Encyclopædia Britannica, Ninth Edition/Grasses

by Henry Trimen

GRASSES (Gramineæ—Gramina) are monocotyledonous flowering-plants, possessing certain characters in common, and constituting the order Graminece. No plant is correctly termed a grass which is not a member of this family, but the word is in common language also used, generally in combination, for many plants of widely different affinities which possess some resemblance (often slight) in foliage to truly graminaceous species; e.g., knot-grass (Polygonum aviculare), cotton-grass (Eriophorum), rib-grass (Plantago), scorpion-grass (Myosotis sea-grass (Zostera). In agriculture the word has an extended signification to include the various fodder-plants, chiefly leguminous, often called "artificial grasses" (see AGRICULTURE). Indeed, formerly grass (also spelt goers, gres, gi/rs, in the old herbals) meant any green herbaceous plant of small size. Yet the first attempts at a classification of plants recognized and separated a group considered even of primary value of Gramina, and this, though bounded by nothing more definite than habit and general appearance, contained the Gramineo? of modern botanists. The older group, however, even with such systematists as Ray (1703), Scheuchzer (1719), and Micheli (1729), embraced in addition the Cyperacece, Juncacece, and some other monocotyledons with inconspicuous flowers. Singularly enough, the sexual system of Linnaeus (1735) served to mark off more distinctly the true grasses from these allies, since very nearly all of the former then known fell under his Triandria Digyniii, whilst the latter found themselves under other of his artificial classes and orders.

I. Structure.—The general type of true grasses is familiar in the cultivated cereals of temperate climates— wheat, barley, rye, oats, and in the smaller plants which make up our pastures and meadows and form a principal factor of the turf of natural downs. Less familiar are the grains of warmer climes—rice, maize, millet, and sorgho, or the sugar cane. Still further removed are the bamboos of India and America, the columnar stems of which reach to the height of forest trees. All are, however, formed on a common type, which we proceed to examine.

Root.—Most cereals and many other grasses are annual, and possess a tuft of very numerous slender root-fibres, much branched, and of great length. The greater part of the order are of longer duration, and have the roots also fibrous, but fewer, thicker, and less branched. In such cases they are very generally given off from just above each node (often in a circle) of the lower part of the stem or rhizome, perforating the leaf-sheaths. In some bamboos they are very numerous from the lower nodes of the erect culms, and pass downwards to the soil around them, whilst those from the upper nodes shrivel up and form circles of spiny fibres.

Stem.—The root-stock of perennial grasses is usually well developed, and often forms very long creeping or sub terranean rhizomes, with elongated internodes and sheathing scales ; it is also frequently short, with the nodes crowded. The very large root-stock of the bamboos (fig. 1) is a striking

Fig. 1.—Rhizome of BambooA, B, C, D, successive series of axes, the last bearing aerial culms.

example of "definite" growth ; it is much branched, the short thick curved branches being given off below the apex of the older ones and at right angles to them, the whole forming a series of connected arched axes, truncate at their ends which were formerly continued into leafy culms. The root-stock is always solid, and has the usual internal struc ture of the monocotyledonous stem. The aerial leaf-bearing branches (culms) are a character istic feature of grasses. They are generally numerous, erect, cylindrical (rarely flattened), and conspicuously jointed with evident nudes. The nodes are solid, a strong plate of tissue passing across the stem, but the internodes are commonly hollow, although examples of completely solid stems are not uncommon (many Andropogons, sugar cane). The general character is thus not unlike that of Umbelliferce. The exterior of the culms is more or less concealed by the leaf-sheaths ; it is usually smooth and of tan highly polished, the epidermal cells containing an amount of silica sufficient to leave after burning a distinct skeleton of their structure. Tabasheerisa white substance mainly composed of silica, found in the joints of several bamboos. A few of the lower internodes may become en larged and sub-globular, forming nutriment-stores, and grasses so changed are termed "bulbous" (Arrhenatherum, Poa bulbosa, <fcc.). In internal structure grass-culms, save in being hollow, conform to that usual in monocotyledons ; the vascular bundles run parallel in the internodes, but a horizontal interlacement occurs at the partitions. Nearly all grasses branch to some extent, and many extensively; the branches are strictly distichous, and many buds are abortive, or the resulting branches short and stunted. In many bamboos they are long and spreading or drooping and copiously ramified, in others reduced to hooked spines. Dinochloa is truly scandent, and climbs over trees 100 feet or more in height. Olyra latifolia is also a climber on a humbler scale. Grass-culms grow with great rapidity, as is most strik ingly seen in bamboos, where a height of over 100 feet is attained in from two to three months, and many species grow two, three, or even more feet in twenty-four hours. Silicic hardening does not commence till the full height is nearly attained. The largest bamboo recorded is 170 feet, and the diameter is usually reckoned at about 4 inches to each 50 feet. A specimen in the British Museum is over 8 inches in diameter. In the same collection are some remarkable monstrosities, in which the septa are oblique and the inter nodes triangular or rhomboidal ; Kurz has also figured one in the Calcutta Museum in which the cavities are confluent into a continuous spiral.

Leaves.—These present special characters usually suffi cient for ordinal determination. They are always solitary at each node and strictly distichous, and consist of two dis tinct portions, the sheath and the blade. The sheath is often of great length, and generally completely surrounds the culm ; as a rule it is split down its whole length, thus differing from that of Cyperacece, which is almost invariably (Eriospora is an exception) a complete tube. In a few grasses (fifelica, Bromus] the edges are soldered together as in the latter order. The sheaths are much dilated in Alopecurus vaginatus and in a species of Potmaochloa, in the latter serving as floats. At the summit of the sheath, above the origin of the blade, is the liyide, a usually mem branous process of small size (occasionally reaching an inch in length) erect and pressed around the culm. It is rarely quite absent, but may be represented by a tuft of hairs (very conspicuous in Pariana). Melica imiflora possesses, in addition to the ligule, a green erect tongue- like process, from the line of junction of the edges of the sheath.

The blade is frequently wanting or small and imperfect in the basal leaves, but in the rest is long and sot on to the sheath at an angle. The usual form is familiar, sessile, more or less ribbon-shaped, tapering to a point, and entire at the edge. The chief modifications are the articulation of the (deciduous) blade on to the sheath, which occurs in all the Bambusecc (except Planolia) and in Spartina siricfa, and the interposition of a petiole between the sheath and the blade, as in Leptaspis, P/iarus, Pariana, Lophatherinn, &c. In the latter case the form of the leaf usually becomes oval, ovate, or even cordate or sagittate, but these forms are found in sessile leaves also (Olyra, Paniciim}. The venation is strictly parallel, the midrib usually strong, and the other ribs more slender. In Anotnocliloa there are j several nearly equal ribs, and in some broad-leaved grasses

(fiambns t >a>, Pharus, Leptaspis) the venation becomes tesse-

lated by transverse connecting veins. The thick prominent veins in Agropyrum occupy the whole upper surface of the ! leaf. Epidermal appendages are rare, the most frequent I being marginal, saw-like, cartilaginous teeth, usually minute, but occasionally (Danthonia scabra, Panicum serratnm) so large as to give the margin a serrate appearance. Alope curus lanatns and one or two Panicums have woolly leaves. Inflorescence.—This possesses an exceptional iirport- ance in grasses, since, their floral envelopes being much reduced and the sexual organs of very great uniformity, the characters employed for classification are mainly derived from the arrangement of the flowers and their investing bracts. The subject also presents unusual difficulties from the various interpretations which have been given to these glumaceous organs and the different terms employed for them by various writers. It may, however, be now con sidered as settled that the whole of the bodies known as glumes and palese, and distichously arranged externally to the flower, form no part of the floral envelopes, but are of the nature of bracts. These are so arranged round the small flowers as to form spikelets (locust), and each spikelet may contain one, two, three, or a greater number of flowers (in some species of Eragrostis nearly 60). The flowers are, as a rule, placed laterally on the axis (rachis) of the spikelet, but in uniflorous spikelets they appear to be terminal, and are probably really so in Anthoxanthum (fig. 9, 2) and in two anomalous genera, Anomochloa and Streptochcete. In immediate relation with the flower itself, and often entirely concealing it, is the palea or pale (" upper pale " of most systematic agrostologists ; " paleola interior," Dumortier; "spathella," Doll). This organ (fig. 3, 1) is peculiar to grasses among Ghimiferce, and is almost always present, certain Oryzece and Phalaridece being the only exceptions. It is of thin membranous consistence, usually obtuse, often bifid, and possesses no central rib or nerve, but is furnished with two lateral ones, one on either side ; the margins are frequently folded in at the ribs, which thus become placed at the sharp angles. This structure points to the fusion of two organs, and the pale was by R. Brown considered to represent two portions soldered together of a trimerous perianth-whorl, the third portion being the " lower pale," to be immediately mentioned. By Bentham the homology of the organ is suggested to be with the two bracteoles found in Hypolytrum pungens and Platylepis, and with the perigynium of the female flower of Car ex in Cyperacece. It is rarely (Triachyrum, Diachy- rium) found split into two separate organs. The flower with its pale is sessile, and is placed in the axil of another bract in such a way that the pale is exactly opposed to it, though at a slightly higher level. It is this second bract which has been generally called by systemat- ists the "lower pale," and with the "upper pale" con sidered to form an outer floral envelope (" calyx," Jussieu ; " perianthium," Brown ; " stragulum," Palisot de Beauvois; " glumella," Dumortier). In the writings of most botanists even though this view is not held, yet, where the term " flower " is employed, it includes these organs. It is, how ever, certain that the two bracts are on different axes, one secondary to the other, and cannot therefore be parts of one whorl of organs. This was made out from the study of so- called "viviparous" grasses, in which the lower pales become transformed into ordinary foliage leaves, first by Von Mohl in 1845, and more clearly by Germain de St Pierre in 1852, who terms the lower pale the "glume fertile." Doll and Bentham have also independently ar rived at the same result, and the latter in 1858 first published the terminology here adopted, and used for the same organ the name flowering glume. The two bracts are usually quite unlike one another, but in some genera (e.g., most Festucacece) they are considerably similar in shape and appearance. The flowering glume has generally a more or less boat- shaped form, is of firm consistence, and possesses a well- marked central midrib and frequently several lateral ones. The midrib in a large proportion of genera extends into an appendage termed the awn (fig. 10, 2), and the lateral veins more rarely extend beyond the glume as sharp points (e.g. t Pappophorum). The form of the flowering-glume is very various, this organ being plastic and extensively modified in different genera. In Leptaspis it is formed into a closed cavity by the union of its edges, and encloses the flower, the styles projecting through the pervious summit. Valu able characters are obtained from the awn. This presents itself variously developed from a mere subulate point to an organ several inches in length, and when complete (as in Andropogone(, Avenece, and Stipece) consists of two well- marked portions, a lower twisted part and a terminal straight portion, usually set in at an angle with the former sometimes trifid and occasionally beautifully feathery. The lower part is most often suppressed, and in the large group of the Panicece awns of any sort are very rarely seen. The awn may be either terminal or may come off from the back of the flowering-glume, and Duval Jouve s observations have shown that it represents the blade of the leaf of which the portion of the flowering-glume below its origin is the sheath ; the twisted part (so often suppressed) cor responds with the petiole, and the portion of the glume extending beyond the origin of the awn (very long in some species, e.g., of Danthonia] with the ligule of the developed foliage-leaf. When terminal the awn Las three fibro-vascular bundles, when dorsal only one ; it is covered with stomate-bearing epidermis. The flower with its palea is thus sessile in the axil of a floriferous glume, and in a few grasses (Leersia (fig. 2), Coleanthus, Nardus] the spikelet consists of nothing more, but usually (even in uniflorous spikelets) other glumes are present. Of these the two placed distichously opposite each other at the base of the spikelet never bear any flower in their axils, and are called the basal or empty glumes (fig. 10, 1). They are the "glumes" of most writers ("palese" of Dumortier), and together form what was called the "gluma" by R. Brown (" tegmen," Palisot de Beauvois). They rarely differ much from one another-, but one may be smaller or quite absent (Panicum (fig. 8, 2), Vulpia,

Fig. 2.—Spikelet of Leersia.

Paspalum, Lolium), or both be altogether suppressed, as above noticed. They are commonly firm and strong, often enclose the spikelet, and are rarely provided with long points or imperfect awns. Generally speaking they do not share in the special modifications of the flower ing glumes, and but rarely themselves undergo modifica tion, chiefly in hardening of portions (Sclerachne, Manisiiris, Antephora, Peltopkorum), so as to afford greater protection to the flowers or fruit. But it is usual to find, besides the basal glumes, a few other empty ones, and these are in two- or more-flowered spikelets (fig. 11, 2) at the extremity (numerous in Lophatherum), or in uniflorous ones (fig. 8, 2) below, interposed between the floral glume and the basal pair. Descriptive writers have been accustomed to call these empty glumes " barren " or " neutral flowers," a misleading use of terms. The axis of the spikelet, when short and rounded, has been termed the calhis, when long the rachilhis. It is fre quently jointed and breaks up into articulations above each flower. Tufts or borders of hairs are frequently present (Calamagrostis, Phragmites, Andropogon), often so long as to surround and conceal the flowers. The axis is often continued beyond the last flower or glume as a bristle or stalk.

Involucres or organs outside the spikelets are not unfre- quent, the morphology of which is various. Thus in Setaria, Pennisetum, &c., the one or more circles of simple or feathery hairs represent abortive branches of the inflor escence ; in Cenchrus these become consolidated, and the inner ones flattened out so as to form a very hard globular spiny case to the spikelets. The cup-shaped involucre of Cornucopice is a dilatation of the axis into a hollow recep tacle with a raised border. In Cynosurus the pectinate involucre which conceals the spikelet is a barren or abortive spikelet. True bracts of a more general character subtend ing branches of the inflorescence are singularly rare in Grarninece, in mirked contrast with Cyperacece, where they are so conspicuous. They however occur in a whole sec tion of Andropogon, in Anomochloa, and at the base of the spike in Sesleria. The remarkable ovoid involucre of Coix, which becomes of stony hardness, white, and polished (then known as " Job s tears "), is also a modified bract or leaf-sheath. It is entirely closed except at the apex, and contiins the female spikelet, the stalks of the male inflorescence and the long styles emerging through the sin ill apical orifice. Any number of spikelets may compose the inflorescence, and their arrangement is very various. In the spicate forms, with sessile spikelets on the main axis, the latter is often dilated and flattened (Paspalum), or is more or less thickened and hollowed out (Stenotaphrum, Rottboellia, Tripsacum), when the spikelets are sunk and buried within the cavities. Every variety of racemose and paniculate inflorescence obtains, and the number of spikelets composing those of the large kinds is often immense. Rarely the inflorescence consists of very few flowers; thus Lygeum Spartum, the most anomalous of European grasses, has but two or three large uniflorous spikelets, which are fused together at the base, and have no basal glumes, but are enveloped in a large hooded spathe-like bract.

Flower.—This is characterized by remarkable uniformity. The perianth is represented by very rudimentary, small, fleshy, hypogynous scales called lodicules (" squamulae,"

Fig. 3—Flowers of Grasses.1, Piptatherum, with the palea; 2, Poa; 3, Oryza.

Kunth ; " nectarium," Schreber) ; they are elongated or truncate, sometimes fringed with hairs, and are in contact with the ovary. Their usual number is two, and they are placed collaterally at the anterior side of the flower, that is, within the flowering glume. They are generally considered | to represent the inner whorl of the ordinary monocoty- ! ledonoas (liliaceous) perianth, the outer whorl of these | being suppressed as well as the posterior member of the inner whorl. This latter is present almost constantly in Stipeoe and Bambusece, which have three lodicules, and in the latter group they are occasionally more numerous (five, spreading and persistent in Pseudostachyum ; six to eight in Beesha). In Anomochloa they are represented by hairs. In Streptochvte, according to Doll, there are six lodicules, alternately arranged in two whorls. They are often quite absent. In some cases lodicules are of the nature of stipules to the palea, and appear as though split off from its sides at the base. Such stipular lodicules often co-exist along with the perianthial ones, and are then either free from of combined with the two anterior ones.

Sexual Organs.—Grass-flowers are usually hermaphrodite, but there are very many exceptions. Thus it is very common to find one or more imperfect (usually male) flowers in the same spikelet with bisexual ones, and their relative position is important in classification. Holcus and Arrhen- a,theruin are examples in English grasses; and as a rule in species of temperate regions separation of the sexes is not carried further. In warmer countries monoecious and dioecious grasses are more frequent. In such cases the male and female spikelets and inflorescence may be very dissimilar, as in the maize, Job s tears, Euchlaena, Spinifex, itc.; and in some dioecious species this dissimilarity has led to the two sexes being referred to different genera (e.g., Antephora axilliftora, Steud., is the female of Buchloe dactyloides, Engelm., and Nenrachne paradoxa, Br., of a species of Spinifex). In other grasses, however, with the sexes in different plants (e.g., Brizopyrum, Distichlis, Era- grostis capitata, Gynerium), no such dimorphism obtains. A mphicarpum is remarkable in having cleistogamic flowers borne on long radical subterranean peduncles which are fertile, whilst the conspicuous upper paniculate ones, though apparently perfect, never produce fruit. Something similar occurs in Leersia oryzoides, where the fertile spikelets are concealed within the leaf-sheaths.

Andrœcium.—In the vast majority there are three stamens alternating with the lodicules, and therefore one anterior, i.e., opposite the flowering glume, the other two being pos- terior and in contact with the palea. They are hypogynous, and have long and very delicate filaments, and large, linear or oblong two-celled anthers, dorsifixed and ultimately very versatile, deeply indented at each end, and commonly exsertecl and pendulous. Suppression of the anterior stamen sometimes occurs (e.g., Anthoxanthum), or the two posterior ones may be absent (Uniola, Cinna, Phippsia, Festuca bromoides). On the other hand there is in some genera (Oryza, most Bambuseai) another row of three stamens, making six in all (fig. 3, 3); and Anomochloa and Telrarrhena possess four. The stamens become numerous (ten to forty) in the male flowers of a few monoecious genera (Pariana, Luziola}. In Beesha they vary from seven to thirty, and in Gigantochloa they are monadelphous.

Gynœcium.—There is but little variation here from a bicarpellary pistil, with a small rounded one-celled ovary containing a single laterally attached or ascending ovule, capped by two styles quite distinct or connate at the base,

Fig. 4—Pistils of Grasses.1, Alopecurus; 2, Bromus; 3, Arrhenatherum; 4, Glyceria; 5, Melica; 6, Mibora; 7, Nardus.

and with densely hairy or feathery stigmaa (fig. 4). Occa sionally there is but a single style (Nardus, Lygeum), and this may attain to a great length (6 inches in the maize); or three styles may be present (some Bambusece, Leptaspis, Streptochaete). Nees has described a case in which three complete carpels were found in Schedonorus elatior. We thus see that, comparing the flower of Graminece with the normal liliaceous plan (fig. 5), it differs in the complete suppression of the outer row and the posterior member of the inner row of the perianth-leaves, of the whole inner row of stamens, and of the anterior carpel, whilst the remain-

Fig. 5—Diagrams of the ordinary grass-flower. 1, actual condition; 2, theoretical, with the suppressed organs supplied. a, axis; b, flowering glume; c, palea; d, outer row of perianth leaves; e, inner row; f, outer row of stamens; g, inner row; h, pistil.

ing members of the perianth are in a rudimentary condition. But each or any of the usually missing organs are to be found normally in different genera, or as occasional de velopments.

Fruit.—The ovary ripens into a usually small ovoid or rounded fruit, which is entirely occupied by the single large seed, from which it is not to be distinguished, tin thin pericarp being completely united to its surface. To this peculiar fruit the term caryopsis has been applied (more familiarly "grain"); it is commonly furrowed longitudinally down one side (usually the inner, but in CW-rand its allies, the outer), and an additional covering is not unfrequently provided by the adherence of the persistent palea, or even also of the flowering glume (" chaff " of cereals). From this type are a few deviations ; thus in Sporobolus, &c. (fig.

Fig. 6—Fruit of Sporobolus, showing the dehiscent pericarp and seed.

6), tli3 pericarp is not unite 1 with the seed but is quite distinct, dehisces, and allows the loose seed to escape. Sometimes the pericarp is mem branous, sometimes hard and brittle, whilst, on the other hand, in some genera of JDambusece, it becomes thick and fleshy, forming a " berry," or rather a drupe. In Meloccuma baccifera this of forms a fruit 3 or 4 inches long, with a pointed beak of 2 inches more ; it is indehiscent, and the small seed germinates whilst the fruit is still attached to the tree, putting out a tuft of roots and a shoot, and not falling till the latter is 6 inches long. A similar germina tion also occurs in Pseudostachyum compactum, which has the largest fruit of the order.

Seed.—The testa is thin and membranous but occasion ally coloured, and the embryo small, the great bulk of the seed being occupied by the hard farinaceous endosperm (albumen) on which the nutritive value of the grain depends.

Fin. 7. A grain of wheat. l,back, and 2, front view; 3, vertical section, showing ( /) the endosperm, and () embryo; 4, commencement of germination, showing (6) the pileola and (c) the secondary rootlets surrounded by their coleorrhizas. The embryo presents many points of interest. Its position is remarkable, closely applied to the surface of the endo sperm at the base of its outer side. This character is abso lute for the whole order, and effectually separates Graminece from Cyperacece. The part in contact with the endosperm is flat and plate-like, and is known as the sciitellum. Some difference of opinion is held on the nature of this ; but it is probably correctly regarded (as by Van Tieghem) as the main portion of the cotyledon, the white epigoeal sheath (pileola) protecting the plumule which is often described as the cotyledon being the ligule only of that organ greatly developed. The radicle is inferior, broad, and blunt ; the primary root is very slightly developed in germination, but several secondary lateral ones burst through above its point, and thus become surrounded at their base with little sheaths (coleorrhizae).

II. Classification.—Gramineæ are thus sharply defined from all other plants, and there are no genera as to which it is possible to feel a doubt whether they should be referred to it or not. The only order closely allied is Cyperacece^ and the points of difference between the two have been alluded to above, but may be here brought together. The best distinctions are found in the position of the embryo in relation to the endo spermlateral in grasses, basal in Cyperacece and in the possession by Graminece of the 2-nerved palea below each flower. Less absolute characters, but generally trust worthy and more easily observed, are the feathery stigmas, the always distichous arrangement of the glumes, the usual absence of more general bracts in the inflorescence, the split leaf-sheaths, and the hollow, cylindrical, jointed culms, some or all of which are wanting in all Cyperacece. The same characters will distinguish grasses from the other glumiferous orders, Restiacece, Enocaulonece, and Desvaiixi- (icece, which are besides further removed by their capsular fruit and pendulous ovules. To other monocotyledonous families the resemblances are merely of adaptive or vege tative characters. Some Commelyuacece and Marantaccce approach grasses in foliage; the leaves of Allium, &c., possess a ligule; the habit of some palms reminds one of the bamboos ; and Juncacece and a few Liliacece possess an inconspicuous scarious perianth. The great uniformity among the very numerous species of this vast family renders its classification very difficult. The difficulty has been increased by the confusion resulting from the multiplication of genera founded on slight characters, and from the description of identical plants under several different genera in consequence of their wide distribution. No characters for main divisions can be obtained from the flower proper or fruit ; though both Kunth and Eeich- enbach have used them especially the form of the styles and stigmas, thelodicules and the caryopsis they have not been generally employed by botanists, who have found it necessary to trust to characters derived from the usually less important inflorescence and bracts. The earlier authors made the general arrangement of the spikelets (spicate, paniculate, &c.) the basis of their classi fications. Palisot de Beauvois s main divisions are founded on the existence in the same inflorescence of similar or diverse spikelets. Fries has proposed a division into Eury- nnthece and Clisantkece, according to the condition of the flowering glume and pale, separated or close during inflor escence. Dumortier gives a classification into Rachidece, Rachilliftorce, and Calliflorce ; the first with the spikelets inserted into excavations of the rachis, the second with the flowers of the free spikelet inserted on its axis (rachillus), and the last with the flower or flowers borne on the short callus of the glumes. Both these systems, and others which have been suggested, possess merits of their own, but they have not as yet been found capable of application to the order as a whole, having been framed for the accommodation mainly of European genera. For such general treatment no better primary divisions have been found than those proposed by Robert Brown in 1810 and further elaborated in 1814, which have been accepted by Munro (1868) and Bentham (1878) in their most recent revisions of the Cape and Australian genera. The basis of Brown s division of the whole order into Panicece and Poacece is the position of the most perfect flower in the spikelet ; this is the upper (apparently) terminal one in the first, whilst in the second it occupies the lower position, the more imperfect ones (if any) being above it. Munro has supplemented this by another character easier of verification, and of even greater constancy, in the articulation of the pedicel in the Panicece immediately below the glumes; whilst in Poacece this does not occur, but the axis of the spikelet frequently articulates above the pair of empty basal glumes. Neither of these two great divisions will well accommodate certain genera allied to Phalaris, for which Brown proposed tentatively a third group (since named P/ialaridece); this, or at least the greater part of it, is placed by Bentham under the Poacece. The following arrangement is based on the writings of these authors, but it cannot be considered very satisfactory. Probably no serial disposition can represent the tangled network of genera related in different ways and various degrees which make up this very natural but complicated family, and though some of the following tribes are fairly natural, the limits of others are but ill-defined, and the position of many genera uncertain. I. PANICACE^E. Pedicel articulated below the glumes. Spikelet with one or two flowers, the more perfect fertile one above, the male or barren one, if present, below it. FIG. 8. Panicaceae. 1, a pair of spikelets of Andropogon; 2, spikeiet of Setaria, with an abortive branch beneath it. 1. Panicece. Spikelets with an apparently terminal hermaphrodite or female flower with or without a male one beneath it. Glumes three or four, the upper flowering one of a firm texture, the lowest usually small, sometimes absent. Awns rarely found ; if present, neither twisted nor kneed. Fruit enclosed in the hardened flowering glume and palea. EXAMPLES : Paspalum, Panicum, Setaria, Pennisetum, An- thephora, Thouarca, Spinlfex. 2. Tristerji iicce. Spikelets as in Panicece, but flowering-glume with a twisted and bent awn. EXAMPLES : Arundinclla, Tristachya, Pohjpogon(t). 3. Andropocjoncce. Spikelets as in the last, but usually in pairs (rarely three or solitary), one being sessile and fertile, the other stalkedandusuallymaleorneuter. Glumes four (rarely fewer), one of the outer ones the largest and enclosing the fruit, tire upper flowering one very thin and transparent, usually bear ing a twisted and bent awn or reduced to the awn. EXAMPLES : Ischcemwm, PoUinin, Andropogon, Imperata, Sorghum, Anthistiria, Erianthus, Eiilalla, Saccharum. 4. RoUloclliece. Spikelets as in the last ; one of each pair fertile, sessile, and sunk in alternate notches or cavities of the jointed simple rachis, the other stalked or absent ; no awns EXAMPLES -.Hemarthria, Manisuris, Rottbocliia, Opliiurus, Pnluna, Lcplarvs. 5. Mayadece. Spikelets monoecious, very unlike, the male numer ous in a terminal panicle, the female few at the base of the male inflorescence or more numerous in a separate one. EXAMPLES : Calx, Zea, Chionachnc, Tripsacum, Sclcrachne. II 6. Olyrccc. Spikelets monoecious, or some hermaphrodite and some male in the same panicle ; the flowering glume large, sometimes utricular. Stamens three, six, or more. Leaves often broad and petiolate. EXAMPLES : Lcptaspis, Diarus, Olyra, Pariana. . PHALAKIDEJE. Pedicel either articulated below the glumes or not, but the rachis of the spikelet articulated above the two lowest ones. Spikelets with (one or) three flowers, the perfect fertile one above, the two male, if present, below it. Two- nerved palea usually absent. EXAMPLES : Phalaris, Anthoxanthum, Ehrharta, Hicrochloe Alopecurus, Phkum, Cryjjsis, Lygcum (?). FIG. 9. Phalaridece. 1, spikelet of HierocJi oe; 2, spikelet of Anthoxanthum. III. POACE.E. Pedicel not articulated below the glumes. Rachis of spikelet often articulated above the two lowest glumes. Spikelets with one, two, or more fertile flowers, the male or imperfect ones (if any) above them. liachis of spikelet often continued as a point or bristle beyond the flowers. 1. Oryzcce. Spikelets one-flowered. Empty basal glumes very small or wanting. Two-nerved palea usually absent. Stamens usually six. EXAMPLES : Oryza, Lcersia, Potamophila, Zizania, Anotno- chloa (?). 2. StipccK. Spikelets one-flowered. Flowering glume with a terminal twisted and bent a vn. Palea small and thin. Lodicules three. EXAMPLES : Stipa, Aristida, Piptathcrum. 3. Agroslidccc. Spikelets one-flowered. Flowering glume usually with a terminal or dorsal bent awn. Palea small and thin. Lodicules two. EXAMPLES: Agrostis, Dcycuxia, Calamagrostis, Cokanthus(t). FIG. lO.Poacea;. 1, spikelet of Aji-usiis; 2, spikelet of Aira. 4. Avcnece. Spikeletsgenerally two- rarely three- or more-flowered. Flowering glumes with a terminal or dorsal bent and twisted awn. Palea large, enclosing with the flowering glume the fruit. EXAMPLES : Aira, TTolcus, Arrhenathcrum, Avena, Trisclum, Danthonia, Lagurus. [2, 3, and 4 together form Bentham s tribe Streptatherce. ] 5. Pappophorccc. Spikelets one- or several -flowered. Flowering glumes rounded on the back, terminating in three or more teeth often carried out into straight awns. EXAMPLES : Pappophorum, Triraphis, Cottca, Triodia. 6. CMoridecp. Spikelets one- or several-flowered, sessile on one sided spikes. Flowering glumes rarery awned.

EXAMPLES : Cynodon, CJiloris, Eleusinc, LeptoMoa, Spartina, Ctenium, Nardus. 7. Miliecc. Spikelets one- or several-flowered, paniculate. Flower ing glumes usually rounded and unawned. Palea large. EXAMPLES : Milium, Coslachne, Sporobolus. 8. Fcstucece. Spikelets several- or many-flowered, stalked, panicu late or capitate. Flowering glumes entire, obtuse or acute, or with a straight awn ; one or more empty ones above the fertile flowers. EXAMPLES : Festuca, Brow/us, Lamarkia, Briza, Poa, Era- grostis, Dactylis, Cynosurus, Arundo, Pliragmitcs, Melica. FIG. 11. Poacece. 1. spikelet of Briza ; 2, spikelet of Tri/icum. 9. Bamlusccc. Spikelets one-, several-, or many-flowered, usually sessile, paniculate or capitate. Lodicules three. Stamens generally six. Stems very large and tall (called arbore ous or shrubby). Blade of leaf articulated with the sheath. EXAMPLES : Arundinaria, CJmsquea, Nastus, Bambusa, Mclo- canna, Beeslia. 10. Hordeae. Spikelets one- or several-flowered, sessile on the opposite sides of the main axis of the spike. Otherwise as in Festucece. EXAMPLES : Ifordcum, Agropyrum, dSgilops, Triticum, Lolium, Lcpturus, Elymus. [5 to 10 compose Bentham s tribe AstrcptcK.]

III. Distribution.—Grasses are the most universally diffused over the globe of all flowering-plants. There is no district in which they do not occur, and in nearly all they are a leading and dominant feature of the flora. In actual number of species Graminece comes considerably after Com posite? and Leguminosce, the two most numerous orders of phanerogams, but in number of individual plants it pro bably far exceeds either; whilst from the wide extension of many of its species, the proportion of Graminece to other orders in the various floras of the world is much higher than its whole number of species would lead one to expect. This number can, however, scarcely be put much below 6000, which is probably somewhat more than a fifth of all monocotyledons. This is only about T^th of the phanero gams as a whole, yet in any given locality, with a very few exceptions, this proportion is largely exceeded. In tropical regions, where Leguminosce is the leading order, grasses closely follow as the second, whilst in the warm and tempsrate regions of the northern hemisphere, in which Composite takes the lead, Graminece again occupies the second position. As the colder latitudes are entered the grasses become relatively more numerous, and are the lead ing family in Arctic and Antarctic regions. The only countries where the order plays a distinctly subor Jinate part are some extra-tropical regions of the southern hemisphere, Australia, the Cape, Chili, &c. The actual proportion of graminaceous species to the whole phanerogamic flora in different countries is found to vary from nearly ^th in the Arctic regions to about ^th at the Cape ; in the British Isles it is about T th. The following are proportions per cent, in various floras, from Decandolle s Geogr. Botanique, which must, however, be taken as merely approximations in most cases :—

In the tropics :—Cape Verd Islands, 13 ; Abyssinia and Nubia, 12 ; Mexico, 10; Hawaiian Islands, 10; Congo, 8; Mauritius, 8; neighbourhood of Quito, 10 ; Barbados, 6 ; Surinam, 6 ; Tunis, 6 ; New Guinea, 4 ; New Grenada, 4. In temperate regions of the northern hemisphere :—Banda and some other districts of India, 15 to 17; Egypt, 12; Texas, 12; Azores, 12; Madeira, 11; Algeria, 9; Canaries, 8; United States, 8; Sardinia, 9; Holland, 10; Sweden, 8J; Great Britain, 8J ; France, 7^; Germany, 7; China, 8 ; Altai, 5

The principal climatic cause influencing the number of graminaceous species appears to be amount of moisture ; it is only in very dry countries that they become distinctly less numerous. A remarkable feature of the distribution of grasses is its uniformity ; there are no great centres for the order, as in Composite?, where a marked preponderance of endemic species exists ; and the genera, except some of the smallest or monotypic ones, have usually a wide distri bution. Speaking generally, however, the Panicacece are tropical and warm temperate plants, whilst the grasses of temperate and colder regions are members of the Poacece. The former are very sparingly represented in Europe by a few species of the vast tropical genera Andropogon and Panicum. Poacece, on the other hand, form a fair pro portion of tropical Graminece, especially in the higher dis tricts where, as in the mountains of Abyssinia, are several endemic genera and many species. The largest tropical genus of Poacece is Eragrostis. The distribution of the tropical tribe Hamliisece is in teresting. There are 170 or more species, which are about equally divided between the Indo-Malayan region and tropical America, only one species being common to both. Apparently there is but a single native species on the African continent, where it has a wide range, and none are recorded for Australia, though species may perhaps occur on the northern coast One species of Ariindinana reaches north wards as far as Virginia, and the elevation attained in the Andes by some species of CJmsquea is very remarkable, one, C. aristata, being abundant from 15,000 feet up to nearly the level of perpetual snow. Many grasses are almost cosmopolitan, such as our common reed, Phragmites communis ; and many range throughout the warm regions of the globe, e.g., Cynodon Dactylon, Eleusine indica, Imperata arimdinacea, Sjiorobolvs indicus, &c., and such weeds of cultivation as species of Setaria, Ec.hinochloa, which are found over both Old and New Worlds. The recent masterly revision of the whole of the Australian species by Bentham well exhibits the wide range of the genera of the order in a flora generally so peculiar and restricted as that of Australia. Thus of the 90 indigenous genera (many monotypic or very small) only 1 4 are endemic, 1 extends to South Africa, 3 are common to Australia and New Zealand, 18 extend also into Asia, whilst no less than 54 are found in both the Old and New Worlds, 26 being chiefly tropical and 28 chiefly extra-tropical. Of specially remarkable species Lygeum is found on the sea-sand of the eastern half of the Mediterranean basin, and the minute Coleanthus has only occasionally occurred at intervals in three or four isolated spots in Europe (Norway, Bohemia, Normandy). Many remarkable endemic genera occur in tropical America, including Anomockloa of Brazil, and most of the large aquatic species with separated sexes are found in this region. The only genus of flowering plants peculiar to the arctic regions is the beautiful and rare grass Pleuropogon Saliinii, B. Br., of Melville Island.

Bibliography.—R. Brown, Prod. Flor. Nor. Holland. (1810); Id., appendix to Flinders s Voyage, p. 580 (1814); Id. and J. J. Bennett in Plant. Javan. Ear., p. 8 (1838); Falisot de Beauvois, Etude d Agrostographie (1812) ; Dumortier, Observations sur les Graminecs de Bclgiquc (1823); Id., Etude Agrostograpliiquc (1868); Trinius, Fundamcnta Agrostographice (1820) ; Id., De Gram. Uniflor. (1824); Kunth, Eniimcratio Plantarum, i. (1833); Id., Distrib. Methodiquc des Gram. (1835); Von Mohl, in Bot. Zcitung, 1845, p. 33; E. Fries, Summa Vcg. Scandinavian (1846); Doll, Flora des Grossh. Badens, bd. i. (1857); Id., in Jahrcsber. des Mannlieimer Ver. f. Naturfcundc, 1868, p. 30; Id., Flora Brasili- ensis, Graminece, i. (1871), ii. (1877), iii. (1878); Bentham,

Handbook of Brit. Flora (1858); Id., in Journ. Linn. Soc. Lond., xv. p. 390 (1877); Id., Flora Australicnsis, vii. (1878); Munro, in Harvey, Genera S. African Plants, ed. 2 (1862); Id., Bambusece, in Trans. Linn. Soc. Lond., xxvi. p. 1 (1868); Duval Jouve, ill Mem. de I Acad. des Sciences dc Montpellier (1871); Van Tieghem, in Ann. des S& Nat., ser. 5, xv. p. 236 (1872); Eichler, Bluthcn- diagramme, i. (1875); Fouruier, iu Bull. Soc. Sot. Belg., xv. p. 459 (1876).

(H. T.)