QK M48 v. 3 [GICAL AND NATURAL HISTORY SURVEY OF CONWAY MACMILLAN, State Botanist :: Minnesota otanical Studies flfll Content^ I. II. III. IV. V. VI. Observations on Egrcgia menzie.£a.:iSr' /:/ • STUDIES PART I. Ik off VOL. III. MINNESOTA BOTANICAL STUDIES. PART I. PLATE VI. HELIOTYPE CO., BOSTON. III. OBSERVATIONS ON PELVETIA. F. L. HOLTZ. Pelvetia fastigiata (J. Ag.) DeToni, is a marine alga found distributed along the western coast of the United States and British Columbia. It grows in beds, attached to the rocks, be- tween high and mid tide, and is, therefore, daily exposed to the air for several hours (PL VIII.}. The material studied for this paper was collected by Miss Josephine E. Tilden on Van- couver Island, in June, August, and December, 1901, and was preserved in formalin. There was originally some doubt in the minds of systematists whether this plant was a Pelvetia. It has been called Fucns fastigiatum (J. Agardh, Symb., L, 3) and Fucodium fasti- giatum (J. Agardh, Sp., L, 203). The difficulty of placing it arose from the uncertainty as to the number of eggs it forms in the oogone, and this point was left undecided by DeToni. Dr. W. A. Setchell seems to have been the first to demonstrate the true generic position.* External appearance. — Pelvetia is one of the smaller wracks. It is 10-20 cm. in height, and springs from a disc-shaped hold- fast with dichotomous branches repeated till it presents a fasci- cled appearance. In well-developed plants the stipe branches immediately above the holdfast, and the branches subdivide again but a short distance farther on, so that at first sight there seems to be several fronds arising from the same holdfast. The regular dichotomy near the base may be further confused by adventitious shoots springing from near the base of the main stipe. In the material at hand but one main stipe was observed arising from a holdfast. The front may undergo dichotomy a dozen times before the terminal laminae are reached. The inter- nodes are longer toward the top. The coordinate branches keep about equal growth, though a few may remain smaller and hence appear like lateral branches (PL VII.}. * Setchell, W. A. Phyc. Bor. Am., No. 176. 23 24 MINNESOTA BOTANICAL STUDIES. The holdfast is a disc about i cm. across, and may be some- what lobed at its margin, due to protruding masses of cells that are somewhat rhizoid in function, tending to clasp the irregu- larities of the substratum. The under surface appears slightly rough and pitted. The mature stipe is elliptical in cross-section, but not winged. It is thicker and rounder at the holdfast, but flattens out into a ribbon-like shape farther up, and widens and thickens slightly toward the top. The average width at the bottom is 3 mm., and at the top 4 mm. It is about i mm. thick at the base and 2 to 3 mm. at the top. It is tough and coriaceous below, soft and fleshy above. The laminal portion is usually two-lobed, and is differentiated externally from the stipe by a rather abrupt thickening and by the fact that it is generally dotted over with the elevated ostioles of the conceptacles, giving it a warty appearance. The lamina is also more translucent than the stipe. The lamina is wedge- shaped ; the lobes into which it is divided in its upper half are tapering with rounded points. The laminae have the softest tissue in the- plant. There are usually laminae in all stages of development, on a . main branch, from cylindrical stipe-like laminae to old, flat, warty, fruiting ones. The color of the plant is nearly uniformly light brown, the older parts being a little darker, especially the lower stipe. The surface, except on the fruiting lobes, is smooth and shining. The plant is very elastic. Adventitious shoots may arise on any part of the surface of the plant. They occur chiefly where old wounds have healed over. For instance, where a branch has been torn of, or a lamina cut off, or where incisions have occurred, here may be found proliferations arising as small outgrowths. Sometimes only one may occur, again a dense cluster. Some of these de- velop into large shoots. The conceptacles may easily be seen by looking through the translucent lamina toward a strong light. They are thickly scattered over the lamina and its lobes. There is a rude ar-' rangement of the conceptacles in rows running approximately perpendicularly across the axes of the lobes. There are 150 to 200 conceptacles on a lamina. They are developed in the younger tissue at the ends of the lamina lobes. Hence the more mature conceptacles are found some distance from the tip Holtz: OBSERVATIONS ON PEI.VETIA. 25 of the lobe. Occasionally conceptacles are scattered over the stipe. These are generally less mature than those on the lamina above. They may be formed here adventitiously after those of the lamina, or else they may have been formed before or at the same time as those on the lamina and were then ar- rested in their growth. The conceptacles cause a small papilla in the surface above them. This can be easily seen with the naked eye, as can also the ostioles themselves, which appear as little pits in the tops of the papillae. A well-developed plant may have half a dozen main branches and fifty to sixty laminae. When placed in fresh water the mucilage of the interior of the plant absorbing the water, causes the laminae to burst. The distending pith pushes its way out and the cortex curls back, showing a state of tension between interior and exterior. As a result the cortex pulls off from the pith. The conceptacles then appear plainly as little spherical masses projecting from the inner side of the cortex. This intimate union of the conceptacle with the cortex might be taken as evidence of the cortical origin of the conceptacles, which is the case, as will be shown. In Plate VII conceptacles are visible on the inside of the cortex in the bursted laminae. Minute anatomy ', tissues in general. — Pelvctia shows con- siderable differentiation of tissues, though not so much as many other algae, not even so much as some of the other Fucaceae. Fucus shows greater differentiation in having a midrib and air vesicles in addition to the structures possessed by Pelvetia fastigiata. There are three principal tissues in the body of the plant. The epidermis, cortex and pith comprise the main bulk of the body. In the holdfast, however, no real pith cells are found. Epidermis. — The epidermal tissue of Pelvetiafastigiata con- sists of a layer of prismatic cells elongated radially to about twice their shorter diameters, which are about equal. The epi- dermis is best developed in the stipe and lamina. Seen here in surface view the cells present a roughly quadrangular or polygonal outline. The epidermis is shown in longitudinal and cross-sections in Figs, i-j, PL IX. The inner end of the epidermal cells and their radial walls are thin, while their free surface walls are convex outward. The surface of the epi- dermis is covered with a cuticle, thick and striated. This 26 MINNESOTA BOTANICAL STUDIES. cuticle is a common sheath to the whole plant. It is depressed into the crevices between the cells and is therefore wavy in sec- tion. It peels off in places. It shows a different, generally weaker, staining reaction from the regular cell wall. The wall underneath the cuticle is thin. The cuticularized epidermis probably is useful in preventing evaporation when the plant is exposed between tides. The hygroscopic nature of the mucilage within no doubt plays a very important part in this respect. The epidermal cells are densely gorged with chromatophores. These are yellow, highly refracting grains of oval shape. As the function of the epidermis is assimilative as well as protec- tive, the question arises, may not the convex outer walls of the cells aid in condensing the light that is necessary for assimila- tion (Kerner)? The epidermal cells have the power of dividing radially and periclinally, cutting off basal cells that are added to the cortex and cause growth in thickness. The division in planes trans- verse to the axis of the plant provides for the elongation of the plant. This cambium-like nature of the epidermal cells is also seen in the origination of a meristematic layer where a concep- tacle is to be produced (PI. XI., Fig. //), and again the grow- ing point is an epidermal cell (PL X.). Cortex. — (PL IX., Figs. 7-^,5.) Below the epidermis are six or seven rows of cells of varying size and nature, differing more or less from the epidermal and pith cells, and agreeing in a general way in not being greatly elongated and in having a large number of chromatophores. This tissue is the cortex. (The epidermis is by some writers included under this name.) The cells of the cortex are arranged with considerable regu- larity in vertical, radial and concentric rows. The regularity of shape and size tends to diminish towards the pith. The row of cells immediately beneath the epidermis is com- posed of the basal cutoffs from the epidermal cells. They are cuboidal cells of a diameter equal to the width of the epidermal cells. They, like the outer cells, are gorged with color bodies. The second and third concentric rows of the cortex are, in cross section, of equal diameter, but a little larger than the row above. In longitudinal section it is seen that these cells are generally respectively two and four times as long as the basal cells of the epidermis. Some of the cells also show this larger size in cross-section. These cells seem still to have the power HoltZ : OBSERVATIONS ON PHLVETIA. 27 of growth ; and to some extent they divide both radially and vertically, but not tangentially. These two rows of cells also have rather thin walls, although there is some thickening at the angles. They are also densely packed with chromatophores. Below the above-mentioned cells are three or four (five) con- centric rows of cells that pass over into the pith on the inner side. Their walls are considerably thickened with a gelatinous substance, which, however, is firmer and denser than that of the pith. These walls stain deeply. The cells of these rows contain color grains but in more loosely disposed masses. The protoplasmic sac is more easily seen around these masses of chromatophores than in the outer cells. These cells have nearly the same radial diameter as the cells in the second and third rows, but are generally twice as wide tangentially and twice as long vertically as those of the third row. There is more or less variation in the number of rows of each of these different sizes of cells. These elements may be diagrammatized as in PL IX. + The original walls between these cells thicken as the pith is approached. The cells lose their rectangular shape more and more towards the pith till at last it is sometimes difficult to dis- tinguish them from the more cylindrical pith cells. The cells remain in communication through pits, the cells anastomosing" frequently. The longer cells form transverse septa, which are often oblique to the lateral walls. These septa are never thickened, but remain very thin and, to all appearance, allow protoplasmic communication (PL IX., Fig- 12). Pith. — (PL IX., Figs. 7-5.) The pith of the stipe and lamina is distinguished by the fact that the cells are separated widely by intercellular jelly, which in the lamina is from two to three times as thick as the diameter of the cells imbedded in it, less thick in the stipe (1-2). The pith is also marked off by the jelly not staining as deeply as the intercellular matrix in the cortex. With some stains, fuchsine for example, the stain may be almost completely removed by washing, leaving the inner wall of the pith cells colored. Pith cells are slightly com- pressed corresponding with the flattening of the stipe or lamina (PL IX., Fig. /). They are nearly as wide as the average width of the cortex cells, and are about as long or a trifle longer than these. They are crossed by delicate septa (PL IX., Figs. 2, 4, /j). Pith cells are joined into vertical rows or filaments- 28 MINNESOTA BOTANICAL STUDIES. which wind about and intertwine somewhat with each other. These filaments anastomose frequently and are often dichoto- mously divided (PL IX., Figs. J, 4* /?)• The cells of the pith in the central part of the stipe or blade are of nearly uniform diameter throughout their length and are regular in shape, except at anastomosing plexi and near the cortex where they are subject to distortion in shape and to displacement. The pith cells contain a few chromatophores collected into a little pellet near the middle of the cell. The protoplasmic con- tents show up well and numerous refracting grains of reserve material are seen. The gelatinous intercellular matrix swells up greatly when the plant is placed in fresh water. This causes the lamina to burst open, beginning at the more tender tip of the lobes in the young laminae. The stipe having a firmer cortex and also pro- portionately less intercellular gelatine, does not burst, though it swells some. Anastomosis and Pits. — Anastomosis is seen best in the pith cells (PL IX., Figs. 2-4., 12-14). Sometimes two filaments will simply be bent toward each other, touching with their con- vexities (Fig. 2). At this point there is no intercellular jelly separating the cells. A thin communicating plate is between the cells in contact, and the protoplasm in these cells sends out branches that meet at the plate. At other times the anastomos- ing cells send out lateral protuberances, which passing through the jelly, meet and form a pit at the point of contact (Figs. 4, /j, 14). Probably these protuberances were not formed before the pit was formed, but are the result of the growth and modi- fication of shape of the cells which took place after the pit already was made. Judging from the conditions in the cortical layers, these pits are simply the original fission walls left un- thickened at these spots (Fig. 12], while at other points the contiguous cells were forced apart by the development of the gelatinous middle lamella between the two cells, however, leav- ing the cells in contact at the pits. The pits at lateral anasto- mosing points are smaller than at the ends of the vertical cells. The pits do not stain as deeply as the rest of the cell wall, but this might be due to their greater thinness. They are sharply marked off in the walls of the pith and the inner cortex cells. They are round or oval plates which in optical section are of uniform thickness, not lenticular. They show less definition of HoltZ : OBSERVATIONS ON PELVETIA. 29 shape as the outer cells of the cortex are approached. Here they appear to be simply the original dividing wall. They can, however, be located by the protoplasm apparently running right through the wall at these places. This apparent communica- tion of the protoplasm of adjoining cells was observed as far out as the second layer of cells below the epidermis. Farther out this could not be seen on account of the chromatophores. But probably even the epidermal cells communicate with each other. The concentrated sulphuric acid test showed that the plates were dissolved as well as the rest of the wall. No positive proof was found that the pits were perforated, no threads of protoplasm having been observed, as would indeed be difficult with the extreme thinness of the plates. But the symmetrical arrangement or attachment of the protoplasm on both sides of the pits leads one to suspect very strongly that there is com- munication. By plasmolysis the protoplasm draws away from the cell wall at all other points than the pits (Fig. /j). It remains attached here and extends in ropes through the cells and seemingly through the pits. The protoplasm often branches to lateral pits (Fig. 14). When the pith cells are swollen in fresh water the protoplasm is frequently torn off from one end of the cell, away from a pit, owing apparently to the elongation of the lateral wall as well as the gelatinous matrix. In such cases the pit curves in toward the loosened protoplasm (Figs. 12, 14). Iodine is the most satisfactory stain to use in studying pits. The protoplasm is stained and its attachments may be studied. Pits and anastomosis may be nicely studied by removing some of the protruding pith from a lamina that has burst in fresh water. By flattening the gelatinous mass under the cover glass the pith cells and their pits show up well, even unstained, though better if differentiated with stains for walls and for protoplasm. Anatomy of Holdfast, Stipe and Lamina. — The above mat- ter on the tissues in the body of P.fastigiata needs some modi- fication and addition when the holdfast, stipe and lamina are considered separately. * Holdfast. — In a vertical section through a holdfast it is seen to be composed of approximately regular, ascending rows of cells ; those near the central part more vertical ; those near the border of the holdfast curving out as they go down. There is a marked difference between the cells in the middle and those SO MINNESOTA BOTANICAL STUDIES. in the peripheral part. The former are irregularly quad- rangular in outline in both vertical and cross-section. Their diameters in cross-section are equal (Fig* 6). They are more regularly disposed in rows than the cells in the mar- ginal parts. The cells of the holdfast have walls of good thickness, composed in part of the usual mucilaginous sub- stance. Taking the central part of the holdfast (Fig. 6), the lowest cells are dead and empty and partially disintegrated into mu- cilage. Clefts arise among the still living cells from this disin- tegration. Gaps are also caused in this way in the body of the holdfast. Here and there individual cells at the bottom have a disc-like lower surface as if they had a holdfast of their own. The decay of the cells near the central part of the holdfast ex- tends not more than one or two cells deep. The next few rows of cells are slightly flattened parallel with the base of the hold- fast. The succeeding rows of cells become gradually elongated in a vertical direction till, at the tenth or eleventh row, a rapid dif- ferentiation begins with cells elongated in the vertical direction to two or four times their horizontal diameter. Evidently the stipe begins at this zone. The peripheral portion (Fig. 7), as was stated above, is com- posed of rows of cells descending obliquely from the axis of the stipe. A vertical section through this part shows that these rows of cells branch dichotomously in the horizontal plane as they go down and outward. The cells decrease in length as one follows the dividing branches, till a zone is reached in what corresponds to the cortex of the stipe. Here the ultimate dichotomous divi- sions of the main strands form a meristem of small cells. The cells of this meristem run in straight rows perpendicularly to the surface of the holdfast. They are in active division. This meristematic layer enables the holdfast to grow in thickness and also to form the rhizoid-flaps on its edge. It is about eight to ten cells deep. In the specimen examined the basal cells, three to five deep in this part of the holdfast, showed advanced disin- tegration. The epidermal layer near the lower edge also was in a similar condition. But the cells of this layer are more re- sisting and persist alive after the two rows beneath are already dead. Probably the mucilage derived from the disintegration of these basal cells is useful in attaching the plant to the sub- stratum. HoltZ : OBSERVATIONS ON PELVETIA. 31 The cortical part of the holdfast passes without any marked change into that of the stipe. The epidermal cells however are not elongated as much radially as those of the other parts of the surface of the plant. It is covered with a cuticle, thicker than on the stipe or lamina. Cross-sections of the central part of the holdfast show (Fig. 8) that the vertical rows of cells seen on vertical section are not disposed in any regular order. The intercellular substance is not nearly as abundant as in the stipe. Toward the margin of the holdfast the cells show power of dividing. Here we find, interspersed with cross-sections of the vertical rows, sections through cell rows slanting up toward the axis of the plant. Still nearer the outside we come upon the meristematic zone. Here are principally slanting rows of cells dividing dichotomously in the radial direction. These divisions repeat the dichotomy, running directly to the surface. All the living cells in the holdfast have chromatophores. The central cells contain but few grains, the cortical are crowded with them. Stipe. — But little need be added here to what has been said under tissues in general. The young stipe has a nearly cylin- drical structure, with a slight notch on the end where a growing point is situated. No differentiation is noticeable between stipe and lamina. Older stipes become flattened, partly on account of the flattening of the cells parallel with the longer axis of the cross-section, but more on account of the greater growth toward the thin margins. A cross-section of an older stipe shows two principal planes of fission by the arrangement of the cells in rows parallel with the major axis of the section and the other obliquely across this axis. This is especially noticeable in the pith. The cortical cells show a distinctly concentric arrangement (Fig. /). The only differentiation seen in cross-section is that the pith and inner rortex cells near the ends of the ellipse are somewhat larger than those of the central part. This differentiation how- ever does not even suggest a midrib. Longitudinal sections of the stipe, cut parallel to the flat surface, show a similar appear- ance, except that the typical pith cells are reached sooner in passing from the surface along the minor axis. The appear- ance of the cells in both cross and longitudinal sections has been discussed under tissues in general. Lamina. — The general tissues of the stipe and lamina are so 32 MINNESOTA BOTANICAL STUDIES. similar that no change is noticeable in passing from one to the other. In the lamina proper, however, the pith cells branch more and the rows of cells have a more meandering course, and there is more anastomosing. The intercellular jelly is de- veloped here more than elsewhere in the plant. Due to this and some to the branching of the cell rows the blade is much thicker than the stipe. The cortex and epidermis are similar to those of the stipe. The crowding growth of the conceptacles disarranges the order- liness of the cell arrangement in the cortex and epidermis. Cross and longitudinal sections of the lamina resemble those of the stalk closely, except for the differences just mentioned, and for the conceptacles (Fig. 3 shows a partial cross-section of a lamina). The growing foint. — In the tip of the maturer laminae no definite growing point can be found. There still is some growth and cell division going on here in the outer cortical cells, and in a mature lamina this is probably the youngest and tenderest portion. It is here that the lamina begins to burst when placed in fresh water. Even at the sinus between the lobes of the lamina no growing point can be found in older laminae. This is the place where the growing point once was. But the growth seems to have stopped here first and continues for a time longer toward the ends of the lobes. If a young frond is examined, one in which there is as yet no difference between stem and blade, a slight notch or dimple is visible at the top. This notch deepens in older fronds, and if a section is made through the somewhat flattened stipe, par- allel with the flat surface and through the axis of the frond, a large apical cell is seen at the base of the sinus (PL X., Figs. /j, 16). This apical cell is an epidermal cell. It is in the shape of a truncated pyramid, with the truncated end to the top or pointing outward. The apical cell is two or three times as large as the other epidermal cells, and is otherwise markedly distinguished by great richness and granularity of contents, and by the absence of chromatophores. The adjacent cells share these characteristics to a less degree. They show a diminution in the granularity of the protoplasm, and color grains begin to appear in all but the latest cutoff. The apical cell, as seen in vertical section, cuts off daughter cells in succession, a lateral, then a basal, and then a lateral Holtz: OBSERVATIONS ON PELVETIA. 33 on the other side (see diagram, Fig. 16). The daughter cells quickly divide again and again, but more frequently in a lateral direction from the apical cell than downward. The cells in these lateral zones divide more rapidly in planes transverse to the axis of the lamina. In this way the zone of most rapid growth extends out laterally and upward from the apical cell and soon grows up ahead of the growing point. As a result there is the bifurcated lobe. Differentiation into the long pith cells begins only three cut- offs below the apical cell. In the wings it does not begin so soon. The zone of cells in the wings retains its power of fis- sion longer than the cells below the apical cell. The cells of the epidermis and the cortical zone attain the characteristics of these tissues but a short distance from the apical cell. The outer cortical cells throughout the plant are capable of dividing and seem to constitute a kind of cambium around the plant. This meristematic nature of the cortex is most highly developed in the lobes of the young lamina near the growing point. It is also well developed where conceptacles form and in the marginal parts of the holdfast. The cuticular sheath that covers the whole plant is very thick over the delicate growing point, being about as thick as the length of the epidermal cells beneath it, no doubt serving as a protection. It is customary to speak of the rows of cells in the plant as hyphae. But when the origin of these cells is considered, that they are derived directly or indirectly from a single apical cell, the idea of their hyphal character seems a little incongruous. On the development of the conceptacle. — As before noted the conceptacles show an intimate connection with the cortex. Sectional views prove the cortical origin and nature of these structures. The first indication of the beginning of a conceptacle is seen to be the cutting off of a basal layer of cells from the lower end of a few adjacent epidermal cells (PL F., Fig. //). These basal cells in turn divide periclinally and radially to form a little pad of meristematic cells beneath the epidermis, around which the cortical cell-rows are deflected. Directly over this mass of basal cutoffs, usually in the center, one or more epi- dermal cells begin to show signs of disintegration and collapse. 34 MINNESOTA BOTANICAL STUDIES. The walls of these cells stain more deeply than those of normal cells, the nuclei disappear and the chromatophores fuse together into a dark mass. The affected cells collapse gradually, be- ginning at the outer end. Often a little conical remnant of the shrunken cell may be seen on its basal cell. The walls and contents of the disintegrating cells change into a mucilaginous substance. Thus far my observations agree with those made by F. O. Bower.1 Bower states that the epidermal cell collapses, but that the basal cell persists, and that it sinks farther and farther into the cavity of the conceptacle, and that the lateral daughter cells of the central basal cell by their division form the lining wall of the cavity. He seeks to limit the disintegration of the epidermis at first to one cell and to make its basal cell the cen- ter of the whole process of the development of the conceptacle. The serial sections made by me for the investigation of this matter do not show that the disintegration is thus confined to one single epidermal cell. Occasionally several will be equally far advanced in decay. Naturally one or the other of these may decay more quickly than the rest, producing thus a line of weakness and apparently a central axis about which the other decaying cells are grouped. Again, it was not found that the basal cell or cells of the dis- integrating epidermal cells persisted. On the contrary, they and several rows of cells below, perhaps five or six, share in this disintegration. It was frequently possible to make out the remains of the disintegrating cells in the mucilaginous mass to which they changed, and with which the cavity formed by their collapse was filled. Neither did it appear that the basal cutoffs of the epidermal cells produced lateral daughter cells to line the cavity. It did appear that they divided chiefly periclinally and somewhat radially, forming five or six rows of meristematic cells, the outer rows disintegrating and forming the cavity ; the deeper ones persisting and finally forming the inner wall of the conceptacle and giving rise to paraphases and the reproductive organs. Bower shows figures like /p, PL XI., in which the two cells, b and c on either side of the central basal cell a, might suggest that they were the lateral daughter cells of this basal cell. But 'Bower. Development of the Conceptacle in Fucacese. Qr. Jr. Mic. Sci. 36 . 1880. Holtz : OBSERVATIONS ON PELVETIA. 35 sections like Fig. 21 are met with in which it is clearly seen that these lateral basal cells are not the daughter cells of the central basal cell «, but that they are the basal cutoffs of the epidermal cells above them. They are, therefore, coordinate with the basal cell a. The cells e and y are later cutoffs which the epidermal cells g and h succeeded in cutting off before be- coming affected by decay. On account of less resistance from the cavity than on other sides these lateral basal cells grow usually in the shape showrn in Fig. 19. To summarize my conclusions on this point, the conceptacle originates by a few contiguous epidermal cells cutting off basal cells, Fig. //, which are meristematic, dividing principally periclinally into half a dozen or more tiers of cells. Directly over this meristematic mass of cells, whether by the tension produced by the growth of the cells below, or otherwise, one or several epidermal cells begin to show signs of decomposition. The disintegration proceeds and the cells collapse (Figs, /p and '• & 0 ^ ^\rV i^*^ ^, O x^. ^Qfe> ^/^ Q^ o o ^ "' c;->^ ^ m -O o CxJ 29 r VOL. III. MINNESOTA BOTANICAL STUDIES. PART I. PLATE XII. HELIOTYPE CO., BOSTON. IV. PETALONEMA ALATUM IN MINNESOTA, DAISY S. HONE. History. — This alga was discovered first in West Scotland by Captain Carmichael, about 1823-1828, and was first figured and described by Dr. Greville as an Oscillatoria. It has since been found in various places in Europe, both in the British Isles and on the Continent. In 1849 Harvey found it growing on dripping rocks under Biddle Stairs, Niagara Falls, in America. It was found by the writer near Minneapolis, Minnesota, in Oc- tober, 1901. Collection and -preservation. — The Minnesota material was collected from the gravel bed of a quiet stream, the outlet of an old tank near the Government Dam works, Minneapolis, on October 12, 1901. It formed a dark chestnut brown stratum. A portion was preserved in a 5-per-cent. formaline solution and the remainder dried for herbarium specimens. The dried ma- terial when soaked regains its original form so that it is as good for study as the preserved. However, the formaline material was used in the work recorded in this paper. Methods. — A small portion of the pickled material was washed in water and then mounted directly in glycerine jelly. If the jelly be raised above the melting point the threads col- lapse. All the drawings used in this paper were made from a .single slide which has been thus preserved. Staining in toto was tried with several fluids, but without valuable result. After washing the material in water for twenty- four hours it was treated with dilute hydrochloric acid for twenty- four hours and then various staining fluids tried. A dilute solution of Kleinenberg's hsematoxylin stained the sheath a beautiful blue, leaving the trichome deep green. Aniline safra- nine stained the trichome red without ^Hecting the sheath. Dahlia colored the trichome a deep blue and slightly affected the sheath. Fuchsin acid stained the whole red. Iodine turned the trichome brown. 47 48 MINNESOTA BOTANICAL STUDIES. Material was also washed with water for twenty-four hours and then passed through the alcohols before staining, but no advantage was gained. The material treated with hydrochloric acid differed only in that it showed a more distinct vacuolated condition, which in the younger active pseudocysts was very ap- parent. Many of these pseudocysts also showed a single very large granule, but this may also be seen in the normal con- dition. Nearly all the stained pseudocysts are constricted along the middle region. OBSERVATIONS. The dark brown color of the stratum is due to the color of the gelatinous sheaths in which the trichomes are imbedded. Filaments. — The filaments are not attached, yet the gravel remained clinging to them when they were detached from the bottom of the stream, due no doubt to the gelatinous nature of the sheath. They seemed to lie horizontally and to be without any definite arrangement. They are in general more or less curved. Many of the filaments are without branches, but pseudobranches are not at all uncommon. Branching occurs either near a heterocyst or at a distance from it (see Figs. 5 and 6). In the first case there is but one branch, that is, the tri- chome being broken off at the heterocyst is thrust out as a single thread which soon secretes a new sheath about itself. In the latter condition both the broken ends of the trichome project, so that there are two branches or twin branches. Sheath. — Harvey's description cannot be improved upon : "When placed under the microscope the filaments present the appearance of a cylindrical central column, containing annu- lated, olive-colored endochrome and a wide, wing-like border at each side of the column. This border or sheath is obliquely striate, the striae running in an arch from the margin toward the center, where they become parallel and are then continued longitudinally downward along the medullary column till lost in the density. The margin of the wing is closely crenulate, and in age transversely striate at the crenatures as though jointed. Such is the apparent structure ; the real structure seems to be that an annulated central filament is enclosed within a number of compressed, trumpet-mouthed gelatino- membranaceous tubular sheaths, one arising within the other and successively developed as the growth proceeds. These sheaths, thus concentrically arranged, are indicated by arching Hone: PETALONEMA ALATUM IN MINNESOTA. 49 longitudinal strise, and the mouths of the younger sheaths, pro- jecting slightly beyond those of the older, form the crenatures of the margin." I find the central cylindrical column containing the trichome to be dense and often very thick near the heterocyst (Fig. 4)<> while near the apex of the filament it generally becomes thin and often scarcely traceable (Fig. j). Thus it would seem to be a second sheath within the larger outer one, or it may be merely a very dense interior folding of the sheath proper, inten- sified near the heterocysts because of the greater rigidity of that portion. The internal striations of the sheath have a beautiful golden or dusky brown feathery appearance. Those farthest within being brown or golden present all shades of brown and yellow as they approach the periphery, where the sheath becomes colorless and transparent. A quite common condition is illustrated in Fig, 5, in which a dense old sheath, contracted for a limited distance, after branching expands into the usual form. Trichome. — The trichome is normally of an olive-green color, cylindrical or somewhat moniliform, separated into dis- tinct pseudocysts or apparently continuous. The apex is often gradually constricted with the tip enlarged. Very often this tip is rose-colored. Pseudocysts. — The pseudocysts are exceedingly variable in size and shape, this depending upon their age. The younger active pseudocysts are globose in shape, 9-15 mic. wide. When ready to divide they lengthen until they are twice as long as wide (Fig. -7). The contents are coarsely granulate. Often a single large granule is seen in each pseudocyst (Fig. j). The older pseudocysts are often rectangular in outline (Fig. 4., a). These are sometimes 20 mic. long and 6 mic. wide. They are more finely granular and densely packed. The apical pseudocyst and often three or four below are coarsely granulate and are of a deep pink or red color. Heterocysts. — The heterocysts are interstitial and sometimes occur at the base of a branch (Fig. 5). They are solitary. In shape they are somewhat globose or oblong. In stained mate- rial the watery contents take a beautiful coTor. In size they are slightly larger than a normal pseudocyst. 50 MINNESOTA BOTANICAL STUDIES. BIBLIOGRAPHY. Greville, R. K. Scottish Cryptogam ic Flora, 222. 1823-28. Berkeley, M. J. Gleanings of British Algae, 23: pi. 7. f. 2. 1833. Harvey, W. H. Nereis Boreali Americana, 3: 99. pi. 48 A. 1858. Wood, H. C. A contribution to the history of the fresh-water algae of North America, 66, 67. 1872. Wolle, F. Fresh-water algae of the United States, I : 267. 1887. 2: pi. i88.f. is, 16. 1887. Bornet and Flahault. Revision des Nostocacees heterocystees. Ann. Sci. Nat. Bot. VII. 5: no. 1887. Kirchner, 0., in Engler and Prantl, Die natiirlichen Pflanzenfamilien, I : (i Abt. a) 79. f. 57 c. 1900. EXPLANATION OF PLATE XIII. Figure i. Apex of a filament showing striae of the sheath, or the " mouths" of the younger sheaths projecting from the older ones, #, trichome, x 154. 2. Same view as Fig. i, but with different focus showing the con- centrically arranged striae with the central portion. #, trichome, X 154. 3. Apex of a filament with well-developed trichome about to slip out from end of sheath, x 154. 4. A portion of a normal filament showing heterocysts and hormo- gone with a very dense central folding of the sheath about the trichome, X 154. 5. A portion of an older filament showing a branch given off at the heterocyst, x 154. 6. A portion of a filament showing four branches not in the neigh- borhood of a heterocyst, x 154. 7. A trichome showing active division removed from sheath, x 154. 8. 9. Stained trichomes removed from sheath. 8. Mature pseudo- cysts. 9. Young pseudocysts, x 154. /OL III. MINNESOTA BOTANICAL STUDIES. PART I. PLATE XIII. HELIOTYPE CO., BOSTON. V. OBSERVATIONS UPON SOME ALG^ WHICH CAUSE " WATER BLOOM." N. P. B. NELSON. An endeavor has been made to collect some of the more im- >ortant known facts concerning " water bloom" as it occurs in [innesota and neighboring states. It seems quite certain that it least some of the algae causing this appearance in water mpplies are of considerable sanitary importance. It is the object of this paper to give, not a complete and scientific treatise, but a general and practical description of the phenome- non and its causes and effects so far as known. Up to this time there is no record of its occurrence to any great extent in the rivers or lakes supplying drinking water to the inhabitants of cities and towns of the state, but in several instances it has apparently caused the death of cattle and other animals. Occasionally persons interested in the matter have sent speci- mens to the Department of Botany at the university, and it would be well if this were more generally done. Such collec- tions with notes accompanying them are of the greatest aid in acquiring a more general and complete knowledge of this kind of vegetation. A small amount of the material, immediately upon being taken out of the water, should be placed in a vial containing a 5-per-cent. solution of formaline. It can then be packed in a small box with cotton and sent at any convenient time. Such a solution of formaline can be purchased for a few cents at any drug store. History. — The first published record of the occurrence of this " water bloom " in the state of Minnesota was that of J. C. Arthur (i) in 1883. "A very fatal disease among cattle and hogs in Waterville, Le Sueur county," was reported on the 8th of July, 1882. Professor Arthur visited the locality and describes the condition as follows: " On June 28, 1882, after two or three days of pleasant weather, the wind gathered a thick scum of algae in the little bay (on the north shore of Lake 51 52 MINNESOTA BOTANICAL STUDIES. Tetonka near the house of Mr. L. H. Bullis). Four calves confined in a pasture near the house, with access to no water but that of the lake were seen at noon apparently well, and at 2 P. M. were dead. On July 5, a number of cattle came down the public road to the lake shore, that partly belonged to Mr. Bullis and partly to neighbors. They were noticed between 8 and 9 A.M. and within an hour afterward three were dead, and before noon three more. . . . The two young cattle were examined shortly after death by Dr. E. B. Case and Dr. J. G. Bemis, resident physicians. Nothing seemed to be abnormal except the stomach, which appeared to have been affected by the algae swallowed by the cattle. . . . The cattle did not appear to suffer pain, but lay down as if enervated and soon expired." The total number of animals thought to have died from the same cause at this time included about twenty head of stock, horses, cattle and hogs. The scum when examined was found to consist of minute balls each made up of a dense colorless jelly in which was embedded a great number of dark-green, whip-like filaments, lying side by side and radiating from a center. The larger ends were at the center and the attenuated ends extended beyond the jelly. Each filament was made up of a row of pseudocysts enclosed in a sheath and at the basal or inner end was attached a spherical heterocyst. When decaying in masses the plant caused a nauseating odor. The plant was determined by Dr. Farlow to be Rivularia ftuitans Cohn. Several weeks later Professor Arthur revisited Lake Tetonka. Rivularia fluitans had disappeared, but in its place was another scum-forming alga, intensely green in color, diffused through- out the water and collected by the wind into a scum two or more inches thick along the shore. Under the microscope it appeared to consist of irregular colonies of minute plants. Each colony was a mass of thin, colorless jelly containing many separate oblong blue-geen cells placed some distance apart. This plant is known as Ccelospkcerium kuetzingianum Naeg. and is not considered injurious. A small quantity of Anabcena circinalis (Kuetz.) Rabenh. was also found associated with this plant. Still another " water bloom " species found at this same time and almost as abundant as the Coclosphczrium was an alga named Aphamzomenon jlos-aquce (Linn.) Ralfs. It consisted of little bundles of thin, delicate filaments. Nelson: SOME ALG^E WHICH CAUSE "WATER BLOOM." 5!} The next reference to the subject was made in 1889 by Dr. Wm. Trelease in an article, "The Working of the Madison lakes, Wisconsin." He observed that every season a greenish- yellow scum occurred in greater or less quantity on the lakes during the hot part of the summer, after the weather had been calm for a number of days in succession. When but little of the scum-forming substance was present it appeared as fine granules suspended in the water. Under the influence of a gentle breeze, continuing in one direction for some time, these particles were carried to the shore, accumulating to form a slimy scum which quickly putrefied, giving off a disagreeable odor. During the process the color of the mass changed to a decided blue-green, which stained the pebbles, sticks, etc., over which it was smeared. This material consisted mainly of Clathrocystis aeruginosa (Kuetz.) Henfr. At different times collections taken from the lakes proved to be, besides the Clathrocystis, Anabcena flos-aquce (Lyngb.) Breb., A. mendota and A. circinalis (Kuetz.) Rabenh. In August, 1897, Miss Elizabeth H.. Foss, a student in the Botanical Department, collected Glczotrichia ptsum, floating in large quantity on the surface of Lake Minnewaska, Glen- wood, Minnesota, and on October 28 of the same year, Miss M. G. Fanning and Mr. H. B. Humphrey found Anabczna Jlos-aquce in abundance in Cedar lake, Hennepin county, Minnesota. In November, 1899, Miss M. G. Fanning, then a student in the Botanical Department, began making a study of the St. Paul water supply. Of the "water bloom "-forming algae she found specimens of Anabcena jlos-aquce (Lyngb.) Breb., and Ccelosfh&rium kuetzingianum Naeg. In August, 1900, Professor Caswell A. Ballard, of Moorhead Normal School, Moorhead, Minnesota, made a collection of a "water bloom" form from one of the shallow lakes in the depressions of the Fergus Falls moraine. A sample of the material was sent to this department and it was determined to be Aphanizomenon jlos-aquce (Linn.) Ralfs. (PI. XI V., Fig. /). Professor Ballard's attention was first called to this locality by the report that several cattle in a pasture adjoining the shore of the lake had died, apparently from poisoning. It was observed that the lake was in a peculiar condition, the water colored by a blue-green scum. A zone from twenty to twenty- 54 MINNESOTA BOTANICAL STUDIES. five feet wide, from the shore out into the lake, was almost thickened by the presence of a great number of colonies or bundles of this plant. This scum being suspected as the source of danger, a temporary fence was put up to prevent the cattle from drinking out of the lake. After that, none of the cattle died or showed symptoms of being poisoned. In a letter describing the circumstances, Professor Ballard states that he is convinced that the death of the cattle was due to the drinking of the lake water, either because of the poisonous characters of the algae, or what seems more likely, because of the stagnant condition of the water which made the growth of the algae possible. On October 13, 1901, the writer collected some " water bloom " on the shores of Lake Minnetonka at Spring Park. The water here did not seem to be very deep for some distance from the shore. Neither was there a clean sandy bottom in the vicinity. Bulrushes grew abundantly in places and there were present some smaller water plants. The water of the lake was quite fresh and clear. The algal scum was gathered at a few places along the shore by a gentle land-ward breeze. It had a pale bluish-green color resembling a mixed paint of that color. Microscopic examina- tion showed the scum to be made up of two species of Anabczna, A. fos-aquce (Lyngb.) Breb. (PL XIV., Fig. j) and A. cir- cinalts (Kuetz.) Rabenh. (PL XIV., Fig. 2). On October 24 another collection was made at the same spot. The scum was about the same in quantity, but was of a grayish- brown color, slightly tinged with blue-green. It was more slimy to the touch. It had an odor similar to mouldy grass or hay. Under the microscope it was seen that this difference was probably due to the presence of a much larger number of gonidia or reproductive bodies than had been found in the previous collection. To the naked eye this scum appeared to be simply a shapeless mass, but examined with the microscope it was found to consist of innumerable filaments, resembling strings of beads. Each of these filaments or trichomes was composed of a number of somewhat spherical pseudocysts, almost uniform in size and shape. Each pseudocyst was filled with a dense, finely gran- ular protoplasm. At intervals in the trichome were seen hetero- cysts, larger than the pseudocysts, with a more distinct wall and Nelson: SOME AI.G^E WHICH CAUSE "WATER BLOOM." 55 dn watery contents. The gonidia were still larger, were sur- mnded by a thick wall and contained numerous granules of different sizes. When floating naturally in the water, each trichome of Anabcena circinalis coils itself into a loose spiral — hence its name. The trichome of A. flos-aquce, on the other hand, is somewhat curved but not in a definite way. These plants in moderate quantities are not supposed to be dangerous, but when they are present in immense numbers in stagnant water they are likely to have an injurious effect. In 1897 Messrs. D. D. Jackson, Assistant Biologist, and J. W. Ellms, Assistant Chemist of the Massachusetts State Board of Health, made a series of chemical experiments on living plants of Anabcena circinalis collected from Ludlow Reservoir, at Spring- field. " It is commonly believed by those who have not investi- gated the subject, that disagreeable odors and tastes in drinking waters are due to the decomposition of organic matter, and are either dangerous or indicative of danger to the public health. Biological investigations already published have sufficed to show that this is not always the case." The plant under investigation proved to contain an essential oil giving the order of mouldy grass which is characteristic of the genus. A chemical analysis was also made of the same plant in a state of decay and showed that " the odor of decomposing An- abcena is evidently not due, to any extent, to the production of hydrogen sulphide, but to the partial breaking down of highly organized compounds of sulphur and phosphorus. The odor is undoubtedly more offensive on account of the high per cent, of nitrogen present. It is true of the whole organic world that those products which give the most offensive odors of decay are partially decomposed, highly nitrogenous compounds, contain- ing sulphur or phosphorus." The investigators concluded that the usual cause for disagree- able odors and tastes occurring in potable water is found in the presence of large numbers of certain microscopical organisms which secrete compounds of the nature of essential oils. When the organisms are living these tastes and oders are as harmless as those of fresh vegetables or fish. When decaying, the plant produces the " pig-pen " odor (characteristic of blue-green algae, Cyanophyceae) due to the decay of highly nitrogenous organic matter in which partially decomposed sulphur and phosphorus 56 MINNESOTA BOTANICAL STUDIES. compounds play the leading part. The sanitary significance of this matter is yet to be determined, but so far analysis indicates that, in large quantities, the effect on general health would be unfavorable. Summary. — Up to the present time there have been found, in or near the state of Minnesota, seven kinds of blue-green algas which form " water bloom." They are : Gloeotrichia pisum (Ac.) THURET. {Rivularia fluitans COHN.) Ccelosphczrium kuetzingianum NAEG. Aphanizomenon flos-aquce (LiNN.) RALFS. Clathrocystis aeruginosa (KUETZ.) HENFR. Anabcena circinalis (KUETZ.) RABENH. Anabcena flos-aquce (LvNGB.) BREB. Anabcena mendotce ( ?) In several instances it has been almost conclusively proved that the presence of one or more of these species in drinking water used by stock has caused fatal results. BIBLIOGRAPHY. Arthur, J. C. Some algae of Minnesota supposed to be poisonous. Bull. Minn. Acad. Nat. Sci. 2: (App.) 1-12. 31 My. 1883. Trelease, W. The working of the Madison lakes. Trans. Wis. Acad. Sci. Art and Letters, 7: 121-129. P^ IO- l%%9- Fanning, M. G. Observations on the algaa of the St. Paul city water. Minn. Bot. Studies, 2: 609-617. pi. 45, 46. 20 Jl. 1901. Jackson, D. D., and Ellms, J. W. On odors and tastes of surface waters, with special reference to Anabcena, a microscopical organ- ism found in water supplies of Massachusetts. Review of Am. Chem. Research, 3: 1897. EXPLANATION OF PLATE XIV. (IN PART). Figure i. A "bundle" of trichomes of Aphanizomenon flos-aquce. Drawn from Professor Ballard's collection, x 700. 2. Anabcena circinalis. a, pseudocysts ; •£, gonidium ; c, heterocyst, X 193- 3. Anabcena flos-aquce. a, pseudocyst ; 3, gonidium ; c, heterocyst, X 193- fOL. III. MINNESOTA BOTANICAL STUDIES. PART I. PLATE XIV. HELIOTYPE CO., BOSTON. r VI. SOME OBSERVATIONS ON THE STAINING OF THE NUCLEI OF FRESH-WATER CATHERINE HILLESHEIM. The material studied comprised several species of the com- moner green algas collected in the stone quarries along the bank of the Mississippi river near the university. Staining. — The only fixing agents tried were chromic and picric acids. Various stains were used, such as hagmatoxylin, fuchsin, anilin safranin, gentian violet, borax and ammonia car- mine. Staining living cells with dahlia was unsuccessful. After the material was stained and thoroughly washed, first in water and then in the alcohols, it was mounted in glycerine jelly or in formaline. The best results were obtained from the following method : Chromic acid, 24 hours, Water, 24 " Alcohol, 10 per cent., 4 4t Alcohol, 30 per cent., 4 u Alcohol, 50 per cent., 4 " Borax and ammonia carmine (one half of each) , 4 days. Glycerine and water (5 per cent, solution), 5 minutes. The slide and cover-glass were then warmed and a small drop of glycerine jelly placed in the center of the slide. When this was melted the stained material was placed in it, the cover- glass laid on and the whole put away to dry. When mounted in formaline the preparation is ringed with Canada balsam to make it air-tight. CELL STAINING. Spyrogyra species. — The nucleus readily took on nearly all of the stains mentioned. It was stained pink by the ammonia- borax, carmine, fuchsin and aniline safranin. Haematoxylin stained it blue. The nucleus was situated near the center of the cell. It was much varied in shape in different species. 57 58 MINNESOTA BOTANICAL STUDIES. Thus, some were polyhedral, some oval, some spherical and some irregular in form. It also varied in size. The nu- cleolus stained much more deeply than the nucleus and was spherical in shape. Radiating in all directions from the nu- cleus are the lighter staining strands of protoplasm which, terminating in the pyrenoids of the chlorophyll bands, suspend the nucleus in the cell. Zygnema species. — A mixture of ammonia and borax carmine gave the best results in staining in this form. Various stages in nuclear division were clearly brought out. PL XIV., Fig. 7, shows two daughter nuclei just after the formation of the cell plate. The nucleus is an elongated, oblong, bean-shaped body situated between the two chloroplastids. The nucleolus is generally situated in one end of the nucleus and is spherical in shape. Dahlia was tried but it stained all the contents of the cell without bringing out the nucleus. Hormiscia zonata (Web. and Mohr) Aresch. — The method of staining was the same as above. The nuclei are somewhat spherical in shape, and occupy different positions in different cells. They lie within the chloroplastid, either at the center or near the wall. Microspora species. — In addition [to the first method, double staining was tried, the material being first stained with anilin safranin and then with gentian violet. This proved to be no more satisfactory than the first method. Stained with hasma- toxylin the nucleolus was brought out much more clearly than in either of the other ways. The method used is as follows : Chromic acid, 33 hours. Water, 22 «* Haematoxylin, 4 " The material was then washed in water acidulated with HC1, then placed in a solution of glycerine, and mounted in glycerine jelly. The nucleus in Microspora is irregularly spherical or oblong in shape. It usually occupies the center of the cell.. CCENOCYTE STAINING. Hydrodictyon rettculatum (Linn.) Lagerh. — The stain used was a mixture of ammonia and borax carmine. The nuclei do not seem to be distributed uniformly throughout the coenocyte,, but most of them occupy a layer or cylinder just within the cell Hillesheim : STAINING NUCLEI OF FRESH-WATER ALG^E. 59 wall with a very few scattered about in the center. The nuclei are extremely numerous, as many as forty-six being counted in a very young coenocyte, while in the mature coeno- cytes there were many hundred. Cladophora species. — Both the borax and the ammonia car- mine stains were taken very readily by the nuclei of these plants. The nuclei, however, were brought out more clearly when hasmatoxylin was used. In a coenocyte of one species thirty-eight nuclei were counted. In another species only six could be made out. The nuclei were mostly spherical in shape. Summary. — The best fixing agent for the algae studied was chromic acid. The most successful stain was a mixture of borax and ammonia carmine. EXPLANATION OF PLATE XIV. (IN PART). All the drawings were made with the camera lucida. Figure 4. Cell of Spirogyra. Length of cell 100 mic., width of cell 50 mic. Diameter of nucleus, 12 mic., x 193. 5. Cell of Spirogyra. Shows relative size of nucleus and cell, X 450. 6. Cell of Zygnema. Nucleus in resting condition. Diameter of cell 37 mic., length 62 mic. Diameter of nucleus 7 mic., length of nucleus 12 mic., x 450. 7. Cell of Zygnema. Daughter nuclei, just after division is com- plete, x 193. 8. Hormiscia zonata (Web. and Mohr.) Aresch. Diameter of cell 20 mic., length 25 mic. Nucleus 5 mic. in diameter, x 450. 9. Microspora sp. Diameter of cell 10 mic., length 45 mic. Diameter of nucleus 5 mic., x 450. 10. Hydrodictyon reticulatum (Linn.) Lagerh. Diameter of coano- cyte ii mic., length 66 mic. Diameter of nuclei i mic., x 193. 11. Cladophora species. Diameter of ccenocyte 63 mic., length 150 mic. Diameter of nucleus 7 mic., X 450. VII. OBSERVATIONS ON DICTYOSPH^RIA. CAROLINE M. CROSBY. NOMENCLATURE AND CLASSIFICATION. The genus Dictyos-phcBria was founded by Decaisne, Valonia favulosa Ag. being chosen as the type. In further investiga- tions by Harvey, Agardh, Kiitzing and Murray, this systematic position has been accepted by all, near Valonia and Anadyomene. The present investigations have been confined to the single species Dictyosphceria favulosa^ and to material collected in the Hawaiian Islands during the summer of 1900. These notes will not attempt to discuss the classification which has been so firmly established, but will merely add some details of structure not noticed at length by Murray, and some possible explanations of certain disputed points. COLLECTION AND PRESERVATION OF MATERIAL. The material was collected in the following portions of the Hawaiian Islands, May-August, 1900: 1. Kapaa, Island of Kauai, most northern point. 2. Waianae, Island of Oahu, most southern point. In all cases the material was collected at or near low tide, in shallow water. The material used was preserved — 1. In 70 per cent, alcohol solution. 2. In 4 per cent, formaline solution. Investigations were made chiefly on formaline material. METHODS OF PREPARATION. i. The material prepared with alcohol was allowed to stand twenty minutes or so in gum-arabic solution and then trans- ferred to gum-arabic solution on the freezing chamber. As this medium necessitated transference to glycerine jelly as a mounting medium, the tissue proved too delicate for successful study. 61 62 MINNESOTA BOTANICAL STUDIES. 2. Treated most satisfactorily as follows : Formaline material first washed in a fresh 4 per cent, formaline solution was put on freezing chamber in 4 per cent, formaline solution, and transferred directly into permanent mounting medium of 4 per cent, formaline. The thallus, after repeated attempts with xylol and cedar oil as clearing media, proved too loosely constructed to cut by microtome. The Osterhout freezing method was used with the best results. Sections were cut 15 to 45 /J. thick, the first prov- ing best for detailed structure, the second for general outlines of thallus and cells. Hand sectioning proved of value only in a general way. The staining of Dictyosph&ria, en masse or in section, proved a difficult matter. The majority of the stains used had little or no value. The loose structure of the thallus could not endure hardening due to alcohol stains — therefore, water stains were used in nearly all cases. Owing to the nature of cell wall and mucilaginous contents the alcohol material was not more satis- factory than the formaline. The following stains gave poor results : 1. Methyl green (saturate solution in H2O). 2. Fuchsin (saturate solution in 50 per cent. Al). 3. Bismarck brown (saturate solution in H2O). 4. Borax carmine (almost saturate solution in H2O). 5. Ammonia carmine (almost saturate solution in H2O). Aniline -water safranine (saturate solution in H2O) for four minutes' time proved the most satisfactory stain, showing clearly structure of cell wall, needles, haptera, staining the mucilagi- nous cell contents a yellow-brown. Sulphuric Acid (very dilute). Twenty-five minutes' time dif- ferentiated clearly starch grains of pyrenoid, a dark brown. Nuclear Stains (alcohol material). DelafielcFs Hcematoxylin five minutes to twenty hours' time stained the walls a bright purple, but only stained cell contents a muddy brown. Gentian Violet (concentrated solution H2O), one to three sec- onds'time. Differentiated pyrenoid centers clearly, but proved too strong for all other structures. Acid Fuchsin (saturate solution in H2O), three hours' time, proved the best nuclear stain. Crosby : OBSERVATIONS ON DicTvospiiyERiA. 63 Mounting Media. 1. Glycerine jelly necessitated a second change from gum- arabic. 2. Pure glycerine proved too strong — drew out the stain and clouded when heated. 3. Formaline proved best. Habitat. — Dictyosphceria favulosa occurs in all tropical seas, /'. e., Hawaiian Islands, Grenada, St. Thomas, Barbadoes, Ceylon, Mauritius, Red Sea and Philippine Islands. In all cases it was found firmly attached by rhizoids to flat- tened coral reefs, as smaller, more rounded plants, or as larger somewhat appressed areas. In shallow water at low tide it grows attached to outer surface of reef or sides of hole in same. It is often mixed with or covered by other algae. GROSS ANATOMY. A typical older thallus consists of an irregular, flattened, hol- low hemisphere, with a single layer of large closely appressed, hexagonal cells, enclosing a hollow center; attached to sub- stratum by central rhizoids on lower surface {Fig. /). The thallus, in early stages somewhat bag-like, later flattens and becomes irregular in shape. Owing to the plasticity of such an undifferentiated thallus, the size and shape of both surfaces are adapted to the position ; expanded if attached to a flat surface ; more bag-like in form with wedge-shaped base, if placed between two surfaces. Later the flattening of the upper and lower surfaces, and the irregu- larity of the upper surface arise as follows : In younger solid plants the cells are of equal size. Soon those in the center enlarge and through the growth of outer cells become torn and disorganized (PL XV., Fig. 4). The hollow thus formed enlarges by the same process. The thallus lacks cohesion, gained by interlacing branches in Struvea, and is bound together by a membrane ; this now splits in all directions causing the thallus to rupture. The membrane, mentioned by Harvey, Kiitzing and Murray, as extending over the plant jbody in younger stages, was not found. The cells divide continuously and replace the torn tissue by a single-celled layer. The resultant form is irregular, expanded, with hollow center, enclosed by upper and lower surface layers, often filled with water (PL XV., Fig. 2). The outer surface of the cells is tough and membrane-like. 64 MINNESOTA BOTANICAL STUDIES. The thallus proved of interest. Murray considers it an aggre- gate of cells loosely bound by tenacula, comparable to the struc- ture of Struvea. Dr. Schmidtz, of Greifswald, considers it an irregularly branched system, equivalent to a congenital branched Cladophora, or a collection of Valonia-\fae cells. Wille, in Engler and Prantl, compares it to a " thickly branched system " senst off from a single layer of cells, which coalesce to form the typical layer. The writer would compare the plant body to a primitive, ir- regular, sessile, branched system, homologous to the elongated branched system of Struvea. Each cell may be considered a sessile detached branch, which coheres by haptera, not by incrustation. The thallus is of a higher type than Valonia, but suggests it in size and structure of cells, and is also a basal type from which still higher branched forms of Valoniaceae can be de- rived. The thallus is not encrusted. Size of thallus. — The specimens collected were small on an average. Length. Width. Depth. Average size, 15 mm. 12 mm. 15 mm. Largest size, 35 " 25 " 4 " (minus rhizoids). Smallest size, 7 " 5 " n " (with rhizoids). Color of thallus. — The thalli were of a light transparent green color, sometimes tinged with brown or pink. The rhi- zoids were vivid green in some cases. It is possible that the strong green or reddish color is due to the plant being intermixed with such forms as Halimeda or some red algae. Comparison with other Valoniacea. — Dictyosph&ria might be considered a low type because : 1. Of the primitive, closely appressed branched system. 2. Of the well-developed rhizoids. HISTOLOGY. The five- to six-sided cells on the external surface differ widely vin size, some being much enlarged and protruding (PL XV., Fig. j). The inner cells and intercellular spaces enlarge toward the center, stretching to abnormal dimensions in older plants. Cell walls. — The cell walls present a fibrillated appearance, due to a varying number of membrane-like layers (never less Crosby : OBSERVATIONS ON DICTYOSPHJERIA. 65 than seven or eight) which compose them. These vary some- what in number, have an irregular course and protrude to form haptera and needles. Chemical tests did not show the sphaero-crystals that Murray found in five Caulerfas^ but not in some Valoniacese also tested. The walls are very refractive and present a finely wrinkled appearance on the inner surface, which is not understood. Inner cell strengthening. — Murray refers to "centripetal membrane point thickenings " in six species of Caulerpa. These have been reported only in leaves of Caulerpa, rhizoids of Mar- chantia and cells of Dictyospharia. They are merely invagi- nations of approximately three-fourths of the wall stratifications, into the cavity of the cell at right angles to its depth (PI. XV., Fig. 5)- These needles are formed from the greater portion of the wall and probably softer stratifications. They are refractive, unseptate, colorless, thin-walled and with a waved -outline (PL XV., Fig. 7). The same, of similar structure and develop- ment, occur in Caulerfa, but differ in being branched many times and interlaced. The present forms are found rarely branched, with either basal (PL XV., Fig. 8) or apical dichot- omy (PI. XV., Fig. p), yet they can be considered as allied in function, and as a primitive condition of the well-developed cross beams of Caulerpa. Their development occurs as follows : From a minimum (PL XV., Fig. 6) an increasing number of stratifications invagi- nate, the inner forming the external wall of the needle. The next stratification passes within this, and this process continues until a varying number have invaginated. Thus the stratifica- tions appear as cross bars, with a lumen between, and a lumen at the base of the needle, the space between the invaginated stratifications and the remaining wall stratifications (PL XV., Figs. <5-p). The plates explain further details and size. A cross-section proves the theory. The needles occur irregularly over the entire inner surface of the cells of a mature plant, ex- cept the upper surface of external cells and the base of rhizoids, where they are absent. The younger plants possess fewer, as there is less strain. From the similarity in structure, origin and branching they can in function be allied to the strengthening structures in Caulerpa, which from its large cells needs both the branching interlaced needles and interlaced branches of thallus. 66 MINNESOTA BOTANICAL STUDIES. Of Valoniaceae exposed to like conditions of wind and wave, Dictyosph&ria needs more firmness. Valonia thalli, on ac- count of their form and structure, need no support, and the re- maining Valoniaceae gain sufficient cohesion through interlacing branches. The presence of the needles may be due to the loose struc- ture of the Dictyosphceria thallus, or to the necessity of having an internal balance to the haptera. There is no stimulus to growth from direct contact, as in haptera, and these may arise from strain on older thallus. Whatever the function, it is sub- sidiary to that of the haptera, as they are less numerous, and chiefly in greater numbers in central cells. External cell strengthening. — Haptera or intercellular organs of attachment are present in Udotea, Boodlea, Microdictyon and Spongocladia, and bind one part of the thallus to another, as in Struvea, where they fasten pinna to pinna, or one cell to another as 'in Dictyosphceria. In all cases they bind a thallus of loose structure together. Origin. — The origin of the haptera is due to the evagination of about one third of the cell wall, similar to the invagination in the case of the needles. Their primary importance as compared with the needles is perceived, for they are never absent, and no young stages of development are present. They are, however, formed through a similar process, i. e., the evagination of the stratifications. The cross beams, caused by stratifications, are nearer their tips, thus leaving a larger lumen (PL XV., Fig* 10). Optical sections near the base appear as dark rings from one to several in number, due to the number of main branches of haptera which are present (PL XV., Fig. //). The haptera are hollow and have no contents. Development. — The evagination continues until a surface is reached to give the needed stimulus. At this stage the haptere consists of an unbranched tube ending in a closed blunt end PL XV., Fig. /j). The tube or stalk now begins to lobe di- chotomously, and the ends flatten out upon the wall. This continues until a branched circle of lobes is formed, convex and radiating (PL XV., Fig. 10). The hollow space thus formed between the opposite cell wall and the concave center causes adhesion by sucking (PL XV., Fig. 14). The base shows from one to three enclosed oval rings, due to the number of main Crosby : OBSERVATIONS ON DICTYOSPHLERIA. 67 branches (PL XV., Fig. //). The main tubes may develop lobes directly (PL XV., Fig. //), or may become branched from one to three times in various directions and levels (PL XV., Figs, ii and 12). Each branch then develops (PL XV., Fig. n) a separate system of radiating lobes, as seen in cross- section, central view (PL XV., Fig. 12). As before, the con- ditions determine size and shape. The haptera are absent from the outer walls of external cells of the thallus, but are abundant elsewhere, and often crowded when developed from the larger cells (PL XV., Fig. 11). Near the exterior, the closely con- nected cells cause the haptera to be short-stalked, and after the opposite wall is reached, continued branching occurs over a varying area, limited only by contact with neighboring haptera. In the central cells, longer stalks arise from the separated cells. In the intercellular spaces their length is often much greater, induced by the greater space (PL XV., Fig. if). A haptere near the edge of the intercellular space is often two or three times branched, and clasps the surfaces in various directions, to meet the added strain at this position (PL XV., Fig. //). The haptera generally extend directly to the opposite cell wall and thus the base from cell I alternates with the lobes of cell 2, but they also extend diagonally and at different levels (PL XV., Fig. //). The numbers, size of the haptera, length and direc- tion of tubes, number of branches, area of adhesion and posi- tion, depend on the distance between the cells. Rhizoids.— -T\\t rhizoids, centrally situated, are elongated, unicellular structures, and are developed from the ventral sur- face of the thallus. They show little differentiation and corre- spond to the normal plant cells. They function as primitive holdfasts, attached to the underlying surface of coral, and are thallus cells, enlarged, elongated, irregularly shaped, and, rarely, budded. To strengthen attachment to substratum, haptera, similar to those above described, are formed from the outer edge of rhizoids. These are few in number (PL XV., Fig. 75).' The relation between the strength' of the rhizoids and the position of the plant is intimate, their function being aided by secondary structures, the haptera. The color is generally a strong green, rarely reddish. The cell wall in form and structure is similar to that of thal- lus cells. In arrangement the rhizoids are scattered or massed together. In size they vary greatly, the longest five mm. by 68 MINNESOTA BOTANICAL STUDIES. one mm., the shortest of barely appreciable length by one half mm. Here as elsewhere the organs are but slightly differen- tiated and vary in size, structure, and number, according to external conditions (PL XV., Figs. 15 and 16). CELL CONTENTS. Endochrome. — The peripheral layer of cell protoplasm con- sists of dove-tailed polygonal chromatophores, plate-like and distinctly separated by colorless thread-like lines. This single layer of thin wall plates in outer cells of thallus, is dense, stains deeply, and forms a compact unbroken layer. The attachment of this to the cell wall does not appear a close one as the layer becomes easily detached, and floats separately in water (PL XV., Fig. 77). The layer (PL XV., Fig. 77) becomes irregu- larly perforated, less solid and finally (PL XV., Fig. 18) in inner cells, consists of widely separated chromatophores, joined by numerous thin granular threads. The endochrome does not project into the cell cavity. In central cells, few chromatophores are present. One kind of chromatophore only can be distinguished, though the size and shape vary somewhat. Pyrenoid. — Centrally placed within each chromatophore, is an irregularly spherical body, with thicker walls, and stronger refraction (PL XV., Fig. 18). Fig. 18 shows in section the thickened wall and hollow center, but does not show the central grain or clefts which are present. The development of the pyrenoid can be clearly traced. In early stages (PL XV., Fig. 19) it is more spherical and solid, except for the beginning of a cleft from the outer edge, which cuts to the central grain. This latter enlarges and becomes more irregular in outline, later. More divisions occur, in one plane only, and four to six lobes are formed, all converging to the inner grain, beyond whose outer edge the cleft does not continue. This latter is spherical, thick-walled, refractive, and contains a minute central grain (PL XV., Fig. 20). Acid fuchsin differ- entiates the pyrenoid and accompanying starch grains clearly. Starch grains. — A weakened iodine solution give this char- acteristic starch reaction. The starch grains show concentric layering and vary in shape from oval and spherical to irregular shapes. They are scattered in older stages near the pyrenoids or throughout the chromatophore (PL XV., Fig. 20). In Crosby: OBSERVATIONS ON DICTYOSPH^:RIA. 6£P younger plants they are formed against and within the pyrenoid (PI. XV., Fig. /p), from which they seem to have origin, and from wrhich they gradually move, until eventually the majority lie beyond or against the pyrenoid. A small chromatophore generally has from one to three starch grains a larger one five to eight, or even more. Oil drops. — The large oil drops are scattered irregularly throughout the endochrome, abundant in number. They are differentiated from other protoplasmic structures by large size, lighter color, regular outline, and stronger refraction. Cell sap. — The cell is entirely filled with a large amount of watery colorless fluid, whose composition was not investi- gated. BIBLIOGRAPHY. Murray, Geo. On Boodlea, a new Genus of Siphonocladaceae. Jour. Linn. Soc. Bot. 25: 243. 1890. Murray, Geo., and Boodle, L. A. A Structural and Systematic At count of the genus Struvea. Ann. Bot. 2 : 265-282. pi. 16. f. 1-8. 1888-1889. Murray, Geo. On the Structure of Dictyosphczria Dene. Phyco- logical Memoirs, Part i : 16-20. pi. 6. 1892. Osterhout, W. J. Van L. Simple Freezing Device. Bot. Gaz. 21 : 195—200. 1896. Agardh, J. G. Till Algernes Systematik. VIII. 113-119. 1886. Harvey, W. H. Phycologia Australia*, pi. 2.f. 4. 1858. Harvey, W. H. Ner. Bor. Amer. P. 7. 1858. Harvey, W. H. Nov. Bor. Am. Pt. Ill: 50. 1858. Wille, N., in Engler, A., and Prantl, K. Die Naturlichen Pflanzen- familien, teil. I : 150. 1897. DeToni, Dr. J. B. Sylloge Algarum, I : 371. 1889. Kiitzing, F. T. Tab. Phyc. 7= 10. pi. 25. 1856. Murray, Geo. Introduction to the Study of Seaweeds, 159. 1895. Rosenvinge, L. K. Sur quelques phenomeries de croissance chez les Cladophore et Chatomorpha. Bot. Tid. 18 : 59-64. 1888-1893. Correns, C. Ueber die Membran von Caulerpa. Ber. Deut. Bot Gesell. 12: 355-367. Taf. 23. 1894. Fairchild, D. G. Ein Beitrag zur Kenntniss der Kerntheilung bei Valonia utricularis. Ber. Deut. Bot. Gesell. 12: 331-338. Taf. 21. 1894. Schmitz, F. Ueber die Zellkerne der Thallophyten. Sitz. d. nied. Ges. f. Natur. und Heilk. 10. 1880. Schmitz, F. Beitr. z. Kerint. der Chromatophoren. Pringsh. Jahrb. f. wiss. Bot. 15: 2-177. pl- 7- 70 MINNESOTA BOTANICAL STUDIES. Stromfelt, H. F. G. Untersuchungen iiber die Haftorgane der Algen. Bot. Centralbl. 33: 381. 1888. EXPLANATION TO PLATE XV. Figure i. Form of typical thallus, two fifths natural size. 2. Irregular form of older thallus caused by rupturing of membrane, two fifths natural size. 3. Thallus showing irregularity of upper surface cells, two fifths natural size. 4. Cross section of thallus showing closely appressed outer cells and increasing separation of inner cells, X 168. 5. Cross section showing interior needles of cell, x 45. 6. Cross section showing early development of needle, x 488. 7. Origin of needles, X 488. 8. Basal dichotomous branching of needle, x 488. 9. Apical dichotomous branching of needle, x 488. 10. Origin of haptera ; development of two main branches; surface extension of lobes and cross beams, X 212. 11. Diagram of intercellular space showing oval rings representing main branches ; wide and narrow expansion of lobes; branching in various directions and at different levels : stalks or main branches of various lengths. 12. Cross section of three main branches, ventral and hollow stalks, X 464. 13. Youngest development of stalk of haptere, x 424. 14. Concave center of haptere from ventral view, x 360. 15. Rhizoid on ventral surface with haptere, x 45. 1 6. Youngest development of foot on ventral surface, x 45. 17. Peripheral protoplasmic layer of chromatophores, X 35. 18. Widely separated chromatophores of inner cell, x 318. 19. Early development of pyrenoid with central grain and forma- tion of first starch grains, X 760. 20. Older stage of pyrenoid ; scattered starch grains and central grain, x 760. VOL. III. MINNESOTA BOTANICAL STUDIES PART I. « 20 16 PLATE XV. H ELIOT YPE CO., BOSTON. p VIII. STAPFIA CYLINDRICA IN MINNESOTA. CHARLES J. BRAND. The material on which these observations are based was col- lected by the writer during August, 1901, in the harbor of Grand Marais on the north shore of Lake Superior. It has also been observed at Tobin and Washington harbors on Isle Royale, Michigan. At Grand Marais the plant was found growing in the water, attached to the smooth diabasic rock, at a depth varying from six inches to eight feet. It was most abundant in a small arm of the harbor which is enclosed on two sides by a reef and on a third by a dock crib. The situation is one that does not ordi- narily require any particular ability or adaptation for resisting very violent wave action. However, there are times when the plant is compelled to undergo considerable strain. In the event of a strong wind shoreward the seas break over the outer protecting barrier reef and cause a very strong current from the small arm into the harbor proper and also a fairly violent wave action. The water in which the collection was made is very fresh and cold, in fact it is the drinking water of the villagers residing about the harbor. It seemed at first that the plant was simply a species of Tetra- spora, but a careful examination leaves no doubt but that it be- longs to the genus Stapjia, established by Chodat in 1897. In the present form the position of the thallus in its usual con- dition of growth is always upright. The mode of attachment is by means of a distinct holdfast. Just above the holdfast there is a short attenuated area, beyond which the thallus assumes its ordinary diameter. This feature can readily be observed by reference to the figure of the entire plant (PL XVI., Figs. i, 2,3\ An examination of the thallus with the dissecting microscope or even without a lens, reveals a much wrinkled and folded sur- 71 72 MINNESOTA BOTANICAL STUDIES. face. The color is uniformly a very dilute green. The gelati- nous thallus is firm and slippery. A species of Bulboch&te was found very commonly growing as an epiphyte on the plant, being attached by a sort of sub- spherical cell imbedded in the gelatinous mass of the Staffia thallus. Diatoms and desmids were also found in g.reat num- bers in the interior of the thallus. The material used for this study was preserved in 2 per cent, formaline and as a consequence was rather unfavorable for cyto- logical investigation. After washing the material for about thirty-six hours in water, it was passed through the usual series of alcohols and xylols into paraffine. The plants were permitted to remain in each for a short time only, as they were very prone to grow hard and brittle, especially if left too long in the higher per cents, of alcohol or in xylol. The material was suitably im- bedded for securing both longitudinal and transverse sections. The sections were mounted in series and cleared in the usual manner. After trying numerous stains, it was found that gentian violet and Bismarck brown were the most useful. A concentrated aque- ous solution of the Bismarck brown was used and the mounted sections were allowed to remain in this for about two hours. Only a 2 per cent, aqueous solution of the gentian violet was used and the slides were left in the stain for about three minutes. Canada balsam was used as the mounting medium. The cross-sections disclosed some very interesting foldings of the gelatinous membrane, which cause the perforate appearance seen in the diagrammatic sketch of the transverse section (PL XVI., Fig. 4) and also in the longitudinal section (PL XVI.y Fig. <5). The cross-section 'reveals no particular method in the grouping of the cells in the gelatinous structure. The cells are distributed in a single peripheral layer. They appear to be in groups of two and four for the most part, but also in threes and singly in the older thalli. The longitudinal section (PL XVI., Fig. 7) does not differ in essential features from the transverse, although the folds are much more clearly distinguishable. This cut also shows clearly the sort of alveolar structure of the interior of the thallus, of which I have been unable to find any mention in the descrip- tions of either Stapfia or Telraspora. This may be the adapta- Brand: STAPFIA CYLINDRICA IN MINNESOTA. 73 lion by which buoyancy is secured and the ordinary upright position of the thallus is made possible. The structure may be produced by the degeneration of the gelatin of the interior. The plant very strongly resembles Enteromorpha intestinalis of marine waters in its pale green color, subtubular thallus, variability in size and attenuate base. The plants of Nordstedt, Wittrock and Lagerheim, no. 1362, distributed as Tetraspora cylindrica (Wahlenb.) Ag. f. enter o- morphoides Lagerh. nov. form, which, according to Chodat, no doubt belongs to the genus Stapfia, resembles the Lake Superior form considerably, though there are some dissimilarities as may be seen by a comparison of the two. The thallus of the former is fistulose, while that of the Lake Superior plant is subfistulose or alveolar. They agree in that both are verrucose, but dis- agree in that the former may sometimes be ramose, the latter never. The former is described by Lagerheim as "fragili," while the latter is firm and quite tough. They differ also in color, the former being of a much darker green. The Lake Superior form, however, is darker in color in quiet, less fresh, water. The former has rather short thalli in comparison to the diameter, while the Lake Superior plant has a thallus long and of relatively small diameter. The former is much like the larger of the type specimens distributed by Stapf as Stapfia cylindrica, while the latter resemble very much the slender and relatively much longer specimens of the same distribution. There is also the great difference in habitat, the one being found in the largest of the fresh-water lakes and the other in a swift-flowing alpine stream of northern Norway. Exteriorly the plant resembles most strongly the specimens distributed by Rabenhorst as no. 2244, Tetraspora cylindrica. There is, however, a greater variance in length. The shortest of my specimens are about as long as the longest of Rabenhorst, while the longest are nearly as long as Wittrock and Nordstedt's Tetraspora cylindrica forma rivularis. The former were col- lected on rocks in Lake Wettern, near Jonkoping, Sweden, by Nordstedt, and the latter were also collected by him near Kongs- vold, Norway, in the river Driva. The Lake Superior form may be described as follows : Thallus 2-5 mm. in diameter, 6 cm. to 3 dec. in length, dilute green in color, erect, gelatinous, cylindrical, • subfistulose, sometimes rugose-verrucose, unbranched, firm, rather tough, 74 MINNESOTA BOTANICAL STUDIES. clavate at distal end, suddenly attenuate at the base into a brief stipe ; holdfast disc-shaped ; cells spherical, solitary or in groups of two, four or rarely three, 4-16 mic. in diameter. Attached to rocks at lake bottom. BIBLIOGRAPHY. Chodat, R. Bull. 1'Herb. Boiss. 5: 939. pi. 23. 1897. Chodat, R. Algues vertes de la Suisse, 112. pi. 49, 52. 1902. EXPLANATION OF PLATE XVI. (IN PART). Figures i, 2, 3. Plants one half natural size. 4. Diagram of transverse section. 5. Cross-section of thallus, X 235. 6. Longitudinal section of thallus showing fold, X 120. 7. Longitudinal section of thallus, x 235. VOL. III. MINNESOTA BOTANICAL STUDIES. PART I. PLATE XVI. HELIOTYPE CO., BO8TON. p IX. OBSERVATIONS ON SOME CALCARE- OUS PEBBLES. CHALMER POWELL. The first account of pebbles formed by algae in the United States is given by Dr. George Murray (i). The pebbles ex- amined by him were found in eight feet of water on the sandy bottom of a Michigan pond separated from Lake Michigan by a sand-bar. The specimens varied in size from one to three and a half inches in diameter, were hollow and showed a strati- fied or concentrically zoned structure. Upon decalcification they were found to be composed of a densely interwoven mass of fila- ments. The predominating kind was a species of Schizothrix, S . fasciculata Gomont. There were also filaments of Stigonema and Dichothrix and a large number and variety of diatoms. Mr. Thiselton-Dyer (2) refers to the occurrence of pebbles on the bottom of Lough Belvedere, near Mullingar, which were "of all sizes up to that of a filbert." The bulk of the algal mass consisted of a Rivularia. The first mention of calcareous pebbles in American literature was made by Professor Conway MacMillan (3), who states that although he had " not yet found any of these algal pebbles in lakes of Minnesota, it is probable that they occur." This prediction was realized in June, 1901. In describing pebbles found along the shore of Littlefield lake, Michigan, Mr. Charles A. Davis (4) states that they are "the result of the development and growth of an alga, Zono- trichia, or a nearly related species. The vegetable origin of these pebbles would not be suspected until one recently taken from the water is broken open, when it is found to show a radial structure of bluish-green lines." In a preceding article the same author (5) describes a blue-green alga concerned in the formation of marl, which had been determined to be a species of Zonotrichia or some closely related genus. "The plant grows in relatively long filaments formed by cells grow- 75 76 MINNESOTA BOTANICAL STUDIES. ing end to end, and as they grow the filaments become encased in calcareous sheaths. The feature of the plant which makes it important in this discussion, however, is its habit of growing in masses or colonies. The colony seems to start at some point of attachment or on some object like a shell and to grow out- ward radially in all directions, each filament independent of all others and all precipitating calcium carbonate tubules. The tubules are strong enough to serve as points of attachment for other plants, and these add themselves to the little spheroid and entangle particles of solid matter, which in turn are held by new growths of the lime-precipitating Zonotrichia and thus a pebble of greater or less size is formed, which, to the casual observer, is in no wise different from an ordinary water-rounded pebble. These algal calcareous pebbles show both radial and concentric structure and might well be taken for concretions formed by rolling some sticky substance over and over in the wet marl on which they occur, but for the fact that a considerable number of them show eccentric radial arrangement and that the shells of accretion are likewise much thicker on one side than on the other, and finally, because the side which rests on the bottom is usually imperfect and much less compact than the others. The pebbles are characteristically ellipsoidal in shape. The radial lines, noticeable in cross-sections of the pebbles, are considered by the writer to be formed by the growth of the filaments, while the concentric lines probably represent periods of growth of the plants, either seasonal or annual." Other forms than the Zono- trichia were found in the pebbles. In June, 1901, Messrs. Freeman and Lyon, of the Botanical Department of the University of Minnesota, found some cal- careous pebbles in Clearwater lake, Wright county, Minnesota. This lake covers an area of about four square miles and is really two lakes connected by a narrow strait. The pebbles were collected from the southern arm. They were found lying in from four to ten feet of clear water on sand-bars, which rose abruptly towards the surface and at their edges sloped almost perpendicularly into deep water. These pebbles range in size from that of a small hickory nut to two inches in diameter. Most of them are flattened, and, though comparatively smooth in some cases, are often rough, coagulated and wave-worn. All are more or less hollow. In section they have a distinctly stratified appearance. The theory Powell: OBSERVATIONS ON SOME CALCAREOUS PEBBLES. 77 of Mr. Davis given above is the most probable one for their formation. A study was made of the pebbles which had been preserved in 5 per cent, formalin. Sections cut with a sharp scalpel were placed in a weak solution of hydrochloric acid until decalcified, after which they were immediately mounted in glycerine jelly. The pebbles were found to be composed of a densely interwoven mass of filaments of which the most common type was those of Schizothrix fasciculata Gomont. The trichomes were 1.4 to 3 mic. broad. The pseudocysts were equal in length to diameter of trichome or twice as long, 1.2 to 3.5 mic. long. A species of Calothrix, two species of Cosmarium, a Nostoc and numer- ous diatoms were also present in the pebbles. In June, 1901, Miss Daisy Hone collected some small peb- bles at Point Douglas, Minn., on a steep bluff side overlooking the Mississippi river. Being high above the water they were not supplied with moisture and seemed perfectly dry at the time of collection. These pebbles had an outside layer of calcareous material, which, when decalcified, showed the presence of an alga, a species of Scytonema. BIBLIOGRAPHY. 1. Murray, G. Phyc. Mem. 74-77. 1895. 2. Thiselton-Dyer, W. T. Ann. Bot. 5: 225. 1890-91. 3. MacMillan, C. Minn. Plant Life, 41. 1899. 4. Davis, C. A. Journ. Geol. 8: 502. 1900. 5. Davis, C. A. Journ. Geol. 8: 495. 1900. EXPLANATION OF PLATES. PLATE XVI. (IN PART). Figure 8. A section of a pebble showing stratified layers, nat. size. 9. A filament of Schizothrix^ x 600. 10. Calothrix sp. Detail of an older filament, X 335. 11. Same. Portion of thallus showing method of branching, x 64. 12. Same. Young free filaments, X 200. PLATE XVII. Figures 1-5. Calcareous pebbles, one half natural size. 1. Pebble showing smooth surface. 2. One showing a rougher surface. 3. One showing a still rougher and more porous surface:, probably wave-worn. 4 and ^. Hollow interiors of two pebbles are shown. VOL. III. MINNESOTA BOTANICAL STUDIES PART I. PLATE XVII. HELIOTYPE CO., BO3TON. r X. NITELLA BATRACHOSPERMA IN MINNESOTA. GENE LILLEY. The first record of Nitella batrachosperma was made in 1833 by Reichenbach who called it Chara batrachosperma. In 1847 A. Braun (i) placed it with the Nitellae. The plant is quite widely distributed in Europe. Migula states that it has been reported from Germany, Switzerland, Sweden, Finland, France, Austria, Italy, Spain and that also it has been reported from Australia and North America. During the summer of 1898 A. J. Pieters, Assistant Botanist of the De- partment of Agriculture, collected Nitella batrachosperma at East Harbor, Ohio. He describes the habitat as follows : " At East Harbor there is a wide stretch of swamp intersected by channels which open into the lake by one deep and- narrow channel protected from severe wave action by a sand-bar. . . A short distance from the entrance, the channel divides, one branch going east, the other west. . . . Just where the channel turns toward the east is a sandy beach covered with two feet or less of water, and here grow several species of Characeae, which are more abundant here than elsewhere in the swamp. Nitella tenuissima and N. batrachosperma grow in about one foot of water with their branches spread out flat on the sand." Nitella batrachosperma was collected in Minnesota by the writer in August, 1901, at Pike lake, a small shallow lake twelve miles west of Duluth. The plants grew on a sandy beach where the water was from three to six inches deep. The east end of the lake is the only part where the beach is sandy, and although the shore around the lake was examined carefully it was only at this one place that the plants were seen. ' Chara coronata Ziz. was found with N. batrachosperma. N. tenuis- sima Desv., which is usually reported with N. batrachosperma ^ was not found at this place. The collection was made in the morning when the sun's rays struck the water at such an angle as to light up the sandy bottom, so that the plant could be easily distinguished from Chara coronata. The previous afternoon 79 80 MINNESOTA BOTANICAL STUDIES. the plant had been overlooked as its dirty brown or gray color and its low prostrate habit gave it the appearance of debris. Very few plants were found. Description of plant. — Nitella batrachosperma is of a dirty brownish-green color, very small, from 1.8 cm. to 3 cm. high. Most of the plants examined were 2 cm. high. There are from two to four main branches springing from the first node. The branches measure .5 mm. in diameter and from 4-7 mm. to the first whorl of leaves. This whorl of leaves consists of six to eight sterile leaves, having two divisions. The first division or main ray is from 55 to 67.5 mic. wide and 685.5 to 75° m^c- long. The leaflets are about the same length and are two- celled, the basal cell being very long, the end cell short and sharp-pointed. The lower sterile leaves are loose and spread- ing and stand at nearly right angles to the main branch. At the tip of the branch the leaves are thick and compressed, giving a bushy-like appearance to the plant. At the apex both sterile and fertile leaves have from two to three divisions. In a fertile leaf, having three leaflets divided and three undivided, the main ray is 50 mic. broad and 312 mic. long. In a sterile leaf with three leaflets divided and three undivided the diameter of the main ray is 62.5 mic. and the length 312.5 mic. In a sterile leaf having three divisions throughout, the diameter is 80 mic. and the. length 250 mic., showing little difference in size of fertile and sterile leaves. The fertile leaves are all borne at the tip of the branch and but lew leaves at the tip are sterile. Rhizoids. — Rhizoids arise from the first and second nodes of the plant. The rhizoids from the first node are much larger than those from the second, being 125 mic. in diameter, while those from the upper are about 62 mic. wide. The rhizoids have long cells and abut upon one another in the characteristic " stocking-foot" manner of the Characeas. Reproduction . — Nitella batrachosperma is monoecious . The antheridia and oogonia are usually borne on the same leaf, though not always. Many of the leaves bear only antheridia, few bear only oogonia. The organs can be distinguished only by careful examination as they are minute and the color, a yellowish-brown, differs but little from the color of the plant. Anther idium. — The antheridia are borne terminally upon the first segment of the leaf. The second and third divisions do not bear antheridia or oogonia. The development of the an- Lillcy : NITELLA BATRACIIOSPERMA IN MINNESOTA. 81- theridium is very similar to that of Nitella flcxilis. It can be distinguished very early in the growing point by its position and size. It is at first a large spherical cell having a large nucleus. The first division after the stalk cell is cut off is vertical (PL XVIII., Fig. 16]. The second division is transverse (PI. XVIII., Fig. //). The unfolding of the walls is seen in the sixteen-celled stage (PI. XVIII., Fig. 12). The shields in the mature state are hyaline and contain few chlorophyll bodies and little coloring matter, so that the contents can be easily made out in optical section. Mature antheridia are 135 by 210 mic. in diameter. Oogonium. — The oogonium arises at the base of the anthe- ridium and takes the place of a leaflet. But one oogonium is borne on a leaf. All stages of development can be found on the same plant. Two " Wendung-zellen " were observed in N. batra- chosperma (PL XVIII., Figs, j and 8]. The filaments sur- rounding the egg and crown cells are colorless and hyaline. The oosperm is almost spherical (270 by 300 mic.). It is of a rich, golden brown color. BIBLIOGRAPHY. Migula, W. Die Characeen, in Rabenhorst, Kryptogamen Flora, 184. 1897. Pieters, A. J. The Plants of western Lake Erie. U. S. Fish Comm. Bull. 64, 78. 1901. DESCRIPTION OF PLATE XVIII. Figure I. Photograph of Nitella batrachosperma. Twice natural size. 2. Rhizoid showing " stocking- foot " cells with branches, X 300. 3. Outline drawing of fertile leaf showing tip of first division bear- ing antheridium and oogonium, x 300. 4. Apex of leaf bearing antheridium, oogonium and leaflet, x 150. 5. Apical cell of branch showing three nodal cells, X 300. 6. Apical cell of branch showing nodal cells and last internodal cell, x 300. 7. Young oogonium showing first Wendung-zelle and first di- vision of crown cells, x 300. 8. Young oogonium showing two Wendung-zellen and two cells of the crown, x 300. 9. Membrane of oogonium, much magnified. 82 MINNESOTA BOTANICAL STUDIES. 10. Young antheridium showing stalk cell cut off and first division of antheridium, x 300. n. Young antheridium in four-celled stage, x 300. 12. Young antheridium in sixteen-celled stage, with young leaflets, X 300. 13. Young antheridium with shields, X 300. MINNESOTA BOTANICAL STUDIES. PART I. 10 11 13 PLATE XVIII. HELIOTVPE CO., BOSTON. r XL CATALOG OF MINNESOTA GRASSES, W. A. WHEELER. The following catalog of the grasses of Minnesota is a par- tial report of the work done on the Minnesota Botanical Survey during 1902. Upham's catalog of the Flora of Minnesota, published in 1884, lists one hundred and thirty-nine species and varieties as occurring within the limits of the state. In the Metaspermae of the Minnesota Valley, published in 1892, Professor Conway MacMillan lists ninety species and varieties indigenous to the drainage basin of the Minnesota river. Other local reports con- tain lists of Minnesota grasses. Most of these lists, however, have been incorporated in the two reports just cited. This catalog of one hundred and seventy-eight species and varieties is based mainly upon a redetermination by the writer of all the specimens of Minnesota grasses in the Herbarium of the University of Minnesota. One hundred and thirty-eight species and varieties have been so determined. The reports of the re- maining forty species listed have been taken from previous publications without examination of specimens. Seven of these have been cancelled by corrections in the determination of the specimens cited. Probably others of the forty so listed have also been reported without sufficient evidence or upon incorrect determination, but the specimens are not at hand for compari- son. These have, however, all been listed, and are accompa- nied by notes stating upon what the report is based. Twenty-four species not previously reported from the state are listed in this catalog. Some of these were determined from lately-published descriptions, without specimens for compari- son, and are therefore somewhat doubtful. This refers espe- cially to the species of Panicum. Some specimens of Elymus remain undetermined because of insufficient material of recently published species for comparison. Synonyms are given where confusion might arise in a com- parison with other Minnesota reports. Herbarium specimens 83 84 MINNESOTA BOTANICAL STUDIES. are cited only when the species is poorly represented by speci- mens in the Herbarium of the University. Andropogon scoparius MICHX. Fl. Bor. Am. i: 57. 1803. Beard grass. Common in dry soil throughout. Herb. : Specimens from all parts of the state. Andropogon furcatus MUHL. Willd. Sp. PI. 4: 919. 1806. Blue stem. Andropogon -provincialis furcatus HACK, in DC. Mon. Phan. 6 : 442. 1889. Common in dry soil throughout. Herb. : Specimens from all parts of the state. Sorghastrum avenaceum (Micnx.) Nash, in Britton, Man. Fl. U. S. and Can. 71. 1901. Indian grass. Chryso-pogon avenaceus BENTH. Journ. Linn. Soc. 19: 73. 1881. Common throughout. Herb. : Numerous specimens. Syntherisma linearis (KROCK.) NASH, Bull. Torr. Club, 22 : 420. 1895. Small crab-grass. Panicum glabrum GAND. Agrost. i: 22. 1811. Waste and cultivated ground throughout. Herb.: Ballard 1168, Goodhue county; Aiton, Hennepin county ; Frost 380, Kandiyohi county ; Sandberg, Hennepin county; Campbell 179, St. Cloud. Syntherisma sanguinalis (L.) NASH, Bull. Torr. Club, 22 : 42. 1895. Large crab-grass. Panicum sanguinale L. Sp. PL 57. 1753. Waste and cultivated ground. Herb. : Aiton, Minneapolis ; Wheeler, Houston county. Echinochloa crus-galli (L.) BEAUV. 1. c. Barnyard grass. Panicum crus-galli L. Sp. PL 56. 1753. Common as a weed throughout. Herb. : Numerous specimens. Echinochloa walteri (PURSH)NASH, in Britton, Man. FL U. S. and Can. 78. 1901. Cockspur-grass. Panicum crus-galli hispidum TORR. Fl. N. Y. 2 : 424. 1843- Panicum walteri PURSH, Fl. Am. Sept. I : 66. 1814 Wheeler: CATALOG OF MINNESOTA GRASSES. 85 In similar locations to E. crus-galli (L.) Beauv., but not so common. Herb. : A. M. Johnson, Hennepin county ; Wheeler, Forest Lake; Aiton, Lake Waseca. Panicum capillare L. Sp. PI. 58. 1753. Witch grass. Common in dry soil throughout. Herb. : Numerous specimens. Panicum cognatum SCHULTES, R. & S. Syst. 2: 235. Diffuse panicum. Panicum autumnale Bosc. Spreng. Syst. I : 320. 1825. Reported from Minnesota in Upham's catalog, but not re- presented by specimens in the University herbarium. Panicum virgatum L. Sp. PI. 59. 1753. Tall smooth panicum. Common throughout. Herb. : Numerous specimens. Panicum agrostoides SPRENG. Pugill. 2 : 4. Agrostis-like panicum. Reported from Minnesota in Upham's catalog and in Mac- Millan's Metaspermag of the Minnesota Valley, but not repre- sented by specimens in the University herbarium. Panicum depauperatum MUHL. Gram. 112. 1817. Starved panicum. Reported from Minnesota by Upham and MacMillan and probably occurs here. Perhaps confused with the two follow- ing species. Panicum linearifolium SCRIBN. in Br. & Br. Illus. Fl. App. 3: 500. 1898. Narrow-leaved panicum. Dry soil, south. The determination of the specimens cited under this and the two following species were made entirely from descriptions without comparison with authentic determinations and are there- fore somewhat doubtful. Herb.: Rosendahl 259, Spring Grove; Holzinger, Winona. Panicum perlongum NASH, Bull. Torr. Club, 26: 575. 1899. Elongated panicum. Dry soil, south. Herb. : Sheldon, Mille Lacs Indian Reservation, 2686, Brainerd ; Ballard 1010, Nicollet county ; Sandberg, Henne- pin county; Wheeler 1192, Winona. 86 MINNESOTA BOTANICAL STUDIES. Panicum werneri SCRIBN. in Br. & Br. Illus. Fl. App. 3: 501. 1898. Werner's panicum. Not previously reported from Minnesota. Herb. : Oestlund, Hennepin county (?). Panicum angustifolium ELL. Bot. S. C. & Ga. i : 129. 1817. Taller narrow-leaved panicum. Panicum consanguineum S. WATS, in A. Gray, Man. Ed. 6, 633, in part. 1890. Reported from Minnesota in Upham's catalog and MacMil- lan's Metaspermge of the Minnesota Valley, but not represented by specimens in the University herbarium. Panicum dichotomum L. Sp. PI. 58. 1753. Forked panicum. The previous reports of this probably refer to some other species, as this species is not known to occur in our range. Panicum boreale NASH, Bull. Torr. Club, 22: 421. 1895. Northern panicum. In moist soil. Not previously reported from Minnesota. Herb. : Sandberg, Hennepin county. Panicum implicatum SCRIBN. in Br. & Br. Illus. Fl. App. 3: 498. 1898. Hairy-panicled panicum. In dry soil throughout (?). Herb. : Leiberg 2103, Blue Earth county (?.); Aiton, Lake Itasca (?). Panicum unciphyllum TRIN. Gram. Panic. 242. Hairy pan- icum. Panicum -pubescens A. GRAY. Common in dry soil throughout. Herb. : Numerous specimens. Panicum leibergii (VASEY) SCRIBN.; Vasey, U. S. Dept. Agric., Div. Bot. Bull. 8: 32. 1889. Leiberg's pani- cum. Common in dry soil south. Herb. : Numerous specimens. Panicum scribnerianum NASH, Bull. Torr. Club, 22: 421. 1895. Scribner's panicum. Panicum -paucijlorum A. GRAY, Man. 613. 1848. Common in dry soil south. Herb. : Specimens collected throughout southern Minne- sota. Wheeler: CATALOG OF MINNESOTA GRASSES. 87 Panicum xanthophysum A. GRAY, Ann. Lye. N. Y. 3 : 233. 1835. Slender panicum. Dry soil throughout. Herb. : Aiton, Lake Itasca ; Sheldon, Mille Lacs Indian Reservation, 2831 Kanabec county; Schuette, St. Anthony Park ; Ballard 1419, Cass county. Panicum porterianum NASH, Bull. Torr. Club, 22 : 420. 1895. Broad-leaved panicum. Panicum latifolium WALT. Fl. Car. 73. 1788. Common in woods south. Herb. : Rosendahl 210 and 487, Wheeler 388, Houston county; Oestlund, Sandberg, Aiton, Hennepin county; Ballard 487, Scott county; Taylor 1599, Lindstrom ; Hvoslef, Lanes- boro. Chaetochloa verticillata (L.) SCRIBN. U. S. Dept. Agric., Div. Agrost. Bull. 4: 39. 1897. Fox-tail grass. Setaria verticillata BEAUV. Agrost. 51. 1812. Reported in Upham's catalog but not represented by speci- mens. Chaetochloa glauca (L.) SCRIBN. U. S. Dept. Agric., Div. Agrost. Bull. 4: 39. 1897. Yellow pigeon-grass. Setaria glauca BEAUV. Agrost. 51. 1812. Abundant as a weed throughout. Herb. : Numerous specimens. Chaetochloa viridis (L.) SCRIBN. U. S. Dept. Agric., Div. Agrost. Bull. 4: 39. 1897. Green pigeon-grass. Setaria viridis BEAUV. Agrost. 51. 1812. Abundant as a weed throughout. Herb. : Numerous specimens. Chaetochloa italica (L.) SCRIBN. U. S. Dept. Agric., Div. Agrost. Bull. 4 : 39. 1897. Hungarian millet. Setaria italica R. & S. Syst. 2 : 493. 1817. Locally escaped from cultivation. Herb. : Foote, Worthington ; Sandberg, Hennepin county. Cenchrus tribuloides L. Sp. PI. 1050. 1753. Sand-bur, Bur- grass. Common in sandy soil throughout. Herb. : Many specimens. Zizania aquatica L. Sp. PI. 991. 1753. Wild rice, Indian rice. 88 MINNESOTA BOTANICAL STUDIES. Common in shallow water and swamps throughout. Herb. : Numerous specimens. Homalocenchrus virginicus (WILLD.) BRITTON, Trans. N. Y. Acad. Sci. 9: 14. 1889. White grass. Leersia virginica WILLD. Sp. PI. I : 325. 1797. Infrequent south. Herb. : Sandberg, Oestlund, Aiton, Hennepin county ; Bal- lard 1172, Zumbrota ; Sandberg, Goodhue county; Wheeler 564, Houston county. Homalocenchrus oryzoides (L.) POLL. Hist. PI. Palat. i: 52. 1776. Rice cut-grass. Leersia oryzoides Sw. Fl. Ind. Occ. I : 132. 1797. Common throughout in moist places. Herb. : Numerous specimens. Homalocenchrus lenticularis (Micnx.) SCRIBN. Mem. Torr.. Club, 5 : 33. 1894. Catch-fly grass. Leersia lenticularis MICHX. Fl. Bor. Am. I : 39. 1803. Rare southeast. Herb. : Lyon 713, Houston county. Phalaris arundinacea L. Sp. PI. 55. 1753. Reed canary grass. Common in moist soil throughout. Herb. : Numerous specimens. Phalaris canariensis L. Sp. PI. 54. 1753. Canary grass. Adventive throughout. Herb. : Well represented by specimens. Anthoxanthum odoratum L. Sp. PL 28. 1753. Sweet vernal grass. Adventive southeast. Herb. : Holzinger 3556, Winona. Savastana odorata (L.) SCRIBN. Mem. Torr. Club, 5 : 34. 1894. Holy grass, Vanilla grass. Hierochloa boreahs R. & S. Syst. 2: 513. 1817. Hierochloa odorata fragrans (WILLD.) RIGHT. PI. Eur.. I : 31. 1890. Common throughout the state especially in the northern part. One of our very earliest grasses. Herb. : Sheldon 175, Blue Earth county, 2062, 2367, Aitkin county; Sandberg, Goodhue and Hennepin counties; Aiton, Wheeler: CATALOG OF MINNESOTA GRASSES. 89 Hubbard county ; Frost, Minneapolis ; Bailey, Vermilion Lake ; Sedge, Detroit. Aristida curtissii (A. GRAY) NASH in Britton, Man. 94. 1901. Curtiss's three-awned grass. In dry soil. Not previously reported from Minnesota. Herb. : Upham, Minneapolis, 1884. Aristida basiramea ENGELM. Vasey, Coult. Bot. Gaz. 9 : 76. 1884. Poverty grass. Three-awned grass. Frequent in dry soil south. Herb.: Sandberg, Wollan, Aiton, Hennepin county; Hol- zinger, Winona county ; Sandberg, Crow Wing county ; Shel- don 6169, Taylors Falls. Aristida longiseta robusta MERRILL, U. S. Dept. Agric., Div. Agrost. Cir. 34: 5. 1901. Long-awned aristida. Aristida -purpurea of authors, not NUTT. Infrequent in dry soil southwest. Herb. : Leiberg, Rock and Blue Earth counties ; Skinner, Heron lake; Sheldon, Lake Benton. Aristida purpurea NUTT. Trans. Am. Phil. Soc. 5 : 145. 1837. All reports of this species from Minnesota should be referred to Aristida longiseta robusta Merrill. Aristida purpurascens POIR. in Lam. Encycl. Suppl. i : 452. 1810. Purplish aristida. Reported from Minnesota by Lapham but probably of doubt- ful occurrence. Aristida tuberculosa NUTT. Gen. i : 57. 1818. Sea-beach aristida. Reported from Minnesota by Lapham but is not represented by specimen in the University herbarium. Stipa macouni SCRIBN. ; Macoun, Cat. Can. PL 5 : 390. 1890. Macoun's stipa. Reported from north shore of Lake Superior by Macoun. No Minnesota specimens known to have been collected. Stipa viridula TRIN. Mem. Acad. St. Petersb. (VI.) 2 : 39. 1836. Green stipa. Rare, west. Herb. : Skinner 41, Heron lake. Stipa avenacea L. Sp. PI. 78. 1753. Black oat-grass. Reported by E. P. Sheldon in the Minnesota Botanical Studies from Poplar Island lake, Ramsey county. The speci- 90 MINNESOTA BOTANICAL STUDIES. men so labelled in the University herbarium is Stipa comata Trin. & Rupr. There are no Minnesota specimens of Stipa avenacea in the University herbarium. Stipa comata TRIN. & RUPR. Agrost. 3: 75. 1842. Western stipa. Not previously reported from Minnesota. Herb. : Sheldon, Poplar Island lake, St. Anthony Park. Stipa spartea TRIN. Mem. Acad. St. Petersb. (VI.) i : 82. 1831. Porcupine grass, Devil's darning-needle. Common on dry soil throughout. Herb. : Numerous specimens. Oryzopsis canadensis (Pom.)ToRR. Fl. N. Y. 2 : 433. 1843. Slender mountain rice. Oryzopsis juncea (Micnx.) B.S.P. Prel. Cat. N. Y. 67. 1888. Common in the vicinity of the headwaters of the Mississippi river and probably extending throughout northern Minnesota. Herb. : Lyon, Rosendahl, Butters and Wheeler, and Aiton, Lake Itasca ; Sheldon 2071 and 2347, Aitkin county, 2012, Brainerd; Anderson 407, Cass county. Oryzopsis asperifolia MICHX. Fl. Bor. Am. i: 51. 1803. White mountain rice. Common north of the twin cities, rare south. Herb. : Sheldon, 4561, Lake county, 2097, Aitkin county, 1926 Hennepin county, 4613 Tower, 6192 Taylors Falls ; Bal- lard, Cass county ; Lyon, Rosendahl, Butters and Wheeler, 38, Lake Itasca. Oryzopsis melanocarpa MUHL. Gram. 79. 1817. Black moun- tain rice. In woods throughout? Herb. : Sandberg, Herrick, Aiton, MacMillan, Hennepin county; Ballard 1794, Cass county; Campbell, St. Cloud; Taylor 949, Glenwood ; Sandberg, Isanti county. Milium effusum L. Sp. PL 61. 1753. Tall millet-grass. Rare in moist woods. Not previously reported from Minne- sota. Herb.: Sheldon 164, Waseca county, 2996, Milaca ; Wheeler, Ramsey county. Muhlenbergia sobolifera (MUHL.) TRIN. Unifl. 189. 1824. Rock muhlenbergia. Wheeler: CATALOG OF MINNESOTA GRASSES. /O \X I / 1 i n n € soLs&j o