conspicuously taller and narrower; the radials, first costals, and interbrachials also are taller; the combined effect is to produce a more slender appear- ance along the lower half of the calyx. A cast of the exterior of this specimen shows that the exterior surface of the plates was convex but smooth. For this second species the term Eucaly- ptocrinus slocomi is proposed. Extended anal tubes, similar to that of Eucalyptocrinus proboscidialis, probably occurred also in Eucalyptocrinus egani, but the calyx of this species has an impressed base, and the interdistichal truncates the second costal. In species of Calli- crinus the concavity at the base usually is conspicuously deeper and wider than in typcial Eucalyptocrinus. Dec, 1920] Cystids and Blastoids 73 20. Periechocrinus cylindricus Foerste. (Plate III, Fig. 4.) Periechocrinus cvlindriciis Foerste, Ohio Jour. Sci. 17, 1917, p. 244, PI. 10, Figs. 1 A, B. In the museum of Wittenberg College, at Springfield, Ohio, there is a calyx of Periechocrinus cylindricus nearly 80 mm. in length. Above the distichals the individual plates are not outlined clearly, but it is evident that in the case of each ray that part of the calyx which is directly above the first pair of distichals is somewhat tumid for a height and width of 13 or 14 mm., thus giving the top of the truncated calyx a somewhat pentagonal outline. Since only casts of the interior of the calyx are at hand, it is impossible to determine how large was the column at its attachment with the base of the calyx, but, as far as may be determined from the form of the base of the cast of the interior, the diameter of this column must have been small, almost too small to support a calyx of such large size. EXPLANATION OF PLATES. PLATE I. Fig. 1. Ccelocyslis suhglohosus (Hall). A, anal side, with the two posterior rays, each branched. Small lateral branches of the main food-grooves alternate from side to side of the latter and lead to the facets supporting the brachioles. The margin of the anal orifice pro- trudes slightly. The attachment area for the column is 10 mm. in diameter. B, viewed from above, with the anal orifice at the top. Both views drawn from a wax cast of the hollow interior of a matrix preserving an impression of the exterior of a complete theca. Only the wax cast is known at present, but the cast shows plainly the presence of the cracks between the original rock fragments. No. 22906, in the Hindshaw collection, in the Walker Museum, at Chicago University. From the Racine dolomite at Chicago, Illinois. Fig. 2. Crinocystites chrysalis Hall. A, lateral view of cast of interior of an anal tube of some Calyptocrinid, regarded by Hall as the theca of some cystid. B, posterior view of the same. Both figures are reproductions of the figures accompanying the original description, in 20th Rep. New York State Cab. Nat. Hist., 1868, p. 318, PI. 12a, Figs. 10, 11, but are published in a position inverted as compared with the original to indicate their position in the Calytocrinid. From the Racine dolomite at Racine, Wis- consin. The type, numbered 2023, is preserved in the American Museum of Natural History. Fig. 3. Crinocystites chrysalis Hall. A, posterior view of cast of interior of an anal tube of some Calyptocrinid. When found this cast of the interior was still attached to the matrix of the specimen used for Figure B, which is an impression of the exterior of the same anal tube. Figure B was drawn from a clay cast of this natural impression. It shows the anterior or slightly concave side of part of the anal tube. The base of this figure shows the upper extensions of the wing-like expansions forming the compartments between which the arms of the Calyptocrinid are folded when at rest. Specimens numbered 22914, from the Hall collection in Walker Museum at Chicago Uni- versity. From the Racine dolomite at Racine, Wisconsin. 74 The Ohio Journal of Science [Vol. XXI, No. 2, Fig. 4. Eucalyptocrinus proboscidialis Miller. Cast of an almost entire specimen, chiefly of the interior, including the dorsal cup, the arching tegmen, constricted between the arms to a narrow tube expanding at the top of the partitions sheltering the arms, surmounted by a long anal tube. Republished from Jour. Cincinnati Soc. Nat. Hist., 5, 1882, p. 224, PI. 9, Fig. 2, for comparison with that specimen of Crinocystis chrysalis which is used on the present plate for Figure 3B. Original figure prepared from a plaster cast of the original specimen, prepared by D. A. McCord, of Oxford, Ohio. Found in the Cedarville dolomite at Pontiac, south of Sidney, Ohio. Fig. 5. Gomphocystites indianensis Miller. A, Viewed almost directly from above, but with part of the base showing. Anal aperture a short distance above and toward the left of the mouth. B, Lateral view, with the anal aperture directly beneath the mouth. The exact shape of the base of the specimen is unknown. From the Osgood formation, in Jefferson County, Indiana. Type numbered 6019, in Walker Museum at Chicago University. Fig. 6. Gomphocystites bownockeri Sp. nov. A, Upper surface of theca, with anal aperture a short distance on the left of the oral aperture. B, Lower part of another specimen showing the ends of two of the rays, drawn so as to indicate the probable form of the complete specimen. There is a possibility that originally these two specimens belonged together, but the broken parts no longer match. Specimens numbered 22943, collected by H. H. Hindshaw, and now in the Walker Museum at Chicago University. From the Racine dolomite at Chicago, Illinois. Fig. 7. Hallicystis imago (Hall). Lateral view showing the pectinirhomb of Plates 12 and 18, the anal aperture being on the right. This probably is the form for which Jaekel proposed the term Hallicystis elongatus. Specimen numbered 22908 in the Walker Museum of Chicago University. From the Racine dolomite at Chicago, Illinois. Fig. 8. Hallicystis imago (Hall). A small theca with 5 series of plates, the plates of the middle series not being in contact with each other. Numbered 21734, from the Van Home collection, in Walker Museum at Chicago University. From the Racine dolomite at Racine, Wisconsin. Fig. 9. Gomphocystites bownockeri Sp. nov. A, Viewed from above, with the anal aperture directly above the oral opening. B, Lateral view, with a glimpse of the anterior ray at the extreme upper left-hand margin of the figure, the tip of the right anterior ray showing at the lower right-hand margin. Numbered 2294.5, collected by H. H. Hindshaw, and preserved in Walker Museum at Chicago University. Foundin the Racine dolomite in the Bridgeport quarry at Chicago, Illinois. Fig. 10. Lysocystites (?) nodosus (Hall). Exterior of theca. Figure prepared from a clay cast of an impression of the exterior of a specimen in a rock fragment. This impression shows chiefly the characteristics of the exterior but some of the structural features near the surface also have left their traces, especially at the nodes and along the coarser radiating ribs. Specimen numbered 2193, from the James collection in Walker Museum at Chicago University. From the Cedarville dolomite at Wilmington, Ohio. Fig. 11. Lysocystites nodosus (Hall). A, Cast of interior of theca viewed from above, with anus. The tips of the inverted cuneate elevations of the cast are seen at the lower angles of the five plates belonging to the third series of thecal plates. A single plate intercalated in the third series is in contact with the lower margin of the anal aperture. No fourth series can he recognized. B, lateral view of the same, the cast sho^^'ing an oblong elevation at the upper end of one of the sutures between the basal plates; Dec, 1920] Cystids and Blastoids - 75 also the characteristic cuneate elevations at the top and bottom of the second series of plates. C, Basal view of the same, showing both the cuneate and the oblong elevations. At the center is a small triradiate structure, of which the median ray points toward the right anterior interradius. The stem appears to have been of remarkably small size. Specimen numbered 21815, from the Van Home collection in Walker Museum at Chicago University. D, Cast of interior of another specimen, with anal aperture at lower margin of figure, left of the median line. Specimen numbered 18943, from the Gurley collection at Chicago University. From the Racine dolomite at Racine, Wisconsin. Fig. 12. Wellerocystis kimmswickensis Gen. et Sp. nov. A, Viewed from above, with anus. B, Lateral view, with anus on upper right- hand side, showing that the arm plates occur in single series; along the arm curving around the anal aperture these arm-plates are seen to line only one side of the main food-groove. The arm on the left side of the anal aperture is curved but appears straight from the point of view seen in the figure. A third arm Hnes the upper left-hand margin of the figure. Specimen numbered 10727, collected by Prof. Stuart Weller, and preserved in Walker Museum of Chicago University. From the Kimmswick limestone near Glen Park, Missouri. Fig. 13. Allocystites hanimelli Miller. A, Viewed from above, with aperture at the top of the figure apparently pentagonal in form and elevated above the general surface of the theca. The transverse ridge near the middle of the figure is interpreted as locating the madreporite. The anal aperture is slightly below the middle of the figure and is more or less quadrangular in outline. B. Specimen numbered 6006, in Walker Museum of Chicago University. From the Osgood formation on Rikers Ridge, northeast of Madison, Indiana. Fig. 14. Troostocrinus reinwardti minimus Var. nov. Lateral view, with left anterior radial in front. Specimen numbered 14791, collected by Dr. H. E. W^ilson, and preserved in Walker Museum of Chicago University. Found 6 miles west of Sb. Marys, Missouri. Fig. 15. Troostocrinus sp. Lateral view, with right posterior radial in front. Specimen numbered 22907, from the Van Home collection,in Walker Museum at Chicago University. From the Racine dolomite at the Bridgeport quarry, at Chicago, Illinois. PiCi. 16. Troostocrinus sanctipaulensis Sp. nov. Lateral view, with right posterior radial in front. Specimen numbered 22909, from the Washburn collection, in Walker Museum at Chicago University. From the top ol the Laurel limestone at St. Paul, Indiana. Fig. 17. Cyclocystoides (?) illinoisensis Miller and Gurley. A, Part of the main ring consisting of large plates, with a trace of the sur- rounding peripheral margin, consisting of small imbricating plates. Type of the species, numbered 6051 A in the collections of Walker Museum at Chicago Uni- versity; the original of figure 27 on Plate 5 of Bull. 6, Illinois State Mus. Nat. Hist., 1895. Found in the Orchard Creek shale, south of Thebes, Illinois. B, A more complete specimen from the Savage collection at the University of Illinois, found at the same locality and horizon. Fig. 18. Savagella ornatus Savage. Specimen badly weathered, some of the lower plates of the ring of large plates considerably displaced, but their position indicated by depressions in the rock. Part of the peripheral margin of small imbricating plates preserved. Original of Fig. 28 on PI. 5, of Bull. 6, Illinois State Mus. Nat. Hist., 1895. From the Orchard Creek shale, south of Thebes, Illinois. 76 The Ohio Journal of Science [Vol. XXI, No. 2, PLATE II. Fig. 1. Ccelocystis snhglobosus (Hall). A, Type, viewed from above. Specimen distorted, with pectinirhomb on plates 14 and 15; apical end crowded toward lower right-hand corner. B, lateral view, showing pectinirhomb on plates 1-5, and 14-15. Specimen numbered 2027 and preserved in the American Museum of Natural History in New York City. From the Racine dolomite at Racine, Wisconsin. Fig. 2. Ccelocystis subglobosus (Hall). A, Lateral view showing pectinirhomb on Plates 12-18, and along the lower left-hand margin a faint indication of the pectinirhomb on plates 1-5. B, Lateral view, showing anal aperture and pectinirhomb on Plates 12-18. From the Welch collection deposited in Wilmington College, at Wilmington, Ohio. Found in the Cedarville dolomite in the Moodie quarry in that city. Fig. 3. Ccelocystis subglobosus -(Hall). A, Viewed from above, showing anal aperature and pectinirhombs on plates 14-15 and 12-18, also the divided madreporite plate 2.3, not numbered in the figure. B, Lateral view, showing the anal aperture and marginal views of the same pec- tinirhombs. C, Basal view, showing the pectinirhomb on plates 1-5; also the tendency toward a quadrangular outline of the impression produced by the invag- ination of the basal plates on the cast of the interior. The size of the attachment area for the column is indicated. From the Cedarville dolomite at Cedarville, Ohio. Fig. 4. Ccelocystis subglobosus (Hall). Oblique lateral view of distorted specimen, showing anal aperture and mar- ginal view of pectinirhomb 12-18. Specimen numbered 1603 in the Illinois State Museum at Springfield, Illinois. From the Racine dolomite at Racine, Wisconsin. Fig. 5. Ccelocystis subglobosus (Hall). A, Theca viewed from above, showing protruding anal aperture, the pectin- irhomb on Plates 14-15, a marginal glimpse of that on plates 12-18, and relatively numerous instances of divided or supplementary plates. B, Same, viewed from in front. D, Same, viewed from the side. C, Basal view of another specimen showing pectinirhomb on plates 1-5; also the quadrangular invaginated base of the cast of the interior. Specimens numbered 35155 and 35061 respectively in the U. S. National Museum at Washington, D. C, the originals of plate diagrams 37 and 36 of Schuchert in his paper on Siluric and Devonic Cystidea and Camarocrinus, Smithsonian Miscellaneous Collections, Vol. 47, Pt. 2, 1904, pp. 248 and 247. From the Racine dolomite at Chicago, Illinois. Fig. 6. Callocystites jewetti-elongata Foerste. Base of another specimen, showing pectinirhomb on plates 1-5; also large size of attachment area for column. From the Cedarville dolomite at Cedarville, Ohio. Fig. 7. Hallicystis imago (Hall). A, Lateral view with anal aperture along the margin on the right of the figure, also the pectinirhomb on plates 12-18, and a marginal view of that on plates 1-5 along the lower left-hand margin. B, Lateral view, showing pectinirhomb on plates 1-5. C, Basal view, showing the same pectinirhomb, also the small area for the attachment of the column. From the Cedarville dolomite at Cedarville, Ohio. Fig. 8. Ccelocystis subglobosus (Hall). Figure prepared from a clay cast of an impression showing part of the width of an entire theca and the attached column. The entire theca was about 4 mm. wider. Even with this increased width the column is relatively very large. Pec- tinirhomb on plates 1-5, and impressions of the recumbent rays, showing branch- ing. Surface pitted. From the Cedarville dolomite at Springfield, Ohio. Dec, 1920] Cystids and Blastoids 11 Fig. 9. Gomphocystites bownockeri Sp. nov. A, Lateral view, showing anal aperture, with the left posterior ray along the upper margin of the figure, the right posterior and right anterior rays occurring at successively lowtr parts of the theca. B, Viewed from above, with anal aperture slightly above and toward the left of the oral aperture. From the Cedarville dolomite at Cedarville, Ohio. PLATE in. Fig. 1. Gomphocystites bownockeri Sp. nov. A, Lateral view of an imperfect specimen showing two of the five food-grooves encircling the upper part of the theca in a dextral direction; the lower of these in the figure is the left anterior food-groove; and the upper one is the anterior food- groove. On the right half of the theca, near midheight of the more globose part, is an approximately horizontal series of small plates connecting on the right with the tip of the left "posterior food-groove. B, Oblique view of the same specimen showing the two food-grooves mentioned before, above which, in succession, are the right anterior food-groove, and the right posterior one, the latter curving toward the right, close to the left margin of the anal opening, as seen from the point of view of the figure. The pentagonal depression containing the oral aperture is best seen in Figure B. From the Cedarville dolomite at Cedarville, Ohio. No. 8736 in the Museum of Ohio State University. Fig. 2. Gomphocystites sp. A, Lateral view, showing in succession, from below upward, the left posterior, left anterior, and anterior food-grooves, with a faint view of the right anterior food-groove at the extreme top of the figure. B, View from above, showing four of the food-grooves, the right posterior one curving downward on the right side of the anal aperture, in the figure; then follow in succession the right anterior, anterior and left anterior food-grooves, with the proximal part of the left posterior one indicated on the left side of the right posterior food-groove. The proximal parts of the two posterior food-grooves are not distinctly preserved in the specimen figured and are added here to assist in orienting the specimen. From the base of the Louisville limestone, two miles east of Anchorage, Kentucky. Fig. 3. Troostocrinus subcylindricus (Hall and Whitfield). A, Type, with the right posterior radial on the left side of the figure. B, Lower half of a theca, with the left anterior radial in the center of the figure. C, Theca with the right posterior radial in the center of the figure. In B and C the truncated basals are angulated along their median lines. From the Cedarville dolomite. A, B, from Yellow Springs, Ohio, are numbered 3306 in the Museum of Ohio State University. C, from Cedarville, Ohio. Fig. 4. Periechocrinus cylindricus Foerste. Calyx, slightly swollen just beneath the right anterior (RA), right posterior (RP), and left posterior (LP) groups of arms. From the Cedarville dolomite, at Springfield, Ohio. In the Museum of Wittenberg College, at Springfield, Ohio. Fig. 5. Ccelocystis subglobosus (Hall). Plate diagram of specimen numbered 1603, in the Illinois State Museum of Natural History, in Springfield, Illinois; from the Racine dolomite at Racine, Wisconsin. Fig. 6. Ccelocystis subglobosus (Hall). Plate diagram of type, numbered 2027 in the American Museum of Natural History in New York City; from the Racine dolomite at Racine, Wisconsin. The dotted lines indicate the parts not distinctly defined in this type. 78 The Ohio Journal of Science [Vol. XXI, No. 2, PLATE IV. Fig. 1. Holocystites alternatus Hall. Specimen numbered 839 in the Milwaukee public museum, of Milwaukee, Wisconsin. From the Racine limestone at Racine, Wisconsin. Figs. 2, 3, 4. Holocystites alternatus Hall. Three specimens from the Cedarville dolomite at Cedarville, Ohio. Fig. 5. Holocystites alternatus Hall. From the Moodie quarry, in the southeastern part of Wilmington, Ohio; in the Cedarville dolomite. In the Welch collection. Fig. 6. Holocystites alternatus Hall. Type, numbered 2020, in the American Museum of Natural History in New York City. From the Racine dolomite at Racine, Wisconsin. Figure copied from 20th Rep. New York State Cab. Nat. Hist., 1868, PI. 12a, Fig. 6. Cystids and Blastoids Aug. F. Foerste Plate I. Cystids and Blastoids Aug. F. Foerste Plate II, 'Cystids and Blastoids Aug. F. Foerste Plate III. 5. CO ELOCYSTIS SUBGL0B0SU5, 1603- ^VJb/<3 S.PLATE DIAGRAM OF COELOCYSTIS SUBGL0B0SU5, TYPE, 2027. Cystids and Blastoids Aug. F. Foerste Plate IV. FORTY-TWO HITHERTO UNRECOGNIZED GENERA AND SUBGENERA OF ZYGOPTERA. Clarence Hamilton Kennedy, Ohio State University. During the past five years the writer has been engaged in a revision of the genera of the Zygoptera. The following new genera and subgenera have been in manuscript form for from two to five years. In nearly every case the characters of the penis have been the primary indicators that a new generic term might be advisable. In nearly every case other char- acters, usually venational, were found to parallel the penis characters. The writer has attempted to give the genus a value which as nearly as possible represents the same amount of differences in whatever part of the Zygopterous series it might fall. This ideal was not altogether attainable, because genera have been split so very fine in the Agrionin^ and some of the other very modern groups. Even in these groups, however, the genera are not as close as in some sections of the Libellulidae, where connecting links have not yet dropped out. No apologies are offered for the series of monotypic genera. These in nearly all cases are annectant forms, the last fragments of faunas preceding the present. Full descriptions of these new genera and subgenera with an extended discussion of their relationships as shown by the genitalia will eventually appear, the author hopes, as a bulletin of the U. S. National Museum. Vestinus genus nov. Type — Vestalis {Calopteryx) gracilis Ramb. This new genus includes Vestalis amoena Selys. It differs from Vestalis (type luctiiosa) in that the lobes of the penis are approximated and parallel, M3 and M4 arise at the same point on the arculus; there are never two complete rows of cells between Cui and Cu2 and the wings are hyaline. Anaciagrion genus nov. Type — Agrion {Calopteryx) cornelia Selys. This new genus includes the single beautiful species cornelia. It differs from Agrion in that Cusa is 5-6 cells long, as against a length of 2-4 cells in Agrion and is directed entad and caudad towards the anal field of the wing. 83 84 The Ohio Journal of Science [Vol. XXI, No. 2, Euchlorolestes genus nov. Type — Chlorolestes fasciata Selys. This genus includes also Chlorolestes tessellata Burm. and Chlorolestes longicanda Burm. Apical soft fold of penis erect and hoodlike; Ac lies slightly distad of the level of the first antenodal. Episynlestes genus nov. Type — Synlestes alhicanda Tilly. Quadrilateral broad, its inner end one-third of the hind side, first segment of Cu2 present. Penis with a long attenuate tip. Ceylonolestes genus nov. Type— Austrolestes analis Ramb. Includes also aridus, colensonis , cyaneus, divisus, gracilis, leda and tenuissimus. Naiad with lateral lobe as in the naiad of Lestes, at least so in analis. Penis with a spiral strap on the terminal lobe. Chalcolestes genus nov. Type — Lestes viridus Lind. Differs from Lestes in that the upper segment of the arculus equals the lower and that the penis lacks the internal fold. Africalestes genus nov. Type — Lestes virgatus Burm. Venation as in Chalcolestes, except that vein Mia is nearly straight throughout its length. Penis with a strap-like inner fold as in Ceylonolestes. Platystigma genus nov. Type — Mecistogaster jocaste Hagen. Penis with a broad, toothed terminal segment. Dense black part of stigma reduced to one cell in hind wing. Xanthostigma genus nov. Type — Mecistogaster ornatus Ramb. Includes ornatus and its varieties. Penis with a broad linear terminal segment without lateral teeth. Black part of stigma lacking in both wings. Haplostigma genus nov. Type — Mecistogaster modes tus Selys. Penis with terminal fold united to the apical segment. Dense part of stigma more than three cells long. Hind wing of male without costal dilation before apex. Dec, 1920] Hitherto Unrecognized Zygoptera 85- Goniostigma genus nov. Type — Mecistogaster amalia Burm. Costal dilation of male hind wing angulate. Penis with a large internal hood. Proplatycnemis genus nov. Type — Platycnemis hova Selys. Includes also agrioides Ris. Differs from Platycnemis in that M2 arises at postnodal 6-7 in front wing and at 4 in hind wing and that the stigma is longer than the cell below it. Leptargia subgenus nov. Type — Argia mollis Hagen. Inclndes fosteri, croceipennis, subapicalis, reclusa, tinctipennis , . chapadcE, sociale, smithiana, botacudo, tamoyo, tupi, hasmani,. sordida, thespis, tinctipennis. Internal soft fold of penis lacking, terminal segment flagel- late. South American. Micrargia subgenus nov. Type — -Argia thisma Calvert. Includes also lilacina. Terminal segment of penis saggitate.- Heliargia subgenus nov. Type — Argia vivida Hagen. Includes also plana, fiinebris, immunda, deami, probably also talamanca, underwoodi and terira. Internal fold of penis small or wanting, terminal segment irregularly triangular or even with a short attenuate tip. External fold present. Cyanargia subgenus nov. Type — Argia lacrymans Hagen. Includes also tonto. Penis with a flagellum attached to- inner surface of terminal segment. Chalcargia subgenus nov. Type— Argia oenea Hagen. Includes also orichalcea, harknessi, barreti, calida, percellulata,. insipida, ulmeca, adamsi, pipila, oculata, difficilis, rogersi, jocosa, tezpi, translata, sedula, gerhardi, frequentida, cuprea, pulla, nigrior, indicatrix, gaumeri, popoluca, cupraurea, johanella. Penis with apex bifid. Argyrocnemis genus nov. Type — Agriocnemis argentea Tilly ard. Penis with edges of terminal segment serrate^ male superior- appendages with hollow tips. 86 The Ohio Journal of Science [Vol. XXI, No. 2, Neoerythromma genus nov. Type — Enallagma cultellatum Selys. Penis characters as in Erythromma, but male appendages resembling those in E?iallagma signatum. Psenderythromma genus nov. Type — Erythromma viridiilum Charp. Like Erythromma except male appendages Psendagrion-Xiko. and wing with only 10-11 postnodals and 3 antenodal ultra- quadrilateral cells. Austrocoenagrion genus nov. Type — Coenagrion lyelli Tilly. Like Coenagrion except penis with shaft spines and the internal soft fold hood-like. Venation not studied. Hawaiiagrion genus nov. Type — Megalagrion {Coenagrion) xanthomelas Selys. Characters as in Coe?iagrion , but colors are reds and yellows, and the male appendages are Psendagrio7i-\ike. Includes deceptor, calUphya, nigrohamatiim, vagabundiim, molokaiense, microdemas and others. Kilauagrion genus nov. Type — Megalagrion {Coenagrion) nesiotes Perkins. Generic characters as in Hawaiiagrion, except that the male superiors are long and forcipate. Oahuagrion genus nov. Type — Megalagrion {Coenagrion) oahiiense Blackburn. Generic characters as in Hawaiiagrion, except that the stigma in hind wing of male is placed one and a half times as far from the wing apex as is the stigma of the front wing, slightly less so in the female. Apanisagrion genus nov. Type — Anisagrion lais Selys. Characters as in Anisagrion, except that the wing is not petioled to Ac by a distance equal to the length of Ac and the apex of segment 10 in the male is not forked. Protallagma genus nov. Type — Amphiagrion titcaccB Calvert. Characters as in Enallagma, except that the colors are largely red and the apex of segment 10 in the male is merely notched; i. e., without the two tubercles. Dec, 1920] Hitherto Unrecognized Zygoptera 87 Oxyallagma genus nov. Type — Oxyagrion dissidens Selys. Characters as in Eyiallagma, except red a dominant color, no postocular spots and penis without lateral basal lobes. Africallagma genus nov. Type^Enallagma glaiicum Burm. Generic characters as in Enallagma, except apex of segment 10 in male is elevated into an apical keel, notched at apex. Includes nigridorsum, ohliteratiim and schultzei as described by Ris, "Od. Sudafrika." Cyanallagma genus nov. Type — Acanthagrion inter r upturn Selys. Characters as in Acanthagrion, except the male superior appendages not decurved from the base and are usually forked. Includes laterale, acutum and perhaps cheliferum. Archaeallagma genus nov. Type — Enallagma ovigerum Calvert. Characters as in Enallagma, except that the hind edge of the prothorax with a rectangular lobe. Mesamphiagrion genus nov. Type — Enallagma occuUum Ris. Characters as in Enallagma, but body colors red, apex of segment 10 elevated and notched, body long haired and stigma one-half cell long. Differs from Amphiagrion in male append- ages being Enallagma-like, in postocular spots, in lacking the metasternal tubercles. Teleallagma genus nov. Type — Telagrion daecki Calvert. Characters as in Enallagma, but the pair of subdorsal apical points of segment 10 are widened laterally into minute lobes, abdomen very slender. Wings petioled to Ac. Ischnallagma genus nov. Type — Ischnura elongata Martin. Venation and stigmas as in Ischnura, apex of ten forked, male appendages and penis as in Enallagma. Proischnura genus nov. Type — Enallagma subfiircatum Selys. Characters as in Enallagma, except stigmas of hind wings smaller than those of front wings. Apex of segment 10 in male 88 The Ohio Journal of Science [Vol. XXI, No. 2^ forked. Penis intermediate between that of Ischnura and that of Enallagma. Homeoura genus nov. Type — Ischnura neops Selys. Characters as in Ischnura, but more Enallagmine. Proximal and distal sides of stigmas rounded, costa slightly indent d at stigma. Penis with large lateral paches of spines on the second segment. Anomalura genus nov. Type — Ischnura prognatha Hagen. Characters as in Ischnura, except the apical fork of segment 10 in the male is elongated into a spine, the paired spines of the penis are external as in Anomalagrion. Nanosura genus nov. Type — Ischnura aurora Braner. Characters as in Ischnura, except male with a pair of mesothoracic hook-like horns. Amphiallagma genus nov. Type — Enallagma parvum. Characters as in Amphiagrion, except post ocular spots present, colors blue and black; body not heavily haired. Seychellibasis genus nov. Type — Telebasis allaudi Martin. Characters as in Teinobasis, except anal plate an male not elongate. Apical lobe of penis linear. Palaeobasis genus nov. Type — Pyrrhosoma tenellum Vill. Characters as in Ceriagrion, except that wings are not petioled to Ac. Diceratobasis genus nov. Type — Agrion macrogaster Selys. Characters as in Metaleptohasis, but male without thoracic horns, while a large pair of horns occur on the seminal vesicle. Aceratobasis genus nov. Type — Metaleptohasis cornicauda Cf.lvert. Characters as in Metaleptohasis, but male without thoracic horns and his superiors longer than the inferiors. Xne Ohio Journal of Science Vol. XXI JANUARY, 1921 No. 3 SEASONAL CHANGES AND TRANSLOCATION OF CARBOHYDRATE MATERIALS IN FRUIT SPURS AND TWO-YEAR-OLD SEEDLINGS OF APPLE.*! SWARXA KUMER MiTRA. Laboratory of Plant Physiology, Ohio State University. At the suggestion of Prof. W. Paddock, of the Department of Horticulture of the Ohio State University, the writer under- took the work of determining the changes in carbohydrate content of fruit spurs and two-year-old seedlings of apple throughout the year. The work was done under the direction of Dr. H. C. Sampson, of the Department of Botny, and with the assistance of Dr. T. G. Phillips and Dr. J. F. Lyman, of the Department of Agricultural Chemistry. This is a part of the larger problem of determining what effect the carbohydrate content of fruit spurs has upon vege- tative growth and reproduction, and to what extent it should be considered in handling fruit trees. The methods used in the analysis of plant materials were as follows: The samples were collected from the campus orchard and nursery every two weeks and a quantitative analysis was made in every case to determine the percentage of starch and the sugars, sucrose, glucose, and maltose. For the determina- tion of starch the method described by Abderhalden^ was followed, and for sugars, that described by Darwin and Acton^ with a slight modification. In both cases the quantity of dextrose was determined by Fehling's solution, titrating against a standard potassium permanganate solution, as has been approved by the Association of Official Agricultural Chemists.^ In the case of two-year-old seedl ngs, separate analyses were made on the one-year-old stem segments, two-year-old stem segments, and roots. All percentages are expressed on a dry weight basis. *Paper read at the Thirtieth Annual Meeting of the Ohio Academy of Science, May 15, 1920, Columbus, Ohio. fPapers from the Department of Botany, Onio State University, No. 121. 89 90 The Ohio Journal of Science [Vol. XXI, No. 3, The earliest work on the study of translocation of starch is that of Mer.i This investigator made a thorough study of the changes of starch content throughout the growing and dormant periods of the year in different parts of the tissues, both in ordinary and girdled stems of various plants, such as beech, oak, pine, etc. Similar investigations were made by Rosenberg. 2 The most recent work is that of Butler, et al,^ who made a thorough analytical study of the carbohydrate, nitrogen, fats, phosphorus, and potassium of different parts of bearing / IS\ I iS ( 1^ I IS I IS I IS I IS I IS I ) S" I IS I IS I IS /*fAy- Jt«*ie, . lu/y. Aua, Se.^ Och t^ov- Dec Ja.n ■ Fe-C-. /^cty. A/^r-. Season oj- tkt- yecir Fig. 1. Curves showing changes in carbohydrates in fruit spurs of apple from May to April, 1919-1920. The ordinates represent the percentage of carbohydrates and the abscissae, the season of the year. app'e trees. The writer's work is similar to that of the latter investigators, so far as the carbohydrate materials are con- cerned. There is, however, some disagreement in the data obtained in the two cases. This is, perhaps, due to the differ- ence in age and in location of the trees in different climatic conditions. The data presented in this report show a seasonal periodicity in the changes of the carbohydrate materials, i. e., starch to sugar and vice versa, and a partial explanation of the factors that bring about these changes. The data also show that changes in temperature play a very important part in the changes of carbohydrates in plant tissues. The fluctuations Jan., 1921] Translocation of Carbohydrate Materials 91 of temperature during the growing and dormant periods are accompanied by changes in acidity in the plant, which in turn affect the activity of the enzymes which have always been found by actual experiment to be present at all seasons of the year. The data of the analyses are presented under three headings : First, curves showing changes in carbohydrates in fruit spurs; second, curves showing changes in carbohydrates in two-year-old seedlings; third, curves showing acidity in both the fruit spurs and seedlings. The curves in each case are followed by a brief summary and discussion. c Hi o L / li I ■ /S I IS ±_JS -LJ"^ J—J^ •' ''^ -^ J7 lit- I _ ij- .X-^JS 'Jr, /Vav fuht. Ju^lj ftu- Oc 4 S on /i" /»;a^• A^» p/ Me y z.air Fig. 3. A curve showing the average minimum temperature of the campus (Ohio State University) from May to April, 1919-1920. The ordinate represents the temperature in F° and the abscissa, the season of the year. 5. In the early spring, especially in March and April, during the swelling of buds and the early growth of leaves and flowers, the starch and sugars disappear very rapidly, apparently being utilised in the formation of new growth and the respiration of the growing cells. 6. The total sugar, especially reducing sugars, increases in December and is at its maximum during December, January and February. This increase of total sugar does not account quantitatively for the decrease of starch during the dormant period. Jan., 1921] Translocation of Carbohydrate Materials 93 7. The translocation of sugars in apple spurs is largely in the form of glucose and maltose. Glucose appears to be the most important sugar of translocation during the dormant period. 8. Maltose appears to be the most important sugar of translocation during starch accumulation in autumn. To sum up, the upward translocation of sugar in spring is largely in the form of glucose, while the translocation of sugar from leaves to stem in autumn is largely in the form of maltose. 9. The quantity of sucrose is very small in comparison to glucose and maltose. In fruit spurs it is found in the early growing and dormant period and is almost lacking at other times. 10. The process of downward translocation of sugar is very slow in comparison to the upward translocation, the latter of which is influenced by the transpiration stream during the early spring. 11. The correlation between the changes in acidity and the resultant effect upon the enzymes involved and the changes in carbohydrates in fruit spurs will be discussed later. The Two-year-old Apple Seedlings. 1. The effect of temperature upon the changes and trans- location of carbohydrates in apple seedlings is similar to its effect on carbohydrates in the fruit spurs of apple. 2. Total carbohydrate accumulation increases rapidly in both stems and roots at the close of the growing period in August, September and October, followed by a slight increase in roots and two-year-old stems, the maximum being reached in October, after which a gradual decrease follows. 3. After December there is a marked decrease in total carbohydrate both in roots and stems, which reaches its mini- mum in January, February and March. This decrease is perhaps due to slow metabolic processes, such as respiration, and also to a change of carbohydrates to non-carbohydrate materials, such as fats. There is also a marked development of latex- like material in the roots during January and February. Butler, et. al.* have found a slight increase of fats in roots and two-year- old branches during the dormant period. The latex material *Loc. Cit. 94 The Ohio Journal of Science [Vol. XXI, No. 3, was not analysed. It disappears by the middle of February, at which time the total carbohydrate material increases slightly. A marked decrease of carbohydrate occurs during the growing period of April, May and June. 4. The starch content of roots and stems increases rapidly with the decrease of growth in autumn, following a course kk (k HI t / \ iyo ; •^ 1 j \ 38 / \ ) 1 ) I 36 ( \ / \ 1 / \ 3*« \ f 1 I < > 1 } ■^ ^, / Sj / ) M ► 32 \ / f- K— < 1 \ \ / i \ r 3* \ ; / \ \ 1 \ f » 1% 1 \\ >-(. H 1—4 f / / \ 1 I 26 \ J / ( < 1 I \ t IH i / k / \ v >.^ "-^ / *"> r ^ ^, 1 " \ V t 1 ' \ \ / V / / >^ ^ r ^ \ k. ^ 20 < I ^\ V /, \ /, / ^ . / ^ sJ / \ ^ /8 \ J M \\ y Ki \ 6 f^ \ I 1 ( < / 1 ^> /^ .-\ ie~i .'■^ \ \ / \ pt /-, jl \ \ J. f'' / / / \ >».'-< r ' ( t i\ \ \ '•:-' »' 'i ■*-< ^,J. N 1 7 \'> 10 c <' < \ / r t-'-H { < g ^ ( 4 lordl LarifohrdYafZ ^.-. Starch --- If 2 0 fifay Tunt, July /^uy. "S?/.. O^t: /V,v. 2)£c. Jan. Pti>. /lor- A/>r. H f^"- y Zcl\r Fig. 4. Curves showing the changes in carbohydrates in two-year-old seed- lings of apple from June to April, 1919-1920. The ordinates represents the per- centage of carbohydrates and the abscissae, the season of the year. parallel to that of total carbohydrate in each case. The increase of starch in roots is much greater than that in stems. In one- and two-year-old stems the maximum is reached in October and November, followed by a gradual decrease, while in the case of roots the maximum accumulation of starch occurs in, December, after which there is a rapid fall. The minimum Jan., 1921] Translocation of Carbohydrate Materials 95 starch content during the dormant period in stems and roots occurs from January to March, after which there is a gradual increase until the inception of growth in spring. This decrease of starch during the dormant period is due o. partly to hydrolyzation to sugar and probably to slow respira- tion, and in the case of roots to the formation of latex-like material, as already mentioned. There is a possibility of some r (J o I a:: o 1. Miry Juii^ Ou,/i S e c; 3 0 n ^ tu yc ar Fig. 5. Curves showing the percentage of total sugar and reducing sugars in two-year-old seedlings of apple from June to April, 1919-1920. The ordinates represent the percentage of carbohydrates and the abscissae, the season of the year. fat or other synthesis. With the swelling of buds in the spring a general decrease of starch and total carbohydrate occurs in all parts of the plant. 6. The role played by the different sugars is similar to that described in the case of fruit spurs. Glucose and maltose are the principal sugars of translocation. Glucose reaches its maximum during the dormant period in all parts of the plant. It decreases in stems very rapidly during the early spring and increases in roots until April and then decreases owing to its 96 The Ohio Joiiriial of Science [Vol. XXI, No. 3, upward translocation and root growth. It is apparently the most important sugar of upward translocation in spring. 7. Maltose reaches its maximum in both stems and roots during summer and autumn at the time of translocation of J..JI, I '^ / /-f ■L^JJ. -Ljil Ar^ 7 /-r ,£.Jj£ ■/ /-T ^ /.f ^•LJ£ ./ /■T. /^. OU tYov- Ate- Tan Fih- May '^^^■ S«.o-son of r/ie. ye.a.r. Fig. 6. Curves showing the percentage of sucrose, glucose and maltose in two-year-old seedlings of apple from June- to April, 1919-1920. The ordinates represent the percentage of carbohydrates and the abscissae, the season of the year. sugars from the leaves. It is very low in stems during the dormant period, but increases slightly with the rise of sap in spring. Jan., 1921] Translocation of Carbohydrate Materials 97 8. Sucrose is very low in comparison to glucose and maltose. Its maximum occurs in all parts of the plants during summer and autumn. Its presence in roots is more pronounced than in the stems, which has also been shown b}^ Butler et al. It is not an important storage sugar in seedlings. 9. It is interesting to note that the amount of sugars in both stems and roots runs a close parallel throughout the year, with the exception of a greater content in roots during early spring. 10. The total sugar and the reducing sugars in both stems and roots show parallel curves. There is a marked increase at the beginning of the dormant season, which reaches its maximum from January to March, when the starch content is at a mini- mum. During the dormant period there is a marked hydrolysis of starch to sugar and a partial resynthesis of this sugar to non- carbohydrate compounds. 11. Total carbohydrate in roots is twice as much as that found in one and two-year-old stems, which are almost identical in their carbohydrate content. The Acidity of Fruit Spurs and Two-year -old Seedlings of Apple. 1. Acidity is high in summer and low in winter. In general during the growing period all parts of the seedlings are distinctly acid, while during the dormant period they approach close to neutrality. The roots and two-year-old stems become slightly alkaline during February and March, after which there is a rapid rise in acidity. 2. Acidity is highest in the leaves and lowest in the roots. The acidity of the leaves is on the decline at the time of abscission. 3. Acidity of the one-year-old stems and the fruit spurs used in the experiments is identical. 4. There is an approximate correlation between the optimum hydrogen ion concentration for the hydrolytic action of plant diastases (about 10~^.^ Sherman et al.)^ and maltase (10~^^ Hober-') and the time 'of the most active hydrolyzation of starch in the fruit spurs and stems, i. e., during the dormant period beginning with November. The correlation is not so clear in the case of hydrolysis in roots. Further data on this point are needed. 98 The Ohio Journal of Science [Vol. XXI, No. 3, 5. There is also a general correlation between the H ion concentration and the relative activity of diastase and maltase and the consequent concentration of glucose and maltose in the tissues. The optimum H ion concentration for diastase is somewhat more acid than for maltase. 6. Maltose is most abundant and glucose is least abundant in summer and autumn, when acidity is highest and nearer the optimum for diastase than for maltase. On the other 3.0 til hi 5.0 ....< !•• ?rri ^ \ 'Jl' Nft wf 1 / (n - \ •" M 1 !^- ::'--:!^^ 1 \ (y J /y f^a y Jum. July Wx.g Se/, Oct- fVo^^ Z>«.t- Tan Feb Nctr Af>r. Fig. 7. Curves showing the seasonal changes in Ph value of acidity in fruit spurs and two-year-old seedlings of apple from November to June, 1919-1920. The ordinates represent the acidity in Ph and the abscissae, the season of the year. Unfortunately, no data were obtained from July to October. hand, maltose is lowest and glucose is highest during the dor- mant period, when acidity is lowest and nearer the optimum for maltase than for diastase. The maltose is thus hydrolyzed to glucose at this time almost as rapidly as it is formed from starch. These facts suggest an explanation of why maltose is the most important sugar of translocation from leaves to stem in summer, while glucose is the most important sugar of translocation from root to stem at the close of the dormant period. 7. The presence of invertase has been found by experiment to be much more in abundance in the tissues of fruit spurs and seedlings than either diastase or maltase. It is for this reason that sucrose is rapidly hydrolysed in October and November, Jan., 1921] Translocation of Carbohydrate Materials 99" when the acidity is nearer to the optimum activity of invertase (about 10^'"", Hober). Sucrose is mostly at its highest during the growing period and early dormant period, when the acidity is rather below the optimum for sucrase. In fact, the data on acidity show in general a correlation of acidity to the activity of enzymes that are involved in the changes of the carbohydrate materials and their accumulation in plant tissues. The writer is indebted to Professors J. F. L^^man and T. G. Phillips for helpful suggestions in analytical work and especially to Prof. H. C. Sampson for numerous suggestions on problems and interpretation of results. LITERATURE CITED. 1. Mer, E. 189 1. Repartition hivernale de ramidon dans les plantes ligneuses. Comp. Rend., T. 1, pp. 964-966. 2. Rosenberg, O. 1896. Die Starke der Pflanzen im Winter. Bot. Centbl. Bd. 66, s. 964-966. 3. 1909. Bui. U. S. Dept. Agri. No. 107. 4. Darwin, F., and Acton, H. 1909. Practical physiology of plants, p. 258. 5. Abderhalden, E. 1910. Handbuch der Biochemischen Arbeits-methoden s. 270. 6. Hober, R. 1914. Physikalische Chemie, s. 271. 7. Butler, O. R., Smith, T. O., and Curry, B. E. 1917. Physiology of the Apple. Xew Hampshire Col. Agr. Expt. Sta., Tech. Bui. No. 13. 8. Sherman, H. C, Thomas, A. W., and Baldwin, M. E. 1919. Influence of hydro- gen ion concentration upon enzymic activity of three typical anvlases.. Jour. Am. Chem. Soc, Vol. 41. No. 2, pp. 231-235. 100 The Ohio Journal of Science [Vol. XXI, No. 3, ANALYSIS OF FRUIT SPURS OF APPLE (DRY BASIS). Date Starch Sucrose Maltose Glucose TorAL Sugar May 15 June 1 " 15 .July 1 " 15 Aug. 1 " 15 ;Sept. 1 " 15 Oct. 1 " 15 Nov. 1 " 15 Dec. 1 " 15 Jan. 1 " 15 Feb. 1 " 15 Mar. 1 " 15 April 1 " 15 7.470 10.580 11.439 11.459 12.459 13.500 13.815 16.214 18.599 17.100 17.399 16.200 15.899 11.459 13.500 13.500 . 8.879 6.300 8.879 11.459 10.899 8.259 6.899 . 190 .252 2.166 2.052 1.531 .502 .507 .126 ^127 .443 .696 .468 .950 .443 1.438 1.531 3.960 3.258 2.628 1.234 1.073 4.020 3.401 1.700 3.403 4.950 1.328 .928 2.628 2.939 2.541 2.011 .928 .772 3.387 2.996 2.560 1.214 1.152 .905 .378 .349 1.218 1.871 .508 1 . 258 .636 .558 .597 2.696 3.696 2.909 4.317 4.092 3.887 6.963 5.323 2.571 3.492 3.688 2.720 2.960 5.333 5.533 5.266 3.086 5.266 5.266 5.266 2.543 4.400 5.800 4.086 4.800 6.133 8.133 7.233 6.933 8.400 6.533 6.133 6.568 6.091 ANALYSIS OF TWO-YEAR-OLD SEEDLINGS OF APPLE (DRY BASIS). Date Starch Sucrose Maltose Glucose Total Sugar June 1-1* " 1-2 20.279 .772 1.111 .152 2.213 " 1-2 " 15-1.;::'; " 15-2 29.399 1.342 2.041 .389 3.920 17.759 .253 ":679' 1.238 2^213 " 15-3 23.999 .251 .752 .608 1.653 July 1-1 10.399 1-2 11.999 .393 2.628 .963 4.133 1-3 29.999 1.077 3.401 .558 5.266 " 15-1 12.699 .840 .958 .526 2.666 " 15-2 13.500 .822 2.889 .267 3.533 " 15-3 24.899 1.647 3.248 .658 5.800 Aug. 1-1 14.088 .750 1.236 1.491 3.533 " 1-2 14.699 .570 1 . 238 1.904 3.800 1-3 20.999 1.648 2.320 - 1.362 5.533 " 15-1 16.204 1.582 2.630 1.370 5.800 " 15-2 15.899 .190 2.628 3.042 6.133 ^' 15-3 22.500 .442 2.630 3.170 6.400 * 1 = One-year-old stem. 2 = Two-year-old stem. 3 = Root. Jan., 1921] Translocation of Carbohydrate Materials 101 ANALYSIS OF TWO-YEAR-OLD SEEDLINGS OF APPLE (DRY BASIS). (Continued.) Date Starch Su CROSE Maltose Glucose Total Sugar Sept. 1-1 13.500 750 3.248 .235 4.400 1-2 13.699 1 647 1.392 1.805 5.000 1-3 26.450 1 077 2.628 1.642 5.533 " 15-1 13.299 1 077 2.784 .943 5.000 « 15-2 17.100 253 2.011 1.178 4.400 " 15-3 29.800 823 3.248 1.258 5.433 Oct. 1-1 13.500 823 .958 .517 2.386 1-2 14.699 409 .958 .944 2.386 1-3 32.699 278 2.909 .455 3.800 " 15-1 20.250 228 3.804 .366 4.500 « 15-2 14.699 126 1.020 1.045 2.546 " 15-3 32.699 236 2.350 .516 3.266 Nov. 1-1 17.100 152 1.670 1 . 355 3.266 1-2 18.900 705 .308 1.630 2.386 1-3 27 . 299 1.331 1.182 2.680 " 15-1 17.100 2.320 1.438 4.533 " 15-2 17.100 1.392 2.672 4.133 " 15-3 35.399 266 2.626 1.384 3.666 Dec. 1-1 15.899 2.011 2.287 4.400 " 1-2 17.100 316 1.236 3.038 4.652 1-3 29.999 316 2.011 2.554 5.000 « 15-1 14.699 1.160 4.281 5.500 « 15-2 17.739 1.934 3.470 5.500- " 15-3 34.538 1.547 4.179 5.800 Jan. 1-1 11.469 570 1.236 4.771 6.866 " 1-2. . 11.999 396 1.545 4.892 6.916 1-3 24.899 712 1.547 5.989 8.333 " 15-1 8.879 127 .467 5.235 5.833 « 15-2 11.999 571 .772 4.409 5.833 « 15-3 24.899 905 .765 5.162 6.916 Feb. 1-1 11.469 348 1.468 3.790 6.133 1-2 13.500 507 1.392 3.988 6.400 1-3 23.599 1 076 2.011 2.551 6.400 " 15-1 12.638 127 .928 4.696 5.800 " 15-2 13.500 507 .772 4.709 6.133 " 15-3 26.100 126 3.712 2.100 6.133 Mar. 1-1 11.459 2.591 5.378 8.000 1-2 11.999 .772 5.943 6.733 1-3 24.899 2.532 4.202 6.916 " 15-1 13.500 1.530 5.320 6.916 " 15-2 11.459 .772 3.656 4.400 " 15-3 27.599 627 2.514 4.696 8,000 April 1-1 14.699 1.663 3.434 4.183 1-2 15.899 2.320 2.312 4.750 1-3 24.899 712 1.531 4.552 6.916 " 15-1 11.455 2.624 3.562 6.390 " 15-2 12.860 2.624 3.560 6.390 " 15-3 23.599 .876 2.610 4.806 7.668 102 The Ohio Journal of Science [Vol. XXI, No. 3, PERCENTAGE OF REDUCING SUGARS IN FRUIT SPURS AND TWO-YEAR-OLD SEEDLINGS OF APPLE (DRY BASIS). Date May 15 June 1 u 15 July 1 15 Aug. 1 a 15 Sppt 1 ii 15 . Ort 1 a 15 NoAr 1 ii 15 Dpr 1 a 15 Jan. 1.. ii 15 Feb, 1 ii 15 ATfir 1 a 15 April 1 ii 15 Spurs 2.100 360 100 360 386 546 666 266 680 666 666 400 4.266 4. 6. 5. 4. .533 133 .000 .800 7.533 .5.500 4.666 3.950 3.450 ]-Yr.-01d Stem .820 1.400 2.253 3.000 253 666 660 720 386 533 533 000 6.616 5.500 5.000 5.266 6.983 6.150 2.466 3.012 Seedlings 2-Yr.-01d Stem 820 653 253 686 666 800 666 3.233 1.533 1.973 1 . 813 3.533 3.533 5.000 5.833 4.750 5.266 5.266 6.400 183 750 183 Roots 1. 1. 9 635 866 666 666 800 800 233 233 253 813 106 253 800 000 7.250 5.666 4.400 4.400 5.833 6.250 5.500 6.000 PERCENTAGE OF TOTAL CARBOHYDRATE IN TERMS OF GLUCOSE. Date May 15. June 1 . " 15. July 1. " 15. Aug. 1 . " 15. Sept. 1. " 15. Oct. 1. " 15. Nov. 1. " 15. Dee. 1. " 15. Jan. 1 . " 15. Feb. 1. " 15. Mar. 1. " 15. April 1 . " 15. Spurs 11 17 16 18 17 17 20 23 25 21 23 23 21 17 21 23 17 13 18 19 18 15 13 .042 .900 .975 .265 .998 .386 .616 .281 .931 .543 .732 .800 .331 .532 .133 .133 .098 .933 .265 .230 .245 .854 .756 Seedlings l-Yr.-Old Stem 16.776 19.186 23.804 19.400 19.776 17.386 27.000 22.266 23.533 22.085 832 21 20.552 15.698 18.876 19.842 20.732 21.916 19.515 17.191 2-Yr.-01d Stem 24.743 21.945 17.465 18.533 20.132 23 . 798 20.201 23.400 18.718 18.876 23 . 386 23.133 23 . 652 25.210 20.248 19.165 21.400 21 . 1.33 20.085 17.132 22.415 20.486 Root 36.525 28.318 28.598 28.198 28.865 31.400 34.921 38.644 40. 132 39.598 33.012 42.999 38.332 44. 175 35.988 35.581 32.621 35.133 34.581 38.665 34.581 34,603 Jan., 1921] Translocation of Carbohydrate Materials 103 DETERIMINATION OF ACIDITY IN FRUIT SPURS AND TWO-YEAR-OLD. SEEDLINGS OF APPLE, EXPRESSED IN VALUES OF Ph. Spurs • Seedlings Date l-Yr.-Old Stem 2-Yr.-01d Stem Root Leaf Ph Ph Ph Ph Ph Nov. 1 3.6 3.2 4.4 6.4 3 0 " 15 5.8 5.6 5.4 6.4 4 2 Dec. 1 5.8 5.8 6.2 6.2 5 0 " 15 6.0 6.0 6.2 6.2 Jan. 1 6.2 5.6 6.2 7.0 " 15 6.2 6.2 6.2 7.0 Feb. 1 6.0 6.0 6.8 7.0 " 15 6.0 6.0 6.6 7.0 Mar. 1 6.2 6.2 7.2 7.4 « 15 7.0 7.0 7.2 7.4 April 1 5.8 5.8 6.2 6.2 " 15 6.0 6.2 6.6 6.2 May 1 5.6 5.8 6.0 . 5.8 5 8 " 15 5.6 5.6 5.6 5.8 5 4 June 1 5.4 5.2 5.2 5.4 5 0 " 15 5.2 5.2 5.0 5.4 4.8 N. B. — The data from July to October were not obtained. NOTES ON THE LIFE HISTORY AND EARLY STAGES OF CORYTHUCHA CELTIDIS O. & D. Harry B. Weiss, New Brunswick, N. J. This lace-bug was described by Osborn & Drake in 1916 in the Ohio State University Bulletin (Vol. XX, No. 35, Bulletin 8, Vol. II, No. 4, p. 227) from specimens taken at Columbus, Ohio, during September, 1903, on hackberry. Later these same authors in the Ohio Journal of Science (Vol. XIX, No. 7, May, 1919) recorded its occurrence in Kentucky, Tennessee and South Carolina. Gibson also treats this species in his paper on the genus Corythucha which appeared in 1918 (Tr. Am. Ent. Soc. XLIV, 69-104). Heretofore nothing has appeared on its life history and early stages and the following is presented as an addition to the knowledge of the biology of this species. Adults were first noted in New Jersey at Riverton on June 1, depositing eggs on the lower surfaces of hackberry leaves. Later, on June 21, adults and eggs were collected at New Brunswick, N. J., and from this it appears that the insect is well distributed in this state. About two weeks are necessary for the eggs to hatch and the various stages each require from 2 to 4 days of very warm weather in which to complete their growth. Cool weather retards development. There are 5 nymphal stages and from 16 to 20 days are necessary for a newly hatched nymph to become an adult. Two generations occur in the lat- itude of New Brunswick, N.J. Newly hatched nymphs feed in colonies on the lower leaf surface but as they become older they scatter and feed more or less independently. Many colonies of nymphs are wiped out completely by spiders and predaceous bugs and their nymphs. The eggs are inserted in the tissue of the lower leaf surface at right angles to the leaf, usually in the angle formed by the juncture of two veins. Only the basal end of the egg is inserted and each egg protrudes well beyond the leaf surface. They are laid close to one another and the clusters may contain anywhere from 4 to 18 eggs. The adults appear to wander over the tree considerably in the spring and as a result the egg clusters are scattered. Only rarely was it possible to find more than one or 104 Jan., 1921] Life History of Corythiicha Celtidis 105 two clusters on a single leaf. In many cases each cluster was found in the basal half of the leaf and the first nymphal feeding took place in this area. Egg. Length, 0.55 mm. Greatest width, 0.17 mm. Sul^elliptical, one side more convex than the other; tapering most at basal end which is acute with rounded end, end slightly constricted where it is inserted in the leaf tissue; sides of apical half tapering slightly; extremity of apical end truncate with rim-like collar and projecting cone-shaped cap. Cap white, remainder of egg brownish to brownish black, shining. First Stage Nymph. Length exclusive of spines, 0.61 mm. Greatest width exclusive of spines, 0.19 mm. Subelliptical. whitish or slightly brownish with a median dorsal, white area which includes part of the metathorax and the first two abdominal segments. Armature apparently similar to that of second stage nymph except that the only apparent tubercles are those on the head and the tubercular bases of the median abdominal spines. Remainder of armature appears to consist of simple spines with slightly tuberculate bases. Ventral surface, legs, antennae whitish, most of spines white. Abdomen margined. Second Stage Nymph. Length exclusive of spines, 0.78 mm. Great- est Avidth exclusive of spines, 0.31 mm. Oval elongate; similar in color and armature to the third stage nymph. Armatrue more pronounced than in first stage. Third vStage Nymph. Length exclusive of spines, 1.10 mm. Great- est width exclusive of spines, 0.40 mm. Oval; armature similar to that of fourth stage nymph except that the smaller spines on the margins of the thoracic lobes are absent and the lateral abdominal spines are single instead of double. Color and markings similar to those of fourth stage nymph. Color slightly darker than that of second stage nymph. Fourth Stage Nymph. Length exclusive of spines 1.18 mm. Greatest width exclusive of spines 0.68 mm. More oval than third stage nymph. Color and markings similar to those of fifth stage nymph. Annature similar to that of fifth stage except that in addition, the second and third abdominal segments bear single lateral spines. vSides of pro- thorax produced laterally. Mesothoracic lobes reaching to second abdominal segment. Fifth Stage Nymph. Length exclusive of spines, 1.75 mm. Greatest width exclusive of spines, 0.78 mm. Oval, dorsal surface dark brown, except for lateral half of prothoracic lobes and tubercles located on this surface, median dorsal portion of mesothorax, metathorax, and first abdominal segment and spines, posterior half of mesothoracic lobes and tubercles on this surface, the median portion of the eighth abdominal segment, the margins of abdominal segments seven and eight, the dorsal and lateral abdominal spines which are light. Antennae, sparsely hairy, four-jointed, third joint about twice as long as the first two combined, fourth joint clubbed. Antennas white except for clubs which are brownish. Eyes prominent consisting of numerous, reddish ommatidia. Head bears a pair of separated spines 106 The Ohio Journal of Science [Vol. XXI, No. 3, between the eyes, posterior to this pair is a median tubercle bearing four spines, posterior to this tubercle and close to anterior edge of prothorax is a large pair of prominent tubercles each bearing five spines. Prothorax . with a median dorsal elevation bearing a tubercle with two long spines and three shorter ones anterior to the long spines, behind this tubercle is a smaller median tubercle bearing a pair of diverging spines; posterior lateral margin of prothorax with a prominent tubercle bearing four spines; anterior to this tubercle on the margin are usually two large and three or four smaller spines (these spines vary in number and size) ; mesothorax with a pair of median separated tubercles each bearing three large spines; each mesothoracic lobe bears a prominent tubercle with four spines on lateral margin just posterior to middle; anterior to this tubercle are usually three large and two smaller marginal spines; centre of mesothoracic lobe slightly elevated; median separated spines on abdominal segments two, five, six, eight and nine; each posterior lateral margin of abdominal segments four, five, six, seven and eight bears a tubercle tipped with two long spines (sometimes a third short spine is present). All spines with tuber- culate bases. All spines tipped with a short probably secreting hair. Abdomen strongly margined. Mesothoracic lobes reaching almost to fifth abdominal segment. Ventral surface brownish black except for the portion of the thorax between the legs and the first and second abdom- inal segments which are white. The lateral edges of abdominal segments six and seven are usually somewhat light. Legs whitish except for dark tarsi and claws. Legs sparsely hairy, hairs short and light brown in color. Rostrum extending to third thoracic segment, whitish except for brownish tip. Scientific Results of the Katmai Expeditions of the National Geographic Society. XIV. HEMIPTERA OF THE FAMILY MIRID^. Harry H. Knight, University of Minnesota. Only thirteen species of Miridas have heretofore been recorded from Alaska. Heidemann (1900) records ten species which were collected by the Harriman expedition. Of these, seven species were European forms which were known to extend their range into North America. Three species, Mecomma gilvipes, Irhisia sertcans, and Lygiis approximatus were originally described from Sitka by Stal (1858). Orthocephalus saltator (Hahn) was recorded from Alaska by Uhler (1886) but appar- ently in error for the writer has seen specimens of an Irhisia sp. taken in Alaska which bear the above name in Uhler's hand- writing. No specimens of Orthocephalus saltator (Hahn), col- lected in North America, are to be found in the U. S. National Museum collection or the Uhler collection which is included there. The present list enumerates eight species of which one has not before been recorded from the Nearctic region, two of which are new records for Alaska, and one new species. With the present list a total of sixteen named species of Miridse are known to occur in Alaska. Mecomma gilvipes (Stal). 1858 Leptomerocoris giloipes Stal. Stett. Ent. Zeit., XIX, p. 187. 1883 Mecomma gilvipes Reuter, Hemip. Gymn. Eur., Ill, pp. 386, 555, PI. 2, Fig. 6. 1886 Mecomma ambidans Uhler, Check List Hemip., p. 20. 1900 Mecomma {Leptomerocoris) gilvipes Heidemann, Proc. Wash. Acad. Sci., II, p. 504. 1909 Mecomma gilvipes Oshanin, Verz. Palae. Hemip., I, p. 835. 1917 Mecomma gilvipes Van Duzee, Cat. Hemip., p. 398. 12c^ 2 9 Aug. 10-20, 1917, Katmai. This species was orig- inally described from Sitka by Stal (1858). Heidemann (1900) records "numerous specimens" from five different localities in Alaska. The male of this species is very similar to the same sex of the European Mecomma ambulans (Fallen) which Uhler 107 108 The Ohio Journal of Science [Vol. XXI, No. 3, (1886) recorded from British Columbia, but evidently in error. The female of gilvipes is easily distinguished by the short, yellowish translucent hemelytra (long-winged females are rare), while the male may be distinguished from amhulans with certainty only by the genital structures. Although Renter (1883) records Mecomma gilvipes as occurring in Siberia, Oshanin (1909) states that the species has as yet been found only in the Nearctic region. Tichorhinus katmai new species. Fusco-brownish to blackish, median line on disk of pronotum pale; narrowly at base of embolium and corium, a small spot at the cuneal fracture, pale or translucent. Closely related to marginatus Uhler, differs chiefly in the structure of the right genital clasper, particularly in the shape of the apical half (Fig.). T. kat ma I Tichorhinus katmai n. sp. a. left genital clasper, lateral aspect. b. right genital clasper, lateral aspect. cf. Length 5 mm., width 1.6 mm. Pubescence and general form similar to marginatus Uhler. Head: width .8,3 mm., vertex .44 mm.; black, narrow tip of tylus and lower margin of bucculje pale. Rostrum (length 1.45 mm.) barely attaining the hind margin of the intermediate coxas, brownish to black, darker at the apex. Antenna: Black; segment I, length .44 mm.; II, 1.58 mm.; Ill, .94 mm.; IV, .74 mm. Pronotum: length .71 mm., width at base 1.25 mm.; black, a pale vitta on the median line of the disk; scutellum black, transversely rugulose; sternum and pleura black, ostiole having a pale streak leading from the orifice. Hemelytra: Brownish black to black; narrow base of corium and along the base of cubitus, pale translucent, apex of coriimi slightly translucent through the brownish black coloration; embolium brownish, translucent, paler toward the base; cuneus brownish black, pale translucent on the margin of the fracture. Membrane uniformly dark fuscous brown, the veins scarcely paler. Legs: Fusco-brownish to black, in paler specimens the brownish may have a greenish tinge; tarsi black. Jan., 1921] Hemiptera of the Family Miridce 109 Venter: Brownish black to black; genital claspers distinctive of the species (Fig. 0). 9. Length 4.G mm., width 1.0 mm.; ovate, more robust than the male ; membrane scarcely extending beyond the tip of the venter ; more broadly pale on disk of pronotum and the base of the corium, also pale along the front margin of the eyes; antennas dark brownish; legs brownish, slightly tinged with greenish; venter slightly pale at the base of the ovipositor. Holotype: cf Aug. 10, 1917, Katmai, Alaska (Jas. S. Hine) ; Ohio State University Collection. Allotype: same data as the type. Paratypes : 3 cf 4 9 taken with the types. Lygus pratensis oblineatus (Say). 1832 Capsiis ohlineahis Say, Heterop. Hemip. N. Amer., p. 21. 1857 Capsus oblineatus Say, Fitch reprint, Trans. N. Y. State Agr. Soc, XVII, p. 784. 1859 Capsus oblineatus Say, Le Conte edition. Compl. Writ., I, p. 340. 1917 Lygus pratensis oblineatus ICnight, Bui. 391, N. Y. (Cornell) Agr. Exp. Sta., p. 562. Five specimens of var. ohlineatiis (Say) were taken Aug. 16-20, Katmai. Heidemann (1900) records pratensis from Alaska, specimens which were probably similar to the above named variety. Plesiocoris rugicollis (Fallen). 1829 Phytocoris rugicollis Fallen, Hemip. Suecia, p. 79. 1861 Plesiocoris rugicollis Fieber, Eur. Hemip., p. 272. 1896 Plesiocoris rugicollis Reuter, Hem. Gymn. Eur., V, p. 70. 1909 Plesiocoris rugicollis Oshanin, Verz. Palce. Hemip., I, p. 733. 30 d' 9 Aug. 2-15, 1917, Katmai; cf July, Savonoski, Naknek Lake. This species has not previously been known from North America. It is' recorded from Siberia and Russia by Oshanin (1909) and is well known in northern Europe and Scandinavia. The writer has compared the present material with European specimens of rugicollis (Fallen), determined by Reuter, and finds them identical. Reuter (1896) records the species as occurring on Salix and rare on Alnus. The present record completes the link in the holarctic distribution of the species. no The Ohio Journal of Science [Vol. XXI, No. 3, Irbisia sericans (Stal). 1858 Leptomerocoris sericans Stal, vStett. Ent. Zeit., XIX, p. 188. 1879 Irbisa sericans Reuter, Ofv. Finska Vet.-Soc. Forh., XXI, p. 58. 1896 Irbisia sericans Reuter, Hemip. Gymn. Eur., V, p. 12, PI. 1, Fig. 4. 1900 Irbisia {Leptomerocoris) sericans Heidemann, Proc. Wash. Acad. Sci., II, p. 504. 1909 Irbisia sericans Oshanin, Verz. Palas. Hemip., I, p. 760. 1915 Irbisia sericafis Essig, Inj. Ben. Ins. Calif., edn. 2, p. 213. 1917 Irbisia sericans Van Duzee, Cat. Hemip., p. 325. cf" 9 July 2-Aug. 16, 1917, Katmai, Alaska. Prof. Hine reports this species as common on rye grass and a few other plants, it being the only Mirid that was taken in considerable numbers. The species was originally described by Stal (1858) from Sitka, and is now known to occur along the western coast from California to the Bering peninsula, thence extending its range to some of the islands bordering the Palaearctic region (Oshanin 1909). Essig (1915) reports the species as injurious to rye and oats in California. Calocoris fulvomaculatus (De Geer). 1773 Cimex fulvomaculatus De Geer, Memoires, III, p. 294. 1861 Calocoris fulvomaculatus Fieber, Eur. Hemip., p. 253. 1875 Calocoris {C.) fulvomaculatus Reuter, Hemip. Gymn. Scand. Fenn., p. 49. 1886 (?) Calocoris fulvomaculatus Uhler, Check List Hemip., p. 18, "Br. Am." 1896 Calocoris fulvomaculatus Reuter, Hemip. Gymn. Eur., V, p. 184. 1907 (?) Calocoris fulvoiiiacidatus Snow, Trans. Kans. Acad. Sci., XX, Pt. 2, p. 159. 1909 Calocoris fulvo-maculatus Oshanin, Verz. Palae. Hemip., I, p. 691. 1917 Calocoris fulvomaculatus Van Duzee, Cat. Hemip., p. 329. 2 cf Aug. 10, 1917, Katmai. Uhler (1886) was the first writer to record this species from North America, merely indicating its occurrence and distribution as "Br. Am." So far as the writer can ascertain there is no specimen extant upon which this record was based. The next record for the species was based on material collected in the desert region of Arizona and published by Snow (1907). If the specimens upon which this record is based are the same as the true fulvomaculatus (De Geer) then the species has a remarkable distribution requiring adaptability, a condition not borne out by a known transitional distribution. The writer has compared the present specimens with material from Finland, determined by Reuter, and finds that the coloration and male genital structures agree in every respect. Calocoris fulvomaculatus is recorded from Siberia by Oshanin (1909) and with the present specimens from Alaska the species would appear to have a holarctic distribution. Jan., 1921] Hemiptera of the Family Miridce 111 Teratocoris saundersi Douglas & Scott. 1869 Teratocoris saundersi D. & S. Ent. Mo. Mag., V, p. 260. 1875 Teratocoris saundersi Reuter, Hemip. Gymn. Scand. Fenn., p. 27. 1892 Teratocoris saundersi Saunders, Het. Brit. Isds., p. 226, PI. 20, Fig. 8. 1895 Teratocoris lon^icornis Uhler, Hemip. Colo., p. 29. 1909 Teratocoris saundersi Reuter, Acta Soc. Sci. Fennica?, XXXVI, No. 2, p. 7. 1917 Teratocoris saundersi Van Duzee, Cat. Hemip., p. 308. 11 cT 9 July 31-Aug. 5, Savonoski, Naknek Lake. This species was described from Colorado by Uhler (1895) under the name Teratocoris longicornis where it was taken on Car ex at Steamboat Springs by C. F. Baker. Reuter (1909) was the first to discover that longicornis Uhler was in reality identical with Teratocoris saundersi Douglas & Scott (1869) described from England. The species was later recorded from Scandi- navia and Russia (Reuter 1875) and now with the present material coming from Alaska the species would appear to be holarctic in distribution. Usually the males and dark females are characterized by having black along the median line of the head, pronotum and scutellum. Certain male specimens have the scutellum entirely black. The females may be entirely green but in such spec- imens the second antennal segment, apices of femora, base and apices of tibiae are distinctly reddish. Teratocoris herbaticus Uhler bears a close resemblance to saundersi D. & S. and after a study of a co-type specimen ( 9 ) from Ungava Bay, Labrador, the writer wishes to remark on one or two distinguishing characters although Reuter (1909) has pointed out the chief differences between the species. In the female herbaticus, antennal segment I is shorter (length .43 mm., width .142 mm.) than in saundersi (length .57 mm., width .128 mm.). In a male specimen of herbaticus from Ft. Chimo, Labrador (L. M. Turner), the length of segment I (1 mm.) is shorter than in the male of saundersi (length 1.23 mm.), the thickness of the segment (.114 mm.) being the same in both specimens. In the male herbaticus there is in addition to the median discal stripe a prominent fuscous stripe on each side of the pronotum which extends back from the anterior angles half way to the basal margin of the disk, being suffi- ciently broad to cover the outer margin of the callus. The pubescence appears heavier and more distinct in herbaticus Uhler than in saundersi D. & S. 112 The Ohio Journal of Science [Vol. XXI, No. 3, Miris ferrugatus Fallen. 1807 Miris ferrugatus Fallen, Monog. Cimic. Suec, p. 107. 1900 Leptopterna ferrugata Heidemann, Proc. Wash. Acad. Sci., II, p. 504. 1909 Miris ferrugatus Oshanin, Verz. Palee. Hemip., I, p. 779. 1917 Miris ferrugatus Van Duzee, Cat. Hemip., p. 302. 13 cf 19 Aug. 10-20, 1917, Katmai. Heidemann records this species from Kadiak (July 20). It is known from Canada and is common in northern Europe and Siberia. Tke Okio Journal of Science Vol. XXI FEBRUARY, 1921 No. 4 ALGAL FOOD OF THE YOUNG GIZZARD SHAD.* Lewis Hanford Tiffany, Ohio State University. Introduction. During a surveyf of the distribution and feeding habits of the Ohio fishes for the State Bureau of Fisheries, opportunity was afforded for the study of the algal food of the gizzard shad, Dorosoma cepedianum Le Sueur, collected at various places over the state. This paper is the result of an examination of some two hundred fishes from Buckeye, Indian, Loramie, St. Mary's and Chippewa Lakes, and the New Reservoir at Akron. While the major part of the discussion is devoted to the algal food of the gizzard shad, attention is called to the animal forms and other material found in the digestive tract, and to the economic importance of algse as a part of the aquatic flora. The writer desires to express his gratitude to those who so materially aided in the preparation of this paper: to Mr. A. C. Baxter, Chief of the State Bureau of Fisheries, for kindly permitting the use of the gizzard shad, collected during the survey, for a botanical study; to Professor R. C. Osburn, for suggesting examination of the stomachic and intestinal content of the fish; and especially to Professor E. N. Transeau, who first introduced me to the study of algee in the summer of 1912, for the loan of algological literature and for helpful criticisms on species determination and identification. ♦Papers from the Department of Botany, Ohio State University, No. 123. fThis survey covered a period of three months (June 15 to Sept. 15, 1920) and was under the direction of Professor R. C. Osburn, Ohio State University, assisted by Professor C. L. Turner, Beloit College, Mr. E. L. WickUff, Mr. W. C. Kraatz, and the writer, Ohio State University. 113 114 The Ohio Journal of Science [Vol. XXI, No. 4', Methods of Collection and Study. The fishes were caught in . a large fine-meshed seine and placed immediately in a ten per cent solution of formalin, thus preserving the stomachic and intestinal content and preventing further digestive action. Collections were made at nearly all hours of the day. An examination of the content of the digestive tract was made with a compound microscope, the oil immersion being necessary in many cases for a final determina- tion of species. The technique of the work is very simple. A longitudinal slit, made along the ventral side of the fish, exposes the stomach and intestine, a small portion of which can then be pinched ofiE and a microscopic mount made in the usual way. Adult fishes are not included in this discussion, examina- tion being limited to specimens 1.5 to 7.5 centimeters in length — measured from the point of the snout to the base of the caudal fin. Food and Feeding Habits — General. Since the excellent work of Forbes, nearly forty years ago, on the freshwater fishes of the United States, very little study has been made upon the nature of the food of the gizzard shad. According to Forbes (3), the shad is a "mud lover par excellence'" ; it swallows "large quantities of fine mud containing about twenty per cent of minutely divided vegetable debris"; and it consumes, when young, food that is approximately ninety per cent microscopic animal organisms and the rest microscopic plants. From data at hand it appears that these statements require considerable modification when applied to young fish within the limits of this study. Mud may form as much as thirty per cent of the contents of the digestive tract, or it may be entirely lacking; in fact, in those fishes taken from St. Mary's Lake it was quite impossible to detect even the smallest quantity. A small portion of the contents is unrecognizable plant debris. It appears that the mud is merely incidental — so much unavoidable non-nutrient material that goes in with the real food. No consideration is given, therefore, to its varying amount in the digestive tract. The number of microscopic algal forms found in the stomach and intestine of the gizzard shad is markedly large. If one can conceive of all the different plankton forms of a given lake Feb., 1921] Algal Food of the Gizzard Shad 115 as being concentrated in one gizzard shad, he will get some idea of the vast number of individuals present at one time in the fish. In fact, the gizzard shad is about the most wonderful tow net that one could desire to get an estimate of the kinds and proportionate numbers of microscopic algae present in a body of water. In a single fish taken at Buckeye Lake on July 1, fifty species and varieties of algae were found; from all the specimens examined to date, one hundred and forty different forms are recorded. The presence of such masses of microscopic material in the digestive tract of the gizzard shad is accounted for in part when the feeding apparatus of the fish is examined. The very numerous fine gill rakers on the gill arches oppose the escape through the gill slits of very small objects. Thus, like a very fine sieve, these allow the water to pass out through the gill slits as the fish swims along with its mouth open, while the minute organisms are retained and pass into its alimentary canal. Animal vs. Pla^it Food. When a comparison of the number of animal and plant forms is made (See Table I below), it is noted that plants make up from seventy to one hundred per cent of the food material of the gizzard shad; animal forms, from zero to thirty per cent, depend- ing upon the particular fish and locality. It should be further noted that even the animal organisms, fed upon by the gizzard shad, depend directly for their food supply on the microscopic algae. The animal forms .include copepods, cladocerans, rotifers, and protozoa. TABLE I. Maximum and minimum percentages of plant and animal food in gizzard shad, based on examination of fish collected June 15 — September 15. Kind of food Buckeye L. Indian L. Loramie L. St. Marys L. Chippewa L. New Reservoir Animal 0- 30% 70-100% 3-20% 80-97% 2-15% 85-98% 0- 10% 90-100% 1-10% 90-99% 5-15% 85-95% Plant The majority of the algal forms belong in the order Proto- coccales. For convenience of comparison each individual identi- fied is placed in one of seven groups: Myxophycece, Peridinece, EuglenidcE, Bacillarice, Desmidiacece, Protococcales, and the filamentous algas. 116 The Ohio Journal of Science [Vol. XXI, No. 4, In the following table is given the relative importance of each group for the six localities named above. The first column under each group represents the total number of algal species and varieties found in the fishes, and the second column gives the total percentage content of the group for the particular locality. The percentages are approximations based on the total number of individuals present, not on the number of species and varieties, belonging in each group. TABLE II. Number of species in, and percentage importance of, algal groups. The first column gives number of species and varieties identified for the group; the second, the total percentages of food content represented by these species. Name of Lake Myxo- phyceae Euglenidae Peridineas Bacil arise Desrnid- diaccae Proto- coccales 1 Filamen- tous algae Buckeye. . 5 2% 4 6% 2 5% 13 10% 10 3% 62 70% 3 4% Indian 0 () 4 c;; 1 i% 4 6% 4 4% 25 80% 0 0 Loramie. . 1 1% 3 2% 1 2% 4 8% 3 2% 22 80% 3 5% St. Mary's 2 1% 1 1% 1 i% 2 1% 7 5% 31 90% 2 1% Chippewa New Reservoir 3 1 3% 10% 3 2 2';o 3% 1 2 1% 2% 7 3 10% 5% 5 3 5% 2% 19 31 77% 76% 1 1 2% 2% The filamentous algae were always present in small quan- tities, if found at all, and no single plant was observed longer than 300^1. It is easily seen how short, detached filaments could enter the fish with the water of respiration. It was not observed that the gizzard shad secured its food in any othpr way. No material was found wadded up, either in the stomach or the intestine, as is so often true among the game fishes. In one shad the remains of some epidermal and palisade cells of a leaf were found, but these may be considered as purely accidental, ingested in the usual manner. It will be noted from Table I that the percentages of plant food present in the digestive tract are very high and relatively constant. Throughout the period during which the gizzard shad is attaining a length of 7.5 centimeters, there is very little change of diet. Nor does the diet appear to change materially with greater age. An examination of two gizzard shad twenty centimeters in length showed practically the same proportion of microscopic alg£e found in the younger fishes, the only variation being a slightly larger amount of unrecognizable debris. Feb., 1921] Algal Food of the Gizzard Shad 117 Algal Forms. In Table III below is given a list of the 140 species and varieties of algae found in an identifiable condition in the digestive tract of the gizzard shad. Their abundance in each locality is noted by the letter x. x= rare, or occasional; XX = common ; and xxx = abundant. TABLE III. Algal Species and their Relative Abundance. Algal Species or Variety Buckeye Lake Indian Lake Loramie Lake St. Mary's Lake Chippewa Lake New Res- ervoir Merismopedia glauca (Ehr.) Naeg X XX X X X X " tpnuissima I^emm X X X X Coelosphaerium kutzingianum Naeg Microcystis marginata (Menegh.)Kuetz. Tptranpdia Reinschiana Arch xxx X X X Euglena spirogyra Ehr XX XX X X xxx X XX XX xxx X XX X X X X Phacus longicauda (Ehr.) Duj " _ pleuronectes (0. F. M.) Duj Peridinium aciculiferum Lemm X X X X X XX XX X Ceratiutrt hirundinella O F M X Melosira varians Ag " roseana Rab. X X X X X X X X X X " crenulata (Ehr. ) Kuetz " distans (Ehr.) Kuetz " granulata (Ehr.) Ralfs . . xxx xxx XX X X V " cvmbiformis (Kuetz.) Breb.. X Cyclotella striata (Kuetz.) Grun X X " comta (Ehr.) Kuetz Asterionella gracillima (Hantz.) Heib. . . X ) X X " salinarum Grun X X X X X X X X X X X X Synedra pulchella (Ralfs) Kuetz Caloneis trinodis (Lewis) Boyer .... XX X X X X " " Lag. var. sub- triangularis (Borge) W. & G. S. West forma triquetra G. S. West X X X X Staurastrum cyclacanthum W . & G. S . West X " leptocladeum Nordst. var. elegans G. S. West . . , i X X X " acerosuiii (Schr.) Eh.r.. X " parvulum Naeg X Cosmarium Regnellii W'olle . ... X XX X X X X X X X ** tenue Arch - - - - " margaritiferum ■ (Turp. ) Menegh . . ** radiosum Wolle X X X Pleurotaenium Trabecula (Ehr.) Naeg... Euastrum elecraus (Breb ) Kuetz X X X Gonium pectorale Muller X X Pandorina morum (Mull.) Bory Eudorina elesans Ehr X X X Pleodorina illinoiensis Kofoid ::::::::::i::::::;;:: 118 The Ohio Journal of Science [Vol. XXI, No. 4, TABLE III— Continued. Algal Species or Variety Buckeye Lake Eurodinella wallichii (Turn.) Lemm Volvox aureus Ehr Gloeocystis gigas (Kuetz.) Lag Chlorella pachyderma Printz Oocystis Borgei Snow " pusilla Hansg " elliptica W. West Micractinium radiatum (Chod.) Wille.. . Lagerheimia genevensis Chod. var. subglobosum (Lemm.) Chod Lagerheimia ciUata (Lag.) Chod. var. amphitricha (Lag.) Chod Chodatella citriformis Snow Franceia ovalis (France) Lemm Tetraedron muticum (A. Br.) Hansg " minimum (A. Br.) Hansg.... " regulare Kuetz " " torsum Turn " " longispinum Reins... " enorme (Ralfs) Hansg ........ " " aequisectum Reins. . hastatum (Rab.) Hansg. palatinum(Schm.) Lemm.. " trigonum (Xaeg.) Hansg. var. gracile Reinsch Ankistrodesmus falcatus (Corda) Ralfs. . " mirabilis G.S.W. Selenastrum gracile Reinsch . . J " acuminatum Lag Actinastrum Hantzschii Lag Kirchneriella lunaris (Kirch.) Moeb " obesa (\V. West) Schmid. . « " major (Ber.) G. M. Smith " subsolitaria G. S. West... . Scenedesmus obliquus (Turp.) Kuetz — " dimorphus (Turp.) Kuetz.. " acuminatus (Lag.) Chod.. . " bijuga (Turp.) Lag " " flexuosus (Lemm.) Coll " " irregularis (Wille) G. M. Smith " arcuatus Lemm " " platydisca G.M.S. " denticulatus Lag " carinatus (Lemm.) Chod. . " abundans (Kirch.) Chod... " " spicatus (W. & G.S.West) G.M.Smith... " longus Meven " " dispar (Breb.) G. M.S " quadricauda (Turp.) Breb " " longispina (Chod.) G. M.S " quadricauda Westii G.M.S. " " parvusG.M.S. " opoliense Richt " armatus (Chod.) G. M. S.. Crucigenia rectangularis (A. Br.) Gay... " quadrata Morren " irregularis Wille " Tetrapedia (Kir.) W. & G. S. W " emarginata (W. & G. S. W.) Schm Tetrastrum tetracanthum (G. S. W.) Brunn Coleastrum microsporum Naeg " sphaericum Naeg " cambricum Archer " reticulatum (Dang.) Senn. . . morus W. & G.^ S. West, var. capense Fritsch X X X XX X X XX X X X X X XX X Indian Lake Loramie Lake St. Mary's Lake X XX XX X X X XX X XX X X Chippewa Lake XXX X X New Res- ervoir XX X X XXX Feb., 1921] Algal Food of the Gizzard Shad 119 TABLE III— Continued. Algal Species or Variety Buckeye Lake Indian Lake Loramie Lake St. Mary's Lake Chippewa New Res- Lake ervoir Ppf1ia*?triim simr>le\ Ralfs XXX XX XXX X X X XXX X XX X XXX XX X X XX X " clathratum (Schr.) Lemm. . " " microsporum Lemm " " punctatum X X " clathratum duodennarium Bailey X " duolex Meven XX X X XX X XX " " reticulatum Lager... . " " subgranulatum Racib X " Boryanum (Turp.) Menegh.. " " perforatum Racib XXX X X X X X XX XX X XX X XX " " longicorne Reins. " " granulatum (Kg.) A Br , " " forcipatum Racib. " tptrns; CFhr 1 Ra1f Xyloterinus politus Say. (Fig. 1(7, work in beech.) Very little seems to have been noted relative to the work and burrows of this species. It has long been known that the insect, like forms in the genera Gnathotrichus, Pterocylon, Trypodendron, etc., rears its young in separate pits or cradles, the cradles projecting in opposite direction at right angles to the main passage way and with the fiber of the wood. The compound ambrosia tunnels of Gnathotrichus, Pterocylon, Trypodendron, etc., have only two rows of larval cradles, one projecting above and the other extending below^ the main egg- gallery. The larval cradles of Xyloterinus politus (Fig. 1, b) are double-compound or quadrifarious, i. e., arranged in double rows or two on each side of the main passage way. Compound ambrosia beetle tunnels should then be divided into two classes : viz. (1) egg-galleries with only two rows of larval cradles, one extending above and the other below the main passage way and (2) egg-galleries with quadrifarious or tetrad-rows of larval cradles, two projecting above and two below the main passage way. The tetrad-rows of larval cradles of X. politus Say were first observed by the writer while collecting Ipidse on the roll-way described above. The insect seems to prefer beech for breeding purposes, but it is also common in maple and frequently in birch. Numerous other food plants have been recorded by Hopkins (Bull. 33, W. Va. Agr. Exp. Stat., 1893, p. 210) and Swaine (1. c, p. 83). The latter (1. c, p. 10) describes the peculiar and characteristic projection of a cylindrical rod-like mass of frass from the entrance hole while the insect is actively engaged in excavation of its tunnels. Schwarz (Proc. Ent. Soc. Wash., Vol. II, 1891, pp. 77-81) publishes notes on the breeding habits of some scolytids, including this insect, but does not describe the larval cradles. The writer examined over thirty different galleries, in most cases on both sides, and found the larval cradles of politus to be arranged in double rows on each side of main passage way. All specimens of work of this insect exam- ined or at hand are from the neighborhood of Cranberry Lake in the Adirondacks. As the larval burrows have not yet been described or figured by other w^orkers, who have published on the insect, it is impossible to state whether the double-^ compound larval cradles are a specific character peculiar to this insect or whether it is a race or variety living in the vicinity^ of Cranberry Lake, New York. NOTES ON ELACHISTA. II. (MICROLEPIDOPTERA). Annette F. Br.\un, Cincinnati, Ohio This paper is a continuation of "Notes on Elachista with Descriptions of New Species," which appeared in the March, 1920, number of this journal. Further rearing from mines during the past season in the vicinity of Cincinnati has resulted in the discovery of the life histories of several species pre- viously known only in the adult stage, and in the addition of two new species. Elachista cucuUata n. sp. Palpi white, second segment dark brown outwardly, third some- times with fuscous shading outwardly. Antennse black. Face and head white, except the collar, which is black. Thorax and extreme base of fore wing black; a silvery fascia almost at base, broadest on the dorsum; remainder of wing very dark brown; a silvery fascia just before middle, curved or slightly angulated on the middle of the wing, extends beyond the fold, but ends abruptly before reaching the dorsal margin; a silvery triangular spot at tornus and a little beyond it, a longer, usually oblique, costal silvery streak. Cilia dark brown. Hind wings and cilia dark grayish brown. Legs dark brown, basal segments in the male, and the tips of segments and a band around the hind tibiae in both sexes silvery. Abdomen blackish, silvery beneath. Expanse: 8-9 mm. Type (cf) and twenty-seven paratypes, reared from larvas mining leaves of Carex Jamesii, Cincinnati, Ohio; imagoes May 13 to June 5. The mine made during the antumn on the overwintering leaves is a narrow linear tract running down alongside the midrib, not at all or but very little enlarged until spring. Toward the end of March, the larva begins to feed actively again, and the mine becomes transparent and occupies most of the breadth of the leaf, which is inflated, due to the elevation of the midrib into a ridge on the upper side. Mines were collected March 31 and April 10; on the latter date most of the larvae were full grown. The larva is red; head brownish red, thorax with mid- dorsal line, abdomen with mid-dorsal and lateral lines pinkish. The coloration of the larva is retained in the pupa, with median 206 April, 1921] Notes on Elachista—II 207 and lateral ridges pinkish. In general, the pupa belongs to the elongate tapering type, but the median ridge is convex from anal end to head when viewed from the side, extending out on to the head, where it divides, a projecting ridge extending on each side to the antennae, thus forming a pointed hood which projects over the face. Lateral ridges also prominent, with prominent lateral thoracic tubercles. The pupa is attached by a median band of silk, and also enclosed in a few strands of silk. The abrupt ending of the median fascia before it reaches the dorsum, easily distinguishes this species from all other described species. In the fore wing, veins 7 and 8 are long stalked and vein 6 arises from the extreme base of 7; other points of the venation as in the figure in Meyrick's Handbook. Elachista enitescens n. sp. Palpi and entire head dark leaden metallic, almost black; antennas grayish black throughout. Thorax and base of fore wing leaden metallic, with a reddish and purplish luster which is most decided at base of dorsum, where the leaden color is sometimes replaced by metallic golden or silvery scales like those of the fascia and spots. Fore wing dark brown, faintly shining; a silvery or golden metallic fascia with reddish and purplish luster before the middle of the wing, is oblique in its costal half, broader and nearly perpendicular in its dorsal half with a slight projection along the fold; at two-thirds a silvery or golden metallic costal and an opposite dorsal spot; beyond them in the middle of the wing near the tip a silvery or golden spot. Cilia dark gray. Hind wings broad, dark brown. Legs dark gray, hind tarsi paler tipped. Abdomen dark gray, underside yellowish. Expanse: 7-7.5 mm. Type (cf) and four paratypes ( cf and 9), reared from larvae mining leaves of the bulrush, Scirpus atrovirens, near Cincinnati; imagoes May 13 to June 8. The larva makes a long transparent mine in a basal leaf, extending from the base of the leaf upwards. In March and the early part of April, they are mining in the old leaves, indicating that feeding began in the preceding autumn. Later the larva enters a new leaf at its base where it is not visible unless the old outer leaves are torn away. . Each mine may be four or five inches in length. The larvae feed at night only at the upper end of the mine, retreating in day time down to the base of the leaf, (sometimes beneath the surface of the water). The larva is yellow, with an ill-defined irregular darker patch 208 The Ohio Journal of Science [Vol. XXI, No. 6, toward the posterior end of the first thoracic segment on each side of the middle. Pupation takes place toward the end of April or in May. The pupa lies on the upper surface of a leaf over the midrib and is covered by a flat cocoon, formed of two series of oblique parallel threads of silk, crossing one another at an acute angle. The pupa is more nearly allied to the stout ovate type, but the dorsal abdominal surface is flattened, without median or lateral ridges, but with a dorso-lateral series of erect short blunt spines, one spine on each abdominal segment except the first; three or four prominent lateral mesothoracic tubercles. This species is closely allied to E. madarella Clemens, agreeing with it in venation and in shape of the hind wings, and differing from it only by the entirely black antennae, the darker head, less golden base of fore wing, and darker legs. It is apparently rather rare and local, as I found the larva in but one locality, although the food plant is very common. Elachista madarella Clemens. Specimens of this species were reared from mines on several species of Carex, very commonly on Carex pubescens and Carex cristata, and on Scirpus atrovirens. The mine is very similar in character to that just described for E. enitescens, and is indis- tinguishable from it on Scirpus. The larva makes several mines, the earlier ones in the outer overwintering leaves, the later ones in the new inner basal leaves. The larva feeds in the upper end of the mine during the night, retiring down into the base of the leaf almost to the rootstock during the day. Mining larvce were collected from April 3 to May 16. Larva whitish or pale green, with the first thoracic segment marked with a pair of prominent dark brown or blackish L-shaped marks. Pupa covered with a flat cocoon formed of two series of parallel silken threads, as in E. enitescens; very similar to that of enitescens, but somewhat broader, with rougher thorax, and across vertex, a transverse beaded ridge, with a broad sinus in its middle. The reared imagoes emerged from May 24 to June 28; a few captured specimens were taken as late in the season as July 10. Often the silvery gray at the base of the wing is almost entirely replaced by the pale golden color of the fascia. April, 1921] Notes on Elachista—II 209 Elachista leucofrons Braun. By some confusion of data which I can not now explain, the mine and larva described as belonging to this species (Ohio Jn. Sci., XX, 170, 1920) belong to Elachista orestella, in which the larva is either grayish or green and marked as described. The mine of Elachista orestella is grayish, with epidermis wrinkled in the middle of the length of the mine, which here only is green. The mine of E. leucofrons is whitish, with epidermis nowhere wrinkled; the mine lies just beneath the upper epidermis, extending usually across the leaf; the underside of the leaf remains green. The mine occurs on both Hystrix and Elymus, most commonly on the latter grass, while that of E. orestella occurs most commonly on Hystrix. The larva of E. leucofrons is pale grayish or greenish, with narrow mid-dorsal and broad lateral lines whitish; first segment of thorax marked posteriorly by a transverse brownish mark, curving forwards at each end. Elachista irrorata Braun. The larvae of this species commonly mine leaves of Glyceria nervata, a tall grass occurring in moist meadows and wet places. The larva mines toward the tip of the leaf, the narrow indistinct pale yellowish green mine usually beginning low down on the leaf sheath, where the larva lies concealed during the day. The larva sometimes makes a short detached -mine near the tip of the leaf; such mines are always untenanted in day time. Even when the larva is full grown, the mine is scarcely wider than the body of the larva. Mines were collected from the middle of April to the early part of May; the imagoes emerged from May 19 to June 10. Larva yellow when young, glaucous above when full grown. The pupa is always attached near the base of the leaf on the upper side with head pointing toward the stem. The pupee may easily be collected on the food plant at the proper season. The pupa, which shows a general resemblance to that of E. leuco- frons, has a broader mesothorax with more tubercles, the median ridge of the abdomen more depressed, the lateral ridges projecting farther. 210 The Ohio Journal of Science [Vol. XXI, No. 6, The occurrence of the mine of the type specimen on another species of grass {Agrostis perennans) was apparently accidental, as no other mines have been found on this grass. Most of the specimens reared on Glyceria are considerably larger than the type, expanding 8.4 to 11 mm. NOTES ON THE POWDERY MILDEWS OF OHIO. Bruce Fink, Miami University. My interest in the powdery mildews dates from 1884, when I was a student of the first class-room teacher of mycology, the late Dr. T. J. Burrill, of the University of Illinois. My first publication on fungi dealt with this group, and my interest has never abated. My collecting in Ohio began shortly after entering the State in the summer of 1906, and two of my students, Mr. E. E. Duncan and Miss Esther Young, have spent a considerable amount of time in studying the group in my laboratory. Miss Freda M. Bachman and Mr. W. G. Stover have also aided considerably with the collecting and determinations. Mr. A. D. Selby pubHshed ''The Ohio Erysipheae" in a Bulletin of the Ohio Agricultural Experimental Station for 1893, and Mr. W. C. O'Kane pubHshed "The Ohio Powdery Mildews" in The Ohio Naturalist for May, 1910. The last paper follows the nomenclature of E. S. Salmon and furnishes- keys and short diagnoses. This publication will still be found useful in studying the powdery mildews of Ohio, and I only hope, in the present paper, to supplement it by additions of localities, hosts, and species not previously reported from Ohio. The mycelia of powdery mildews usually occur on the leaves or the small stems or twigs of seed plants, and these fungi are easily collected by those who are accustomed to observe small fungi, provided that the mycelia are fairly con- spicuous. In summer and autumn, these parasites may be observed on the leaves of goldenrods, asters, sunflowers, yard grass, ragweeds, verbenas, roses, willows, oaks, lilacs, and other herbs, shrubs, and trees. Less conspicuous mycelia occur on yellow sorrel, grapes, hackberry, tulip poplar, Ohio buckeye, maples, elms, chestnut, and other seed plants. In order to know whether the perithecia (Fig. 1) are in good condition and to see the inconspicuous forms at all, one should collect with a hand lens at hand. Previous consultation of a host index will add greatly to the success of the collecting trip. 211 212 The Ohio Journal of Science [Vol. XXI, No. 6, The most common powdery mildew in Ohio is Erysiphe dehor aceanim DC. This species grows on a large number of hosts and may be collected from mid-summer until late autumn. We get this fungus about Oxford, in June, for study in the lab- oratory at Miami University. From this time until the host plants are killed by frost, the number of hosts increases, and material for study may be had directly from the field. Nearly as abundant and much better known is Micro- sphcsra alni (Wallr.) Salm, which comes to maturity a little later and continues in condition for study from some of its Fig. 1 Fig. 2 Pig. 1. A perinthecium of Sphaerotheca humuli (DC.) Burr, on leaves of Rosa blanda — a form with short appendages. X 150. Fig. 2. Mycelium, cohidiophores and conidia of Erysiphe polygoni, (DC.) on leaves of garden peas — largely diagramatical. X 150. hosts quite as late in the season. This species is a beautiful object for microscopic study and is the form of powdery mildew most commonly used for laboratory study. However, the appendages, which are wonders of natural beauty, defy the best powers of the artist to reproduce. No drawings that I have seen do them justice, and the sketches ordinarily made by students are far from satisfactory. Some Americans who have known this species from the studies of Dr. T. J. Burrill are doubtless reluctant to follow the revision of Dr. E. S. Salmon. Yet the studies of this worker bear the stamp thoroughness, and it seems best to part company with our American pioneer at points of divergence between him and the later student of the group, with respect to this and other species of the Erysiphacece. April, 1921] Powdery Mildews of Ohio 2ia The powdery mildews should be collected as soon as the perithecia are mature, and if possible, while the mycelium is still fairly conspicuous. Of course the combination can not be had in species which have very evanescent mycelia. Finally, the collecting should be done before wind and rain have caused nearly all of the mature perithecia to disappear, or these fruits have become so ripe that they fall off shortly after being collected. More than once, such material collected in quantity for study in the laboratory has proved satisfactory soon after being brought in from the field, but has been found so nearly devoid of perithecia by the next year as to be practically useless for any purpose. For successful study, the conidial stage must be taken in its prime (Fig. 2), and this is before the perithecia are mature. Young mycelia with such conidial conditions may be found at any time from the middle of June until late autumn. By seeking silvery-white, often glistening stages, devoid of perithecia, any species with a reasonably conspicuous mycelium will serve for this purpose. Erysiphe graminis on grasses, the species on roses, and some forms on plaintain are likely to give good conidial conditions late in the season, while quite as good material may be found on asters and other hosts. Mr. E. E. Duncan made the collections recorded from Montgomery County. All others were made by the writer unless otherwise stated, in the additions to hosts, distribution, and species newly reported for Ohio to follow. Sphaerotheca humuli (DC.) Burr. Butler, Hamilton, and Montgomery Counties. This seems to be the species that causes injury to our roses most commonly throughout Ohio and other portions of the United States. Spcerotheca pannosa (Wallr.) Lev. is said by Salmon to grow commonly on the same hosts in Europe, but to be replaced largely on the roses in America by the other species. Someone might well try to ascertain the facts regarding the cause of the disease of our roses. Sphaerotheca humuli fuliginea (Schlecht.) Salm. Montgomery County. Known on three or four hosts from about as many localities in Ohio. :214 The Ohio Journal of Science [Vol. XXI, No. 6, Sphaerotheca phytoptophila Kell. & Swingle. Butler and Hamilton Counties. On witches brooms of Celtis occidentalis. Erysiphe cichoracearum DC. Butler, Montgomery, and Lake Counties. Also collected in the first county by Freda M. Bachman. Very common and found on an unusually large number of host plants. Erysiphe graminis DC. Butler County, collected and determined by Freda M, Bachman. Probably in good condition when collected, but the herbarium specimens are practically worthless now. I have seen the conidial condition late in autumn, but have not found the perithecia in Ohio. Erysiphe polygoni DC. Butler and Montgomery Counties. Also collected in the first county by W. G. Stover and E. E. Duncan. This mildew is found on several hosts in Ohio. It is very common on yard grass (Polygofium aviculare). Uncinula circinata Cooke & Peck. Butler, Hamilton, and Lake Counties. Also collected in the first county by Freda M. Bachman and V. E. Lantis. The mycelium usually disappears early, but it sometimes accom- panies the perithecia on the maple leaves. Uncinula clintonii Peck. Butler County. Also collected here by E. E. Duncan. Seen but rarely. Previously reported from two localities in Ohio, on Tilia americana, to which it may be confined in America. Uncinula fiexuosa Peck. Butler County. Also collected here by E. E. Duncan. Con- fined to species of Aesculus, and little known in Ohio. Uncinula geniculata Gerard. Butler County, collected and. determined by Freda M. Bachman and later by E. E. Duncan. On Morns rubra. Not previously reported from Ohio. This species is known only in North America and on the single host. The mycelium is inconspicuous. April, 1921] Powdery Mildews of Ohio 215 Uncinula macrospora Peck. Butler and Hamilton Counties. Also collected in the first county by Freda M. Bachman and W. G. Stover. Ususally found on elms in Ohio. Uncinula necator (Schw.) Burr. Butler and Hamilton Counties. Also collected in the first county by Freda M. Bachman and E. E. Duncan. This mildew sometimes damages grapes considerably. Uncinula parvula Cooke & Peck. Butler County. Also collected here by E. E. Duncan. On Celtis occidentalis . Not previously reported from Ohio. Known only in North America and on members of the genus Celtis. Uncinula salicis (DC.) Winter. Butler and Lake Counties. Also collected in the first county by Freda M. Bachman, W. G. Stover, and E. E. Duncan. A beautifully zonate condition, not otherwise known to me, was collected on a species of Populus in Lake County. Unfortunately the zonation has disappeared in the herbarium. Well known in Ohio, and confined mainly to the Salicacece. Podosphaera oxycanthe (DC.) de Bary. Butler County. Also collected here by W. G. Stover. Pre- viously collected in Ohio from several localities and on three or four hosts. Microsphaera alni (Wallr.) Salm. Butler, Highland, Ross, and Montgomery Counties. Also collected in the first county by Freda M. Bachman, W. G. Stover, E. E. Duncan, and J. R. Wright. Probably the best known mildew in Ohio. Common on Syringa vulgaris and found on many other hosts. A variable and confusing species. Microsphaera alni vaccinii (Schw.) Salm. Butler County. Reported from a few other localities in Ohio, with species of Vaccinium and Catalpa as hosts. Microsphaera diffusa Cooke & Peck. Butler County. Collected here also by Freda M. Bachman and E. E. Duncan. On Symporicarpus vulgaris. Recorded for Ohio from several other localities and on other hosts. 216 The Ohio Journal of Science [Vol. XXI, No. 6, Microsphaera euphorbiae (Peck) Burk. & Curt. Butler County. Also collected here by E. E. Duncan. Known now from but three localities in Ohio and on the single host, Euphorbia corollata. Microsphaera grossulariae (Wallr.) Lev. Butler, Montgomery, and Lake Counties. Also collected in the first county by V. E. Lantis. Recorded for Ohio from only two other localities and only on Sambucus canadensis. Microsphaera russellii Clinton. Butler County. Collected here also by E. E. Duncan. Com- mon enough here, but reported from only three other localities in Ohio. Easily overlooked as the mycelium is inconspicuous and usually evanescent. Found only on species of Oxalis, and known to me only on Oxalis stricta. Phyllactinia corylea (Pers.) Karst. Butler and Montgomery Counties. Also collected in the first county by Freda M. Bachman. On Fraxinus americana, Juglans nigra, and Morns rubra. Known from several localities in Ohio, but inconspicuous and usually overlooked. Formerly known under the synonym, Phyllactinia stiff ulta (Reb.) Sacc. Tke Oliio Journal of Science Vol. XXI MAY, 1921 No. 7 THE FIGWORTS OF OHIO.* Mary A. Tavlok. The following study of the species of the Scrophulariaceee of Ohio is based upon the material in the Ohio State Herbarium and upon personal collections. Although the Figworts have for some time been grouped more or less in agreement with their phylogeny, a special attempt has been made to bring the various groups into their natural sequences, proceeding in each case from the generalized to the more specialized species, according to the principles followed by Prof. John H. Schaffner, under whom the study was made, and to whom indebtedness is gratefully acknowledged. The nomenclature follows that of Britton & Brown's Illustrated Flora, second edition. The later important work of F. W. Pennell on the Scrophulariaceas has been carefully considered, and all changes of nomenclature advocated by him have been indicated. It was thought best, however, for the present, in order to facilitate easy reference, to retain the older names for the list as given in the Ohio Catalogue of Vascular Plants. The keys are based upon the characters most evident at the time of blooming and apply only to the local flora. ScROPHULARiACE.^. Figwort Family. Herbs, shrubs, or trees, with opposite or alternate, and usually simple leaves, without stipules, and with perfect, zygomorphic, mostly complete flowers, in clusters or in the axils of the leaves; flowers hypogynous, tetracyclic, pentamerous or the parts reduced; corolla sympetalous, nearly regular, or commonly more or less two-lipped, sometimes spurred or saccate; calyx persistent, with four or five united sepals. Andrecium with five, four, or two fertile stamens, united with the corolla, and alternate with its lobes; if less than five, corn- Papers from the Department ot Botany, The Ohio State University, No. 124. 217 218 The Ohio Journal of Science [Vol. XXI, No. 7, monly with vestiges of the missing ones, didynamous or nearly equal; anthers with four microsporangia, sometimes con- fluent. Gynecium with two united carpels ;■ ovulary usually bilocular; ovules mostly numerous, rarely few, borne on axile placentae; style simple, stigma two-lobed; fruit mostly a septicidal or loculicidal capsule; seeds usually numerous and small, the testa reticulate, pitted, striate, ribbed, or nearly smooth. SYNOPSIS OF THE GENERA. I. Stamens .5, (ours) all with fertile anthers; corolla more or less rotate; upper lip of corolla covering the lower in the bud; leaves alternate. VERBASCAT^. Verba scE^. 1. Verbascum. II. Stamens 4 or 2, with fertile anthers, the fifth and sometimes the two lateral ones sterile or reduced to mere vestiges, or sometimes entirely absent; corolla usually tubular, sometimes more or less rotate. A. Capsule opening by valves, either loculicidal or septicidal; cor.dla not spurred, sometimes gibbous or saccate, corolla usuallj^ with- out a palate in the throat, not with 2 sack-like lips. 1. Upper lip or lobe of the corolla usuallv covering the lower in the bud; capsule usually septicidal.' SCROPHULARIAT^. a. Fifth or sterile filament present, sometimes longer than the fertile, sometimes gland-like or scale-like; herbs; inflorescence usually compound, cymose, or if single, the peduncle mostly 2-bracteate. Chelone^. 2. Penstemon. 3. Chelone. 4. Scrophidaria. 5. Collinsia. b. Fifth stamen absent; trees with opposite leaves; inflorescence large panicle. Paulownie^. 6. Paulownia. c. Fifth stamen vestigial, small; herbs; flowers solitary in axils of bracts or leaves. (a). Stamens 4, all anther-bearing and similar. MlMULE.-E. 7. Mimidus. S. Conobea. (b). Anther-bearing stamens 2, sometimes 2 addi- tional filaments present. Gratiole^. 9. Gratiola. 10. Ilysanthes. 2. Under lip or the lateral lobes of the corolla enfolding the upper in the bud; capsule commonlv loculicidal. RHINANTHAT^. (1). Fertile stamens 4 or 2, not didynamous, or if so, their anthers not enclosed by the upper lip. a. Stamicns 4, all anther-bearing; corolla cam- panulate, salverform, or funnclform, scarcely 2-lipped. (a). Ours with the leaves alternate; cavities of the anthers finally confluent at the top; not parasitic. Digitale^. 11. Digitalis. (b). Ours with the leaves, at least the lower, opposite; cavities of the anthers con- tinuously distinct; plants commonly more or less parasitic. BucHNERE.^i. 12. Afzelia. 13. Dasystoma. 14. Agalinis. 15. Otophylla. 16. Buchnera. May, 1921] The Figworts of Ohio 219 b. Fertile stamens only 2, rarely 4; corolla rotate, salvcrform, tubular, or none. Veronice^. 17. Leptandra. 18. Veronica. 19. Synthyris. (2). Fertile stamens 4, ascending, their anthers enclosed by the upper lip of the corolla. a. Cavities of the ovulary with several to numer- our ovules. Euphrasies. 20. Pedicularis. 21. Castilleja. b. Cavities of the ovulary each with 1 or 2 ovules. 22. Melampyntm. B. Capsule usually opening by chinks or holes; corolla spurred or prominently saccate, or with a palate in the throat, or decidedly 2-lipped, the lips broad and sack-shaped; upper lip of the corolla covering the lower in the bud. ANTIRRHINAT^. 1. Corolla tubular, prominently spurred or saccate, its throat usually closed by the palate; fertile stamens 4, a fifth, minute, vestigial stamen sometimes present. Antirrhine^e. 23. Antirrhinum. 24. Linaria. 25. Chaenorrhimim. 26. Kickxia. 27. Cymhalaria. 2. Corolla 2-lipped, sack-shaped, or the lower one large and slipper-shaped; stamens only 2. Calceolarie.'e. 28. Calceolaria. KEY TO THE GENERA. 1. Anther-bearing stamens 5, the fifth smaller; corolla rotate; leaves alternate. Verbascum. 1. Anther-bearing stamens 4 or 2 ; 2 2. Anther-bearing stamens 4 3 2. Anther-bearing stamens 2 23 3. Corolla definitely spurred on the lower side 4 3. Corolla not spurred, although it may be gibbous or saccate 7 4. Flowers in terminal racemes; leaves narrow Linaria 4. Flowers solitary in the axils 5 5. Leaves linear-spatulate to linear; flowers blue or bluish; throat of the corolla not closed by the palate Chaenorrhimim 5. Leaves broad, pinnately or palmately veined ; the throat of the corolla nearly or quite closed by the palate ; flowers yellowish, variegated, or lilac 6 6. Leaves pinnately-veined, entire Kickxia G. Leaves palmately-veined, coarsely toothed; creeping vine Cymhalaria 7. Trees with large, heart-shaped, opposite, entire leaves; violet, panicled flowers Paulownia 7. Herbs 8 8. Stamens not enclosed in the upper lip of the corolla, or apparently enclosed in the young condition, either included in the throat, exserted, or enfolded by the lower petal; upper lip of the corolla not galeate, or if slightly so, the vestigial stamen large 9 8. Stamens ascending under the upper lip of the corolla; the corolla decidedly 2-lipped, the upper lip galeate; the fifth stamen either minute or wanting. . 21 9. Corolla decidedly saccate or gibbous on the lower side, with a prominent palate closing the throat; the fifth stamen very minute; leaves mostly alternate, entire; upper lip of the corolla covering the lower in the bud A ntirrhinnni P. Corolla tubular or only slightly gibbous, without a palate; frequently with the fifth, sterile stamen prominent 10 10. Fifth or sterile stamen prominent, either a filament, scale, or large gland; upper lip of the corolla covering the lower in the bud : .... 11 10. Fifth stamen minute or wanting 14 220 The Ohio Journal of Scietice [Vol. XXI, No. 7, 11. Corolla 2-cleft, the middle lobe of the lower lip conduplicate, enclosing the stamens and filiform style; odd sterile stamen gland-like Collinsia 11. Lower lip of the corolla not enclosing the stamens and style; the fifth stamen not gland-like, either a long filament or a scale 12 12. Fifth or sterile stamen reduced to a scale on the upper lip of the inflated corolla; stem 4-angled Scrophularia 12. Fifth or sterile stamen a filament, prominently bearded or glabrous; stem cylindrical or only slightly 4-sided 13 13. Corolla somewhat 2-lipped, the upper lip 2-lobed, not arched or keeled; sterile stamen nearly as long, or sometimes longer than the fertile ones, frequently prominently bearded; inflorescence thyrses, panicles, or racemes, sometimes spike-like at the tip Penstemon 13. Corolla decidedly 2-lipped, the upper lip arched, keeled in the middle and notched at the apex; sterile stamen shorter than the fertile ones; fiowers in terminal and axillary dense spikes Chelone 14. Calyx prismatic; leaves opposite; stem square; flowers axillary. . . .Mimulus 14. Calyx not prismatic 15 15. Leaves alternate; flowers in one-sided racemes Digitalis 15. Leaves, at least the lowest, opposite 16 16. Sepals free nearly to the base, linear; upper lip of the corolla covering the lower in the bud, emarginate or 2-lobed; stamens included; annual, diffusely branched, pubescent herbs, with pinna tely-par ted leaves. .Conobea 16. Calyx with a considerable tube, campanulate or turbinate 17 17. Corolla salver-form, purple; flowers in a long, peduncled spike. . . .-. .Buchnera 17. ' Corolla campanulate or funnel-form; flowers not in a long, peduncled spike . . 18 18. Filaments glabrous or nearly so; leaves all, or some of them, auricled at the base; stamens didynamous, the anthers of the shorter stamens smaller Otophylla 18. Filaments wholly, or in part, pubescent or villous: leaves not auricled.. . . 19 19. Stamens nearly equal; calyx lobes as long as the tube; corolla yellow; lower leaves long-petioled and pinnately-parted Afzelia 19. Stamens unequal, strongly didynamous; calyx teeth shorter than the tube. .20 20. Anthers awned at the base; corolla yellow; leaves parted, lobed, or toothed, or if the upper are entire, then lanceolate or ovate-lanceolate. . . .Dasystotna 20. Anthers awnless; corolla purple, pink, or rarely white or yellowish; leaves narrowly linear or linear-lanceolate Agalinis 21. Leaves opposite, entire; ovules 2-4 Melampyrum 21. Leaves alternate, or if opposite or whorled, then compound or lobed; ovules numerous 22 22. Anther-sacs dissimilar, the inner one pendulous by its apex; floral bracts brightly colored; leaves parallel-veined Castilleja 22. Anther-sacs similar and parallel; bracts of the inflorescence not brightly colored; leaves pinnately-veined Pedicularis 23. Corolla strongly 2-lipped, large and showy, the lower lip sack-sliaped and broad; cultivated; sepals 5 or frecjuently the two lower more or less united Calceolaria 23. Corolla nearly regular, or if 2-lipped, the lips not sack-like 24 24. Corolla 2-3 lobed or none; leaves alternate; flowers in dense, terminal, elongated spikes or racemes Synthyris 24. Corolla 4-5 lobed; at least the lower leaves opposite or whorled 25 25. Calyx 4-lobed, rarely 5-lobed, the under lip or lobes of the corolla enfolding the upper in the bud ; flowers solitary in the axils or in racemes or spike- Hke 26 25. Calyx 5-lobed; upper lip of the corolla covering the lower in the bud 27 26. Corolla tubular or salver-form, much longer than the calyx; stamens much exserted, nearly parallel; leaves opposite or verticillate, or both; tall herbs, 2-7 ft. high Leptandra 26. Corolla wheel-shaped, tube short; leaves usually opposite below and alternate above; stamens divergent; low or spreading herbs, 3-30 in. Veronica 27. Peduncles 2-bracteate at the summit; flowers yellow or whitish Graliola 27. Peduncles not bracteate at the summit; flowers purplish Ilysanthes May, 1921] The Figworts of Ohio 221 1. Verbascum (Tourn.) L. Mullen. Mostly biennial, erect herbs, with alternate leaves, and prominent winter rosettes. Flowers pentamerous, in terminal spikes, racemes, or panicles. Corolla rotate; the fifth stamen anther-bearing; some or all of the stamen filaments pilose; ovules and seeds numerous. 1. Plants glabrous or sparingly glandular-pubescent; flowers racemose. V. hlattaria. 1. Plants densely woolly; flowers in dense spikes or spike-like racemes. V. t haps us. 1. Verbascum blattaria L. Moth Mullen. vStem slender, erect, usually simple, glabrous or sparingly glandular- pubescent, 2-6 ft. high. Leaves oblong to ovate, or lanceolate, dentate or pinnatifid, the upper ones more or less clasping, J^-2i^ in. long, the basal ones sessile or short-petioled, up to 12 in. or more in length, forming rosettes, and in late autumn strongly geo tropic, pressed close to earth, and with much anthocyan present. Corolla yellow or white, about 1 in. broad; filaments pilose, magenta. General and abundant in lawns, fields, and waste places. June-Nov. Naturalized from Europe. Flowers and leaves used as medicine. 2. Verbascum thapsus L. Great Mullen. A stout, erect, usually simple biennial, 2-7 ft. high, densely woolly all over, and with large winter rosettes. Stem leaves thick, 4-12 in. long, prominently decurrent, the rosette deaves up to 18 in. long. Flowers yellow, i/2~l in. broad, sessile or nearly so, in dense spikes. Fields, waste places, and pastures, especially on hillsides. June-September. General and abundant. Naturalized from Europe. 2. Penstemon Mitch. Beard-tongue. Erect, perennial herbs with opposite leaves and large, usually showy flowers in terminal thyrses, panicles, or racemes. Corolla tubular, inflated, 2-lipped, the upper lip not arched. Stamens 5, included, 4 of them didynamous, anther-bearing, the fifth sterile, frequently bearded, nearly as long as or longer than the fertile ones. Seeds numerous. 1. Plants more or less glandular or pubescent; leaves dentate or serrate 2. 1. Plants glabrous throughout and glaucous; leaves entire P. grandiflorus. 2. Corolla-tube not prominently enlarged; J^-IM ii^- long; leaves serrate or denticulate; thyrsus elongated and open, usually branched, panicle-like. . .3. 2. Corolla-tube much enlarged above, 2 in. long; thyrsus short; leaves dentate. P. cobaea. 3. Only inflorescence or calyx pubescent, or if pubescent to the base, the upper leaves ovate-lanceolate and usually tapering from near the broad base; throat of corolla slightly bearded 4. 3. Stems pubescent or puberulent nearly to the base; upper leaves usually narrowly-lanceolate; corolla bearded in the throat P. hirsuius. 4. Corolla-tube not gibbous above the point of enlargement; corolla purplish; stems usually puberulent; anther-sacs usually glabrous P. penstemon 4. Corolla- tube gibbous above the point of enlargement; corolla white or purplish; stems usually glabrous; anther-sacs barbate usually... P. digitalis. 222 The Ohio Journal of Science [Vol. XXI, No. 7, 1. Penstemon penstemon (L.) Britt. Tall Purplish Beard- tongue. Usually puberulent, the inflorescence glandular-pubescent, 2-3 ft. high. Leaves 3-6 in. long, the lower ones narrowed into margined petioles, the upper ones sessile or slightly clasping, acute, oblong or lanceolate, denticulate. Inflorescence usually a many-flowered, open thyrsus; corolla purple or purplish, %-% in. long. Anther-sacs usually glabrous. In fields and thickets. May-July. General. 2. Penstemon digitalis (Sweet) Nutt. Foxglove Beard-tongue. Usually glabrous, 2-5 ft. high, the inflorescence glandular-pubescent. Leaves 2-S3^ in. long, the lower and basal ones oblong or oval, nar- rowed into margined petioles, the upper ones sessile or clasping, lanceo- late or ovate-lanceolate. Inflorescence a many-flowered, open thyrsus; corolla white or purplish, 1-13^ in. long, the limb somewhat 2-lipped, the throat open. Anther-sacs usually barbate. In fields and thickets. May-July. General. This is not very definitely separated from the preceding species, and possibly does not deserve specific rank. Specimens collected from the same limited area show smooth stems and barbate anther-sacs, puberulent stems and glabrous anther-sacs, and puberulent stems and barbate anther-sacs. 3. Penstemon hirsutus (L.) Willd. Hairy Beard-tongue. Stems slender, erect, sometimes tufted, downy-hirsute and more or less glandular to the base, frequently purplish, 1-3 ft. high. Leaves puberulent or glabrous, denticulate or the uppermost entire, the basal ones oblong or ovate, narrowed into petioles, the upper ones oblong to lanceolate, sessile or slightly clasping. Inflorescence thyrsoid, rather loose, glandular-pubescent; corolla purplish or violet, about 1 in. long, the tube gradually dilated above, the throat nearly closed by the villous palate. In dry woods and thickets, also on exposed limestone cliffs. May-July. General. 4. Penstemon cobaea Nutt. Cobaea Beard-tongue. Stem stout, densely and flnely pubescent below, glandular-pubes- cent above, 1-2 ft. high. Leaves firm, oblong to ovate, 3-5 in. long, mostly sharply serrate, the lower ones mostly glabrous with margined petioles, the upper ones sessile or cordate clasping, usually pubescent. Thyrsus short, several-many-flowered; flowers about 2 in. long; corolla dull reddish-purple or paler, puberulent without, glabrous within, its tube narrow up to the top of the calyx, then abruptly dilated and campanulate, the limb slightly 2-lipped, the lobes short, rounded, and spreading. Capsule ovoid, acute, pubescent. On prairies. May-July. Lake County. 5. Penstemon grandiflorus Nutt. Large-flowered Beard-tongue. Glabrous and somewhat glaucous with stout stem, 2-4 ft. high. Leaves thickish, entire, the basal ones obovate, narrowed into broad May, 1921] The Figworts of Ohio 223 petioles, the lower stem-leaves sessile, oval or oblong, \-2\^ in. long, the upper stem-leaves nearly orbicular, cordate-clasping, shorter. Thyrsus open, the bracts leaf-like, orbicular, cordate; flowers 2 in. long; corolla lilac or lavender-blue, the tube rather abruj^tly dilated above the calyx, the limb somewhat 2-lipped. Cajjsule acute, considerably longer than the calyx. On prairies, especially on flood-plains. Hamilton County. 3. Chelone (Tourn.) L. Turtlehead. Glabrous perennials, with opposite, serrate leaves, and large white, red, or purple flowers in dense, terminal and axillary spikes or spike-like racemes. Calyx of five distinct sepals, subtended by sepal-like bracts. Corolla inflated-tubular, two-lipped, the upper lip arched, the lower densely pubescent, the two lateral lobes larger than the middle one. Andrecium with five stamens, included, the filaments bearded, four of the stamens didynamous, with densely woolly, heart-shaped anthers, the fifth sterile and shorter than the others. Seeds numerous, flattened, winged. 1. Chelone glabra L. Smooth Turtlehead. A slender, erect, smooth-stemmed perennial, 1-3 ft. high. Leaves opposite, linear-lanceolate to ovate-lanceolate, 3-6 in. long, ^-134^ in. wide, narrowed at the base into a short petiole; leaf margins serrate with sharp, appressed teeth. Flowers white, sometimes delicately tinged at the tips with pink, about 1 in. long, and very striking in appearance. In swamps, wet places, and along streams. July-Septem- ber. Leaves used as medicine. 4. Scrophularia (Tourn.) L. Figwort. Perennial, strong-scented herbs, ours with four-angled stems, usually with opposite leaves, and with small, green or yellow flowers in loose, terminal panicles or cymes. Corolla with a short, somewhat globular tube, the two upper and two lateral lobes erect, the lower one spreading or reflexed. Andrecium with five stamens, four anther-bearing and declined, the fifth one reduced to a sterile scale on the upper lip of the corolla. Seeds numerous, wingless, wrinkled. 1. Sterile stamen deep purple; corolla dull outside; panicle broad; petioles slender, scarceh' margined 5. marylandica. 1. Sterile stamen greenish-yellow; corolla shining outside; inflorescence nar- rowly elongated; petioles stout, evidently wing-margined 5. leporella. 224 The Ohio Journal of Science [Vol. XXI, No 7, 1. Scrophularia marylandica L. Maryland Figwort. Perennial, glabrous below, somewhat glandular-pubescent above, with slender, 4-sided, grooved stem, 3-10 ft. high, and light green leaves, membranous, usually puberulent beneath, ovate or ovate- lanceolate, sharply serrate, long-petioled, 3-12 in .long, the petioles slender, scarcely margined. Flowers small, on pedicels ]{-! in. long, clustered on a long, nearly leafless, broad panicle. Corolla y^ in. long, green, dull without, and brownish purple and shining within. Sterile stamen deep purple. Capsule subglobose with a slender tip. In woods and thickets. July-September. General. 2. Scrophularia leporella Bickn. Hare Figwort. Simple or somewhat branched perennial, 3-8 ft. high, with sharply 4-angled stem with flat sides, the lower part of the plant puberulent, the upper part viscid-glandular. Leaves ovate to lanceolate, narrowed at the base or sometimes subcordate, glabrous on both sides when mature, usually incised-dentate, 2-10 in. long, short-petioled, the petioles evidently, wing-margined. Flowers %-}/2 in- long, in a narrow, elongated panicle. Corolla green to purple, shining without, dull within. Sterile stamen greenish-yellow. Capsule ovoid-conic. In woods and along roadsides. May-July. Cuyahoga, Ashtabula, and Belmont Counties. 5. CoUinsia Nutt. Collinsia. Winter-annual or biennial herbs, with opposite or verticillate leaves. Flowers blue, white, pink, or variegated, verticillate or solitary in the axils. Corolla two-lipped, the upper lip two- cleft, the lobes erect or recurved, with a slight palate; the lower one three-cleft, the middle one conduplicate, enclosing the four stamens and filiform style. Vestigial stamen gland-like, short, with a green tip. Seeds few, large. 1. Collinsia verna Nutt. Blue-eyed-Mary. A slender, branching herb, with weak stem, 6 in. -2 ft. high, glabrous or puberulent. Leaves opposite or verticillate, the lower broadly ovate or orbicular, obtuse, rounded, 'narrowed, or subcordate at the base, crenate or entire, slender-petioled, the upper ones sessile or clasping, 1-2 in. long, ovate or oblong. Corolla slightly pubescent within, the upper lip white, with a slight, purple-spotted palate, the lower one dark blue or sometimes nearly white, with scattered hairs on the outside. Upper pair of stamens glabrous or nearly so, the lower pair pubescent below. Moist woods and hillsides. April- June. General. G. Paulownia S. & Z. Paulownia. A large tree, with broad, opposite, entire or three-lobed, petioled leaves, superposed, axillary buds, prominent lenticels, and more or less diaphragmed pith. Flowers large, violet, in May, 1921] The Fig-worts of Ohio 225 terminal panicles. Stamens four, didynamous, the lateral ones shorter, the fifth one absent. Capsule ovoid, acute. Seeds numerous, winged. 1. Pauiownia tomentosa (Thunb.) Baill. Paulownia. A tree, with thin, flaky bark, up to 70 ft. high, with a trunk diameter of 4 ft., and with broad, heart-shaped leaves, persistently pubescent beneath, 6-15 in. long, 4-8 in. wide, and long, often hollow petioles. Sepals 5, very thick and tomentose; corolla glandular-pubescent on the outside; ovulary densely glandular-pubescent. May-July. Cultivated in Southern Ohio, and escaped in Lawrence County. 7. Mimulus L. Monkey-flower. Erect or creeping, perennial herbs, with opposite, usually dentate leaves, and showy, pink, violet, or yellow, peduncled flowers, solitary in the axils, or raceme-like. Calyx prismatic, 5-toothed or lobed. Corolla tubular, 2-lipped. Stamens 4, didynamous, all anther-bearing. Capsule many-seeded, enclosed by the calyx. 1. Leaves sessile, clasping, prevailingly lanceolate; peduncles considerably longer than the calyx M. ringens. 1. Leaves petioled, prevailingly ovate; peduncles mostly shorter than the calyx M- alatus. 1. Mimulus ringens L. Square-stemmed Monkey-flower. Perennial, glabrous plant, with erect, 4-sided, or somewhat 4-winged stem, often considerably branched, 1-3 ft. high. Leaves oblong or lanceolate, acuminate at the apex, clasping by a heart-shaped base, or the lower ones merely sessile, 2-4 in. long, 3^-1 in. wide. Flowers solitary and axillary; peduncles considerably longer than the calyx; corolla 1-13^ in. long, violet, or rarely white, with 2 yellow spots near the narrow throat, the upper lip erect, the lower spreading; calyx lobes lanceolate. In swamps, along streams, and in wet places. June- September. General. 2. Mimulus alatus Soland. Sharp-winged Monkey-flower. Perennial, glabrous, similar to the preceding species, but with sharply 4-angled stem, the angles more or less winged. Leaves ovate, ovate-lanceolate, or oblong, narrowed at the base, petioled, 2-5 in. long, yi-XYi in. wide; petioles \i-\ in. long. Calyx-lobes setaceous-tipped. Peduncles shorter than the calyx. In swamps and wet places. June- September. Rather general. 8. Conobea Aubl. Conobea. Low, branching herbs, with opposite, pinnately-parted or serrate leaves, and small, blue or white flowers, solitary or two together in the axils. Calyx of 5, equal, linear sepals, free nearly to the base. Corolla 2-lipped. Stamens 4, fertile, didynamous, included. Anther-sacs parallel. Seeds numerous. 226 The Ohio Journal of Science [Vol. XXI, No. 1. Conobea multifida (Mx.) Benth. Conobea. A viscid-pubescent, diffusely spreading, much branched annual, 4-S in. high. Leaves opposite, i^^-l in. long, pinnately-parted, the segments linear or linear-oblong. Flowers small, blue; corolla slightly longer than the calyx. Along streams and rivers. June-September. Ham- ilton, Greene, Madison and Ottawa Counties. 9. Gratiola L. Hedge-hyssop. Low herbs inhabiting wet or damp places, with opposite, sessile, entire or dentate leaves. Flowers white or yellowish, peduncled, solitary and axillary, usually with two bractlets at the base of the calyx. Corolla more or less two-lipped. The two upper stamens fertile, the lower pair vestigial, minute, or sometimes represented by two slender, capitate filaments. Seeds numerous, striate. 1. Plant glabrous or nearly so; peduncles much shorter than the subtending leaf-bracts; corolla within throat pubescent with knobless hairs. G. sphaerocarpa. 1. Plant glandular-puberulent; peduncles as long as or longer than the subtend- ing leaf-bracts; corolla within throat pubescent with knobbed hairs. G. virginiana. 1. Gratiola sphaerocarpa Ell. Round-fruited Hedge-hyssop. Glabrous, ascending or erect, annual, simple or branched, G-12 in. high. Leaves oblong or obovate-oblong, sessile, toothed, 3-5 nerved, narrowed at the base, 1-2 in. long, M~/^ in. wide. Peduncles stout, much shorter than the subtending leaf -bracts. Flowers small, the corolla tube yellow, the limb paler; corolla within throat pubescent with knobless hairs. Capsule globose. In wet places. June-September. Erie County. (Gratiola virginiana L., according to Pennell.) 2. Gratiola virginiana L. Clammy Hedge-hyssop. Annual, stem glandular-puberulent above, widely branched, 3-12 in. high. Leaves oblong or oblong-lanceolate, sessile, narrowed at both ends, entire or sparingly toothed, 1-2 in. long, 'k-3^^ in. wide. Peduncles slender, glandular, nearly as long as or longer than the subtending leaf- bracts. Corolla-tube yellowish; limb almost white; corolla within throat pubescent with knobbed hairs; sterile filaments minute or none. Capsule broadly ovoid, as long as the calyx. In wet places. May- October. General. (Gratiola neglecta Torr., according to Pennell.) 10. Ilysanthes Raf. False Pimpernel. Small, smooth, annual or biennial herbs, with opposite, sessile, usually dentate leaves. Flowers small, purplish, peduncled, without bracts, solitary in the axils. Corolla two-lipped. The two upper stamens fertile, and included, the two lower ones sterile, two-lobed, the one lobe capitate and glandular, the other smooth and shorter. Seeds numerous, wrinkled. May, 1921] The Figworts of Ohio 227 1. Peduncles longer than the subtending leaf-bracts; calyx-segments shorter then the capsule /. dubia. 1. Peduncles shorter than the subtending leaf-bracts; calyx-segments as long as the capsule or longer /. attenuata. 1. Ilysanthes dubia (L.) Barnh. Long-stalked False Pimpernel. Stem square, at first simple and ustially erect, later much branched and diffusely spreading, 3-1 (J in. long. Leaves ovate, obovate, or oblong, usually sessile, clasping, entire or sparingly toothed, 3^-1 in. long. Coiolla pale lilac, 34^3^ i^- long, the peduncles usually considerably longer than the subtending leaf -bracts; calyx-segments usually shorter than the capsule. In wet places. July-September. Meigs, Fairfield, Licking, Franklin, Huron, Auglaize, Defiance and Scioto Counties. 2. Ilysanthes attenuata (Muhl.) Small. Short-stalked False Pimpernel. Stem square, erect or ascending, 3-16 in. long, the branches spread- ing. Leaves ovate, obovate, or oblong, 3^-1% in. long, sparingly toothed. Corolla pale lilac to nearly white, ]i-}/2 in. long; peduncles mostly shorter than the subtending leaf -bracts; calyx-segments as long as the capsule or longer. In wet places. May-October. Cuyahoga, Stark, Summit, Madison, Scioto, and Highland Counties. The two species are evidently closely related as inter- mediates show peduncles longer than the subtending leaf- bracts, but with calyx-segments as long as the capsule. 11. Digitalis (Tourn.) L. Foxglove. Tall, erect herbs, with alternate, entire or dentate leaves, and showy, yellow, purple, or white flowers in long, terminal, usually one-sided racemes. Calyx 5-parted. Corolla declined, slightly two-lipped, the tube contracted above the ovulary, then abruptly inflated. Stamens four, didynamous, included. Seeds numerous, rugose. 1. Corolla purple to white, 13^-2 in. long; stem pubescent; leaves slender- petioled D. purpurea. 1. Corolla yellow, about % in. long; stem smooth; leaves sessile D. lutea. 1. Digitalis purpurea L. Purple Foxglove. Stout, erect, pubescent herb, usually biennial, 2-5 ft. high. Lower leaves ovate or ovate-lanceolate, 6-10 in. long, acute at the apex, nar- rowed at the base into petioles, dentate; upper leaves smaller, sessile. Racemes 1 ft. or more in length, dense, one-sided; flowers purple to white, 13^-2 in. long, drooping, the corolla spotted within; upper calyx- lobe narrower than the others. Escaped from cultivation. June- August. Cuyahoga and Lake Counties. From Europe. Leaves of the second year's growth used officially as medicine. 228 The Ohio Journal of Science [Vol. XXI, No. 7, 2. Digitalis lutea L. Yellow Foxglove. Glabrous perennial, 2-3 ft. high. Leaves oblong or lanceolate, denticulate, 4-4j^ in. long, % in. wide, ciliate only on the margin, sessile, narrowed at the base. Raceme many-flowered; corolla yellow to white, glabrous outside; calyx-segments lanceolate, acute. June- August. A waif in Cuyahoga Count^^ From Europe. 12. Afzelia Gmel. Mullen Foxglove. (Dasistoma Raf., according to Pennell). Erect herbs with opposite leaves, and yellow flowers solitary in the axils of leaf-like bracts. Corolla rotate-campanulate, the lobes nearly equal, and longer than the tube. Stamens 4, slightly unequal, not exserted; anthers glabrous; filaments short, villous; anther-sacs parallel and distinct. Seeds numerous, reticulated. 1. Afzelia macrophylla (Nutt.) Ktz. Mullen Foxglove. Puberulent or glabrate, simple or sparingly branched, 4-6 ft. high, perennial by buds in the crown. Lower leaves long-petioled, pinnately- divided, 6-15 in. long, the segments dentate or pinnatifid; the upper leaves short-petioled or sessile, entire, 1-3 in. long. Flowers 3^-/8 in. long, sessile and solitary in the axils of the upper leaves; calyx-lobes lanceolate or ovate; corolla yellow, 2-3 times as long as the calyx, the corolla-tube, except the lower part, very woolly inside. Capsule globose, pointed. In thickets, and along streams. August-October. General in western Ohio, as far east as Huron, Noble, and Vinton Counties. (Dasistoma macrophylla (Nutt.) Raf., according to Pennell.) is. Dasistoma Raf. False Foxglove. (Aureolaria Raf., according to Pennell). Large, erect herbs, more or less parasitic on the roots of other plants, with opposite, or whorled, or some alternate leaves, and large, showy, yellow flowers in terminal, usually leafy-bracted racemes. Calyx campanulate or turbinate, 5-lobed, the lobes sometimes foliaceous. Corolla funnel-form or cam- panulate-funnelform, with five rather unequal, spreading lobes, the tube villous or pubescent within. Stamens 4, didynamous, included, villous or pubescent, the anther-sacs parallel and awned at the base. Capsule oblong, acute. \. Leaves all toothed or pinnatifid 2. 1. Leaves, at least the upper ones, entire 3. 2. Plants, especially the sterns, glaucous or glabrous; calyx and corolla glabrous outside; perennial ". D. virginica. 2. Plants glandular-pubescent or hirsute; calyx rind corolla pubescent outside; annual D. pedicidaria. 3. Plants jjubescent; calyx pubescent outside D. flava. 3. Plants glabrous; calyx glabrous outside; leaves all entire or the lower ones dentate D. laevigata. May, 1921] The Figworts of Ohio 229 1. Dasystoma virginica (L.) Britt. Smooth False Foxglove. Glabrous and glaucous perennial with stout, usually branched stem, 3-G ft. high. Leaves usually all petioled, the lower ones pinnatifid, 4-G in. long, the upper ones pinnatifid or deeply incised. Flowers very striking; corolla pure yellow, glabrous outside, about 2 in. long, in leafy- bracted racemes. Capsule glabrous. In woods. July-September. Adams, Fairfield, Clarke, Cuyahoga, Fulton, and Wood Counties. (Aureolaria flava (L.) Farw., according to Pennell.) 2. Dasystoma pedicularia (L.) Benth. Fernleaf False Foxglove. A much branched, leafy annual or biennial, 1-4 ft. high, more or less glandular and viscid. Upper leaves sessile, the lower ones usually petioled, pinnatifid, 1-3 in. long. Corolla 1-13^ in. long, pubescent outside; calyx-lobes foliaceous, usually pinnatifid or incised. Capsule pubescent. In dry woods and thickets. August-September. Fulton County. (Aureolaria pedicularia (L.) Raf., according to Pennell.) Our two specimens belong to the variety A. ambigens (Fern.) Farw., densely glandular-hirsute . 3. Dasystoma flava (L.) Wood. Downy False Foxglove. Erect, usually simple, sometimes branched perennial, 2-4 ft. high, pubescent with a fine, grayish down. Leaves oblong, lanceolate, or ovate-lanceolate, usually opposite, entire, or the lower ones sinuate- dentate, or sometimes pinnatifid, 3-G in. long, short-petioled, the upper ones much smaller and sessile, becoming bract-like. Corolla pure yellow, 1^-2 in. long, glabrous outside. Calyx and capsule pubescent. In dry woods and thickets. July-August. Eastern Ohio, as far west as Erie, Clarke and Adams Counties. (Aureolaria virginica (L.) Pennell.) 4. Dasystoma laevigata Raf. Entire-leaf False Foxglove. Simple or sparingly branched perennial, 1-3 ft. high, glabrous or nearly so, but not glaucous. Leaves lanceolate or ovate-lanceolate, 1H~^ in. long, usually petioled, the upper ones entire, the lower ones dentate or incised. Corolla yellow, glabrous without, hairy within, l-l^/i in. long, the limb fully as broad. Capsule glabrous. In dry thickets. July-August. Jackson, Adams, Vinton, Hocking, Fairfield, and High- land Counties. (Aureolaria laevigata (Raf.) Raf., according to Pennell.) • 14. Agalinis Raf. Agalinis. Erect, branching herbs, some shrubby, with opposite, entire, sessile leaves, and large, showy flowers in racemes or panicles, or solitary in the axils. Corolla slightly two-lipped, campanulate or funnelform, five-lobed. Stamens four, didynamous, included; filaments pubescent. Capsule ovoid or globose. Seeds numerous, mostly angled. 230 The Ohio Journal of Science [Vol. XXI, No. 7, 1. Pedicels shorter than or but slightly longer than the calyx; plants of moist ground 2. 1. Pedicels much longer than the calyx, usually exceeding the corolla; plants of dry ground ;3. 2. Corolla %-\}4 in- long A. purpurea. 2. Corolla Yz-'^/i in. long A. paupercula. 3. Plants leafy; leaves flat, linear to lanceolate, ^-1/2 in. long; pedicels often shorter than the subtending leaf-bracts A. fenuifolia. 3. Plants very slender; leaves few and distant, filiform or with revolute margins, three-eighths to five-eighths in. long; pedicels usually longer than the subtending bracts A. skinneriana. 1. Agalinis purpurea (L.) Pennell. Large-flowered Agalinis. vSmooth or roughish annual, with slender stem, l-2i^ ft. high, and with long, rigid, widely spreading branches. Leaves narrowly linear, usually widely spreading, 1-23^ in. long, rarely with clusters in their axils, rough-margined. Flowers rose-purple or rarely white, '%-\}/i in. long; pedicels shorter than or but slightly longer than the calyx; corolla much expanded above, often downy, the lobes all spreading. Capsule globose. In low ground, moist fields and meadows. August-October." Rather general. 2. Agalinis paupercula (Gr.) Britt. Small-flowered Agalinis. Annual, glabrous or nearly so, 6-18 in. high, the whole plant very similar to the preceding species, but smaller, apparently intergrading with it. Leaves narrowly linear, 3^-1 in. long. Corolla rose-purple, the lobes all spreading, ^2 /i in- long. Capsule globose-oblong. In bogs and low meadows. July-September. Stark, Ottawa, Logan, Cham- paign, and Gallia Counties. 3. Agalinis tenuifolia (Vahl.) Raf. Slender Agalinis. Glabrous annual with slender stem, 6-24 in. high; plant very leafy. Leaves flat, linear to lanceolate, acute, Yi-XYi in. long. Flowers light rose-purple, spotted, rarely white, %-^ in. long, the two upper lobes ascending over the stamens and style; pedicels often shorter than the subtending leaf-bracts. Capsule globose or slightly obovoid. In dry woods and thickets. August-October. General. Our specimens include both the typical form and A. tenuifolia macrophylla (Hook.) 4. Agalinis skinneriana (Wood) Britt. Skinner's Agalinis. Very slender, roughish annual, 6-18 in. high. Leaves few and distant, ){-% in. long, filiform or with revolute margins, commonly erect and appressed. Corolla light rose-purple or white, up to Y^ in. in length; pedicels usually longer than the subtending leaf-bracts. Cap- sule oblong. In dry, sandy woods and thickets. August-October. White-flowered form from southeastern part of Fulton County. 15. Otophylla Benth. Otophylla. Annual, hirsute-pubescent herbs, with opposite, sessile, entire or pinnately-divided leaves, all or some of them auricled at the base. Flowers purple or white in terminal spikes. May, 1921] The Figworts of Ohio 231 Corolla-tube broadly dilated at the throat; lobes spreading, stamens four, didynamous, included; filaments glabrous or nearly so; anthers awnless, those of the shorter stamens much smaller than the others. Seeds angled. 1. Otophylla auriculata (Mx.) Small. Auricled Otophylla. Rough-hairy annual, with slender and usually simple stem, 1-2 ft. high. Leaves 1-2 in. long, lanceolate or ovate-lanceolate, acviminate at the apex, sessile, the lower leaves usually entire, the others entire but commonly with a short, oblong-lanceolate lobe on each side at the base. Flowers solitary in the upper axils, purple, sessile, about % in. long, denselv puberulent outside, glabrous within. Filaments glabrous or sparingly hairy; anther-sacs very unequal. In moist, open soil, low grounds, and prairies. July-September. Ottawa County. 1(). Buchnera L. Blue-hearts. Erect, hispid or scabrous perennials or biennials, mostly with opposite leaves and large, white, blue, or purple flowers in dense, terminal, bracted spikes. Corolla salverform; tube somewhat curved; lobes five, somewhat unequal, spreading. Stamens four, didynamous, included; anther-sacs confluent. Capsule oblong. Seeds numerous, reticulated. 1. Buchnera americana L. Blue-hearts. Perennial; stem slender, stiff, hispid and rough, 1-23^ ft. high. Leaves prominently veined, usually all opposite, the lower ones obovate or oblong, obtuse, narrowed into short petioles or sessile, sparingly and coarsely toothed, the upper ones lanceolate or linear-lanceolate, entire or nearly so. Spike peduncled, 6-10 in. long in fruit; flowers mostly opposite, about 1 in. long; subtending bractlets shorter than the calyx; corolla purple, its lobes obovate, obtuse. Capsule oblique, a little longer than the calyx. In sandy or gravelly soil. June-September. Fulton County. 17. Leptandra Nutt. Culver's-root. (Veronicastrum Heist.) Tall, erect, perennial herbs, with verticillate or opposite leaves. Flowers small, blue or white, in dense, spike-like racemes. Calyx 4-parted. Corolla tubular or salverform, four-lobed, only sHghtly two-lipped. Stamens two, much exserted. Seeds numerous, oval, minutely reticulated. 1. Leptandra virginica fL.) Nutt. Culver's-root. Perennial, with simple, erect stem, 2-7 ft. tall, glabrous or nearly so. Leaves verticillate, or some of the uppermost opposite, lance-shaped, finely serrulate, 3-6 in. long, ^-1 in. wide. The small, white or nearly white flowers are in dense, spike-Hke racemes, 3-9 in. long. In woods, thickets, and open places. June-Sei)tember. General. (Veronicastrum virginicum (L.) Farw.) Rhizome and roots used officially as medicine. 232 The Ohio Journal of Science [Vol. XXI, No. 7, IS. Veronica (Tourn.) L. Speedwell. Chiefly herbs, with small, blue, pink, or white flowers, in terminal or axillary racemes or spikes, or solitary in the axils. Calyx usually four-parted, sometimes five-parted. Corolla wheel-shaped, the tube short, four-lobed, rarely five-lobed, the lower lobe commonly the narrowest. Stamens two, exserted, divergent, the anther-sacs confluent at the apex. Capsule flattened, obtuse or notched at the apex. Seeds numerous, smooth or rough. 1. Flowers in axillary racemes, their bracts small _ 2. 1. Flowers solitary in the axils, the subtending bracts leaf-like and similar to the leaves, usually becoming smaller toward the top S. 2. Calyx 4-parted '^• 2. Calyx 5-parted, the upper point small, all the sepal points with bristles; leaves ovate to lanceolate, sessile; racemes compact and showy. V. teucrium. 3. Plants hairv; leaves ovate to obovate; species of dry soil 4. 3. Plants glabrous or minutely glandular; if hairy, then the leaves linear or linear-lanceolate; species of low ground and brook margins, or aquatic. . .5. 4. Stem pubescent in 2 lines; leaves ovate, pointed; pedicels longer than the calyx V. chamaedrys. 4. Stem hairy all over; leaves oval or obovate; pedicels shorter than the calyx V- officinalis. 5. Leaves all short-petioled V. americana. 5. Leaves sessile and clasping,. or only the upper or lowermost petioled .6. 6. Leaves ovate-oblong or oblong-lanceolate V. anagallis-aqiiatica. (1. Leaves lanceolate to linear ' • 7. Leaves linear or nearly so, three-eighths to five-eighths in. wide; pedicels not glandular; capsule much wider than long, strongly 2-lobed , . V. scutellata. 7. Leaves lanceolate, broadest near the base, or the lowest elongated, lanceo- late, clasping, five-eighths to one in. wide; rachis and pedicels sparsely pubescent with glands; capsule broad, globose, emarginate. . . . V. glandifera. 8. Pedicels shorter than the subtending bracts 9. 8. Pedicels longer than the subtending bracts IL 9. Corolla white or pale blue, sometimes streaked with dark blue; leaves of the oblong type, glabrous or short-pubescent; calyx and bracts glabrous or slightly short-pubescent 10- 9. Corolla dark blue; leaves of the ovate type, 5-7 palmately nerved; long glandular-pubescent, including the bracts and calyx; annuals. . . F. arvensis. 10. Pedicels equalling the calyx; stem glabrous or pubescent; flowers in narrow racemes, more or less peduncled, the bracts becoming abruptly smaller than the uppermost foliage leaves; perennials V. serpyllifolia. 10. Pedicels much shorter than the calyx; stem glabrous or short-pubescent; corolla white or very pale blue; annuals V. peregrina. 11. Sepals not heart-shaped at maturity, prominently veined; leaves of the ovate or oblong type; rather short petioles; leaves crenately cut-toothed, 12. 11. Sepals heart-shaped at maturity, without prominent veins; leaves of the orbicular type or very broad, 3-5 lobed or 3-5 crenate; rather long petioles V- hederaefolia. 12. Corolla not longer than the calyx V. ngrestis. 12. Corolla lon.ger than the calyx V. tournejortii. May, 1921] The Figworts of Ohio 233 1. Veronica teucrium. L. Germander Speedwell. Pubescent perennial, with erect stem, uj) to 20 in. high. Leaves ovate to Hnear-lanceolate, nearly entire, crcnate to even bluntly dentate- incised, mostly sessile. Racemes opposite, elongated, many-flowered; flowers large, blue or violet, rarely rose or white; calyx-segments oblong- linear to lanceolate. Capsule obovate. Rare in grass or waste land. August. Medina County. From Europe. 2. Veronica chamaedrys L. Bird's-eye Speedwell. Sim]jle or branched perennial, with slender, ascending stem, pubes- cent in 2 lines, 4-12 in. high. Leaves ovate, sessile or nearly .so, pubes- cent, truncate, rounded, or cordate at the base, incised-dentate, obtuse at the apex, Yr-\}/i in. long. The loose racemes, 2-6 in. long, are long- peduncled, 10-20 flowered, and either in opposite or alternate axils; flowers light blue or violet-blue, about % in. broad; pedicels longer than the calyx, and usually longer than the subtending bractlets. Capsule obcordate, narrowed at the base. In fields and waste places. May- July. Lake County. From Europe. 3. Veronica americana Schwein. American Speedwell. Glabrous perennial, at first decumbent, later erect and branching, rooting at the lower nodes, 6 in. -3 ft. long. Leaves oblong or oblong- lanceolate, short-petioled, sharply serrate, truncate, rounded, or sub- cordate at the base, 1-3 in. long, 1<^-1 in. wide. Racemes loose, elongated, sometimes 6 in. long, usually 10-25 flowered, peduncled, and in most of the axils ; bractlets shorter than the pedicels ; flowers blue or nearly white, usually striped with purple. Capsule nearly orbicular, compressed. In brooks and swamps. Plant emersed. April October. General. 4. Veronica anagallis-aquatica L. Water Speedwell. Perennial, with stout stem, glabrous or glandular-puberulent above, erect or decumbent, usually branched, often rooting at the lower nodes, 1-3 ft. high. Leaves of the sterile, autumnal shoots orbicular to obovate, serrulate, narrowed into margined petioles; leaves of the flowering stems lanceolate to oblong, 1^-4 in. long, }/4-2 in. wide, sessile and more or less clasping, or the lowest ones short-petioled, finely serrate or nearly entire. Racemes 2-G in. long, peduncled; flowers blue, often purple- striped. Capsule nearly orbicular, 2-lobed. In brooks and swamps. May-September. Butler, Champaign, Auglaize, Lucas, Erie, Miami, Clark, and Highland Counties. 5. Veronica scutellata L. Skullcap Speedwell. Perennial, usually glabrous, sometimes pubescent or hairy. Stem slender, decumbent or ascending, leafy, simple or branched, 6 in.-2 ft. high. Leaves sessile, slightly clasping, linear or linear-lanceolate, remotely denticultae, 1-33^ in. long, }s to nearly % in. wide. Racemes axillary, equalling or longer than the leaves; peduncles slender; flowers small, blue, scattered, on very slender, spreading pedicels; bractlets shorter than the pedicels. Capsule much wider than long, strongly 234 The Ohio Journal of Science [Vol. XXI, No. 7, 2-lobed. In brooks and swamps. May-September. Cuyahoga, Ottawa, Erie, Perry, Lucas, Crawford, Licking, and Franklin Counties. 6. Veronica officinalis L. Common Speedwell. Pubescent perennial, with prostrate but finally erect stem, 3-12 in. high. Leaves oblong, oval, or obovate, 3^-2 in. long, obtuse, serrate, narrowed at the base into short petioles. Racemes spike-like, narrow, dense, elongated, and axillary, much longer than the leaves; bractlets longer than the pedicels; flowers pale blue, striped with violet. Capsule obovate-cuneate. In dry fields and on hills. May-August. General and abundant. In colonial times grown as a medicinal plant. 7. Veronica serpyllifolia L. Thyme-leaf Speedwell. Perennial, glabrous or puberulent, with slender, branching stem, 2-10 in. high. Leaves oval, ovate, or oblong, crenulate or entire, short- petioled, or the uppermost sessile. Flowers in short, narrow, loose racemes, at the end of stem or branches; bractlets leaf-like, becoming abruptly smaller than the uppermost leaves; pedicles shorter than the bracts; corolla whitish or blue, with deeper stripes. Capsule rounded, obcordate or emarginate at summit, broader than high. Fields, lawns, roadsides, and thickets. April-August. General. 8. Veronica peregrina L. Purslane Speedwell. Erect, glabrous or glandular- puberulent, branching annual, 3-13 in. high. Leaves oblong, oval, linear, or only slightly spatulate, /^-/^ in. long, the lowest ones opposite, sessile, usually denticulate, the upper ones alternate, sessile, mostly entire, longer than the flowers. Flowers solitary, axillary, usually white or very pale blue; pedicels shorter than the calyx. Capsule nearly orbicular, usually a little shorter than the calyx, glabrous. Waste and cultivated grounds, in damp soil. May- October. General. Our specimens include the glandular-pubescent form. .V. peregrina xalapensis (H. B. K.) Pennell. 9. Veronica arvensis L. Field Speedwell. Simple or diffusely-branched, pubescent annual, 3-11 in. long. Lower leaves opposite, oval or ovate, crenate, the lowest ones petioled, the uppermost sessile, alternate, ovate or lanceolate, commonly entire. Flowers small, dark blue, solitary in the axils; pedicels shorter than the calyx. Capsule broadly obovate, obcordate. Fields, lawns, and woods, and cultivated soil. March-September. General. From Europe. 10. Veronica agrestis L. Garden Speedwell. Slender, pubescent annual, with creeping or procumbent stem, 3-S in. long, and ascending or spreading branches. Leaves broadly ovate, obtuse at the apex, truncate or subcordate at the base, crenate, short- petioled, the lower ones opposite, the upper ones alternate. Flowers small, blue, solitary in the axils, long-peduncled, the peduncles equalling or longer than the leaves. Corolla not longer than the calyx. Capsule broader than high, not very flat, narrowly emarginate at the summit. In fields and waste places. ' May-September. Hamilon, Montgomery, and Franklin Counties. From Europe. May, 1921] The Figworts of Ohio 235 11. Veronica tournefortii Gmel. Tournefort's Speedwell. Diffusely-branched, pubescent annual, 6-18 in. long. Leaves oval or ovate, short-petioled, crenate-dentate, %-\]i in. long, the lower ones opposite, the upper ones alternate. Flowers blue, comparativeh' large, on slender peduncles, solitary in the axils of the upper leaves; peduncles as long as or longer than the leaves ; petals exceeding the sepals. Capsule twice as broad as high, with a wide, shallow indentation at the summit. In waste places. May-September. Madison, Franklin, Jefferson, Lorain, Cuyahoga, Lake, and Belmont Counties. From Europe. 12. Veronica hederaefolia L. Ivy-leaf Speedwell. Slender, diffusely-branched, pubescent annual. 3-18 in. long. Leaves broadly cordate, 3-5 lobed, 3-5 crenate, petioled, 3^-1 in. wide, the lower ones opposite, the upper ones alternate. Flowers small, blue, axillary, peduncled, the peduncles often longer than the leaves; corolla scarcely longer than the calyx; sepals densely ciliate, becoming heart- shaped at maturity. Capsule only slightly 2-lobed, scarcely notched at the apex, but little compressed. In thickets, fields, and waste places. Erie and Hamilton Counties. From Europe. 19. Synthyris Benth. Synthyris. Perennial herbs, with simple, erect stems and a thick rhizome, with alternate leaves, the basal leaves large and petioled, the upper stem leaves smaller, sessile or partly clasping and bract-like. Flowers small, pink, purple, or greenish- yellow, in dense, elongated spikes or racemes. Corolla 2-4-lobed or cleft, or sometimes wanting. Stamens two, sometimes four, exserted; anther-sacs parallel or divergent. Seeds numer- ous, flat, oval or orbicular. 1. Synthyris buUii (Eat.) Heller. Bull's Synth^^ris. Pubescent perennial, with a stout stem, 1-23^2 ft. high. Basal leaves ovate or orbicular, rounded at the apex, truncate or cordate at the base, crenulate, 2-5 in. long, 5-7 nerved, with petioles usually shorter than the blades; stem leaves small, crenulate, sessile oj slightly clasping, gradually becoming smaller. Inflorescence a dense spike, elongating in fruit; flowers small, greenish -yellow; the corolla, if present, commonly 2-lobed. Capsule emarginate, slightly exceeding the calyx. On dry prairies. May-July. Montgomery County. 20. Pedicularis (Tourn.) L. Lousewort. Erect herbs, with pinnately-lobed, cleft, or pinnatifid leaves, and rather large flowers in spikes or spike-like racemes. Corolla strongly two-lipped, the upper lip arched, laterally compressed and often beaked at the apex, enclosing the four didynamous stamens. Anther-sacs similar and parallel. Ovules and seeds, numerous. 236 The Ohio Journal of Science [Vol. XXI, No. 7, ]. Leaves parti}'' opposite; leaf-blades deeply toothed or somewhat pinnatifid; upper lip of corolla truncate. Aug. -Oct P. lanceolata. 1. Leaves scattered; leaf-blades pinnately-parted; upper lip of corolla with 2 lateral teeth. Apr. -June P. canadensis. 1. Pedicularis lanceolata Mx. Lanceleaf Lousewort. A stout, erect, usually simple perennial, 1-3 ft. high, glabrous or nearly so. Leaves mostly opposite, lanceolate, or linear-lanceolate, 2-5 in. long, finely cut. Spikes short; calyx 2-lobed, with foliaceous mar- gins; corolla pale yellow, about % in. long. Capsule ovate, scarcely longer than the calyx. Swampy places. August-October. Rather general, but no specimens from south of Montgomery and Hocking Counties. 2. Pedicularis canadensis L. Wood Lousewort. Pubescent or hirsute perennial, (J-18 in. high, with simple stem, commonly tufted, the whole plant more or less tinged with purple. Basal leaves clustered, 9-14 in. long, slender-petioled. Stem leaves scattered, 3-5 in. long, decurrent; leaves of the inflorescence gradually reduced to bracts. Flowers in short, terminal, dome-shaped spikes; calyx cleft on the lower side, with several tooth-like lobes on the upper side; corolla pale greenish-yellow, the upper lip tinged with brown or brownish-purple, 's-1 in. long. Inflorescence decidedly elongating in fruit, 5-12 in. long; capsule flat, about twice as long as the calyx. Hill- sides and thickets. April-June. General. 21. Castilleja. Mutis. Painted-cup. Erect herbs, parasitic on the roots of other plants, with alternate leaves and usually brightly-colored, inflorescence bracts. Flowers in spikes or racemes. Calyx laterally com- pressed, deeply cleft above or also below^ Corolla decidedly zygomorphic and two-lipped, the upper lip arched and enclosing the four didynamous stamens. The two lobes of the anther unequal, the outer attached to the filament at the middle, the inner one by its apex. Capsule ovoid or oblong, many-seeded. 1. Castilleja coccinea (L.) Spreng. Scarlet Painted-cup. Annual or commonly biennial, villous-pubescent, rather slender plant, 1-2 ft. high. Stem leaves prominently parallel-veined, divided into 3-5 linear, obtusish segments, sessile; the basal leaves in a rosette, parallel- veined or some at least, with two prominent side ribs from the base, entire. Inflorescence bracts 3-5 cleft, the lower green, the upper ones orange-scarlet, very conspicuous. Corolla ^-1 in. long, greenish- yellow, the three lower lobes plicate. Ovulary with a green gland and prominent protuberance at its base. Moist meadows and hillsides. May-July. Franklin, Knox, and Madison Counties. May, 1921] The Figworts of Ohio 237 22. Melampyrum (Tourn.) L. Cow-wheat. Erect, branching, annual herbs with opposite leaves, and small flowers solitary in the axils or in terminal, bracted racemes. Calyx four-toothed, the two upper ones somewhat longer. Corolla two-lipped, the upper lip arched, compressed, enclosing the four didynamous stamens. Capsule flat, oblique, 2-4-seeded ; seeds smooth. 1. Melampyrum lineare. Lam. Narrow-leaf Cow-wheat. A delicate annual with a slender, puberulent, wiry, somewhat 4-sided stem, 6 in.-lj/^ ft. high. Leaves light green, opposite, short-petioled, the floral ones with bristle-pointed teeth near the base or entire, the lower ones lanceolate or linear-lanceolate to ovate, 1-2 3^^ in. long, 3^ in. wide. Flowers about 3^ in. long, short-peduncled; corolla white and yellow, puberulent. In dry woods and thickets. May-August, Lorain, Cuyahoga, Lake, Ashtabula, Geauga, Portage, and Hocking Counties » 23. Antirrhinum (Tourn.) L. Snapdragon. Annual or perennial herbs, with mostly opposite, entire leaves, and large, showy flowers in terminal racemes or solitary in the upper axils. Corolla decidedly saccate or gibbous on the lower side, with a prominent palate closing the throat. Andrecium with four didynamous, anther-bearing stamens, included, the lateral pair longer, and a very minute, sterile, fifth stamen. Capsule many-seeded. 1. Antirrhinum ma jus L. Great Snapdragon. Perennial, 1-3 ft. high, glabrous except near the inflorescence, which is usually more or less glandular-pubescent. Leaves lanceolate, linear, or oblong-lanceolate, entire, acute at both ends, short-petioled, 1-3 in. long. The flowers are of a variety of colors ranging from white, pale yellow, and pink, to very deep orange, brown, and purplish-red, 1-2 in. long, arranged in simple racemes, 3-12 in. long. Style and ovulary cov- ered with glandular hairs. The fifth or vestigial stamen very small and 2-lobed. Sparingly escaped from gardens. June-September. Madison and Highland Counties. From Europe. 24. Linaria (Tourn.) Mifl. Toadflax. Herbs, or somewhat shrubby plants, with alternate leaves, or the lower opposite or verticillate, and flowers in terminal, bracted spikes or racemes. Corolla two-lipped, usually spurred on the lower side, its throat nearly closed by the palate. Stamens four, didynamous, included. Seeds numerous, angled or rugose. 238 The Ohio Journal of Science [Vol. XXI, No. 7, 1. Flowers yellow, %-lM in- long; spur of corolla subulate, nearly as long as the body L. linaria. 1. Flowers blue to white, J^-J 2 in. long; spur of the corolla filiform . .L. canadensis . 1. Linaria linaria (L.) Karst. Yellow Toadflax. A very common but beautiful perennial, with slender, erect, glabrous stems, sometimes glandular-pubescent above, and slightly glaucous, 1-3 ft. high. Leaves linear, pale green, entire, Yi-^Yi in. long, about )i in. wide, mostly alternate, but growing close together on the stems. Flowers pale yellow with darker spur and orange-colored palate, about V/i in. long, crowded in dense racemes. Capsule ovoid. Seeds rugose, winged. In fields and waste places. June-October. General, but no specimens from the north-western counties. From Europe. Commonly called Butter-and-Eggs. Used in medicine; flowers used as a dye. 2. Linaria canadensis (L.) Dum. Blue Toadflax. Glabrous biennial or annual, 6 vci.-lYi ft. high, the flowering stems erect or ascending, the sterile shoots spreading or procumbent, and very leafy. Leaves linear or linear-oblong, \i-\Yx in. long, 's in. wide, those on the sterile shoots usually opposite. Flowers blue or white, M^^s in. long, in long, slender racemes; spur of corolla filiform, curved, as long as the tube or longer; palate white. Seeds angled, wingless. In dry or sandy soil. May-September. Richmond Township, Huron County. 25. Chaenorrhinum (D. C.) Lange. Toadflax. Herbs, with alternate, usually entire leaves, and blue, violet, or white, axillary flowers. Corolla two-lipped, definitely spurred on the lower side, the throat not closed by the palate. Stamens four, didynamous. Capsule inequilateral, one carpel longer than the other. Seeds ovoid or cuneate, ribbed. 1. Chaenorrhinum minus (L.) Lange. Lesser Toadflax. Annual, glandular-pubescent all over, often branched, 5-13 in. tall. Leaves linear-spatulate to linear, mostly obtuse, narrowed at the base, 3^~lM in. long. Flowers blue or bluish, about ^g in. long, shorter than the pedicels; spur short and stout, much shorter than the body of the corolla. Capsule globose-ovoid. Waste grounds and ballast. August- October. Portage, Huron and Belmont Counties. From Europe. 26. Kickxia. Dum. Cancerwort. Spreading or creeping herbs, with pinnately-veined, short- petioled leaves, and solitary, white, yellow, or variegated, axillary flowers. Corolla two-lipped, spurred on the lower side, the throat closed by the palate. Stamens four, didynamous. Seeds numerous, ovoid, mostly rough or tubercled. 1. Leaves cordate or rounded at the base, ovate-orbiculai ; corolla spur curved. K. spuria \ . Leaves hastate; corolla spur straight K. elatine. May, 1921] The Figworts of Ohio 239 1. Kickxia spuria (L.) Dum. Roundleaf Cancerwort. Pubescent annual, with simple or branched, prostrate stems, 3 in.- 2 ft. long. Leaves ovate-orbicular, entire, or sometimes dentate, cordate or rounded at the base, ]4,-l in. long, with short petioles. Flowers small, solitary in the axils; peduncles very pubescent, often much longer than the leaves; corolla yellowish, the upper hp purple; spur curved; calyx-lobes ovate or cordate. Capsule subglobose, shorter than the calyx; seeds rugose, not winged. In waste places and ballast. June-September. Lake County. From Europe. 2. Kickxia elatine (L.) Dum. Sharp-pointed Cancerwort. Pubescent annual, with prostrate stems, u.sually branched, G in.- 2 ft. long. Leaves 3^-1 in. long, short-petioled, the apex acute, the base hastate or subcordate. Flowers small, solitary in the axils; peduncles glabrous, or somewhat hairy, filifomi, usually longer than the leaves; calyx-lobes lanceolate, acute; corolla yellowish, purplish beneath; spur slender, straight. Capsule subglobose, shorter than the calyx; seeds wingless. In sandy waste places. June-September. Lake County. From Europe. 27. Cymbalaria. Kenilworth-ivy. Perennial, creeping herbs, with long-petioled, palmately- veined leaves, and solitary, long-peduncled, bluish, violet, or white flowers, in the axils. Corolla two-lipped, definitely spurred on the lower side, the throat nearly or quite closed by the palate. Stamens four, didynamous, included. Seeds numerous, small. 1. Cymbalaria cymbalaria (L.) Wettst. Kenilworth-ivy. A dainty perennial, with smooth, trailing stem, 3 \n-A]/2 ft- long, often rooting at the nodes. Leaves reniform-orbicular, }4r -1 in. in diameter. Petioles usually as long or longer than the blade. Corolla pale lilac or white, streaked with purple, J^-3^ in. long, the palate yellowish. In waste places and along roadsides. June-August. Mont- gomery, Crawford, Highland, Cuyahoga, Fulton Counties. From Europe. 28. Calceolaria L. Calceolaria. Tender herbs or shrubs, mostly with opposite leaves, a four-parted calyx, and a two-parted corolla, the two lips sack- shaped. Flowers in cymes or clusters. Andrecium with two stamens and no vestiges. 1. Calceolaria crenatiflora Cav. Calceolaria. Herbaceous, soft-hairy plant, with simple leaves, and yellow, variously-spotted flowers. Corolla 2}^ in. broad, 1^ in. long. This species is apparently the main source of the more showy, herbaceous, garden varieties and hybrids. Cultivated. A REVISION OF THE RACES OF DENDROICA AUDUBONI. Harry C. Oberholser. Washington, D. C. Identification of Arizona specimens of Dendroica auduboni in the collection of the Biological Survey has made necessary this investigation into the geographical forms of the species. As these results are of some interest, it may be well to place them on permanent record. Three subspecies of Dendroica auduboni have hitherto been recognized — Dendroica audubo?ii auduboni, Dendroica auduboni nigrifrons, and Dendroica auduboni goldmani — but to this number another should apparently now be added. The following conclusions are based chiefly on the considerable series (248 specimens) in the United States National Museum, including the Biological Survey collection, and comprising types of three of the forms here recognized. The type of the fourth, Dendroica nigrifrons Brewster, has also been examined. Dendroica auduboni auduboni Townsend. S[ylvia\. Auduboni Townsend, Journ. Acad. Nat. Sci. Phila., Ser. 1, VII, pt. ii, November 21, 1S37, p. 191 ("forests of the Columbia River"). Chars. Subsp. — Size smallest, and with a minimum extent of black on the lower parts. Measurements. — Male-.^ wing, 71.5-77 (average, 74.9) mm.; tail, 55-60 (5S.2); exposed culmen, 9.5-lO.S (10.3); tarsus, 18.5-20 (18.6); middle toe with claw, 10-12.8 (11.8). Female:^ wing, 69.8-73.8 (average, 71.6) mm.; tail, 55.5-58 (56.6); exposed culmen, 10-11 (10.3); tarsus, 19-19.8 (19.3); middle toe with claw, 10.5-12.5 (11.6). Type Locality. Columbia River, near Fort Vancouver, Washington. Geographic Distribution. Central British Columbia, through the western United States, to southwestern Mexico. Breeds north to central British Columbia; west to Vancouver Island in British Columbia, western Washington, western Oregon, and western California; south to southern California; 1 Ten specimens, from Oregon,. Washington, and British Columbia. 2 Six specimens, from Washington and British Columbia. 240 May, 1921] 'A Revision of- Dendroica 241 and east to central eastern California, central Oregon, central Washington, and central British Columbia. Winters from southern British Columbia, south through California and Arizona, to the States of Guanajuato and Michoacan, Mexico. In migration it occurs casually east to Wyoming and New Mexico. Remarks. The type of Townsend's Sylvia aiiduboni was an adult male obtained by him on the Columbia River near Fort Vancouver, and is now in the United States National Museum. It proves to belong to the smallest race of the species, so that the specific name aiiduboni applies without doubt to the breeding bird of the Pacific coast region of the United States. This subspecies reaches its minimum size in Washington and British Columbia. Average wing measure- ments of adult males from different localities compare as follows: Washington and British Columbia, 74.9 mm.; Cali- fornia, 76.6 mm. Individuals from central Oregon and from California (excepting the southeastern portion of the state) are somewhat intermediate between typical Dendroica aiidubo7ii auduboni and the Rocky Mountain race, but are evidently nearer the present form. The 116 specimens examined came from the following localities : British Columbia. — Comox (May 31, 1895; June 1 and 8, 1895); Stuart Lake (June 4, 1889; eggs and nest); Wellington (May 25, 1895) ; Agassiz (Dec. 7, 1895). Arizona.— Fort Verde (Jan. 23, 1888; Dec. 30, 1887); Apache (Oct. 26 and 29, 1874) ; Cochise (May 5, 1902) ; Tucson (Jan. 29, 1884). Califorfiia. — San Francisco (Oct. 29, 1895) ; San Bernardino (Jan. 5, 1886); Red Bluff (Dec. 22, 1883; May 6, 1884; April 7 and 25, 1884; March 29, 1884); Warner Mountains (Aug. 9, 1878); Poway (Feb. 24, 1888); Nevada (October, 1872) Berryessa, Santa Clara Co. (Jan. 20, 1889; April 4, 1890) Santa Clara County (Oct. 5, 1896) ; Lassen Peak (Aug. 26, 1898) Chico (Dec. 21, 1905; Jan. 5, 1906); Burney (June 9, 1906) South Yolla Bolly Mountain (July 29, 1905); Oro Grande (March 18, 1905); San Diego (Oct. 9, 1893); Riverside (Jan. 2, 1889); Carberry's Ranch (May 20, 1894); Strange Camp, 5800 feet, San Gabriel Mountains (July 16, 1905); Preston 242 The Ohio Journal of Science [Vol. XXI, No. 7, Peak, northeastern slope of Siskiyou Mountains (Oct. 7, 1909) Cahto (May 6, 1889); Los Angeles County (Feb. 13, 1915) Camp Bidwell (July 24, 1878); Big Trees (July 7, 1878) Picard (Sept. 27, 1905); Mount Shasta (July 24, 1883; Aug. 10, 1898; Aug. 26, 27, and 31, 1883; Sept. 2, 1883); Sacramento Valley. Nevada.^Avc Dome (May 22, 1898). New Mexico. — Corner Monument No. 40, Mexican Boundary Line, 100 miles west of El Paso (May 3 and 5, 1892). Oregon. ^Beavevton (April 24, 1890); Portland (May 24 and 25, 1905); Lapush (June 12, 1897); Salem (March 25, 1888; Sept. 22, 1891); Fort Klamath (Sept. 5 and 6, 1882; Oct. 1, 9, and 13, 1882; July 16, 1882); Howard (June 10, 1915); Paulina Lake (Aug. 18, 1914); Home (June 27, 1916); 10 miles S. W. of Silver Lake (Sept. 2 and 4, 1914) ; Fremont (Aug. 23, 1914). Washington.— Kivkland (May 11 and 18, 1911); Columbia River (April 24, 1836, type; April, 1836; May 31, 1835); Fort Steilacoom (about April 1, 1856; July 28, — ) ; Steilacoom; Mt. St. Helens (Aug. 11, 1897). Wyoming.— Fort Bridger (Aug. 27, 1858). Chihuahua.— Chihuahua City (Oct. 21, 1893). Guanajuato. — Guanajuato. Jalisco.— OcotXan (Dec. 24, 1902); Tonila (October, 1865). Lower California. — La Laguna (Jan. 26, 1906); El Sauz, Sierra Laguna (Jan. 21, 1906); San Jorge (Nov. 25, 1859); Gardiner's Laguna, Salton River (April 20, 1894) ; Seven Wells (April 15, 1894); Mouth of Colorado River; La Paz (December, 1881); Mouth of Hardy River (April 2, 1905). Michoacan. —Zamora (Jan. 20, 1903) ; Mt. Tancitaro (March 3, 1903). San Luis Potosi. — San Luis Potosi (Feb. 3, "1891). Sinaloa. — Sierra de Choix, 50 miles northeast of Choix (Oct. 19, 1898). Sonora.—^onoyta (Jan. 14, 1894); Guaymas; Taronato Creek, near U. S. and Mexican boundary line (Nov. 4, 1892). Tepic. — Maria Cleofa Island, Tres Marias Islands (May 30, 1897); Tepic (December, 1865). May, 1921] A Revision of Dendroica 243 Dendroica auduboni memorabilis, subsp. nov. Chars. Subsp. — Similar to Dendroica auduboni auduboni, but larger; male with breast and jugulum nearly always more solidly and extensively black; and sides of head also witli more of blackish. Description. — Type, adult male; No. 137415, U. S. National Museum, Biological Survey Collection; Ward, Colorado, June 12, 1893; J. A. Loring. Upper parts between neutral gray and slate gray, the forehead and sides of crown thickly, the cervix sparinglv, the inter- scapulum broadly, streaked with black; upper tail-coverts black, edged with the gray of the back; center of crown gamboge yellow; rump lemon chrome;. tail brownish black, narrowly margined exteriorly with gray like that of the back or with whitish; wings rather light chaetura black, the tips of the quills dark fuscous, the secondaries edged and tipped with mouse gray, the primaries with pale gray or whitish; superior wing-coverts black, the median series broadly tipped, the greater series broadly tipped and margined externally with white, the lesser coverts margined and tipped with the gray of the back; sides of head and of neck gray like the back, the latter, together with the auricu- lars, flecked with black; a spot in front of the eye and of the suborbital region black ; upper and lower eyelids white ; throat and chin, gamboge yellow; jugukmi and sides of breast, black, a little flecked laterally with the gray of the back; a spot on each side of the breast lemon chrome; remainder of lower parts white, the sides and flanks broadly streaked with black; lining of wing white, the under wing-coverts mottled with brownish black. Measurements. — Maler^ wing, 77.3-83.5 (average, 80.5) mm.; tail, G0-G6.S ((33); exposed culmen, 9.5-11 (10.3); tarsus, 18-21 (19.0); middle toe with claw, 10.3-13.2 (11.9). Female:^ wing, 72-78 (average, 74.5) mm.; tail, 55-61 (58.1); exposed culmen, 9.5-11 (10.1); tarsus, 18-20 (19.2); middle toe with claw, 11-12.2 (11.7). Type Locality. — Ward, Boulder County, Colorado. Geographic Distribution. — Central southern Canada, western United States, Mexico, and Guatemala. Breeds north to southwestern Saskatchewan and central Alberta; west to eastern Washington, eastern Oregon, central Nevada, and southeastern California; south to southeastern California, central Arizona, southeastern Arizona, and central western Texas; and east to central western Texas, central New Mexico, central Colorado, northwestern Nebraska, eastern Wyoming, western South Dakota, and central Montana. Winters north to southern California, southern New Mexico, and south 1 Fifteen specimens, from Arizona, New Mexico, Colorado, and Montana. 2 Twelve specimens, from Arizona, New Mexico, Colorado, and Wyoming. 244 The Ohio Journal of Science [Vol. XXI, No. 7, central Texas; south to southern Mexico and Guatemala. In migration it occurs east to Iowa. Accidental in Pennsylvania and Massachusetts. Remarks. — The best character separating this new race from Dendroica auduboni audiiboni is that of size, although this is less marked in the female. There is considerable individual variation in the amount of black on the lower parts, but this is nearly always of greater extent than in Dendroica auduboni auduboni. Birds from Arizona and Montana are of the same size, and of practically the same coloration. Individuals from Nevada and Idaho are intermediate between Dendroica auduboni memorabilis and Dendroica auduboni auduboni, though they are nearer the present form. The same remarks apply also to breeding birds from the mountains in southeastern California near the State border, as, for instance. Mount Whitney. All the breeding birds from the mountains of Arizona, except those from the Huachuca Mountains, belong to the present race, but those from the southern part of the State verge a little toward Dendroica auduboni nigrifrons. Comparative wing measurements of adult males from different localities are as follows: Arizona and New Mexico, 80.3 mm.; Wyoming and Montana, 79.2 mm.; Idaho, 78.6 mm. The 104 specimens that we have examined represent the localities given below. Arizona. — Graham Mountains (April 24 and 25, 1914 May 12, 1914); Fort Verde (May 5 and 22, 1884; May 3, 1887 March 6, 1886); Pinal County (March 19, 1885; Oct. 28,1884) Nantan Plateau (May 15, 1916); Horseshoe Cienega, White River, White Mountains (July 9 and 15, 1915) ; Disaster Peak (June 14, 1915) ; Chiricahua Mountains (June 14, 1894) ; Camp 150, Coco Mengo Rancho (March 19, 1854) ; Apache (Oct. 26, 1874; Sept. 1, 1873); Mt. Graham, Graham Mountains (Sept. 21, 24, and 25, 1874; Oct. 29, 1874); San Pedro slope, Santa Catahna Mountains, Pinal County (May 6, 1885); Santa Catalina Mountains (June 10, 1906); Tucson (May 18, 1884 Nov. 18, 1890; April 13 and 19, 1884); Fort Whipple (May 5 1865; May 8, 1866; Oct. 3, 1864; April 24, 1865; April 27, 1866) Yuma (March 10, 1894); Willow Spring (July 12 and 13, 1874) Monument No. 89, Mexican Boundary Line (vSept. 24, 1892) Sawmill, at 5,600 feet, 25 miles northeast of Rice (May 10, May, 1921] A Revision of Dendroica 245 1916); Tanks, 7 miles from Strawberry, Mogollon Mountains (July 3, 1886); Marsh Lake, White Mountains (July 20, 1915); Lakeside (July 5, 1915); Huachuca Mountains' (May 26, 1903; April 8, 24, and 27, 1903; Sept. 10 and 21, 1893). California. — San Denias Canyon, Los Angeles Co. (Oct. 24, 1915); Riverside (April 16, 1887); Berryessa (Dec. 5, 1889); Cahto (May 15, 1889); Mount Shasta (Aug. 31, 1883); Oro Grande (March 22, 1905); Mendota (Oct. 2, 1907); Placerita Canyon, Los Angeles Co. (Dec. 10, 1915); Mt. Whitney (June 19, 1891; July 7, 1891); Santa Barbara (October, 1888); Laguna Station, San Diego Co. (May 5, 1874); Death Valley (Feb. 1, 1891); Fullerton (Dec. 4, 1900); Southern sierra Nevada (July 26, 1891). Co/ora^o.— Denver (May 7, 9, 10, and 17, 1873); Pueblo (Oct. 14, 1874); Estes Park (July 17, 1893). Idaho. — Swan Lake (July 5 and 7, 1911);.Lardo (July 18, 1913); Edna (June 21, 1910); Tamarack (July 6, 1913); Inkom (June 25, 1911); Idaho City (June 16, 1910). Montana. — Chief Mountain Lake (Aug. 22, 1874) ; Jefferson River (Sept. 15, 1888) ; Sioux National Forest, 8 miles east of Sykes (June 2, 1916); Madison River (Sept. 23, 1888); 5 miles southeast of Ekalaka (May 28, 1916) ; 5 miles south of Ekalaka (May 28, 1916). Nevada. — Lake Tahoe (Sept. 18, 1876) ; Toyabe Mountains (Aug. 16, 1915) ; Arc Dome, Toyabe Mountains (May 24, 1898) ; Carson City (April 4 and 18, 1868). New Mexico.— Vecos (July 20, 1883); Tres Piedras (July 31, 1904); Aug. 1, 1904); Rinconada (May 3 and 31, 1904); Fort Cummings (Oct. 15 and 25, 1873); Gila River (Oct. 11, 1873); Corner Monument, No. 40, Mexican Boundary Line, 100 miles west of El Paso (May 3 and 5, 1892) ; Zuni Mountains (June 16, 1909); Camp Grant, 6 miles east of Tucson (March 10, 1867); Dog Spring, Grant Co. (May 24, 1892) ; EHzabethtown (Sept. 17, 1903); southwestern slope of Capitan Mountains (July 10 and 13, 1903); Big Hatchet Mountains, Grant Co. (May 18, 1892). Oregon. — Strawberry Mountains (July 13, 1915). South Dakota.— Hot Springs (Oct. 17, 1892); Redfern (June 2, 1910). 1 Not breeding. 246 The Ohio Journal of Science [Vol. XXI, No. 7, Texas.— Yort Clark (April 6, 1893); Eagle Pass (Oct. 27, 1890); Henrietta (April 19, 1894); Fort Stockton (April 20, I860); Marathon (May 15, 1901). Utah.— Month of Bear River (May 23, 1915; Sept. 10, 1914). Wyoming. — Teton Pass (Sept. 15, 1910); Laramie; Horse Creek, 8,000 feet, near Merna (Aug. 12, 1911); Casper Moun- tains (Aug. 28, 1909); Jackson (May 17, 1911); 14 miles south- east of Laramie (July 15, 1915); Fort Steele (May 22, 1911); Fremont Peak (July 18, 1911); Sierra Madre Mountains (June 16, 1911); Fort Bridger (May 20, 1858); western side of Wind River Mountains (June 6, 1860) ; Bridgers Pass (May 9, 1890); Laramie Peak (May 10, 1864); Pahaska (July 30, 1910) ; Medicine Bow Mountains (June 30, 1911) ; Fossil (May 7, 1912). Chiapas.— [^o further locality] (Jan. 24, 1869). Chihuahua. — Colonia Pacheco (May 22, 24, and 31, 1909). Coahiiila. —SsXtWXo (April 17, 1902). Guanajuato. — Guanajuato. Hidalgo.— Real del Monte (May 5, 1891). Lower California. — Comondu (Nov. 9, 1905) ; Ensenada (Feb. 27, 1906); Cape San Lucas (Nov. 12, 1859). Mexico.— Salazar (Oct. 26, 1892) ; Tlalpam (Dec. 8, 1892). Michoacan. — Patamban (Jan. 29, 1903) ; Mt. Tancitaro (March 4, 1903); Los Reyes (Feb. 8, 1903). Nuevo Leo;?..— Monterey (Feb. 21 and 26, 1891). Ptiebla. — Chalchicomula (April 13, 1893). Sinai oa. — Mazatlan (December, 1867). Sonora. — Mouth of Colorado River. Yucatan. — Temax (1884). Guatemala. — [No more definite locality]. Dendroica auduboni nigrifrons Brewster. Dendroica nigrifrons Brewster, Descriptions of Supposed New Birds from Western North America and Mexico, Jan. 31, 1889. [The Auk, VI, No. 2, April, 1889], p. 94 ("Pinos Altos, Chihuahua, Mexico"). Chars. Subsp. — Similar to Dendroica auduboni memorabilis, but male with upper parts and sides of neck darker, the back with much more black, the forehead and sides of head entirely of this color; breast, iugulum, flanks, and sides of bod}' solidly black; female darker than the May, 1921] A Revision of Dciidroica 247 same sex of Dciidroica audiiboni mcniorabilis, with more black both above and below. Measurements. — Male:^ wing, 78-85. 5 (average, 81.8) mm.; tail, 58-68 (62.8); exposed culmen, 10-10.5 (10.2); tarsus, 18.5-20 (19.4); middle toe with claw, 11.3-12.8 (12). Female:- wing, 75-77.5 (average, 7().l) mm.; tail, 57-61 (58.5); exposed culmen, 9.5-10.8 (10.3); tarsus, 18.5-20 (19.3); middle toe with claw, 11.7-12.8 (12.2). Type Locality. — Pinos Altos, Chihuahua, Mexico. Geographic Distribution. — Southern Arizona and north- western Mexico. Breeds north to the Huachuca Mountains in central southern Arizona, and southeast through western Chihuahua to southern Durango. Remarks. — This race of the Audubon warbler occurs in the United States only in the Huachuca Mountains, Arizona. Birds of this species from all the other adjacent ranges are, as already explained, referable to Dendroica aiidiiboni memorabilis. Those from the Huachuca Mountains are, furthermore, in color somewhat intermediate between Dendroica auduboni nigrifroyis and Dendroica auduboni memorabilis, as would be expected from their geographic" location, but are decidedly nearer the former. Comparative average wing measurements of males are as follows: Chihuahua, 81.3; Huachuca Moun- tains, Arizona, 82.3. Of this subspecies 29 specimens have been available, from the subjoined localities. Arizona. — Huachuca Mountains (Aug. 19, 26, and 30, 1902); July 1 and 13, 1902; July 31, 1893; May 11, 20, and 26, 1903; June 21, 1902; April 5, 1903; May 9, 1902). Chihuahua. — Colonia Pacheco (June 3 and 9, 1909) ; Colonia Garcia (July 4, 1888; July 3, 9. and 24, 1899; Aug. 5, 1899); Pinos Altos (June 5, 1888, type). Durango.— Cerro Prieto (Sept. 10, 1898); El Salto (July 11, 1898). Dendroica auduboni goldmani Nelson. Dendroica goldmani Nelson, The Auk, XIV, No. 1, January, 1897, p. 66 ("Hacienda Chancol, Guatemala"). 1 Nine specimens, from the State of Chihuahua, Mexico, and from Arizona. ■- Eight specimens, from the States of Chihuahua and Durango, Mexico, and from Arizona. 248 The Ohio Journal of Science [Vol. XXI, No. 7, Chars. Subsp. — Similar to Dendroica aiidubonl nign'frons, but with still more black on the upper parts, the whole top and sides of head being solidly black (excepting, of course, the yellow crown space and the white cervical spot) ; flanks not solidly black ; and black of posterior lower parts not so extensive; a large white spot on each side of the posterior border of the yellow throat; and white occipital patch much larger. Measurements. — Male: (type) wing, S3 mm.; tail, 66. 2; exposed culmen, 10.6; tarsus, 20.7; middle toe with claw, 13. Type Locality. — Hacienda Chancol, at about 10,000 feet altitude, Huehuetenango, western Guatemala. Geographic Distribution. — Mountains of western Guate- mala. Remarks. — The only specimen of this subspecies that, so far as we are aware, has been obtained is the type. Nevertheless in this the characters are such as indicate its subspecific distinct- ness from Dendroica anduhoni nigrifrons, since the differences that characterize the latter as distinguished from De?idroica aiiduboni memorabilis are in Dendroica auduboni goldmani carried still further. Xke Oliio Journal of Science Vol. XXI JUNE, 1921 . No. 8 STEEL MOLDING SAND IN OHIO. J. A. BOWNOCKER, Ohio State University. There are two types of molding sands, (1) those for steel castings and (2) those for iron castings. The difference between the two is primarily their refractory nature. Steel castings require a sand that will not melt below 2,900° Fahrenheit, while for iron castings 2,400° is Sufficient. Ohio is a leading producer of both grades of molding sand, but this article considers only those suitable for steel purposes. The chief requisites for a molding sand are refractory nature, strength when in mold, vent or porosity and the surface left on the castings. In steel casting, however, all of these properties except that of fusibility may be regulated by the sand producer or in the steel mill, and they need not, therefore, be discussed in this paper. The minerals which compose the sandstones and con- glomerates from which steel molding sands are made are numerous, but silica or quartz in some form constitutes 95 per cent or more of the mass. Other common minerals present, are feldspar, mica, and oxide of iron. The fusion temperatures of these are about as follows: Silica 2,678° F. Orthoclase (feldspar) 2,167° Hematite 2,206°^ Magnetite 2,507°i Muscovite (mica) 2,246° Manifestly as the proportion of silica increases the fusion point of the sand rises, while as the proportion of other common minerals grows larger the fusion point is lowered. A high silica percentage is therefore the first requisite of a sand for steel molding purposes. 1 Hsmatite and magnetite have no definite melting temperatures. 249 250 The Ohio Journal of Science [Vol. XXI, No. 8, In Ohio the steel molding sands are at present all derived from rocks of Pennsylvanian or Carboniferous age. These rocks are subdivided on the basis of their age as follows: Pennsylvanian : Monongahela formation. Conemaugh formation. Allegheny formation. Potts villa formation. At present only the Pottsville and Allegheny formations yield steel molding sand in Ohio, and the Pottsville is much the larger source. KENTUCKY Map I. This shows the outcrop of the Pottsville formation and the location of the plants which produce steel molding sands in Ohio. June, 1921] Steel Molding Sand in Ohio 251 THE POTTSVILLE FORMATION. Since the Pottsville is the basal formation of the Penn- sylvanian rocks, its Hne of outcrop Hes along the junction of the Pennsylvanian and Mississippian systems. It is shown in detail on the Geologic Map of Ohio for 1920, and in a general way in Map I of this article. As there sketched it enters Ohio in Trumbull County, extends north nearly to Lake Erie, thence west of south to the Ohio River, which it reaches in Lawrence County. The Pottsville formation, which has a thickness of about 225 feet in Ohio, consists of conglomerate, sandstone, shale, coal, clay, limestone and flint. The following section shows the principal members of the formation. Homewood sandstone Tionesta or No. 3b coal Upper Mercer limestone, flint and iron ore Bedford coal Shales Upper Mercer or Webster Block or No. 3a coal Lower Mercer limestone and iron ore Lower Mercer or No. 3 coal Upper Massillon sandstone Quakertown or No. 2 coal Clay and shale Lower Massillon sandstone Sharon or No. 1 coal Sharon conglomerate THE SHARON CONGLOMERATE. The Sharon conglomerate, the basal member of the Potts- ville, is the main source of steel molding sand in Ohio. While called conglomerate, it varies greatly in its physical makeup and may be conglomerate, coarse sandstone or shale. Where the rock is conglomeratic the pebbles are of quartz which vary in size from a small fraction of an inch up to three inches or even more. The color is usually of light shades. In places pink is common. Near the top of the rock, buff is the characteristic color, while below, light gray is the usual shade. In places dark brown patches, due to iron, were noted. The pebbles are well rounded and have smooth surfaces, which suggest abrasion, due to water action. The characteristic shape is oval. The proportion and size of pebbles vary rapidly 252 The Ohio Journal of Science [Vol. XXI, No. 8, in places, both horizontally and vertically. Not infrequently they form pockets in coarse sandstone. In thickness the Sharon conglomerate also shows marked variation due largely to the unevenness of its lower surface. G. F. Lamb reports the maximum thickness of the rock in northern Ohio to be 90 feet,^ while in the southern part of the State, W. Stout records as much as 200 feet.- Thicknesses in excess of 50 feet are common in Trumbull, Geauga, Portage, Summit, Jackson, and doubtless in other counties. As a rule, the rock is poorly cemented and hence is easily crushed. Silica appears to be the chief bonding agent. The rock is massive and cross-bedding is common. The base of the Sharon conglomerate is very uneven. Before it was deposited the underlying Mississippian rocks were extensively eroded and valleys 200 feet in depth were formed in places. Lamb states that in northern Ohio the Sharon is restricted to these valleys and that they have a north-south trend.'' On such a surface the Sharon was deposited. The irregularity of this contact in the central and southern part of the State is well shown on Plate III of Bulletin 21, Geological Survey of Ohio. It forms the most striking uncon- formity in the State. In northeast Ohio, all of the Mississippian above the Cuyahoga appears to have been removed by erosion, and the Sharon, therefore, rests directly on the Cuyahoga. In the central and southern part, however, the Sharon lies in many places on the Maxville limestone, the top of the Mississippian strata.^ This indicates that before the Sharon was laid down there was more extensive erosion of the Mississippian rocks in the northern part of the State than there was in the southern part. From what has been recorded concerning the thickness and length of outcrop of the Sharon in Ohio, it is apparent that the quantity of rock is enormous, and regardless of what use may be made of it, the supply is ample for centuries. Brief descrip- tions will now be given of the principal plants, and data on the chemical and mineralogical composition of the sand. For 1 Personal letter dated August 30, 1920. 2 W. Stout, Bull. 20, Geol. Survey of Ohio, p. 42. 3 G. F. Lamb, Jour, of Geol., Vol. 19, p. 105. * W. Stout, Bull. 20, Geol. Survey of Ohio. June, 1921] Steel Molding Sand in Ohio 253 the chemical work the Geological Survey is indebted to Prof. D. J. Demorest, of the Ohio State University, and for the microscopical examination of the sands the Survey is equally indebted to D. D. Condit, formerly of the Geological Survey of Ohio and later of the United States Geological Survey. The Trumbull Stone and Sand Company. This plant is about b^'o miles west of Warren, Trumbull County, on the Baltimore and Ohio Railroad. The company owns 105 acres and in 1920 was working about 45 feet of the Sharon con- glomerate. Pebbles, the largest about 1)4. inches in diameter, were observed near the base of the quarry and were reported to be found occasionally above the base. The mass of the rock, however, is a coarse sandstone with a buff or light brown color. The rock is loaded on cars with a steam shovel, hauled to the mill and crushed and screened to three grades of sand. No. 1 is used for furnace bottoms in steel and tube mills, No. 2 for steel castings and furnace bottoms, and No. 4 for lining Bessemer converters. The principal market is Lorain, Youngs- town, Pittsburgh and Sharon. The product is unwashed and the output averages about 100 tons per day. The composition of the sand is shown below: Silica, SiOa • 95 Alumina, AI2O3 1 Ferric oxide Calcium oxide, CaO Magnesium oxide, MgO Titanium oxide, Ti02 Loss on ignition 99% 97% 35% 09% /c 62^! '>?% Microscopic examination of this sand showed the following minerals which are named in the order of their abundance: 1. Quartz. o. Feldspar. 2. Limonite. 6. Zircon. 3. Kaolinite. 7. Apatite. 4. Tourmaline. 8. Rutile. Kaolinite is thought to be the cementing material. Portage Silica Company. This company operates the largest plant in Ohio, located on the Erie Railroad in the extreme eastern part of Portage County, about midway between Gar- rettsville and Phalanx. The company owns about 1,100 acres of land, approximately one-half of which is reported to be underlaid with Sharon 254 The Ohio Journal of Science [Vol. XXI, No. 8, conglomerate. The quarry is about \]/2 miles north of the mill and has a face three-fourths of a mile long, with a maximum height of 60 feet. The rock is a mass of loosely cemented quartz pebbles, the largest measuring about two inches in diameter. The color of the rock face is buff, except near the base, where it is gray. The rock is crushed, screened and washed and in the process between 2 and 3 per cent of the material is reported to be lost. Steel molding sand is the principal product. Only one grade is made and that must pass through an eight-mesh screen. The market for this material includes Ohio, West Virginia, Pennsylvania, New York, Michigan, and Indiana. Other products of this plant are core sand, sand blast sand, filter sand and gravel, roofing gravel, traction sand and gravel for highway construction. Sand blast sand is shipped to New England, Alabama, Iowa and intermediate states. Filter sand and gravel have even a wider market. Sand blast sand is dried by artificial heat, re-screened and divided into five grades. About 200,000 tons of material are shipped per year during normal times. Three samples of sand gave the following results: Silica, Si02 Alumina, AI2O3 Ferric oxide Calcium oxide, CaO. . . . Magnesium oxide, MgO Titanium oxide, Ti02. . . Loss on ignition Steel molding sand, washed. 98.14 .16 .35 .38 .05 .08 .31 Fine-grained blast sand, washed. 98.46 .17 .23 .17 .00 .03 .28 Coarse-grained blast sand, washed. 98.04 .24 .28 .21 .01 .03 .30 A microscopical examination of steel molding sand from this plant showed the following minerals, named in order of their abundance: 1. Quartz. 2. Zircon. 3. Kaolinite. 4. Limonite. 5. Muscovite. 6. Tourmaline. 7. Microcline 8. Monazite. 9. Hematite. 10. Chlorite. 11. Apatite. June, 1921] Steel Molding Sand in Ohio 255 Geauga Silica Sand Company. This plant is located at Geauga Lake, on the Erie Railroad, in the northwest corner of Portage County. A ledge of coarse Sharon sandstone 40 feet thick is worked and the superintendent of the plant claims that 45 feet of good stone lies below the base of the present quarry. The face of the quarry has a buff color, due to the oxidation of the iron content. No pebbles were seen, but they were reported to be present in places. The rock is broken in a gyratory crusher and then reduced to sand in a disintegrator. It is neither screened nor washed and only one grade is marketed. This is for steel molding and the market extends from Cleveland to Pittsburgh. About two cars of sand are produced per day during summer and one-half as much during winter. It is proposed, however, to dis- continue crushing the rock hereafter during winter and to supply the trade during that season from the stock pile. The plant was opened in 1911 or 1912. Bedford Silica Products Co. This plant is situated in the northeast quarter of Northfield Township, Summit County, where the company owns 20 acres and is now working a ledge 33 feet thick. It is about one mile east of the Pennsylvania Railroad with which it has switch connection. The Sharon conglomerate is covered with about 4 feet of mantle rock which is removed by a wheel scraper drawn by a tractor. The rock has the usual buff color, but in places has small black spots, probably due to iron. Pebbles occur near the base of the quarry and the largest measure about 13/^ inches in diameter. A gasoline well-drilling outfit prepared the rock for shooting. The rock is loaded on cars by a steam shovel, drawn to the mill by horse power and elevated by cable. It is broken in a gyratory crusher, screened and the coarse material passed through a disintegrator. The sand is not washed. The product finds a market for steel castings and the coarse material for furnace bottoms. Northeast Ohio and adjacent parts of West Virginia and Pennsylvania provide a market. The company also makes a specialty of sand for plaster, which finds a ready sale at Cleveland. For this purpose a drying plant is now being constructed. About 20 men are employed when the plant is operating to capacity and the output is from 175 to 200 tons of sand per day. Work continues throughout the year, except in the 256 The Ohio Journal of Science [Vol. XXI, No. 8, coldest weather. This quarry is reported to have been opened about 30 years ago and to have supplied much heavy stone for a breakwater at Cleveland. The sand plant was erected in 1909 and has been in possession of its present owners for approxi- mately 10 years. Following are two analyses of the sand, the first from a bin sample and the second. from chips from the lower part of the quarry : Silica, Si02 Alumina, AI2O3 Ferric oxide Calcium oxide, CaO. . . . Magnesium oxide, MgO. Titanium oxide, Ti02. . . Loss on ignition Chips from Sand from bin. lower part of the ( i^uarry. 98.00 98 .29 .36 .69 .53 .18 01 .00 .08 .00 .00 .04 .34 .28 Microscopic examination of the first sample showed the following minerals, named in order of their abundance : 1. 2. 3! 4. 5. Quartz. Limonite. Kaolinite. Feldspar. Muscovite. 6. Zircon. 7. Sericite. 8. Apatite. 9. Titanite. Summit Silica Company. This is located just south of Barberton, Summit County. A ledge of Sharon conglomerate, 45 feet high, is the basis of the industry. In places the pebbles make up the mass of the rock, while elsewhere they may be restricted to the upper part and in other places to the lower part of the quarry. Most of the pebbles are less than one inch in diameter, but one measuring 3 inches was found. Practically all colors occur, but light shades prevail. Large cracks filled with clay were observed and some of these extended to the base of the quarry. Moreover, chunks of clay were noted in places in the rock. After blasting, the rock is loaded on cars with a steam shovel and hauled to the mill with mules, where it is reduced to sand by a gyratory crusher, disintegrator, and rolls. The material is then washed and that for sand blasting passed through a cylinder drier, after which it is screened. The coarse material from the screens is put through the rolls, after June, 1921] Steel Molding Sand in Ohio 257 which it is re-screened. The sand is used for steel castings, blasting sand, glass making, and for concrete. The pebbles are the source of sand for blasting, while the sand proper yields molding and glass sands. Three grades of molding sand are made and five grades of blasting sand. The market extends from Pittsburgh to Chicago and Milwaukee. A sample of unwashed and unscreened sand from this plant had the following composition : Silica, SiOa. 97.41 Alumina, AI2O3 58 Ferric oxide 31 Cilcium oxide, CaO • . 11 Magnesium oxide. MgO 00 Titanium oxide, Ti02 09 Sodium oxide, Na20. . 04 Potassium oxide, KoO 08 Loss on ignition 44 Microscopical examination showed the following minerals present. These are named in order of their abundance : 1. Quartz. 5. Feldspar. 2. Magnetite. 6. Muscovite. 3. Zircon. 7. Apatite. 4. Kaolinite. Franklin Industrial Company. This plant is located on the Pennsylvania Railroad about one mile west of Warwick, Wayne County. A ledge of 50 feet of Sharon conglomerate is the basis of the industry. The rock is reduced in a jaw crusher and then passed through three sets of rolls, after which it is screened. Part of the material is put through a rotary drier. The product is used for steel castings, furnace bottoms, and by traction lines. The output is about 100 tons per day and the plant operates throughout the year. A sample of unwashed and undried sand from this plant gave the following analysis: Silica, SiOa 97.47% Alumina, AI2O3 72 Ferric oxide 38 Calcium oxide, CaO 00 Magnesium oxide, MgO 06 Titanium oxide, Ti02 09 Loss on ignition 60 Microscopical examination showed the following minerals, which are listed in order of their abundance : 1. Quartz. 6. Microcline. 2. Tourmaline. 7. Sericite. 3. Zircon. 8. Hematite. 4. Limonite. 9. Rutile. 5. Kaolinite. 10. Zenotime. 258 The Ohio Journal of Science [Vol. XXI, No. 8, Oliver Silica Sand Company. This plant is located about one-fourth of a mile east of that of the Franklin Industrial Company. About 40 feet of the Sharon is worked. The rock is treated in much the same manner as in that of the adjacent plant and the product, of course, is similar. Its principal use is for steel castings, furnace bottoms and cores. Chalfants Plant of the Central Silica Company. This plant is located between Chalfants and Glenford, on the Baltimore and Ohio Railroad, in the northern part of Perry County. The rock is the Sharon and it there shows more variation than was found in the quarries of northeast Ohio. The old quarry just south of the mill has been abandoned, except for the ganister, which ranges from 18 inches to 12 feet in thickness and which lies at the top of the Sharon. The ganister is finer grained than the rock below, is well cemented and has a color which ranges from light gray to brown. The new quarry which is now the source of rock, except for ganister, is situated nearly a half mile south of the mill. A maximum of perhaps 35 feet of rock, exclusive of 4 to 6 feet of stripping, is worked. The rock is coarse-grained and in places pebbly, but these lie in pockets rather than in beds. At present the stripping is run through the mill, but the company is now removing this waste by a drag line system. The rock is loaded on cars with a steam shovel and transported to the mill by a dinky engine. The rock is broken in an oscillating crusher and is carried by gravity to a dry pan which is operated wet. This reduces the rock to sand, which is transported by gravity to a 6-mesh rotary screen. Any coarse material which cannot pass through the screen is carried back to the dry pan. From the screen the sand runs into a sluice box and is pumped from there to the washer. Here the coarse sand settles to the bottom and is removed, while the find sand passes over the top with water into a settling tank, where the sand collects on the bottom and the clay is carried with water over the top and flows into the creek. This sand is used for steel molding purposes. The coarse sand (6-mesh) referred to in the last paragraph is conveyed by gravity to a pile outside of the mill and in that form is used for furnace bottoms and in brick making. Much of the greater part of this pile, however, is transported by a drag to an elevator, which lifts it to the top of the drier, through. June, 1921] Steel Moldiu<^ Sand iu Ohio 259 which the sand falls, a distance of about 35 feet. From the drier the sand is conveyed by an elevator to a screen, where it is separated into three sizes. The finest passes through a 20 mesh, the medium through a 16 mesh, while the coarsest passes from the end of the screen. The 20 mesh, or finest sand^ is raised by an elevator to a bin above the tube mill into which it runs. This mill measures 20 by 5 feet and has a "silica" lining. It is filled half full with flint pebbles and rotated. The sand remains in this mill about 45 minutes and is from 140 to 200 mesh fineness when it emerges, the difference being dependent on quanity of sand in the mill. The main use of this material, known as Silica Wash, is for painting molds for steel castings; minor uses are for soap, paint, and in rubber works. The market for this product at present extends from Ohio to California. The medium sand (16 mesh) is used in glass. making and on tracks to prevent slipping. The market for this is restricted to Ohio. The coarsest material, or that which does not pass through the screen, is used as steel blast sand. The sand which collects in the settling tank is used for steel moldings. The ganister is reduced in the dry pan without water and is shipped in that condition. The plant reduces about 175 tons of rock per day, except in very cold weather, and the product is about as follows : Silica Wash 15 tons. Glass sand 80 " Steel blasting 5 " Steel molding 23^" Furnace bottoms 47 " Waste 25 " When the stripping is removed in the quarry the quantity of waste will be very greatly reduced. Four samples of sand from this plant were analyzed and the results follow: Blast sand from pebble, washed. Ganister sand, unwashed. Steel mold- ing sand, washed. Glass sand, washed. Silica, Si02 Alumina, AI2O3 Ferric oxide Calcium oxide, CaO Magnesium oxide, MgO Titanium oxide, Ti02 Loss on ignition 97.99 1.08 .22 .00 .07 .08 .53 98.61 .39 .20 .00 .06 .20 .27 95.66 1.84 .19 .00 ,16 22 '87 99.43 .18 .13 .00 .05 .04 .22 260 The Ohio Journal of Science [Vol. XXI, No. 8, Microscopic examination of the steel molding sand showed the following minerals which are listed in order of their abundance : 1. 2. 3^ 4. 5. 6. Quartz. Kaolinite. Zircon. Limonite. Hematite. Tourmaline. 7. Magnetite 8. Titanite. 9. Muscovite 10. Feldspar. 11. Sericite. 12. Apatite. Examination of three additional varieties of sand from this plant disclosed quite a variation in the relative abundance of the minerals, though of course quartz in all was by far the most plentiful. Jackson Sand Mining Company. This, the southernmost steel molding sand plant in Ohio, is located on the Cincinnati, Hamilton & Dayton Ry., 2 miles north of Coalton, Jackson County. The rock is the Sharon, but it is a sandstone, rather than a conglomerate. The sandstone measures about 70 feet in thickness and forms two benches, separated by a thin layer of clay. The lower bench is the coarser grained. The rock is reduced by a hammer pulverizer and is screened but not washed. Two grades of sand are marketed, the white and the light yellow. The product is used as molding sand for steel and iron castings, furnace bottoms and brick kiln sand. This plant has been in continuous operation for about 25 years. Two samples of sand and one sample of uncrushed rock were analyzed with the following results: Silica, SiOa Alumina, AI2O3 Ferric oxide Calcium oxide, CaO. . . . Magnesium oxide, MgO Titanium oxide, Ti02. . . Loss on ignition Sand from top ledge. 96.79 2.00 .20 .00 .08 .17 .55 Sand from bottom ledge. 98.50 .70 .22 .10 .00 .10 .35 Uncrushed rock. 96.19 2 22 ^20 .00 .03 .18 .65 A sample of uncrushed rock was examined microscopically and the following minerals identified. They are listed in order •of their abundance. 1. Quartz. 2. Zircon. 3. Muscovite. 4. Kaolin. 5. Microcline. 6. Limonite. 7. Hematite. 8. Rutile. 9. Xenotime. June, 1921] Steel Molding Sand in Ohio 261 THE MASSILLON SANDSTONE. As the section of the Pottsville formation given on a pre- ceding page shows, the ]\lassillon sandstone is divided into two parts by the Quakertown or No. 2 coal. At Massihon, the type locality, the Upper Massillon sandstone, GO feet thick, is well shown in the Everhard quarry on the west bank of the Tuscarawas River, while at Pauls, about 4 miles farther up stream, the Lower Massihon sandstone was formerly quarried. The following section in the Everhard quarry shows very well the rock succession: Ft. In. Upper Massillon — Stripping, rejected 12 0 Sandstone, much broken, iineven bedded, coarse- grained, not as good as sandstone below with which it is mixed in proportion of 1 to 2 25 0 Clay shale, rejected 10 0 Sandstone, massive, used for silica sand 35 0 Shale, part siliceous 33 0 Anthony coal, bony 0 2 Sciotoville clay, siliceous, light-colored 8 6 Shale, dark-colored 3 0 In Holmes County, southwest of Massihon, the two sand- stones are well developed. The lower division is represented partly by shales, but the upper division appears to consist more largely of sandstone.^ The rock is well developed in Muskingum County where in places it is slightly con-' glomeratic." Along the western border of the Hocking Valley coal field, Orton reports both members of the Massillon sand- stone present, the upper in places very pure and from 10 to 20 feet thick, while the lower is "often heavy. "•'^ In Scioto County in the extreme southern part of Ohio, thick sandstones on the horizon of the Massillon are shown in numerous sections.* From what has just been said, it appears that the Massillon sandstone extends across the State much as does the Sharon conglomerate, though the latter by reason of its texture is much the more conspicuous. The Massillon sandstone is used in a large way at Massillon for steel molding sand and several other purposes. It was formerly quarried for building stone and many structures of it may be seen in Massillon and near-by cities. 1 A. A. Wright, Geol. Survey of Ohio. Vol. 5, p. 818. 2 W. Stout, Geol. Survey of Ohio, Bull. 21, p. 60. 3 Edward Orton, Geol. Survev of Ohio, Vol. 5, p. 991. ^ W. Stout, Geol. Survey of Ohio, Bull. 20. 262 The Ohio Journal of Science [Vol. XXI, No. 8, The Everhard Company. In the Everhard quarry, which was opened in 1884, a maximum of 60 feet of sandstone is quarried. The rock is broken by two jaw crushers and is further reduced in rolls. It is then passed through a rotary drier and finally over screens. The coarse material caught by the screens is returned to the finishing rolls and is again screened. During the milling, dust is withdrawn by high speed fans and the product marketed. The capacity of the plant is about 200 tons per day. The products are steel molding sand, furnace sand, core sand, and formerly glass sand. The molding sand is passed through a 5 mesh screen, while the furnace sand is passed through a 2 mesh. The molding sand is well adapted for large steel castings and has a market at Pittsburgh, Cleveland, and Cincinnati, as well as at intermediate places. Massillon is the chief market for core sand. In recent years but little sand has been marketed for glass making and the rock must be carefully sorted for this purpose. Four samples from this plant were analyzed and the results follow : Silica, Si02 Alumina, AI2O3 Ferric oxide Calcium oxide, CaO Magnesium oxide, MgO Titanium oxide, Ti02. . Sodium oxide, Na20. . . Potassium oxide, K2O. . Loss on ignition Furnace bottom sand. 96.29 1.63 .33 .03 .00 .20 .43 Core sand. 95 1 . 10 .60 .81 .00 .02 .15 .12 .26 .68 Dust from mill. 8-1.02 8.78 2.09 .03 .02 .72 2.65 Selected chips from quarry face. 96.51 1.90 .58 .07 .01 .12 .56 Carefully selected chips from the quarry face, when examin- ed with a microscope, showed the following minerals present. They are named in order of their abundance. 1. 2 3; 4. 5. Quartz. Limonite. Kaolinite. Feldspars. Muscovite. 6. Hematite. 7. Sericite. 8. Zircon. 9. Magnetite. June, 1921] Steel Molding Sand in Ohio 263 Newman Silica Sand Company. This plant is located on the Baltimore and Ohio and Pennsylvania railroads, about one- half mile above Pauls, Stark County. The rock is broken in a jaw crusher and then passed through two sets of rolls, after which it is washed and some of it screened. It is then ready for shipment. The sand is used for steel molding and furnace bottoms. It was formerly used in glass manufacture. The capacity of the plant is about 100 tons per day. A ledge of 30 feet of sandstone is worked, above which is 4 to 6 feet of stripping. The sandstone is coarse, has a buff color and parts are impregnated with iron. A dark shale lies below the sandstone and still lower the Massillon or No. 1 coal is reported. The rock is loaded by hand and hauled to the mill by gravity. For many years the Lower Massillon sandstone was worked at Pauls for foundry facings, furnace bottoms, and to a small extent for glass, but the plant was destroyed by fire a few years ago and it has not been rebuilt. THE DUNDEE SANDSTONE. The Dundee sandstone lies near the middle of the Pottsville formation. It cannot be said to be steady or persistent, however, for its place is frequently occupied by shales, but it is well developed locally and in a few places is of value. The one locality where the Dundee sandstone is worked in the large way is the valley of Sugar Creek in the northwest corner of Tuscarawas County. The stream has there cut a deep trench in the sandstone and has thus made it readily available. The plant at Barrs Mills, operated by the Massillon Sand Stone Company, is the best equipped and the largest producer in the valley. The sandstone worked averages about 50 feet in thickness and has a maximum of about 67 feet. Most of the rock has a buff color, but the lower part is in places light gray. It is coarse grained, but without pebbles. The rock is transported to the mill by horse power and elevated by cable. It is broken in a gyratory crusher and reduced to sand in a disintegrator. Most of the sand is dried and screened, the coarser material then passing through a set of rolls and being re-screened. Dust is blown from the sand 264 The Ohio Journal of Science [Vol. XXI, No. 8„ while it is in the drier and is used for small steel cores, glazing tile, and plaster. The product of this plant is used principally for steel castings and furnace bottoms. Minor uses are on traction lines, in brick making, and to a smaller extent in the manufacture of glass. Pittsburgh is the principal market. The plant produces 200 tons of sand per day on the average. It is said to have been in operation about 30 years as a sand plant and prior to that time the quarry was worked for dimension stone. The composition of the sand from this and adjacent plants in Sugar Creek Valley is shown by the following analyses: Silica, Si02 Alumina, AI2O3 Ferric oxide Calcium oxide, CaO. . . . Magnesium oxide, MgO Titanium oxide, Ti02. . , Loss on ignition Massillon Sand & Stone Co. Unwashed, Barrs Mills 98.28 .32 .24 .00 .13 .07 .32 National Malleable Castings Co. Unwashed, Dundee 96.63 2.00 .42 .00 .04 .20 .60 Beach City Silica Sand Co. Unwashed. 97.76 .69 .34 .00 .11 .05 .40 Microscopical examination of a run-of-quarry sample of sand from the plant of the Massillon Stone Company at Barrs Mills showed the presence of the following minerals which are listed in order of their abundance: 1. Quartz. 2. Microcline. 3. Feldspars. 4. Limonite. 5. Kaolinite. 6. Tourmaline. 7. Chlorite. 8. Zircon. 9. Titanite. 10. Serpentine. Dundee, which is situated about 3 miles north of Barrs Mills, has long been an important sand producer. Here are located the plant of the National Malleable Castings Company and that of the American Sand Company. Both work the Dundee sandstone, the rock being broken in gyratory crushers and then passed through rolls and over screens. Steel castings and furnace bottoms are the principal uses of the product. June, 1921] Steel Molding Sand in Ohio 265 The plant of the Beach City SiHca Sand Co., which is located about midway between Dundee and Beach City, was opened in 1910. A ledge of about 50 feet of the Dundee sand- stone is quarried and the rock hauled to the mill by gravity. The rock is broken in a jaw crusher and then run through three sets of rolls, after which it is screened, and where desired, washed and dried. Very fine sand is produced by grinding the material in a rotating steel cylinder which contains rounded flint pebbles. The market is chiefly for steel castings, furnace bottoms, cores, potteries, brick yards, and for glass making. From 4 to 5 cars a day are shipped on the average. THE ALLEGHENY FORMATION. The Allegheny formation contains extremely valuable beds of coal and clay and less valuable deposits of limestone, iron ore, and sandstone. Only one sandstone of this formation is now quarried for steel molding sand and that at a single locality — Strasburg, Tuscarawas County. The territory is notable for its resources, as it contains a good bed of the Lower Kittanning or No. 5 coal and below it the most refractory clay of Ohio. Following is a composite section: Ft. In. Lower Kittanning, No. 5 coal ,3 3 Plastic fire clay 3 6 Flint fire clay 3 0 Plastic fire clay -1 6 Sandstone, measured! 58 0 The White Rock Silica Sand Co. This is the one plant which produces steel molding sand from the Allegheny formation. The sandstone which is shown in the above section measures, where quarried, 58 feet, but the top 8 feet are shaly. The rock is coarse grained, massive, gray to buff in color, and. carries in places black coaly material, which is sorted out by the workmen. A gasoline engine hauls the rock to the mill, where it is broken in a jaw crusher and then reduced to sand by two sets of rolls. The sand is screened for steel castings, cores, and brick yards. Furnace sand is not screened. The plant does not have a washer. A great quantity of raw material is available. 266 The Ohio Journal of Science [Vol. XXI, No. 8, The following analyses show the composition of the sand: Silica, SiOo Alumina, AI2O3 Ferric oxide Calcium oxide, CaO. . . . Magnesium oxide, MgO Titanium oxide, Ti02. . . Loss on ignition Furnace Steel bottom sand. molding sand. Unwashed. Unwashed. 96.87 . 96.52 1.20 1.83 .45 .31 .00 .00 .06 .03 .11 .17 .43 .39 Microscopical examination of the steel molding sand showed the following minerals which are named in the order of their abundance : 1. Quartz. 2. Kaolinite. 3. Limonite. 4. Tourmaline 5. Microcline. 6. Muscovite. 7. Zircon. 8. Magnetite 9. Hematite. 10. Epidote. 11. Xenotime. 12. Rutile. THE COMPARATIVE RESISTANCE OF DIFFERENT SPECIES OF EUGLENIDAE TO CITRIC ACID.* W. J. KOSTIR, Ohio State University. In the course of his experiments on the nutrition of Euglena gracilis, Zumstein (1900) discovered that this species is able to tolerate surprisingly high concentrations of certain organic acids. He made use of this fact to obtain cultures free, or nearly free, from bacteria. Adding various proportions of the acid to the culture medium employed, he found that while the multi- plication of bacteria was effectively inhibited, rich cultures of apparently normal Euglenae could be obtained. Ternetz (1912) working with the same species, also made use of this method. Of several organic acids experimented with, Zumstein found citric acid to be the least harmful to this organism. He reports that solutions of 0.5 to 2% are "not injurious," and that cul- tures to which these percentages of the acid had been added were quite successful. Whether all the individuals used in inoculating these cultures survived the transfer to the acid medium is unfortunately not clear from Zumstein's account. In the case of "3 and 4%," he states that many individuals remained living even after 88 hours. In "4 and 5%" many were actively moving after five days. In "5 and 6%" a few remained alive even after 17 days. Higher concentrations were apparently found to be uniformly fatal, though Zumstein makes no definite statement as to the lowest percentage sufficient to kill all the individuals exposed to it. Unfortunately, the results of these experiments ' have been interpreted by the writers of certain text-books and manuals in a way not at all justified by the facts. Results obtained on a single species have been made the basis for generalizations concerning the whole genus Euglena, and even the family Euglenidas. Prowazek, in his "Einfuehrung in die Physiologic der Einzelligen" (1910), Lemmermann, in the section on the Euglenidai in "Die Suesswasserflora Deutschlands, Oester- reichs, und der Schweiz" (1913), and Doflein, in the fourth * Contribution Xo. 66 from the Department of Zoology and Entomology, Ohio State University. 267 268 The Ohio Journal of Science [Vol. XXI, No. 8, edition of his text-book (1916) all either definitely state or imply that a high degree of resistance to organic acids is char- acteristic of Euglense (or even of Euglenidas) in general. The present writer became interested in this matter when, having successfully cultivated Euglena gracilis in strongly acid media, he attempted to use similar methods with Euglena deses. These attempts met with failure in every case, and even rel- atively weak solutions of citric acid proved fatal to this species. How was this to be interpreted? Is the case of Euglena deses merely a striking exception to the general rule for the group? Or is Euglena gracilis the exceptional form? Or, lastly, may it be that no general rule for the group is justified? Good cultures of several species of Euglena and of one species of the closely related genus Phacus were at hand at the time, and it occurred to the writer that a comparatively simple series of experiments w4th them might throw light on these questions. Such a series of experiments was actually planned and carried out, as described in the following paragraphs. A quantity of clear surface water was procured from a near-by pond (known to be a habitat of several species of Euglena), and filtered. Tests showed it to be decidedly alkaline in reaction. By careful titration, using neutral red as an indi- cator, it was found that 1 c. c. of 1% HCl (chemically pure) was needed to bring 190 c. c. of this water to the point of neutrality corresponding to that of distilled water. Accordingly, the original quantity of filtered water was now divided into two parts, one of which was left unchanged, while the other was made neutral by the addition of the appropriate quantity of 1^ HCl. With this neutralized pond water the following solutions of citric acid were then made up (chemically pure citric acid being employed): .025%, .05%, .1%, .25%, .5%, 1%, 2%, 3%, 4%, 5%.* As seven different species were available for the experiment, seven series of these solutions were now put into Syracuse watch *It may be noted that the hydrogen ion concentration of these solutions is by no means proportional to the concentration of the acid, since the degree of ioniza- tion decreases as the concentration of the acid becomes greater. But the work of Collett (Jour. Exp. Zool., 1919) and others has shown that other factors besides hydrogen ion concentration are involved in the toxicity of organic acids; hence, for the purposes of these experiments, known concentrations of the acid are pref- erable to known concentrations of hydrogen ions. Per cent solutions were used rather than molar solutions in order to make the results directly comparable to those of Zumstein. June. 1921] Resistance of Eiiglenidce to Citric Acid 269 glasses, 5 c. c. of liquid in each watch glass. To each series were added, in similar glasses, both neutralized and unneutralized pond water without citric acid, these to serve as controls; and each series was then inoculated with a dififerent species, about five or six individuals being introduced into each watch glass. The quantity of culture medium introduced with these was so slight as to be negligible. As the watch glasses were kept stacked (close to a north window and away from the direct rays of the sun) except when actually being handled, it is evident that evaporation was also a negligible factor. vSince the object was purely to get a measure of the com- parative resistance to the acid, no attempt was made to deter- mine the length of time that the organisms could live in a given concentration of the acid. Instead, all the watch glass cultures were very carefully examined just 24 hours after inoculation, and the condition of every individual was noted. In every case all the organisms in the controls (both neutralized and unneutralized pond water) were found to be alive and normal in every way, showing that the results observed in the other solutions were actually due to the addition of the citric acid, and not to anything in the original pond water, nor to the HCl introduced in neutralizing it. The results from the citric acid solutions are concisely expressed in the accompanying table. In each case where all the individuals in the watch glass were still alive after 24 hours, the proper space has been left blank. Where part of the indi- viduals were dead, an asterisk (*) is used; and where all were found dead, a large X. In the case of each species, therefore, the column with the last blank space (counting from left to right) indicates the highest percentage of the acid successfully withstood by one hundred per cent of the individuals tested; while the column with the first X indicates the lowest per- centage causing the death of one hundred per cent. The criterion used in determining whether an individual was dead or alive was the unmistakable change in color and appearance which ensues shortly after death has occurred. Movement can not be used as a criterion in this group, since individuals may temporarily remain perfectly rnotionless for a considerable length of time. 270 The Ohio Journal of Science [Vol. XXI, No. 8, It will readily be seen from the table that the tolerance of different individuals of the same species is not necessarily the same. Some of the Euglena deses, for example, succumbed to the .05%, while the rest were killed by the .1%. In Euglena gracilis the variability in this regard is very great, a fact shown by Zumstein's results also. Similar physiological variability among the individuals of a species and even in the same culture is a phenomenon which the writer has repeatedly observed in connection with factors other than acidity. Its significance remains to be shown. TABLE I. Percentages of Citric Acid Causing Death Within 24 Hours. * — Some, but not all, dead. X— All dead. .025% .05% .1% .25% .5% 1% 2% 3% 4% 5% E. deses Ehrenb. * X X X X X X X X E. acus Ehrenb. X X X X X X X X E. geniculata Duj. (?) X X X X X X X X E. ehrenbergii Klebs * X X X X X X X E. oxyuris Schmarda X X X X X X X Ph. anacoelus Stokes X X X X X X X E. gracilis Klebs * * * X The most striking fact brought out by a perusal of the table is the remarkable difference between Euglena gracilis and all the other species. Thus the species first studied in this connection, and assumed by various writers to be typical of the other species of its group, is seen to be decidedly exceptional when compared with the six other forms that have been tested. One discrepancy between these results and those of Zum- stein may be noted. All the individuals of Euglena gracilis in the 5% solution were found dead; while Zumstein states that a few remained alive in 5 and 6%. In view of the great variabil- ity which we have noted in this species, and the relatively small June, 1921] Resistance of EtiglenidcE to Citric Acid 271 numbers of individuals tested, this difference is undoubtedly without significance. Further study on the physiology of these interesting organ- isms, to include, it is hoped, a more detailed investigation of their behavior toward acids, is in progress. LITERATURE CITED. Doflein, F., 1910. Lehrbuch der Protozoenkunde. 4te Aufl. Jena. Lemmerman, E., 1913. Eugleninae. In: Die Suesswasserflora Deut.schlands, Oesterreichs, und der Schweiz. Heft 2. Jena. Prowazek, S. von, 1910. Einfuehrung in die Physiologie der Einzelligen. Leipzig und Berlin (Teubner). Ternetz, Ch., 1912. Beitraege zur Morphologie und Physiologie der Eiiglena gracilis Klebs. Jahrb. f. wiss. Bot., Bd. LI. Zumstein, H., 1900. Zur Morphologie und Physiologie der Euglena gracilis Klebs. Jahrb. f. wiss. Bot., Bd. XXXIV. NEW FORMS OF OEDOGONIUM* L. H. Tiffany, Ohio State University. Some algal collections made from lakes in the vicinity of Akron, Ohio, in the summer of 1920 contained a species and a variety of Oedogonium apparently undescribed. In accordance with the rules of the Vienna Congress a Latin and an English diagnosis of each is given. Oedogonium exocostatum nov. sp. Oedogonium dioicum; macrandrium; oogoniis singulis vel binis, ellipsoideis vel globoso-ellipsoideis, rarius terminalis; poro superiore apertis; oosporis eadem forma ac oogoniis, hasc plane complentibus, membrana duplici ; episporio longitudinaliter costato (in sectione optica transversal! undulato), costis integris, raro anastomosantibus, in medio oosporae circa 13-15, endosporio Isevi; cellulis suffultoriis tumidis; plantis masculis paululo gracilioribus quam femineis; antheridiis 3-7 cellularibus ; spermatozoidis binis, divisione horizontalis ; cellula fili basali forma, ut vulgo elongata; Crassit. cell. veg. plant, fern. (13-) 18-25 ^ altit. 72-140 m Crassit. cell. veg. plant, masc. (13-) 16-20 fx altit. 48-100 fj. Cras.sit. cell. suff. 22-30 u altit. 60-90. m Crassit. oog. 40-52 /u altit. 68-96 m Crassit. oos. 38-41 m altit. 60-68 ^ Crassit. cell, antherid. ] 2-16 ju altit. 7-12 ^ Dioecious, macrandrous, oogonia single or often two, ellipsoid to ellipsoid-globose, occasionally terminal, pore superior; oospores of the same form as the oogonia which they very nearly completely fill, mem- brane double; the outer spore wall marked by 13-15 longitudinal ribs, inner wall smooth ; suffultory cells swollen ; male filaments a little more slender than the female, antheridia 3-7 celled, sperms two, division horizontal; basal cells elongate. Diam. veg. cells, female plant (13-) 18-25 m length 72-140 fx Diam. veg. cells, male plant (13-) 16-20 ^ length 48-100 m Diam. suffultory cells 22-30 n length 60-90 m Diam. oogonia 40-52 m length 68-96 m Diam. oospores 38-41 ^ length 60-68 n Diam. antheridial cells 12-16// length 7-12 n * Papers from the Department of Botan^', Ohio State University, No. 128. 272 June, 1921] Neiv Forms of Ocdogonium 273 In appearance this species is near Oedogonium cyathigeriim Wittr. It differs, however, in being macrandrous, in having a smaller number of ribs, and in having the ribs on the outer wall instead of the inner wall of the oospore. It is easily distin- guished from Oedogonium crenidato-costatiim Wittr. by its swollen suffultory cells, its spore markings, and its larger dimensions throughout. Among the operculate species it bears some resemblance to Oedogonium pauco-costatum Transeau. Found rather abundant in Dollar Lake and Summit Lake, near Akron, Ohio, August, 1920. Type in L. H. T. collections, Nos. 196, 198. Plate I, Fig. A-F. Oedogonium p^uco-costatum Transeau var. gracilis nov. var. Var. omnibus partibus gracilior; oosporis ellipsoideis vel globoso- ellipsoideis, oogonia fere complentibus vel non complentibus ; ceterum ut in typo; Crassit. cell. veg. Crassit. oog. Crassit. oos. Crassit. cell, antherid. 15-20 M altit. 66-120 At 48-52 u. altit. 70-88 m 44-48 M altit. 60-70 m 17-20 M altit. 8-12 m Somewhat smaller than the species; the oospore is ellipsoid or occasionally globose-ellipsoid, completely filling the oogonia or not filling the oogonia; otherwise similar to the type; Diam. veg. cells Diam. oogonia Diam. oospore Diam. antheridial cells 15-20 M length 66-120 n 48-52 M length 70-88 fj. 44-48 M length 60-70 fj. 17-20 M length 8-12 fx This variety bears some resemblance to Oedogonium Aus- traliamim Hirn. It differs, however, in having largely ellip- tical oospores, in the similarity of the vegetative cells of the male and female filaments, and in the smaller number of ribs on the median spore walls. It seems most closely related to Oedogonium pauco-costatum Transeau in its general habit of growth. It was found associated with Oedogonium taphro- sporum Nordst. and Hirn. Collected at Turkeyfoot Lake near Akron, Ohio, August, 1920. Type in L. H. T. collections Nos. 177, 182. 274 The Ohio Journal of Science [Vol. XXI, No. 8, EXPLANATION OF PLATE. Figures A-F Oedogonium exocostatum nov. sp. A — Female filament with two oogonia and two oospores, one of the latter immature. B — Elongated basal cell, showing habit of attachment. C — Female filament, showing terminal oogonium with mature oospore. D — Female filament with oogonium, oospore mature. E — Male filament with antheridia containing sperms. F — Portion of vegetative filament showing variability in cell diameter and elongation. Camera lucida drawings. New Forms of Oedogonium L. H. Tiflfany Plate I. FUNDS FOR SCIENTIFIC RESEARCH. The Research Information Service of the National Research Council has recently compiled information about funds for scientific research. From this compilation it appears that there are hundreds of special funds, trusts, or foundations for the encouragement or support of research, in the mathematical, physical and biological sciences, and their applications in engineering, medicine, agriculture and other useful arts. The income from these funds, which amounts annually to at least fifty million dollars, is used principally for prizes, medals, research scholarships and fellowships, grants and sustaining appropriations or endowments. So numerous have been the requests to the Research Council for information about sources of research funds, availability of support for specific projects and mode of administration of particular trusts or foundations, that the Research Information Service has created a special file which it is proposed to keep up to date in order to answer the questions of those interested in such funds. . Furthermore, in order to give wider publicity to the immediately available information about research funds, the Council has issued a bulletin under the title ''Funds avail- able in 1920 m the United States of America for the encouragement of scientific research.'' Inquiries concerning the bulletin or for information about research funds should be addressed. National Research Council, Information Service, 1701 Massachusetts Avenue, Washing- ton, D. C. 276 SUBJECT INDEX. Academy of Science, Annual Meeting, Ohio, 1. Additions to the Catalog of Ohio Vas- cular Plants for 1920, 128. Adirondack's, A New Ambrosia Beetle from the, 201. Alg:al Food of the Young Gizzard Shad, 113. Ambrosia Beetle, A New, 201. A New Ambrosia Beetle, 201. Annual Meeting of the Ohio Academy of Science: 1. Apple Seedlings, Carbohydrate Ma- terial in, 89. A Preliminary -General Survey of the Macro-Fauna of Mirror Lake, 137. Beetle from the Adirondacks, A New Ambrosia, 201. Blastoids, and Cystids of Racine and Cedarville, 33. Carbohydrate Materials in Fruit Spurs and Apple Seedlings, 89. Catalog of Ohio Vascular Plants, 1920, 128. Citric Acid, Comparative Resistance of Euglenidae to, 267. Colonies of Pectinatella, 123. Comparative Resistance of Euglenidae to Citric Acid, 267 Concerning "Larval" Colonies of Pec- tinatella, 123. Corythuca celtides, Life History, 104. Cj^stids and Blastoids, of Racine and Cedarville, 33. Dendroica auduboni. Revision of Races of, 240. Dragon-flies, Phylogeny of Zygoter- ous, 19. Early Stages of Corythuca celtidis, 104. Echmoderms, Notes on, 33. Elachusta, Notes on, 206. Euglenidae, Comparative Resistance of, to Citric Acid, 267. Family Miridae, Hemiptera of, 107. Figworts of Ohio, 217. Food of the Young Gizzard Shad, 113. Forms of Oedogonium, New, 272. Forty-two Hitherto Unrecognized Gen- era and Subgenera of Zygoptera, 83. Fruit Spurs, Carbohydrate Materials in, 89. General Survey of the Macro-Fauna of • Mirror Lake", 137. Gizzard Shad, Algal Food of Young, 113 Hemiptera of the Family Miridae, 107. Inflorescences, Sexual State in Certain Monecious, 185. "Larval" Colonies of Pectinatella, 123. Macro-Fauna of Mirror Lake, 137. Microlepidoptera, Notes on, 206. Mildews, Notes on Powderv of Ohio, 211. Miridae, Hemiptera of the Family, 101. Mirror Lake, Macro-fauna of, 137. Molding Sand, Steel in Ohio, 272. Monecious Inflorescences, Sexual State in, 185. New Ambrosia Beetle, 201. New Forms of Oedogonium, 272. Notes on Elachista II (Microlepidoo- tera), 206. Notes on the Life Historv of Corythuca Celtidis, 104. Notes on the Powdery Mildews of Ohio, 211. Notes on the Work of Xyloterius Pol- itus, 201. Oedogonium, New Forms of, 272. Ohio Academy of Science, 30th Meet- ing, 1. Ohio, Figworts of, 217. Ohio, Powdery Mildews of, 211. Ohio State University, Macro-Fauna of Mirror Lake on the, 137. Ohio, Steel Molding Sand in, 272. Ohio Vascular Plants for 1920, 128. Pectinatella, "Larval" Colonies, 123. Phylogeny of the Zygo]iterous Dragon- flies, 19. Plants', Ohio Vascular for 1920, 128. Powdery Mildews of Ohio, 211. Preliminary Survey of the Macro- Fauna of Mirror Lake, 137. Racine and Cedarville Cystids and Blastoids, 33. Report of the 30th Annual Meeting of the Ohio Academy of Science, 1. Reversal of the Sexual State in Mone- cious Inflorescences, 185. 277 278 Subject hidex Revision of the Races of Dendroica auduboni, 240. Sand, Steel Molding, in Ohio, 272. Science, Ohio Academy of, 1. Seasonal Changes and Translocation of Carbohydrates in Fruit Spurs and Seedlings of Apple, 89. Seedlings of Apple, Carbohydrates in, 89. Sexual State in Monecious Inflorescen- ces, Removal of, 185. Shad, Algal Food of Young Gizzard, 113. Survey of the Macro-Fauna of Mirror Lake, 137. Translocation of Carbohydrate Ma- terials on Fruit Spurs and Apple Seedlings, 89. Vascular Plants for 1920, Ohio, 128. Work of Xyloterinus Politus, 201. Xyloterinus Politus, Work of, 201. Young Gizzard Shad, Algal- Food of, 113. Zygoptera. Unrecognized Genera and Subgenera of, 83. Zvgopterous Dragon-flies, Phvlogeny 'of, 19. MBL WHOl LIBRARY H nps 3 ;j;-i:,:i;.;;;i;.;;.;i;5^:;'5::^r,i;l:uj?v2;;;o!i!fil;;