, न »७०२ ] পরিশিষ্ট । 8o Montpellier, Toulouse, Poitiers, L'Orient, &c., is simply ridiculous. We can only transliterate them as they are pronounced by Frenchmen,—i.e., Blwa, Valé, Mongpellié, Toolooz, Pwatie, Loreeawing. To look for how “a mere Englishman would pronounce them' is to search for what is ever changing and always wrong. It has not the sanction of scholars, and is not therefore Worthy of notice. And if this be admitted, the principle laid down in the rule under notice is fully vindicated, for I do not think anybody would deny to the classical languages of the East, such as the Arabic and the Sanskrit, the consideration which is so readily conceded to the French. The rule contemplates the strictest attention to the native sound of Words as opposed to the sounds apparent in Roman letters. It might be said that as our book makers are not generally familiar with foreign languages, except the English, they cannot be expected to conform to this rule. But I do not apprehend any practical difficulty in this respect. In Webster's large Dictionary there is an Appendix giving the pronunciation of all foreign geographical names, both ancient and modern, and reference to it will at once obviate the difficulty. The pronunciations given in it are not always strictly accurate, but they are sufficiently so for all practical purposes. The School Book Society's large wall maps in Bengali, which have been got up on the principle here advocated, will also afford material help. It is observable also that the Society's maps are the only ones available in the market, and by following them the text-books will prove much more convenient for reference than what they now are. Il. The second rule I would propose would be that in transliterating every attention should be paid to the powers of the Sanskrit letters. The object of this rule is to prevent the use of short vowels where in the original long vowels exist, of cerebral letters in lieu of dentals, of proper sibilants, and of B, Bh, and V. In a text-book now in use I find the Arabic word hadranat written शgभा. Now, in the Arabic, the vowels after the h, r and m are short ones, but they have been made long, and the lingual S and have been used for the dental and 5. In Arabic, as in French, cerebral letters are unknown. The word should have been Written years. In Shah the sibilant is deeply guttural, but it is frequently written with the dental . The letter B has a wide phonetic range. When aspirated it becomes Bh, but when it is weakened it first becomes P, then W, then W, and then U, the W occasionally changing into F, as in fry for tery in the Inouth of an uneducated German. V, however, is two steps removed rom Bh, and should not be interchangeable; but in many Bengali books
