especially during the cold season, when the seeds ripen, contains a notable proportion of oxalic acid, and is said to be very injurious to the leather shoes of those who walk through fields of gram. In Mysore the dew containing it is collected by means of cloths spread on the plant over night, and is valued as a remedy for dyspepsia, indigestion, and costiveness. The steam of water in which the fresh plant is immersed is in the Deccan resorted to by the Portuguese for the treatment of dysmenorrhœa. The
seed of Phaseolus Mungo, Linn., or green gram (Hind., and Beng., moong), a variety of which plant, P. Mungo melanospermus (P. Max, Roxb.), is termed black gram, is an important article of diet among the labouring classes in India, and is annually exported in large quantities from Madras. Soup made from it is considered to be especially suited to sick persons. The meal is an excellent substitute for soap, and is stated by Elliot to be an invariable concomitant of the Hindu bath. P. Roxburghii, W. and Arn., or P. radiatus, Roxb. (Hind., urid ; Beng., máshkalái), which also is known as green gram, is perhaps the most esteemed of the leguminous plants of India, where the meal of its seed enters into the composition of the more delicate cakes and dishes, and is used in medicine both externally and internally. P. aconitifolius goes by the name of Turkish gram. Horse gram, Dolichos uniflorus, Lam. (Hind., and Beng., kulthi), which supplies in Madras the place of the chick-pea, affords seed which, when boiled, is extensively employed as a food for horses and cattle in South India, where also it is eaten in curries, and, made into poultices or pastes, is applied to medicinal purposes. Turkish gram is the Dolichos Catjang of Roxburgh. White gram, Glycine (Soja) hispida, produces the beans from which soy is made. The quantity of gram exported from India in 1876-77, chiefly to Mauritius, Ceylon, and the Straits, amounted to 316,592 cwt., against 322,661 cwt. in the previous year.
See W. Elliot, “ On the Farinaceous Grains and the various kinds of Pulses used in Southern India,” Edin. New Phil. Journ., 1862, vol. xvi. p. 16 sq. ; H. Drury, The Useful Plants of India, 1873; U. C. Dutt, Materia Medica of the Hindus, Calcutta, 1877.
GRAMMAR. By the grammar of a language is meant
either the relations borne by the words of a sentence
and by sentences themselves one to another, or the sys
tematized exposition of these. The exposition may be,
and frequently is, incorrect ; but it always presupposes
the existence of certain customary uses of words when
in combination. In what follows, therefore, grammar
will be generally employed in its primary sense, as denot
ing the mode in which words are connected together in
order to express a complete thought, or, as it is termed in
logic, a proposition.
The object of language is to convey thought, and so long as this object is attained the machinery for attaining it is of comparatively slight importance. The way in which we combine our words and sentences matters but little, pro vided that our meaning is clear to others. The expressions " horseflesh " and " flesh of a horse," are equally intelligible to an Englishman and therefore are equally recognized by English grammar. The Chinese manner of denoting a genitive is by placing the defining word before that which it defines, as in kouejin, " ma a of the kingdom," literally " kingdom man," and the only reason why it would be incorrect in French or Italian is that such a combination would be unintelligible to a Frenchman or an Italian. Hence it is evident that the grammatical correctness or incorrect ness of an expression depends upon its intelligibility, that is to say, upon the ordinary use and custom of a particular language. Whatever is so unfamiliar as not to be generally understood is also ungrammatical. In other words, it is contrary to the habit of a language, as determined by com mon usage and consent. In this way we can explain how it happens that the grammar of a cultivated dialect and that of a local dialect in the same country so frequently disagree. Thus, in the dialect of West Somerset, thee is the nominative of the second personal pronoun, while in cultivated English the plural accusative you (Anglo-Saxon, eow) has come to re present a nominative singular. Both are grammatically correct within the sphere of their respective dialects, but no further. You would be as ungrammatical in West Somerset as thee is in classical English ; and both you and thee, as nominatives singular, would have been equally un grammatical in Early English. Grammatical propriety is nothing more than the established usage of a particular body of speakers at a particular time in their history. It follows from this that the grammar of a people changes, like its pronunciation, from age to age. Anglo-Saxon or Early English grammar is not the grammar of Modern English, any more than Latin grammar is the grammar of modern Italian ; and to defend an unusual construction or inflexion on the ground that it once existed in literary Anglo- Saxon, is as wrong as to import a peculiarity of some local dialect into the grammar of the cultivated speech. It further follows that different languages will have different grammars, and that the differences will be more or less according to the nearer or remoter relationship of the languages themselves, and the modes of thought of those who speak them. Consequently, to force the grammatical framework of one language upon another is to misconceive the whole nature of the latter, and seriously to mislead the learner. Chinese grammar, for instance, can never be understood until we discard, not only the terminology of European grammar, but the very conceptions which underlie it, while the polysynthetic idioms of America defy all attempts to discover in them " the parts of speech " and the various grammatical ideas which occupy so large a place in our school-grammars. The endeavour to find the dis tinctions of Latin grammar in that of English has only resulted in grotesque errors, and a total misapprehension of the usage of the English language.
It is to the Latin grammarians, or, more correctly, to the Greek grammarians, upon whose labours those of the Latin writers were based, that we owe the classification of the subjects with which grammar is commonly supposed to deal. The grammar of Dionysius Thrax, which he wrote for Roman schoolboys in the time of Pompey, has formed the starting-point for the innumerable school-grammars which have since seen the light, and suggested that division of the matter treated of which they have followed. He defines grammar as a practical acquaintance with the language of literary men, and as divided into six parts, accentuation and phonology, explanation of figurative expressions, definition, etymology, general rules of flexion, and critical canons. Of these, phonology and accentuation, or prosody, can properly be included in grammar only in so far as the | construction of a sentence and the grammatical meaning of I a word are determined by accent or letter-change ; the accentual difference in English, for example, between incense and incense belongs to the province of grammar, since it indicates a difference between noun and verb ; and the changes of vowel in the Semitic languages, by which various nominal and verbal forms are distinguished from one an other, constitute a very important part of their grammatical machinery. But where accent and pronunciation do not serve to express the relations of words in a sentence, they fall into the domain of phonology, not of grammar. The explanation of figurative expressions, again, must be left to the rhetorician, and definition to the lexicographer; the grammarian has no more to do with them than he has with the canons of criticism.
In fact, the old subdivision of grammar, inherited from
