TO SIR ALEX. -JOHNSTON

meagre in the extreme , and better calculated to set the learner out wrong, and to keep him so, than to benefit bim in acquiring the Arabie language. The admirable works of Colonel Baillie and Mr. Lumsden are unfinished, and likely to remain so. In this case the learner must hâve recourse either to the (Mammaire Arabe of M. de Sacy, or to one of the gramtnars published in Latin by the Catholic missionaries ; in the latter of which , however , he will hâve lhe mortification to find very great defects , and , in some cases , views on the subject quite foreign to the genius of that language. M. de Sacy lias supplied inany of the deficieneies. and corrected many of the errors of precieding writers on Arabie grammar, but valuable as his work is , it leaves something to be desired through the omission of the prosody, and his paradigm of the verb does not quite agréé with the Views of the Arabian grammarians. In the Latin grammar of Guadagnoli, indeed, a prosody is to be found, but this is full of mistakes, as Clarke has shewn. If he hâve recourse to the work of Mr. Gladwin on this subject, he will here find endless difïîeulties. The only work of much value on this subject, is the litlle book published by Clarke at Oxford, about 150 vears ago. But this requires the greatest stretch of attention to understajid the rules, and of principles nothing is said. It is surely much to be regretted, that we hâve no good elementary work of this kind in English : and still more so to find that there is not the least prospect of having one, until some step be taken, either by the Royal Asiatic Society , or some other body capable of bearing the expenses incident to such an undertaking. t

In the next place, what hâve we in lexicography? If we except the lexicons of Golius and Castell , we hâve nothing we can recommend as a general dictionary. Wilmet, indeed, has corupiled. a very useful work for a few partieular hooks : but then that work is scarce , not to insist on its uselessness in a general way. But this objection will go in a, great degree against the lexicons of Golius, Castell, the Kamoos, and the Soorah ; for in these we find scarcely one of the ternis of art, without which hardly a single book in Arabie can be made out. Were it necessarv herc to go into the detail, I could shew, that scarcely a translator is to be named , from Pocoeke down to the présent day , who has not had his labor greatly increase(\ through the omission of technical terms in those dletionnaries. This remark extends to every science, to works on theology, and even to the commonest expressions in use among the Arabians. Again, let a inan take any book of poetrv, or of proverbial expressions, such as' the work of Meidani, and try his hand with any of the dictionaries just mentioned. I hâve no doubt be will make out a