Digitized by the Internet Archive in 2010 with funding from University of Toronto http://www.archive.org/details/bulletin190113v31v37unit Ewe 'nteru Ly i 7 ee F AU; >) DEPARTMENT OF AGRICULTURE. = BUREAU OF PLANT INDUSTRY + BULLETIN NO. 31. B. T. GALLOWAY, Chief of Bureau. CULTIVATED FORAGE CROPS OF THE NORTHWESTERN STATES. I, Sh IBUMNCIELCOLORE ASSISTANT AGROSTOLOGIST, IN CHARGE OF CooPERATIVE EXPERIMENTS, GRASS AND FORAGE PLANT INVESTIGATIONS. Issurp DECEMBER 13, 1902. — ~ WASHINGTON: GOVERNMENT PRINTING OFFICE. 1902. — : ‘ hi), ] \ 4 fay | Vie ee TE LETTER OF TRANSMITTAL. U. S. DeprarTMENT OF AGRICULTURE, Bureau oF PxLant I[Npustry, OFFICE OF THE CHIEF, Washington, D. C.. October 17. 1902. Str: I have the honor to transmit herewith a paper on ** Cultivated Forage Crops of the Northwestern States,” and respectfully recom- mend that it be published as Bulletin No. 31 of the series of this Bureau. This paper was prepared by Mr. A. S. Hitchcock, Assistant Agros- tologist, in Charge of Cooperative Experiments, Grass and Forage Plant Investigations, and has been submitted with a view to publica- tion by the Agrostologist. Respectfully, B. T. GatLoway, Chief of Bureau. Hon. James WIzson, Secretary of Agriculture. el eae Bann JB During the summer of 1901 Professor Hitchcock, under instruc- tions from the then Agrostologist, Prof. F. Lamson-Scribner, visited the States of Kansas, Nebraska, Colorado, Wyoming, Utah, Nevada, California, Oregon, Washington, and Idaho for the purpose of study- ing conditions with reference to cultivated forage crops. In the course of his investigations he visited the experiment stations of the above States and interviewed many farmers and ranchmen, from some of whom he received much valuable information. Considerable informa- tion was also obtained from seedsmen and from dealers in grain and hay and farm machinery. The accompanying paper is a résumé of the information thus obtained. It isrecognized that in a large section of country rather sparsely settled, and particularly one in which agri- culture is a recent development, many farmers and others have learned much that would be valuable to others in the same section of country. The principal object of this paper is to make common property of the individual knowledge of various farmers, ranchmen, and others, so that each may benefit by the experience of others. This is particularly important in a new country such as the region described herein. The paragraph relating to the ‘* Inland Empire” and the last para- graph of the section devoted to velvet grass were written by the Agrostologist; otherwise the paper is entirely the work of Professor Hitchcock. W. J. SPILLMAN, Z Lore »stologist. OFFICE OF THE AGROSTOLOGIST, Washington, D. C., October 14, 1902. Tie. : r loa? ¥ ’ Mt t F i . ; ; s * - . > - i « dee A ~~ — ~ CONTENTS. BRCM OURO TsOh OMeD ONS mam settee ieee atime = <1 senile enn so (Guigntt IANS cocoate ddonceasecséccen = ees SCS OCS eee eee eae ee eae Rainy MoyaaniMin GA Ofils = one cha ae Connec 6 Co See ee ee Hee a ae ae (GiREGhi TEIN S55 Soandcee Caso c ee ede Os dl 00 see OSes Cee see ana lintemoumallevs on Caloris =r e= ees manana sca a- cc c= <== ~~ -=- Up pene aaGtn cre maetienl Olea em eae eee see eee ane clem ~ aiemieitmean -=- = Thee: MC Ibni real Siri occ coke sa nseosn5. 5s Sees e oes eas eSenE aS ee eee INoPHe GRO soaocatecssoes oon ee eusobeo sac odo Sooner Gee eae coe Spe pee a aaa (MPI 2 oS cede ches eece ene SES POO SSeS See eee eo eee UMN <5 oo bonsooselondedeseneacstCecneeck! ples =seS Sosa Seats (Graimhayoc. 2-4 = = Aeeere 1 Beco onocoSop IRC eee a S Seon IRQOIMOD <.ogencs seeing Sadee oes co see elec SSS 00r (eee Rese Sener Arnillege: Inraniet Rae os Sapecseooecdqbee coe e eee epee eee EE Saeeoee WEIN Gi GRIER c ood soc pakdeasen dead MECN OSCE DoCS =e ASE Eee ae (HaWESE onc ce od GOS ORE CORE EEO Se OB SEE CR OEe Conc nO Sees e See Hondcercrops OtsmMimon lm pORtanee esac ase = ese —= === ¥ries, Elias. Systema Mye., 1: 421; Rabenhorst’s Kryptogamenflora, 1: 421, 1SS4. ¢Peck, C. H. Thirty-Fifth Report, New York State Museum, 1881, p. 136. 14 A DISEASE OF THE WHITE ASH. The fungus under discussion is one of the most distinct forms of the Fomes type of Polyporus, and considering the great variability of form of many species of this genus it can be said to be remarkably constant in most of its characters. MICROSCOPIC CHANGES IN THE WOOD. The minute changes which the wood cells undergo are marked by great distinctness and regularity. The wood of the ash forming the bulk of the trunk serves as a repository for large quantities of starch. Even in trees which are 75 to 100 years old one will find starch almost at the center. In the ash the starch occurs in the form of small grains (Pl. LV, fig. 6), filling the cells of the medullary rays and wood paren- chyma. Fig. 1, Pl. Il, represents a cross section of wood (cut in March), stained with iodine. The medullary rays appear almost as black lines. One of the first changes noticeable in the wood when attacked by the ash fungus is in connection with this starch. The region where the starch changes is just outside of the dark line seen in Pl. I. The large grains (PI. IV, fig. 6) appear to break up into numerous smaller ones (Pl. IV, fig. 7), and finally even these disappear. The change is a very ‘apid one, and transition stages are very rare. No such regular gradual dissolution of the grains occurs as is deseribed by Hartig as taking place in oak wood attacked by Polyporus sulphureus and Polyporus igniarius. When stained with iodine one finds large grains now and then, with channels through them (PI. LV, fig. 6), or more frequently some which look as if the center had been dissolved out. In several instances grains were found which stained brown with iodine at the edges. This brown color then gradually passed in toward the center of the grain. No hyphe are present in the wood where the starch is breaking up. This would indicate that a diastatic enzyme given off by the mycelium precedes the latter for some distance. The first hyphie are generally several rings farther toward the middle of the trunk. ‘The even extent of the solution strengthens this supposition, for in a limited area of one wood ring one and the same stage of dissolution is found at about the same distance from the point where the fungus begins its growth. After the disappearance of the smallest grains the cells formerly filled with starch appear empty for several cell rows inward. Shortly after the disappearance of the starch they become filled with a bright- colored substance, which is probably liquid at first and hardens after infiltration into the cells (Pl. IV, fig. 2). This substance, which is very soluble in alkalis, is probably some humus compound which must be regarded as a decomposition product. It is distributed throughout the medullary rays and the woody parenchyma, occupying almost the identical cells which had harbored the starch. This will MICROSCOPIC CHANGES. 15 readily be comprehended by a comparison of Pl. ILI, figs. 1 and 2. Fig. 2 is from a photograph of an unstained section taken from the region of brown wood at the outermost edge. It is rather difficult to determine the origin of this decomposition product. It is possibly the last product of a change in the starch grains, possibly also a substance derived from wood cells farther inward, which infiltrates into the medullary ray cells and wood paren- chyma in advance of the fungus hyphe. The latter is the probable explanation, for one finds the- humus compound in the summer wood cells, which had very little starch originally. The humus compound appears to form in many of the wood cells, however, as a product of the walls. Figs. 4 and 5 of Pl. IV show various stages of this change. The cells a are sound wood cells, which have very thick walls anda very smalllumen. The walls of cells marked / are very much thinner, and at these points they are coated with the humus compound. Such walls when stained with phloroglucin show no very sharp dividing line between the yellow humus compound and the apparently sound ligni- fied wall. Cell ¢ is completely filled with the humus mass. This eyi- dence that the wall actually changes into the yellow mass is not very conclusive. The humus compound does not seem to be formed from the walls of the medullary ray cells, where it is found ultimately, for no signs of change are evident in the walls of these cells. The local- ized distribution of the humus substance is very striking. It is always absent from the wood cells of the spring wood (Pl. III, fig. 2) and from the large vessels. In the cells it appears to be as a solid mass, sometimes completely filling the lumen (Pl. IV, figs. 2 and 5), or in globules or plates adhering to the walls (PI. 1V, fig. 2). It i§ this sub- stance which gives the brown color to the early stage of diseased wood. The next stage in the dissolution of the wood cells takes place abruptly, and is rapid after it has once set in. The hyphe of the fungus first evident in the medullary rays spread through the wood of both the spring and summer bands, branching in all directions. They give off an enzyme which attacks the inner parts of the wood cells, extracting the lignin. A transverse section of wood in this stage (Pl. IV, fig. 1) stained with phloroglucin presents a most striking picture. Here and there, in irregular groups and in all stages, one finds wood cells from which the hadromal has been removed: the extracted parts remain white and stand out in sharp contrast to the unaffected parts of the walls. In the figure the unaffected parts are shaded. The white parts represent delignified walls. The middle lamella is dissolved last and then the individual cells fall apart. When this takes place throughout larger areas, for instance, one or more wood rings become separated from one another, and this gives rise to the plates spoken of above. The white areas which are evident in the figures on Pl. I represent wood thus destroyed. The individual tibers 16 A DISEASE OF THE WHITE ASH. remain intact for some time, and are then gradually dissolved. In the oldest parts of diseased wood they are no longer present. Wood partially destroyed in the manner just mentioned was stained with potassium permanganate, HCl and NH,OH, according to the method recently described by Maule.” A dilute solution of the permanganate is allowed to act on the wood for a minute. The wood is then treated with strong HCl until no color is visible. A drop of ammonia is thenadded. The lignified walls stain a deep red, which in many respects defines the various parts of the walls more sharply than the phloroglucin reaction. The parts (Pl. LV, fig. 1), which do not stain with phloroglucin do not stain with the permanganate. The contrasting color between the lignified and delignified parts is even sharper. Maule claims that the permanganate reacted with an ether compound in the walls even after the removal ot Czapek’s hadromal. In the ‘‘ delignified” wood cells of the ash even this compound (if there be a separate compound which reacts with the permanganate) is therefore absent. In the ash wood the white fibers are not pure cellulose. The same is true of many similar fibers from oak wood destroyed by species of Hydnum, ov Polyporus igniarius, and probably of other white fibers resulting from fungus action on wood. With chloriodide of zinc, the best cellulose reagent we have, these fibers stain a yellow brown, not blue. This would indicate that the change in the wall is not the same as in many of the conifers, where the so-called lignin is destroyed, leaving a comparatively pure cellulose, as determined by staining reaction and macrochemical analysis. This subject is simply referred to in this connection, as it will form the subject of a separate paper. The change to an impure cellulose takes place locally, and generally very early in the course of the destructive action of the fungus. The mass of wood destroyed changes somewhat differently. The first changes noticeable are in the medullary rays and immediately adjoin- ing cells. Very fine fungus hyphe invade these cells, and shortly after the middle lamelle disappear. Small cayities occur in thicker parts of this layer, i. e., where several cells touch (PI. IV, fig. 3, 0), and these increase in size (7), spreading laterally, until two or more join. Ultimately the individual cells become entirely isolated. The wood cells proper are gradually destroyed from within outward, the middle lamelle remaining longest. The change from perfectly sound wood to wood entirely dissolved is a very abrupt one (Pl. IV, fig. 8). The hyphe invade a cell and dissolve the wall. So rapid is this that no intermediate changes can be found. A piece of completely rotted wood, such as occurs in the center of a diseased trunk (PI. 1), 1s repre- sented in Pl. IV, fig. 8. A more resistant piece of summer wood is «Maule, C. Das Verhalten verholzter Zell membranen gegen Kalium permanganat, eine Holzreaction neuer Art. (Beitriige zur wissenschaftlichen Botanik, Vol. IV. Stuttgart, 1901.) (Reviewed in Bot. Cent., 89. 328, 1902.) GROWTH OF THE FUNGUS. iw shown at one side. It is surrounded by an intricate mass of hyph, in which pieces of undissolved wood are held in much the relative position which they occupied in the sound wood. It will be seen that the wood is practically destroyed entirely. The mass of fungus hyph gives a soft, leathery, yielding consistency to the rotted material. The young hyphe are exceedingly fine, so much so that it requires a strong immersion lens to detect them. They are perfectly colorless, and remain so when older. Clamp connection occurs frequently. GROWTH OF THE FUNGUS IN DEAD WOOD. could be ascertained. It will grow out from infected wood when the latter is kept in a moist place, but only to a very small extent. A number of pieces of diseased ash trunks, each about a foot long, were placed in the mushroom cellar of the Missouri Botanical Garden, some with the cut surface in contact with the soil, others exposed to the moist air. In order to test whether dead wood could be infected, several healthy pieces of ash trunks, recently cut and of about the same diameter as the diseased pieces, were placed in contact with the smoothed end surfaces of the diseased pieces. After two or three days the hyphe in nearly all the pieces began to grow out from the diseased areas (PI. V), both from the brown areas and from the parts entirely decayed. This indicates that the fungus is equally active all through the diseased parts. In the pieces where the cut surfaces were exposed to the moist soil or air‘the hyphe grew for some weeks, making a thick, tough felt. “They gradually ceased growing after about three weeks. The sound ash trunks were firmly united to the diseased ones after three days, and after a week the fungus had so thoroughly united the two pieces that they could not be pulled apart, using a moderate amount of force. After three months the healthy pieces were examined. The hyphe of the fungus had grown into the wood for a very short distance only. They had effected practically no change. A hard cushion of mycelium had formed between the two pieces, and this was turning brown and had evidently ceased growing. These tests show that under the conditions of temperature and moisture which permit of vigorous growth of several of the wood-destroying fungi growing on dead wood the mycelium of the ash fungus will not grow for any length of time. The sound wood placed in contact with the diseased wood was full of starch at the time, so it could not have been lack of food which prevented the growth of the hyphe. A piece was removed from a sporophore immediately after it was brought in from the woods. The sporo- phore remained attached to a section of the trunk about a foot long. For several weeks hyphe grew out from the injured surface, making a new rounded edge, doing so almost as rapidly as in the natural state. 12163—No. 32—03 9 The mycelium of the fungus grows only in livine trunks. so far as C > => “ > 18 A DISEASE OF THE WHITE. ASH. REMEDIES. The white ash is becoming more valuable as a lumber tree, and it is being grown extensively as an ornamental tree in parks and grounds. In limited areas it will pay to adopt measures which will tend to pre- vent the disease described in the foregoing pages, or at least to recog- nize diseased trees and use them for lumber, so as to save the parts still sound. A disease such as the white rot of the ash is a difficult one to combat after a tree is once badly diseased, for the fungus grows in the interior of the trunk, where it can not be reached. Trees which grow in forest tracts should be cut down when badly diseased, so as to prevent the spread of fungus spores. That a persistent cut- ting out of diseased trees will in a comparatively short period reduce the number of newly infected trees has been demonstrated repeatedly in European forests, where it is now often impossible to find many well-known forms of disease which were formerly comparatively common. In parks and grounds diseased trees, when they appear healthy otherwise, need not necessarily be cut down, for the trees may remain alive and vigorous even when the heartwood is partially decayed. The only danger is that trees weakened in that way are liable to be broken off by windstorms. A diseased tree can be recognized as soon as the white punks or sporophores appear at a knot hole. As soon asa punk appears it can be cut out, and some of the diseased wood with it. The hole should then be filled with tar oil and left open for atime. Taroil should be added from time to time, as a good deal will soak into the decayed wood, and thereby arrest the further growth of the fungus to some extent. If the hole made by removing the punk is a large one it should be covered with tar paper, so that no opening is left for water or dust to enter. A sure method of combating this disease is by a careful system of pruning and the coating of all wounds with an antiseptic substance. Vigorously growing ash trees heal wounds rapidly, and after three or four years any ordinary-sized wound will be completely occluded. In treating trees planted in parks or gardens the pruning had best be done in the winter. Care should be taken to cut all branches as close to the trunk as possible, and after trimming the ragged edges of a cut the whole surface should be coated. Ordinary gas tar is the best sub- stance for this purpose. If too hard it should be heated so as to be fairly liquid and then applied with a brush. The gas tar, especially when warm, penetrates for a considerable distance into the wood and prevents the development of the ash fungus. It forms an air-tight and water-tight cover which is not destroyed by weathering, and which at the same time is objectionable to insects. Where the coating of wounds is carried on with care it will be entirely practicable and possible to prevent this ash disease. 19 DESCRIPTION OF PLATES. Pate I. (Frontispiece. ) Sections of living white ash trees (Fraxinus americana) attacked by Polyporus fraxinophilus Pk. The upper figure shows an early stage; the lower, a later stage of the decaying process. Piate II. Fig. 1.—Fruiting body of Polyporus fraxinophilus Pk. growing out from a dead branch. This is a rather exceptional form of sporophore, which is found only on branches. Fig. 2.—T wo young sporophores of Polyporus fraxinophilus Pk. growing on living ash. Fig. 3.—An old sporophore of Polyporus fraxinophilus Pk. growing on living ash. Puare lil. Fig. 1.—Transection of healthy ash wood, stained with iodine so as to show the distribution of starch in the medullary ray cells and in the wood paren- chyma surrounding the large ducts. This section is made just outside the dark line dividing sound from diseased wood (see Pl. I). Fig. 2.—Transection of diseased ash wood, not stained, showing the distribution of a humus compound in the medul- lary ray cells and in the wood parenchyma surrounding the large ducts. This sec- tion is made just inside the dark line dividing sound from diseased wood (see Pl. I). Puate TY. 1.—Transection of ash wood, showing one form of change in the wood cells caused by the fungus hyphe. The darkly shaded parts are sound wood cells. The white parts are wood parts which do not stain with phloroglucin. (Magnifica- tion same as for fig. 2.) 2.—Transection of medullary ray from the brown wood layer, showing how the cells become filled with a brown humus compound, here shown by the dotted areas. In two cells the dry compound has cracked. 3.—A medullary ray, showing a later stage of fungus attack. The middle lamell are dis- solyed out, separating the individual cells from one another. Note the absence of ‘he humus compound. (Magnification same as for fig. 2.) 4 and 5.—Transection of wood cells (highly magnified), showing various stages of change of wood into a brown humus compound. Note the great thickness of walls of neighboring sound cells. The humus compound is shown by the shaded parts. 6.—Starch grains from medullary ray cell. Normal grains and several grains showing how grains are now and then dissolyed. The short line equals 10”. 7.—Starch grains from diseased wood, showing how the large grains are broken up into smaller ones. (Magnifica- tion same as for fig. 6.) 8.—Transection from entirely rotted wood. The sound wood cells at one side belong to a small piece of more resistant wood. (Magnifi¢a- tion same as for fig. 2.) ; Piare V. Cross section of diseased trunk of the white ash kept in a moist place for several weeks. The fungus hyphie. have grown out from the diseased wood, forming a white felt. 20 O Bul. 32, Bureau of Plant Industry, U. S. Dept. of Agriculture Fic. 1.—Fruitinc Bopy oF PoLyPorus FRAXINOPHI ~~ ant Industry, U >) of Bureau FIG. 1.—HEALTHY ASH WOOD SHOWING STARCH IN THE MEDULLARY RAYS. FIG. 2.—DISEASED ASH WOOD SHOWING HUMUS COMPOUND. Mv i: PLATE IV. Bul. 32, Bureau of Plant Industry, U. S. Dept. of Agriculture. re j DISEASE CAUSED BY POLYPORUS FRAXINOPHILUS. medullary ray, showing later tarch grains from medullary ray ntirely rotted wood wood; 8, transection from e transection of wood cells: 6, s transection of medullary ray: 8, eased 9 starch grains from dis of fungus attack: 4.5, t 7, 1 1, Transection of ash wood; sta. cel Bul, 32, Bureau of Plant Industry, U. S. Dept. of Agriculture PLATE V. CROSS SECTION OF DISEASED TRUNK OF WHITE ASH KEPT IN A MOIST PLACE FOR SEVERAL WEEKS (SHOWING GROWTH OF MYCELIUM FROM THE ROTTED PART HELIOTYPE ©O., BOSTON, Us] DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY—BULLETIN NO. 33. B. T. GALLOWAY, Chief of Bureau. NORTH AMERICAN SPECIES OF LEPTOCHLOS, BY A. 'S: HITCHCOCK, ASSISTANT AGROSTOLOGIST, IN CHARGE OF COOPERATIVE EXPERIMENTS, GRASS AND FORAGE PLANT INVESTIGATIONS. IssunD Fepruary 10, 1903. Yysricultune; WASHINGTON: GOVERNMENT PRINTING OFFICE 1908. et LETTER OF TRANSMITTAL. U. S. DEPARTMENT OF AGRICULTURE, BUREAU OF PLANT INDUSTRY, OFFICE OF THE CHIEF, Washington, D. C., October 18, 1902. Sir: [have the honor to transmit herewith a technical paper entitled “‘North American Species of Leptochloa,” and respectfully recommend that it be published as Bulletin No. 33 of the series of this Bureau. This paper was prepared by Mr. A. 8. Hitchcock, Assistant Agros- tologist, in Charge of Cooperative Experiments, Grass and Forage Plant Investigations, and has been submitted by the Agrostologist with a view to publication. Respectfully, B. T. GALLOWAY, Chief of Bureau. Hon. JAMES WILSON, Secretary of Agriculture. PRBPACE. There is much confusion in the names applied to our North Ameri- can grasses. This is partly due to the fact that much new material has been collected since the revision of some of the important genera. The practice, formerly more prevalent than at present, of erecting new species on the basis of a single specimen or of a very few speci- mens at most, has added to this confusion. The economic importance which the grasses have assumed in the last two decades has made this confusion all the more embarrassing. It therefore seems desirable that the bibliography, synonymy, and systematic relationships of American grasses be worked out as rapidly as possible. The present paper by Professor Hitchcock is an attempt to do this for the genus Leptochloa. It is based chiefly upon the material in the herbarium of the U.S. National Museum and that of the U.S. Department of Agri- culture, but all the important public herbaria in this country were consulted during its preparation. The descriptions of the species are diagnostic rather than complete, but it is hoped that these will serve the purpose of students of systematic botany. Much time has been spent in working out the proper relationship of the species and it is hoped that the short descriptions, the text figures illustrating the spikelets of each species, the plates taken from herbarium specimens of several species, and the key to our United States species will take the place of more complete descriptions and render this paper valua- ble to students of this genus. The species of Leptochloa are inhabitants of the warmer regions, only one or two of our species extending as far north as New York and Illinois. One of the species, Leptochloa dubia, called sprangle, is an important range grass in the Southwest, and recent experiments indicate that it will prove a desirable grass for cultivating in semiarid regions. W. J. SPILLMAN, Agrostologist. OFFKICE OF THE AGROSTOLOGIST, Washington, D. C., October 14, 1902. cl ES WS TRACT OuNese PLATES. Page. Puate I. Fig. 1.—Leptochloa mucronata, Fig. 2.—Leptochloa viscida_.- 24 Il. Fig. 1.—Leptochloa domingensis,from Florida. Fig. 2.—Lepto- chloadomingensis, from’ Texas 2-22 225 s5-5 = == 24 Ill. Fig. 1.—Leptochloa scabra. Fig. 2.—Leptochloa nealleyi _____- 24 IV. Fig. 1.—Leptochloa fascicularis (Diplachne procumbens Nash). Fig. 2.—Leptochloa fascicularis (Diplachne tracyi Vasey) ---- 24 V. Fig. 1.—Leptochloa fascicularis (ordinary form). Fig. 2.—Lep- tochloatmoricatd — 2-2-2 2 sees 11 2. Spikelet of Leptochloa mucronata 22522 =a) a es 11 3. Spikelet of Leptochloa virgata. — 5. == ==) sees ae eee 12 4, Spikelet of Leptochloa domingensis, from Texas___.__. -...-...---- 13 5. Spikelet of Leptochloa domingensis, from Florida ------.-----.---- 13 6. Spikelet of Leptochloa domingensis, from Central America -_--___- 13 7. Spikelet (of Leptochlow nealleyt — —— 22 == aa 14 8 Spikeletiof Zeptochlow scabr.a: = = 223 14 9. Spikeletiof eptochlow viscida@ = — 2-8 eee 15 10: Spikeletof Geptochtow dubia == =- = = sa 15 11. Spikeletiof Leptochloa floribunda. 22 see 16 12. Spikelet of Leptochloa aquatica____- cane eeectd soe) coe eee 7 13. Spikelet of eptochloa fascieulams: - = 25 — = s.5s= =e ale 14. Spikelet of Leptochloa imbricata... 2. 2252552 a2 eee 19 15: Spikelet of Geptochloa spicata =—- 3 == =e. = 19 16: Spikelet of (Gouwwmua brandege === ese e a == --- 21 8 B. P. I.—42. G.F.P.1.—08 NORTH AMERICAN SPECIES OF LEPTOCHLOA. INTRODUCTION. In presenting the following review of the genus Leptochloa I have been able to bring together our knowledge of this group of grasses without describing any new species. In regard to the latter, botanists will probably be thankful. But, on the other hand, I have been con- strained in several cases to unite species kept separate by others. All will not agree with me in the course I have taken in this respect. It is always difficult to decide where specific lines shall be drawn, but I have been governed by this rule: When two or more forms are con- nected by numerous intergrading specimens they are to be considered as the same species, although typical specimens of the extreme forms may be easily distinguished. The notes are based mainly upon the Herbarium of the U. 5. Department of Agriculture, but through the kindness of those in charge I have had the opportunity to examine the collections at the Missouri Botanical Garden, the Gray Herbarium, the New York Botanical Garden, and the Philadelphia Academy of Natural Science. I have also examined the specimens in the larger European herbaria, to the directors of which I wish to express my thanks for the privilege. For the purpose of this paper it seemed not worth while to enumer- ate all the specimens examined, but a number of representative specimens from numbered sets have been indicated for easier refer- ence. KEY TO SPECIES OF THE UNITED STATES 1. Spikelets usually short-pediceled (sessile in L. spicata, but flowers several), arranged somewhat distantly along the branches of the panicle, not so con- spicuously one-sided as in the following group; 4 to several flowered (2-flow ered in some forms of L. dubia) 2 1. Spikelets nearly sessile in two or more rows on one side of the branches of the panicle, 2 to 4 flowered and usually closely imbricated (more distant in L. mucronata) - : 7 2. Panicle simple or often reduced to a single branch or spike spicata. 2. Panicle compound 3 3. Spikelets 4 (2) to 6 flowered : { 3. Spikelets many flowered, elongated 6 10 NORTH AMERICAN SPECIES OF LEPTOCHLOA. 4. Flowering glume broad, truncate and more or less emarginate; sometimes slightly awned from the protrusion of the mid-nerve -__._.--------- dubia. 4, Flowering glume rounded at apex and short-awned or mucronate____-- ---- 5 5. Panicle 2 to 3 inches long. Plant with numerous culms, a few inches to a foot hich:) leaves dion 4inches long: 39> = 2.22 oe eee aa i eee viscida. 5. Panicle larger, culms 2 to 3 feet tall, leaves a foot or more ene. floribunda. 6. Flowering glume awned- PPE Se Se nn oS fascicularis. 6. Flowering glume awnless or mucronate _-_-.--.----------.-------- imbricata. . Spikelets usually 2-flowered, sometimes 3 or even 4 flowered, 1 to 2 mm. long, branches of panicle very slender, upper empty glume as long as or longer than the first flowering glume, latter obtuse __.._..._.. -_.--- mucronata. 7. Spikelets usually 3 to 4 flowered, rather closely imbricated, spikes shorter and close set on the axis. forming a narrow panicle; empty glumes shorter than the first flowerinsvelumer —: 2-2 = ee a eee 8 8. Sheaths scabrous, glumes/sharp-poimted ----.2-2 222222232222 eee scabra. 8, Sheaths smooth; flowering glumes rounded or truncate at apex_----------- 9 9. Sheath ciliate on margin above; flowering glume more or less awned. domingensis, a eee eee nealleyi. 9. Sheath not ciliate; flowering glume awnless HISTORY OF GENUS. The genus Leptochloa was established by Palisot de Beauvois.“ To his new genus he refers Cynosurus capillaceus, Hleusine filiformis, and FE. virgata. The last of these species is figured? and in the description’ of plates he uses the name Leptochloa virgata. Tt may be inferred that he intends to make the new combination for the other two species, as in the index, page 166, he indents under Lep- tochloa the three names, capillacea, filiformis, and virgata. It may be remarked that if one intends to be very accurate in regard to cita- tions these three species of Leptochloa should be referred to page 166 (the index) rather than page 71 in the body of the work, where the genus is described. The same remark would apply to the most of Beauvois’s species. Beauvois also established the genera Diplachne,” to which he refers Festuca fascicularis Lam., and Rabdochloa,’ to which he refers Cyno- surus monostachyos, virgatus, domingensis, cruciatus?, mucronatus?. Kuntze substitutes Rabdochloa for Leptochloa because Beauvois assigns five species to the former and only three to the latter. Professor Seribner unites these under the genus Leptochloa.’ Pro- fessor Gray also placed Diplachne under Leptochloa as a section.’ Nuttall” proposed the genus Oxydenia to include O. attenwata (Hleusine mucronala). I have accepted the genus as delimited by Seribner, U.S. D. A. Div. Agros. Bul. 20:110. Our species Ben are annuals except L. dubia, a Maca dune Hoarele Wererioeraanses qile “1812. €l.c..p. 84. bl. c., Atlas, pl. xv, fig. 1. Jf Proc. Acad, Phil., 1891; 303. Cl, c., Atlas, 10. 9 Man., Ed. I, 588. a1. ¢., 80. 2 Gen. 1: 76, 1818. NORTH AMERICAN SPECIES OF LEPTOCHLOA. 11 NORTH AMERICAN SPECIES. A.—LEPTOCHLOA proper. Spikelets 2 to 4 flowered, arranged close together on one side of the branches of the panicle. LEPTOCHLOA MUCRONATA Kunth. Rey. Gram. 1: 91. 1835. Transfers Eleusine mucronata Michx. (P1. I. fig. 1; text fig. 2.) Eleusine mucronata Michx. Fl. 1: 65. 1803. ‘* Hab. in chltis Dlinoensibus.” Festuca filiformis Lam. Il. 1: 191.n. 1044, 1791. **Ex Amer, Merid. Comm. D. Richard.” Eleusine filiformis Pers. Syn. 1: 87. 1805. ‘* Hab. in Americ. meridion.” Eleusine sparsa Muhl. Descr. Gram. 135,1817. ** Habitat in Carolina et Georgia.” Oxydenia attenuata Nutt. Gen. 1: 76. 1818. **On the banks of the Mississippi near New Orleans.”’ Mr. Nuttall says: *‘ To this genus belongs the Eleusine filiformis of Persoon, growing in the tropical regions of America, nearly allied to the present species,’’ and is often quoted as the author of Orydenia filiformis, but he does not make this combination. Leptochloa filiformis Beauy. Agros. 71 and 166, 1812. Transfers Hleusine fili- formis Pers. Roemer and Schultes (2: 580. 1817), also transfer Eleusine jili- formis Pers. Presl, Rel. Haenk. 1: 288, 1830, gives as the locality ** Hab. in Mexico, ad Sorzogon Luzoniae.*’ In the herbarium of the U.S. Department of Agriculture are several specimens from India. I am unable to distinguish these from the American plant. Hooker includes these under L. filiformis R. & S. (Flora Br. India, 22: 298. 1896.) I have examined the Asiatic material in European herbaria and feel satisfied that L. mucronata occurs in southern Asia. It can be distinguished from the allied L. chinensis by the papillose sheaths. Fia.1.—L. attenuata. Fic. 2.—L. mucronata. Eleusine elongata Willd. ex. Steud. Nom. ed. 2, 1: 549, 1840. Labelled ** Habitat in America meridionalis Humboldt.** ypes of this and the next examined in herbarium Willdenow. Eleusine stricta Willd. 1. c. Labelled ** Habitat in San Domingo.” Leptochloa attenuata Steud. Syn. 209. 1855. Transfers Oxrydenia attenuata Nutt. This is kept separate by Mr. Nash in Britton’s manual, but the char- acters do not seem to me to be sufficiently constant for separation. This form is represented by Bush, Nos. 590, 403, 792, 793, and Eggert, 219a. from Mis- souri, and Palmer, 392, 401, from Indian Territory. Leptochloa pellucidula Steud. 1, c. ** Duchaissing legit in Panama.” Leptochioa pilosa Scribn. U.S. D. A., Div. Agros. Cir. 32: 9. 1901. “Type specimen collected in sandy soil, Dappan, Travis County, Tex., 2%. J. E, Bodin, September, 1891."’ Professor Scribner states that ‘* This species is closely related to Leptochloa mucronata, but it is at once distinguished by its rigid leaves and papillate-pilose sheaths.’ The leaves are somewhat more rigid than is usual in this species, but the papillate-pillose sheaths are found com- monly in L. mucronata, Stems tufted 6 to 10 dm. high, erect or occasionally more or less decumbent at base and rooting at the nodes. Leaves numerous, flat and rather soft, vary- ing from 1 to 3 or more dm. in length and as much as } cm. wide. Sheaths more or less pilose from a papillate base. Panicle often 3 dm. or more in length, consisting of numerous slender spikes, arranged along a central axis; 12 NORTH AMERICAN SPECIES OF LEPTOCHLOA. spikes usually 8 to 15 cm. long. Spikelets 3 to 4 flowered, 1 to 2 mm. long, rather distant on the axis, that is, scarcely overlapping. Empty glumes about equal, lanceolate, acute or acuminate, nearly as long as the spikelet, or some- times longer, lower slightly narrower. Flowering glumes thin, awnless, smooth or somewhat pilose on the nerves. The form separated as L. attenuata has large panicles, with acuminate empty glumes and flowering glumes pilose on nerves. 765; Bush, 468, 590; Curtiss, 5998; Coulter, 785; Lindheimer, 212. Meaico: Palmer, 248, 22, 694, 749, 1364, 117, 50 (in part); Rose, 1542; Schott, 739, 590. Yucatan: Gaumer, 853. Cuba: Wright, 740 (in part), 741 (in part). Porto Rico: Sintenis, 3550. Var. PULCHELLA Scribn. Bull. Torr. Bot. Club, 9: 147. 1882. ‘Santa Cruz Valley, near Tucson.” DISTRIBUTION.—Tewas to Arizona: Heller, 1884; Hall, 777, 778; Coues & Palmer, 511; Jones, 4176. Mexico: Palmer, 50 (in part), 503, 694, 8; Wright, 1316. Differs from the type in the short branches of the panicle, 2-3 cm. long, and the short narrow leaves. LEPTOCHLOA VIRGATA Beauy. Agrost., 166; Atlas, p. 10. 1812. Refers Eleusine virgata to his new genus Leptochloa (1.c. p. 71). (Fig. 3.) Fic.3.—L. virgata, from St. Croix. Cynosurus virgatus L. Syst. Nat., Ed. X: 1759. No locality is given, but he refers to Sloan jam., t. 70., f. 2, which is probably this species. In Spec. Pl., Ed. 2. the locality is *‘ Habitat in Jamaica.*’ See Munro, ‘**The Grasses of Linnzeus’s Herbarium,”’ Proc. Linn. Soc. Bot. 6: 33-35. 1862. Linnzeus mentions that the lower flowers are subaristate. Festuca virgata Lam. Tl. 1: 189. 1791. ‘‘ Ex ins. Domingi.’’ States that the spikelets are aristate and ** floscul. ultimis submuticis.”’ Eleusine virgata Pers. Syn. 1:87. 1805. Description taken from Lamarck, l.c. Oxydenia virgata Nutt. Gen. 1:76. 1818. This is the citation often given, but is an error, as Nuttall merely says, ** Tothis genus belongs Eleusine filiformis of Persoon . . . and we may probably add the Eleusine virgata of Jamaica.” Chloris polystachya Lag. Nov. Gen. 4. 1816. The short description scarcely suffices to determine this plant. ‘*Spicis pluribus, patentibus: calycibus flos- culisque glabris, muticis: culmo compresso. H. in N. H. unde semina missit D. Sesse.”” Chloris poeformis H. B. K. 1: 169. 1815. ** Crescit in calidissimis humidis flu- minis Migdalene prope Mompox: item prope Guayaquil et San Bowndon Quitensium.’* As synonyms are given Cynosurus virgatus L., Eleusine vir- gata Willd., and Leptochloa virgata Beauy., but a new specific name is applied because there is already a Chloris virgata Sw. In the description it is stated that the awn is very short. Leptochloa procera Nees in Syll. Ratisb. 1: 2. 1828. Type examined at Berlin. Leptochloa digitaria Willd,, ex Steud. Nom. Ed, 2. 1: 549. 1840. Types of this and the next examined in herbarium Willdenow. Both specimens labelled ‘* Habitat in America Meridionalis, Humboldt.”’ Leptochloa wnioloides Willd., 1. ¢. —_ aN) NORTH AMERICAN SPECIES OF LEPTOCHLOA. Leptochloa mutica Steud. Syn. 1: 208. 1854. ‘‘Surinam Am. Anstr.”” Type examined. DistRiBsuTion: Ruatan Island: Gaumer. Mexico: Liebmann 251, 252: Nelson 2768, 2483. Cuba: Rugel 193; Wright 3436, 740 (in part). 741 (in part); Combs 256. Porto Rico: Heller 4535: Sintenis 844. Martinique: Bourgean 2375; Hahn 163. St. Vincent: Smith 577. St. Croix: Ricksecker 258. St. Thomas: Eggers 68. Galapagos: Anderson 44. Brazil; Riedel, Traill 1274. Paraguay: Morong 970. LEPTOCHLOA DOMINGENSIS Trin. Fund. Agrost., 133. 1820. Transfers Cynosurus domingensis Jacq. (PI. II, figs. 1, 2; text figs. 4, 5, 6.) Cynosurus domingensis Jacq. Misc. 2: 363, 1781. ‘‘ Facie infra medium pilosa dorsa glabra.” Bromus capillaris Moench. Meth.194,1794. ‘Sub nomine Poae capillaris semina accepi,’’ no locality given. Kunth refers this to L. domingensis (Enum. 1: 269) and the description applies, especially, *‘ Folia lata infra glabra. supra deorsum scabra, basin versus pilosa,** but Moench also says, ** vaginee glabrze.”* However, the pubescence is confined to the margin of the sheath. Eleusine domingensis Pers. 1: 87. 1805. ** Hab. in Jamaica, St. Domingo.” Rabdochloa domingensis Beauy. Agrost.176. 1812. Transfers Cynosurus dom- ingensis, p. 84. He also refers Poa domingensis Pers. Syn. 1: 88 to his genus Rabdochloa, and in this is followed by Kunth (1. c.). Fig. 4.—L. domingensis, Fig. 5.—L. domingensis, Fig. 6.—L. domingensis, from Hidalgo, Tex. from Florida. from Central America. Leptostachys domingensis Meyer. Esseq. 74. 1818. Transfers Hleusine domin- gensis Pers. Leptochloa gracilis Nees. Syll. Ratisb.,1: 4. 1824. Transfers Chloris gracilis H. B. K. See note under L. dubia. Nees in Agrost. Bras., 433. 1829, gives “Habitat in Brasiliis . . . (Sellow. Vidi in Herb. Reg. Berol.)*’ Chloris gracilis H. B. K. Nov. Gen. 1: 168. 1815. ‘*Crescit in calidis Pro- vinci Jaen de Biacamoros prope Tomependa, alt.. 207 hex.”* Leptostachys gracilis Meyer. Fl. Esseq., 74. 1818. Transfers Chloris gracilis to his new genus Leptostachys. Our plants have the rigid, glaucous appearance of L. virgata, with involute leaves, but resemble L. domingensis in having the margin of the sheaths and the upper surface of the lower part of the blades ciliate or pilose. The awns are almost the length of the flowering glume. Grisebach distinguishes these by the length of the spikes and of the awns (Fl. Br. W. I.). thus. L. virgata with spikes 3-6 in. long and awns short or none; var. gracilis, awns about as long as glume, spikes 14-2 in. long; var. domingensis, spikes 8-5 in. long and awns longer. The length of the awn can not be depended upon to distinguish these forms. Stems | to 1 m. high, smooth and somewhat shining or glaucous, leaves long and narrowed to a slender point, involute; the tropical specimens have softer, flat leaves. Our specimens are probably introduced as the plant is not common within our borders. The drier climate would account for the involute leaves. The upper surface of blade near base is sparsely pilose with long weak hairs, the margin of the sheath is more densely ciliate. Panicles 1 to 2 dm, long 14 NORTH AMERICAN SPECIES OF LEPTOCHLOA. with numerous ascending branches 4to8 cm. The tropical specimens often have more ample panicles. Spikelets crowded, about 2 mm. long, 3 to 5-flowered. Empty glumes acute, lower narrow and shorter, about 14 mm.; lower flowering glumes bear awns about their own length, upper with shorter awns or awnless. Distrisution: Florida along the coast south of Tampa, Simpson. Texas, Cor- pus Christi, and Hidalgo,4 Nealley. South America and West Indies. LEPTOCHLOA NEALLEYI Vasey. Bull. Torr. Bot.Club, 12: 7. 1885. ‘* Col- lected in Texas by Mr. G. C. Nealley, for whom it is named.” Leptochloa stricta Fourn. Pl. Mex. 2: 147. 1886. I have examined the type in Paris. ‘‘ Vera Cruz (Gouin, n. 73).” Fie.7.—L. nealleyi. Stems 4 to 14m. high, smooth. Leaves elongated or on the smaller plants only 5 to 10 em. long, 3 to 5 mm. wide, involute, somewhat scabrous; sheaths smooth or very slightly scabrous. Panicles narrow, 2 to 4 dm. long, branches numer- ous, crowded, appressed, 2 to 6 cm. long. Spikelets crowded, about 2 to 3 mm. long, 3 to 4-flowered. First empty glume about one-half the length of the second and narrower; flowering glumes obtuse. Distrisution: Texas: Nealley 2501: Bush 1363; Buckley, Drummond 291; Tracy 7368. This has the aspect of L. scabra, but the glumes are rounded at the apex, while in the iatter they are acuminate or slightly awned. (PI. III, fig. 2; text fig. 7.) LEPTOCHLOA SCABRA Nees. Agrost. Bras. 435. 1829. ‘‘ Habitat in ripa inundata fluaminum Amazonum, Tagipuru et Tocantins, provincize Paraensis (Mart.).”’ Nees remarks that this differs from L. virgata in having the leaves and sheaths very scabrous and the small, whitish, slender spikelets entirely unawned. (PI. III, fig. 1; text fig. 8.) Fic. 8.—L. scabra, L. langloisii Vasey. Bull. Torr. Bot. Club, 12: 7. 1885. ‘‘ This large and showy species was found in Louisiana by Rey. A. B. Langlois, for whom it is named.”’ Resembles L. nealleyi in habit. Differs in having distinctly scabrous sheaths: the branches of the panicle longer and more or less curved; the spikelets 3 mm. or more long, the glumes acute or acuminate. Our plants are probably intro- duced from further south. Distrisution: Louisiana: In ditches and fields, Station Michaud, 13 miles from New Orleans, Langlois. Brazil: Rusby 235. British Guiana: Jenman 4441. Costa Rica: Tonduz 2604; Spruce 424. aThe specimen from Hidalgo (fig. 4) differs from the others in having the flowering glumes awnless. It isin an unsatisfactory condition, but may be L. virgata, Beauv. NORTH AMERICAN SPECIES OF LEPTOCHLOA. 1s) B.—Intermediate between Leptochloa and Diplachne LEPTOCHLOA VISCIDA Beal. Grasses N. A. 2: 434. 1896. Transfers Diplachne viscida Scribn. (PI. I, fig. 2; text fig. 9.) Diplachne viscida Scribn. Bull. Torr. Bot. Club, 10: 30, 1883. ‘Santa Cruz Valley, near Tucson, Arizona.’ Collected by Pringle. Growing in tufts in moist places, 1 to 3. dm. high. Leaves a few em. long, 2 to 3 mm. wide. Panicle short, 1 to 4cm. long, more or less enclosed in the sheaths. Spikelets 3 to 4 mm. long, 5 to 7-flowered. First glume about one-half the second, #mm. long. Flowering glumes short awned, somewhat viscid on the back. Fria. 9.—L. viscida, DISTRIBUTION: Arizona: Pringle; Mearns 793, 833; Griffiths 1988. New Mesico: Wright 2041, 2044. Mexico: Pringle 814; Palmer 748, 7484, 692, 1789: Brandegee 5; Wright 1086. LEPTOCHLOA DUBIA Nees in Syll. Ratisb. 1: 4. 1824. In an article entitled ** Nove plantarum species in horto botanico Bonnensi cult.’ Nees ab Esen- beck. who signs the portion relating to Leptochloa, describes L. procera, and states that it differs from ‘** Leptochloa gracile, Humb. et Kunth n. gen. et sp. I. p. 168 (sub chlori), vaginis glabris, valvulis corollinis nudis, nec ciliatis, apice integris, mucronatis, nec aristatis, flosculorum numero minore .. . A Leptochloa (Chlori) dubia Humb. et Kunth 1. ¢c. p. 169; panicula aequali, nec subfastigiata, flosculorum numero minore, valvulis nudis, nec ciliatis 33 He thus incidentally transfers these two species of Chloris to Leptochloa. (Fig. 10.) Fig. 10.—L. dubia. Chloris dubia H.B.K. Noy. Gen. 1: 169. 1815. ‘ Crescit in apricis suabhumitdis prope rupem porphyriticam el Penon, in convalle Mexicana, alt. 1168 hexap.”’ Leptostachys dubia Mey. Fl. Esseq. 74. 1818. Refers Chloris dubia doubtfully to Leptostachys. Festuca obtusiflora Willd. in Spreng. Syst. 1: 356. 1825. ** Mexico.” Type seen. Uralepis brevispicata Buckley. Proc. Acad. Phil. 1862: 98. 1863. ** Northern Texas."’ I have examined Buckley's specimen in the herbarium of the Philadelphia Academy. 16 NORTH AMERICAN SPECIES OF LEPTOCHLOA. Diplachne dubia Seribn. Bull. Torr. Bot. Club. 10: 30. 18838. Transferred to the genus Diplachne. Leptochloa pringlei, Beal Grasses N. A. 2: 436. 1896. ‘*D. pringlei Vasey ined. Arizona, Pringle, 1884.°° In the Department herbarium is a specimen collected by Pringle in 1884 in Tucson (No. 13), which answers to the description given in Beal’s Grasses, but seems to me to be a small form of L. dubia. This is figured in U.S. D. A. Div. Agrost. Bull. 7: 224, fig. 218. Diplachne dubia Pringleana O. K. Rev. Gen. Pl. 37: 348. 1898, transferred to Leptochloa by Scribner and Merrill, U.S. D. A. Diy. Agrost. Bull. 24: 27, 1901, is a robust variety from Chihuahua, Mexico (Pringle 422). Stems 3 to 10 dm. high from a perennial root. Leaves long and narrow, tapering to a slender point as in L. fascicularis Gray, usually not over one-half cm. wide. Panicle, consisting of several or many more or less spreading spikes, 5 to 15 em.long. Spikelets,5 to 10 mm. long,5 to 8 flowered, or in the smaller forms only 2-flowered. Empty glumes acute, upper 4 mm. long, lower a little shorter and narrower; flowering glumes broad and obtuse or emarginate at apex, the midrib sometimes extending into a short point. This species is readily distinguished by the broad, scarious emarginate apex of the flowering glumes. This is a valuable forage plant in the Southwest. where it is called *-sprangle.”” Experiments indicate that it may prove valuable under culti- vation in the arid regions of our Western States. DISTRIBUTION: Arizona: Lemmon 368. New Mexico: Wooten 418. Texas: Jones 4210; Wright 767. Florida: Garber 33; Curtiss 3450; Simpson 302; Tracy 6453. Mexico: Palmer 270, 273, 530, 381, 482, 468: Bourgeau 533; Brandegee 6; Schaffner 671, 1079, 933; Pringle 422; Xantus 119; Botteri 690. C.—DIPLACHNE. Spikelets several flowered, arranged more distantly on the branches of the panicle and not conspicuously one-sided. LEPTOCHLOA FLORIBUNDA Doell in Mart. Fl. Bras. 2*: 89. 1878. Type locality: ‘ad ripas fluminis Amazonum inter Manaos et Santarem (Spruce).”’ (Pl. VI, fig. 1; text fig. 11.) Diplachne halei Nash. Bull. N. Y. Bot. Gard. 1: 292. 1899. Type collected in Louisiana by Hale. Co-type in herbarinm U. 8. D. A. Fic. 11.—L. floribunda. Leptochloa halei Scribn. & Merr. U.S. D. A. Div. Agrost. Bull. 24: 27. 1901. Transfers Diplachne halei. The relation of L. halei to L. floribunda is dis- cussed in the article last cited. Going over the same evidence I believe that we are safe in making the present disposition. Plant with the aspect of L. fascicularis Gray. Panicle oblong, rather compact, with numerous branches 4 to 6 cm. long. Spikelets 4 to 5 mm. long, 5 to 7 fiowered. Empty glumes slightly unequal, upper about 2 mm., lower shorter. Flowering glumes with a very short point. Probably introduced in the United States from farther south. DistrinuTion: Texas to Brazil. Key West: Blodgett; Mississippi: Tracy 7451; Louisiana: Hale; Texas: Drummond 322; Brazil: Spruce 1112. NORTH AMERICAN SPECIES OF LEPTOCHLOA, Li LEPTOCHLOA AQUATICA Scribn. & Merrill. U.S. D. A. Div. Agrost. Bull. 24: 26.1901. ‘‘ Type specimen collected in shallow water near Cuernavaca, State of Morelos, altitude 1700 m., C. G. Pringle, 6664 August 22, 1897.” Resembles L. floribunda, but differs in having more unequal outer glumes, longer spikelets, with more distant flowers and obtuse flowering glumes. In L. floribunda the flowering glumes are distinctly short-awned. (Fig. 12.) Fia. 12.—L. aquatica. LEPTOCHLOA FASCICULARIS Gray. Man. Ed. 1.588. 1848. Festuca fascicularis Lam. Tabl. Enc. 1: 189. 1891. ** Ex. Amer. merid. Comm. D. Richard.’ (Pl. IV, figs. 1, 2; Pl. V, fig. 1; text fig. 13.) Bromus poeformis Spreng. Nach. Bot. Gart. Halle 15. 1801. Dr. Dammer, of the Royal Botanical Museum of Berlin, has kindly sent me a transcript of Sprengel’s description. ‘* Bromus poeformis mihi Pyrenien,”’ with refer- ence to a footnote which says ‘‘Poa digitata Michaux. Sed est certissime Bromus, utut repugnet habitus: namque ariste manifesto infra apicem glume oriuntur. Br. panicula erecta stricta composita, spicatis sex floris sub secun- dis, fol. longissimis involutis.”’ Fig. 13.—L. fascicularis. Festuca polystachya Michx. Fl. 1: 66, 1803. ‘In arvis Tllinoensibus.’’ Type seen, Diplachne fascicularis Beauy. Agrost. 80 and 160. Atlas, p. 11, pl. xvi, fig. 9. 1812. Made type of new genus without description of species. Festuca procumbens Muhl. Gram. 160. 1817. A prostrate form with longer awns, but the characters are not constant, and it does not seem best to sepa- rate this as a species, as is done by Mr. Nash. Diplachne procumbens Nash in Britton Man, 128. 1901. Transfers Festuca pro- cumbens Muhl. There is a South American species by this name, Diplachne procumbens Arech. Gram. Urug. 354. 1894. 11068—No. 33—03 » a 18 NORTH AMERICAN SPECIES OF LEPTOCHLOA. Leptochloa polystachya Kunth. Rev. Gram. 1: 91. 1835 (orearlier?). Transfers Michaux’s Festuca polystachya. Under the rule once a synonym always a synonym the Australian species should receive another name (Leptochloa polystachya Benth. Fl. Austr. 7: 617. 1878). Bentham says (p. 618), ‘I have been able to retain Brown's specific name, as the American Diplachne panicularis | fascicularis| named Leptochloa polystachya by Kunth is gener- ally retained under the former genus. Syn. Cynodon polystachya R. Br. Prod. 187. C. virgatus Nees in Steud. Syn. 1: 213. C. Neesii Thw. Enum. Pl. Ceyl. 371.” Diplachne acuminata Nash in Britton Man. 128. 1901. Represented from Nebraska, Rydberg 1713; Arkansas, Coville 87; Colorado, Clements 263. Uralepsis composita Buckley. Proc. Acad. Phil. 1862: 94. 1863, ‘‘ New Mexico. Dr. Woodhouse.’’ I have examined this specimen in the herbarium of the the Academy. Diplachne tracyi Vasey. Bull. Torr. Bot. Club. 15: 40. 1888. ‘In clumps growing in ditches at Reno, Nevada.’* Tracy No. 216. Dr. Vasey remarks that this is ‘‘ Near D. fascicularis..”’ In the type specimen which is in the herbarium of the Department of Agriculture the lateral nerves are more con- spicuously excurrent than is usual in D. fascicularis, but there seem to be no constant characters by which this form can be separated. It is a large form, with more exserted panicles, found from Nevada to Mexico, Pringle 813; Palmer 691. Leptochloa tracyti Beal. Grasses N. A. 2: 436. 1896. Transfers Diplachne tracyi. Festuca multiflora Walt. FI. Car. 81. 1788. ‘**Repens, paniculis erectis ovatis, spiculis 8 ad 40-floris acutis, floris angustis, acutis, fauce subplumosis.** This may refer to L. fascicularis, but the description is scarcely sufficient. This plant is not represented in Walter's herbarium, which is at the British Museum. Stems tufted, smooth. 3 to 12 dm. high, erect or procumbent. Leaves narrow, usually involute, 1 to 3 dm. long, 3 to 5 mm. wide; sheaths smooth or slightly scabrous. Panicles from a few cm. to 2 dm. long, more or less included in the upper sheath; branches of panicle few or several and of variable length, in the larger forms as much as 1 dm., appressed or ascending, or at maturity spreading. Spikelets usually somewhat overlapping, 7 to 12 mm. long, 6 to 12 flowered. Empty glumes narrow, acute, lower 2 to 3 mm. long, about one-half the upper; flowering glumes 4 to5 mm. long. with an awn of variable length, sometimes, especially in the procumbent form, as long as the glume; lateral nerves pubescent below. DistrRipuTion: Maryland to Florida and west to South Dakota and New Mexico. Texas: Jones 4203; Drummond 387. Kansas: Hitchcock 920. Florida: Nash 2306. St. Croix: Ricksecker 306. Cuba: Wright 3822, 3812. Meaxico: Pringle 813; Palmer 254, 691; Schaffner 683 (D. procumbens). LEPTOCHLOA IMBRICATA Thurb. Bot. Calif. 2: 293. 1880. ‘* Larkins Station, San Diego County (Palmer No. 404); Fort Yuma (Major Thomas); and through the Gila Valley to the Rio Grande.’ (PI. V, fig. 2; text fig. 14.) Diplachne imbricata Scribn. in Vasey Til. N. A. Grasses 1°: No.42. 1891. Trans- fers Leptochloa imbricata and gives a plate. Diplachne verticillata Nees & Mey. Noy. Act. Nat.Cur.19. Suppl. 1: 158. 1843. (Not Leptochloa verticillata Kunth, 1835.) ‘Ad Copiapo in republica Chilensi, Martio 1831,et ad Aricam Peruvie.’’ The authors remark that this species differs from Diplachne virens of Brazil (presumably Tridens virens Nees) and D. fascicularis in having the glumes not awned from the apex but very shortly mucronate and from the first in its larger spikelets. I have examined T. virens Nees and think it is not identical with L. imbricata Thurb. NORTH AMERICAN SPECIES OF LEPTOCHLOA. 19 Leptochloa virletii Fourn. Pl. Mex. 2: 147, 1886. San Luis de Potosi” (Virl., n. 1404). Type specimen examined at Paris. Rabdochloa imbricata Kuntze. Rev. Gen.3: 788, 1891. Transfers Leptochloa imbricata Thurb. Resembles in habit L. fascicularis Gray. The panicle is more oblong in outline, being more compact and with shorter branches, and often dark colored and more exserted. Spikelets also resembling L. fascicularis. but the empty glumes are broader and more obtuse, and the flowering glumes are somewhat apiculate but not awned. Fig. 14.—L. imbricata, DistrRiBuTion: Arizona: Palmer 548, 51; Lemmon 360; Vasey 540. California: Wright 2118; Coulter 776. Texas: Tracy 7367. Mexico: Palmer 47, 134, 331, 216,5; Mearns 2741. Argentina: Hieronymus 1088. Paraguay: Morong 981. There is a Leptochloa verticillata from the East Indies (Kunth Gram. 1: 91, 1855. Eleusine verticillata Roxb., Hort. Beng. 8. 1814). Diplachne tarapacarum Philippi from Chili appears to belong here, judging from the specimen in herbarium U.S.D.A. (Anal. Mus. Nac. Chili. Bot. 88. 1891.) LEPTOCHLOA SPICATA Scribn. Proc. Acad. Sci. Phila., IS91. 304, 1891. Transfers Diplachne spicata. (Fig. 15.) Fia@. 15.—L. spicata. Bromus spicatus Nees. Agrost. Bras., 471. 1829. ‘*‘ Habitat in campis, campo mimoso dictis, provincie Piauhianw.’? Nees observes that in habit this forms a transition to Brachypodium or Agropyron, but differs in the few nerved glumes; nor does it fit in Diplachne any better, since the native species has the glumes not at all apiculate, and foreign species differ much otherwise, 20 NORTH AMERICAN SPECIES OF LEPTOCHLOA. Tricuspis (Triplasis) simplex Griseb. Mem. Acad. Sci. and Arts. N. Ser. 8: 532, 1862. Plant. Wright. 2. ‘* In rupibus aridis,’’ Wright, 1551. Diplachne simplex Doell in Mart. Fl. Bras., 2*: 97. 1878. ** Habitat in proy. Piauhy (Gardner n. 2367).” Diplachne spicata Doell 1. ¢., 159. 1878. This isa correction. *‘Pag.97. Delea- tur Diplachne simplex; legatur Diplachne spicata, ut conservetur nomen specificum. ** Triodia schaffneri Wats. Proc. Am. Acad., 18: 181. 1883. ‘‘In the Escabrillos Mountains, San Luis Potosi (1077 Schaffner) closely resembling in habit the Cuban Tricuspis simplex of Grisebach and Diplachne spicata Doell of Brazil. It is clearly a Triodia as the genus is defined by Mr. Bentham.” Diplachne reverchoni Vasey. Bull. Torr. Bot. Club. 18: 118. 1886. ** Collected on granitic rocks, Llano Co., Texas, by Mr. J. Reverchon.”’ Triplasis setacea Griseb. in Goett. Abhandl., 24: 304. 1879. Plant Lorentz- jane). ‘Pr. la Merced. S.: ad fl. Juramento.”” In his remarks upon this species Grisebach says: ‘* Species 7. simplici Gr. (Pl. Wright. Cub., IT, p. 532) proxima.” Stems tufted, slender, 1 to 3 dm. high. Leaves usually about one-half the height of the flowering culm, numerous, narrow, slender, andinvolute. Infloresence reduced to a single spike, 5 to 10 cm.long. Spikelets 4 to 7 mm. long, several flowered. Empty glumes acute, flowering glume short awned. DISTRIBUTION: Terwas: Reverchon, 1613; Nealley, 78. Mexico: Pringle, 3267. Argentina: Hieronymus, 337. Brazil: Gardner, 2367. Diplachne loliiformis F. yon M. of Australia closely resembles this. Besides those species mentioned above are three described by Four- nier, which I have not seen. Copies of the original descriptions of these are here appended. LEPTOCHLOA LIEBMANNI Fourn. Pl. Mex., 2: 147, 1886. Culmo elato 2-3 pedali, valde ramoso, stramineo, glabro; foliis infra longe vaginan- tibus mellibus lanceolatis, 4-5” latis, ligula fimbriata; panicula longa stricta, radiis appressis. secundifloris; spiculis 4-floris, glumis inzequalibus, inferiore aucta dimidio breviore, superiore obtusa obscure trilobata, lobo medio mucro- nato; palea exteriore acuta carinata. Antigua, februario (Liebm., n. 248); absque loco (Liebm., n. 244). DIPLACHNE PATENS Fourn. Pl. Mex., 2: 148. 1886. Culmo a basi ramoso, ramis circulariter ascendentibus glabro, striato, stramineo, nodis brunneis. ligula hyalina acuta szepe laciniata, foliis longis linearibus angulo recto divergentibus, acutis; panicula invaginata, radiis alternis patulis flexuosis scabris, spiculis 7-floris; gluma inferiore dimidiam superiorem non zquante, exteriore violacea acuminata carinata scabra: rhachi inter flores flexuosa; palea inferiore carinata, nervo medio prominente acuminata, supe- riore duplo minore, bicarinata, obtusa, apice integra. Vera Cruz (Gouin, n. 93). LEPTOCHLOA PANICULATA Fourn. Bul. Soc. Bot. France (ser. 2), 27: 296. 1880. Culmo 3-pedali, cum nodis glabro: foliis latis brevibus, acuminatis, ligula brevi laciniata; inflorescentia pedali, axi panicule et radiorum scabro; radiis primariis primum patulis, dein divaricatis, in dimidia inferiore parte radiolos semipollicaresemittentibus; spiculis 3-4 floris muticis, floribus remotis, glumis wqualibus. palea exteriore bidentata mutica. Absque loco (n. 1079). NORTH AMERICAN SPECIES OF LEPTOCHLOA. 21 SPECIES EXCLUDED. LEPTOCHLOA BRANDEGEI Vasey =GOUINIA BRANDEGEI Hitche. (Fig. 16.) Fa. 16.—Gouinia brandegei. Callus on right. This was first described by Vasey (Proc. Calif. Acad., ser. 2,2: 213. 1889). This agrees with the other species of Gouinia in habit and in general floral strue- ture, such as the 1-nerved unequal empty glumes, the 3-nerved flowering glume, the rather long-pediceled rudimentary flower, and the hairy callus of the lower flower. It differs from the other species chiefly in the very short awn to the flowering glume. DISTRIBUTION: Lower California; Brandegee 7,9,11,38. Carmen Island, Mexico: Palmer, 362. Leptochloa rigida Munro= Eragrostis sessilispica Buckley. Leptochloa palmeri Vasey ined.= Gouwinia virgata Scribn. Leptochloa mexicana Scribn.=Gouinia mexicana Seribn. PLATE I. aT Ii. V. Wile DESCRIPTION OF PLATES. Fig. 1.—Leptochloa mucronata Kunth. Athens, Il. The usual form. Fig. 2.—Leptochloa viscida Beal. Mexican Boundary Survey, Mearns No. 793. Fig. 1.—Leptochloa domingensis Trin. Florida, Simpson. Fig. 2.— Leptochloa domingensis Trin. Hidalgo, Tex., Nealley. Fig. 1.—Leptochloa scabra Nees. Louisiana, Langlois. This is the specimen upon which was based Leptochloa Langloisii Vasey. Fig. 2.—Leptochloa nealleyi Vasey. Texas, Nealley. Type specimen. . Fig. 1.—Leptochloa fascicularis. The prostrate form that has been named Diplachne procumbens Nash. Denver, Colo., Letterman. Fig. 2.—Leptochloa fascicularis. The western form which has been named Diplachne tracyi Vasey. Reno, Ney., Tracy, 216. Type speci- men of D. tracyi Vasey. Fig. 1.—Leptochloa fascicularis Gray. Sheffield, Mo. Bush No. 804. The ordinary form. Fig. 2.—Leptochloa imbricata Thurb. Culti- vated in Grass Garden, U. 8. Department of Agriculture. Leptochloa floribunda Doell. The cotype of Diplachne halei Nash. Louisiana, Hale. A fragmentary specimen, but interesting because of its history. 24 Agricultur f Bureau of Plant Industry, U. S. Dept Fic. 1.—LEPTOCHLOA MUCRONATA. Fic. 2.—LEPTOCHLOA VISCIDA. PLATE Bul. 33, Bureau of Plant Industry, U. S. Dept Fic. 1.—LEPTOCHLOA DOMINGENSIS, FROM FLORIDA. Fic. 2.—LEPTOCHLOA DOMINGENSIS, FROM TEXAS. TE Ill. PLA Agr of Dept S U dustry Bureau of Plant In Fia. 1.—LEPTOCHLOA SCABRA. Fic. 2.—LEPTOCHLOA NEALLEYI. al ie tS Sa) we ane Bul. 33, Bureau of Plant Industry, U. S. Dept Fia, 1.—LEPTOCHLOA FASCICULARIS (DIPLACHNE PROCUMBENS Nash). Fic. 2,—LEPTOCHLOA FASCICULARIS (DIPLACHNE TRACY! VASEY). *(WHO4 AYVNIGHO) SIYVINDIOSVS VOTHOOLd3]— } “SI4 “WLVOINSW! VOIHOO1d3a—'S ‘SI4 Bul. 33, Bureau of Plant Industry, U. S. Dept of Agr PL ATE V. . ry eo pat i i an Bul. 33, Bureau of Plant Industry, U. S. Dept. of Agricultur Fia. 1.—LEPTOCHLOA FLORIBUNDA. <= ‘ = . 2 ‘ - r - .! = ~ a FRONTISPIECE. Bul. 34, Bureau of Plant Industry, U. S. Dept. of Agriculture ‘pAVMOH “OT ta AG ‘ZOGE ‘9S SUSY poydussoyoyd "ATV L| ‘AGHVEWO7 ‘SHSHSHLV5D 4v3q GNV SSauL AYYSSTNW iis DEPARTMENT, OF AGRICULTURE. BUREAU OF PLANT INDUSTRY—BULLETIN No. 34. B. T. GALLOWAY, Chief of Bureau. SILKWORM FOOD PLANTS: CULTIVATION AND PROPAGATION. BY GEORGE W. OLIVER, Expert, SEED AND PLANT INTRODUCTION AND DISTRIBUTION, Issuep JANUARY 15,:1903. ys . WASHINGTON: GOVERNMENT PRINTING OFFICE. 1908. LETTER OF TRANSMITTAL. U. S. DEPARTMENT OF AGRICULTURE, Bureau or Prianr Ixpusrry, OFFICE OF THE CHIEF, Washington, D. C., October 27, 1902. Sr: I have the honor to transmit herewith a paper entitled **Silk- worm Food Plants: Cultivation and Propagation,” by George W. Oliver, Expert, Seed and Plant Introduction and Distribution, and respectfully recommend its publication as a bulletin of this Bureau. The paper has been prepared at the request of Dr. L. O. Howard, under whose direction the funds appropriated at the last session of Congress for an investigation into the subject of silk culture in this country are expended. Dr. Howard has made a number of sug- gestions in regard to the scope and character of the paper, and has furnished the illustration used as a frontispiece, selected from a large number of photographs taken by him during the past summer while investigating the silk-cultural industry in Italy and other countries. Respectfully, B. T. Gattoway, ( hief of Bureau. Hon. James WiLson, Secretary of Agriculture. CONDE WES: Page. ILgnvawhintin@yal os-a4 sake asset seccd seco 4 Ake Aes ce Cae o ee SSOSsoe aes 7 Meno Os Orme PLOCUCHOUE seas ase eae ena eee cece ee eee ane 8 BrOpapaiOnlM vaCUiUCS Smee moe So oe san eee ean wanes =-- anc. 9 SUMO? OMAN ee ae de Scasce Base see CCl sete eases 9 VAY ILE GINS SS Cece sacocericg -US ced EeScee == 5o Rass eos Enea eee 10 Sere Uh y See ae eee eee Eee nc ge econ 10 Preparations for planting cuttings .-.......---.-----..---.----- 10 Thiayslovoyr i Oumbe re (RMN home ve coe agebe7 eRe ene ae CeCe ee 11 TAO (oye HOWL JOhy SCG Plo oehSee oe cor be Gabe Seo ee CES ee Soe See ee eee eee ll (Greetings eral LoVe Katey = oe oe Seon ne eR Se OO SEE See 13 RGU ERENNMNS A-Re origbectobocenol 0b net OU ee Coe Eee ae 13 PSC LOMON CGN tat pi OTA CLL erate ere eels reine Selanne een cca cases l4 SHNGHl Joe Ubilie oo ese edcisc ne sete cS pees SSE ane ee 15 Raising stocks for grafting and budding ...--..--.------------------ 16 lel) nosanctocuSS see eee psoodbet t GOS nUe nee nen ee See 16 TP Bitte =< Ge oo Boge Scie ee ake Aor SEP a Sh Os nae AIEEE 16 Prdunowel? » 55 2c bends seeee Sek Sane boo Soc so OSS psa ee 17 ESCnr Ts MON Olip AtOS be seater te eet seem wine ae ne c's aindww eines asiencee--- 20 TE LUST RAT TONS: Page. Mulberry trees and leaf gatherers, Lombardy, Italy............------ RronGeuiens Puate I. Branch of the white mulberry, Morus alba, with large undivided leaves ...-- 20. ceci55seecee sess eseeccce sess See ee nee eee 20 II. Branch of the white mulberry, Morus alba, with divided leaves. ---- 20 III. An ornamental variety of mulberry, Morus alba, variety venosa-.--- 20 IV. Leayes of seedling Russian mulberry, Morus alba, variety tatarica.- 20 V. The'native red\mulberry, Morusirubna 2 sss >= ee eee 20 VI. Paper mulberry, Broussonetia papyrifera. A.—Leat from old tree. B.—Leat from 2-year-old seedling. C.—Twig with female flowers- 20 VII. The Persian or black mulberry, Morus nigra....-...--------------- 20 VIII. Osage orange, Toxrylon pomiferwm. Leaves, fruit, and bark..--...-- 20 IX. Summer cuttings of the white mulberry, with leaves shortened. ---- 20 X. Winter cuttings of l-year-old shoots of the white mulberry, ready for planting... ..- J... .c25 .cucise 6 gee eee ee 20 XI. Root grafting the mulberry. A and B.—Scions fitted on stocks, ready to be tied. C.—Stock and scion wrapped and ready to be planted ........2.sseesccete ones cece: See ee eee 20 XII. Scion or sprig budding. A and B.—Scions prepared for inserting. C.—Stock with bark raised, ready for scion. D.—Scion in posi- tion, ready to be wrapped. E.—Stock with scion held in place by wrapping. F.—Stock waxed to exclude air and moisture. ---. 20 6 B. P. 1.—43. Ss. P. I. D.—30 SILKWORM FOOD PLANTS: CULTIVATION AND PROPAGATION. INTRODUCTION. There is a small family of plants closely allied to each other, a few of which supply the silkworm with food. This family is called Moracee. There are three genera of trees in the groun— J/orus, the mulberry (Pls. I, I, U1, 1V, V, and VII); Zozylon, the Osage orange (Pl. VUI), and Broussonetia, the paper mulberry (Pl. VI). The last named, being unsuitable for silkworm food, will not again be referred to here. The Osage orange provides palatable food for the silkworm, and if the worms were free to select the leaves for themselves the tree would be satisfactory; but the leaves are selected for them often with bad results, for the young and immature leaves have a tendency to sicken the worms. Ignorance of this fact renders the use of the Osage orange dangerous. Of the mulberry there are many so-called species and a great many varieties, but there are only one or two species and a few varieties which are of importance in silkworm propagation. Chief among these for producing silkworm food is the white mulberry, Jorus aha (Pl. I). This is thought by some to be a native of China. It has long been known that the white mulberry and its varieties are hardy over a large area of the United States. The uninitiated should not be left to their own devices in growing mulberry trees, especially if the enterprise is to be an extensive one, for if failure results, silkworm propagation in the particular section of the country where the experiment is conducted will receive a seri- ous setback. It is not the purpose of this paper to discuss the question of the most suitable varieties of the white mulberry, as this could only be done from a European point of view. Bureau, in his monograph, describes 27 varieties of the white mulberry alone. In Italy, silk- worm growers favor J/orus alba, variety moretti, and forms raised from it. France and Spain have each its favorite kinds. Japan has 8 SILKWORM FOOD PLANTS. close upon 100 forms, one or two of which would probably answer all purposes, while most of the silkworms reared in China are said to be fed upon Morus multicaulis. This mulberry was largely planted in the United States many years ago. Few, if any, of the original trees remain, but specimens which are thought to be wild seedlings of these are very plentiful in the Southern States. These trees are thoroughly acclimated and free from disease. It is therefore probable that there is now in the United States an abundant supply of material for propa- gating purposes, at least. It is intended to show in these pages how the mulberry may be prop- agated and grown so as to provide the maximum amount of leaves for the food supply of the worms. The white mulberry, under good cultivation, is a low-growing tree, seldom attaining a greater height than 25 or 30 feet. It will reach this height in a comparatively few years after planting. Although it will live to a good old age, its growth, like that of most other trees, is most rapid when young. As the trees attain their full height they become stocky and make a mul- titude of small growths, from which flowers and fruit are produced. The fruit, which is usually abundant, is not a favorite in this country, being generally considered too sweet and insipid. In shape it may be said to resemble more or less that of an elongated blackberry. In the vicinity of Washington the trees flower about the middle of May and ripen their fruit during June. METHODS OF REPRODUCTION. The usual methods of propagation in use for fruit trees are employed with varying degrees of success in the case of the mulberry. These methods consist of budding, grafting, layering, cuttings, and seeds. Grafting and budding are by far the most expensive methods, and it is doubtful if the results justify their use, so far as raising mulberry trees is concerned. Part of the work connected with budding and grafting consists in raising stocks, which are seldom large enough for use until they are two years old. At this age, the buds or grafts are inserted, and then troubles previously undreamed of present themselves to the inexperienced cultivator. Were the mulberry tree as easily managed so far as budding or grafting is concerned as is the peach, the use of these methods would be feasible, but unfortunately the mulberry is far from being an easy subject in this respect, and a few failures are apt to produce disappointment and disgust. It will frequently happen that old trees must either be removed or desirable varieties worked on them; budding or grafting may be resorted to in such cases. Layering consists in bending down a portion of a branch so that its stem after being notched may take root in the ground while still attached to the parent tree. It is a cumbersome method, however. METHODS OF REPRODUCTION, +) Although good-sized plants can be raised in a short time by its use, it is seldom employed when any other method will produce the same results. Raising young trees from cuttings of the 1-year-old ripened wood is a method which requires but little skill. As with budding and grafting, this method is instrumental in perpetuating varieties, as every rooted cutting will eventually be a reproduction of the tree from which it was taken. This is not the case with plants raised from seeds, which always vary considerably from the parent. For this reason some mulberry growers in Europe object to the seed method. Some of the seedlings, even from a single parent tree, will vary greatly in the value of the leaves for feeding purposes. Some will be thin in texture and lacking in the necessary chemical constituents; some, very hairy; others thick, smooth, and in every way desirable. However, experi- enced mulberry growers can readily tell the value of a seedling tree for feeding purposes, and it is therefore possible to make a selection in this respect without much loss. PROPAGATION BY CUTTINGS. SUMMER CUTTINGS. In any group of seedlings there will always be found individuals the leayes of which possess great adaptability for feeding purposes. These should certainly be propagated to perpetuate this desirable character- istic. Propagation should be started after the seedlings have made considerable growth in order to insure a good supply of wood. These plants should be increased by cuttings during the summer months. At this season it is advisable to retain some of the leaves on the cutting and give treatment which will prevent shriveling during the process of rooting. The cuttings should be made from wood as ripe as possible: the leaves, besides being well matured, should be healthy and free from noxious insects. During July the lower parts of the current season’s shoots will be found in good condition for propagating. Trim the cuttings similarly to those shown in Pl. IX. At least two leaves shortened to one-half their length should be allowed to remain on the cutting. When placed in the propagating bed, the slips should be inserted in the sand ina direction sloping from the operator. Good results will follow if a cool propagating house is used, with clean sand as the rooting medium. When a propagating house is not avail- able, a wide frame provided with sash will answer the purpose. The frame should face north, and if in the shade of trees, so much the bet- ter. The sash should be kept closed, so that a humid atmosphere may be maintained until the cuttings take root. After they have made a considerable quantity of roots in the sand they should be transferred 10 SILKWORM FOOD PLANTS. to beds in the open. The beds should be 5 feet wide. Place the rooted cuttings about 6 inches apart each way and water copiously until established, when they must be freely exposed to air and sunshine. WINTER CUTTINGS. The cutting.—Vhe principal supply of plants may be secured by propagating from cuttings, which should be made from dormant wood taken from the trees just after the leaves have fallen. In no case should the cutting wood be less in diameter than a quarter of aninch. The cuttings (Pl. X) should beabout 10 inches in length, mak- ing the upper cut about one-half inch abovea bud. The position of the lower cut is immaterial. The cuttings should now be tied in bundles of fifty and either stored for the winter or be immediately put out where they are to root. Where the winters are not too severe, or in the Eastern States south of the thirty-ninth parallel, they should be put in the ground during autumn. North of this it will be found best to keep them under coyer until the ground is in a condition to he worked in early spring. If they are kept even for a short time ina dry place, they will lose their sap and become shriveled. Therefore they should be buried in moderately moist sand or sand and ashes. Under such conditions a good callus will have formed around the lower cut surface before the time arrives when they are to be put in the open. If sphagnum moss be easily procurable, it can be used very successfully as a substitute for sand or ashes; but in this case the bundles of cuttings should be smaller and they should be placed with the buds pointing upward, the moss to be packed tightly around them, with the top part uncovered. This is an excellent method for indue- ing the formation of a good callus. Preparations, for planting cuttings.—Previous to putting the cut- tings in the open the soil should be plowed deeply, then harrowed and rolled until well pulverized. A furrow is made with a spade toa suflicient depth, a little sand placed in the bottom, and the lower ends of the cuttings placed on top. Fill in the soil to half the depth of the furrow, firm well with the feet, then fill in the remainder of the soil, leaving only enough of the cutting exposed to view to keep the top bud from being covered. Where there is danger of hard freezing weather after fall planting, cover the surface with rough stable litter or dead leaves, this covering to be removed before the buds begin to swell during the latter part of March. The rows of cuttings can be arranged in beds of any convenient width, leaving spaces between the beds; this arrangement will facili- tate covering, watering, and hand-weeding. If plenty of good ground is available, enough space should be left between the rows to permit of horse cultivation. During the summer the plants shouid be gone over several times and all superfluous shoots removed, leaving only ‘ CUTTINGS AND SEEDS. 1] one shoot to each plant. If large enough, the rooted cuttings should be removed to nursery rows the following fall. In no case should the plants be removed from the cutting beds to permanent locations. If the plants make sufficient growth the first season, they should be severely cut back; otherwise the operation should be deferred until the following season. The length of stem to remain as the future trunk must be regulated according to whether a dwarf or tall specimen is wanted. It must be taken into consideration that the leaves are much more easily gathered from dwarf trees than from tall ones; in fact, they are more easily managed, not only so far as leaf gathering is concerned, but also in pruning and in keeping noxious insects and fungus diseases under control. The leaves ona tall tree are not all developed alike; those on the side fully exposed to the sun will naturally be in perfect condition, while on the opposite side they are softer and probably not so well adapted to the purpose for which they are intended. Medium-sized trees are therefore preferable for all purposes. INDOOR SPRING CUTTINGS. Another method of propagation from cuttings, and a very success- ful one, consists in selecting medium-sized shoots about the beginning of November. These, before being made into cuttings, are sorted into bundles of different lengths, tied, and heeled in ashes or sand, or in a mixture of both, and protected by a frame having a northern exposure. During the winter they are taken out and cut into lengths of about 5 inches. These are tied in bundles and buried in moist sand or moss. In early spring they are untied and put quite thickly in a propagating bed having a mild bottom heat, where they will root rapidly. When such a bed is lacking, wooden flats about + inches deep may be used for the reception of the cuttings; but they must have the protection of a frame covered with sash. If a little loamy soil is placed in the bottom of the flats the cuttings will remain in good condition for a considerable time after rooting and until a favorable opportunity arrives for planting them out in nursery rows. If those rooted indoors are given plenty of air after being rooted in the bed, they can be transferred to the open ground with safety during dull weather. PROPAGATION BY SEEDS. The most convenient and rapid method of propagation is undoubtedly from seeds, as they are quick to germinate and the seedlings make growth about as rapidly as plants raised from cuttings. Seeds sown shortly after being harvested will germinate in a few days. If kept over winter and sown in early spring, the seedlings will appear within fourteen days. When the seed is spring sown, the seedlings will, if the weather be propitious, attain a height of from 12 to 18 inches in 12 SILKWORM FOOD PLANTS. one year; but during dry seasons they will only grow from 6 to 12 inches. Seedlings from seeds sown immediately after the fruit ripens are always small at the end of the season, but they produce strong plants the season following. Seed is usually produced in great abundance by nearly all of the Species and their varieties. The mulberry, like the strawberry, black- berry, and raspberry, does not ripen all of its fruit at one time; con- sequently several gatherings are necessary before a crop is harvested from any one tree. The earliest fruits can be gathered immediately after they are ripe and the seed sown if desired. It should be remem- bered that seedlings thus raised have comparatively little time to make their growth; therefore, every day counts. In gathering the fruit, it will be found easiest to shake the tree and pick the fruits from the ground. To remove the seeds from the sur- rounding pulp, put the fruit into a large bucket or tub and squeeze with the hands until it becomes a jelly-like mass. Add water and stir well until the contents are thinned sufficiently to allow the seeds to sink to the bottom. The remaining material can be poured off. The seeds should be exposed to the air until dry. If it is desired to sprout them the same summer, they should be sown in beds in the open, the soil of which should previously be well worked by deep plowing and gone over several times with a harrow and a roller. When the soil is sufliciently pulverized the ground should be marked off into beds 5 feet wide and of any convenient length, leaving a space of 2 feet between the beds. To prevent washing of the soil and also to mini- mize the evil effects of drying winds, drive some stout stakes into the ground along the sides and ends of the beds, and to these nail eight or twelve-inch boards. The surface of the bed should be leveled and all stones and roots of plants removed with a hand rake. Sow the seeds broadcast, taking care not to sow them too thick, as there is a danger of the seedlings crowding each other. Crowding produces weak plants, because even the best soil is capable of sup- porting only a certain number of plants to the square foot. Press the seeds into the soil with the back part of a spade and cover lightly with soil screened through a quarter-inch sieve. In order to have the best results, the seed beds should not be exposed to the sun until a considerable time has elapsed after germination. This condition may be arranged as follows: Procure some pieces of 2 by 38-inch scantling; place two of the pieces parallel to each other 54 feet apart. Nail laths from one to the other, using the 2-inch surface. in which to drive the nails. Leave l-inch spaces between the laths. The slats are put lengthwise over the beds, and can be used with or without the side boards. Over the slats spread archangel mats, or can- vas, until germination takes place; these coverings should be fre- quently dampened. After the seedlings show above the ground, the GRAFTING AND BUDDING. 13 cloth coverings are to be kept on during the hottest part of the day only, and when the first true leaf appears they may be removed alto- gether and the shade necessary thereafter supplied by the lath slats. Water must be supplied if the soil needs it. With spring-sown seed, the coverings over the lath slats may be dispensed with, but the sur- face of the bed should not be allowed to become dry until the seedlings are large enough to take care of themselves. GRAFTING AND BUDDING. In Italy and other silk-raising countries it is claimed that the leaves of trees raised from cuttings and seeds are superior for silk produe- tion, but that the quantity of leaves produced by trees so propagated is only about one-half the bulk of those from grafted or budded trees. Therefore, to produce a large quantity, grafting and budding methods of propayation are practiced to a great extent. Before the beginner undertakes these expensive methods of propagation in the United States, however, he should consider that land rentals are high in Europe and that land is cheap in the United States; therefore the American can afford to grow more trees by the methods which are instrumental in giving the best grades of silk. This is an important point to consider, and the writer is inclined to the belief that in the propagation of plants giving the highest grades of silk there will be little danger of a scarcity of material, as the mulberry thrives as well, if not better, in most parts of the United States as anywhere in Europe. For those who decide to try propagating by grafting and budding two of the most successful methods of performing the operation are here described. ROOT GRAFTING. This is performed in February and March. The stocks, which are two-year-old seedlings of the Russian mulberry (Jorus alba, variety tatarica), should show a diameter of at least three-eighths of an inch to give a satisfactory union. The stocks should be lifted in the fall and ‘* heeled in” out of the reach of frost. The scions should be cut while in a dormant state and buried in damp sand in a protected place. In the latter part of February the work of root grafting (Pl. XI) may be started. The preparatory work consists in securing a quantity of strong tidy cotton, and of grafting wax made of beeswax two parts, of resin two parts, and of mutton tallow one part. Put the ingredi- ents in a small tin bucket, place on a hot stove, and when melted drop in one or more balls of the cotton, allowing them to remain in the melted wax for five minutes; remove with a pointed stick. When cool they are ready for use. Procure a deep box in which place the stocks, keeping them covered with a dampened sack; another box 14 SILKWORM FOOD PLANTS. should be provided for the scions similarly protected, and a third one for the grafted roots. These precautions are necessary, as a little exposure to dry air is always detrimental. In beginning work with the stocks sever the top from the root at the collar; this can be done best with a pair of pruning shears. Take a scion atleast 8 inches long and attach by the tongue method, as shown in Pl. XI. Select stocks and scions of as nearly the same diameter as possible; make a slanting cut at the bottom of the scion and a similar cut at the top of the stock. In the case of the scion, make an upward incision at a point about one-third of the length of the cut surface from the base; this will forma tongue. Next make a corresponding incision downward near the top of the slanting cut on the stock. The idea is to have the tongue of the scion take the place which the knife blade occupies when making the incision in the stock. When the two parts are neatly fitted so that the bark of stock and of scion come neatly together at one side, or at both if possible, bind firmly with the waxed cotton. This material should be used in preference to raflia, because when the grafted stock is buried in the ground, raffia would be certain to rot before the union took place, while cotton will remain in good condition for a long time. After the fitting and tying have been done, the grafted stocks should be tied in bundles of twenty-five, the first tie to be made rather firmly near the upper part of the scions; secure them again near the base of the scions, but not as firmly as before. Care must be taken so as not to displace the fitted parts. The bundles should now be buried in sand in a frame or other protected place until planting time arrives. The grafted stocks should be planted out just as soon as the condition of the soil will permit. Plant them deep enough so that only the top bud is exposed to the light. The subsequent treatment is in all respects similar to that given for cuttings. Mark the kinds, with the dates of grafting and planting, on large labels which will not be easily displaced. SCION OR SPRIG BUDDING. Scion or sprig budding, as shown in Pl. XII, is perhaps the most successful and easiest to accomplish of all methods. It is practiced on stocks which have not been transplanted for at least one year previous to the time when it is desired to bud. The stocks should be larger than those used for root grafting. The most desirable time for the operation is in spring, when the bark lifts easily; this will necessarily be after the stocks come into leaf. The scions must be selected from shoots of the previous season’s growth, short and stocky, with two buds present (Pl. XII, A and B). They should be cut from the parent plants in the falland kept dormant until the opportune moment arrives when the stock plants are in a receptive condition. SCION AND SHIELD BUDDING. 15 In preparing the stock for the scion the preliminary work is similar to that in shield budding the peach, cherry, or rose. Ata point a little above the collar of the stock a transverse cut is made through the bark for a distance of half an inch or more around the stem (PI. XII, C.) This is followed by a longitudinal cut, beginning in the middle of the first cut and extending downward for about an inch. Prize up the bark at each side of the long cut (PI. XII, C) and it is ready for the scion, which is prepared for insertion by making an oblique cut through the base, so as to leave a cut surface about an inch long (PI. XI1, A and B). The scion is then fitted in place so that its cut surface is neatly placed against the wood of the stock (Pl. XII, D) laid bare by the raising of the bark. The next operation is shown in Pl. XII, E, and consists in tying the parts together so that they will be held firmly while the union is taking place. In order to exclude air and moisture, grafting wax or clay should be applied, as shown in Pl. XII, F. Within two weeks from the time of budding, the union will be effected, if everything has gone well. The ligature should not be removed, however, until there is danger of its cutting into the bark. The most essential part of the subsequent treatment consists in head- ing back the stock, so that the future head of the tree will be formed by the growth of the scion, and to do this successfully good judgment must be exercised. Cut off only a part at first, leaving some foliage on the stock until the buds on the scion begin to push, when that part of the stock above the union should be removed with a sharp knife. Cover the wound thus made with grafting wax. SHIELD BUDDING. The shield system of budding may be used, but only in the spring, as the mulberry does not take kindly to shield buds inserted during the season suitable for most of our fruit trees. Shield budding consists in selecting a stock, either a branch or stem, from which the bark slips readily. In raising the bark of the stock for the reception of the bud, the work is similar to that described for scion or sprig budding. The bud is usually selected from dormant wood kept over winter in ashes or sand; but for this there exists no necessity, because there is always present an abundance of dormant buds on a growing plant, and these answer the purpose much better than buds from dormant wood. To remoye them, with a sharp knife make an incision in the stem about five-eighths of an inch below the bud; bring the blade up under the bud, severing a section of bark three- eighths of an inch in width, with the bud in the center; bring the blade out a little above the bud. If this operation is neatly performed the bud will require no further trimming before being inserted under the bark of the stock. The bark of the stock is then firmly bound over that of the bud and the parts kept in position with raflia. No 16 SILKWORM FOOD PLANTS. waxing is necessary. The union should take place within fifteen days, after which the ligature should be loosened or removed as proves necessary. RAISING STOCKS FOR GRAFTING AND BUDDING. In grafting and budding from any particular yariety which it is desired to perpetuate, the Russian mulberry, J/orus alba, variety tutarica, is the one used as stocks. It is of a robust-growing nature and has been found well adapted to the soils and climates of all the agricultural belts of the United States. It is this variety that is so much used in the West and Northwest for hedges, as it is the hardiest of all the mulberries. Stocks are best raised from seeds, and a supply for this purpose should be obtained from a reliable source, to avoid unnecessary delay and disappointment. The sowing and the subsequent management of the seedlings are the same with stocks as with seedlings for general planting, except that when planted in nursery rows they should be placed about a foot apart, so as to give an abundance of space for the operator to work. SOIL. So far as has been ascertained, the mulberry is not particular as to the character of the soil. It seemingly grows equaliy well in a great variety of well-drained soils. Even in sandy and gravelly situations it holds its own. In shallow soils over hardpan the mulberry thrives after most of our fruit and ornamental trees have given up the struggle. Under the same conditions the Persian mulberry has been found to fruit abundantly. Notwithstanding its behavior under what would be supposed adverse conditions, there are few plants which respond more vigorously to applications of manure. In Japan it has recently been shown that by liming alone the percentage of fiber in the leaves decreased very per- ceptibly. Again, by liming and also manuring with sodium nitrate and calcium sulphate a still further reduction in the fiber was appar- ent. The trees operated on were 14 meters (5 feet) high. Each tree was treated with 500 grams (1.1 lbs.) of lime, 400 grams (.9 lb.) of sodium nitrate, and 200 grams (.44 lb.) of calcium sulphate. How the cater- pillars fared asa result of this change in the composition of the leaves is not stated. PLANTING. This all-important operation may be performed either in the fall or spring. After the leaves have fallen or are matured, no delay should oceur in transplanting to permanent positions. When this period is ‘selected, it gives good opportunities for the formation of new roots. PLANTING AND PRUNING. 17 In spring the trees may be transplanted any time after the ground is in a workable condition and up to the period when the buds are about to burst into growth. Spaces intended to be planted should be deeply worked beforehand by plowing and harrowing, and after planting the weeds should be kept down. The distance between the trees should not be less than 10 feet in the rows, and the rows should be the same distance apart. If the field devoted to the trees is more than 2 or 3 acres in extent, wider spaces should be left at intervals for wagons, etc. It is certain that trees planted 10 feet apart will eventually occupy all the space; but when there is danger of their becoming too much crowded, enough of the plants may be rooted out and burned to allow the remainder abundant space to develop. If this is done, those which are to remain permanently should be trained accordingly. The above arrangement is the best for trees nearly all the branches of which can be reached from the ground, not only for pruning, but also for leaf gathering. In planting trees similar precautions should be taken to those in the case of ordinary forest trees; that is, not to allow the roots to become in the least dry from the time they are lifted from the nursery rows until planted in the field. As soon as they are lifted the roots should be dipped in a mixture of soil and water and kept covered until planted, so that they will not become dry. If the ground is naturally hard and the soil is poor, dig large holes, even for very young trees, as they grow rapidly and should be encouraged to make good, stout growths from the beginning. Put some good soil in the hole, spread out the roots on this, and cover with several inches of tine soil before firming with the feet. Allow the roots to be about the same depth in the hole as they were in the nursery rows. Prune back the growth of young trees one-half in the fall, and if necessary cut back to strong buds in early spring. PRUNING. The pruning of the trees presents no special difficulties so long as it is done early enough in the season to avoid late growth, which, if caught by cold weather before ripening, will perish during the winter. The principal pruning should be done in winter and should consist of shortening back strong growths so as to forma low, spreading tree. Keep the central part of the tree as free of growth as possible, to admit light and air. After the first cutting back, select three or more of the strong shoots to form the principal branches. If they are strong and show a disposition to grow upright, they may be kept apart by using three sticks tied in the shape of a triangle; place these in the center of the tree and tie the branches to them until they grow in the desired 11805—No. 34—02——2 18 SILKWORM FOOD PLANTS. direction. By careful attention to cutting out the undesirable growths the tree can be made to assume any desired shape. In gathering the leaves always allow enough to remain on the tree to insure its perfect health. If some of the trees show signs of fail- ing vigor as a result of excessive leaf gathering, it is advisable to allow them to grow for a season without picking, and by early prun- ing out of unnecessary growth permit those growths which are desir- able to become ripened. 19 DESCRIPTION OF PLATES. Frontisprece.—Old mulberry trees, showing Italian method of pruning, with a group Pratt I. Il. Ii. IV. Vil. Vill. of embryo silk culturists (leaf gatherers) in the foreground, Lombardy, Italy. By this method of pruning, tall trunks from 8 to 10 feet from the ground are produced, necessitating the use of ladders for leaf gath- ering. From a photograph taken August 26, 1902, by Dr. L. O. Howard. Branch of the white mulberry, Morus alba, with large undivided leaves, of thick texture and smooth surface. The leaves of this variety are pre- eminently adapted for silkworm food. From photograph of a tree in the grounds of the U. 8. Department of Agriculture. ; Branch of seedling white mulberry, Morus alba, with divided leaves. Seed- lings from the same parent will sometimes have leaves of the divided form, others assuming the undivided shape shown in Plate I, while some may have both forms on the same tree. An ornamental variety of mulberry, Morus alba, variety venosa. Of no value as food for silkworms. Leaves of seedling Russian mulberry, Morus alba, variety tatarica. This mulberry, owing to its extreme hardiness, is used for stocks on which to graft or bud the most valuable varieties in order to perpetuate their characteristics, propagation from seed being altogether unreliable for perpetuating varieties. The native red mulberry, Morus rubra. From aspecimen in the Herbarium of the U.S. National Museum. The varieties of this species are usually prized for their fruits, being of little value as food for silkworms. . Paper mulberry, Broussonetia papyrifera. Valueless in silk culture. A.— Leaf from old tree. B.—Leaf from 2-year-old seedling. C.—Twig with female flowers. The Persian or black mulberry, Morus nigra. This species is cultivated in Europe and Asia for its fruit. From photograph of a tree in the grounds of the U. S. Department of Agriculture. Osage orange, Toxylon pomiferum. Leaves, fruit, and bark. The mature leaves of this native tree provide excellent food for silkworms. . Summer cuttings of the white mulberry, with leaves shortened to prevent excessive evaporation. . Winter cuttings of 1-year old shoots of white mulberry, ready for planting. . Root grafting the mulberry. A and B.—Scions fitted on stocks, ready to be tied. C.—Stock and scion wrapped and ready to be planted. . Scion or sprig budding. This method of propagation can be used on strong seedling stocks or on branches of trees. A and B.—Scions pre- pared for inserting. C.—Stock with bark raised, ready for scion. D.— Scion in position, ready to be wrapped. E.—Stock with scion held in place by wrapping. F.—Stock waxed to exclude air and moisture. 20 O Bul. 34, Bureau of Plant Industry, U. S. Dept. of Agriculture PLATE BRANCH OF WHITE MULBERRY (MORUS ALBA), WITH LARGE UN BRANCH OF WHITE MULBERRY (MORUS ALBA), WITH DiviIDED LEAVES BRANCH OF WHITE Mt LBERRY Mort ‘ i $ i. Oy _- Bul, 34, Bureau of Plant Industry, U. S. Dept. of Agriculture, PLATE IV. LEAVES OF SEEDLING RUSSIAN MULBERRY (MORUS ALBA), VARIETY TATARICA, "9 Bul 34, Bureau of Plant Industry, U. S, Dept. of A ulture BRANCH OF THE NATIVE RED MULBE TRY Mc -— i :0a1] PlO WO1y Bey “¥ “(WHSSINAdWd VILSNOSSNOYG) AYYSEINA YadVd ‘9 ‘sullpaas P[O-1BAA-O.N7 WIOIT TROT “Ef “SIOMOY OPBULOS WIEN SrA Bul. 34, Bureau of Plant Industry, U. S. Dept. o f Agr AE PLATE Vi. * Bul. 34, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE VII. BRANCH OF THE PERSIAN OR BLACK MULBERRY (MORUS NIGRA Vill. PLATE Bul. 34, Bureau of Plant Industry, U. S. Dept. of Agriculture LEAVES, FRUIT, AND BARK OF THE OSAGE ORANGE (TOXYLON POMIFERUM). ab Bul. 34, Bureau of Plant Industry, U. S, Dept. of Agriculture SUMMER CUTTINGS OF WHITE MULBERRY, WITH LEAVES SHORTENE Bul. 34, Bureau of Plant Industry, U, S. Dept. of Agriculture PLaTE X. WINTER CUTTINGS OF ONE-YEAR-OLD SHOOTS OF WHITE MULBERRY, READY FOR PLANTING opie! + Bul. 34, Bureau of Plant Industry, U. S. Dept. of Agriculture Plate Xl. Root GRAFTING THE MULBERRY A and B, scions fitted on stocks, ready to be tied; C, stock and scion wrapped 4 PLATE All. J Bul. 34, Bureau of Plant Industry, U. S, Dept. of Agric SCION OR SPRIG BUDDING. A and B, scions prepared for inserting; C, stock with ark raised: D, scion ready to be wrapped: E, scion wrapped; F, stock waxed. U.S. DEPARTMENT OF AGRICULTURE, BUREAU OF PLANT INDUSTRY—BULLETIN NO, 35. B. T. GALLOWAY, Chief of Bureau. RECENT FOREIGN EXPLORATIONS, AS BEARING ON THE AGRICULTURAL DEVELOPMENT OF THE SOUTHERN STATES. BY S. A. KNAPP, Specian AGENT. SEED AND PLANT INTRODUCTION AND DISTRIBUTION. Issuep Fesruary, 14, 1903. ts \ —= Iss WASHINGTON: GOVERNMENT PRINTING OFFICR. 1903. LETTER OF TRANSMITTAL. U. S. DEPARTMENT OF AGRICULTURE, Bureau or Piant INpusrry, OFFICE OF THE CHIEF, Washington, D. C., September 18. 1902. Str: I have the honor to transmit herewith a report on ** Recent Foreign Explorations, as Bearing on the Agricultural Development of the Southern States,” by Dr. S. A. Knapp, Special Agent, Seed and Plant Introduction and Distribution, and recommend that it be published as Bulletin No. 35 of the series of this Bureau. This report has been submitted by the Botanist in Charge of Seed and Plant Introduction and Distribution with a view to publication. Respectfully, B. T. GaLtLoway, Chief of Bureau. Hon. JAMES WILson, Secretary of Agriculture. oo ke Rea CE The introduction of Kiushu rice by the Section of Seed and Plant Introduction of the United States Department of Agriculture in 1899 was the first step taken toward improving the conditions of rice grow- ing in southern Louisiana and Texas, and the marked development of the rice industry since that time is in a large measure due to the value of this variety. There were still other problems connected with the rice industry, however, as well as those which concerned the improve- ment of extensive tracts of pine lands occurring in many of the Southern States, which remained unsolyed. These problems could be best approached by first securing all the information available in foreign lands, and Dr. 5. A. Knapp was commissioned to go to Asia to make a careful study of the rice industry and to secure such seeds as he might decide were valuable. Dr. Knapp’s report deals with the life of the peoples among whom he traveled, as well as with the methods and cost of rice production and the cultivation and production of certain other crops, and alto- gether it constitutes a unique contribution to our knowledge of the agriculture and the condition of the farming communities of these countries. The report is submitted for publication as Bulletin No. 35 of the Bureau of Plant Industry. A. J. PIeTers, Botanist in Charge. OFFICE OF SEED AND PLANT INTRODUCTION AND DisTRIBUTION, Washington, D. C., September 12, 1902. CONTENTS), UENO wp oc enim sosane Coca cad ec ee PESO Ren S RCO BBA CRS Cease Oe eae LN ariiegeUl SORT ENTE 0) Oe ae ee OSS CC) BOO LOO 2a eee eee Creare anolevie OuOltOOU COPS! = se saesee= mae eae a= o--~-------25--—= (Chika WG2, det ee eee aor sneee a enbeee Soccer WIDE. const esc0en car Cece OEE Bene EBERT a Eee oa IN Taal WEP Sa cen ose ScaceteHeesence nconn edt assess ec c= eeeeeee nae (Glefiton Tere Ce) Sk eee Gs See AS Sa See Bee Eee Shima WG: 3. Be Sg eSeSe ABR aa Be eee ROOn Cone Ce SE eee (Geral Minn ee Roo bee eee oop ae Et CE EOE ees (ChagllOl coaceSacsioe soetecn SoC see Shee Ceo DD Er Hoe Co enSe se Cee Agriculture ncsonseicecaeccteteebel Sect! cece Se speee Witay SOFAS) we oa sk OSG E RQ RR ESE CERES CSN C006 Ca SOCe BE 6 SE soe EES eee IRamiial ye REE = aoc Coepsonpe coor eeC Ee doe Se Sor SOC E Ee See TRteviilittiyy (ont Hae) GOW as coe Seige Saas eS Ose SEC CH OR ee COC et eae eee REE CMMI TET INCH MP ae a eee ee Pane ce leans ce Seo e cee ee eee (Clasavencicetisll RerIWTTS Cage cic oGtn OEE DOSES BE OA CSO Re nee ae ee (Civoya) TOLGHMONN ~ Sosegace ace ban SOC HOPES RES ae eee Ee at ollie nGiil) se sSon lobo. ada ee Seo DOR Eee noe ee eee OAV GMI ES) 2 canned ond Sa CaSp en ee So Sonn Ons ne cee es IDIN 228... shonomtoce eiccts SScedee RSS Ses = (Chitin) IN@UEES 4 naacnc se borsSelO SSS Ser se ease Sees WEES) oc to an SO SEE GoEE aOR SESE Spe SiC =e aes Tone intl! ARC oe See be Ogee SS SOO Eo Ea Siciabiny Gin AGIA TEEN) oS Seep ee eee EE eee eee eee SRT ESE CSE eee ele es Ce he ee IU oie vec ecics Swiss ~ann= Treatment of the seed bed and manuring.............--.------------ Dele wun oven CLaLeM LEMP eee oo tec ee eee coo ae cece arc cs ease sence RVPeH OCS ON OUI URE DOM Meee en nee oe aan se ncew cn cct cane cose naw es Io hoe PARE io So So eee AE OAR OEE I ee THERA AME seine SoS ce Seg iG 5 SACRO AEE ee NAb Sot Hea oe Sek Cas OR ERE WING? Scots Geet esac Sodas RSE cee ee (CHislE he ABTA Ce Se Set eS ee NOHUETERIINLETNT YG OLC ni nGene eee Meena oo oy Ceca et ew se econ (Clovnstayany oliloyevcay NUVI ei le a Cea INR GLE A ETO NG Be SS SS CE SS ENV) wha a Han EO A Se Sess sis Ae SE es bo bo bo te to to ty lS bY bo bo bo bo ty bg be ~' wo OF Go Go Oo to lo le bo to Si) CONTENTS. India—Continued. Siveistock and ‘farmimplementssea2 cen eae aasieeiecen ae ose eee eee seers Wells: 22. 2nceec< «ssa eo eee atins see eee eee a See ae nos eee Rice produced...) -.-S. cs Ss ask kee So aaa nae ee oe eee Agriculturein the: Punjabeocssss- se o-ccees =e eee eee eee ee eae Cost of living: = 2-22 s222 3 iss cb ces2 sc ons cb eicee sees = sce See ee eee Rice farmingyin Lower-Burmass---¢ -o2-6-e 5-2 - eee ae eee Rice milling'232: 25-222 20 see choos oo eeee eee eee eee Rice for foreipmimankets tas. 4- saa2cs 225 coe eee eee ee ee Selection of seedsusce. 22 anos sae hese tee eee ae eee (0) cht 0): Ee eee Net eee eee PCE n mms ha Hn eed nosccs et Agricultural conditions!ss=-5.— 2... 30 aoe ee eee ae eee Tillage of the s0Uli22 22-2 socc.222 Sacens cedede ees seat eee see eee Irrigation s..2:2:2 56 j2cce sss oss seb s ets eels = See eee eee ee Cultivating, harvesting, and thrashing rice. ......------522-2. 2. --s2sse= Hulling rices.< 22465 a5 ss a5 see ons aoe, = eee eee Production ‘and! cost;of malling. mcCe==- 4... 242-4 ae ee Cost of ‘building sete. 5.3.22 26-6 scatesaeteins ae ee eee Ex portationiotagricultural prod Gts 2/5 aj = oe eee eae The Philippine Tslands: 5s. ~-.42 4 s<25 1S ase cis eee eee eee Rainfall cose soe eee so 2 ss eee ae Ce ee See - Temperatures. 2a 3sscie © Sas soe Sere a ee ee Rangeiofproducts:. 222. csc coi ee eee sesSStos Stock and! pasture lands. <<< oThe statement regarding rice refers to this product with hulls removed, and for comparison with paddy about 20 per cent should be added. The acreage devoted to rice in Japan can not be very much increased, The islands are of volcanic formation, and in a general way it may be stated that a rather bold range of mountains traverses them from the southwest to the northeast, occupying seven-eighths of the terri- tory. The remaining one-eighth consists of fertile valleys, widening toward the sea until they gradually expand into coastal deltas of con- siderable extent. The narrow valleys are terraced on each side; at the base of the mountains canals are made to receive the descending rivulets and conyey the water to the various fields as required for irrigation. Frequently the surplus water is used to turn an overshot wheel for milling rice or for manufacturing purposes in the native villages, or it may be allowed to flow into some creek or river, but as far as possi- ble sufficient mountain water for irrigation is conducted by canals ata level somewhat higher than the rice field. (PI. I, tig. 1.) The inge- nuity displayed in devising the elaborate system of irrigating canals 14 RECENT FOREIGN EXPLORATIONS. and the amount of patient industry required to construct them are simply marvelous. The extent of the retaining walls constructed to prevent the washing of the terraces, or to arrest mountain slides, or as barriers against a river bent on destroying a field, is inconceivable. These are the works of a patient and industrious people throughout many generations. Occasionally water for irrigation is elevated from a creek or river, but almost invariably by the simplest machinery, such as has been employed for hundreds of years. One of the simplest machines for elevating water in common use is a wooden wheel 6 to 8 feet in diam- eter and 12 inches wide, with buckets on the perimeter, or rim. The power that raises the water is the weight of a man traveling on the buckets on the side of the wheel opposite the buckets lifting the water. It is so adjusted that the weight of the man on one side of the wheel is a little more than the weight of the water raised by the buckets on the other side; hence the wheel revolves. When the water reaches the required elevation it is discharged into a spout. METHODS OF RICE CULTURE. Rice production in all oriental countries is conducted upon the same general plan, but the methods differ so materially from those employed in the United States that they should be carefully noted. The lands are divided by levees into small fields. These are of no regular form, and generally the inclosing levees are gracefully curved to represent some ideal of beauty in the mind of the planter. In the small valleys among the mountains these curved embankments were doubtless nec- essary to conform to the mountains and thus to inclose a larger area, but as the improvements encroached upon the lowlands curves contin- ued to be used. The levees vary in width from 1 foot for field divi- sions and paths to 4 feet wide for main embankment roads. This system of levees and fields has precluded the use of domestic animals in the preparation of the soil and harvesting of the rice. The Japanese are fully aware of the disadvantages of having such small and irregular fields, and have made strenuous efforts to relieve the situation. Many of the rice fields in Japan average scarcely more than 35 feet square, and the boundary levees haye such wavy lines that they look as if made by hogs in a frolic. Under modern conditions the horse and the ox could be used in tillage, but there are no paths which such animals can traverse to these minute fields; and if there were, the tracts are too small for the use of plow or harrow, because there is not room to turn, much less to follow the angular boundary lines. If a farmer owns several tracts it is seldom that they are adjacent, and hence he is helpless to institute reform. Many progressive Japanese farmers have tried to institute reforms, but under the old law changes in land boundaries required the unanimous consent of the owners, LAND TRAOTS, JAPAN. 15 which it was practically impossible to secure. This was precisely the situation of the lands belonging to the yeomanry of England until about the commencement of the Nineteenth century. Three years since a law was passed by the Japanese Parliament that if two-thirds of the owners of a tract of land agreed to reform the boundaries the minority must concur. Still the farmers of Japan were conseryative, and only two or three provinces have made any considerable progress. The accompanying diagrams present a striking example of the land situation and the reform accomplished in one locality. LS SSS ee Ye SS PETE sue ge el B (| [] Fie. 2.—The same tract shown in fig. 1, redivided into 188 regular fields. Fig. 1 is a plat of a tract of land at Masuda village, near Sendai, and shows the little fields as they have been forages. Fig. 2 is of the same tract readjusted under the reform movement. Mr. J. H. De Forest, of Sendai, who furnished the maps from which these illustrations were made, states that this tract as platted contains only 25 acres and for- merly had 409 irregular fields in it. (See fig. 1.) There are now (see fig. 2) only 138 regular fields, with perfectly straight water courses and roads wide enough for two loaded carts to pass. Even 16 RECENT FOREIGN EXPLORATIONS. thus enlarged these fields are small indeed as compared with those in the United States, but it is a great advance for Japan. Such reform as this will greatly facilitate the use of cattle in plow- ing the wet fields and in carting out the crops. But, more than this, the area of arable land is greatly increased by breaking down the numerous grass ridges and throwing their space into productive soil. About one-tenth is thus gained, or 2 acres in the plat figured; and as 1 acre averages about $175 in value, the entire gain is over $350. But the whole expense of this reform was only $400, so that it almost paid for itself in the value of new space gained, to say nothing of the less- ening of manual labor. Japanese farmers are beginning to see that American methods must be more and more considered if they are to keep pace with agricul- tural advance all over the world. FIELD WORK. The fields of the Japanese farmers are generally well drained and thoroughly tilled, mostly with the spade or mattock. Both of these implements differ from those used in the United States. The mat- tock has a blade about 16 inches long and 5 inches wide, with a handle tor 5 feet long. The implement weighs 7 or 8 pounds. With a quick, powerful blow the blade is driven into the soil about 14 inches; then, using the handle as a lever, the soil is disintegrated and partially inverted. The spade is a wooden blade about 2 feet long with an ordinary handle; the lower end of the blade is cased with steel, and upon the back of the upper end is a block the width of the spade. The spade is thrust into the soil by the foot at an angle of about 30°, and, using the block for a fulerum, the soil is rolled to one side, as in plowing, but it is more thoroughly disintegrated. All the trash, straw, or grass upon the field is turned under, together with such an amount of lime, ashes, fish manure, or human excreta as the farmer may be able to secure. Where a winter crop is raised the manure is generally applied in the fall. If the rice field remains fallow during the winter the manure is applied at the time of spring working, in March or April, according to conditions. The seed bed is prepared as early as convenient in the spring, about April 1, thoroughly manured, and is given the care of a bed in the garden. It is spaded 8 inches deep and worked until the manure is thoroughly incorporated and all clods pulverized, after which it is surrounded by a low ridge and water is admitted to fill the soil until the spaded earth becomes consistent mud. The seed, which had been previously selected for purity, size of grain, and flinty character, is then soaked in pure water till well sprouted, which usually requires two days, and is then sown on the bed broadcast as thickly as admis- sible for strong plants. Prior to sowing the bed is covered with water JAPANESE RICE FIELDS—CUTTING RICE. 17 to the depth of 24 inches. In five or six days the rice is well started. It is then left dry in the daytime and is flooded at night. Covering with water at night keeps it warm, and allowing the bed to become dry in the daytime admits air and prevents sun scalding, which fre- quently occurs when the rice is young and the covering of water is shallow. Early in June, when the rice is 8 or 10 inches high, it is pulled up, tied in bundles of 6 to 10 plants, and transplanted into fields, which have been prepared and flooded to the depth of 14 to 2 inches. (PI. 1, fig. 2.) The rice plants are set in rows about 1 foot apart and at a distance of 10 to 12 inches in the row, on the richest lands, making ! bunches to the yard. On poor lands double that number might be set. They are so set that the soil covers the root. Thereafter the flow of water is not continuous. After a few days it is drawn off, and if the farmer is able to make the investment an application of rape-seed oil cake or fish scraps is made to the surface. As soon as the fertilizer has had time to become incorporated with the soil, water is again applied and withdrawn to allow the crop to be hoed. Every weed is cut out, and in some cases the roots are slightly pruned. Each field is given the minute attention of a garden. When the growing period is well advanced the water is allowed to remain permanently upon the field, care being taken to renew it by gentle inflow and escape, till a slight change in color indicates that the period of ripening is approaching. It is then withdrawn. While the slight change of color is given as the guide, the time when the milk in the seed has become dough is more correct, for the Japanese cut their rice when the straw is scarcely turned. Both the straw and the rice are better when the harvest occurs before the grain is dead ripe. CUTTING RICE. The grain is cut close to the earth, with a small sickle-like knife set in a handle. Four hills or bunches are bound together with two straws, making a bundle 3 or + inches in diameter. These are gen- erally laid crosswise in small piles, and are allowed to dry during the day. At evening they are hung with heads down on bamboo poles, which, by means of cross sticks, are made into a structure like a fence. The lower pole is high enough to allow a space of about a foot between the suspended bundles and the ground. The upper pole is 18 to 20 inches above this, the rice bundles on the upper pole overlapping the bundles below. After the bundles hang upon the poles long enough to become dry they are taken down by women and the grain removed by drawing the heads through a hatchel. The grain is then placed upon mats and exposed to the sun till thor- oughly dry. Before it is sent to market the hulls are removed by 11084—No. 85—03 » 18 RECENT FOREIGN EXPLORATIONS. passing the grain through a pair of burrs made of cement and bamboo and worked by hand. Winnowing is done by the open-air process, or by a simple fanning mill. (PI. I, figs. 1, 2.) After winnowing the milled product is placed in sacks deftly made of rice straw, each sack holding about 133% pounds. In these the rice is transported to market and the sacks are afterwards sold for paper material. MANURE. The extent to which night soil is used for fertilizing is scarcely conceivable. Whether in city or country, it is practically all saved in earthen receptacles and remoyed once or twice daily, according to the weather. The night soil is carried in wooden buckets, balanced ona pole across the shoulder. In cities the collectors sell to fertilizer com- panies what a man can carry (about 8 gallons) for 10 cents in silver. The companies transport it on flatboats to the rural districts, where it is applied in liquid form. In one corner of almost every garden and field may be found a cistern for storing liquid manure. FARM WAGES. Common laborers on the farm in Japan receive on an average 6 cents (gold) per day for women and 10 cents for men, with board, except in harvest time, when they are paid about double these amounts. Harvesting is expensive, considering the price of labor. On one occa- sion while in Japan a field was passed where two men were cutting rice. They stated they were paid 2 yen ($1 gold) for cutting, bind- ing, and hanging on poles the rice in a small field by the roadside. On measuring it there was found to be two-elevenths of an acre, the cost being at the rate of $5.50 (gold) per acre. Still, it is difficult to see how there could be any change in the methods of managing the rice industry in Japan. The present system of transplanting insures the best results and allows time to take off the winter crop. By the hand process the straw, which is quite valuable, is preserved, the grain is cut at the right time, even where there is a variation of matu- rity in the same field, and there is no loss from the cracking of kernels by the hatchel. COST OF RAISING RICE. A farmer near Tokyo furnished the following data in regard to the profits of rice farming, the estimate being for 1 acre of land: Case 1.—Where the owner of the land hires the work done: Cost of seed, 16'sho, or nearly 36\pounds!-c222. 222-0 -- aoe eee $0. 62 Costiof manurew: .2. 2. 222.6 cot eo ek oe ee ne aoe eee ee 10. 00 Cost /of labor, 120 days) work =... <= --0~ = ae ns Eee eee .18 54. 00 Total income, $54, as above. The foregoing statement, taken from the account book of a practical Japanese farmer, is full of interest and throws some side lights on their agricultural system. The small amount of seed used is due to transplanting. Consider- able expense is incurred for manure, but a crop of 205 barrels per acre is large for old land. One is chiefly impressed by the number of days’ work, one hundred and twenty, expended on 1 acre, and the amount of the Government taxes, $8. Eight hundred dollars taxes on a hundred acres of rice would stagger the American farmer. Where the tenant does the farming it will be noted that one-half of the grain produced is allowed for the use of the land and that there is no real profit. He simply receives pay for his labor. FARM LIFE. How the Japanese farmers live can best be understood by giving a description of some particular farmhouse. While visiting the distin- guished statesman, K. Mochizuki, at his country estate, a visit to the dwellings of some of his tenants was made. The following is a description of an average farmhouse on this estate: In the rear of the house was a garden of about half an acre, planted to field crops, beans, barley, ete., and in front was a garden of about one-fourth of an acre, artistically laid out and planted to vegetables, with oceasional flowers. The main building was one story high, about 24 by 48 feet in size, with the kitchen, 14 by 24 feet, across one end. Here was the usual clay stove, similar to those of Mexico, and a dirt floor, which by some process had been made as hard as cement. The remainder of the house was floored with mats. The family stores were packed in tubs, of which there were a dozen or more stacked at one side of the kitchen, all scoured to appear as if just brought from 20 RECENT FOREIGN EXPLORATIONS. the shop. The farmer’s wife was cooking at the stove. On the left ot the kitchen, in front of the house, was a room 10 by 12 feet, covered with the customary mats and used for a sitting room. Each mat was 3 by 6 feet in size and 2 inches thick. Back of the sitting room and opposite the stove was a room, 10 by 12 feet, used for a dining room. Beyond the sitting room, in the front of the house, was a private room, 12 by 16 feet, for lodging. From the dining room a hallway extended to and along the end of the house. The partitions of the rooms, which are generally removed during the day to give more venti- lation, were made of light sash, with strong white paper instead of glass. On the right of the kitchen was an addition, 20 by 24 feet, for the sery- ants’ quarters and general storage. Each servant had a small sitting room and a lodging room, with mats on the floor. There was no fur- niture, as we use the term, in the house; no chairs, tables, bedsteads, or mirrors. The members of the household sit, eat, and sleep on the matted floor. How everything can be kept so perfectly clean, without soil or stains, belongs to the mysteries of Japanese housekeeping. In front of the servants’ quarters a servant was cleaning grain and spreading it on the mats to dry in the sun. The tub and pounder for cleaning rice was in front of her. She did not like to be photo- graphed in her ordinary garb, but was satisfied when told to turn her back and appear to be at work. Adjoining the house on the left was a beautiful Japanese garden or tiny park, possibly 40 feet square, containing the usual landscape, trees, and statuary. In the center of this park and about 20 feet from the farm dwelling stood an artistic little one-story house, about 14 by 16 feet in size. It looked like a large playhouse for children, but we were informed that this was a special house for receiving guests and serving tea. The Japanese paper windows were slid back, revealing -a beautiful little parlor about 10 feet square, with the usual seat or bench of honor on one side, and a tiny waiting room. The house was a frame building, cross lathed and plastered, with posts exposed, boarded up and down on the outside, and ceiled overhead. In the rear of the house was a barn, 18 by 20 feet. The house here described is a typical Japanese farmhouse, one story, with thatched roof. The laborers’ cottages are built upon the same plan, but are smaller. The residences of wealthy country gentlemen are somewhat larger and with more elaborate grounds, but they retain the same simple arrangements and general style of living. There is gant caste in Japan. ‘The rich and the poor, the landlord and the tenant, the employer and the employed, live on the most intimate and friendly terms. Among the farmers of Japan, rice is considered quite a luxury and many can not afford to eat it regularly. Among the poorer farmers barley, millet, and sweet potatoes are substituted for rice. Among the no arro AGRICULTURAL CONDITIONS IN JAPAN AND CEYLON, Z1 better nourished Japanese the following constitutes the ordinary bill of fare: Boiled rice, boiled rape and daikon (half radish and half turnip), bean soup, and barley tea for breakfast and dinner. Lunch at noon is the same without the bean soup. A little salt fish is added ocea- sionally. GENERAL REMARKS. Japan has an area of 147,655 square miles, exclusive of Formosa, about one-tenth of which, or 15,000 square miles, is tillable. The population is now not far from 45,000,000, which gives a ratio of 3,000 persons to the square mile of arable land. At this ratio the State of Iowa could sustain 156,000,000 people and Texas more than 600,000,000, This statement is sufficient to refute the claim that Japanese agricultural products may at some future time compete with America in our home markets. Japan is rapidly becoming a great manufacturing and commercial nation, for which she is, by virtue of the genius of her people, exceedingly well adapted. The trend of events indicates that when that time arrives Japan will be a large con- sumer of American food and fiber products. CEYLON. The island of Ceylon, a British dependency, in latitude 6~ north, contains 25,365 square miles and has a population of 3,391,443, com- posed of about two-thirds Cingalese and one-third Tamils, with a few Moormen and Malays. The Cingalese are the primitive inhabitants and occupy mainly the southwestern portion of the island. They are medium sized, well formed, rather light colored, intelligent, and digni- fied. They are inclined to play the gentleman even in the roughest work, but are honest and make good clerks. The Tamils have been imported from the mainland, presidency of Madras, and bear a strik- ing resemblance to the American negro. They do a large part of the farm work and furnish most of the servants. There is not, however, much general farming done in the island, the central portion of which is occupied by mountain ranges, though the valleys are fertile. Only about 4,400 square miles are under cultivation of any kind. The thin sandy soil of the coast does not appear to be adapted to any crops except the cocoanut palm, which grows with amazing luxuriance, and the nuts constituting an important article of export. In the higher lands and on the mountain sides are large plantations of tea and cotfee, with occasional groves of cinnamon and other spices. AGRICULTURE. Rice is the main crop, but not enough of this is produced for home consumption, large quantities being imported from Penang, Singapore, India, and Burma. When preparing the ground for rice, a kind of 22 RECENT FOREIGN EXPLORATIONS. wooden drill, shod with iron and drawn by oxen or water buffaloes, is used. Two crops are produced, of which the principal or maha crop is sown in July, just in time to catch the late summer rains, and is harvested in December or January. The small or yala crop is planted in February and harvested in June. About 15 bushels per acre is considered a fair crop on the west coast, but in Anuradapura Province 30 to 50 bushels per acre are frequently obtained, depending on con- ditions. The Ceylon rice is rather inferior in quality. IMPORTS. The imports of cleaned rice at Colombo, Ceylon, from January 1 to November 10, 1900, were 486,652,390 pounds; from January 1 to November 1, 1901, 459,229,540 pounds. This shows that Ceylon, with a population of about 3,500,000, imports more rice than the entire product and annual imports of the United States. FARMHOUSES. The farmhouses are one story generally, with about three rooms, and are commonly built of brick or sun-dried clay, with mud-plastered walls. Some houses are built of poles, lathed with bamboo or bamboo matting, and are plastered with clay outside and inside. The floors are of tile or clay, and the roof is covered with grass, palm leaf, or tile. The usual cost of a house is $50 gold. Farm laborers receive about 8 cents (gold) per day, without board, but generally prefer to work for a share of the crop. One-half is given to the laborer. (Pl. III, figs. 1, 2.) INDIA. India (including Burma) has an area of 1,800,258 square miles and a population a little short of 300,000,000, This population is not uni- formly distributed. It is very dense in the valleys of the Ganges, the Brahmaputra, and the Indus and its tributaries. Bengal, with an area of 151,543 square miles (less than three-fifths of Texas), has a popula- tion of about 75,000,000. TIMBER. The absence of timber in India strongly impresses the traveler. No fences, rarely woodlands, and no barns in a country almost exclusively devoted to agriculture indicate a peculiar people. Inthe government reports considerable forest lands are mentioned. They are, however, in remote sections and quite inaccessible as a source of supply of wood and timber for the centers of a dense population. The price of wood for fuel is from $16 to $40 per cord and not very good wood at that; hence the masses must live without tire, except the little that is used tor cooking. AGRICULTURAL CONDITIONS IN INDIA. 23 EXTENT OF ARABLE LAND. The large proportion of the whole country that is arable is one of the first and most noteworthy observations of the traveler in India, In Japan one-tenth of the entire area can be tilled, and in China a large part of the country can never be subjected to the plow, although China as a whole ranks high in fertile lands; but in India, out of the 544,993,192 acres of surveyed land in 1899, seven-eleyenths were available for cultivation and 196,487,658 acres were actually sown with crops. FERTILITY OF THE SOIL. One of the most suggestive items to be noted is the fertility of the soil, after a tillage of so many thousand years, with little manure of any kind. With few exceptions all the dung of animals is used for fuel, and as far as observed those exceptions were limited to the goy- ernment farms. Many good farmers are said to use some cattle excreta on the land, but in all the small villages visited dung, made into patties and dried in the sun, was almost the only fuel. In the vicinity of cities the preparation and sale of cattle dung for fuel is quite an industry, and as far as observed it is all used in this way. GREEN MANURES. Inquiry at all the government agricultural stations visited and observations throughout India failed to develop a single case where green manures had been used to fertilize the soil. A further evidence that it is not used is found in the fact that the plows used simply stir the soil, but can not turn anything under. COMMERCIAL FERTILIZERS. It is difficult to use commercial fertilizers among Hindu farmers, for they suspect that all such preparations contain bone, blood, or some refuse of dead or slaughtered animals, and they declare it will defile them to handle it. An English gentleman in Caleutta told me that he had purchased some commercial fertilizer for his | his Hindu gardener refused to put it on the land. He employed a low-caste man to apply it to the vegetables, and after it was applied the gardener made no objection to working the soil on which it had been scattered, warden and CROP ROTATION, Rotation of crops is well understood and practiced. This gives a partial relief in case of continuous cropping. To some extent sum- mer fallowing has been employed as a renovating method. On the whole the present fertility of the soil is marvelous. 24 RECENT FOREIGN EXPLORATIONS. PUBLIC ROADS. The main highways are models of excellence, broad, well graded, and bordered with lovely shade trees, such as the banyan, the tamarind, and the sacred neem. At suitable distances wells have been made, and near them are located rest houses for weary travelers. Generally the rest houses are unfurnished and without any resident care-takers, but all day and all night they are occupied by weary travelers fora shorter or longer rest,as the case may be. Here and there may be seen a single man or woman; but generally the people travel in fami- lies or small groups, carrying their more cumbersome bundles upon their heads and their wealth upon the ankles of their women in the form of silver bangles. Mingled with the country people are numerous pack oxen and don: keys, with immense loads of all kinds of products. The oxen are noted for their docility and the donkeys for their diminutive size, being not more than 30 inches tall; but they are sturdy little animals and for their size they carry enormous loads. CONVEYANCES. In addition to the native families and village groups traversing the principal highways, there may be seen numerous carts drawn by oxen with a peculiar hump on their shoulders, the straight yoke resting on their necks and tied firmly to their horns. The carts are crude affairs. In some cases the wheels are merely two thicknesses of 2-inch plank, crossing each other at right angles, while in other cases the wheel con- sists of a large hub through which spokes are mortised to support a wooden felly 5 or 6 inches deep and 5 inches wide. The carts invariably have large wooden axles, which soon wear the hubs and allow the wheels to stand at considerable angles. Occa- sionally a native ofticial or the family of some village headman rides in an ekka or a tonga drawn by a trotting ox. DRESS. The clothing of the country people is exceedingly simple. In warm weather the men wear a turban and a single loin cloth so wrapped as to form a sort of breeches, extending to the knee; generally they have neither shoes nor sandals. In cold weather the cotton loin cloth is supplemented by a thick cotton bedquilt worn like an Indian’s blanket. The women wear short skirts and a thin cotton waist without sleeves, and in addition a long shawl or wrap of thin cotton stuff is thrown over the head and twined about the shoulders or allowed to hang loose. COUNTRY HOUSES. There are no country houses, in an English sense, in India. The ryots (farmers) live in a collection of dwellings called a village for PLOWS AND SCRAPERS. Zz» INDIAN VILLAGES want of a better term. These houses are of one story, having a single room, or occasionally two. In the mountain regions the walls are of stone, while on the plains they are made of brick or dried mud. There is usually a small yard in the rear of the house. There are openings, but no windows, and the doorway, if closed at all, simply has a bamboo- mat curtain. The roofs are made of tile and the floors of clay hardened by repeated washings with cow dung. VILLAGES. Between the houses in the small villages are narrow, tortuous alleys, but rarely regular streets. The village is surrounded by a high wall of stone, brick, or adobe, which answers for a fence against depreda- tors, the cattle being brought within this inclosure at night. Each village has its customs and unwritten laws, and it and not the indi- vidual is the political and social unit. It has its blacksmith and car- penter, its doctor, and its headman or chief, and generally its banker. The government taxes for the village are paid by the headman, who assesses them among the inhabitants in proportion to their property or income. Local matters are settled by the village, though in impor- tant cases there lies an appeal to the British courts. The village doc- tor, the carpenter, and the blacksmith are paid in rice at the harvest, not for specific work done, but as a sort of annual salary. PLOWS AND SCRAPERS. The plows used in different provinces vary somewhat, but have a general resemblance in that there is no moldboard and the instrument is simply one for stirring the soil. It consists of three pieces —the standard, the tongue, and the steel drill at the tip of a wood support or shoe. (Pl. IV, fig. 2.) The standard is usually 3 by + inches and about 5 feet long, into which, about 12 inches from the lower end, the tongue is mortised at an angle. The standard stands a little less inclined than ordinary plow handles. Near the upper end isa single pin used fora handle. A steel bar about 1 inch square at one end and brought to a point at the other passes through the lower end of the standard and is supported by a V-shaped shoe. This steel bar stands at such an angle that the sharp point penetrates the soil 3 or 4 inches or more, as may be required. It amounts to nothing more than a sharp-tooth drill, and costs 60 cents complete. This plow is drawn by two oxen. (PI. IV, fig. 1.) In use, the steel tooth cuts from the land a cloddy strip from 4 to 6 inches wide, and this is then broken up by the wedge- shaped wooden shoe. Afterwards men and women pass over the fields and smash the lumps with their mauls. Some ryots use a crude clod crusher made of wood and drawn by oxen. The harrow is much like ours, but crude. After the harrow has been used the routine of labor depends upon the crop to be planted. 26 RECENT FOREIGN EXPLORATIONS. In some cases where the farmers were planting wheat they used a wood scraper to prepare wide, flat furrows for the seed. This scraper consists of a board 1 by 6 inches and 3 feet long, with a handle 4 feet long attached to one edge at the center. The lower edge of the board is sharpened. It requires two men to operate it—one holding it.on the ground by means of the handle and the second standing about 8 feet in front and pulling it from the holder by means of a rope. In this slow way a shallow furrow is formed for the water of irrigation. (PLY, fig. 1.) It must not be inferred from the inferior implements used that Indian lands are not well tilled; the farmers make up for the defects of tools by additional labor. SEEDING AND HARVESTING. Seeding is done in a variety of ways, one method being for the dropper to follow the plow and drop the seed into the drill-like fur- row through a tube behind the plow, the next furrow covering it. Or the seed may be sown broadeast and harrowed. Or, in case of rice, the plants may be set into the flooded field from a seed bed pre- viously prepared. The grain is all hand cut, and when dry, thrashed by tramping with oxen. RICE FARMING. The experience of the practical and scientific farmers of India has shown that rice does best on a deep clay or clay-loam soil, but the sub- soil should not be so stiff as to prevent all natural drainage and cause stagnation of water, since rice is more luxuriant where fresh water is constantly added. Sandy-loam soils, if manured, produce an excellent quality of rice; the more manure the better the rice. More seed per acre should be used on sandy-loam soils than on clay loams. Rice sown late in the spring when the weather is hot requires more seed than if sown in the early spring. If sown. ina seed bed and transplanted the least seed is required—about 85 pounds per acre. Drilled rice requires about double this quantity, and if broadcasted 15 to 20 pounds more per acre are needed than when drilled. While there are many hundred varieties of rice, for practical pur- poses only three general classes need be recognized, i. e., early, medium, and late ripening. TREATMENT OF THE SEED BED AND MANURING, The site for the seed bed is usually selected on land more or less ele- vated to insure drainage. If water is allowed to stand on the field between crops it produces a ferment which is unfavorable to the future production of the plants. The use of green stable manure on rice fields just before planting is not recommended. — It is of little value, due to the fact that where ordi- nary manure is kept very wet it undergoes no chemical changes by RICE CULTIVATION IN INDIA. 27 which useful plant food is liberated. Therefore manure should be well rotted and applied long enough before planting to have some effect; better still, in case of a winter crop on the same field the manure should be applied to the winter crops. It is a common prac- tice after plowing to burn trash on all seed beds from which rice plants are to be transplanted. Coarse grass, dead leaves, brush, rice husks, straw tramped under the feet of the oxen, dust piles, and occa- sionally some cattle dung are piled on the plowed land, and on top of thisa thin layer of soilis spread to prevent rapid burning. The trash is then fired. The effect of this on the seed bed is the production of an ash for the support of the young plants and the destruction of weed seeds and injurious roots near the surface. The action of the heat on the surface soil also tends to liberate potash and phosphoric acid and to make the soil more porous. PLOWING AND FERTILIZING. Plowing and other heavy field work are generally done by bullocks, water buffaloes, or camels. Great emphasis is placed on repeated plowing. In India most of the rice lands receive no manure and have not received any for centuries, yet they continue productive, and when well tilled yield fair crops. One writer states: ‘‘All that is necessary to produce a bumper crop is timely and abundant rain.” Some writers seem to think that the fertility of the rice lands of Ben gal is due to the overflowing of the Ganges and Brahmaputra rivers. But these streams do not overflow and deposit silt to the same extent that this is done by the Nile. Moreover, this would not explain the fertility of the terraced rice land. The continuous fertility can not be due to the use of manure, for practically no commercial fertilizers are used, and almost all the droppings of cattle are used for fuel. It is mainly due to great natural strength of soil, good tillage. and rotation of crops. — METHODS OF CULTIVATION, In December the old straw and trash are raked into pil md burned on the land. The field is then plowed, and at intervals it is given two more plowings, after which it is left until the lattet part of March or early part of April, when the clods are crushed, and advantage is taken of the first rains to plow it twice more. The field is harrowed after each of the later plowings. Harrowing is done with a ladder having pins on the under side. The cultivator rides on the ladder, which also serves in a measure to break the clods. When the rice is up few inches it is raked. This stirs the soil and to some extent thins the plants. The average product of a field sown and cultivated in this way is 64 barrels per acre. Where rice is sown in a bed or nursery and transplanted into the field, the field is first plowed three or four times in water, thoroughly 28 RECENT FOREIGN EXPLORATIONS. mixing the soil into thin mud. After the mud has settled the ground remains covered by about 2 inches of water. Where the fields depend on rainfall for moisture the plants are transplanted during a shower. The plants are set in hills 6 inches apart each way, two or three plants being set in each hill. In this way about 28,000 plants are set per acre. Transplanting for the main or aman crop is done in May, and for the spring or boro crop in December and January. It is possible in some parts of India to raise five crops of rice in one year. The first crop is called aus and is the summer harvest from July to August; the second crop, or kaida, from September to October; the third, chatan aman, from October to November; the fourth, called boran aman, from December to January, and the fifth or boro crop from April to May. In the sub-Himalayan districts labor is very cheap, and it is cus- tomary to dig over the fields for rice with the mattock to the depth of 6 inches. This costs 80 cents per acre. PRODUCT PER ACRE. It is difficult to arrive at any correct estimate of the yield peracre from direct statements by native farmers. By dividing the total product in a given season by the total number of acres planted it has been ascer- tained that the average yield of rice per acre for all India is 823 pounds for the principal crop and 558 pounds for the spring or boro crop, making 1,381 pounds, or about 83 barrels, for the year, as only two crops in one year are generally raised. This is not a large showing for two crops, and it is quite evident that if one crop should be raised and the land devoted to green-manuring crops the remainder of the season, the one rice crop for the year would exceed the amount at present secured from two crops. HARVESTING. Rice is cut with a small sickle or hook knife and bound at first in bundles about 3 inches in diameter. After it has cured a while, the small bundles are made into larger ones and drawn to the thrashing place, where they are placed in hollow stacks, one tier of straw deep, with the heads on the inside. ‘Twenty women can on an average har- vest 1 acre ina day. One binder, four horses, and two men in the United States daily do the work of two hundred women in India. THRASHING. The usual mode of thrashing is to clear and level a small space of ground, wash it with cow dung until hard, and to pile on this circular form the rice to be thrashed. Five bullocks are tied to a rope tandem, and driven around on this pile of unthrashed grain. Sometimes, to expedite the work, a second line of bullocks is used. ‘Two men drive the two lines of bullocks and two men sift the straw with forks. In this way four men and ten bullocks will thrash the grain from an acre RICE CULTIVATION IN INDTA. 29 in six hours. When the straw is to be kept whole the rice is thrashed by beating the heads over the edge of a plank. During the harvest and thrashing time the farmer has to be con- stantly on the watch to see that the paddy is not stolen by dishonest laborers. He frequently builds a straw hut close to the thrashing floor in which he can sit and sleep. It is a regular custom to surround the pile of paddy with a ring of ashes so that it can not be approached without evidence. WAGES, Money wages are not usually paid. In some cases the reaper gets 1 load out of every 21 he cuts. In other cases he gets 10 or 12 pounds of paddy for a day’s work. Usually he receives 6 pounds of paddy and half a pound of cleaned rice. Laborers are generally employed by the year, and the wages paid are much less than the above, averag- ing about 2 cents per day. The ordinary plan upon which crops are raised is to form a farmers’ club. For this purpose five to ten farmers, each the owner of a pair of bullocks and a plow, form a club to help each other plow their lands. COST OF CULTIVATION. The ryot never keeps any account of his expenses, and hence it is difficult to estimate the cost of cultivating an acre of rice; but allow- ing customary wages and estimating the time required for the work performed, the following is an approximation of the cost on an acre of land where rice is sown broadeast: Plowing 4 times, 12 days’ work for 1 man and a pair of bullocks, at 3 cents ew Ole secre coseeceCoeBe Ste cee Senet OC EE Oe nat a: ee $0. 36 Carrying and spreading manure, 4 men | day ......-.-..------------------ 08 20) jovereinshs Gatal qoptelGhe 2 oe en Ss Cee Seer Gree CREE Sa eCe Ses 32 One plowing and harrowing after seeding, 3 teams 1 day -.......-----.-.---- O84 Onenweedinny 20! women il day, ati2 cents’. .-.-.....-.-----.---..--+.------ - 40 Repaummplevecsmloumen Uday, ab 2 cents.-.--.2-.---..------------..----- .32 Gain Oa LORW OI OMUENUS Ware == satel edio om «cee ewe eon rcs a weet oss sse---- -32 Carrying the bundles of paddy to the thrashing place or floor -........-.---- . OF Thrashing, 4 men and 10 bullocks 1 day, at 2 cents for each man and | cent nore @ixday fowler. 5 soe Gee ade Ge CREE EERE COSA OC .18 Gleamnprand wwinmowang, SumMenh Gye 52 — 622 coon se ee eee scenes eee 06 Eve iC acai CLASH SCN Ol AOLGhs acme ean ese noe oon eon wee eos ee 96 PAU OLUEC Here GR OI AONG mise ene ee ie oe Seen ieee ce cce cee see ces nes 12 3.283 Yield of first-class land, 1,010 pounds of paddy, valued at............-.----- 3. S4 Uovile joe CMR OAS apoE Sa SSeS sete cee i .594 The foregoing estimate, obtained from the most reliable authority, is impressive because it shows the low condition of agriculture in this Himalayan district. The wages of a man one day—2 cents—and the charges for the use of an ox one day—l1 cent conception of values of labor. are prices below our 30 RECENT FOREIGN EXPLORATIONS. It is noted that no account is made of manure and straw. Very little manure is generally used, and in many districts none. In the interior, where the above estimates were received, the straw and manure have no commercial value. While wages are low the price of rice is also low, only 32 cents for 823 pounds of paddy, or 614 cents per bar- rel of 162 pounds. When the rice crop-is handled in the usual way— the plants grown in a seed bed and transplanted to the field—there is an additional cost of 64 cents for preparing the seed bed; and the cost of pulling plants and transplanting into the field, which requires five men and twenty-eight women one day, is 66 cents. There is, how- ever, a saving of 40 cents for weeding and also a saving in spreading manure and other small items, which reduces tbe total cost of an acre of transplanted rice to 30 cents more than that of broadcast, leaving a net profit of 29 cents per acre on the crop. In the above estimates no account is made of the Government assess- ments on rice, which are considerable. These are sufficient at least to wipe out all profits in this class of farming. The following estimates of the cost of raising rice under high-class conditions are furnished by Hon. James Mollison, inspector-general of agriculture for India: Preparing ‘and: tilling seed bed.. =. 2522-122. 45-s eames =e aoe eee ee $0. 64 Manure used on seed bed, 6 loads; on an acre, 20 loads .............-------- 4.16 Cost of. seed, 80'pounds!:.-. ... 22. .20esst2aes2en ese eee See eee - 80 Plowing, puddling, and leveling: -..- 2322 -=-sse-5 see) es eee 1.52 Transplanting... -2-.).ssce0- 2stp22+-2 52325228633 e eee . 80 Weeding seed bedi. < .. -.3- 02 c065 d0 50 ede cone en Ce eee eee ee -48 Top dressing with castor cake, 200 pounds per acre ...-.-------------------- - 96 Cutting, thrashing, and winnowing......-.-----.---.-- jeoee ater eee eee Tying and jstacking bundles of straw - .< 22 5.- ees ee Cost of irrigation ic oc sejo cece oe aie oes 2 cen eee ere ere Add Government'tax per acre__- .---- .5- sce. «= =e ee =e Total, cost periacre.---s-=-. 2-1 sote see) oe Probable crop, 3,000 pounds, valued at Value of straw . 22 gsc. occ oc. 0 an ecole ne doe e ee eee eee eee Net profitsiper acre. << <<< <<< - sa ten eee atte The above estimates are based on wages in the Surat district, which are higher than in the Himalayan, but still very low. Under good cultivation the cost per acre is equal to that in the United States. NORTHERN LIMIT OF CULTURE. The question is frequently asked how far north rice can be produced profitably. Hon. C. L. Dundas, director of agriculture for the Punjab, stated that he could not tell, but assuredly as far north as his adminis- tration extended, 34° 15’ north latitude. PRINCIPAL CROPS OF INDIA. 31 CONSUMPTION OF RICE AS FOOD. The people in India do not keep account of farm products, except as they are compelled to by law; hence it is impossible to arrive at any exact data except through Government sources. In some provinces of India rice is the principal food; in others, less rice is produced and it constitutes only a portion of the food supply. In Bengal the 75,000,000 people on an average consume 1 pound of rice per capita each day, or 365 pounds per year, as determined by the Government reports. This would appear to be large, but in the way this amount is obtained it covers all losses, wastage, etc. The following table gives a comprehensive statement of the food crops produced in India and the relative proportion of rice to other grains: TABLE 1.—Area (in acres) under crop of principal products in each province of British India, 1897-1900. [Native States not included.} @ Total yield, 618,966,312 barrels of 162 pounds each. ¢ Total yield, 2,110,562 bales of 400 pounds each. + Total yield, 266,250,560 bushels. Province. Rice. Wheat. | Barley. | inet, | corn. aeearete UPPerBUrMe sos see oe | 1, 818, 962 | 15, 813 120 804, 950 76, 300 195, 523 Hower Burma... ....0.0-..-- 6, 277, 678 } Nan eeeonee 748 11, 492 15,010 JANIS, See erinpnb atop OpoCooon | 3,653, 583 223) | 59 | 2, 823 86, 283 BONPAl a cceccceeesteerece--s| 39,656,800 | 1,541,400 | 1,448, 200 1, 802, 200 5, 141, 200 Northwest Provinces ......-.. 4,592,608 | 4,601,392 | 3,154,323 1, 143, 430 3, 913, 905 (0) ils Ea Ree ee eee ECOnRAre 2,899,792 | 1,619,583 | 1,020, 830 | 462,575 | 2,224, 635 Ajmer-Merwara ... 296 1, 888 27,214 | 91, 109 | 52 14, 240 ‘Pargandé Ménpur .. 167 Nad) ES a ee ee 2,504 | 509 482,795 | 5,488,598 898,443 | 1,243,605 | 1, 1, 047, 568 898,853 | 347,445 8,910 | 1,014,678 | 165,678 1, 251, 143 811, 590 44,328 | 11,035, 141 1, 86), 980 4,708,624 | 1,633,777 | 6,005 80, S87 2,921, 603 44,138 21,192 | 234 60, 102 6, 429, 045 20, 636 3,318 | 4,155,425 | aaa be Sem ceets |oceeeseeeses ibstah ES See eee 972, 808, 952 (616,104,793 | 6,611,984 | 22,633,756 | 5,195,472 23, 338, 758 Province. Sugar cane.| Cotton. Jute Theres in si Rees Upper Burma .......200..00. 2, 236 That || See 3, 167, 791 Lower Burma ..............- 9, 330 UPL os SS 4, 603, 103 28, 315 3,399 100, 168 5, 433, 668 73, 000 144, 000 1,970, 500 70, 414, 425 1, 028, 851 NGS S02) eet cucw'3s ss 33, S01, SM4 235, 326 sonic eV) Re epee 12, 650, S31 199 35, 453 | 5 MS, 258 16 Dee = G,T65 |. wes sesceccdk 360, 978 iobalien yeeeeasenes << 20, S61, 061 2,615 91, 091 2,871,774 IBGUIDE Viewers ses. o. 70, 515 2, 050, 251 15,135, 27 Central Provinces . 25, 588 712, 836 10, 734, 264 Berir.. 2,471 RUAN st) SES Soe 3,097, 782 2, 897, 40 Madras 58, 688 nO ira Li eee 20, 694, 679 35, 690, 440 Rata eee PRM aC Naw iie]'s vitaGe os caconafeacexdavhecasleCaakneccstcas ye i) Peer a ee 8 ATGUAl <3, -0-2 05 weeas-| 2,608,088 | ©8,875,841| 2,070,678 | 164,878,789 |... .......... 32 RECENT FOREIGN EXPLORATIONS. In the more populous provinces the area planted to food crops is so small in proportion to the population that even the slightest failure results in disaster. Nearly all the tilling of the soil is done with the plow, and oxen, buffaloes, and sometimes cows furnish the motive power. The small number of carts (wagons are not used on the farms) is explained by the fact that a large part of the transportation of produce is done on the backs of oxen or donkeys. Taste 2.—Area (in acres) irrigated in British India, 1899-1900. Irrigated from— ae So nes s | Net area Area Province. Gover Private | Tanks. | Wells. one imtoeteal| gone Be uce Grennil canals. year. once. | Upper Burma ..... 252,161) 307,198, 129, 864! 7,211] 102,587) 799, 021| 3,695,206} 260, 036 Lower Burma ..... 310 EGY) eee odocd peooocste.c | St 434) 5,069) 6,857, 898 846 Assam .. 4, 552, 210 565, 146 Bengallsesesee ace 53, 253, 600) 10, 618, 100 Northwest Proy- | | | INCeN =. ee 1, 981, 373 5,692 1,215, 683) 4, 478, 507) 553,595) 8, 234,850) 24, 402,658) 4, 461. 342 (O00 ee gansesa Sebonssenc bocaeteaas 976, 394| 1, 643, 178 80, 453) 2,700,025) 8, 624, 254, 2,427,975 Ajmer-Merwara -.-}.--. 2.2. -|2.-- ~~ = 7, 228) 43, 776 116) 51,120 230, 773 17, 400 Pargand Manpur é. 924) -cen see 324 6, 786) 136 Punjab - 4,‘ 20, 049) 4, 154, 598 134, 083) 9,375, 983) 28,275,728) 2,017,570 Sind}s os: -7-eeeeee 2,35 38} 140,595). --....-.. 41, 005, 110, 414) 2,644,447) 2,781,014 215, 474 Bombay --/.....:.- 99, 829 5,013 30,413) 667, 789 78,149 871, 223) 19, 278, 203 320, 293 Central Provinces .|\-...--..--- 810, 176,187 64,118 14,079 255, 264) 14, 762, 603) 164, 340 OTA acral tolalefaie = chet | ral ateiataeeers 1B 4 oreo 66, 838 107, 67,017 5,408, 758) 1, 495 Wee Aba coesoca 2, 648, 160 26, 289 1,129,804) 146,986) 5,783, 766 23,122,215) 2, 674, 229 COOr Ras saocctace eee 1-6 Y()) DoSemeeeenl Hecorecsec soc soomecll>otSocccos 1,370 200, 117 701 Total). .-<---3 12, 333,717 1,310,723 4, 388, 345/12, 297, 148) 1, 224, 003 31, 544, 036 190, 447, 023) 28, 745, 083 Table 2 shows the number of acres irrigated as in Table 1, native states not being included. Of course lands subjected to natural over- flow or on which there is a heavy rainfall are not included in this table. The irrigated lands are principally planted to wheat and food crops other than rice, although in some provinces the rice crop depends entirely on artificial irrigation. In the best rice districts, however, the rainfall is very heavy, amounting to over 200 inches in a year in Lower Burma, which with the annual deposits from the overflow of the Irawadi River makes it ideal rice land. Table 3 shows the number of head of live stock and number of farm implements in the same area as that covered by Table 1. LIVE STOCK, ETC., IN INDIA. 33 Taste 3.—Live stock and farm implements in British India. Billa and Buffaloes. Province. | PUT Cea: (gar i ee Sheep. Goats. (Opoyertel stb el oeeeoeensccesod 687, 823 697, 989 109, 953 25,013 Lower Burma - 578, 821 397,338 | 276, 620 12,374 PARSE sy oe hiner taletn alas os' 1,102,938 | 1,006,305 | 91, 136 FR Erip illic ete peice ar seta 0 1,195,736 |° 880,754 | 105,597 | Northwest Provinces........| 7,045,630 | 4,567,777 | 565, 835 OY ets eerste elem at alm Siierelatciniats. 3,148, 842 | 2,065, 569 219, 357 866, 282 Ajmer-Merwara ........----- 64, 390 60, 233 2,731 21,142 Pargané Minpur........-.-. 1,996 2, 009 | 27 727 Punjab ..... 4,681,729 | 2,566, 047 592,137 | 1,903,070 SING eeermstettctn cies ia iste eine ie sais 528, 744 515, 559 5, 502 190, 093 SOE DBs aitetarataie cjein'g == <== 2,295,886 | 1,139, 843 197, 780 669, 469 Central Provinces.........-. 2,831,655 | 2,524,616 354, 531 511, 467 697, 791 623, 870 87, 909 | 211, 774 4,411,350 | 8,858, 408 846,679 | 1,560,104 8, 234, 262 5, 181, 639 34, 629 26, 674 11,931 | 7,690 629 1,755 29, 257,910 | 20,927, 441 3,417, 725 | 9,133,896 | 17, 932, 237 19, 059. 649 | y Proyince. | Homes avd aules ae | Plows Carts, ‘nd battazo [MppeniBuxmMa--...ssee..c es | 28,197 2, 692 | 415, 630 239, 101 638, 724 Lower Burma ....-......---- TOIL ON | esa anon ane | 478, 388 199, 181 594, 034 Assam 9, 908 133 | 821,570 11, 883 1,390, 361 Bengal | 35, 913 458, 211 60, 337 246, 227 Northwest Provinees........ 434, 426 | 3, 162, 668 | 5,13 6, 408, 351 (DW Nile s6 tee aeee aoe REeaaEE | 181, 084 | 50, 213 1, 464, 406 2, 427, S24 Ajmer-Merwara........... <7 * 2,486 5, 029 33, 069 21,040 Pargan’ Ménpur........--.. 83 | 283 865 1,411 Iaith Ee SAS eae eee | 312, 746 612, 887 2, 229, 768 3, 712, 614 i 76, 799 84, 442 243, 042 352, 797 90, 867 51, 207 891, 451 1,550, 194 94, 542 17, 628 1, 295, 953 1, 886, 250 29, 627 | 20,488 136, 041 29, 49 40, 239 120, 086 2,749, 701 4,383, 639 401 274 26,979 19, 096 BINS GEC Unrate acta ooo, 1, 548, 880 | 1, 240, 101 14, 407, 742 2,925, 849 23, 882, 111 WELLS. The wells are open, 5 to 7 feet in diameter, and 30 to 60 feet deep. On one side of the well an embankment is made about 5 feet high. This slopes at an angle of 20° from the well and frequently terminates ina pit a few feet deep. This embankment forms a descending road for the oxen to travel when hoisting the water, A bullock’s hide is used for a bucket; the corners are attached to a rope, which passes over a single pulley at the top of the well and is tied to the yoke of the oxen. Each hoist carries about one barrel. (PIV. tig. 2.) Two yoke of oxen are required, as one yoke can be used only six hours consecutively, and there must be one man to drive the oxen, one to 11084—No. 35—03——-3 B4 RECENT FOREIGN EXPLORATIONS, manipulate the bucket, and one in the field to distribute the water. , Three men and four oxen will water 10 acres of wheat during the cropping season. RICE PRODUCED. In 1900 there were in the provinces of Bengal, Burma, and Madras 49,915,913 acres in rice, which produced 435,822,000 barrels. If we place the product of the remaining 22,893,039 acres in rice at 183,144,312 barrels, the total for India would be 618,966,312 barrels of rough rice, or about 177 times more than the entire rice product of the United States. AGRICULTURE IN THE PUNJAB. Hon. C. L. Dundas, director of land records and agriculture for the Punjab, stated, in reply to inquiries, that— Unirrigated rice can only be grown in the submontane tracts, where there is heayy rainfall. The average yield is about 550 pounds per acre. On irrigated lands the average yield is about 900 pounds. A good crop would be 1,200 pounds, and 1,500 can be obtained by careful cultivation. In the Punjab this is produced almost invariably by owners with small holdings. If the holding is large, part is culti- vated by the tenant on the share plan, the tenant paying one-fourth to one-half the gross product. Hired labor is employed sometimes in transplanting and generally in harvesting. This is paid for in kind. Throughout the Punjab, women of the agricultural class are employed in the lighter kinds of outdoor field labor, such as harvesting, picking cotton, ete. The women of certain tribes of high sovial or religious character never work in the field, but generally women work on the lands of their male relatives. Compensa- tion consists in their food and a small present in kind at the close of the harvest, practically subsistence and nothing more, but differing from the starvation wages of civilized countries by the patriarchal customs of India, which forbid a man from filling his own stomach while leaving his employee hungry. Hence harvest wages depend entirely on the harvest. If this is good, the laborer, male or female, may get enough grain to keep him or her two or three months. Unless foreed by famine, women will not work in the field except for their male relatives. In the Himalayas the women do all the farm work, including plowing. Windmills being unknown and water mills impossible on the plains, all the grain used as food in India is ground on handmills (small stone burrs) by women. Spin- ning is universal, and much of the coarse cloth used for clothing is manufactured at home. The cost of labor necessary to produce a crop of rice is about 45 per cent of the total product grown, including the straw. To give a definite cash estimate of cost is practically impossible. A landlord would, in a typical case, pay some 8 per cent customary dues and divide the balance with his tenant, paying one-half his own share in water rates and land revenue to the Government. The revenue or tax to the Government varies from $1.50 to $3 peracre. As a rule, the landlord works his own farm. The highly flavored rices are regarded as choice, but the people prefer to plant the coarser varieties, as giving less trouble. There is apparently great obseurity in the scientific names of rices, and it is difficult to distinguish varieties. Wheats, millets, and gram (peas) form the staple crops, wheat being the chief article of export. Considerable cotton is produced. About 110 pounds of lint cotton is an average crop for an acre. It sells at about 5 cents per pound. LIVING AND RICE FARMING IN INDIA. 35 a) The practice of plowing under renovating crops I believe is unknown in the Punjab. Cattle for plowing or lifting irrigating water range in value from $15 to $25 per head. Buffaloes are worth from $20 to $35 per head and camels about $15 each. The price of cattle for work varies with the provinces. At Poona a good buffalo for ‘field work is worth $6.50; an ox $16 to $17. At Delhi a buffalo is worth $8 to $10; an ox $16 to $26, according to size. Native plows generally sell for 60 cents each. COST OF LIVING. Among the ryots no cash estimate of the cost of living could be obtained. The following statement made by an educated Hindu may be assumed to be correct as regards cost of living in the city: A laborer needs 1 pound of rice, worth 2 cents; one-half pound of dahl (split peas), 0.75 cent; one-half pound of barley, 0.875 cent; condi- ments, 0.17 cent; fuel, 0.5 cent; making a total of 4 cents for a day’s living. Better living for laborers earning higher wages costs about 6 cents per day, divided as follows: Rice, 1 pound; mutton, one-half pound; barley, one-half pound; vegetables, condiments, oil or butter, and fuel. The retail price of rice, low grade, is here given at 2 cents per pound. The wholesale price in India for this grade is about 1 cent per pound and in Burma 90 cents per hundred. RICE FARMING IN LOWER BURMA. Rice farming in Lower Burma varies somewhat from that in Bengal. The lands are richer, and the rains are more abundant. The cultivator commences to plow about the Ist of June and continues to work the soil till he secures an even surface of mud, which is kept soft by the heavy rains. In July women transplant the rice from the seed bed and receive for this work at the harvest a certain number of bundles per hundred plants set. The harvest commences in November, and cutting, curing, thrashing, and winnowing are done in much the same manner asin Bengal. Rice cultivation in Lower Burma comes nearer being on a commercial basis than in India. Wages are regulated by each village and are frequently paid in money. Laborers who are imported from Madras in harvest time usually receive 23 cents per barrel of product for cutting and binding. A large portion of the crop is cultivated on the tenant system, the landlord furnishing land and seed every other year and receiving one-third to one-half the product. He furnishes no house nor other buildings and does not fence the land. A yoke of cattle will work about 10 acres of land. RICE MILLING. Very little rice milling, as the term is commonly understood, is done in India proper, except for resident Europeans. In the rural districts, where the rice is wanted for local consumption or for export, the hulls are removed by pounding, using a pounder worked by the foot. Pounding and winnowing in the open air or by a fanning mill 36 RECENT FOREIGN EXPLORATIONS. complete the milling process. There is no charge for milling, the hulls and bran being considered by the natives full compensation. As late as 1891 there were only two modern power mills in India. Most of the rice exported to Europe from Bengal is cargo rice, four- fifths husked and one-fifth paddy. It is claimed by shippers that cargo rice is not as liable to heat on shipboard as that completely milled. In Burma the grower markets all his rice in the paddy and in bulk, except such as goes by rail, which must be sacced. The larger part is delivered by boat, and is carried to the mills in baskets by coolies. It is weighed and delivery actually takes place in the mills. At first the mills were merely husking mills to prepare the large crop of paddy for export, but gradually other processes were added until complete modern milling plants were equipped. The hulling stones in the best mills are made of emery. Some of the machines are cruder than similar machines in the United States, but they appear to do the work satisfactorily. Permission was freely granted to inspect the Kemen- dine mill in Rangoon, which has a daily milling capacity of 500 tons of rice for native use or 300 tons for Europeans. 80.6 March ... SUcOr MORO MHRee ace ok. coc sne nce ac. pause 80.4 DLE MGt Sm Me owiehe bene kn ov Eb wie wcnie's one Gee FINO VOU DOR an econ 5 cacsacacns ‘ 79.0 RMU ECRMR Meh Katee nie. ba Ceca Gyacnen cae BOCGH I MACCOMNIEE ee cnc p as nancucanee 77.3 SIC RCRGEON CREE es Us a ccekiuwep’s buss 82. is June . Pky ee = sew So. ; MUR metRRReee rat osteo tut sesteuewens 80.9 11084— No. 85—03 + 42 RECENT FOREIGN EXPLORATIONS. The yalleys are broad and well drained, while the mountains are approached by a gradual elevation and frequently by table-lands, and are generally fertile to the top. Neither on the coast nor in the lowest valleys of the interior is the heat at any time oppressive, and within a short distance from any point on the islands it is possible to reach an altitude where the climate is perfectly delightful, even in the warmest season of the year. RANGE OF PRODUCTS. Taking all the islands and the fertile mountains into consideration, there is possible a very wide range of products, from the most delicate spices to the hardy cereals. The chief commercial products haye been rice, sugar, tobacco, coffee, and fiber plants, but the islands can pro- duce cattle, wheat, corn, oats, the legumes, and the grasses. STOCK AND PASTURE LANDS. Like Porto Rico, the Philippines furnish admirable conditions for stock raising. The mountain sides have frequent streams of pure water and produce an abundance of grasses, somewhat coarse and lack- ing in flavor, but which if cropped closely are relished by domestic animals. Softer and sweeter grasses can readily be introduced. Ber- muda grass and several of the Paspala and some cloyers do well. Stock raising has been profitably carried on for many years by natives, often on quite a large scale. The native horses are small, but are hardy and of immense energy, showing their descent from Andalusian stock. There is a good demand for dairy products, and few lines of husbandry would be found more profitable. FODDER PLANTS. The soil and climate of the Philippines are especially adapted to the production of a great variety of fodder plants. Among the many may be mentioned alfalfa, esparcet, serradella, vetch, lupine, pea, soy bean, Lespedeza bicolor, Pueraria thunbergiana, Astragalus latoides, cow peas, Panicum colonum, guinea grass, and Panicum macimun, Dur- ing the rainy season it would be necessary to use these plants for soiling, as the almost daily showers prevent curing. From December 1 to May hay could be made in most parts of the islands. SUGAR CANE. Conditions are very favorable for raising sugar cane. The heavy rainfall during the growing period, followed by the dry months of December, January, February, March, and April, are ideal conditions, so far as climate is concerned. This gives a full year for growth and five months for manufacturing the sugar. The sugar mills are very RICE FARMING, ETC., IN THE PHILIPPINE ISLANDS. 45 crude, except some in Negros, Panay, and Cebu. In Luzon the sugar factories are mainly of the open-kettle sort, and with machinery cruder than is generally used in farm sorghum manufacture in America. Some stone rollers for crushing the cane are used, and many factories have only large wooden tubs with iron bottoms for boiling the cane juice. In Panay and Cebu the mills are of a higher type, although crude as compared with American up-to-date milling plants. (PL. VI, fig. 2.) RICE FARMING. The method of raising rice in the Philippines is practically the same as in India, except that the plowing is almost exclusively done with water buffaloes, and a larger proportion of the land is sown broadcast. Rice planting is usually done in June, and harvesting in November and December. Only one crop is raised each year. With artificial irrigation two crops could be produced annually, one in the summer and one in the winter and early spring. The area deyoted to rice could be considerably enlarged, but it is doubtful whether in the eyo- lution of the islands under American conditions such will be the result, as a number of other farm products are more profitable and are culti- vated with less labor. The natives much prefer to plant and work manila hemp (J/usa teatilis), as when once planted it produces a crop for several years with slight attention. Coffee and some of the spices are favorite products in certain sections. Plowing the land and setting rice plants in the mud is a disagreeable task, even to Filipinos; conse- quently the general trend of agricultural industries in case of expansion will be away from rice and toward crops more easily handled and more profitable. FRUITS. Nearly every known variety of fruit can be produced on these islands, from such as require extreme tropical conditions to the hardy fruits of the temperate zone, like the apple and the cherry, for the islands possess a great range of climate. There are valleys where the temperature never falls below 70° and there are table-lands where it drops nearly to the frost line in the winter. These extremes are found on the same island. At Manila 65° F. above zero would be extraordi- nary weather. A hundred and thirty miles north, in the province of Benguet, the grains and fruits of northern New York can be pro- duced. TIMBER. It is estimated that only about one-fifteenth of the land has been brought under cultivation. A large portion of the remainder is timber land, and nearly all of it belongs to the Government. Many very 44 RECENT FOREIGN EXPLORATIONS. valuable varieties are found, among which is mahogany. Except the teak forests of Upper Burma, now under complete Government con- trol, these are the most valuable timber lands in eastern Asia, and if cutting is properly regulated they will remain a source of profit for many years. At present the only method of obtaining this wood is to cut and hew it into square timbers, which are then dragged down the mountains by oxen. By this method fully one-third is wasted and many valuable young trees are destroyed. O Bul. 35, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE I. Fia. 1.—RicE MILL AMONG THE MOUNTAINS, JAPAN. Fia. 2.—PLANTING RICE, JAPAN. Bul. 35, Bureau of Plant Industry, U. S. Dept. of Agriculture PLATE Il. Bul. 35, Bureau of Plant Industry, U. S. Dept. of Agriculture PLaTE III Fic. 2.—CARTS WITH BAMBOO COVERS, CEYLON. Bul. 35, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLaTe IV. FiG. 2.—ENGLISH PLOW AND INDIAN PLOW. Bul. 35, Bureau of Plant Industry, U. S. Dept. of Agriculture PLATE V. FIG. 1.—WoOoDEN SCRAPERS USED IN PREPARING FOR IRRIGATION, INDIA FiG. 2.—WELL USED FOR IRRIGATION, INDIA Bul, 35, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE VI. Fic. 1.—WASHING RICE, CHINA. Fia. 2.—SUGAR-BOILING HOUSE, LUZON > @ Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture . Plate | Cross SECTION OFA DYING TREE OF THE BULL PINE, SHOWING BLUE COLOR U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY—BULLETIN NO. 36. B. T. GALLOWAY, Chief of Bureau. THE “BLUING” AND THE “RED ROT” OF THE WESTERN YELLOW PINE, WITH SPECIAL REFERENCE TO THE BLACK HILLS FOREST RESERVE. BY HERMANN VON SCHRENK, SprecraL AGENT IN CHARGE OF THE Mississtppr VALLEY LABORATORY, VEGETABLE PATHOLOGICAL AND PHYSIOLOGICAL INVESTIGATIONS. Issuep May 5, 1903. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1908. ma LETTER OF TRANSMITTAL. U.S. DerartMENT OF AGRICULTURE, Bureau oF Pianr Inpustry, OFFICE OF THE CHIEF, Washington, D. C., December 24, 1902. Str: I have the honor to transmit herewith a technical paper on The ‘‘Bluing” and the ‘‘Red Rot” of the Western Yellow Pine, with Special Reference to the Black Hills Forest Reserve, and respectfully recommend that it be published as Bulletin No. 36 of the series of this Bureau. This paper was prepared by Dr. Hermann von Schrenk, Special Agent of this Bureau in Charge of Timber Rot Investigations, a line of work being conducted jointly by this Bureau and the Bureau of Forestry, and it was submitted by the Pathologist and Physiologist with a view to publication. The illustrations, which comprise 14 full-page plates, several of which are colored, are considered necessary to a full understanding of the text. Respectfully, B. T. GaLtoway, ( hief of Bureau. Hon. Jamms .Witson, Secretary of Agriculture. ERE PACE. The report submitted herewith, entitled The ** Bluing” and the “Red Rot” of the Western Yellow Pine, with Special Reference to the Black Hills Forest Reserve, covers in part an investigation under- taken by the Bureau of Plant Industry in cooperation with the Bureau of Forestry in the broad field of the diseases of forest trees and the means of controlling them, as well as the causes of and methods of preventing the decay of all kinds of timber, especially that valuable for construction purposes. At the present time an immense quantity of dead and dying timber of the bull pine is standing in the Black Hills Forest Reserve, South Dakota. The amount has been variously estimated, but will probably approach 600,000,000 feet. The death of the trees was caused by the pine-destroying beetle of the Black Hills. as shown by investigations conducted by the Division of Entomology of the United States Department of Agriculture. Following attack by the beetles the wood of the tree is invaded by various fungi, one of which causes the blue coloration of the wood. Dr. yon Schrenk has demonstrated, however, that the fungus which causes the bluing does not injure the strength of the wood. The rapid decay or ‘‘red rot” of the timber is caused by another fungus, and its ravages can be forestalled by a proper use of the wood. A series of recommendations is made, which, if followed, will result in the saving of a very large part of the dead wood. ALBERT F. Woops, Pathologist and Physiologist. OFFICE OF THE PATHOLOGIST AND PHYSIOLOGIST, Washington, D. C., December 23, 1902. @Bull. 32, n. s., Division of Entomology, U. 8. Dept. of Agriculture, 1902. 5 CONTENTS. Tntroduction -..<--<---- nana nn eb ne nnn we ence nee nn ne wenn e nnn nae n none Death of the trees .----------------- en ww wn ne nen a nnn en eee === When are the trees dead......------------------------ -------+-------- Mhet' blue? wood... ----s-< wo O° 7 v2 to bo wow yw Co Co OO Puate I. . Dying trees of the bull pine. Fig. 1.—Green, ‘‘sorrel-top,’ . Sections showing early stages of the [LEUSTRATIONS: Cross section of the trunk of a dying tree of the western yellow or Page. bull pine; showing blueicolor 22-2 2252-0. -— eee Frontispiece. ” and “‘red-top”’ trees. Fig. 2.—Green and “‘sorrel-top”’ trees -...-.-- . Color change in leayes of the bull pine. 1. Leaves from healthy ‘ ‘sorrel-top’’ tree. 3 and 4. Leaves from ” stage. 22.5< nat ocensesee asa a eee tree. 2. Leaves from trees turning to the ‘‘red-top’ . Fig. 1.—‘‘Red-top”’ tree in a group of healthy trees near Elmore, S. Dak... Big: 2:—<"Black-top”’ trees. ~~ ss - 5 ee . Figs. 1 and 2.—Sections of trunks of the bull pine, showing early stages\of ““blueidisease”’ -....2....13.-5.. 5 o-e -eee . “Blue”? sections from dead trees. Fig. 1.—Sections from tree dead five months. Fig. 2.—Sections from tree dead eighteen months - - . Mycelium and fruiting bodies of the ‘‘blue’’ and ‘‘red-rot’’ fungi. 1. Tangential section of ‘‘blue’’ wood. 2. Crosssection of ‘‘blue”’ wood. 3. Crosssection ofa medullary ray. 4. Young perithecium of the ‘‘blue”’ fungus. 5. Mature perithecia of the ‘‘blue”’ fungus. 6. Two perithecia of the ‘“‘blue’’ fungus. 7. Two asci with spores of the ‘‘blue’”’ fungus. 8. Spores of the ‘‘blue”’ fungus. 9. Top of beak of perithecium of Ceratostomella pilifera just after the dis- charge of the spore mass. 10 and 11. Median sections of sporo- phores ofthe ‘red-rot”” Tun pus 222-2 .e = - eee eee . Sections of ‘blue’? wood. Fig. 1.—Radial section. Fig. 2.—Tan- gential) section \.5-- 24 2 5-2. ee se ee . Pieces of wood from the bull pine, showing blue fungus starting from holes made by a wood-boring beetle.............---------- “red rot.’’ Fig. 1.—Section taken 35 feet from the ground from a dead tree. Fig. 2.—Section showing more adyanced stage of decay. Fig. 3.—Section from tree shown in fig. 2, made 15 feet higher up...--.....---...-.-----... . Sections from ‘‘black-top’’ bull pines, showing advanced stages of decay. Figs. 1 and 2.—Sections from the top of a fallentree. Fig. 3.—Section from a standing pine 4 feet from the ground.....-.--- . Group of broken’ */black-top” trees-- <<< ooo seec sence . Fig. 1.—Top of ‘‘black top’? broken off. Fig. 2.—Polyporus pon- derosus growing on dead pine stump ...-..-.--------------------- . Sections of rejected cross-ties. Fig. 1.—Wood affected with ‘‘red rot.’’? Fig. 2.—Diseased wood from living tree ....-.-..--------- 8 40 40 40 40 40 40 40 40 B. P. I.—46. V. P. P. L.—100. ™HE ~ BLUING” AND THE “RED ROT” OF THE WEST- ERN YELLOW PINE, WITH SPECIAL REFERENCE TO THE BLACK HILLS FOREST RESERVE. INTRODUCTION. The present investigation was undertaken to determine— (1) The cause of the blue color of the dead wood of the western yellow pine, commonly known as the bull pine (Pinws ponderosa), and the effect of the coloring on the value of the wood. (2) The reason for the subsequent decay of the wood, the rate of decay, and whether the decay could be prevented. (8) Whether it would be possible to use the dead wood before it decayed; first, to reduce the fire danger; second, to prevent the decay and thereby save an immense quantity of timber. DEATH OF THE TREES. The physiological changes which take place in the bull pine (2%nws ponderosa) as a result of the attack of the pine-bark beetle (Dendroc- tonus ponderose Hopk.”) are intimately connected with the fungus diseases under consideration, and may therefore be referred to briefly. According to Hopkins, the beetles enter the bark of the living trees in July, August,and September. The primary longitudinal burrows or galleries are excavated by the adult beetles, and the transverse, broad, or larval mines (Bull. 82, n. s., Division of Entomology, U.S. Depart- ment of Agriculture, Pls. I and II] and fig. 1) through the inner bark and cambium of the main trunk have the effect of completely girdling the tree, and by September the cambium and the bark on the lower portion of the trunk are dead. The foliage of the trees thus attacked, however, shows no change from the normal healthy green until the following spring, when the leaves begin to fade. The first signs of disease noticeable in an affected tree are visible in the spring of the year following that of the attack by the beetle. Here @Hopkins, A. D. Insect Enemies of the Pine in the Black Hills Forest Reserve. Bull. 32, n. s., Division of Entomology, U. S. Dept. of Agriculture, pp. 9, 10 uv 10 THE ‘‘BLUING”? AND THE ‘'RED ROT” OF THE PINE. and there one will find the needles of affected trees turning yellowish. The bright green fades almost imperceptibly, starting near the tip of the needle. The needles first affected are those on the lowest branches (Pl. II), and on these branches the discolored leaves will be more or less scattered. By the end of May most of the leaves on an affected tree will be pale green or yellowish. (PI. II; Pl. III, 2.) This yellow color increases in intensity during the summer and makes the affected trees a conspicuous mark among the healthy green trees. Trees in this stage are locally known as ‘‘sorrel tops” or ‘‘ yellow tops.” When standing on a hillside, groups of ** sorrel tops” can be easily detected at a distance of several miles. It is rather a difficult matter to show the contrast in a photograph. The middle tree on PI. II, fig. 1, shows the contrast with the green trees on the left to some extent. The yellow needles are drier than the green ones and show a marked disintegration of the chlorophyll. As they continue to dry the color changes gradually through various intermediate stages (Pl. III, 3) to areddish brown. This color (Pl. III, 4) becomes very marked after the trees have passed through the second winter. The needles are then dry and they begin to fall off. Such trees are known as ‘‘red tops.” (See Pl. II, fig. 1; Pl. IV, fig. 1.) The leaves finally fall off completely, leaving the branches bare. Such trees without any leaves are known as ‘‘ black tops.” (PI. IV, fig. 2.) The group of trees on Pl. Il, fig. 1, shows the green trees and the ‘‘sorrel tops” and ‘‘red tops” (rapidly becoming ‘* black”) side by side. To summarize the foregoing: One finds the living trees attacked in July and August; the following spring the leaves turn yellow (‘‘sorrel tops”) and gradually red (*‘ red tops”), and the third year they drop off altogether (‘black tops”). It is a difficult matter to say at what point the trees are dead. Girdled trees die with different degrees of rapidity, depending upon the species. The black gum (Vyssa sylvatica) will live—i. e., will have green leaves—for two years after being gir- dled; so also several species of oak. Pines and spruces rarely live more than a year, and generally not so long. The reason for the different behavior of these trees is probably to be found in the different power to conduct water through the inner sapwood. The subject is one about which little is known as yet. In the case of the bull pine, after the girdling by the beetles certain changes take place in the cambium and the newer sapwood which leave no doubt as’ to the death of those parts. By September, as described below, the cambium and bark are actually dead and par- tially decayed for 30 feet or more from the ground. The leaves are still green and full of water the following spring. The only way in which this can be accounted for is by assuming that sufficient water passes through the inner sapwood to keep the crown of the tree supplied THE ‘‘ BLUE’? WooD. alal WHEN ARE THE TREES DEAD? The question as to when a tree is dead is one of considerable prac- tical importance in determining which trees in the forest should be cut. For this purpose it is safe to assume that a tree may be pro- nounced dead when the bark is loose at the base of the tree for con- siderable distances up the trunk. A tree with its bark in this condi- tion can not possibly recover. The wood under this loose bark will always be found to be dark in color and will appear covered with shreds of bark when the bark is pulled off. It must be remembered that such trees will have green leaves. The criterion of green or yel- low leaves is not a safe one to follow, and ought not to be considered in making specifications for cutting dead timber. Attention is here called to the recommendation (4) made on page 35. THE ‘‘BLUE”’ WOOD. Very soor after the attack of the bark beetles (Dendroctonus pond- eros) the wood of the pine turns blue. The color at first is very faint, but it soon becomes deeper. A cross section of a trunk several months after the beetle attack will appear much as shown on Pl. V, fig. 1. Lines of color extend in from the bark toward the center of the tree, and increase rapidly in intensity until the colored areas stand in sharp contrast to the unaffected parts. The color appears in small patches at one or more points on the circumference of the wood ring. At first it is a mere speck, but this gradually spreads laterally and inward, eventually forming triangular patches on cross section. The color likewise spreads up and down the trunk from the central spot. As the time passes after the first attack of the beetles, several color patches may fuse. Their progress laterally and upward toward the cen- ter of the trunk may be equally rapid on all sides of the tree, or more rapid on one side than on another (PI.V, fig. 2). The intensity of the color may vary considerably on the two sides of one and the same trunk. After a certain period of time the whole sapwood will have a beautiful light blue-gray color, as shown on Pl. I. The wood which adjoins the inner line of the ‘*blue” wood is of a brilliant yellow color, which con- trasts sharply with the blue outside and the straw yellow of the heart- wood. This yellow areais in the form of a ring of more or less irregular shape. Sometimes it is formed of one annual ring very sharply defined; then, again, it may include all or only parts of several annual rings. As the wood grows older, the blue color becomes deeper and the yellow ring more sharply defined. RATE OF GROWTH OF THE BLUE COLOR. The first signs of the blue color are usually found several weeks after the attack by the beetles at points on the trunk in the immediate ’ 12 THE ‘‘BLUING” AND THE ‘‘RED ROT” OF THE PINE. vicinity of the attack. The first signs of the blue color are found in the base of the trunk. On PI. VI, fig. 1, three sections of a tree which was attacked the latter part of July, 1901, are shown. The sections were cut in November, 1901, at points 5 feet, 16 feet, and 36 feet from the ground. The sapwood of the first section, 5 feet up, is entirely blued; the second section, 16 feet up, is blue here and there; while the section made in the top, 36 feet up, is without a particle of blue color. Note in this connection that the sections with blue color show the cross sections of the galleries of the bark beetles (Dendroctonus ponderose) in the layer formed by the cambium layer, the outer wood, and the inner bark. The sections on PI. V1, fig. 1, show some of these galleries filled with sawdust. A more adyanced stage is shown on PI. VI, fig. 2. In this tree the sapwood is blue from the ground up into the extreme top. The smallest section, cut from the tree in the upper part of the crown, is blue with the exception of the innermost rings, 1. e., the beginning of the heartwood. The blue color develops very rapidly when once the tree is attacked. Standing trees attacked by the beetles in July, 1902, showed signs of blue color in three weeks. Three months after the attack the sapwood of the lower part of the trunk is usually entirely blue, as shown on Pl. I. The year following the attack, i. e., when the trees have reached the ‘‘sorrel-top” stage, the bluing has reached the top, and late that year, when the ‘‘red-top” stage is reached, the entire sapwood is blue (PL. VI, fig. 2). An experiment was made during the past summer to see whether the blue color would appear in trees felled before being attacked by the pine-bark beetle. It may be said at this point that they did ** blue” just as the standing ones did. NATURE OF THE ‘‘ BLUE” WOOD. Some weeks after the attack by the bark beetles, changes take place in the bark and the newer wood which ultimately result in the bark becoming loose and separating from the tree. When the first flow of resin into the galleries has stopped, the air enters into the galleries, and channels of communication with the outside are established through which the water in the cambium and newer wood can escape. The result of this is that a moist atmosphere prevails in the air chambers, very favorable to the growth of fungi. As the cambium and bark cells lose water they shrivel and break from one another, so that after a few months the bark breaks away from the wood proper. On the south and southwest sides of the trees the bark dies most rapidly, and here, contrary to the general occurrence, it frequently adheres firmly to the tree. On the shaded sides of the trunk the bark becomes loosened, as described, before six months have elapsed. The surface of the wood is moist, very dark in color, and feels somewhat clammy. THE ‘‘BLUE”’ WOOD. 13 Numerous white strands of fungus mycelium make their appearance after six months or more. As the wood of the trunk dries, the bark, loose at first, tightens, so that in the ‘‘black-top” stage it adheres quite firmly to the trunk. When cut into, it peels off in large sheets very readily, however. The ‘* blue” wood differs very little from the sound wood in general appearance, except its color. It is full of moisture at first, but loses this rapidly, so that in two years after the beetle attacks the wood it may be almost perfectly seasoned, even when completely covered with its bark. The ‘‘blue” wood is said to be very much tougher than the green wood, so much so that the tie makers in the Black Hills can be induced to cut wholly blued wood only with difficulty. This toughness and a possible reason therefor are discussed hereafter. STRENGTH OF THE “‘ BLUE” TIMBER. Ever since its first appearance there has been considerable discussion as to the strength and durability of the ‘“‘blue” timber when com- pared with sound timber. It was universally believed that it would prove very much inferior in both respects. A test was made in the testing laboratory of the department of civil engineering of Washing- ton University, St. Louis,” to determine the comparative strength of the ‘‘blue” and the healthy timber. Sections of tree trunks 5 feet long were cut from trees at points 10 to 15 feet from the ground, and were shipped to St. Louis, where they were sawed into blocks of sey- eral sizes. For the compression tests, blocks 2 by 2 by 4 inches and 3 by 3 by 6 inches were cut and planed to the exact dimensions, or as nearly so as possible. For the cross-breaking strength, sticks 2 by 2 inches by 4 feet, and 3 by 3 inches by 4 feet were prepared. The blocks for these tests were kiln-dried at a temperature of 172° F. until an approximately constant weight was reached. It was found that completely dried blocks would not shear atall. The moisture content of the green blocks was slightly higher than that of the ‘** blue” blocks. Three kinds of timber were used: A—Green timber; B—‘** Blue” tim- ber taken from ‘‘sorrel-top” trees, i. e., trees dead about one year; C—“Blue” timber taken from ‘‘red tops” and ‘* black tops” (mostly the latter), i. e., trees dead about two years. The tests were made with the machinery described by Johnson in early reports’ of the Division of Forestry. Every block was carefully measured. The results, reduced to the average crushing strength and the average cross-breaking strength per square inch, are @The machinery was put at the writer's disposal through the courtesy of Prof. J. L. Van Ornum. bTimber Physics, Bulls. Nos. 6 and 8, Division of Forestry, U.S. Department of Agriculture, 14 THE ‘‘BLUING” AND THE ‘‘RED ROT” OF THE PINE. given in the following table. The number of pieces used for each test is given in a separate column. It will be noted that the heart- wood pieces were kept distinct from the pieces cut from the sapwood. Compression strength in pounds per square inch. Heartwood. Sapwood. Kind of timber. q : IN or pieces) AVEC [Or pieces AYerage | ‘tested, | STCDS tose texteden nes Pounds. Pounds. ACT GT@CTs ELM BOY eee re eater ieee nln ate 210 3, 919. 74 } 1,575 5, 089. 98 Be “Blue! timber) L yearold ee eats eee = aeons 190 | 3,876.44 | 649 5, 130. 95 CG. (Blue? timber, 2 yearsvold) hoc eceee ese ccs o et -cae 131 4, 017. 48 | 770 5, 308. 32 Cross-breaking strength in pounds per square inch. Heartwood. Sapwood. Kind of timber q y of pieces| AYeT@EE | of'pieces | Average tested. | Senet. | ‘testea. | Strensth. Pounds. Pounds. fe spy sttlin tin) or eee oe Aan ee eee oe See pe emocaae ca 338 5, 375. 26 553 5, 832. 66 Bo Blue” timber, dyear'qld/2o=~ opens - oon econ eee enes 317 5,361.17 | 242 5, 818. 84 GiSBine?) timber, years oldscseseces=~s25s 1 - oe eae 322 5, 665 272 6, 843. 31 The figures given in this table show that the ** blue” timber is slightly stronger, both when compressed endwise and when broken crosswise. This result is probably due to the fact that the ‘* blue” wood was slightly drier than the green wood when the tests were made. It is scarcely probable that the presence of fungus threads in the cells of the wood in any way strengthens the fiber. However that may be, these tests show beyond doubt that for all practical pur- poses the ** blue” wood is as strong as the green wood. Under the con- ditions now existing in the Black Hills Forest, the ** blue” wood is cer- tainly very much stronger than the green wood. It is in effect sea- soned timber. The trees have stood in the most favorable position possible for drying, with thousands of holes in the bark made by the beetles through which the water could escape, assisted by the winds which constantly sweep by the trunks. Where wood is used, as it unfortunately is in these days, almost immediately after it is cut from the forest, the ** blue” wood is certainly as good so far as its strength is concerned as the green wood, and ought not to be discriminated against because of supposed weakness. LASTING POWER OF THE ‘* BLUE” WooD. The wood of the bull pine is one which is not very resistant to decay-producing fungi. Under ordinary conditions, such as are found THE ‘‘ BLUE” FUNGUS. LS in the State of Nebraska outside of the arid belts and in the Black Hills, the wood will last from four to six years when placed in the ground in the form of a cross-tie, for instance. Dead trees may stand in the forest for many years without decaying, especially when killed by fire, but ordinarily when the bark remains on the trees they begin to decay after the third year. From observations made on the *‘ black-top” trees now standing in the forest it would seem that the lasting power of the ** blue” wood would be very small. It is perhaps not fair to compare these trees with sound ones, for their bark is full of holes, giving fungus spores every opportunity to enter, as described below. When placed in the ground this wood rots very fast, if one can draw conclusions from the dead tops lying around in the forest. There is every reason why it should rot rapidly. The hyphe of the ** blue” fungus have opened pas- sageways for the rapid entrance of water and for other fungi in almost every medullary ray. Dried wood will probably last a long while, especially if properly piled, so as to allow the air to circulate between the separate pieces. When sawed and split for cord-wood, the ** blue” wood should keep just as long as the green wood. The tendency to rapid decay can be largely done away with by treating the wood with some preservative. Ties were cut during the past spring from green timber and from dead trees. These were shipped to Somerville, Tex., where they were impregnated with zinc chloride. These ties were laid in the tracks of the Santa Fe Railroad and are now under observation. A second lot of ties has been cut during the past summer from green trees and from ‘‘sorrel tops,” ‘‘red tops,” and **black tops.” These will be treated within a short time and laid in the track of a Mexican railway so as to determine the relative resistance of the various grades of ‘*blue” timber in a tropical climate as compared with the green tim- ber. On the particular road chosen for this experiment the life of very resistant timbers is short. THE ‘‘BLUE” FUNGUS. The blue color of the wood is due to the growth of a fungus in the wood cells. The staining of wood due to fungi has been known for many years, especially the form known as ** green wood” (ozs verd’). In Europe this green coloration attracted the attention of foresters and investigators as early as the middle of the last century, and a number of descriptions and discussions appeared from time to time (particularly in France), in which an attempt was made to account for this phenomenon. Oy- ok +h x : oe > o- 7 > ——=* : =e a pe ’ > DESCRIPTION OF PLATES. Prater I.—Frontispiece. Cross section of the trunk of a dying tree of the western yellow or bull pine (Pinus ponderosa) from the Black Hills, South Dakota. This tree was attacked by the beetles in August, 1901. The section was cut at a point 6 feet from the ground during the early part of November, 1901. Note the beetle holes in the bark; also the yellow ring between heartwood and sapwood. Piare I1.—Dying trees of the bull pine. Fig. 1 shows several trees; at the left two live, green trees, a “‘sorrel-top’’ tree in the center, and a ‘‘red-top’’ tree at the right. Photographed August 5, 1902. Fig. 2 shows several live, green trees at the left and a “‘sorrel-top’’ tree toward the right. Note that this tree is still green at the top. Photographed August 5, 1902. Piare III.—Various stages showing the gradual color change of leaves of the bull pine (Pinus ponderosa) after they have been attacked by the bark beetles (Den- droctonus ponderose). 1. Leaves froma healthy tree. 2. Leaves from a ‘‘sorrel- top’’ tree. 3 and 4. Leaves from trees changing to the ‘‘red-top”’ stage. When the leaves have reached the stage of 4 they fall off and are completely dead. Puare IV.—Fig. 1. Group of bull pines (Pinus ponderosa) near Elmore, 8. Dak., showing a “‘red-top”’ tree in the center and healthy trees on both sides. Fig. 2 shows a group of ‘‘black-top”’ trees from which all leaves have fallen. This photograph was made in November, 1901, and it is probable that these trees were attacked by the beetles in August, 1899. Piare V.—Sections of trunks of the bull pine (Pinus ponderosa), showing the “blue” disease. Fig. 1 shows an early stage. This section was cut in Noyem- ber, 45 feet up in the trunk, from a tree attacked by the beetles in August of the same year. The tree is still alive at this point. The blue color has started at two separate points. Fig. 2. A later stage, showing the blue color spread out over one-half of the section. Note the yellow ring at the border of heartwood and sapwood. Prats VI.—Fig. 1. Three sections from a bull pine made in November, 1901. This tree was probably attacked by the beetles the latter part of July, 1901. The sections were made at points 5 feet, 16 feet, and 36 feet, respectively, from the ground, i. e., the largest section was cut from the butt, the second one about half way up, and the third in the top. The healthy wood photographs white, and all darker shades represent blued wood. Note the beetle holes in the bark. Fig. 2. Three sections from a bull pine made in November, 1901. This tree was prob- ably attacked by the beetles in July, 1900. It isa ‘‘black-top”’ tree. The sec- tions were made at points 4 feet, 26 feet, and 40 feet from the ground. All are blue. The section near the ground shows ‘‘red rot.’? This happens frequently where the bases of the trees are shaded by long grasses and bushes. In most trees the base will be found sound. The whole tree was dead. Pratre VII.—Mycelium and fruiting bodies of the ‘‘blue’”’ and ‘“red-rot’’ fungi. 1, Tangential section of ‘blue’? wood; m, cross sections of hyphve of the blue fungus ( Ceratostomella pilifera (Fr.) Winter), growing in the medullary rays; h, hyphe growing longitudinally in the wood fibers. These hyphie are brown. 2. Cross section of ‘‘blue’’ wood, showing longitudinal section of medullary ray with hyphee of the ‘‘ blue” fungus (/) growing in the ray and into adjoining cells; the 38 DESCRIPTION OF PLATES. 39 ray cells have been destroyed; m, cross sections of hyphz of Ceratostomella pilifera. 3. Cross section of a medullary ray, with resin duct showing the internal cell walls wholly dissolved out. Masses of brown hyphe, m, of the ‘‘blue”’ fungus extend longitudinally through the ray. 4. Young perithecium of the ‘‘blue”’ fun- gus (Ceratostomella pilifera (Fr.) Winter), grown on pine agar culture. 5. Mature perithecia of the “‘ blue’”’ fungus ( Ceratostomella pilifera (Fr.) Winter), grown on pine agar culture, showing the spores, s, discharging from the top of the beak. The line at the side equals 0.1mm. 6. Two perithecia of the ‘‘blue’’ fungus ( Ceratostomella pilifera ( Fr.) Winter) just before the discharge of the spores. Peri- thecia from culture on pine wood. 7. Two asci with spores of the ‘‘ blue’’ fungus ( Ceratostomella pilifera (Fr.) Winter). 8. Spores of the ‘ blue’’ fungus ( Ceratoxto- mella pilifera (Fr.) Winter). 9. Top of beak of perithecium of Ceratostomella pilifera (Fr.) Winter, just after the discharge of the spore mass. The hyphae composing the tip of the beak have spread out, forming a sort of support for the spore mass. 10 and 11. Median sections of sporophores of the ‘“‘ red-rot’’ fungus ( Polyporus ponderosus, n. sp.), natural size. Puare VIII.—Photomicrographs showing the structure of ‘blue’? wood. Fig. 1. A radial section, showing how the hyphe of the ‘‘blue”’ fungus grow in the medul- lary rays, being confined almost entirely to the rays. Magnification, 80 diame- ters. Fig. 2. A tangential section, showing how the hyphz completely fill the medullary rays. Numerous small hyphe grow out into adjoining cells in a tangential direction. This makes the wood cells in the photograph look as if they were septate. The apparent septa are hyph. Magnification, 80 diameters. Puare IX.—A number of pieces of wood from the bull pine (Pinus ponderosa), show- ing holes made by wood-boring beetles. The trees from which these pieces were taken were in most cases dead, either standing or felled. The “blue’’ fungus has started to grow in the wood cells bordering on these holes, and is gradually spreading to other cells from these holes asa center. Note that these wood pieces show both radial and tangential surfaces. The piece of wood in the center at the bottom of the plate is western hemlock. Priate X.—Sections of “‘black-top’’ trees of the bull pine (Pinus ponderosa) , showing early stages of the ‘“‘red rot’’ caused by Polyporus ponderosus, n. sp. Fig. 1. Section of a dead tree 35 feet up from the ground. This tree had probably been dead for eighteen months to two years. The decay has just started in at several points on the north and northwest sides of the tree. Note that the larger part of the wood is blue. The healthy, unaffected wood is white. Note also the beetle holes in the bark. Fig. 2. A section from a similar ‘‘ black-top’’ tree, showing a more advanced stage of decay. The whole section was blue. The decay started on the side where the bark prevented the rapid evaporation of moisture from the wood and had reached the heartwood. Note the radial and tangential sheets of white mycelium. Fig. 3. A section from the same tree from which fig. 2 was taken, made some 15 feet higher up. The section is blue, but shows few signs of decay. This shows how the ‘‘red rot’’ usually attacks the tree somewhere below the crown. Piare XI.—Sections of ‘‘black-top’’ trees of the bull pine, showing advanced stages of decay caused by Polyporus ponderosus n. sp. Figs. 1 and 2. These two sections were cut from a fallen top of a ‘‘ black top’’ such as is shown in Pl. XIV, fig. 1, one near the point where the top broke off, the smaller one near the top of the crown. Both show how completely the wood has been destroyed. This stage was probably reached about three years after the beetle attack. Fig. 8. The lower figure shows a section cut 4 feet from the ground from a standing “* black- top’”’ pine. On one side a fruiting body of Polyporus ponderosus is to be seen, ? ? 40 THE ‘‘BLUING” AND THE ‘‘RED ROT” OF THE PINE. which is probably two years old. The sapwood is wholly converted into a brown, brittle mass. Such a tree is liable to be blown over at any time. Pirate XII.—A group of “‘black-top”’ trees of the bull pine near Elmore, S. Dak., showing how the tops break off after the trees have been dead for some time. Many of the tops are visible, lying near the base of the trees. A single ‘“‘ black top’’ from which the top has not fallen is seen at the left. The standing trunks are decayed for several feet downward from the point where the top broke off. The base ot these trunks is generally sound, and contains enough timber to make a good cross-tie. Puate XIII.—Fig. 1. View of a broken top, showing how it has broken off almost straight across. Near the middle of the figure a fruiting body of the ‘‘red-rot”’ fungus ( Polyporus ponderosus, n. sp.) is growing out. Fig. 2. Base of a dead bull pine (Pinus ponderosa) near Elmore, 8. Dak., showing a number of fruiting organs of the ‘‘red-rot’’? fungus ( Polyporus ponderosus, n. sp.) growing out from the wood. These are the bodies variously known as “ punks,’’ ‘‘toadstools,’’ ““mushrooms,’”’ or ‘‘frogstools.’’ The double one to the left is very old. Note the cracked upper surface. A section of the trunk made at the point where these bodies are growing out would appear much like Pl. XI, fig. 3. PLate XIV.—Sections of the ends of two cross-ties cut from dead timber, showing defects which are so serious that ties of this kind should be rejected. Fig. 1. Defective because of the ‘‘red rot.’? Fig. 2. Defective because of a disease of the living timber. O DYING TREES OF THE BULL PINE. . 2.—“GREEN” AND “SORREL-T OP” TREES. Plate Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture. CoLoR CHANGES IN LEAVES OF THE BULLPINE L Leaves fram healthy tree. 2. Leaves from “Sorret-top” tree Gand 4. Leaves from trees turning of he Red-top” stag yvd ‘S ‘“SYOW19 YVAN S334] AHLIV3H 40 dNOYD V NI 334] «dOl-3ay ck Die] so ‘Old "SS35N1 «dO1-NOvI1g, Bu! 36, Bureau o f Plant Industry, U.S, Dept. of Agriculture, PLATE IV. eed pan ~s esi no WY 7 A Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agnculture Fis. I] SECTIONS OF TRUNKS OF THE BULL PINE, SHOWING EARLY STAGES OF BLUE DISEAS Bui. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE VI. Fila. 1.—SECTIONS FROM TREE DEAD FIVE MONTHS. Fia@. 2,—SECTIONS FROM TREE DEAD EIGHTEEN MONTHS. “BLUE” SECTIONS FROM DEAD TREES. PLATE VII. Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture WEA Of), “a eR yy UE Yusha Ni ‘ Y | ' Ant N MYCELIUM AND FRUITING Boobies OF “BLUE” AND “RED-ROT” FUNGI. blue’ wood; 3, cross section ¢ a); 5, mature peri whit blue 1, Tangential section of “ blue’? wood; 2, cross section of 1, young perithecium of the ‘blue’? fungus ( Ceratostomella pilife fungus; 6, two perithecia of the ‘blue’ fungus; 7, two aseci with s of the blue” fungus; 9, top of beak of perithecium of Ceratostom the spore mass; 10 and 11, median sections of sporophores of the yres of the pilifera just after t *red-rot’’ fungus 4 osus, 1. Sp.). PLaTe VIII. Fic. 1.—RADIAL SECTION. SECTIONS OF “BLUE” WOOD. ulture Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agnc YT a ar ki Bigs LUE FUNGUS ST SHOWING B BuULLPINE PIECES OF WOOD FROM THE Le ORING BEET 00-BC HOLES MADE BY A Wo a = = “8 rs oy 4 - A », — ware ° - i ae — Bul, 36, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE X. Fia. 2.—SECTION SHOWING MoRE ADVANCED STAGE Fic. 3.—SECTION FROM TREE SHOWN IN oF Decay. Fic. 2, MADE 15 FEET HIGHER UP. EARLY STAGES OF “RED ROT.” ' ; ‘ fi i ‘ . ‘ / : i A « i ' ‘ ie ; 4 / A | ' < e % 4; ALY eo ea . i , Gly “ ‘ : { 1 - os { f . Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE XI. Fig. 3.—SECTION FROM A STANDING PINE, 4 FEET FROM THE GROUND. SECTIONS FROM “BLACK-TOP” WESTERN YELLOW PINE TREES, SHOWING ADVANCED STAGES OF DECAY. TREES *BLACK-TOP” Group OF BROKEN Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture PLATE XIII. i> > wa Ss FiG. 2.—POLYPORUS PONDEROSUS GROWING ON DEAD Pine STUM Bul. 36, Bureau of Plant Industry, U. S. Dept. of Agriculture. PLATE XIV. Fic. 1.—Woop AFFECTED WITH “© RED KOT.” Fic. 2.—DISEASED WOOD FROM LIVING TREE, SECTIONS OF REJECTED CROSS-TIES. U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY—BULLETIN No. 37. B. T. GALLOWAY, Chief of Bureau. FORMATION OF THE SPORES IN THE SPORANGIA OF RHIZOPUS NIGRICANS AND OP PHYCOMYCES NITENS, BY DEANE B. SWINGLE, ASSISTANT IN ParHoLtoey, LABORATORY OF PLanr Parnonoey. VEGETABLE PHYSIOLOGICAL AND PATHOLOGICAL INVESTIGATIONS. IssuED JuNE 27, 1903. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1908. BUREAU OF PLANT INDUSTRY. B. T. Gattoway, Chief. VEGETABLE PATHOLOGICAL AND PHYSIOLOGICAL INVESTIGATIONS. SCIENTIFIC STAFF. Atsert F. Woops, Pathologist and Physiologist. Erwiy F. Sarva, Pathologist in Charge of Laboratory of Plant Pathology. GerorGce T. Moorg, Physiologist in Charge of Laboratory of Plant Physiology. Hersert J. Wesser, Physiologist in Charge of Laboratory of Plant Breeding. Newron B. Prerce, Pathologist in Charge of Pacific Coast Laboratory. HERMANN VON Scurenk, Special Agent in Charge of Mississippi Valley Laboratory. P. H. Roues, Pathologist in Charge of Sub-Tropical Laboratory. M. B. Warre, Pathologist in Charge of Investigations of Diseases of Orchard Fruits. Mark A. Carveron, Cerealist in Charge of Cereal Investigations. Water T. Swrxair, Physiologist in Charge of Life History Investigations. C. O. Townsenn, Pathologist. P. H. Dorserr, Pathologist. T. H. Kearney, Physiologist, Plant Breeding. Cornetius L. Suear, Assistant Pathologist. Wiiuiam A. Orton, Assistant Pathologist. Frora W. Parrerson, Mycologist. JoserH S. CHAMBERLAIN, Expert in Physiological Chemistry. R. E. B. McKenney, Expert. Crarues P. Harriey, Assistant in Physiology, Plant Breeding. Deane B. SwinGie, Assistant in Pathology. James B. Rorer, Assistant in Pathology. Lioyp 8S. Tenny, Assistant in Pathology. Jesse B. Norron, Assistant in Physiology, Plant Breeding. A. W. Epson, Scientific Assistant, Plant Breeding. Karu F, Ketiterman, Assistant in Physiology. Gerorce G. Hepecock, Assistant in Pathology. 9 LETTER OF TRANSMITTAL. U.S. DerpartMENT OF AGRICULTURE, Bureau or Pranr INpustry, OFFICE OF THE CHIEF, Washington, 10», (OLR February 20, 1903. Smr: I have the honor to transmit herewith a technical paper entitled ** Formation of the Spores in the Sporangia of R//zopus Nigricans and of Phycomyces Nitens,” and respectfully recommend that it be pub- lished as Bulletin No. 37 of the series of this Bureau. This paper was prepared by Mr. Deane B. Swingle, of the Pathological Laboratory of Vegetable Pathological and Physiological Investigations, and was sub- mitted with a view to publication by the Pathologist and Physiologist. Respectfully, B. T. GatLtoway, a hief of Bure au. Hon. JAMES Wrison, Secretary of Agriculture. PROB aes The following paper by Mr. Deane B. Swingle, entitled ** Forma- tion of the Spores in the Sporangia of RA/zopus Nigricans and of Phycomyces Nitens,” throws a new light on certain intricate processes in two important genera of fungi. The question of spore formation is one of vital interest to the study of the reproduction and distribu- tion of fungi, both parasitic and nonparasitic. Mr. Swingle’s paper corrects an erroneous idea that has received wide acceptance both in this country and abroad. The inherent properties and behavior of protoplasm must be the basis of work in pathology and physiology. This paper is a contribution to our knowledge, especially in regard to the mechanics of this type of cell-division, and to the nature and functions of the vacuole and the relation of the activities of the nucleus to those of the rest of the protoplasm. The results of this study are in a large measure applicable to many of the other fungi, including a number that are parasitic. The paper is technical and is intended for the use of investigators in pathology and physiology. ALBERT F. Woops, Pathologist and Physic logist. Orrick or THe PArHoLoGisr AND PHystoLocis?, Washington, D. C., February 7, 1903. on CONTENTS. Plate I. Il. 1H 0 VE LECUST RATIONS: Rhizopus nigricans. Fig. 1.—Group of sporangiophores with sporangia. Fig. 2.—Longitudinal section of young stolon.. Fig. 3.—Longitudi- nal section of old stolon. Fig. 4+.—Disintegrating nuclei from old stolon. Fig. 5.—Longitudinal section of young sporangium. Fig. 6.—Longitudinal section of nearly full-sized sporangium --....-.-- Rhizopus nigricans. Fig. 7.—Longitudinal section of full-sized spor- angium before the columella is formed. Fig. 8.—Longitudinal sec- tion of sporangium in which the columella is being formed. Fig. 9.—Section of small part of sporangium showing cleavage furrows. - Rhizopus nigricans. Fig. 10.—Longitudinal section of sporangium showing spore formation. Fig. 11.—Nuclei from columella that has just been formed. Fig. 12.—Sporangium in which the spores are completely formed. Fig. 13.—Nuclei from columella of old spor- angium. Fig. 14.—Ripe spores in their living condition -.....--.- Phycomyces nitens. Fig. 15.—Longitudinal section of young sporan- gium. Fig. 16.—Small part of young sporangium very highly mag- nified. Fig. 17.—Formation of zones in sporangium. Fig. 18.— Layer of vacuoles in/sporangium! 2. == <2. cess ess eee Phycomyces nitens. Fig. 19.—Formation of columella. Fig. 20.— Structure of vacuoles and nuclei. Fig. 21.—Formation of the spores. Fig. 22.—Furrows cutting outward from columella cleft. Fig. 23.—Furrows from the vacuoles cutting out to the periphery. Fig. 24.—Section showing nearly ripe spores. Fig. 25.—Ripe spores in their living condition. Fig. 26.—Peculiar-shaped spores. Fig. 27.—Very irregular-shaped spore -..-..-..----=--- BRO ESS eae Ss Figures illustrating mechanics of cleavage. Fig. 28.— Pilobolus before formation of columella. Fig. 29.—Pilobolus during formation of columella. Fig. 30.—Pilobolus after formation of columella. Fig. 31.—Pilobolus during formation of spores. Fig. 32.—Synchitrium during formation of spores. Fig. 33.—Fuligo during formation of spores. Fig. 34.—Egg of squid during segmentation -...... ae 8 40. 40 P.P.1.—47. V.P.P.1.—10L. FORMATION OF THE SPORES IN THE SPORANGLA OF RHIZOPUS NIGRICANS AND OF PHYCOMYCES MITENS. HISTORICAL. Although the life history and gross anatomy of nearly all the species of the Mucorine have been carefully worked over and described, yet in regard to the cytological details there are the widest differences of opinion, chiefly owing to the fact that only a few forms have been studied with the aid of the most recent methods. It seems desirable, therefore, that others should be critically examined. The present paper is a contribution toward that end. The earliest account that deals specifically with the formation of the spores in the Mucorinew is that of Corda (1838). He investigated the development of the sporangia of 2?A/zopus nigricans, but was able to discover little of the real nature of the process. After the formation of the columella in the lower part of the sporangium, he describes the spores as being formed in rows radiating from the columella, but just how they originate he does not make clear. Van Tieghem (1873, 1875, 1876) in a series of classic papers has covered practically the entire group, describing the structure and development of a very large number of forms with much accuracy and minuteness of detail. He believed that the method of spore formation was the same in all the genera having a spherical sporangium. In these forms the sporogenous protoplasm separates itself into two very different substances—the sporal protoplasm which is always granular, and the intersporal protoplasm which is homogeneous and brilliant. The sporal protoplasm has the form of small polyhedrie portions, and these are separated from each other by the intersporal protoplasm. Soon the polyhedric masses round themselves off, secrete a cellulose wall, and acquire the homogeneous refringent appearance which char- acterizes the spores of the greater number of the Mucorinew. At the same time the intersporal protoplasm distributes itself so that it oceu- pies all the space between the spores, and forms a layer between the peripheral spores and the sporangium wall. Van Tieghem considers this a process of free formation similar to that which occurs in the ascus, differing chiefly in the amount of intersporal protoplasm. v 10 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. Strasburger (1880) has given an account of the more general features of the spore formation in Mucor imucedo. He considers that the spo- rogenous protoplasmic mass is cut up by cell plates analogous to those formed in cell division in the higher plants. This account is, how- ever, very brief and incomplete. Shortly afterwards, Biisgen (1882) studied the formation of the spores in J/ucor. His conclusion is that the protoplasmic mass is cut up into blocks by cell plates, and that these blocks are subdivided until the final spores are reached. In this he adds little to Stras- burger’s account. Léger (1896) published a paper intended to fill the gaps in our knowledge of the spore formation in the Mucorinex. This paper is quite comprehensive, dealing with nearly all the principal genera. Léger studied the spore formation partly by means of sections of material embedded in collodion, but largely by examining the sporan- gia in toto or by crushing them under a cover glass. His results agree entirely with those of Van Tieghem. He finds that all the forms investigated agree in having the protoplasm divided at once into gran- ular portions separated by nongranular plates. Later, the granular masses are surrounded by walls and become the spores, while the nongranular plates form the intersporal protoplasm. In the case of Rhizopus nigricans, Léger finds that when the spores are first formed they are separated by thin membranes only. How these membranes originate he does not make clear. The intersporal substance appears a little later after the spore walls are formed. In this respect /2A¢zopus differs from all the other forms investigated. In his description of the formation of the columella, Léger states that the contents of the sporangium are easily seen to be differentiated into a lighter anda denser portion. These are then separated by a columella wall, the lighter part being included in the columella and all the denser part remaining outside. Just how the protoplasm is divided and the wall formed he does not tell us. He states that the nuclei in the spores are oyal, while those in the columella are spherical. As the spores ripen the cytoplasm disappears from the columella and the nuclei, reduced to nucleoli [sic], remain adhering to the inner sur- face of the columella wall. The nucleus is essentially the same in all the forms which Léger describes. It consists of a nucleolus surrounded by a clear zone which does not stain, and outside of this by a distinct nuclear membrane. The nucleoli are described as so many times larger relatively than I have found them that I am entirely unable to credit his results. The nuclei, also, as he figures them, are much too large and contain no chromatin. Thaxter (1897) has done the most to clear up our knowledge of the spore formation in the Syncephalidw. He states that he was earlier HISTORICAL. 1] inclined to accept the view of Fischer (1892), which is that in such forms as Syncephalis the spores borne in a single row are formed exogenously by constriction like conidia, the wall of the fruiting body forming part of the spore wall, and that this body can not, therefore, be considered as a sporangium homologous with that of Mucor. After athorough study of Syncephalastrum and Syncephalis, however, he accepted the ‘‘sporangial” theory,and brings very con- clusive evidence to support his results. In Syncephalastrum race- mosum he finds that the contents of the cylindrical cells that are to form the chains of spores are divided into spores, not by gradual con- striction from the surface inward, but simultaneously by a hyaline intersporal substance. Walls are then formed around the individual spores entirely within and distinct from the wall of the mother cell. By crushing these spore rows under a cover glass he was able to force the spores out ina perfect condition, leaving the walls of the sporangia empty and intact except for their ruptured tips. This is conclusive evidence of the endogenous formation of these spores. Furthermore, in many cases he finds that the spores are borne, not in single rows, but more or less irregularly, the diameter of the sporangium being somewhat greater than that of a single spore. In such cases the planes of separation are oblique, or even parallel, to the long axis of the sporangium. In such a form as this Thaxter finds an inter- mediate stage between the spherical sporangium of J/ucor and the eylindrical one of Syncephalis, the supposed absence of which was used by Fischer as evidence against the homology of the two. In Syncephalis, Thaxter finds that the separation of the protoplasm into spores is quite different from that in Syncephalastrum. He investigated an undescribed species from Liberia, and also Syncephal/s pycnosperma, and finds that in both cases the protoplasm is cut pro- gressively from the surface inward by *‘intermediary zones,” each of which is made up of an inner nonstainable part, and an outer one that takes stains readily. The spore wall in both species is distinct from the sporangium wall and forms close around the protoplasm, exclud- ing the intermediary zones. In the undescribed species these zones remain until the spores are ripe and then deliquesce, while in Syv- cephalis pycnosperma the stainable portion breaks up into a refractive oily substance and the nonstainable part forms a thick permanent layer around the spore wall and gives to the spores their peculiar shape. Harper (1899) has described the spore formation in ///odo/us and Sporodinia of the Mucorinew, and also in Synech/trium of the Chytri- diacex. The processes in these widely separated forms show many interesting points of similarity. In Synehitriwm, Harper finds that the ** initial cell” contains at first one comparatively large nucleus, which, as the cell reaches nearly its 12. FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. full size, divides rapidly to form a vast number of smaller daughter nuclei. This multinucleated mass of protoplasm is then divided into comparatively large blocks by narrow furrows, cutting progressively inward from the periphery. These furrows cut inward at nearly right angles to the periphery, but, as seen in surface sections, they intersect each other at almost every angle. They are so narrow that they appear in section as single lines which push aside the vacuoles, arrang- ing them in a row on either side. In case the sporangium is slightly shrunken in fixing, however, they appear as slightly separated sur- faces. As these cleavage furrows grow deeper they branch, curve, and intersect each other until the whole mass is divided into multi- nucleated pieces. These are then divided into uninucleated pieces by furrows cutting inward from their surfaces. The nuclei then divide until there are usually from 8 to 12 in each piece. Without further cleavage these multinucleated protoplasmic masses then enlarge somewhat, secrete a protective wall, and become the spores. They then go into a resting condition until germination. In Pilobolus, Harper traces the entire development of the sporan- gium. He finds that when it has reached a considerable size its con- tents are divided into three parts—a central vesicle of cell sap, which, from the absence of a smooth, rounded surface, can not be consid- ered as a central yacuole; outside this, a thin layer of spongy proto- plasm with numerous nuclei; and outside this layer, extending to the sporangium wall, a much denser mass of protoplasm, also containing many nuclei and a few rounded vacuoles. In the spongy protoplasm, and running parallel to the sporangium wall except at the lower side where it extends to the periphery, 2 dome-shaped layer of vacuoles then appears. These vacuoles are at first round, but later they be- come flattened parallel to the surface of the sporangium until they are disk-shaped. They finally fuse, edge to edge, to form a cleft, which, with the aid of a circular furrow cutting upward through the spongy protoplasm until it meets the lowest vacuoles in the series, cuts out the columella. This columella is bounded at first by only a plasma- membrane, outside of which is a more or less open cleft. Later the columella wall is formed in this cleft. It has its dome-shaped outline from the first, and does not begin as a cross wall at the base of the sporangium, being rounded upward later by pressure of turgor from below, as is described for J/vcor in most standard text-books. (See Bessey’s text-book, p. 236.) The spore plasm is then invaded by surface furrows cutting pro- gressively inward. These are much like those in Synehitrium, but wider, owing to the more shrunken condition of the protoplasm dur- ing the process. While this is going on, the vacuoles in the spore plasm become sharply angular, and these angles, continuing outward as furrows, cut into each other and into the furrows from the surface, HISTORICAL. 13 thus aiding in the cleavage. The whole mass is thereby reduced to blocks of varying sizes which are, as in Synchitrivin, progressively cut down to uninucleated pieces. As in Synchitriwin also, these pro- tospores are pressed tightly together by turgor. The nuclei then divide until there is a considerable number in each piece of protoplasm. This division is followed by successive con- strictions of the nature of bipartitions until a binucleated stage is reached. Each piece then surrounds itself with a wall and is a mature spore. The later phases of the process—i. e., from the protospore to the mature spore—Harper regards as an embryonic development. In the subdivisions of the protospores, Harper notes that the pro- toplasm in advance of the cleavage furrows becomes clear and non- stainable, forming a hyaline zone in the plane of constriction, as though the denser part of the protoplasm drew away from this region toward the nuclei, leaving only a clear liquid substance behind. In the earlier stages of cleavage, however, both in PJobolus and in Synchitrium, such a differentiation of protoplasm in advance of the cleavage furrows does not take place. Here, asin Synechitrium, the entire protoplasm is included in the spore, there being no intersporal protoplasm. There is a slime excreted to fill the spaces between the spores, but it is not protoplasm. In Sporodinia the process is in many respects much like that in Pilobolus, Wat there are some striking differences. The sporangia here are much smaller and are composed of two parts, the outer and upper part being filled with dense protoplasm, while the central and lower portion is occupied by a foamy protoplasm, there being no large opening filled with cell sap as in Pilobolus. The vacuoles that cut out the columella are much larger than in P7/Jobo/us, and are arranged on the line, as it appears in section, between the two kinds of proto- plasm. They fuse laterally to form a curved cleft, but no surface furrow cutting in to meet them has been observed. The spore plasm is then divided into blocks by furrows cutting from the columella cleft outward and from the surface inward, but here the cleayage process ceases. No uninucleated stage is ever reached. These protoplasmic blocks contain numerous nuclei, and round off and are covered with a cell wall. They are then the mature spores. This is a considerable abbreviation of the process in /%//obolus, and there is a corresponding shortening in the time required for developing the spores in Sporodinia. The nuclei in all three forms are made up of the same parts as those in the higher plants. There is a nucleolus surrounded by a zone filled with nuclear sap and chromatin, the whole being enveloped in a nuclear membrane. A. point well worthy of consideration is that the nuclei are in a resting condition during cleavage. Hans Bachmann (1899) has described the entire structure and 14. FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. development of a new species of J/ortierclla. Though the paper was published yery recently, little improvement over the older writers is shown in the matter of technique. He has not, so far as he states, made any sections of the sporangia. By a study of entire sporangia he finds that the surface comes to be marked out into polygons separated by rather broad bands of an even width. These markings he interprets as representing a surface view of polvhedrie masses of protoplasm which are destined to become spores, separated by layers of intersporal protoplasm. Plasmolyzing agents in some cases cause the sporangium to contract as a unit and not as individual polygons, showing that each is not yet entirely surrounded by an osmotic membrane. Gentian violet stains the material between the polyhedrons; the formation of the violet lines is progressive, as is shown by the fact that in some cases they are short and do not extend over the entire sporangium, but radiate from yarious points. In this, Bachmann makes a decided advance over Léger, but still he apparently fails to grasp the most important point—that these blue-staining lines represent cleavage furrows filled with the stain. METHODS. The mold RA/zopus was first obtained in mixed cultures by exposing moistened bread for a few minutes to the air of the laboratory. To obtain pure cultures, a few sporangia were carefully transferred from the original cultures to slightly moistened bread, which had been exposed an hour or so on two or more successive days to a temperature of from 60° to 65° C. in a steam sterilizer. In from one to two days after inoculation the stolons began to appear on the surface of the bread, and in another day there were a considerable number of sporangia formed. The cultures of Phycomyces were obtained from Ann Arbor, Mich., through the kindness of Dr. J. B. Pollock. This mold was grown either upon sterilized bread or nutrient agar. From these cultures small bits of mycelium were cut out (below the surface of the sub- stratum in the case of /’%/ycomyces) and instantly immersed in the fixing fluid. After remaining in this about twenty-four hours, they were washed for a few hours in running water, dehydrated hy run- ning through grades of alcohol, cleared in xylol or chloroform, and embedded in paraftin. The sections were cut on a Jung, or a Reinhold-Giltay microtome, usually 4 ye thick, but sometimes 2 44, and were fastened to slides with albumen and glycerine. They were then stained with Flemming’s triple stain (safranin, gentian violet, and orange G), then dehydrated, cleared with clove oil or bergamot oil, and mounted in Canada bal- sam. If the right exposures are given to these stains, the cytoplasm RHIZOPUS NIGRICANS. 15 appears orange, the chromatin blue, the nucleolus and proteid erystal- loids red, and the cell wall either blue or orange. For fixing fluids the mixtures of Flemming, Hermann, and Merkel were used with yery good results. Eisen’s fluid gave some very fine results, but was little used. An exposure of one hour to Flemming’s fluid, followed by twelve to twenty-four hours in Merkel’s fluid or chrom-acetic acid, gave especially fine preparations, not being so much blackened as when exposed longer to the osmic acid. I am deeply indebted to Dr. Robert A. Harper of the University of Wisconsin, and to Dr. Erwin F. Smith, Dr. Rodney H. True, and Mr. Karl F. Kellerman of the United States Department of Agriculture for many valuable suggestions and criticisms given during the prog- ress of the work. RHIZOPUS NIGRICANS Ehrbg. The general morphology of R/7/zopus has been yery well described by the earlier authors. The spore in germinating sends out a tube which branches until a tangled mycelium is formed in the substratum. This mycelium sends up from various points aérial hyphae, which are erect at first and form a delicate white growth in the cultures. After these hyphx reach the height of one or two centimeters they bend oyer and grow horizon- tally along the surface of the substratum. When one of these stolons has grown in this direction for a short distance, it forms a swelling at the apex two to four times the diame- ter of the stolon, and out of this grow from two to six branches, one of which is in reality a continuation of the stolon, while the others grow into sporangiophores. (PI. I, fig. 1.) If this swollen portion of the stolon comes in contact with the substratum or the sides of the cul- ture dish, a few rhizoids are sent out which firmly anchor it, and, in case they penetrate any nutritive substance, these doubtless aid in nourishing the sporangiophores. The stolon continues growing out and forming these groups of sporangiophores at intervals, and finally ends with such a group at the apex. Each sporangiophore bears a single spherical sporangium, In healthy stolons, especially if they are growing rapidly, the pro- toplasm is almost continually streaming in one direction or the other. This has been fully described by Arthur (1897), who considers that it is principally due to evaporation of moisture from the surface of exposed parts, together with the constant taking in of water by the hyph that are in the substratum. In his conclusion he expresses the opinion that *‘ the moyement is an incidental feature in the life of the plant.” Further mention of this paper will be made in connection with the distribution of the protoplasm in the sporangium. 16 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. The growing ends of the stolons are densely crowded with proto- plasm containing many nuclei. This condition prevails for some dis- tance back in the stolons (PI. I, fig. 2), but as we follow back toward the older part the protoplasm is more and more permeated with cell sap, and at last we find a region where there is nothing but a wall filled with cell sap, so far as we can distinguish from a surface view of liy- ing material. In stained sections, however, as shown in PI. I, fig. 3, it can be seen that there is still a thin layer of protoplasm lining the wall, and strands or even small masses of it in the center. In parts as old as that shown in the figure, the nuclei haye begun to disinte- grate somewhat, and appear as tiny red-staining masses of various shapes. (PI. I, fig. 4.) The young sporangiophores, like the ends of the stolons, are densely crowded with protoplasm and nuclei, and even the lower part of the _older ones is never entirely devoid of protoplasmic contents, as is stated by Léger, but retains a structure very much like that in the stolons. As the sporangiophore reaches its full length it begins to swell out at the tip into a tiny round body, the future sporangium. The con- tents of this are at first evenly distributed, being equally dense in the center and at the periphery, but before it has reached half its final size the protoplasm begins to be decidedly dense toward the sporan- gium wall, while in the center it is of a much looser structure. PI. 1, t=) fig. 5, shows the distribution of the cytoplasm and nuclei at this stage. There are also present a few crystalloids. They seem often to be in tiny clear vesicles, but whether or not these are ordinary yacuoles I can not be certain. These crystal-like bodies vary much in size, and as a rule increase in number as the sporangium gets older. It is quite noticeable, however, that they are entirely confined to the central part of the sporangium. The nuclei are so small that they appear only as dots in a drawing of the size of Pl. I, fig. 5. Their structure can, however, be clearly made out with higher magnification, and it is to all appearances pre- cisely like that of those shown in PI. IT, fig. 9, which will be described later. The cytoplasm in young sporangia, it will be observed, is quite dense next the sporangium wall, but gradually becomes less dense toward the center, where it is of a very loose spongy structure, con- taining many vacuoles of considerable size. There is at this stage no sharply defined boundary between the denser and the less dense parts of the cytoplasm, but a gradual transition from center to periphery. The denser layer does not, however, extend quite to the sporangio- phore at the base of the sporangium. (PI. I, figs. 5 and 6.) At this time also there is a very marked streaming of the protoplasm up the sporangiophore into the sporangium. These currents appear RHIZOPUS NIGRICANS. 17 as a bundle of strands, which in optical section spread fan-like as thcy enter the sporangium and extend toward the periphery. Many of these streams, particularly at the sides, extend nearly to the sporan- gium wall, as seen in PI. I, fig..5. Harper (1899) has described each individual current in Pé/obolus as having “‘marked a path for itself through the protoplasmic structure. It is marked by continuous deli- cate films, quite distinct from the spongy structure of the adjacent plasma.” These surrounding films, as the writer has seen them in Rhizopus, are of a more hyaline and homogeneous appearance than either the currents or the surrounding cytoplasm. The nuclei in these currents are much elongated in the direction in which the currents run and I have not been able to differentiate the parts, the nucleolus and chromatin both staining red. As the streaming continues the protoplasm at the periphery becomes denser, and there appear clearly differentiated layers within the spo- rangium. The beginning of this differentiation is not simultaneous throughout the protoplasm. It appears at certain points approxi- mately equidistant from the periphery, between which and the periphery the thickening of the protoplasm forms a dense zone lining the sporan- gium except at the base, where its inner boundary line gradually extends to the periphery. In the stage shown in PI. I, fig. 6, this boundary line is not perfect, but somewhat broken, admitting thin streams of loose protoplasm from the interior of the sporangium. Inside this zone and of about one-third its thickness is a semitransparent layer consist- ing of loose protoplasm like that which fills the interior of the sporan- gium, but clearer and less granular, and taking the orange stain less strongly than the latter. In structure it resembles the thin films about the streams previously mentioned. The cytoplasm inside this semi- transparent zone and occupying the central and lower part of the sporangium is of a loose, spongy, much-vacuolated structure, contain- ing scattering nuclei and a considerable number of proteid bodies. There are no marked protoplasmic strands indicating currents in the center of the sporangium, though the writer has often found them in later stages; but radiating from the central part of the sporangium and passing from it through the clear zone to the denser plasm are many very slender strands marking the paths of currents. These currents bear nuclei and seem to represent a very late stage of the migration of cytoplasm and nuclei toward the periphery. Some of these streams enter the openings in the denser plasm, while others run against its inner surface. This streaming goes on until the inner boundary of the denser plasm is at all points sharply detined. This boundary does not consist of a membrane or of any differentiated layer. The denser plasm at this time contains only a very few vacuoles of any considerable size, but under very high magnification it can be seen 20844—No. 87—03 » 18 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. that there are very many exceedingly small ones, with definitely rounded outlines. Most of these are scarcely larger than the nuclei, and some are much smaller. They can not, therefore, be shown in a drawing on so small a scale as Pl. 1, fig. 6. They are, however, essen- tially the same in size, number, and distribution as those shown in Plies) Thus far, except for the arrangement of the cytoplasm and nuclei, we have had no phenomena in the sporangium that even suggest cell divi- sion, unless possibly it be the clear zone. The greater part of the more solid portion of the cytoplasm has formed itself into a layer at the periphery. Nearly all of the nuclei also have migrated into this por- tion of the sporangium, and are distributed irregularly throughout the dense cytoplasm. They are not even approximately equidistant from each other, nor are they often, if ever, in actual contact, though Léger states that such is very frequently the case. How he could determine the normal distribution of the nuclei from crushed sporan- gia is difficult to-comprehend. As soon as the protoplasm is distributed as has been described, the separation of that which is to be included within the columella from that which is to form the spores begins. The columella is not at first a flat cross wall at the base of the sporangium which is later pushed up by turgor to its characteristic dome shape, as it is currently described as doing, but is laid down in essentially the same fashion as described by Harper (1899) for Pilobolus. There first appears in the denser plasm a single layer of spherical vacuoles (Pl. I, fig. 7) running par- allel to its inner surface. ‘The layer of the denser plasm inside the system of vacuoles is usually from one-fifteenth to one-twentieth as thick as the layer outside. Apparently these vacuoles are formed by the enlargement of the very minute ones already mentioned that lie in this region, rather than by the migration of previously enlarged vacuoles. In sporangia in which this layer of vacuoles is only partly formed there are usually a few large vacuoles arranged in the layer, and between them are smaller ones, varying in size down to the small- est in the sporangium (PI. II, fig. 7). This leads one to believe that the vacuoles in this layer are essentially like the others in the sporan- gium and in the mycelium. ‘These vacuoles and all others in the spo- rangium agree with those of 2//obo/us and Sporodinia in being deyoid of all stainable contents (PI. II, fig. 7), in which respect they differ strikingly from those of Phycomyces, described later. The vacuoles are at first spherical, or nearly so, but soon begin to flatten, their long axes being parallel to the inner surface of the denser plasm. By this flattening they become disk-shaped, as in Pl. I, fig. 8, and the edges of adjacent ones come in contact and fuse, forming a narrow curyed cleft in the protoplasm, At the same time a circular furrow begins to cut upward from the surface of the protoplasm at RHIZOPUS NIGRICANS. 19 the base of the sporangium through the denser plasm (PI. II, fig. 5). This furrow increases in depth until it reaches and fuses with the lowest vacuoles in the layer. Thus the protoplasm of the sporangium is divided into two distinct portions destined to perform radically different parts in the further life of the plant. That-outside the cleft is to be entirely cut up into spores, while that inside is later to be surrounded by the columella wall and plays no direct part in repro- duction. The former I shall distinguish as the spore-plasm and the latter as the columella-plasm. It will be noted from what has been already said and from PI. II, figs. 7 and 8, and Pl. IIT, figs. 10 and 12, that the columella-plasm includes all the looser plasm in the sporan- gium and also a thin layer of the denser plasm. One might have expected from Pl. 1, fig. 6, that the columella wall would be laid down in the clear zone shown in that figure, but that such is not the case there is no room for doubt. The writer has preparations in which this zone is still almost as marked as in the fioure mentioned, while the columella cleft is forming in the denser plasm. Pl. Il, fig. 8, and PJ. III, fig. 10, show that the outer part of the looser plasm is still somewhat clearer than that in the center, though the paths of the currents have become almost obliterated. The time for the disappearance of the currents varies greatly in different sporangia. There is no visible difference while cleavage is going, on between the denser plasm inside the layer of vacuoles and that outside, nor is there any differentiation of the cytoplasm between the vacuoles or in advance of the surface furrow, such as Harper found in the late sub- divisions of the protoplasm of //obo/us and in the last stages of cleay- age of Fuligo (1900). While the cutting out of the columella is going on, the sporangium gives every appearance of having only slight turgidity. The cleft in the protoplasm is always quite wide—at least in certain places. When, however, the cleavage is complete, the protoplasmic masses increase in volume and become strongly turgid again, causing the two proto- plasmic surfaces lately separated to become pressed together so tightly that only by the closest study can one follow the cleft throughout its entire extent. In case the spore cleavage, which will be described later, begins before the columella cleft is completed, as often occurs, this period of turgidity is postponed until after the spores are entirely cut out. It will be noted that when first formed the cleft around the colu- mella is bounded by two protoplasmic surfaces. When these surfaces become tightly pressed together by the turgor in the sporangium, one might expect them to fuse into a continuous mass of protoplasm again, there being no wall between them at this time. Indeed, sueh a phe- nomenon was described by Biisgen (1882) in the formation of the 20. FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. spores of the Saprolegniexw. It is not, however, surprising that with the technique used in those days he should fail to see that there was still a distinct boundary between the closely packed spores. When the period of turgor relaxes a little the two surfaces generally separate slightly, but at irregular intervals points are often found where they still adhere, forming tiny conical projections, whose apices are for a short time in contact. In the behavior of these two protoplasmic surfaces we have consid- erable additional evidence for the existence of a definite plasma-mem- brane. ; Even before the cutting out of the columella takes place the nuclei of the looser protoplasm begin to disintegrate. In very young spo- rangia all the nuclei have the same normal structure, but in the one shown in PI. I, fig. 6, for example, they are clearly suffering disinte- gration in the center of what is to become the columella-plasm, though out near the denser plasm they retain their characteristic structure, often until the spores are nearly ripe. (PI. ILI, fig. 13, a.) It might be suggested that the nuclei in the center of the sporangium are not well fixed, but these sporangia are*so small and thin-walled that I can not believe, with all the cytoplasm and the greater part of the nuclei having a perfectly normal structure, that the difference in appearance of these nuclei is to be attributed to poor fixation, espe- cially as it is essentially the same for all the best fixing fluids used. The first sign of disintegration is the appearance of a red-staining mass on one side. As the process goes on, the whole nucleus comes to appear as a slightly shrunken, homogeneous mass, often irregular in shape, and staining the same shade of red as the crystalloids. It might be argued that these red-staining bodies are crystalloids whose substance is being dissolved, but I have found very good evidence that such is not the case. As shown in PI. III, figs. 11 and 13, there are all stages of disintegration between the almost perfect nuclei and the most shrunken and angular ones. On the other hand, all the erystalloids in these sporangia, so far as could be observed, are perfect in shape, none showing notches or marks of corrosion, such as we should expect to find if they were being dissolved. Furthermore, the crystalloids seem to be forming rather than dissolving, judging from their greater number and size in the older sporangia. In Pl. ILI, figs. 11 and 13, represents a nucleus with normal structure lying just inward from the denser plasm, while 4, ¢, and d lie nearer the center and are breaking down. In no sporangia as old as that shown in PI. I, fig. 6, have I found nuclei in or near the center of the looser plasm in which nuclear membrane, chromatin, and nuele- olus could be distinguished. These nuclei do not entirely disappear during the life of the plant, nor would it be at all accurate to say, as Léger has done, that they are ** reduced to a nucleole.” RHIZOPUS NIGRICANS. yi The formation of the spores usually begins after the columella cleft is complete, although in some instances (as in PI. I], fig. 5) somewhat previous to that, but always before the laying down of the columella wall. Spore formation does not take place in the manner described by Van Tieghem and Léger—by the simultaneous differentiation of plates of hyaline nongranular protoplasm cutting the spore-plasm into polyhedric blocks—nor by the progressive differentiation of such plates from lines on the surface of the protoplasm, as described by Bachmann (1900). In the scores of sporangia sectioned in all stages of development the writer has not found at any time even the slightest indication of such a differentiation of the protoplasm into granular polyhedric masses with nongranular plasm between. ‘The first indica- tion of the division of the spore-plasm is the formation of furrows at the surface, which cut progressively inward. (PI. II, figs. 8 and 4.) These furrows are not broad,as in 2//obo/us, nor are their sides closely pressed together, as in Synechitrium. They cut in at very different angles to the surface of the sporangium, and pass between, and often very close to, nuclei and vacuoles. (PI. I, fig. 9.) They usually branch or curve at a short distance inward from the surface, and by cutting into and fusing with neighboring furrows cut out small pieces of the surface layer of the protoplasm of the sporangium. These pieces are almost always the definitive spores, lacking only the walls. Only a few of the larger ones are further divided up. There is no uninucleated stage in the spore formation of 2/A7/zopus, as in Pilobolus, it being like Sporodinéa and Phycomyces in this respect. These spores are at first somewhat angular in shape and contain exactly the same number of nuclei (2 to 6) as when ripe, there being no nuclear division at any stage of their existence previous to germination. The nuclei of the spore-plasm during all stages of cleavage are ina resting condition. (PI. IJ, fig. 9.) Each consists of a nucleolus, or occasionally two nucleoli, which in my preparations is stained a deep red, surrounded by a zone of evenly granular, blue-staining chromatin, the whole being bounded by a definite nuclear membrane. Both in the spore-plasm and in the columella the nuclei are spherical or very slightly ovoid until they begin to disintegrate. They are relatively more numerous in some sporangia than in others, which may possibly be due to differences in the moisture supply, wet cultures making looser and more bulky cytoplasm than drier ones. The vacuoles of the spore-plasm, which are for the most part exceedingly minute, as can be seen by a comparison with the nuclei in Pl. II, fig. 9, do not become angular and assist in dividing the pro- toplasm here as in Pilobolus and Phycomyces. They retain their rounded form throughout the entire process of cleavage, even when furrows cut very close to them. As previously stated, they contain nothing but ordinary cell sap. 22. FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. After the surface furrows have cut inward for a considerable dis- tance, a few similar furrows begin to cut outward from the columella cleft, which as yet contains no wall. (Pl. III, fig. 10.) With the meeting of these two systems of furrows the cleavage is practically complete. During the process of spore cleayage the protoplasm is slightly shrunken, apparently hecause of the giving off of water. The furrows are more or less open and filled with clear cell sap only. (Pl. II, fie. 9.) As soon, however, as the cleavage is complete, the spore mass becomes strongly turgid again, and each spore so increases in volume that all are pressed tightly together and the furrows are entirely closed, so that with the Zeiss 2 mm. immersion objective, 1.30 aperture, and No. 18 compensating ocular, they appear in optical section as single lines and are very hard to trace through the dense spongy cytoplasm. The spores are thus made sharply angular, but later they round off, leaving little spaces between them. The formation of the columella wall usually begins before the spores are entirely cut out, but it does not reach its definitive thickness until they are nearly ripe. As seen in Pl. III, fig. 12, these spores have no regular system of arrangement whatsoever, and the writer can not find the slightest ground for Corda’s view that they are in radial rows. ; As already stated, the spaces between the spores contain at first absolutely nothing except cell sap. There is no trace of any inter- sporal protoplasm, such as has been described by the earlier authors and considered as homologous with the epiplasm of the ascus. The spores of RA/zopus are at first angular and covered by only a plasma-membrane, but soon round off and a firm wall is formed about them. During this process of ripening a homogeneous slime is excreted by the spores, which fills up the spaces between them. In such exposures to the triple stain as best bring out the cytoplasmic and nuclear structures this intersporal slime does not stain at all, and for this reason the writer has left it an empty space in Pl. II], fig. 12. By a longer exposure to the violet it is readily brought out as a smooth bluish mass filling up the spaces between the rounded spores. There is no special mechanism for the discharge of the spores in Rhizopus as in Pilobolus. There is, however, an inner layer of the sporangium wall that can not readily be differentiated from the rest of the wall in specimens fixed in the killing fluids of Flemming and Merkel; while in those fixed in Kisen’s fluid and stained in the triple stain it is very readily distinguishable from the outer layer by its lighter blue color, the boundary between the two being sharply defined. PI. I, fig. 7, is therefore the only one in the writer’s series in which he could show the separate layers of the sporangium wall. PHYCOMYCES NITENS. 25 The inner layer is somewhat thicker than the outer, both being of an even thickness except for a little space around the sporangiophore where the inner one thins out and disappears. Whether or not this is homologous with the ‘‘collar” of Pilobolus, the writer can not be certain. The spores are set free by the bursting of the sporangium wall, without its being thrown off. Whether or not the inner layer of the wall swells by the absorption of water and bursts the outer layer the writer has not determined. The writer has never found this inner layer on sporangia as young as that shown in Pl. I, fig. 5. nor in the walls of the mycelium. The ripe spores as they escape from the ruptured sporangia are mostly ovoid in shape and of varying sizes. Their walls are marked with longitudinal ridges, as may be seen in PI. II, fig. 14. PHYCOMYCES NITENS Kunze. Unlike LAzzopus, the sporangiophores of Phycomyces are borne singly, springing directly from the mycelium. When the sporangio- phore is yet only a few millimeters long, the apex begins to swell out into a sporangium in the same manner as that described for P//zopus nigricans. As the sporangium enlarges the sporangiophore elongates, pushing up the former farther and farther from the surface of the substratum. The spores are formed when the sporangiophore is about 2 em. long, and it is then that the sporangium has its maximum diameter. As shown in PI. LV, fig. 15, there is the same streaming of cytoplasm and nuclei up the sporangiophore and out toward the periphery of the sporangium as in LAizopus nigricans. As can be seen by a comparison of Pl. LV, fig. 15, with Pl. I, tig. 5, the cytoplasm in the young sporangium of /7/iycoiiyces is more coarsely granular than that of RA/zopus and takes the stain much more deeply. ’ The most noticeable difference between the young sporangium of Phycomyces and that of RA‘zopusis that in Phycomyces there are many more large round vacuoles which, as they move outward toward the periphery of the sporangium, become tilled with a visible content. (PI. IV, fig. 15.) This content appears in sections stained with the triple stain asa bluish homogeneous body of the same shape as the vacuole but somewhat smaller in diameter, lying in the middle of the vacuole, with a clear zone between it and the vacuolar membrane. (Pls. 1V and V, figs. 15 to 22.) This content begins, not asa very minute, sharply-staining body which grows larger and larger in diameter, but as a faintly-staining mass which, as it grows older, becomes more dense and takes the stain more strongly. In the youngest stage it appears quite as large in proportion to the size of the vacuole in which it lies as when it becomes older. (PI. IV, tig. 16.) It forms in 24 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. the vacuoles w7ter they have entered the sporangium—neyer, so far as I have observed, appearing in those of the mycelium or sporangio- phore, and rarely in those of that part of the sporangium which lies close to the mouth.of the sporangiophore. The younger stages of their formation are shown in Pl. IY, figs. 15 and 16, while in Pls. IV and V, figs. 18, 19, and 20, they have reached their maximum density. As the protoplasm streams up into the sporangium and out toward the periphery, there is at first a gradual transition in density from the center outward, precisely as in ?//zopus at the same stage. (PI. IV, fig. 15.) A little later, however, as in PI. LV, fig. 17, it is divided into three regions, differing in density. The outer region or layer is very dense and takes the stain strongly. Inside this is a second layer, which is considerably less dense and stains less strongly. Inside this second layer and occupying the central part of the sporangium is a region of very loose and much yacuolated protoplasm which takes the stain scarcely at all. Between the interior region and the second layer the the differentiation becomes very sharp, but, as in RAzzopus, there is no wall or membrane of any kind between them. Between the second and the outer layers, however, the transition is at first very gradual (Pl. TV, fig. 17), but becomes more and more sharp as the sporangium grows older. I haye never found in Phycomyces a stage such as is shown in PL. I, fig. 6, which occurs regularly in RA‘zopus. It is pos- sible that this second layer is homologous with the semitransparent zone that has the same relative position in the sporangium of Rh/zopus. Ihave not regarded it as such, however, as it is of so much greater relative density and contains no delicate strands representing currents. It is interesting in this connection to compare PI. I, fig. 6, with Pl. IV, figs. 17 and 18. The nuclei are at first about evenly distributed in the outer and second layers, but in the interior there are very few, or for a short period in the development of the sporangium none. (PI. LV, figs. 17 and 18.) None of the vacuoles in the interior region of very loose protoplasm or in the inner part of the second layer has the stainable content men- tioned above. Practically all of the larger ones in the outer dense layer contain this substance, however, as also do most of those in the outer part of the second layer. (PI. IV, fig. 17.) Between these larger vacuoles are yery small ones which contain nothing that takes the stain. (Pls. [V and V, figs. 15-24.) The difference in the destinies of these two kinds of vacuoles will be seen later. As may be seen from Pl. 1V, figs. 15, 17, and 18, and Pl. V, fig. 19, the vacuoles that contain the stainable substance are very numer- ous, taking up a considerable portion of the space in the sporangium and lying very close together, often two or more being in actual con- tact, their clear zones being separated by only the yacuolar mem- branes. (Pls. IV and V, figs. 15, 17, 18, 19, and 20.) In such cases PHYCOMYCES NITENS. 95 the vacuolar membrane is isolated from the remainder of the proto- plasm for a little space, and may readily be seen end studied by itself. (Pl. V, fig. 20.) It is very thin and homogeneous, taking the violet stain very slightly, which gives it a faint blue color. When two vacuoles are thus in contact they are usually flattened against each other, so that the membrane between appears in optical section asa thin, straight line. In such cases the contents are often flattened on that side to conform to the shape of the vacuole. (PI. V, fig. 20.) A considerable number of the nuclei that are in the second layer when it is first formed migrate into the denser plasm, and the differ- entiation between the two layers becomes more distinct. Then a layer of vacuoles, practically all having stainable contents, becomes arranged in a dome shape in the denser plasm and running parallel to its inner surface. (Pl. IV, fig. 18.) These vacuoles flatten out, become disk-shaped, and fuse edge to edge to form a dome-shaped cleft in the denser plasm, as in PAzzopus and Pilobolus. (Pl. V. tig. 19.) It is interesting to note that as the vacuoles flatten, the content flattens also, so that its surface remains always more or less parallel to the yacuolar membrane. (PI. V, fig. 19.) So far as I have been able to observe, there is never a surface fur- row that cuts inward to meet the lowest of the layer of vacuoles, as is the case in /%/obolus and Rhizopus. In this respect Phycomyces appears more like Sporodinia. The layer of vacuoles begins so very near the surface of the protoplasm (Pl. V, fig. 19) that if there is such a surface furrow it must be very shallow indeed. I have never found any evidence of its existence. When the vacuoles of this layer have entirely fused, edge to edge, the separation of the columella is complete. There is at first no wall— simply a cleft bounded by plasma-membranes. The contents of all the vacuoles that make this cleft have now fused, forming a layer of slightly uneven thickness separating the outer surface of the columella plasm from the inner surface of the spore-plasm. All the very loose interior protoplasm, the second layer, and a small part of the denser plasm are included within the columella, while the greater part of the denser plasm goes to form the spores. As soon as the differentiation of the columella is complete, or in exceptional cases a little before, the formation of the spores begins. Here we get a most striking difference between Phycomyces and Rhizopus. The large round vacuoles in the spore plasm begin to lose their rounded form and become angular. (Pl. V, figs. 21 and 22.) These angles become sharper and sharper, and appear to cut through the cytoplasm between the nuclei, and when they encounter each other fuse to form irregular clefts. The cytoplasm in advance of these vacu- olar furrows shows no visible differentiation, but remains of an even density throughout the entire spore-plasm during the whole process of 26 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. cleavage. (PI. V, figs. 21 and 22.) So far as the writer has been abie to observe after a most diligent search in a very large number of spo- rangia in all stages of spore formation, there are never surface furrows cutting into the spore-plasm at any point. The angles from the vacu- oles may often be seen cutting out to the surface of the spore-plasm. (Pl. V, figs. 21 and 23.) Furrows also cut into the spore-plasm from the columella cleft and fuse with the vacuolar furrows in the spore- plasm, and thus aid in dividing the protoplasm into spores. (PI. V, fig. 22.) During the whole process of spore formation the nuclei are in a resting condition. They are spherical, or nearly so, and are made up of one or two nucleoli and finely granular chromatin within the nuclear membrane. (Pl. V, figs. 20-24.) They are a little larger than those of RAvzopus. The furrows often cut very close to them, but they give no visible sign of being in any way affected by the cleay- age of the cytoplasm in which they lie. (Pl. V, figs. 21-23.) I have never observed a single case of nuclear division in the sporangium of Phycomyce Ss. The very small vacuoles deseribed above that have no stainable con- tents do not take any part in the cleavage. They remain round through- out the process, even when the furrows from the larger vacuoles cut very close to them. (Pl. V, fig. 22. As the yacuoles that take part in the cleayage become angular the content becomes angular also, taking approximately the shape of the vacuoles, so that its surface is parallel to the vacuolar membrane, but seldom in contact with it, there being still the clear nonstainable zone between. (PI. V, figs. 21 and 22.) As the angles of adjacent vacuoles fuse, the contents are brought in contact and fuse also, thus forming amass filling up the spaces between the spores. (PI. V, fig. 21.) It will clearly be seen that this mass is not protoplasm, as it originates as a secretion from the vacuolar membrane deposited inside the vacuole. It is homogeneous at the time the spores are formed, staining bluish- brown in Flemming’s triple stain and containing no nuclei or other inclusions. All the cytoplasm and nuclei of the spore-plasm are included within the spores themselves. (PI. V, fig. 21.) There appears to be a considerable shrinkage of the protoplasm while the cutting out of the columella and the spore formation are going on, and this is followed by an increased turgidity of the protoplasmic masses, but this is not so marked as in RA/zopus and Pilobolus and the spores do not become sharply angular. ‘This increase in turgidity of the protoplasmic masses is followed by a very marked enlargement of the small vacuoles, which did not take part in spore formation. They still, however, contain only ordinary cell sap and no stainable contents. The columella wall begins to form while spore cleavage is going on, and continues to thicken until the spores are nearly ripe. PHYCOMYCES NITENS. 27 Up to this time the spores are surrounded by only a plasma-mem- brane, the spore wall not yet having been formed. They now begin to round off and contract, the vacuoles become very much smaller, and the whole spore is thereby much reduced in size and surrounds itself with a wall of considerable thickness. At the time the spore yall is formed the plasma-membranes of the adjacent spores are not in contact, but are separated by the intersporal slime from the yacuo- lar contents. The plasma-membranes of the spores, except in the peripheral layer, originate entirely from the vacuolar membranes, without visible change except in form. Only a part of the plasma-membrane of the spores in this layer is made up of the original plasma-membrane of the sporangium. In this respect there is a marked difference between Phycomyces wd Rhizopus. The spores vary greatly in size and in the number of nuclei. Every spore has at least one nucleus, and some have as many as twelve or per- hapsmore. Asa rule, thereare about sixoreight. In PI. V, fig. 25. and ¢ show the extreme sizes of the spores and / the usual size and shape. Unlike RA‘zopus, the walls of these spores are smooth. Occasionally the cleavage is interrupted before it is complete, and walls are built around partially divided masses of protoplasm before they haye rounded off sufficiently to obliterate the furrows. This results in peculiar-shaped spores, such as are shown in PI. V, fig. 26, ¢, and fig. 27. After the spores have been formed the intersporal slime becomes foamy in appearance. (Pl. V, fig. 24.) If the sporangium be allowed to dry out and is then placed in water, this intersporal substance swells considerably and probably aids in breaking the sporangium wall. This wall is made up of two layers from a stage even younger than that shown in PI. IV, fig. 15, though the walls of the mycelium and the sporangiophores show only one layer in stained preparations. Owing to the great shrinkage of the spores in ripening and to the partial collapse of the columella, the sporangium is very much smaller in diameter when mature than at the time the spores are formed. From the time of the cleavage to the ripening of the spores the sporangiophores elongate very rapidly, often reaching a length of 10 em. or more. The ripening usually requires only a few hours. As in Rhizopus, the old mycelium is not entirely empty, but con- tains a very thin layer of protoplasm lying close to the wall, and in this protoplasm are embedded a few nuclei. The columella also in the ripe sporangium contains a loose network of protoplasm with scattered nuclei. The nuclei, while the mycelium is young, have essentially the same structure as those in the spores, except that they often have as many as three nucleoli. As the mycelium grows older, however, they disintegrate like those of R//zopus. 28 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. The writer has also studied the cutting out of the columella and the spores in Pilobolus crystallinus and Sporodinia grandis, but, as his investigations agree with Harper’s account (1899), he has simply given a fuller review of his work than he should otherwise have done, and shall treat these two genera in his general discussion. GENERAL CONSIDERATIONS. From a consideration of the preceding pages, we find that the processes by which the spores are formed in Ri/zopus and Phycomyces appear very different, and that both these forms differ from P//obolus and Sporodinia, which two are different from each other. In other words, of the four genera of the Mucorinex that have been most care- fully studied no two are alike. In Phycomyces the spore-plasm is divided into spores by vacuoles alone.“ The furrows cutting outward from the columella cleft form no exception to this statement, this cleft being simply a fused system of vacuoles. In /Joholus both vacuoles and surface furrows take part in the process. In RA/zopus and Sporodinia the work is done entirely by surface furrows and furrows from the columella cleft, the vacuoles in the spore plasm playing no part in the process. Sporodinia, how- ever, differs from all the other forms in having none of the denser plasm included inside the columella, and from /%/obo/us and Rhizopus in having no surface furrow to assist the vacuoles in cutting out the columella. /%/obolus differs also from the other three forms in that the spores in this genus only are cut down toa uninucleated stage, fol- lowed by an embryonic development consisting of nuclear and cell division. There are some respects, however, in which all four genera agree. In all cases the protoplasm is divided progressively, the nuclei during the cleavage are ina resting state, and a// the protoplasm in the spo- rangium outside the columella is included within the spore walls, the substance between the spores not being protoplasm but a slimy mate- rial excreted by it through osmotic membranes. Harper (1899) has pointed out that this is not a process of free cell formation in the sense that the cells are cut out entirely within a larger mass of protoplasm, as in the ascus of Lachnea and Erysiphe, but is an entirely different type of cell division. He uses this as evidence against the homology of the sporangium of the Zygomycetes with the ascus. Juel (1902), however, in a very recent paper on Ziphridium (a new genus of the Protomycetes) seems to have entirely missed this part of Harper's distinction. He refers to the action of the kinoplasmic rays as being Harper’s whole distinguishing characteristic of free cell forma- tion, and considers this insufficient grounds for such a distinction. He «By this is meant not that the vacuoles are the sole and active agents of division, but that they are not assisted by surface furrows. See note to page 31. GENERAL CONSIDERATIONS. 29 calls the division of the protoplasm in Zaphridium free spore forma- tion, though he confesses that he does not understand the stage in which the spores are being cut out, and gives us no conciusive evidence that the substance between the spores is protoplasm. Timberlake (1902) has described a process of cleavage in the forma- tion of the swarm-spores of //ydrodictyon urtriculatum similar to that which I have described. In this alga the protoplasm forms a layer of an even thickness around a central vacuole, and this protoplasm is divided into a single layer of spores by narrow furrows cutting from the central vacuole outward and meeting similar furrows from the surfaces. The mechanics of this process of division present a very perplexing problem. Sections like those shown in PI. IT, fig. 8, and Pl. HI, fig. 10, where cleavage is only partly complete, havean appearance that suggests the effect of cracking on the surface due to drying. If acolloidal sphere were allowed to dry by evaporation from its surface, it would crack and split ina manner much like the sporangia of //zopus. That such an explanation is not adequate for this cleavage phenomenon is clearly evident from the fact that the furrows are filled with cell sap in living specimens throughout the entire process of division. One can scarcely imagine any body cracking from drying out when the crevices are filled with a watery liquid. In any case such an explana- tion would not account for cleavage by angles cutting out from yvacu- oles embedded in the protoplasm. An explanation that would in a measure account for the angles being pushed out from the vacuoles is that the vacuoles take up water from the surrounding cytoplasm by osmosis through their membranes, which would cause an outward pressure against the latter. If now cer- tain parts of this membrane should become weaker than other parts, these weaker parts would be pushed out by the internal pressure. Such an explanation, however, would not account for the surface fur- rows, as they are not surrounded on all sides by an osmotic membrane, there being no membrane across the mouth of the furrow at the periph- ery of the sporangium. (PI. II, figs. 8 and 9.) If such a membrane be present, it is so thin that it is not visible with the highest powers of the microscope, and hence it is doubtful whether it would be more resistant to outward pressure from within the furrow than the plasma- membrane and cytoplasm at the inner edge of the furrow. That the plasma-membrane and vacuolar membrane should possess sufficient, rigidity to cut into the protoplasm after the fashion of a knife is entirely foreign to our conception of these membranes. The most probable explanation the writer has found for the mechanies of the cleavage is on the basis of local contractions of the cytoplasm, somewhat comparable to the phenomena exhibited in the naked proto- plasm of amoebae and pseudopodia. In Pl. VI, figs. 28-31, the writer 30 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. has attempted to demonstrate diagrammatically the way in which such localized contractions would cut up the protoplasm in exactly the same manner actually occurring in these sporangia. The type of cleayage represented in P//obolus has been chosen so that the same diagrams may be used to explain yacuolar and surface-furrow cleavage. For the sake of clearness the diagrams were made much simpler than the actual sporangia, but without changing any essential fact of strue- ture. The lines of force caused by the contraction of the cytoplasm haye been represented by arrows—red indicating a locality in maxi- mum contraction; green, a locality that has not yet reached its maximum; and blue, a locality that has passed its maximum. Where there are wide spaces between arrows there is assumed to be little or no contraction. Dotted black lines represent planes where cleavage will take place. For the cutting out of the columella, let us assume that after the system of vacuoles shown in Pl. VI, fig. 28, is formed, the cytoplasm at such points between but close to the vacuoles, as shown by the red arrows, begins to contract in a direction at right angles to the future columella cleft, the spore-plasm pulling toward the periphery and the columella-plasm toward the center of the sporangium. This would tend to pull the cytoplasm away from the points at the rear ends of the arrows and also to draw the general masses of the spore-plasm and the columella-plasm toward each other. This would cause pres- sure against the sides of the vacuoles and cause them to flatten out to fill the spaces from which cytoplasm is being withdrawn, as is shown in Pl. VI, fig. 29. At the base of the sporangium where the surface furrow is to cut in, the cytoplasm contracts in a direction approxi- mately radial to the curve in which the furrow is to cut. This pull- ing causes a rift beginning at the surface of the sporangium, and the viscid plasma-membrane, ever adhering to the surface of the cyto- plasm, folds in to line this rift. As the furrow cuts inward, the points of greatest contraction moye inward also (PI. V1, fig. 29), keep- ing always close in front of the furrow until the latter fuses with the lowest vacuoles in the system. The principle involved in the cleavage of the spore-plasm is essen- tially the same. At the points indicated by red arrows in Pl. VI, fig. 80, viz, on the periphery, on the columella cleft, and on the yacu- oles, the cytoplasm begins to contract in a direction approximately toward the centers of the masses of protoplasm that are to become the spores. ‘This pulling away of the protoplasm causes rifts or furrows running into the spore-plasm from the periphery, the columeila cleft, and the yacuoles, as shown in Pl. VI, fig. 81. The width of the fur- rows depends on the continuation of the contraction after the furrows have progressed beyond the points of contraction, i. e., on the amount of contraction that takes place at the points marked in the diagrams GENERAL CONSIDERATIONS. 31 by green-colored arrows. If the pulling at the sides of the furrows continues, as in Pilobolus, the furrows are wide, but if it soon ceases, as in Synchitrium, they are narrow. It is not improbable that in the last stages of cleavage, where the spores are connected by only a slender neck, constriction like that which cuts off conidia may play a part in finishing the process. There is little evidence that the nuclei directly influence the con- traction. The direction of the contraction seems to be in general toward the center of the protoplasmic masses that are to be the spores, without regard to the distribution of the nuclei. The nuclei do, how- ever, seem to determine to some extent just what protoplasm shall constitute each individual spore; otherwise we might have spores formed of enucleated pieces of protoplasm, and none such has ever been observed in these forms. Viewing the cleavage from the basis of localized contraction of the cytoplasm, we do not find such radical differences in the processes involved in Pilobolus, Sporodinia, Rhizopus, and Phycomyces as appeared at first sight. In Piloholus and Phycomye s there are large racuoles in the spore-plasm, in the vicinity of which cytoplasmic con- tractions take place in such a way as to cause angles to cut outward from the vacuoles, while in Sporedinia and Rhizopus such is not the case. On the other hand, there are no cytoplasmic contractions on the periphery of the sporangium in /Aycomyces as in the other three gen- era. Otherwise these four genera exhibit no essential differences in the manner of formation of the columella and spores. The difference is simply in the location of the cytoplasmic contractions. The explanation offered for the mechanics of the cleavage in the sporangia of the Mucorinex seems equally applicable to other cases of surface cleavage, e. g., Synchitrium, Fuligo, and some animal eggs. To illustrate this extended application of the theory the writer has made diagrams of Synchitrium, Fuligo, and the egg of the squid, indi- cating, by means of arrows, as in Pl. VI, figs. 28-31, the location, direc- tion, and duration of the cytoplasmic contractions that would produce such furrows as have been observed in these forms. Pl. VI, figs. 32 and 33, are based on Harper’s (1899 and 1900) figures, and Pl. VI, tig. 34, on Watasé’s (1890) figure. If this view of the mechanics of cleav- age be the correct one, we must regard the vacuoles as passive rather than active agents in cutting the protoplasm.” They have, however, avery detinite and important mission to perform. In all four genera under discussion they form the greater part of the plasma-membrane for the columella and for the surface of the spore-plasm next to the columella, and in Pilobolus and Phycomyces they form the greater part of the plasma-membrane for the spores. As I have already stated, this is done by the vacuolar membrane becoming directly a See note at the bottom of page 2s. 32. FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. part of the plasma-membrane without any visible change except in form. The protoplasmic surface that abutted against the vacuole is the same that is later in contact with the cell sap in the clefts. The boundary of the vacuole has become directly the boundary of a part of the cleft. We have good reason, therefore, to believe that the vacuolar membrane is identical with, or at least very similar to, the plasma-membrane, and may serve the same purpose if opportunity is offered. This homology is further substantiated by the fact that the columella wall is laid down in the dome-shaped vacuolar cleft by the plasma-membranes, formed for the most part by the vacuolar membranes, and, in the case of Phycomyces and Pilobolus, the walls of most of the spores are formed by what was once a number of vacuolar membranes. If, with Strasburger (1898), we regard the plasma-membrane as kinoplasmic, we find here very strong reasons for believing that the vacuolar membrane is of a kinoplasmie nature also. The vacuoles are, then, openings in the protoplasmic mass, less resistant to the contraction of the cytoplasm, and from which clefts may originate. In the higher plants and in the ascus of the Ascomy- cetes we have the new plasma-membrane of the daughter cells formed by the kinoplasmic fibers. In most animal cells and in many of the alew, as Cladophora, and in the formation of conidia in fungi, the new plasma-membrane originates from the old by following the con- striction furrow from the surface inward. In Phycomyces there are neither spindle fibers nor surface furrows present during spore forma- tion, and the kinoplasm which forms the plasma-membranes for the spores seems to be located entirely in the vacuolar membrane. The behavior of the vacuoles in the sporangia of /%/obolus, Sporo- dinia, Rhizopus, and Phycomyces is of considerable interest in its bearing on the question of whether or not the vacuole can be consid- ered asa permanent organ of the cell. Though, as already suggested, the vacuoles are probably not active agents in the division of the pro- toplasm, yet there can be no doubt that they do have a part to play in the process by offering places of slight resistance to the contractions of the eytoclasm, and by supplying material for the formation of new plasma-membranes around the spores and the columella. In the eut- ting out of the columella it is evident that the vacuoles are arranged in their definite dome-shaped system for the distinet purpose of being where they can best do their part in the process. In /’hycomyces the arly formation of the stainable substance in some vacuoles, while others remain empty, and the fact that the former go to form plasma- membranes for the spores and the columella, while the latter do not, indicate that certain vacuoles are predestined from a very young condition of the sporangium to take part in columella and spore formation. GENERAL CONSIDERATIONS. 33 The idea that the vacuolar membrane has special properties not possessed by the general body of the cytoplasm is by no means a new one. De Vries (1885) has shown, by treating living celis with plasmo- lyzing agents containing coloring matter, that the vacuole wall is an osmotic membrane like the hautschicht. He has also been able to isolate the vacuoles from the cytoplasm without breaking them, show- ing the wall to have some strength and elasticity, and that it retains its identity even when not surrounded by a viscid cytoplasm. The vacuoles of Spirogyra were often seen to divide by constriction when treated with a saltpeter solution. By long immersion in a saltpeter solution followed by eosin the vacuole wall was hardened, so that it would be broken by pressure without collapsing. De Vries concludes that there is a very strong similarity between vacuole wall and hautschicht. Went (1888) holds that all living plant cells, with the possible exception of bacteria, Cyanophycer, and spermatozoids, contain vacuoles, which by division furnish all the vacuoles for the succeed- ing generations of cells. In Aspergillus oryze he saw both division and fusion of vacuoles. Ina cell of Dematium pullulans he observed nine vacuoles fuse into two large ones. These then fused to form one; but before the constriction left at the point of fusion had disappeared another constriction had begun to form in another part of the same vacuole, which increased in depth until it had cut the vacuole in two ‘again. Went expresses his belief that the wall of the vacuole plays an active part in this division. In Cladosporium herbarum and in the hairs on the epidermis of Cucurbitu pepo he found that the vacuoles divide just before cell division. Went concludes that the vacuole wall is an organ of the protoplasm ranking with the nucleus and the chromatophores, originating hy the division of a previously existing vacuole, and never forming de noyo in the protoplasm. Bokorny (1898) treated living cells with a weak caffein solution and found that the vacuole wall was not killed by it, but that it contracted without losing its rounded outline, precisely as De Vries describes for vacuoles when the cell is treated with a 10 per cent saltpeter solution. Bokorny points out that, as a dilute caffein solution has but very weak plasmolyzing power, the phenomenon in this case is one of irritability, the caffein solution being the stimulus and the vacuole wall being the receptive part of the cell. A caffein solution as weak as 0.01 per cent will cause the reaction. The work of these authors offers very strong evidence that the vac- uolar membrane is at least a differentiated and specialized portion of the protoplasm, differing molecularly from ordinary cytoplasm, and having many properties in common with the plasma-membrane. 20844—No. 387—03——3 34. FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. Pfeffer (1890) contirms the conclusions of De Vries and Went that the vacuole wall is an osmotic membrane yery much like the haut- schicht, and that it reproduces itself by division. He is able to form new vacuoles in plasmodia, however, by introducing very small parti- cles of asparagin, and finds these to agree in all essential particulars with normally produced yacuoles. He also holds that by extensive vacuolization nearly all of the cytoplasm may be changed to ** plasma- haut” (vacuole wall and hautschicht). Pfeffer, seems, however, inclined to regard both the vacuole wall and the hautschicht as the result of surface tension, and a precipitation of the surface of the cytoplasm by contact with water. Biitschli (1892) also refuses to accept the view that the vacuole is bounded by a definite, permanent wall, and would, with Pfeffer, refer it to surface tension between the watery liquid of the vacuole and the viscid, semiliquid protoplasm; or, at most, he regards this boundary as only a precipitation membrane formed by the action of the yacuo- lar contents on the adjacent protoplasm. In the part played by the vacuoles in the formation of the spores in the Mucorine we have additional evidence that the vacuolar membrane is a more definite structure than Biitschli regards it. The vacuolar membrane so clearly is able at times to perform the functions of the plasma-membrane that the structure and composition of the two seem identical. The exact composition of the stainable substance in the yacuoles of Phycomyces is not easy to determine. In living sporangia crushed under a coyer-glass it is easy to see the vacuoles more or less isolated from the cytoplasm, but no contents can be seen at any stage of development. Neither can the intersporal substance be seen in spo- rangia where cleavage is only partially complete. In older sporangia, however, this is clearly visible. This would suggest that this sub- stance is in solution or yery transparent in living sporangia and is precipitated in the process of dehydrating or clearing—probably by the alcohol. It appears in sections of old sporangia eyen before stain- ing, no matter what fixing fluid is used. It is not readily soluble in water, as may be seen from the fact that it is visible in sections that have been soaked several hours in water. In seeds, Wakker (1888) found that the aleurone grains inside the vacuoles begin as very minute, dense bodies, much smaller than the vacuoles themselves, afterwards increasing in size till the vacuoles are nearly or quite filled by them. As TI have already pointed out, how- ever, the contents of the yacuoles of Phycomyces are at the moment they first become visible quite as large in proportion to the size of the vacuole as when they become older. The substance seems to be evenly distributed in the cell sap of the vacuole, simply increasing in density as the sporangium grows older. ‘There can be little doubt that it is a GENERAL CONSIDERATIONS. Bd secretion of the cytoplasm through the vacuolar membrane, and the fact that the substance is secreted only in those vacuoles which are to take part in the cleavage seems to indicate a difference, in function at least, between the membranes of the two kinds of vacuoles. The clear zone between the body inside the vacuole and the vacuolar membrane seems to be due to the contraction of the substance in dehydration. The fact that this substance takes so readily the shape of the vacuole or the furrow that contains it would show that in the living state it is not solid, but very plastic, if not in actual solution. Stevens (1899) describes a gelatinous, stainable substance in the yacu- oles of the oogonium of A/bugo b/iti, and seems to regard it as being used to form the walls of the oospore. He says of it: ** It appears to be trans- ferred directly from the vacuoles to the exterior of the protoplasm, there to be changed to true cellulose.” Whether or not this substance is the same as that in Pycomyces I can not be certain. Stevens’s description agrees very well with my own in that the substance takes the stain only slightly when first formed, and stronger in later stages. In Albugo, however, it leaves the vacuoles and goes to form cellulose walls, while in Phycomyces it never disappears, but forms the inter- sporal substance in the clefts made by the yacuoles and apparently plays no part in the formation of walls. Stevens describes this sub- stance as occurring in figs. 91, 92, 93, and 94 of his Pl. XV, but I have been unable to find any representation of it in the places referred to. Neither does he describe the method by which it is transferred to the periphery of the oospore. Trow (1901) also has figured a similar content in the vacuoles of Pythium ultimum, but does not describe it so as to give any idea of its true nature. A problem that has been most perplexing to me is how the proto- plasm in the sporangium comes to be differentiated into a very dense layer at the periphery containing many nuclei, and a very loose struc- ture in the interior with few nuclei. The purpose of such a differen- tiation is very evident, viz: That as much of the protoplasm as possible may be included within the spores, but just what propels the protoplasm up the sporangiophore and out toward the periphery of the sporangium is not so easy to determine. Arthur's (1897) explana- tion that it is due to evaporation of water from the surface, combined with absorption of moisture from the substratum, seems entirely inade- quate. If this were the cause, we should expect to find the layer of denser protoplasm at the base of the sporangium as well as on the sides and top, as we have no evidence that evaporation does not go on from the part of the sporangium just around the sporangiophore as well as from the rest of the surface. Furthermore, from Arthur's explanation we should expect a gradual transition between the denser 36 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. and the less dense protoplasms, but, as has been already pointed out, the transition is quite sudden. Still further, the layer of dense pro- toplasm is not of the same absolute thickness for all sporangia, nor does it bear a constant relation to the size of the sporangiuy. The writer has been rather inclined to regard the thickness of this layer as dependent on the amount of available protoplasm, though by no means certain on this point. Arthur’s conclusions bear more specifically on the streaming in the hyphz than in the sporangium, and yet he gives us no intimation that he wishes to separate the two processes and refer them to different causes. In the formation of the oosphere in some of the Peronosporexe we have, according to Wager (1896), Stevens (1899), and others, a differ- entiation of the protoplasm into ooplasm and periplasm, but this differentiation is not characterized by such a marked difference in the density of the two protoplasms as in the Mucorinez. The wall about the oosphere is described as forming on the boundary between two protoplasms. The question as to just how the protoplasm is divided and the wall formed has been pretty carefully avoided by all these authors. Stevens states that there is a thin film formed between the two protoplasms, and that this film seems to develop into the wall of the oospore, but his account of the process is very incomplete. Trow (1901) figures a stage in Pythium ultimum, in which the oosphere is only partially cut out, but, unfortunately, he does not describe it suffi- ciently to give us a clear conception of the real nature of the process. The fact that the columella cleft forms just inside the denser plasm, rather than between it and the looser plasm, accords well with the idea that the cleft is formed by cytoplasmic contractions. The layer of denser plasm inside the columella cleft seems to be for the specific purpose of aiding in the cleavage by its contraction, a function that the looser plasm is probably unable to perform. SUMMARY. The essential processes inthe formation of the spores in the sporan- gia of Rhizopus and Phycomyces may be summarized as follows: 1. Streaming of the cytoplasm nuclei and vacuoles up the sporangi- ophore and out toward the periphery, forming a dense layer next the sporangium wall anda less dense region in the interior, both containing nuclei. 2. Formation of a layer of comparatively large, round vacuoles in the denser plasm parallel to its inner surface. 3. Extension of these vacuoles by flattening so that they fuse to form a curved cleft inthe denser plasm; and, in the case of A‘zopus, the cutting upward of a circular surface furrow from the base of the sporangium to meet the cleft formed by these vacuoles, thus cleaving out the columella. SUMMARY. 37 4. Division of the spore-plasm into spores; in P//zopus, by fur- rows pushing progressively inward from the surface and outward from the columella cleft, both systems branching, curving, and intersecting to form multinucleated bits of protoplasm, surrounded only by plasma- membranes and separated by spaces filled with cell sap only; in Phy- comyces, by angles forming in certain yacuoles containing a stainable substance and continuing outward into the spore-plasm as furrows, aided by other furrows from the columella cleft and dividing the protoplasm into bits homologous with and similar to those in 2//z0pus, and separated by furrows partly filled with the contents of the vacu- oles that assist in the cleavage. 5. Formation of walls about the spores and columella, and, in the case of RAizopus, the secretion of an intersporal slime. 6. Partial disintegration of the nuclei in the columella. 38 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. INDEX TO LITERATURE. Artur, J. C. (1897): The Movement of Protoplasm in Ccenocytic Hyphee. Ann, of Bot. 11, 1897, p. 491. BacuMmann, H. (1899): Mortierella van Tieghemi. Beitrag zur Physiologie der Pilze. Jahrb. fiir wiss. Bot. xxxiv, 1899, p. 279. Boxorny, Tu. (1893): Die Vakuolenwand der Pflanzenzellen. Biol. Centr. 13, 1893, p- 271. : Biscen, M. (1882): Die Entwicklung der Phycomycetensporangien. Jahrb. flr wiss. Bot. XIII, 1882, p. 253. Birscuit, O. (1892): Untersuchungen iiber mikroskopische Schiiume und das Proto- plasma, 1892, p. 145 et al. Corpa, A. C. J. (1838): Iecones Fungorum, II, 1838, p. 19. De Vates, H. (1885): Plasmolytische Studien tiber die Wand der Vacuolen. Jahrb. fiir wiss. Bot., XVI, 1885, p. 465. Fiscuer, A. (1892): Phycomycetes. Rabenhorst’s Kryptogamen-flora. Band I, Abt. IV, 1892, p. 161 et al. Harper, R. A. (1899): Cell-Division in Sporangia and Asci. Ann. of Bot., 13, 1899, p. 467. — (1900): Cell and Nuclear Division in Fuligo varians. Bot. Gaz., 30, 1900, p. 217. Jve., H. O. (1902): Taphridium. Bihang-till K. Sv. Vet. Akad. Handl, 1902. Lécer, M. (1896): Recherches histol. sur la structure des Mucorinées, 1896. Prerrer, W. (1890): Zur Kenntniss der Plasmahaut und der Vacuolen. Abh. d. k. siichs. Ges. d. Wiss., XVI, 1890, p. 187. Srevens, F. L. (1899): The Compound Oosphere of Albugo Bliti. Bot. Gaz., 28, 1899, p- 149. SrraspurGer, E. (1880): Zellbildung und Zelltheilung, 3d ed., 1880. (1898): Die pflanzlichen Zellhiiute. Jahrb. fiir wiss. Bot., XXNI, 1898, p. 511. Trow, A. H. (1901): Observations on the Biology and Cytology of Pythium ultimum. Ann. of Bot., 15, 1901, p. 269. Tuaxrer, R. (1897): New or Peculiar Zygomycetes. 2. Syncephalastrum and Syn- cephalis. Bot. Gaz., 24, 1897, p. 1. TimperLake, H. G. (1902): Development and Structure of the Swarm-spores of Hydrodictyon. Trans. Wis. Acad. Sei., XTIT, 1902, p. 486. Van Trecuem, Ph., et G. Le Monnrer (1873): Recherches surles Mucorinées. Ann. d. se. nat., 5° sér., Bot., 17, 1873, p. 261. Van Trecuem, Ph. (1875): Nouvelles Recherches sur les Mucorinées. Ann. d. se. nat., 6° sér., Bot., 1, 1875, p. 5. (1876): Troisitme Mémoire sur les:Mucorinées. Ann. d. se. nat., 6° sér., Bot., 4, 1876, p. 312. Wacer, H. (1896): On the Structure and Reproduction of Cystopus candidus. Ann, of Bot., 10, 1896, p. BOD, Wakker, J. H. (1888): Studien iiber die Inhaltskérper der Pflanzenzelle. Jahrb. fiir wiss. Bot., XIX, 1888, p. 425. Warasé, 8. (1890): Studies on Cephalopods: I. Cleavage of the Ovum. Jour. of Morph., IV, 1890, p. 247. Went, F. A. F. C. (1888): Der Vermehrung der normalen Vacuolen durch Theilung. Jahrb. fiir wiss. Bot., XIX, 1888, p. 295. DESCRIPTION OF PLATES. [All the figures were drawn with the aid of a Leitz or a Zeiss camera lucida, with objectives and oculars, as follows: Fig. 1, Leitz No. 1 objective, No. 0 ocular: figs. 5, 6, 7, 8, 10, and 12, Leitz 44; oil-immersion objective, No. 0 ocular; fig. 14, Leitz oil-immersion objective, No. 3 ocular; figs. 15, 17, 18, and 19, Zeiss 2 mm. 1.30 aper- ture, oil-immersion objective, No. 1 Huyghenian ocular; figs. 2, 3, 21, 24, 26, and 27, Zeiss 21m. 1.30 aperture, oil-immersion objective, No. 6 compensating ocular; figs. 22, 23, and 25, Zeiss 2 mm. 1.30 aperture, oil-immersion objective, No. 12 compen- sating ocular; figs. 4, 9, 11, 13, 16, and 20, Zeiss 2 mm. 1.30 aperture, oil-immersion objective, No. 18 compensating ocular. } Prare I. Rhizopus nigricans. Fig. 1.—Group of sporangiophores bearing sporangia, showing how they grow out from the stolon. % 12. Fig. 2.—Longitudi- nal section of young stolon, showing distribution of cytoplasm and nuclei. X 750. Fig. 3.—Same, except that the stolon is much older; wall very thick, and nuclei disintegrating. ™ 750. Fig. 4.—Disinte- grating. nuclei from stolon shown in fig. 3.» 2,250. Fig. 5.—Young sporangium, showing cytoplasm and nuclei streaming up the sporan- giophore into the sporangium and out toward the periphery. There are a few crystalloids in the center. 520. Fig. 6.—Sporangium that has attained nearly its full size. The differentiation between the looser and the denser plasms is sharply marked, except at a few places. Just inside the denser plasm is a clear zone of protoplasm that does not take the orange stain, and through this run strands of orange-staining cytoplasm bearing nuclei. > 520. Il. Rhizopus nigricans. Fig. 7.—Full-sized sporangium, showing layer of vac- uoles nearly formed in the denser plasm. The two layers of the wall are here shown. X 520. Fig. 8.—Section cut through sporangium a little to one side of the sporangiophore. The columella cleft is being formed by fusion of the vacuoles shown in fig. 7, and by a surface furrow. The spores are also being cut out by progressive surface furrows. ™ 520. Fig. 9.—A small part of the same sporangium as shown in fig. 8, drawn from another section, showing in detail very early cleavage furrows, and structure, size, and distribution of nuclei and vacuoles. > 2,250. MI. Rhizopus nigricans. Fig. 10.—Cleayage much farther advanced than in figs. Sand 9. Furrows cutting outward from the columella cleft. Sec- tion not cut through sporangiophore. » 520. Fig. 11.—Nuclei from columella of same; a, very close to columella cleft; }, c, and d, nearer the center; a has a normal structure, while }, ¢, and d show stages in disin- tegration. X 2,250. Fig. 12.—Sporangium in which the spores are completely formed, rounded up, and surrounded by thin walls. The columella wall is also formed. 520. Fig. 13.—Nuclei from columella of same; a lies near columella wall and still retains its normal structure; b lies near it but is beginning to disintegrate; ¢ and d lie near the center and are reduced to homogeneous angular masses. % 2,250. Fig. 14.— Ripe spores in their living condition, showing variations in size and ridges on walls. x 950. IV. Phycomyces nitens. Fig. 15.—Young sporangium, showing cytoplasm nuclei and vacuoles streaming up the sporangiophore and out toward the periph- ery of the sporangium. Vacuoles in the denser protoplasm have a visi- ble content. 550. Fig. 16.—Small part of young sporangium very highly magnified, showing early stage in the formation of the visible 2 oy 40 FORMATION OF SPORES OF RHIZOPUS AND PHYCOMYCES. content of the vacuole; also two much smaller vacuoles with no such contents; nuclei in resting condition. > 2,250. Fig. 17.—Portion of cross section of sporangium at a somewhat-later stage than fig. 15, show- ing distribution of protoplasm into an outer dense layer, an interior region of very loose protoplasm containing empty vacuoles and no nuclei, and between these two a layer of intermediate density. 550. Fig. 18.—Part of longitudinal section of sporangium, showing layer of vacu- oles forming in denser protoplasm where the columella is to be cut out. < 550. Prate V. Phycomyces nitens. Fig. 19.—Layer of vacuoles in the denser plasm, flatten- ing out toward each other to form the columella cleft by their fusion. The contents flatten out also, taking the shape of the vacuoles. > 550. Fig. 20.—Small part of section very highly magnified, showing three vacuoles in contact, separated only by their membranes; also three very small empty vacuoles and six nuclei in resting condition. 2,250. Fig. 21.— Spore plasm being cut up into spores by vacuoles becoming angular, and the angles cutting through the protoplasm as furrows; cytoplasm in front of furrows undifferentiated; nuclei in a resting condition. The contents of the vacuoles extend out into the furrows and fuse as the furrows fuse, to form the intersporal substance. 750. Fig. 22.—Furrows cutting outward into the spore plasm from the columella cleft; eytc plasm in front of furrows undifferentiated. 1,500. Fig. 23.—Furrows from the vacuoles cutting out to the plasma-membrane at the periphery of the sporangium. 1,500. Fig. 24.—Nearly ripe spores containing resting nuclei and empty vacuoles, and embedded in intersporal slime. 750. Fig. 25.—Living, ripe spores; walls smooth; a, very large; b, average size; c, very small. » 1,500. Fig. 26.—Very large peculiar-shaped spores; e, probably due to arrested cleavage. 750. Fig. 27.—Very large, irregular-shaped spore due to arrested cleavage. > 750. VI. Pilobolus erystallinus. (Diagrammatic and much simplified.) Fig. 28.—One- half of longitudinal section of sporangium just before the cutting out of the columella. The arrows indicate lines of contraction of the cytoplasm to form the columella cleft. Green arrows indicate points where the con- traction is just beginning and red arrows points where the contraction is at its maximum strength; dotted black lines represent planes where cleavage is to take place. Fig. 29.—Same, but somewhat older stage; vacuoles flattened to fill the spaces where the cytoplasm has been pulled away; also surface furrow at the base of the sporangium. Blue arrows indicate points where contraction has passed its maximum strength. Fig. 30.— Columella cleft completed, spore formation just ready to begin. Fig. 31.— Vacuoles in the spore-plasm becoming angular, and furrows cutting inward from the periphery and outward from the columella cleft, due to the cytoplasm pulling away at these points. Fig. 382.—Synchitrium decip- iens. (After Harper.) Two cleayage furrows cutting into the sporan- gium. These are slightly open at the inner extremity where the cyto- plasm is contracting, but closed nearer the periphery of the sporangium where contraction has ceased. Fig. 33.—Fuligovarians. (After Harper. ) Two furrows cutting into the spore plasm; furrows slightly open through- out their entire extent. Fig. 34.—Squid. (After Watasé.) Surface view of egg, showing cleavage furrows cutting into the cytoplasm between the nuclei; furrows very narrow at the extremities. O kha OO xy ON OF THESPORES IN RHIZOPUS NIGRICANS AND PHYCOMYCES NITENS A J QM Bul. 37, Bureau of Plant Industry,U.S.Dept of Agriculture. Bul.37, Bureau of Plant Industry,U.S.Dept. of Agriculture PLATE I FORMATION OF THESPORES IN RHIZOPUS NIGRICANS AND PHYCOMYCES NITENS Bul.37, Bureau of Plant Industry,U.S. Dept.of Agriculture PLATE Ill ey) Tn Ae are Ag eA ele h eos ¥ iy FORMATION OF THESPORES IN RHIZOPUS NIGRICANS AND PHYCOMYCES NITENS v Bul.37, Bureau of Plant Industry,U.S. Dept. of Agriculture. PLATE IV. AS Bel JULES BEN & CONS FORMATION OF THESPORES IN RHIZOPUS NIGRICANS AND PHYCOMYCES NITENS Bul. 37, Bureau of Plant Industry, U.S. Dept. of Agriculture PLATE V. Bul. 37, Bureau of Plant Industry,U.S.Dept of Agriculture PLATE VI Cas oe LL , ee) FORMATION OF THESPORES IN RHIZOPUS NIGRICANS AND PHYCOMYCES NITEN = a _ i. 4 . 7 ie } : « ion “ i } ~ A a " - - . os = » > - FA ‘ . t = : 7 ? s , : : 7 - Tye U Friant 19 Industry, Soils, and Agri- A3 cultural Engineering no. Bulletin 3137 —~S/ PLEASE DO NOT REMOVE SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY Badr nen — Cate Boo oe. ethereal en pau sve Nate beanie netiens arc . tay woe hyn yep rene wena te ene oma Soda eves nals ryonay ane heagrots opmakdyee ota receapesh ees tea