Hull ALU Che Mural CertBook Series Epirep sy L. H. BAILEY Bailey: Scuoot-Boox or Farmina. Carleton: Tor SMALL GRAINS. Chase: First Book or GRASSES. B. M. Duggar: THe PuysioLoay or PLANT PRODUCTION. J. F. Duggar: AGRICULTURE FOR SOUTHERN ScHOOLs. J. F. Duggar: SourHerRN Frecp Crops. Fisk: Tue Book or Icn-Cream. Gay: Breeps or Live-Stock. Gay: PRINCIPLES AND PRACTICE OF JUDGING Live-Srock. Goff: PRINCIPLES OF PLANT CULTURE. Gourley: Text-Book or PoMo.Loey. Guthrie: Taz Book or Burrer. Harper: ANIMAL HUSBANDRY FOR SCHOOLS. Harris and Stewart: Tam PRINCIPLES OF AGRONOMY. Hitchcock: Text-Book oF GRASSES. Jeffery: Text-Boox or LAND DRAINAGE. Jordan: FrrpiIna oF ANIMALS, REVISED. Livingston: Frxtp Crop PropuctIon. Lyon: Sorts AND FERTILIZERS Lyon, Fippin and Buckman: Sos; THER PROPERTIES AND MANAGEMENT. Mann: BrcinninGs IN AGRICULTURE. Montgomery: Toe Corn Crops. Morgan: Firrutp Crops FoR THE CoTTon- BEtr. Mumford: Tan BREEDING oF ANIMALS. Piper: FoRAGE PLANTS AND THEIR CULTURE. Sampson: ErrectivE FARMING. Smith: AGRICULTURAL METEOROLOGY. Stubenrauch, Wood and Booth: HorticuuTURE FOR SCHOOLS. Thom and Fisk: Tar Book or CHEESE. Warren: ELEMENTS OF AGRICULTURE. Warren: Farm MANAGEMENT. Wheeler: MANURE AND FERTILIZERS. White: PrinctpLes oF FLORICULTURE. Widtsoe: PrincrpLes oF IRRIGATION PRAC- TICE. FIRST BOOK OF GRASSES THE STRUCTURE OF GRASSES EXPLAINED FOR BEGINNERS PY AGNES CHASE ASSISTANT AGROSTOLOGIST, UNITED STATES DEPARTMENT OF AGRICULTURE New York THE MACMILLAN COMPANY 1922 ae All rights reserved Copyrricut, 1922 By THE MACMILLAN COMPANY Set up and electrotyped. Published November, 1922. Printed in the United States of America me 4 Rateeuay.> PREFACE Of all plants grasses are the most important to man. The different kinds are known by very few even among botanists. This is largely because they are supposed to be very difficult. When the struc- ture of grasses is clearly understood, they are not more difficult to study than are other plants. The method here offered has been used by the author for some years in teaching special students. The introduction explains the method followed. The body of the primer consists of twelve lessons, graded from the simplest to the most complex. Each lesson is accompanied by figures bringing out the facts in the text. The difference in the size of the flowering organs is so great that the figures are not drawn to the same scale. The organs are enlarged as much as necessary to illustrate the char- acter discussed. Grasses of the United States are used for the lessons and the commoner ones are selected so far as possible; however, our native grasses cover so wide a range that the principles laid down in the lessons are ap- plicable to grasses generally. AGNES CHASE. Washington, D. C., August 10, 1922. ssO gra rstbookof Is/fi org/detai ive arch //\WWW. > http TABLE OF CONTENTS BP MOD NOTION oo fc za cecae ON eben pea EC Hee eRe nd & 1-7 Use of technical terms, 2; Use of Latin names, 4; Tools needed, 6. RIDMNON De sa Sie Sad FLAT ae RL GR OER» Wale Ly combos 8-16 The grass family, 8; Summary, 16; Review, 16. RMA BE 053 cos ag Meas urine case ely Oe RS 6 aR ES eA 17-20 The spikelet and the inflorescence, 17; Summary, 19; Review, 20. CS 1 SEP ne ee ag an aie Ler trp ee oly 21-24 Modifications of the spikelet, 21; Summary, 24. BAO SW coy oak SIE ae) CL REE RUSSEL Ro es 25-33 Pediceled spikelets of few to many florets, 25; Sum- mary, 33; Review, 33. Sessile spikelets in two-sided spikes, 34; Summary, 41; Review, 41. BOSON VE Gy civics oe es cha cain Use rete emee ets 42-45 Pediceled spikelets with large glumes and other modi- fications, 42; Summary, 45; Review, 45. en Gk SOU Tr Sm RCS PR AR fa Pe > 3A dla 46-53 Pediceled one-flowered spikelets, 46; Summary, 52; Review, 53. BONN BES 5 Gio Ou dads ois + oe VOR ee Pees AA 54-60 Sessile spikelets in one-sided spikes, 54; Summary, 60; Review, 60. MOMOUOIET Baia) se weg So 6.005 hin 5-4 pis Vie es Re RR ia at ek 61-67 Diversely specialized spikelets, 61; Summary, 67; Re- view, 67. Vii vill TABLE OF CONTENTS Spikelets with membranaceous glumes and hardened fruits, 68; Summary, 77; Review, 78. EO RL soo5' 5 oo seule ois + 0 cu MRSA Oren Se nee ete aie 79-88 Paired spikelets with hardened glumes and thin lemmas, 79; Summary, 88; Review, 88. PRBS Eg E55. is. ds 5 ccale, ohscn sau batos aa ee ad pn tae ees aca 89-94 Highly specialized unisexual spikelets, 89; Summary, 94; Review, 94. DIAGRAMMATIC SUMMARY OF THE PRIMARY CHARACTERS OG THE ERIBWRY ovules, which when fer- tilized develop into the seeds. The style serves to lift the stigma into the air. The stigma is more or less expanded, has a 7 viscid surface to which Fic. 3. Grass flower, showing sta- the pollen-grains adhere _ mene and pil and the rudl- and upon which they ger- minate, sending their con- tents in a minute tube which pushes down through the style to the ovules, fertilizing them. The grass flower (Fig. 3) is reduced to the essential organs, the floral envelopes being represented by the minute lodicules. Each flower is borne in the axil of a small green bract (the lemma) and is subtended and enveloped in a second bract (the palea). The flower with its lemma and palea is termed the floret (Fig. 4). The ripened ovary (the grain, or caryopsis) (Fig. 5) consists of a small embryo lying at the base THE GRASS FAMILY 11 of a mass of starchy endosperm. [Endosperm means within the sperm or seed. It is the store of food used by the infant plantlet when it begins to grow.] The ‘‘germ”’ of a kernel of corn is the embryo, while the remainder of the kernel ox: is starchy endosperm. The grain yj lies in the palea with the hilum _ ,|| (the scar of the point of attach- ‘ ment) toward it, and the embryo _ \\\ on the side toward the lemma. °°” Fig. 5 gives two views of a grain, 1% i Bow Howe one showing the hilum, the other the embryo. In Fig. 81, A (page 91), are two ker- nels of corn showing the embryos. The grain with very few exceptions is permanently in- closed in the lemma and palea, the ma- ture floret being the fruit, that is, the seed with its permanent envelopes. The florets are borne in two ranks and alter- nate upon an axis (the rachilla). Below them are two bracts without flowers (the glumes). The glumes, rachilla, and florets together form the spikelet. Fig. 6 is a diagram of a branch with Fic. 5. Two leaves and flowers arranged as are the aed of 2 glumes, lemmas, paleas, and flowers of a grass spikelet. Fig. 7 is a diagram of a spikelet for comparison with Fig. 6. [The hy- pothetical flower-bearing branchlet is never elongate, 12 FIRST BOOK OF GRASSES as here shown for the sake of comparison. The palea is immediately above the lemma and the flower immediately above the palea.| It will be seen that the spikelet is theoretically a leafy flowering branch with a jointed main axis, the flowers, except Fig. 6. Diagram of a flowering Fic. 7. Diagram of a grass branch. spikelet. for the minute lodicules, reduced to the essential organs. In Fig. 8 a typical grass spikelet is shown, the lemmas’and paleas nearly closed together and concealing the flowers. The spikelet is characteristic of grasses and is not found in any other family except that of the sedges. ES A Sey a THE GRASS FAMILY 13 In the spikelets of sedges the florets are commonly, but not always, spirally arranged, there never is a palea, and the fruit is an achene or nutlet. [The “seeds” of buttercups and mints are achenes or nut- lets. | In grasses specialization takes place mostly in the \\ spikelet. By its vegetative \ \W\ iif characters we knowa given \\\\\\ #////// , plant to be a grass, but it is by its spikelets and their fi ff rcorer arrangement that we know what kind of a grass it is. The genera of grasses and the groups of genera called tribes are based on the structure of the spikelets “com OQ and their arrangement in Wont the inflorescence. ) Before studying the spike- let we must observe the relatively few specializations of the vegetative parts. As in other plants, stems or parts of stems may be underground. These underground stems (rhizomes, or rootstocks) are borne at the base of the main culm under the earth, spread out horizontally, and in due time send up shoots which form young plants at a distance from the parent. Sod-forming grasses have this kind of underground stem. Kentucky blue- PEDICEL Fie. 8. Generalized spikelet. 14 FIRST BOOK OF GRASSES grass (Poa pratensis, Fig. 9) furnishes an excellent example. Sometimes the rhizome is thick and woody, sending up shoots from its nodes, the whole forming a dense colony, as in gama-grass (T'ripsacum dacty- loides). A rhizome, being a stem, is jointed and bears scales, which are reduced leaves. By these it may always be distinguished from a root, which is not jointed, and never bears scales. In some grasses the shoots borne at the base of the culm are on the surface of the earth instead of beneath it. = ———— Such shoots are OSG RCS SS S=__ stolons, or run- eee HAW ners. These, like rhizomes, are Fig. 9. Base of plant and underground parts, jointed and bear roots and rhizomes, of Poa pratensis. scales or, some- times, well-developed leaves. Rhizomes and stolons both bear roots at the under side of the nodes. Ina few species, Bermuda-grass for one, a plant may produce either rhizomes or stolons according to the conditions under which it is growing. There is no real difference between a rhizome and a stolon, the one is below ground and colorless, the other above ground and green. Culms are hollow in most grasses, but in corn, sugar-cane, sorghum, and related grasses they are THE GRASS FAMILY 15 pithy. They may be erect, spreading, or creeping; they may be simple or freely branching. A branch is borne only at a node in the axil of a sheath, that is, between the sheath and the culm. It either grows up parallel with the parent culm until it emerges from the sheath or the young branch splits the sheath and grows outward. In manuals of botany these two methods of branching are called intravaginal, that is, inside the vagina (Latin for sheath), and ex- travaginal, outside the sheath. In bunch-grasses, like orchard-grass and the wheat-grasses of the West, the branching is intravaginal; in Kentucky blue-grass, quack-grass, and others producing rhi- zomes or stolons, the branching is extravaginal. The branches borne at the middle and upper nodes of a culm are nearly always intravaginal. If they spread from the parent culm they do not burst through the sheath but carry it with them. Leaves are always borne at the nodes and are always 2-ranked (see Fig. 1, page 9). In corn and other large grasses the leaves sometimes appear to be all on one side instead of 2-ranked. This is due to a twisting of the culm inside the sheath. Sometimes in large grasses, particularly in sugar- cane and in bamboos, the leaves fall, leaving the culm naked. In relatively few grasses the edges of the sheath are grown together, forming a tube. Some- times the blade of the leaf is not developed. This is always the case in the leaves or scales of rhizomes (Fig. 9), and often in those of stolons and in the 16 FIRST BOOK OF GRASSES lowermost leaves of a culm, especially in bamboos. Blades may be flat or folded (often called con- duplicate) or involute; that is, rolled lengthwise. Rarely the inrolled edges are grown together, so that the blade really has no upper surface. Such leaves are called terete (meaning rounded). Blades vary greatly in size, shape, and texture, smoothness or hairiness. In broad leaves there is sometimes a narrow neck, or petiole (leaf-stalk), between the sheath and the blade. SUMMARY Grasses are distinguished by jointed, round or flattened, usually hollow, culms, with solid nodes; 2-ranked leaves, composed of sheath and blade, with a ligule at their junction; and by the spikelets with 2-ranked glumes and florets. REVIEW (1) How is a grass distinguished from all other plants? (2) Break a cornstalk and note the arrangement of the torn fibers standing out of the pith. Compare this with the cut end of any twig of a tree or shrub. Cut across a wheat, oat, or rye straw and compare with the cornstalk and with the twig. (3) Examine the culm (nodes and internodes) above and below ground and leaves (sheath, ligule, and blade) of any available grass. (4) How is a rhizome distinguished from a root? (5) What is the difference between a rhizome and a stolon? LESSON II THE SPIKELET AND THE INFLORESCENCE THEORETICALLY the spikelet is a reduced leafy branch. In the generalized spikelet shown in Fig. 8 the likeness to a jointed culm with 2-ranked leaves (Fig. 1) is readily seen, the glumes and lemmas corre- sponding to sheaths, their blades not developed. The palea, with two nerves and with its back to the axis, corresponds to a minute bract (the prophyllum) borne at the base of a branch in the axil of a sheath. The prophyllum is always 2-nerved, with its back (that is, the space between the nerves) against the main axis and its margins clasping the young branch. The flower, also, is theoretically an ultimate branch- let. In the flower-bearing lemmas, therefore, the palea is developed, while in the glumes, bearing no flowers, there are no paleas. Glumes and lemmas are, morphologically, reduced leaves, the lower pair, not flower-bearing, being termed glumes, the flower- bearing ones being termed lemmas. (See Fig. 7.) The jointed axis of the spikelet (the rachilla) corresponds to the jointed culm and, like it, usually breaks at the nodes, the internode (the part of the rachilla between two nodes) remaining attached to the floret at its base (Fig. 4), just as in a broken grass stem the internode of the culm remains with 17 18 FIRST BOOK OF GRASSES the sheath that surrounds it; that is, the break normally comes just under the node. Rachis (which means the spine, or backbone) and axis (the imaginary central line of any body) are often used inter- changeably as bo- tanical terms. In most recent works on grasses and in these lessons, axis is used for the main axis of a com- pound inflores- cence, rachis for the axis or sup- port of the spike- lets. In Fig. 10 are shown the axis of a panicle (A) \|4 ae : and the rachis of Fia. 10. Forms of inflorescence: A, panicle; @& Taceme (B); the Ys FOR coe rachis of the spike (C) is concealed by the overlapping spikelets. Spikelets are borne pediceled (that is, on a pedicel or foot stalk) or sessile (without a pedicel) in leafless panicles, racemes, or spikes (Fig. 10). These different types of inflorescence insensibly grade into each other. The axis and branches of a panicle and the rachis of a raceme or spike may be as elaborately modified and specialized as may be the parts of a THE SPIKELET AND THE INFLORESCENCE 19 spikelet. The axis, rachis, or branches may be continuous (not jointed), or articulate (jointed) and usually disarticulating (breaking up). The break- ing up takes place at definite points and has to do with scattering the seed. The point of disarticula- tion is the same in grasses of the same kind (or genus) and is usually alike in related genera, and for this reason is of great importance in the classifica- tion of grasses. When there are no joints in the axis or branches, the disarticulation comes in the spikelet, either above the glumes and between the florets or below the glumes. Modifications of these two ways of disarticulating will be met with later. The study of the form of inflorescence and the modification of its parts will be carried on together with that of the spikelet. SUMMARY A spikelet consists of glumes and florets, in two ranks and alternate on the rachilla, the florets consist- ing of lemma, palea, and the inclosed flower. Every organ found in the most highly specialized spikelet is to be interpreted as an elaboration or a reduction of one of these parts. The spikelet is the unit of the inflorescence; the floret is the unit of the spikelet. The spikelet is always simple; that is, the rachilla never branches. The floret is always 1-flowered with never more than one lemma and one palea; the glumes and florets are always alternate, two consecu- tive ones never being borne one above the other. 20 FIRST BOOK OF GRASSES These basic facts kept in mind will aid in the recogni- tion of complicated or congested spikelets and in the correct interpretation of their parts. REVIEW (1) Examine a spikelet of a brome-grass, such as cheat or chess. Separate the florets from each other and from the glumes. Note that these disjoint without tearing. Open out the lemma and palea. Note that these permanently adhere at their base; that they can only be torn or cut apart. (2) Name the parts of a spikelet. (3) What is their arrangement? (4) What is the difference between a rachis and a rachilla? / LESSON III MODIFICATIONS OF THE SPIKELET Taxinc the generalized spikelet as a beginning, examine Figs. 4 and 8. Note that the glumes and lemmas have nerves or veins (fibro-vascular bundles) running from the base to the apex or nearly to it, one nerve in the middle and an equal number on each side, there being an odd number of nerves in the glumes and lemmas. In the palea there are only two nerves; none in the middle. The modifications and variations of spikelets will be studied, as nearly as possible, in the order of their increasing com- plexity. No attempt should be made to fix in mind the forms of specialization here enumerated. This lesson is only meant to put one on the lookout for modifications, so that one may be prepared to recog- nize a given organ under various guises. (1) Spikelets differ in size. (Bromus or brome- grass, meadow fescue, and blue-grass are examples.) (2) The number of florets may be reduced to one or increased to twenty or more. (3) The parts of the spikelet vary in their relative sizes. The glumes may be large and the florets small or the florets large and the glumes small. (4) The rachilla joints may be slender or thick, so short that the florets appear to be opposite, or nearly 21 22 FIRST BOOK OF GRASSES as long as the florets themselves; the node at the base of the floret (the callus) is sometimes prolonged into a sharp point; the uppermost rachilla joint may bear no floret and may extend into a little bristle. (5) The glumes may vary in shape, in texture, and in the number of their nerves and may be gla- brous (smooth) or pubescent (hairy) ; sometimes they are reduced to rudiments and sometimes they are suppressed. (6) The lemma is subject to such great modifica- tions that we shall now note only the simplest ones. As in the glumes, the shape and texture and the number of nerves vary. The summit of the lemma may be acute (pointed) or acuminate (long-pointed) or obtuse (blunt) or it may be lobed or cleft. The nerves may be faint or strong, or may extend into awns (bristles) beyond the body of the lemma. The lemma is sometimes minute, but it is never sup- pressed. (7) The palea is always 2-nerved, but in a few grasses the nerves are so close together as to appear like a single one. In others they are far apart and the palea may split between them. The palea differs in size and texture. In a few genera it is reduced or even suppressed. (8) Spikelets are compressed (flattened) laterally (sidewise), as in Figs. 11-14, or dorsally (on the back), as in Figs. 60-66. The side or back of a spikelet is recognized from the position of the florets on the rachilla. The palea side of the floret is always MODIFICATIONS OF THE SPIKELET 23 toward the rachilla. Taking this as the: ‘‘front”’ of a floret, the back (or dorsum) is the back of the lemma. Spikelets with many florets, or those in which the glumes or lemmas are folded on the mid- nerve, are generally laterally compressed; those with a single fertile floret and with convex (rounded) glumes and lemma are generally dorsally compressed. (9) Spikelets are sometimes unisexual (of one sex) instead of perfect (having both stamens and pistil). The two kinds, pistillate (the ovule-producing) and staminate (pollen-producing) may be borne on a single plant, as in corn, with pistillate spikelets in the ear and staminate spikelets in the tassel, or in differ- ent plants, as in salt-grass and buffalo-grass. Grasses with staminate and pistillate spikelets on the same plants are moneecious (which means dwelling in one _ house); those with staminate and pistillate spikelets on different plants are dicecious (dwelling in two houses). The unisexual spikelets of a single species may be similar in appearance, or they may be very unlike, as in corn. (10) Some grasses with perfect spikelets bear in addition staminate spikelets or neuter spikelets (hav- ing neither stamens nor pistils). These additional spikelets may be similar in appearance to the perfect ones or very unlike them. They are borne in the same inflorescence as the perfect spikelets and are usually paired with them. (11) The florets of a single spikelet may be of two kinds, perfect and sterile, the sterile being either 24 FIRST BOOK OF GRASSES staminate or neuter. In spikelets like those of Figs. 11-14, with several to many florets, the uppermost florets are commonly sterile (not perfecting seed), though they are like the fertile florets in appearance and have rudimentary stamens and pistils. In some grasses the sterile florets are very different from the fertile ones and may be borne above or below them on the rachilla. The position of sterile florets in the spikelet is the same in large series of related grasses; hence it is of great importance in classifying genera. The type of modification of the sterile floret is also uniform within a genus, and generally in related genera. It commonly consists of a lemma without a palea, but there may be a rudimentary or even a well-developed palea. These types of modification will be brought out and illustrated in the succeeding lessons. SUMMARY All spikelets are built on the simple plan of 2- ranked florets with a pair of glumes at the base. The spikelet as a whole and each of its organs is subject to modification. The palea and one or both glumes may be suppressed; the lemma may be re- - duced but is never suppressed. The position and type of modification of sterile florets are of impor- tance in classification. The parts of a spikelet, however complex, are recognizable as rachilla, glumes, lemma, or palea. LESSON IV PEDICELED SPIKELETS OF FEW TO MANY FLORETS TAKING the spikelet of cheat or chess (Bromus secalinus), Fig. 11, as a starting point, we have one but little different from the diagrammatic spikelet (Fig. 8). [Names incidentally mentioned should not be memorized.] The lemmas are convex on the back, several-nerved, 2-toothed at the apex and bear an awn from be- tween the teeth. The awn is the midnerve extending beyond the body of the lemma. The rachilla joints are short, bring- ing the florets (Fig. 11, B) close together. The articulation is above the glumes and between the florets. The palea is grown fast to the grain. All species of Fig. 11. A, several-flow- ered spikelet of Bromus secalinus; B, single floret. Bromus have spikelets of this character, differing in size, texture, length of the awn, which may be much longer or reduced to a mucro (a minute point) or even suppressed, in being glabrous, as in cheat, or pubescent. The pubescence may cover the lemma 25 26 FIRST BOOK OF GRASSES — or be along the margins only. These differences distinguish the species. A genus is composed of one to many species having few to several important characters in common, and presumably descended from a common ancestor. Groups of species having less important characters in common form related genera. Festuca (Fig. 12) is related to Bromus, having few to several-flowered spikelets, disarticulating above the glumes and between the florets and with several-nerved lemmas; but the lemmas are awned from the tip, or pointed only, and not toothed, and the palea is not grown to the grain. A large number of grasses have laterally compressed spikelets of this general type, aya Leen of oe. Cisarticulating above the tuca ovina; B, lemma de- glumes and between the few boat to many florets. They are sorted into genera and the genera separated from each other chiefly according to the modifications of the lemma. In Panicularia (Fig. 13) the lemmas are broad and obtuse with strong parallel nerves. In Poa (Fig. 14) the lemmas are keeled on the back. and have five nerves converging toward the acute but never awned apex. The species figured (Ken- tucky blue-grass, Poa pratensis) and many others PEDICELED SPIKELETS 27 are villous (having soft curly hairs) on the lower part of the midnerve and the marginal nerves and have a tuft of white cottony hairs at the base, but this pubescence is not found in all the species. In all the grasses mentioned so far, the spikelets are borne in panicles (see Fig. 10, A). Spikelets much like those of Panicularia (Fig. 13) but borne in a raceme and having awned lemmas are found in Pleuropogon (shown in Fig. 10, B). In these spikelets the palea is crested or winged on the nerves (Fig. 15, showing a three quar- fo. 14. Spike- ter view of a palea removed [tof Poa pra- from the floret). , Returning to Fig. 11, A, we tes of note the midnerve of the lemma Panicularia extending as an awn beyond the sepenie- minutely toothed apex. The mid- nerve and the two lateral nerves as well are extended into awns in T'riodia flava (Fig. 16, the floret seen from the back) ; : the apex of the lemma is toothed and the 5%.,'?; nerves are villous below. In Fig. 17 (floret Pleuropogon of Cottea pappophoroides opened out and neat seen from the back) the lemma is lobed and nine to eleven of its many nerves are extended into awns. 28 FIRST BOOK OF GRASSES After the foregoing the spikelet and its parts will be recognized in most of the genera of the group having few to many-flow- ered pediceled spikelets. As stated in Lesson III the florets in a single spike- let may be of two kinds. The simplest spikelet of aces this type is found in the Floret of Teed (Phragmites commu- Triodia nis). In this, the lowest flava. floret is staminate or neu- Fie. 17. Lemma of ter and its lemma is much longer ©? ”?PoPhoratdes. than in the other florets (Fig. 18, A). In the per- fect floret (Fig. 18, B) it will be seen that the palea is very much shorter than the lemma, that the Fia. 18. A, spikelet of Phragmites communis; B, floret. rachilla bears copious long soft hairs, and that in- stead of disarticulating at its summit and remaining PEDICELED SPIKELETS 29 attached to the floret next below, it disarticulates at its base, remaining as a tiny feathery stem to the floret next above, its copious long hairs carrying the floret before the wind, dispersing the seed. [The hairs are much more co- \\\ pious than shown in the figure; they \\ are slighted to avoid obscuring the | difference in the florets.] Another spikelet with two kinds of florets is shown in Fig. 19 (Melica mutica). In this the lemmas of the Fic. 19. Spikelet of F * Melica mutica. upper florets are reduced in size, changed in shape, and contain no flower. Two or three of them are crowded together in a little club- shaped body. In the species figured, this modifica- tion is more marked than in most of the as, In Fig. 20 is shown a side view of the florets of a spikelet of | ( Pappophorum vagina- UX), \ Uy set etme OY / floret is perfect. Its ‘ broad lemma is cut into ’ f many spreading awns Wy | (compare with Fig. 17). NZ The two to four other Fic. 20. Florets of Pappophorum florets are crowded on eee the very short rachilla and are sterile; their lemmas are similar to that of the one fertile floret. .The 30 FIRST BOOK OF GRASSES rachilla does not disarticulate, the sterile florets re- maining permanently attached to the fertile one, the numerous awns of all together forming a pappus-like crown which carries the seed before the wind. [Pap- pus is the ‘‘down”’ on the seed of a dandelion, thistle, or other plant of their family.] Next we shall examine the inflorescence of a grass having spikelets entirely of sterile florets in addition to spikelets of fer- tile florets. Ex- 4) amine Fig. 21, A “4A (Cynosurus crista- -El4 tus), which shows yj] a small part of a 17 spike-like panicle. The spikelets are borne on minute Fig. 21. A, part of a panicle of Cynosurus pedicels on Very cristatus; B, sterile spikelet; ©, fertile short compound isa om branches. The lower one to three spikelets of each little branch are sterile, the lemmas containing no flowers (Fig. 21, B). The upper one to three spikelets are smaller and fertile (Fig. 21, C). When the bracts of the sterile spikelet are all alike empty, why are all but the lower pair called lemmas, instead of glumes? In many cases the nature of modified organs can only be recognized by their correspondence to or- gans in the same relative position in allied but more simple forms. In all the spikelets examined PEDICELED SPIKELETS 31 so far (and in all but a very small number of grasses) the bracts above the lower pair are flower- bearing or have a palea, which indicates their struc- tural identity. Corresponding parts in a modified spikelet are, therefore, regarded as lemmas. In the grass just examined the sterile spikelets remain on the panicle branches after the fall of the ripened fertile florets from their glumes. In Fig. 22, A (Achyrodes aureum), is showna fascicle # of one fertile and three sterile spikelets of another grass. In this the fascicles hang from the short slen- der branches of a narrow panicle and disarticulate from them, falling entire. This is the first example we have had so far of disartic- ulating branches of the in- florescence. In the figure the fascicle is seen from the fre. 22. A, fascicle of three sterile inner face to show the fer- pier ed nt Meg ee tile spikelet, which from _ spikelet. the outside is nearly hid- den by the sterile ones. Fig. 22, B, shows a sepa- rate fertile spikelet. It will be seen that except for the glumes, the two forms are strikingly dif- ferent. The fertile spikelet is reduced to one fertile and one rudimentary floret, both awned, me Sp “a ~ 5 ie | oe a fe coke &. Waiet 32 FIRST BOOK OF GRASSES while the sterile spikelets are many-flowered and awnless. We shall next examine a grass having unisexual spikelets (see Lesson III, page 23), the two forms borne on different plants (dicecious). (In the group with relatively simple | spikelets which we are now studying 4H there are no moncecious grasses). In Fig. 23 are shown the pistillate 7 and staminate spikelets of salt-grass | (Distichlis spicata). They differ but /] little in appearance and are both Fic. 23. Pistillate and _ staminate spikelets of Dis- tichlis spicata. borne in narrow panicles. [¢ sig- nifies female, ¢ male. These signs arecommonly used to indicate pistillate and sta- minate plants, respectively. | In Fig. 24 are seen thestrikingly Fig. 24. Staminate and pistillate spikelets of diverse staminate Scleropogon brevifolius. and pistillate spikelets of another dicecious species PEDICELED SPIKELETS 33 (Scleropogon brevifolius). The lemmas of the stam- inate spikelet are merely pointed; those of the pistillate spikelet bear three long slender twisted spreading awns. The pistillate florets fall from the glumes as a whole (the rachilla not disarticulating between: them) and roll before the wind as tiny tumble weeds. SUMMARY The inflorescence and the two to several-flowered spikelets of the brome-grasses, blue-grasses and their relatives are comparatively simple. In a few genera sterile spikelets are developed and in a few others the spikelets are unisexual. REVIEW Collect specimens of orchard-grass, meadow fescue, any species of brome-grass, or of Poa, or of any available grasses having laterally compressed, few to several-flowered spikelets. Identify the different parts of the spikelets. Lemmas may be spread out for examination by cutting off the very base with a sharp knife or scalpel. Spikelets of tough or rigid texture if soaked in water for a few minutes may be dissected without tearing. If very tough or hard, boiling the spikelets in water with a little glycerine (a drop of glycerine to about a teaspoonful of water) will make them manageable and keep them from drying out during dissection. LESSON V SESSILE SPIKELETS IN TWO-SIDED SPIKES RETURNING again to the spikelet of Bromus secalinus (Fig. 11), we shall strike out in another direction. Differ- entiation among living beings does not follow a line, but radiates like waves following the falling of a peb- ble in the water, or rather like waves of sound, in all directions. Hence we ean not follow an unbroken line in studying the increasing complexity of the inflorescence of grasses. We can only return to the center and start out on another line. Compare Fig. 10, A and B, with C, and with Fig. 25 (couch-grass or quack-grass, Agropyron repens). A raceme is a panicle reduced to its lowest terms. Eliminating the pedicels of the spike- lets of a raceme we have a spike, the spikelets set directly upon the rachis. _ Fig, 25. A, partofa Jn such an inflorescence the rachis spike of Agropy- , A ron repens; B, is usually more or less thickened. In part of ns Fig. 25, A, part of a spike is shown edge, all but from the flat side of the spikelet. t ikelets re- Sey pies . . ire aie The rachis is jointed and a spikelet is 34 SESSILE SPIKELETS IN TWO-SIDED SPIKES 35 borne at each joint, alternating on opposite sides. The rachis is thickened and the joints slightly hol- lowed on alternate sides. (See Fig. 25, B, a diagram- matic illustration of the rachis seen from the edge with all but two spikelets removed.) The spikelet is not very different from that of Bromus (Fig. 11) and as in that, the ripened florets fall from the i glumes. 4 In the group of ¥ grasses taken up in this \ , lesson the specialization is mostly in the rachis and in the position of the relatively simple spikelets. In Fig. 26, a spikelet of cultivated wheat (Triticum estivum), we have the same type of spikelet as in Fig. 25 but with fewer and much plumper florets, with broader lemmas slightly toothed at the apex, and with long awns that are scabrous (rough, like a file). [Awns form the ‘‘beard”’ of wheat.]| The spikelets are borne on a jointed rachis as in Fig. 25 (Agropyron repens) but the joints are shorter, bringing the spikelets closer together and hiding the rachis. In this, cultivated wheat, the florets do not readily fall from the glumes but re- Fig. 26. Spikelet of Triticum estivum. 36 FIRST BOOK OF GRASSES main in the spike and the ripened grain is thrashed from them. This: persistence of the lemma and palea has been fixed by selection in cultivation. In the closely related emmer (Triticum dicoccum) the rachis breaks at the joints, each joint remaining attached to its spikelet. Compare Fig. 27 (Lolium multi- - | \ florum) with Fig. 25. Note that in Vy /, Fig. 25 the spikelets are borne “flat side against the rachis, while Uf in Fig. 27 they stand with their (, edges against the rachis. The f y Yj rachis itself is of the same type as GEZAA ,_———— iy yy ———— ——— pe Z/la- VY Y that in Fig. 25. The diagram- matic rachis with two spikelets \\e | Hi | Y ° . . \V i Y seen edgewise (Fig. 25, B) is shown } { See from the same position as is Fig. 27. In this, as in Agropyron, the rachis is continuous (not disartic- . ulating) and the florets fall. If peda leol he spike we separate a spikelet from the rachis we find that the first glume (the one that would be against the rachis) is sup- pressed, the first floret lying directly against the rachis. In the single spikelet borne at the summit of the rachis the first glume is developed and is about as large as the second. When but one glume is present, we know which glume it is and which is suppressed by the position of the first floret, which Zi Ji —— SESSILE SPIKELETS IN TWO-SIDED SPIKES 37 is always above the first glume. When no glume is found below the first floret, it is obvious that it is the first glume which is suppressed. In Fig. 28 (Lepiurus cylindr- cus) we have a greatly thickened, strongly nerved rachis with spike- lets placed as in Fig. 27 but reduced to the second glume and a single floret. This little spikelet is sunken in the hollow of the rachis joint, the second glume fitting snugly over the hollow, the whole forming a long, slender, wiry cylinder. At maturity the rachis disarticulates with the spikelets firmly embedded in the joints. (See diagram of - rachis, Fig. 28, B, and, above, a joint with spikelet removed, show- ing the hollow, and a second with the spikelet in position.) The plant bearing these spikes grows along mud flats near the sea. The rachis joints are cylindrical and readily roll down the slope to the water. Being corky, they are carried by the lightest ripples and are thus spread over wide areas. The grain Fig. 28. A, part of spike of Lepturus cylindricus; B, dia- gram of _ rachis, spikelets removed; above, diagram of single joint. germinates within its little cell, and the young roots and leaves push aside the water-soaked glume. 38 FIRST BOOK OF GRASSES Returning to Fig. 25 with its one spikelet at each joint, flat against the rachis, compare with it Fig. 29 (Elymus vir- ginicus, or rye-grass). In this there are two spikelets at each joint of the rachis, the first glumes back to back, the spikelets somewhat distorted, each pair reaching around the edges of the rachis. The / figure shows a pair of spikelets and two joints of the rachis, with the pair of spikelets next above, on the opposite side of the rachis, lightly sketched in behind; two more internodes of the rachis, with spikelets re- moved, are shown by dotted lines. A dia- grammatic sketch of a pair of spikelets, the dis- Fie. 29. A, pair of spikelets of Ely- cornon reduced, isshown mus virginicus; B, diagrammatic above. A comparison of figure of the pair of spikelets. the diagrammatic spike- let with the spikelet in Fig. 25 will show the structural similarity. In their natural position the spikelets, a SESSILE SPIKELETS IN TWO-SIDED SPIKES 39 _ overlapping on the short joints and extending around the edges of the rachis, so that at least one is seen nearly edge-wise, form a spike that may well be con- fusing to the beginner, especially when, as in the species shown in Fig. 29 and several others, the glumes stand out like a 4-rayed involucre below the ap- pressed florets. However, a single joint with its spikelets attached cut out of the spike readily dis- closes the structure. In some species of Elymus there are three spikelets and occasionally four or five at a node, the distortion being correspondingly greater. In several species the glumes are so narrow as to appear like bristles or awns only. In most of the species the rachis is continuous and can not be disjointed. In a closely related genus, Sitanion, the rachis disarticulates at the base of each joint, the slender rigid joint remaining as a tiny sharp-pointed stem below the cluster of long-awned spikelets. The awn-like glumes of Sitanion commonly split between the nerves, sometimes to the very base, appearing like a cluster of awns below the florets. In Elymus and Sitanion the spikelets are all alike (or some occasionally variously aborted) and all sessile (set directly on the rachis). In Fig. 30 (Hor- deum nodosum, one of the wild barleys), a group of three spikelets and a joint of the rachis are shown. As in Sitanion the rachis disarticulates at the base of the internode, the joint remaining attached to the spikelets above it. Note that the central spikelet is sessile and the lateral ones pediceled, that the lower 40 FIRST BOOK OF GRASSES floret of the central spikelet is well developed while those of the lateral florets are rudimentary, and that the back of the floret is turned from the rachis with the glumes (bristle-like in this species) at the sides or back, contrary to the arrangement characteristic of grass spikelets. The problem of the glumes in Hordeum has not been satisfactorily solved. It appears prob- able that the reduced rachilla joint between the second glume and the floret is twisted and bent inward, bring- ing the glumes at the side or back of the floret. In the culti- vated barley the rachis does not break up, as in the wild species, the continuous rachis having been fixed by selection. The florets fall from the spike in thrashing, or in naked or hull-less barley the grains fall from the lemma and palea, as in wheat. In cultivated 2-rowed _ barley the lateral spikelets are pediceled and sterile, as in the wild species, but in 4-rowed and 6-rowed barley the lateral spikelets are sessile and fertile, characters fixed by selection. The grasses characterized by the spicate in- Fic. 30. Joint of spike of Herdeum nodosum. SESSILE SPIKELETS IN TWO-SIDED SPIKES 41 florescence dealt with in this lesson form the barley tribe, which from the standpoint of man is the most important group of grasses, if not of all plants, in the world, containing wheat, barley, and rye. SUMMARY The specialization in spicate inflorescence is chiefly in the rachis and next in the position of the relatively simple spikelets. When the glumes are distorted, standing side by side, as often found in species of Elymus, or when one of them is suppressed, we can tell which is which from the fact that the first, or lowermost, floret is always above or on the same side of the spikelet as the first glume. REVIEW Collect heads of wheat, rye, barley, quack-grass, species of Elymus, Lolium, or squirrel-tail grass (one or more of these will be found anywhere in the United States). Note whether the rachis readily disjoints. If so, separate out a single joint with the spikelets attached. Note where the rachis breaks, at the summit or base of the joint; note the number of spikelets at a joint and the number of florets to a spikelet. Distinguish the individual spikelets and their parts. If the rachis does not dis- joint, cut across the middle of the internodes, taking out a single joint with attached spikelets. Note the number of spikelets to a joint; whether there are one or two to several fertile florets to the spikelet and whether the spikelet is placed flatwise or edgewise to the rachis. Note how spikes bearing spikelets with scabrous awns push themselves forward when handled, LESSON VI PEDICELED SPIKELETS WITH LARGE GLUMES AND OTHER MODIFICATIONS TURNING again to the spikelet of Bromus secalinus (Fig. 11), compare with it Fig. 31 (wild oats, Avena fatua). The glumes are greatly enlarged and the rachilla joints are - so short that the florets appear to be almost oppo- site. The awn, instead of extend- ing from the apex of the lemma, pro- trudes from the back and is twisted for about half its length. We noted in Les- son IV that the awn is an exten- sion of the mid- nerve. This fact is well shown in the floret of wild oats (Fig. 31, B, the floret seen from the back), in 42 Fia. 31. A, spikelet of Avena fatua; B, floret. PEDICELED SPIKELETS WITH LARGE GLUMES 438 which the midnerve leaves the body of the lemma about the middle of its back and becomes a free awn, while the lemma above the departure of the awn is nerveless. This is invariably the case when Fig. 32. Spikelet of Trisetum spicatum. the awn is dorsal (that is, pro- truding from the back); the lemma is always nerveless above it. Compare Figs. 31 and 32 (Trisetum spicatum). It will be seen that they are the same type of spikelet. In Trisetum the awn is loosely twisted Fia. 33. Spikelet of Danthonia and is borne npn tea’ the ap bs spicata; floret above. of the lemma, which is tipped with two slender teeth. In Danthonia spicata (Fig. 33) the florets are more numerous and are smaller in proportion to the glumes. The broad strongly twisted awn arises from 44 FIRST BOOK OF GRASSES between the teeth or lobes of a bidentate (2-toothed) apex (see floret above, seen from the back). In these three spikelets, as in all but a few of the grasses of the oat tribe, the florets fall from the glumes which remain on the pedicel. In velvet grass, Notholcus lanatus (Fig. 34), the spikelet falls entire. 7 The articulation of the spikelet, taken throughout the grass family, is so nearly uniform for related genera that it is relied on to differentiate large series. (See Lesson IT, on inflorescence). There are exceptions to the Fic. 34. A, spikelet of mode of articulation character- Notholeus lanatus; B, istic of the group as a whole pair of florets. A ; in the case of a few genera in three of our tribes. Such exceptions are puzzling to the beginner, leading him astray in using keys. We must learn to observe all the characters of the inflorescence and base our judgment on the sum total of the characters, remembering that ‘‘Nature does as she pleases” and rejoicing that in grasses she at least pleases to keep invariably to the 2-ranked arrangement of the spikelets. (See the summary of Lesson II). In Lesson IV, Figs. 18 and 19, we observed spikelets in which the florets were of two kinds. In Notholcus lanatus the lower floret is per- fect and awnless and the upper is staminate and bears a hook-like awn from the back (Fig. 34, B, PEDICELED SPIKELETS WITH LARGE GLUMES 45 florets removed from the glumes. Note the curved and exceptionally long lowermost rachilla joint.) In Sphenopholis, closely related to Trisetum (Fig. 32), the spikelets fall entire, as in velvet-grass. In tall oat-grass (Arrhenatherum elatius) the spikelets bear one perfect awnless floret and one staminate awned floret, as in velvet-grass, but their position is © reversed, the staminate being below and the perfect above. SUMMARY In the oat and its relatives the large glumes and the awn of the lemmas are the most prominent characters, although in species of some genera the awn is wanting. Pubescence is commonly con- spicuous. The inflorescence is an open or contracted panicle. REVIEW Collect panicles of wild oats or awned specimens of cultivated oats (cultivated oats growing wild commonly bear awns), of Danthonia, one or more species of which are to be found through- out most of the United States, and any of the related grasses available, and examine the spikelets. Place a floret with a twisted awn in a drop of water and observe the result. In awn- less florets of cultivated oats note that the midnerve of the lemma, if it is not at all produced into an awn, stops abruptly at the point where the awn would arise normally and that the lemma is nerveless above this point, just as if it were awned. LESSON VII PEDICELED ONE-FLOWERED SPIKELETS Turn to the spikelet of wild oats (Fig. 31) and in imagination eliminate all but the lowest floret and the glumes. Better still, with a spikelet of oats in hand break off all above the lowest floret. Now we have a large model or pattern of the reduced spikelet of a very large number of grasses, red-top, timothy, and their kind. Examine Calamagrostis canadensis (Fig. 35, the floret raised from its glumes) and note how it corre- sponds to the pattern obtained by reducing the spikelet of oats to a single floret. The only vestige remaining of the other florets is the minute rachilla joint back of the palea (shown, exaggerated somewhat, in Fig. 35). Inall but a few genera the Fig. 35. Spikelet of Cala- ° : P magrostis canadensis, Yachilla is entirely suppressed. | Compare Fig. 35 with Fig. 32 and note that Trisetum reduced to a single floret would closely resemble Calamagrostis. In Agrostis (Fig. 36) the rachilla is normally suppressed. In two species in the far West it is present as a minute rudiment. In most of the species 46 = PEDICELED ONE-FLOWERED SPIKELETS 47 the palea also is suppressed or represented by a rudiment only (Fig. 36, Agrostis hiemalis, the floret with palea wanting, raised from the glumes). Here we have specialization through elimination, the spikelet reduced almost to its lowest terms, one floret with no palea, no awn, no callus hairs. In this particular species the very open panicles break off and roll before the wind as tumbleweeds, scattering the fe. 36. spike- seed. Several species of Agrostis have ‘ is reduced to a mi- Fig. 65. A, part of raceme of Axonopus furcatus; B, nute sheath inflorescence; ©, cross- ™é section of rachis. \ around’ the | enlarged rachilla joint below the |} second glume and grown fast to it. / The fertile lemma is tipped with a minute awn, which breaks off ; readily. Fic. 66. A, spikelet of | In barnyard-grass, Echinochloa Eriochlo ctata; i B, fertile floret. Crusgalli, is the same type of spike- 74 FIRST BOOK OF GRASSES let as in Panicum, but the second glume and sterile lemma are awned or awn-tipped, and the pointed tip of the palea is not inclosed in the lemma, which also is sharp-pointed. The glumes and sterile lemma bear spine-like hairs, but not so thick as those of Nazia. Observe the diagrammatic panicle in Fig. 67, A. In imagination, remove all the terminal spikelets \ Fic. 67. A, diagrammatic panicle; B, two branches of panicle with terminal spikelets removed. \ Jie Fic. 68. Small part of bristly Le panicle of Chetochloa Grise- [> bachii. from a branch. The result is Fig. 67, B. Compare Fig. 68 with this. It will be seen that the bristles of the latter are branches of the panicle which are without spikelets at their tips but which bear sub- sessile spikelets at or toward their base. This SPIKELETS WITH HARDENED FRUITS 75 modification of sterile branches and branchlets into rough bristles is found in the millets (Chetochloa), common millet, yellow foxtail, and green foxtail. These branchlets are the “‘involucre of bristles” — referred to in many manuals and descriptions. The spikelets fall from their pedicels, as in Panicum, and the bristles remain on the axis. In most of the species the pani- cles are dense and spike-like, the spikelet- < bearing branches fas- “dy cicled and very short ° and the sterile ones, or bristles, long and slen- der. In yellow fox- tail, Chetochloa lute- scens (Fig. 69), the fas- cicle consists of several branches, only one of them spikelet-bearing, Fia. 69. A, fascicle from panicle of the others transformed Chetochloa lutescens; B, spike-like : - panicle of same. into slender bristles. Between the ordinary panicle represented in Fig. 67 and the bristly spike-like one of yellow foxtail (Fig. 69, B) there is every degree of gradation. A further specialization of sterile branches is shown in the sand-burs (Cenchrus). Compare Fig. 70 with Fig. 69, A. Instead of the nearly simple fascicle of 76 FIRST BOOK OF GRASSES branchlets found in the millets, the sand-burs have a complex fascicle of many compound branches. The _ primary branches disarticulate from the axis, and the whole fascicle, or bur, falls entire with the spike- let permanently inclosed, the grain eventually germinating within it. The simplest form of sand- bur is in Cenchrus myosuroides (Fig. 70). The strue- ture of this will be comprehended if we conceive of Fig. 70. Bur of Cenchrus myosuroides. Fie. 71. Bur of Cenchrus pauciflorus. a dense fascicle of branches with but one of them bearing a single sessile spikelet, the others branching at the base and surrounding the spikelet-bearing one. In the common inland sand-bur, Cenchrus pauci- florus (Fig. 71), the sterile branches are much thick- ened and flattened and are grown together below, their free summits sharp and spine-like. The nu- merous secondary branchlets, in the form of stout spines, spread from the body of the bur formed by the cohesion of the main branches surrounding the one or few sessile spikelets. a ———————— SPIKELETS WITH HARDENED FRUITS 77 In all the seed-bearing spikelets so far examined, the stamens and stigmas project at flowering time, resulting in cross fertilization of the ovules. A few grasses bear cleistogamous (close-fertilized) as well as openly fertilized spikelets. Such spikelets do not open, and there is usually but a single small stamen which empties its pollen directly on the short stig- mas. The grain is larger than that of open-fer- tilized spikelets. Two of our native grasses, form- 4 ing the genus Amphicar- Q pon, related to Panicum, '* 7? paar Fayre a are remarkable in that they produce large cleistogamous spikelets under- ground from subterranean branches (Fig. 7 2). The plants also bear terminal panicles of ordinary and much smaller spikelets, but these seldom perfect seed. SUMMARY In Panicese, the millet tribe, the spikelets fall entire. They bear one perfect floret with a sterile floret below it. The lemma and palea of the perfect floret are indurate. The sterile lemma resembles the second glume, the two simulating a pair of glumes, while the first glume is small and looks like an addi- 78 FIRST BOOK OF GRASSES tional one (in some older works it is called the acces- sory valve or glume) or is entirely suppressed. The genera differ chiefly in the arrangement of the spike- lets in the inflorescence. In a few genera some of the panicle branches are without spikelets and are transformed into bristles or burs. REVIEW Collect the inflorescence of broom-corn millet, old witch-grass, or of as many species of Panicum as are available, and dissect the spikelets. Collect heads of barnyard-grass and compare the spikelets with those of Panicum. Compare the spine-like hairs . with those of Nazia, Fig. 54. Collect the inflorescence of crab-grass and of any species of Paspalum available; note the arrangement of spikelets. and com- pare the form and texture of the fruit (fertile floret) with that of Panicum. Collect heads of common millet or of yellow or green foxtail. Remove enough branches from the axis to show clearly the form and arrangement of the few remaining. If available, examine a sand-bur, splitting the bur with a sharp knife. LESSON XI PAIRED SPIKELETS WITH HARDENED GLUMES AND THIN LEMMAS In the sorghum tribe, which we are about to study, the spikelet, as in the millet tribe, falls entire and is dorsally compressed, but the glumes are hardened and the lemmas thin, while in the millet tribe the glumes are thin and the fertile lemma and palea are hardened. The glumes entirely inclose the two florets. The midnerve of the first glume is com- monly suppressed, while a pair of nerves near the margin is often prominent and sometimes keeled, or even winged. The second glume may be like the first or the midnerve may be slightly keeled. The lower floret consists of an empty lemma only, the upper of a perfect flower with a small thin often awned lemma (sometimes so small as to appear like a bit of membrane at the base of the awn) and a minute palea (sometimes suppressed). Examine Fig. 73, A, and compare it with Figs. 21 (p. 30), 22 (p. 31), 30 (p. 40), 40 (p. 49), 52 (p. 61), and 53 (p. 62). In all the figures referred to, fertile spikelets are associated with sterile ones. These are grouped in various ways. In the sorghum tribe the typical arrangement is a jointed raceme with a sessile perfect spikelet and a pediceled sterile spikelet 79 80 FIRST BOOK OF GRASSES (staminate or neuter) at each joint, the rachis dis- articulating at the summit of each joint, this and the pedicel of the sterile spikelet remaining attached at their base to the per- \J. fect spikelet as a pair } of little stalks. In / this group of grasses specialization consists chiefly in modifications of the axes of inflores- cence and secondarily in the modification of the spikelets. In Fig. 73, A (Johnson-grass, Holcus halepensis), are two views of a single joint, consisting of the sessile perfect spikelet , with the attached ra- ‘4 chis joint and the ped- icel with its staminate oy , spikelet. Fig. 73, B, is 2 WE «4 NWI a diagram of a raceme Fie. 73, A, two views of single joint of four such joints and of raceme of Holcus halepensis; B, 73, C, a diagram of the diagram of raceme of four joints; fe i C, diagram of rachis and pedicels; D, rachis and pedicels, the sieaeeores aie spikelets removed and the points of disarticulation shown by dotted lines. In sorghum, or Holcus, these little racemes are borne on the ultimate branchlets of a panicle (Fig. 73, D). PAIRED SPIKELETS 81 In the closely related Indian-grass, Sorghastrum, the racemes are reduced to one or two joints, and the sterile spikelets are wholly suppressed, the slender hairy pedicel alone being developed. Compare Fig. 74, A, a diagram- matic single joint of Erianthus sac- charoides, with Fig. 73, A, and Fig. 74, B, three joints of a long raceme, with Fig. 73, B. It will be seen that, while the spikelets are paired, the pedi- celed spikelet is as large as the sessile one and, like it, is awned. In this genus the pediceled spikelet is usually perfect. The ra- cemes are long and slender and _ co- piously hairy, and are crowded on a Fic. 74, A, diagram of single joint of raceme of Erianthus saccharoides; B, three joints of raceme. stout main axis forming large, dense, woolly panicles. In sugar-cane the inflores- cence is like that of Erianthus, but the spikelets are awnless. 82 FIRST BOOK OF GRASSES Examine Fig. 75, A, a joint of a raceme of one of the broom sedges (Andropogon scoparius). It will be seen to have the same general plan as a joint of sorghum, and the racemes, shown in Fig. 75, B, though slender and flex- uous, are in structure similar to the racemes of sorghum. The entire in- florescence, however, is very different. In sorg- hum and Erianthus the , inflorescence is a leafless terminal panicle, such as is common in grasses. The diagrammatic inflores- cence (Fig. 75, B) shows that the racemes of the broom-sedge are borne on Fic. 75. A, single joint of raceme ae meses a slender : leafy of Andropogon scoparius; B, branches arising in the oe ert of eompound, into- "sails of.) leaves 66 ae main culm or branches, the whole forming a compound inflorescence. The leaves, especially the ultimate ones immediately below the racemes, are mostly reduced to blade- less sheaths and are sometimes bright colored. Such transformed leaves subtending or surround- ing single inflorescences are commonly termed spathes. In some species the racemes are partly inclosed PAIRED SPIKELETS 83 in the spathes. The racemes may be solitary (that is, one to a branch, as shown) or two to several, digitate, on a single branch. In some species the sterile spikelet is suppressed, only the pedicel being developed. In a few species of Andropogon the racemes are in leafless panicles, as in Erianthus, but the pedi- celed spikelets are sterile and awnless. Examine Fig. 76, A (Man- suris cylindrica), and compare it with 73, A. In Holcus and Erianthus the rachis joint and the pedicel of the second spike- let are about equal in thick- ness. In Fig. 75 the rachis joint is stouter than the pedi- cel. In Manisuris the rachis joint is greatly thickened and hollowed out below on the inner face. (See Fig. 28, p. 37, for an earlier example of a thickened rachis.) The pedicel of the ster- ile spikelet is also thickened, but much less so, and the two lie close together (instead of Fia. 76, A, single joint of raceme of Manisuris cy- lindrica; B, another view of two joints; C, inner face of rachis joint and pedi- cel, spikelet removed; D, part of cylindrical many- jointed raceme. spread apart, as in Figs. 73 to 75) and entirely cover the second-glume side of the spikelet which fits into the cavity formed by the rachis joint and pedicel. Fig. 76, C, shows the inner faces of the rachis joint 84 FIRST BOOK OF GRASSES and pedicel, the perfect spikelet removed. The spikelets are awnless and the marginal nerves of the Fic. 77. A, single joint of raceme of Rytilix granu- laris; B, reverse view; C, diagram of rachis and pedicels of three joints of raceme, spike- lets removed; D, view of inner face of first glume of sessile spikelet; E, raceme; F, reverse view. first glume are winged at the summit. This inflorescence and that of Lepturus (Fig. 28) are so similar in appearance and func- tion, disjointing with the mature seed permanently attached, that the student may wonder why they are placed in tribes so re- mote as the barley grasses and sorghum. The two _ illustrate what is not seldom shown in nature, that very different struc- tures may arrive at the same function, although the forms from which they are derived were very remote. The inflorescence of Lepturus is a spike, and its spikelet shows it to be a reduced and specialized relative of Lo- lium. The inflorescence of Mani- suris is a raceme with two spike- lets at a joint, one pediceled, and the perfect spikelet not greatly different from that of sorghum. We have just observed the rachis joint and pedicel lying closely pressed together. In Rytilix (Fig. 77, A) they are adnate (grown to- gether). Examine the diagram of three joints with PAIRED SPIKELETS 85 the spikelets removed (Fig. 77, C) and note that the rachis joints (the thicker parts) form a central axis with the pedicels borne on alternate sides, just as they are in Holeus (Fig. 73, C). In this raceme we have just the reverse of the arrangement of Manisuris. In that, the fertile spikelet fits into a cavity formed by the rachis joint and pedicel while in Rytilix the rachis joint and pedicel fit into a cavity formed by the fertile spikelet. (See Fig. 77, D, showing the inside face of the first glume of the perfect spikelet, and examine the two views of the joints, A and B, and the two views of the raceme, E and F). The sterile spikelet is well developed but usually contains no flower. The perfect spikelet with its peculiar gray, ridged and pitted, subglobose first glume sug- gests the achene of some species of Scleria (a sedge). The whole inflorescence is very unlike that of any other known grass. Return to Fig. 73, A, and compare with,it Fig. 78, A (Rhaphis pauciflora), which shows a raceme reduced to a single joint consisting of the sessile perfect spikelet and two pediceled sterile spikelets, the rachis jomt (found in Holcus and the others studied) often being replaced by a second pedicel and sterile spikelet. This second sterile spikelet is not always developed. In Rhaphis this 1-jointed raceme has a sharp hairy callus at the base., Fig. 78, B (the base of a raceme and the summit of a branch, the hairs removed), shows the source of this callus. The racemes are borne on the long slender branches of a 86 FIRST BOOK OF GRASSES spreading panicle (Fig. 78, D) which disarticulate obliquely below the raceme, the line of articulation bordered by a dense brush of hairs (Fig. 78, C). The Fic. 78. A, one-jointed raceme of Fie. 79. A, single joint of raceme Rhaphis pauciflora; B, base of of Heteropogon contortus; B, per- raceme and summit of branch; fect spikelet from which sterile C, hairy summit of branch from spikelet has fallen; C, base of fer- which raceme has fallen; D, in- tile spikelet and its callus; D, ra- florescence. ceme; E, diagram of raceme; F, diagram of rachis and pedicels of four joints of raceme. sterile spikelets with their pedicels fall off before the maturity of the perfect spikelet, which, with, its hairy callus and long awn, closely resembles a mature floret of Stipa (Fig. 45). PAIRED SPIKELETS 87 In Heteropogon (Fig. 79) is a more complicated arrangement. In A is shown a joint of the raceme, consisting of the sessile perfect awned spikelet and the pediceled large pale sterile spikelet, with the short hairy rachis joint below forming a pointed callus. In Fig. 79, C, is seen the base of the fertile spikelet and the callus, with enough hairs removed to expose the very base of the pedicel, which has fallen with the sterile spikelet, and (to the left) the oblique scar from which the rachis joint next above has disarticulated. The articulation is at the base of the rachis joint instead of at the summit, as in Holcus and the others (Figs. 73-77), and forms a callus to the spikelet next above. The rachis joints are very short and the fertile spikelets (except their awns) are hidden by the overlapping sterile spikelets, the long flexuous awns forming a tangle beyond the end of the raceme (Fig. 79, D). Besides having this modified rachis and large unsymmetrical sterile spikelets, the raceme of Heteropogon presents a further modifica- tion in that its lower two or three pairs of spikelets are all sterile, instead of each pair consisting of a perfect and a sterile one, as typical for the sorghum tribe. In these lower pairs the sessile spikelet is staminate and the pediceled one empty, but both have large unsymmetrical glumes and appear to be alike. Examine the diagram of the raceme (Fig. 79, E, the rachis greatly lengthened to show the structure and arrows indicating the points of dis- articulation) and contrast it with Fig. 73, B. Fig. 88 FIRST BOOK OF GRASSES 79, F, is a diagram of the rachis of four joints, show- ing the points of disarticulation. Contrast it with Fig. 73, C. The mature fertile spikelets of Rhaphis and Heteropogon closely resemble mature fruits of Stipa and function in the same way, disseminating the seed by attaching themselves by their sharp calluses to passing animals and securing their hold by the un- twisting and twisting of their hygroscopic (moisture- sensitive) awns. They afford another example of the same function performed by different organs. SUMMARY The grasses of the sorghum tribe have spikelets with hardened glumes and thin lemmas. Typically they are arranged in pairs, one sessile and perfect, the other pediceled and sterile on a jointed rachis, forming racemes. In several genera the rachis is elaborately modified. REVIEW Collect the inflorescence of Johnson-grass and of any of the broom-sedges. Break the raceme into single joints, noting the points at which it readily separates. Examine the spikelets and the rachis and distinguish the rachis joint from the pedicel. If you had a panicle of Rhaphis (Fig. 78, D) from which all the sterile spikelets with their pedicels had fallen, how would you know it was not a species of Stipa? Make diagrams of the inflorescence of any available species of the sorghum tribe. LESSON XII HIGHLY SPECIALIZED UNISEXUAL SPIKELETS ExAMINE Fig. 80, gama-grass (T'ripsacum dactyloides). is moncecious (see p. 23). The pistillate spikelets are borne on the lower | part of the one to three stout digitate racemes and the staminate spike- lets on the upper part of the same racemes (Fig. 80, D). The part of the rachis bearing the stami- nate spikelets is rela- tively slender and falls off after flowering time; the part bearing the pis- tillate spikelets is greatly thickened and disarticu- lates with the spikelets permanently embedded in the joints. Fig. 80, A, shows two joints with spikelets embedded; B, a This grass |\ Fig. 80. A, two joints of pistillate part of raceme of Tripsacwm dac- tyloides; B, spikelet removed from the joint; C, rachis joint from which spikelet has been removed; D, inflorescence; E, diagram of rachis of pistillate part; F, pair of staminate spikelets. 89 ‘ 90 FIRST BOOK OF GRASSES spikelet removed from its joint; C, the joint from which a spikelet has been removed; and E, a diagram of the rachis. Compare the joints with those of Manisuris (Fig. 76). Note that in Tripsacum the pistillate part is a simple spike with no vestige of a second spikelet. The structure of the spikelet is similar to that of spikelets of the sorghum tribe, hay- ing firm glumes (the first hard and like the rachis joint in texture), a thin sterile lemma, and a very thin fertile lemma and palea. The staminate spike- lets are in pairs (Fig. 80, F) and one of each pair is usually on a very short pedicel. The glumes are much like those of different species of Andropogon; both florets are staminate, their lemmas and paleas thin. Indian corn, or maize (Zea Mays), the most highly specialized grass known, is believed to be related to Tripsacum. It is moncecious, but the staminate and pistillate spikelets are borne in distinct inflorescences. The staminate spikelets are in pairs on a slender rachis forming racemes, these arranged in a panicle, the ‘‘tassel,”’ at the summit of the culm. The pistillate spikelets are in pairs, crowded in 8 to 16 rows (rarely more) always an even number, on a greatly thickened compound axis, the ‘‘cob,’”’ borne in the axils of the leaves and enveloped in numerous leafy bracts (‘‘husks’’), the long styles (the “‘silk’’) protruding from the summit, the whole called the ‘ear.’ The staminate spikelets (Fig. 81, C) are much like those of Tripsacum and contain two HIGHLY SPECIALIZED UNISEXUAL SPIKELETS 91 staminate florets. The pistillate spikelets stand at right angles to their axis, the cob (Fig. 81, A). The glumes are minute, scarcely covering the ovary at flowering time (Fig. 81, B), and in the ripened grain remain as chaff on the cob, the greatly enlarged grain being naked. This grain, ‘‘a kernel of corn,”’ remains on the axis until loosened by force. Indian corn is . i known only in culti- (\Y’ vation. Theproblem of its origin has not yet been solved. We : (A geen oteag of arar ete originated somewhere __pistillate spikelets standing at right in America, probably _ sy¥lesto tr ats (the cob): Bi on the uplands of C, part of raceme of staminate spike- the tropics. It was si widely cultivated by the aborigines when the western continents were discovered. The problem of the morphology of the ear of corn has not yet been satis- factorily solved, but it is not at all difficult to recog- nize the individual spikelets, both staminate and pistillate. The student who has come this far will find it pos- sible to dissect the inflorescence of any of our grasses, with the possible exception of buffalo-grass, and to distinguish the spikelets and the different organs. 92 FIRST BOOK OF GRASSES Buffalo-grass (Bulbilis dactyloides) is placed in the same tribe as the grama-grasses (page 57), but it has been left until the last, so that the student might have greater experience to draw on. It is dicecious (p. 23), and the staminate and pistillate inflorescences are strikingly different. (Recall Fig. 24, p. 32). Examine Fig. 82, B, showing a_ pistillate spikelet with a small narrow first glume and a very large, broad, 3- toothed second glume entirely infolding the body of the floret. The floret, removed from the glume, is shown in Fig. 82, C. Its 3- toothed lemma _ ap- proaches in form the ae . lemma of some of the Fie. 82. A, pistillate inflorescence of Bulbilis dactyloides; B, pistillate QTAaMA-grasses. (Com- gilt nt rom the machi: Coie pare with Fig. 50, 8) head, showing one of the two rows This pistillate spike- cence; F, staminate spikelet. let as presented is not at all puzzling, the large thick second glume with overlapping edges en- tirely inclosing the floret being the only remarkable feature. These spikelets, however, are borne in the curious inflorescence shown in Fig. 82, A. These little hard white heads are borne, mostly two together, on HIGHLY SPECIALIZED UNISEXUAL SPIKELETS 93 a very short slender axis, in the axils of broadened sheaths at the summit of short culms and very much overtopped by the blades. If we cut off the base of one of these little heads we have three to five such spikelets, as shown in Fig. 82, B, which had been held rigidly together by the hard thickened base, the overlapping backs of the second glumes forming a thick white wall surmounted by their green-toothed summits. The base can be nothing but a rachis, shortened, broadened, and thickened, though all trace is lost of the junction of the spikelets and the rachis. Fig. 82, D, is a diagram of half a head, show- ing one of the two rows of spikelets. Compare this with Fig. 51, E, showing a great number of spikelets’ on an elongate rachis. As in Bouteloua gracilis the spikelets stand nearly at right angles to the rachis and the first glume is inward, that is, it would be against the rachis, as in Bermuda-grass, if the spike- lets were appressed. The difficulty the student en- counters in comprehending the pistillate inflores- cence of buffalo-grass is not so much due to its complexity as to the difficulty of dissecting the rigid little structure, and also to the suppression or deformity of some of the organs. When spike- lets are closely crowded, as in the Cenchrus bur, in some species of millet, and in buffalo-grass, some of them are nearly always deformed from pressure. In buffalo-grass one or two of the spikelets in a head are not fully developed. The first glume is commonly reduced to a minute scale in two or 94 FIRST BOOK OF GRASSES three of the spikelets, and sometimes it is sup- pressed. The staminate inflorescence (Fig. 82, E) resembles the spikes of grama-grass (Fig. 51, E) but the spike- lets (Fig. 82, F) are awnless, and the second floret is well developed and contains a staminate flower. SUMMARY In the case of greatly modified structures the different organs are to be interpreted by their relative position and by their analogy to corresponding organs in related grasses. REVIEW Collect the inflorescence of Tripsacum if available, disjoint it and dissect the spikelets. Examine a very young ear of corn and note that the spikelets are always in pairs (consequently an ear of corn always has an even number of rows). Distinguish the glumes and lemmas. Examine the staminate spikelets. If available, collect buffalo-grass and dissect the staminate and pistillate inflorescence, preparing the latter by boiling it a few moments in water to which a few drops of glycerine have been added. When no glume but the large thickened one is found in the pistillate spikelet of buffalo-grass, how do you know which glume that is? DIAGRAMMATIC SUMMARY OF THE PRIMARY CHARACTERS OF THE TRIBES SERIES I.—POATA Spikelets laterally compressed; florets mostly falling from the persistent glumes 1. Festucem, fescue tribe Spikelets many-flowered. Glumes relatively small. Vy ZL Lemmas awned from summit \ Ms or awnless. aN Inflorescence a panicle. Vf [Contains fescues, brome- Fic. 83. grasses, and blue-grass.] 2. HorpEm, barley tribe Spikelets 1- to many-flowered, sessile on opposite sides of a jointed rachis, forming a spike. Rachis rather than spikelets specialized. Inflorescence a solitary spike. [Contains wheat, rye, and barley.] 3. AVENE, oat tribe Like Festucez, but glumes en- larged and florets fewer in number. Rachilla joints short. Lemmas awned from the back. Inflorescence a panicle. Fic. 85. [Contains oats.] — 95 96 FIRST BOOK OF GRASSES 4, AGROSTIDE®, timothy tribe Like Festucex reduced to its lowest terms; spikelets 1- flowered. Lemmas awnless or awned, the awn from back or summit. Glumes sometimes awned, sometimes suppressed. Inflorescence a panicle. [Contains redtop and timothy.] 5. CHLORIDED, grama tribe Spikelets 1- to several-flowered, sessile on one side of a con- tinuous rachis. All but the lowest floret com- monly sterile and variously modified. ’ Inflorescence of two to several 1-sided spikes, solitary, race- mose or digitate. [Contains grama-grasses.] 6. Nazim, curly-mesquite tribe. Spikelets 1- or 2-flowered, in fascicles, the whole falling from the main axis entire. [Not a natural tribe, some of the genera included not being closely related.] AAAG Fie. 88. SUMMARY OF PRIMARY CHARACTERS 97 7. PHALARIDEZ, canary-grass tribe Spikelets with one perfect floret and two sterile florets below, these falling attached to the fertile floret. Inflorescence a panicle. [Contains canary-grass and sweet vernal-grass.] 8. OryzEm, rice tribe Spikelets 1-flowered, falling from the pedicel entire. Glumes reduced or suppressed. Inflorescence a panicle. [Contains rice.] 9. ZizANtE#, Indian rice tribe Plants moneecious. Spikelets 1-flowered, falling entire. Glumes suppressed. Inflorescence a panicle. [Wild or Indian rice.] Fie. 89. 98 FIRST BOOK OF GRASSES SERIES II.—PANICATA Spikelets dorsally compressed, falling entire, singly or together with parts of the axis. 10. Pantcem, millet tribe Spikelets with one perfect ter- minal floret and a sterile floret below. Rachilla joints very short. Glumes membranaceous, the first small, suppressed in some genera; sterile lemma like the second glume, the (| two appearing like a pair of glumes. Fertile lemma and palea in- durate. Inflorescence a panicle or of one to many racemes, these digitate or racemose on the main axis. [Contains millets and crab- grasses.] SUMMARY OF PRIMARY CHARACTERS 11. ANDROPOGONE, sorghum tribe Spikelets paired, one perfect and sessile, the other sterile and pedicellate, borne on a jointed rachis. Fertile spikelets with one per- fect terminal floret and a _ sterile lemma below; falling with joints of rachis and sterile pediceled spikelet at- tached. Glumes indurate, inclosing the florets; lemmas very thin; palea often suppressed. {Contains sugar-cane, Johnson- grass and broom-sedges.] 12. Tripsacem®, corn tribe Plants moncecious. Pistillate spikelets with one perfect terminal floret and a sterile lemma below, falling embedded in the joints of the rachis. Persistent on the thick compound axis in Zea.) Staminate spikelets 2-flowered in pairs, racemose. |Contains gama-grass and Indian corn.| 99 GENERAL INFORMATION FOR THE BEGINNER It is hoped that after completing the twelve les- sons the student will be eager to study and identify the grasses of his region. To aid the beginner in the study of the plant, the following outline is offered, suggesting points to be observed: OUTLINE FoR Stupy or A Grass PLANT Duration—Annual: Winter annual, summer annual. Perennial: Without rhizomes (with or without winter rosette) ; with rhizomes (these short or long, thick, knotty, or slender). Habit—Hrect, ascending, prostrate, geniculate, creeping, rooting at nodes, or stoloniferous. Tufted or culms few or solitary. Culm—Height; slender or robust; simple or branching (from upper nodes, lower nodes, all nodes); glabrous, scabrous, or pubescent (throughout or below nodes or inflorescence). Nodes: Glabrous or pubescent (hairs appressed, spreading, retrorse, i. e., pointing backward). Sheaths: Close, loose, or spreading; glabrous or pubescent (note margin and summit); open (as common) or closed. Ligule: Length; membranaceous; hairy (stiff or delicate); obsolete. Blades: Erect or spreading, length and width. Shape: Rounded, clasping or narrowed at base; flat, folded, or involute. Texture: Thin, thick, rigid, lax. Surface: Glabrous, pubescent (pilose, villous, hispid), scabrous (note each surface and margin). 100 GENERAL INFORMATION 101 Inflorescence—Terminal only on main culm and branches, or axillary also. Simple or compound. Panicle, size (length and width), loose or compact, few-flowered or many- flowered, nodding or erect. Raceme, size, etc. Spike, | size, etc. Axis of panicle: Size, pubescence, etc.; branches solitary, fascicled or whorled, flexuous or stiff, ascending, spread- ing or reflexed. Axis of raceme or spike: Continuous or disjointing; slender, stout, narrow, winged; pubescence. Spikelet—Pediceled or sessile; laterally or dorsally compressed; falling entire, alone, or with joints of axis; florets falling from the glumes; number of florets. Glumes: Similar or unlike; size (compared to spikelet, one- half, one-third, etc.); shape, awned, toothed, etc.; tex- ture, nerving, pubescence. Lemma: Fertile or sterile, size, shape, texture, pubescence. Palea: Flower bearing or empty, size (sometimes obsolete), shape, texture. It will further the student’s self-training if he writes down the characters of the first few grasses studied and of any particularly puzzling or interesting later ones, making drawings of the spikelet and its parts or diagrams of complicated inflorescences. One should not hesitate to make drawings merely because he “‘can not draw.’”’ However crude a sketch may be it is of great value, not only as a record of observa- tions, but for its training in powers of observation. The one who ‘“‘never could draw” is the very one who should make frequent drawings, however crude they may be. Anyone wishing to make a serious study of grasses 102 FIRST BOOK OF GRASSES should prepare an herbarium, that is, a collection of plants pressed and dried in such a way that they may be mounted by gummed straps to sheets of heavy paper and arranged in folders for reference. Grasses © so prepared keep indefinitely. A circular giving ‘Directions for preparing herbarium specimens of Grasses” (Bureau of Plant Industry Document 442, 1909) may be obtained without charge on request to the office of the Systematic Agrostologist, Bureau of Plant Industry, United States Department of Agriculture, Washington, D. C. In studying grasses the first thing is to understand the structure of the plant in hand, particularly its inflorescence; the second is to learn its relationship to other grasses and what name has been applied to it. The latter is what is meant by ‘‘identifying”’ or ‘“‘determining”’ a plant. Practically all the grasses of the United States are described in one or more manuals of botany or state floras. Most of these manuals give ‘‘keys” to genera and species. [The word key is used figuratively—an instrument by means of which a way is opened.] A key is an ar- rangement of contrasting characters by which, choosing one and rejecting the other, the student is led to the name which applies to the plant in hand. For example, take Figs. 23, 27, 45, 48, 65, and 76, and trace them through the following key, read- ing both of the lines having like indention, these giving the contrasting characters from which to choose. GENERAL INFORMATION 103 Spikelets laterally compressed, the florets falling from the per- sistent glumes. Inflorescence a panicle, spikelets pediceled. Plants dicecious; spikelets several-flowered. ... . DIsTICcHLIs. Plants perfect; spikelets 1-flowered............... Srrpa. Inflorescence a spike, spikelets sessile. Spike solitary; spikelets several-flowered, borne on opposite Meee OL CHG TARNIEC ey mes coe ond dad Oe hee a Louium. Spikes several , digitate; spikelets 1-flowered, borne on one GG OF. ChE TASHIE S86 ys aid ys oes ae lee wes CaPRIOLA. Spikelets dorsally compressed, the spikelets falling entire or to- gether with joints of the rachis. Spikelets all alike, solitary (borne singly) on a continuous PROMISE 566 aes ene EUS ER SRO Ok ane AXONOPUS. Spikelets of two kinds borne in pairs, one sessile and per- fect, the other pediceled and sterile, the rachis articu- PEE ay ORS ith hs ies Saree ay Teg ma ANDROPOGON. In actual work, of course, one never has so simple a problem as that. With exceptions (such as are noted pages 44 and 59) and closely related genera and species, it often demands careful weighing of all the facts to use an extended key successfully, 'but it is excellent training in judgment. The name reached ‘in the key is verified or rejected by reading the description in the text and noting how the characters of the plant agree with those specified. When the description does not apply to the plant in hand, one must return to the key and try again. 104 FIRST BOOK OF GRASSES BOOKS The plants one obtains for himself, and he studies them by means of the few tools mentioned before (p. 6). To identify the plants requires books. A recently published work, “‘The Genera of Grasses of the United States,” by A. 8. Hitchcock, Bulletin 772, United States Department of Agricul- ture, giving keys to tribes and genera, descriptions and illustrations of all the genera, and notes on the more important species, as well as indications of exceptions, most helpful to the beginner, is the first book which the student should acquire. It may be obtained from the Superintendent of Documents, Government Printing Office, Washington, D. C., at 40 cents a copy. The farmer, the agricultural student, or the botanist who wishes only to be able to identify any conspicuous or useful, weedy or injurious grass of his neighborhood requires nothing more than this and the manual of botany covering his region. The following are the current manuals: From Maine to Virginia and west to Minnesota and Missouri, inclusive: Gray’s New Manual of Botany, 7th edition, 1908 (grasses, by A. S. Hitch- cock) ; Britton’s Manual of the Flora of the Northern States and Canada, 3d edition, 1907 (grasses, by G. V. Nash), extending west to the western boundary of Kansas and Nebraska; Britton and Brown’s Illustrated Flora of the Northern United States, Canada, and the British Possessions, second edition, GENERAL INFORMATION 105 1913, 3 volumes (grasses, by G. V. Nash, in vol- ume 1). From North Carolina to Florida and west to Oklahoma and Texas: Small’s Flora of the South- eastern United States, 1903 (grasses, by G. V. Nash). The Rocky Mountain region and east to the Black Hills of South Dakota: Coulter and Nelson’s New Manual of Botany of the Central Rocky Mountains, 1909 (this does not cover the Mexican flora that extends into southern New Mexico and Arizona); Rydberg’s Flora of the Rocky Mountains and Adja- cent Plains, 1917. There is no manual covering the flora of the South- west, but Wooton and Standley’s Flora of New Mexico (Contributions United States National Her- barium, Vol. 19, 1915), includes also most of the species of Arizona. This work gives keys to the genera and species and descriptions of the genera but not of the species. California: Jepson’s Flora of California, 1912. (Grasses, by A. S. Hitchcock). Washington: Piper’s Flora of the State of Wash- ington (Contributions United States National Her- barium, Vol. 11, 1906), giving keys but not descrip- tions; Piper and Beattie’s Flora of the Northwestern Coast, 1915, with keys and descriptions; and Piper and Beattie’s Flora of Southeastern Washington and adjacent Idaho, 1914, with keys and descriptions. Oregon, Utah, and Nevada, and most of Idaho 106 FIRST BOOK OF GRASSES are not yet covered by manuals, but manuals of ad- joining regions include most of the species. Besides these general manuals there are: Hitch- cock’s Text-Book of Grasses, 1914, with illustrations, treating of economic grasses and their uses and also of the morphology and the classification of grasses; Grasses of Iowa, Part II, by Pammel, Ball, and Scribner (Iowa Geological Survey, Supplementary Report, 1905), with keys, descriptions, and’ illustra- tions; Grasses of Illinois, by Edna Mosher (Univer- sity of Illinois Agricultural Experiment Station Bulletin 205, 1918), with keys, descriptions, and illustrations; Manual of Farm Grasses, by A. 8. Hitchcock, 1921. A helpful little bulletin is Lyman Carrier’s ‘‘Iden- tification of Grasses by Their Vegetative Char- acters’? (Bulletin 461, United States Department of Agriculture, to be obtained from the Superintendent of Documents, Government Printing Office, Wash- ington, D. C., for 5 cents). One should always have the inflorescence for accurate identification, but there are occasions when one would be glad to iden- tify a seedling if possible. One who finds interest and delight in the study of grasses will want to accumulate a working library. A bibliography of even the most important works would be too long to be included here and would be out of place in a first book of grasses. How- ever, it may be well to mention a number of pa- pers on grasses that have been issued in the past GENERAL INFORMATION 107 few years by the United States Department of Agriculture and the United States National Her- barium which can be purchased for a small amount from the Superintendent of Documents or, the few out of print, from dealers in second-hand books. Hitchcock, A.S. North American Species of Leptochloa. U.S. Dept. Agr., Bur. Pl. Ind. Bull. 33, 24 pp., with plates. 1903. North American Species of Agrostis. U.S. Dept. Agr., Bur. Pl. Ind. Bull. 68, 68 pp., with plates. 1905. Mexican Grasses in the United States National Herbarium. Contr. U. 8. Nat. Herb. 17, pp. 181-389. 1913. The North American Species of Ichnanthus; The North American Species of Lasiacis. Contr. U. 8. Nat. Herb. 22, 31 pp., with plates. 1920. Revisions of North American Grasses: Isachne, Oplis- menus, Echinochloa, and Chaetochloa. Contr. U. S. Nat. Herb. 22, pp. 115-208, with text figures. 1920. and Chase, Agnes. The North American Species of Panicum. Contr. U. 8. Nat. Herb. 15, 396 pp., with text figures. 1910. Tropical North American Species of Paniear. Contr. U. S. Nat. Herb. 17, pp. 459-539, with text figures. 1915. Grasses of the West Indies. Contr. U.S. Nat. Herb. 18, pp. 261-471. 1917. Griffiths, David. The Grama Grasses: Bouteloua and Related Genera. Contr. U.S. Nat. Herb. 14, pp. 343-428, with plates. 1912. Merrill, Elmer D. North American Species of Spartina. U.S. Dept. Agr., Bur. Pl. Ind. Bull. 9, 16 pp. 1902. Piper, C. V. North American Species of Festuca. Contr. U. S. Nat. Herb. 10, 48 pp., with plates. 1906. 108 FIRST BOOK OF GRASSES Scribner, F. L., and Merrill, E. D. The Grasses of Alaska. Contr. U. 8. Nat. Herb. 13, pp. 47-92. 1910. Shear, C. L. A Revision of the North American Species of Bromus Occurring North of Mexico. U.S. Dept. Agr. Div. Agrost. Bull. 23, 66 pp., with text figures. 1900. Chase, Agnes. The North American Species of Brachiaria; The North American Species of Cenchrus. Contr. U. 8. Nat. Herb. 22, pp. 33-77, with text figures. 1920. The North American Species of Pennisetum. Contr. U.58., Nat. Herb. 22, pp. 209-234, with text figures. 1921. BOTANICAL NAMES In the introduction, the reasons were given for using Latin names of plants. It will have been noticed that these names are made up of two words, the generic name (a noun) and the specific (an adjective or a noun in the possessive case or in apposi- tion), and that the generic name is placed first, like the surnames of persons in a directory. Both words of the name are generally supposed to refer to some characteristic or property of the plant to which it is applied, as Lepturus (slender tail) cylindricus (cylindric) and Erianthus (woolly flower) saccharoides (like Saccharum, that is, sugar-cane), but often botanical names do not fit any better than do names of persons—Paul (meaning small) Baker ~ may be a tall blacksmith, or Martha (meaning bitter) Stern may be sweet and gentle. Many of the Lin- neean genera bear the ancient classic Greek or Latin names, such as Quercus for the oak and Ulmus for the elm. In many other cases Linnzeus used classic GENERAL INFORMATION 109 names but applied them, not as anciently used, but to a different group. Bromus to the Greeks was the oat, but Linnzus used it for the brome-grasses; Zizanion was the tares (supposed to be Lolium temulentum) sown by the enemy among the wheat, in the parable of Scripture, but Linnzeus used the feminine form of the word for our wild rice. On the whole, however, botanical names are more or less descriptive, and it is helpful as well as interesting to be mindful of their meaning, especially of the names of species. Any plant with the name ‘‘asper”’ or ‘“‘scaber’’ will be rough in some part; ‘‘pubescens”’ will be hairy; “‘alba’”’ will have white flowers or bark probably, and ‘‘rubra” red. The name Bromus secalinus implies that the species was a weed in the rye (Secale) fields of Europe, as it is in our wheat fields. Holcus halepensis was known in the early ages from Aleppo (Haleb) in Syria, and Phalaris canariensis from the Canary Islands. The present system of botanical nomenclature dates from 1753, when Linnzeus’s Species Plantarum was published, using binomials (names of two words) for all the species. Before this, plants were given phrase names, more or less descriptive, such as ‘Panicum with lax drooping panicle, the sheaths of the leaves pubescent,”’ for proso millet (Panicum miliaceum) or ‘“‘Grass with a very long spike, like cat-tail,” for timothy. These phrase names were in Latin, of course. Linneus’s binomial plan so simplified the hitherto cumbersome system that 110 FIRST BOOK OF GRASSES it was everywhere adopted in less than a gen- eration. In botanical works it will be noted that plant names are followed by the initial, abbreviation, or the full name of a person, as Poa pratensis L., Phragmites communis Trin. This stands for the name of the person who gave the plant the name, L. for Linneus, Trin. for Trinius. It has often hap- pened that the same species has been given different names by men working in various places and not in touch with each other. In such cases the name first given is the one now generally used. It has also happened that two species in a genus have been given the same name. In this case the name stands for the earlier one and the second is renamed. These superabundant names constitute what is termed synonymy. (Synonyms are two or more names for the same thing.) Superabundant names are also due to the fact that many species when first de- scribed were placed in genera to which more intensive study shows they do not belong. Triodia flava, for example, was first named Poa flava by Linnzus, and Danthonia spicata was named Avena spicata by him. When these species are placed in Triodia and Dan- thonia, respectively, the name of the original author is given in parenthesis with the name or abbreviation of the person who made the change following it; thus, Triodia flava (L.) Hitche., Danthonia spicata (L.) DC, GENERAL INFORMATION 111 CLASSIFICATION OF PLANTS As stated in the introduction (page 2), the clas- sification of grasses is based on the characters of the spikelet. The classification of plants is itself but a human attempt to show natural relationship, and the attempt is based on knowledge of but an infinitesimal part of the plant kingdom. The plants occupying the earth today are the survivors of millions of generations. Countless forms have become extinct. Some of these are known from fossils, but far more have vanished, leaving no record. Linnzus said “Nature never makes jumps.” Connecting links exist or have existed between the greatest extremes. When no such link is known, we conjecture that the missing intermediates are among the countless ex- tinct forms. Human minds approaching a given problem from various angles form different con- jectures; hence it is, that botanists of different periods or even of the same period, have diverse ideas of relationships. The species that Linnzus described under Poa, for example, are today recog- nized as belonging in three genera and his species of Panicum in nine genera. As a vastly greater number of plants are known today, botanists have wider knowledge on which to base conclusions, but the main difference between the Linnean and the present-day idea of genera is due to the modern concept of a genus as a network of related species as contrasted with the earlier concept of a genus as a 118. FIRST BOOK OF GRASSES sort of receptacle built of certain characters and open to any species having those characters. Lin- neeus’s idea of Cenchrus, for example, was primarily a genus of bur-grasses, and, besides what is now recog- nized as Cenchrus (represented by Figs. 70 and 71), he included in it a species of Nazia (Fig. 54) and a Mediterranean grass in which the ‘‘bur”’ is composed of the rigid lobes of the lemmas of the dense head of spikelets, as well as a plant that is not a grass. This brief explanation is offered to save the begin- ner undue bewilderment when he finds in using different books that the standard excuse offered by botanists of all times and places, that the Latin name of a plant is the same throughout the world, is not as true as could be wished. It will be noted in using botanical works that cer- tain characters are accepted as generic, that is, as indicating that species having these characters in common belong to a single genus. Such are the 5-nerved keeled lemmas of Poa. Other characters, sometimes quite as conspicuous, such as the pu- bescence or want of it in Poa, and the presence or absence of an awn in Bromus and Festuca, are re- garded as specific, that is, as differentiating species. It might seem as though by some revelation certain characters are known to be generic and others to be specific. Such is not the case. A species consists of a group of individuals presumably capable of freely inter-breeding. A genus is a group of species which in the sum total of their characters are so much alike GENERAL INFORMATION 113 as to warrant the assumption that they have had a common ancestor. Characters taken as generic and specific are analogous to theories or working hypoth- eses. We use them as far as they work; if they do not work, we discard them for other characters. Botany is a science of living things, and its problems can never be settled once for all. It is this enduring interest that makes it so fascinating. POSITION OF GRASSES IN THE PLANT KINGDOM Flowering plants (excluding the pines and their allies) fall into two rather distinct groups, (1) mono- cotyledons, characterized by an embryo having a single cotyledon and by stems having woody fibers not in layers but distributed through them (as seen in the cornstalk) and not increasing in thickness by age, and (2) dicotyledons, with an embryo having two cotyledons and stems with their woody fibers forming a zone between pith and bark and increasing in thickness by annual layers. Cotyledons are the seed leaves. Anyone will have observed that sprout- ing corn, rye, and other grasses send up a single leaf first, while squash, radishes, lettuce, and morning- glories, for example, have a pair of opposite seed leaves. Grasses belong in the first class, with sedges, rushes, lilies, and the like. They form a highly specialized family with a greater number of species than any other except the orchids and the com- posites (asters, dandelions, thistles, and their kinds). we L oy — 114 FIRST BOOK OF GRASSES Grasses have been so successful in the struggle for existence that they have a wider range than any other © family, occupying all parts of the earth and exceed- ing any other in the number of individuals. They reach the limits of vegetation (except for lichens and alge) in the polar regions and on mountain tops, endure both cold and torrid desert conditions, form the main part of the vegetation of vast prairies, plains, savannas, and steppes of both hemispheres, and occupy great stretches of marsh and tide flats, where they are building up dry land. Bamboos, the largest of grasses, form extensive forests and dense jungles. Grasses range in height from less than an inch, full grown, to over a hundred feet, and they have developed all manner of contrivances for the dissemination of their seeds. GRASSES IN RELATION TO MAN. Of all plants, grasses are by far the most important toman. The grains of wheat, barley, rye, oats, rice, corn, sorghum, and millet form the staple food of the greater part of mankind, while the animals that furnish food and labor, wool, mohair, and leather live principally on grasses. The grains are also sources of starch, alcohol, and glucose. Sugar and sirup are obtained from sugar-cane and varieties of sorghum, and of late years enormous quantities of cooking oil are secured from the germ of the corn. Grasses furnish the material for brooms and brushes GENERAL INFORMATION 115 and are an important source of fiber for paper-making and cordage. Tons of essential oils used in per- fumeries are annually extracted from grasses. In a great part of Asia and Oceanica bamboo forms the principal or only timber for buildings and bridges and furnishes the material for all manner of tools, utensils, and furniture. Achyrodes, 61 aureum, 31 Aigopogon, 62 tenellus, 61 Agropyron repens, 34 Agrostider, 96 Agrostis, 46, 47 hiemalis, 47 Alopecurus, 49 geniculatus, 48 Amphicarpon Purshii, 77 Andropogon, 83 scoparius, 82 Andropogonee, 99 Anthers, 9 Anthoxanthum, 64, 65 odoratum, 64 Aristida, 51, 52 dichotoma, 51, 52 Arrhenatherum elatius, 45 Avena fatua, 42 Avenez, 95 Awns, 22, 35 Axonopus furcatus, 73 Bamboos, 114, 115 Barley, cultivated, 40 tribe, 95 wild, 39 Barnyard-grass, 73 INDEX Beckmannia, 59 Bermuda-grass, 14, 55, 93 Binomials, 109 Blade, 8, 9, 16 Blue-grass, Kentucky, 13, 15, 26 Botanical names, significance of, 108 Bouteloua, 57, 58 curtipendula, 57, 58 gracilis, 58, 93 Brachiaria, 73 Bromus, 25, 26 secalinus, 25 Broom sedges, 82 Buffalo-grass, 60, 92, 93 Bulbilis dactyloides, 92 Bunch grasses, 13 ' Calamagrostis, 46, 47 canadensis, 46 Callus, 22: Campulosus, 59 Canary-grass, 65 reed, 65 tribe, 97 Capriola Dactylon, 55 Caryopsis, 10 Cathestechum, 59 Cenchrus, 75 117 118 Cenchrus, continued bur, 93 myosuroides, 76 pauciflorus, 76 Cheetochloa, 75 Grisebachii, 74 lutescens, 75 Cheat, 75 Chess, 75 Chloride, 96 Chloris, 57 latisquamea, 56 Cinna, 48 Classification, 111 Cleistogamous spikelets, 77 Corn, Indian, 90 tribe, 99 Cottea pappophoroides, 27, 28 Couch-grass, 34 Crab-grass, 70 Culms, 8, 14 Curly-mesquite tribe, 96 Cynosurus cristatus, 30 Danthonia spicata, 43 Dieecious, 23, 32 Distichlis spicata, 32 Echinochloa Crusgalli, 73 Eleusine, 54, 55, 56 indica, 54, 55 Elymus, 39, 41 virginicus, 38 Embryo, 10, 11 Emmer, 36 Endosperm, 11 INDEX Erianthus, 82 saccharoides, 81 Eriochloa punctata, 73 Fescue tribe, 95 Festuca ovina, 26 Festucez, 85 Filaments, 9 Floret, 10, 11 Flower, of grass, 9 typical, 9 Foxtail, green, 75 yellow, 75 Fruit, 11 Gama-grass, 14, 89 Genus, 5 Genera, 5, 111 Generic characters, 112 Glumes, 11, 17 Grain, 10, 11 Grama-grass, 57, 92, 94 tribe, 96 Graminezx, 8 Grass, description of, 8 flower, 10 Grasses, bunch, 13 in relation to man, 114 position of, in plant king- dom, 113 sod-forming, 13 Green foxtail, 75 Herbarium, use of a, 102 Heteropogon, 87, 88 contortus, 86 INDEX Hilaria Belangeri, 62 Hilum, 11 Holcus, 83, 85 halepensis, 80 Holy-grass, 65 Homalocenchrus, 65 oryzoides, 66 Horde, 95 Hordeum, 40 nodosum, 39, 40 Indian corn, 90, 91 -grass, 81 rice tribe, 97 Inflorescence, 18 Johnson-grass, 80 Kentucky blue-grass, 13, 15, 26 Leaves, 15 Lemma, 10, 11, 17 Leptochloa, 59 Lepturus, 84 cylindricus, 36 Ligule, 8, 9 Lodicules, 9, 10 Lolium multiflorum, 36 Luziola, 67 Lycurus phleoides, 49 Maize, 90 Manisuris, 83, 84, 90 cylindrica, 83 Manuals of botany in current use, 104 119 Melica mutica, 29 Milium effusum, 50 Millet tribe, 77, 98 Millets, 75 Moncecious, 23 Morphology, 2 Muhlenbergia foliosa, 50 Schreberi, 50 Nazia aliena, 63 Naziez, 96 Nodes, 8 Notholceus lanatus, 44, 49 Oat, 45 tribe, 95 Oats, wild, 42 Oryza sativa, 65, 66 Oryzee, 97 Oryzopsis, 50 racemosa, 51 Ovary, 9, 10 Ovules, 10 Palea, 10, 11 Panicate, 98 Panicee, 77, 98 Panicle, 18 Panicularia septentrionalis, 27 Panicum, 69, 70, 71 miliaceum, 69 Pappophorum vaginatum, 29 Paspalum, 71, 73 leve, 71 Pedicel, 18 Pericarp, 47 120 Petiole, 16 Phalaridexz, 97 Phalaris arundinacea, 65 canariensis, 65 minor, 65 Phleum pratense, 48 Phragmites communis, 28 Pistils, 9, 10 Pleuropogon californicus, 27 Poa, 26 pratensis, 14, 26, 27 Poaceze, 8° Poate, 95 Prophyllum, 17 Quack-grass, 15, 34 Raceme, 18 Rachilla, 11 Rachis, 18 Reed, 28 canary-grass, 65 Reimarochloa, 72 Rhaphis, 85, 88 pauciflora, 85, 86 Rhizomes, 13, 14 Rice, 65, 66 tribe, 97 wild, 66 Rootstocks, 13 Rye-grass, 38 Rytilix, 84, 85 granularis, 84 Saint Augustine-grass, 72 Salt-grass, 32 Sand-bur, 75, 76 INDEX Scleropogon brevifolius, 32, 33 Seneca-grass, 65 Sheath, 8, 9 Sitanion, 39 Sod-forming grasses, 13 Sorghastrum, 81 Sorghum, 80, 82 tribe, 99 Spartina, 59 Spathes, 82 Sphenopholis, 45 Spike, 18 Spikelet, 11, 17, 19 Sporobolus airoides, 47 heterolepis, 48 Stamens, 9, 10 Stenotaphrum, 72, 73 secundatum, 72 Stigma, 9, 10 Stipa, 51, 52, 88 spartea, 51 Stolons, 14 Styles, 9, 10 Sugar-cane, 81 Sweet vernal-grass, 64 Synonyms, 110 Syntherisma, 73 sanguinalis, 70 * Timothy, 48 tribe, 96 Torresia, 65 Tribes, primary characters of, 95 Triodia flava, 27, 28 Tripsacese, 99 INDEX 121 Tripsacum, 90 Wheat, 35 dactyloides, 14, 89 Wild oats, 42 Trisetum, 43, 46 rice, 66 spicatum, 43 Yard-grass, 54 ; Triticum zstivum, 35 Yellow foxtail, 75 dicoccum, 36 Zea Mays, 90, 91 Vanilla-grass, 65 Zizania palustris, 66 Velvet-grass, 44 . Zizaniee, 97 Zizaniopsis miliacea, 67 Cat. No. 1137 _ PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY ike; ? ort, ‘ rele aie. as ee