Erskine continues: “The Persian system it is unnecessary to particularise; but it is worthy of attention that, among the words of Indian origin, not only are many Sanscrit, which might happen in a work of a remote age, but several belong to the colloquial language of Hindustán: this is suspicious, and seems to mark a much more recent origin. Many words indeed occur in the Desátir that are common to the Sanscrit and to the vulgar Indian languages (the author quotes thirty-four of them); many others might be pointed out. But the most remarkable class of words is that which belongs to the pure Hindi; such I imagine are the word _shet_, ‘respectable,’ prefixed to the names of prophets and others (twenty-four are adduced). Whatever may be thought of the words of Persian descent, it is not probable that those from the Hindustaní are of a very remote age; they may perhaps be regarded as considerably posterior to the settlement of the Muselmans in India.” Strongly supported by the opinion of respectable philologers, I do not hesitate to draw a quite contrary conclusion from the facts stated by Erskine. It should be remembered that, in the popular or vulgar dialects are often found remains of ancient tongues, namely, roots of words, locutions, nay rules of grammar which have become obsolete, or disappeared in the cultivated idioms derived from the same original language. It was not without reason that the illustrious William Humboldt recommended to the Royal Asiatic Society of Great Britain and Ireland,[45] to examine, on behalf of general Oriental philology, the different provincial dialects of India. Even the gibberish of gypsies is not to be neglected for that purpose.[46]