In grasses the spikelets form the ultimate divisions of a type of branched raceme often vaguely described as a panicle. It is very difficult to define a grass panicle, but the features which give it its characteristic appearance are, firstly, the absence of axillant leaves, and, secondly, the abbreviation of the basal internodes of the branches, associated with precocious development of lateral axes; this second factor results in a confused crowding of branches of different orders. The lowest of the secondary branches may arise at the very base of the primary branch, and may, again, have a tertiary branch at its base, so that three branches, arising apparently as a triplet, may yet be, in reality, of three different orders. Owing to this precocious branching, the secondary and later axes may form a pseudo-half-whorl at each node, these half-whorls alternating along the primary axis. Examples of secondary and tertiary axes arising at the same node, and thus giving ‘twinning’ of branches of different orders, may be seen in Fig. 54, A1 and A2 (Briza media L.). The hair-like lateral axes, at the ends of which the spikelets dance in the slightest breeze, have earned for this grass such dialect names as Earthquakes and Doddering Jockies. In some Gramineae, for instance, the Fescues, the panicles are one-sided; for though the primary branches stand alternately to right and left, they do not lie in one plane, as one would expect of branches axillary to distichous leaves, but their planes meet at an angle which is less than 180°, so that they all stand out to one side of the axis.

In the preceding chapter, we considered the structure and anthesis of the reproductive shoot in grasses, omitting the more aberrant forms. Even when the field was thus limited, we met with a series of instances in which the divergence from the ‘typical’ Monocotyledonous floral diagram—or even from the more restricted scheme developed in the Bambuseae—followed the direction of fusion or suppression of parts. Examples which may be recalled are: fusion of the outer empty glumes; absence of the third lodicule; fusion of the remaining pair; absence of one whorl of stamens; loss of two of the three stamens of the remaining whorl; and the uniovular character of the gynaeceum. In this chapter we shall continue to study the subject of reduction, both in typical grasses and in the more peculiar examples in which the factor is specially apparent. It is remarkable that this reduction-trend—though itself, in the last analysis, a negative character—has led to an exceedingly varied series of forms.

Certain modifications in the outer empty glumes may first be discussed. The inflorescences of the curious little grass, Cornucopiae cucullatum L., have delicate cupules, each of which encloses one spikelet; they are drawn in Fig. 72, D and E, while the structure is shown in section in F–H. Unlike certain former writers, I interpret the cupule as consisting of the first and second outer empty glumes in a state of fusion.

It is a matter of common observation that no family of flowering plants includes so great a number of individuals as the Gramineae—a fact which seems in part accounted for by their extraordinarily prolific and effective seeding. There are a number of detailed records confirming this general impression. As long ago as 1660, Sir Kenelm Digby related that the “Fathers of the Christian Doctrine at Paris doe still keep by them for a Monument (and indeed it is an admirable one) a Plant of Barley consisting of 249. stalkes, Springing from one Root or Grain of Barley, in which they counted above 18000. Grains or seeds of Barley”. Later in the seventeenth century, Thomas Everard recorded that from one grain of Wheat he had obtained eighty ears and above 4000 grains. Seeding on this scale may lead to a rapidity of increase which seems almost miraculous. A few years ago the chief kind of Wheat grown in Canada was that known as ‘Marquis’. One grain of Marquis was planted in an experimental plot at Ottawa in the spring of 1903, and in 1918, from the progeny of this grain after fifteen years, 300,000,000 bushels of Wheat were produced in Canada and the United States.

In the grasses there is normally a period of seed-rest before germination. The whole subject of seed-rest, and the related problem of delayed germination, seem to need a thorough comparative study, for our present knowledge of these questions is only complete enough to reveal much obscurity.

Among the Gramineae, the distinction between the towering, woody bamboos, and the herbaceous grasses and cereals, is so striking that an effort of the mind is needed before we can think of them together, as members of a single and seemingly natural family. In Chapter iv we treated the tree habit of the bamboo merely as an observed fact; in this chapter we shall make an attempt to understand something of its significance. It cannot, however, be interpreted at all, if it is considered as an isolated phenomenon; in order to get any glimmering comprehension of it, we shall be obliged now to turn to the problem of the dendroid and herbaceous type in the flowering plants as a whole. From the earliest times at which a desire to pigeon-hole the facts of the world led man to attempt some kind of classification of the plants which he saw around him, the difference between the tree and the herb has seemed to him an important one. Theophrastus (born circa 370 .c.), the earliest scientific botanist of whose opinions we have a record, divided plants into the tree, the shrub, the undershrub and the herb. His classification thus makes a fundamental distinction between the arboreal and herbaceous habits, though his shrubs and undershrubs to a certain extent form a transition between the two.

The first point which strikes one about the mode of occurrence of the grasses on the face of the earth is their gregariousness. All meadows and many gardens in this country bear witness to the way in which the Gramineae, while living sociably among themselves, oust other plants. The same herd behaviour, coupled with exclusiveness, reappears among the bamboos. An American botanist, who has made a special study of the trees and shrubs of Japan, writes that in that country, “the forest-floor is covered, even high on the mountains, and in the extreme north, with a continuous, almost impenetrable, mass of dwarf Bamboos of several species, which makes traveling in the woods, except over long-beaten paths and up the beds of streams, practically impossible. These Bamboos which vary in height from three to six feet in different parts of the country…prevent the growth of nearly all other under-shrubs, except the most vigorous species”. The same writer attributes the climbing habit, which characterises so many Japanese plants, to this dense undergrowth of bamboos, which renders any other mode of life impracticable for relatively small species, if they are to keep a footing in the forests. The Gramineae are indeed formidable competitors in the struggle for space, not only because of their faculty for adopting a mass formation, but because they show an amazing tolerance about the external conditions of their lives.

In considering the arboreal habit, we noticed that the flowers of the Bambuseae were developed on a fuller plan than those of the other grasses, and approached more nearly to the complete Monocotyledonous type. This contrast will be realised on comparing Fig. 33, which represents the flower of a bamboo, with Fig. 34, p. 111, which shows the relatively reduced flower of Rye. It is thus obvious that the best way to understand the flower of the Gramineae is to start with the bamboo, but, unfortunately, the historical process has followed the reverse direction. Because analytical botany began in temperate regions, there has been a tendency to treat the exiguous European grass as the type form for the Gramineae. This has led to an exaggerated idea of the degree of peculiarity of the flower in this family, which has found expression in a special and elaborate terminology. Much of this terminology, and also of the controversy which has raged about the interpretation of the ‘palea’ and “lodicules’, might have been spared, if the study of the Gramineae had proceeded from the tropics to the temperate regions, and not vice versa. So, in order to follow the logical course, we will now examine the flowers of the bamboos, leaving those of the ‘grasses’, in the limited sense, to be dealt with in later chapters.

In the bamboos which flower annually, the inflorescence terminates the leaf-bearing culm; but in those which are gregarious and periodic flowerers, the leaves fall, and the whole culm becomes one huge reproductive shoot.

The flowering of bamboos, like their vegetative growth, is liable to be on a grand scale. Large quantities of pollen may be formed, which sometimes induce a kind of hay fever. The scale of the seed production can be gauged from the statement that, from one clump of Dendrocalamus strictus Nees, the crown of which covered an area of about 40 square yards, grain was collected to the amount of 160 seers (330 lb.), besides a quantity naturally shed, which resulted in a dense mass of seedlings around the clump. Another record describes the wild tribes of the Assa forest gathering the seed of the same species in March 1901; the outer culms of each clump were cut, one by one, at a height of about 4 ft., and each was laid on the ground which had previously been cleared and swept. The culm was beaten with stout sticks until all its caryopses had fallen; they were then carefully winnowed by children. One adult could collect 2–3 seers (4–6 lb.) of seed in a day.

The periodicity of flowering in the bamboos seems to vary within wider limits than in any other homogeneous group of plants. In South America, annually flowering species are common; they belong to Arundinaria, Bambusa, Guadua, and other genera. In India, on the other hand, only a limited number of species flower every year, e.g. Arundinaria Wightiana</i. Nees, Bambusa lineata Mun. and Ochlandra stridula Thwaites. Many of the Asiatic bamboos have a more prolonged life-cycle, and show themselves, in Riviére's phrase, “assez avares de leurs fleurs”.

Our knowledge of the root systems of grasses falls far short of that of their visible parts—not because there is less to know, but because it is so troublesome to get at facts when they are hidden under layers of earth. There is no short cut to this knowledge; merely uprooting the plants is of little use, for it partially destroys some of the underground organs and displaces the rest. An American botanist, J. E. Weaver, however, has worked out a laborious and thorough technique, by the aid of which he has arrived at a much clearer picture of root systems than had been obtained before. The method he employed was to dig a trench by the side of the plant to be examined, about 5 ft. in depth and of convenient width. This afforded an open face which could be explored with a hand-pick. The original trench often had to be deepened to 8 or 10 ft. and sometimes more. To ensure certainty as to the maximum depth of the root endings, the soil was usually undercut for about a foot below the deepest of them, and was carefully examined as it was removed. It was found that when the root system thus exposed was photographed, many of the finer branches and root ends were always obscured; experience showed that a more accurate record of the extent, position and minute branching of the root system could be obtained by means of a drawing to scale on a large sheet of squared paper.

Before considering certain general points arising out of the study of cereals, we have to discuss two other important crop plants—Rice and Maize—as well as a few minor members of the Gramineae, whose seeds supply food for man. Rice, Oryza sativa L. (Fig. 14), is probably the staple food of more people in the world than any other cereal. It is essentially a swamp plant, and the ‘deep-water Rices’ will succeed in 5 or 6 ft. of water. Certain varieties called ‘mountain Rices’, may, however, be cultivated with no more water than other cereals. The distribution of Oryza, like that of other aquatic plants, is very wide. At the present day it is apparently native to India, Australia and Africa, and it thus becomes very difficult to decide what was its country of origin. The earliest record we have of it is in old Chinese writings, and it has been cultivated continuously in that country since the remote past. Indeed there are fields in China where Rice is believed to have been grown for four thousand years uninterruptedly—a state of things rendered possible by the high pitch to which the use of manures has been brought in that country. The traditional Chinese ceremonies associated with the sowing of the five kinds of ‘corn’ at the vernal equinox, have a history stretching back into antiquity. The first spring sowing is attributed to the Emperor Shên-nung, the Father of Agriculture and Medicine, who reigned about 2700 b.c.

It would be difficult to find a parallel in any other corresponding group of flowering plants for the range in individual size met with in the Gramineae. At one end of the scale are the bamboos, which, as we have shown, may exceed 120 ft. in height. Although the bamboos arrive at greater statures than the other Gramineae, the examples of tall forms among grasses outside this tribe are more numerous than is generally realised by those accustomed only to the European flora. Spruce, in his account of the forests of Chimborazo, describes the Uva, Gynerium saccharoides H. et B. (Festuceae, related to Arundo), as attaining “its maximum of development on stony springy declivities, at an elevation of about 1500 feet above the sea, where a forest of Arrow-cane [Uva], with its tall slender stems of 30 to 40 feet, each supporting a fan-shaped coma of distichous leaves, and a long-stalked thyrse of rose and silver flowers waving in the wind, is truly a grand sight”. Spruce goes on to say— “The longest stem I ever measured was one I met a man carrying on his shoulder at Tarapoto. From that stem had been cut away the leaves and peduncle, and the base of the stem, which is generally beset with stout-arched exserted roots (serving as buttresses), to a height of 1 to 3 feet; yet the residue was 37 feet long, so that the entire length must have been at least 45 feet”.

In an earlier chapter something has been said about the agricultural history of Maize (Zea Mays L.). Its peculiarities call, however, for further study, since—as was already recognised in Lyte's Nievve Herball of 1578— “This corne is a marveilous strange plante, nothing resembling any other kinde of grayne”. Maize is a vigorous annual, whose tall shoot bears a succession of broad distichous leaves; the lowest of these leaves arise at or below the surface of the ground, the buds in their axils producing suckers or tillers. In its general vegetative growth, Maize does not differ essentially from the Gramineae of the Old World, but it was its mode of reproduction which perplexed the botanists of Europe when it was first introduced from America: “for it bringeth foorth his seede cleane contrarie from the place where as the flowers growe, which is agaynst the nature and kindes of all other plantes, which bring foorth their fruite there, where as they have borne their flower”. The main shoot terminates in the male inflorescence, or, in Lyte's words, “at the highest of the stalkes growe idle and barren eares, which bring foorth nothing but the flowers or blossome”. Since in the sixteenth century the function of the stamens was still undiscovered, these blossoms, which were not succeeded by fruits, naturally seemed objectless and “idle”. The female reproductive shoots (Fig. 182, B, p. 356) are borne laterally in the leaf axils at some distance below the male inflorescence.

The impulse to analyse the plant into component members seems, in the first instance, to have arisen out of the desire to establish a comparison between construction in the animal and the vegetable body; for the existence of a close analogy between the two was a fundamental postulate with the biologists of ancient Greece. The first extant attempt at such an analysis is, in some respects, strikingly alien to modern botanical thought. It is that of Theophrastus who, in the fourth century b.c., stated that “the primary and most important parts… are these—root, stem, branch, twig; these are the parts into which we might divide the plant, regarding them as members, corresponding to the members of animals: for each of these is distinct in character from the rest, and together they make up the whole”. Theophrastus then proceeds to distinguish, as subsidiary parts, the leaf, flower, fruit, etc. He was influenced in this discrimination by the fact that, in the tree, which he took as the standard of plant life, the trunk and its branches persist permanently, whereas the leaf, flower and fruit are ephemeral. The importance of the leaf was destined to remain for long unrecognised, and it was not until Goethe turned his attention to botany, more than two thousand years later, that the equivalence of the foliage leaves and the parts of the flower came fully into the light.

In the preceding pages, attention has been concentrated on those Gramineae whose seeds are used by man for food. In this chapter we must turn to those from whose vegetative shoots he takes toll, indirectly or directly. The significance of grass in the life of man long ago received symbolic recognition. Pliny tells us that in Rome no Coronets were “better esteemed…to give testimonie of honour and reward for some notable service performed for the Commonweale, than those which were made simply of greene grasse”. He also refers to an ancient tradition that the greatest sign “of yeelding to the mercie of the enemie, was this, If the vanquished did take up grasse, and tender it unto the conqueror: for this served as a confession and protestation, That they rendered up all their interrest which they might challenge in the earth (the mother that bred and fed them)”. Grass means as much to man today as it did in antiquity, but town life tends to dull his consciousness of the part it plays in his existence. This, however, becomes sufficiently clear when we turn to statistics and learn that in 1932 the number of arable acres in England and Wales was less than 10 millions, while the acreage of pasture and rough grazing exceeded 21 millions.

Considering the immemorial antiquity of cereal culture, it might have been expected that the culture of chosen species of grass would also date back to the remote past. We do not, however, find evidence of it.