that their influence extended thither; but, however that may be, it seems clear, from the following pagodas, of perhaps the very oldest construction to be found in India. They are simply pyramids formed by piling stone on stone; the apex is cut off at about one seventh of the whole height of the complete pyramid, and four of them have small ornamental buildings on the top, evidently of more modern work, which are finished by an ornament made of copper and gilt, perfectly resembling the trident of the Greek Neptune." To which I take occasion to add, (premising, however, that I speak from memory, having no present access to the book,) that in the drawing given either by Mr. Hodges or Mr. Daniel, of the superb building called the Toja Mahal, at Agra, there runs along the frise a continued line of stars of six points, like the one drawn in Fig. 171; ✡ and I think I have met with the same ornamental symbol in drawings of other architectural remains in India. This star I take to have been intended as a peculiar mark of passage, that they had established themselves so high up the river Ganges as the point where the connexion subsisting between India and the British Islands, the latter as well as the former resembling a triangle in shape, but as their apexes point contrary ways, the one to the north and the other to the south, the union of the two together would of course constitute the figure of such a star. The present may be no unfit occasion to notice, that in the Introduction to these Chapters are many observations tending to shew that the Latin is an artificial language, and composed probably of portions of various other languages. From the superior advancement in literature and science, which the survey of the works of the ancients taken in these volumes, as well as the conclusions drawn from every other consideration, would lead one to ascribe to the ancient inhabitants of those islands, there is reason to suppose, that if they were not the sole framers of such a language, they at least had some share in the undertaking; and if the evidence above offered be deemed sufficient to prove the close connexion that subsisted anciently between these islands and India, the whole together furnishes a simple solution of a fact which must otherwise its fresh water ceases to be affected by the salt of the sea, statements. appear very extraordinary, namely, that there should be such a multitude of words in the Persian and Sanscrit languages, which (allowing only for such shadows of difference as arise from the several grammatical constructions of the languages,) are evidently appellatives of the same things in the Latin likewise. In the 13th vol. of the Edinburgh Review, is given a very able critique upon Mr. Wilkinson's learned Grammar of the Sanscrita language, containing the following among other valuable "The analogy we propose to demonstrate is, in the first place, that of the words composing the languages treated of; and secondly, that of their structure. In the execution of the first part, we shall merely open the Persian Dictionary Farhang Jehangiri; and when we meet with a word that has the same signification in Sanscrit, Latin, or German, write them in a line. Others will certainly occur in the course of the operation. The number of words to be exhibited (of which the reviewer has given about two hundred,) will amply demonstrate the analogy; but it is not their number, merely, to which we would direct the attention of our readers. That is una Αυταρ ο εκ ποταμε χροα νίζετο διος Οδυσσευς voidably limited by the nature of our review; and, with a very inconsiderable portion of time and labour, might easily be extended to ten times its present amount. It is the sort, and not the number of similar words that attest the affiliation or consanguinity of nations and languages. There are things which must have been named in the very infancy of society, and before the first dawn of civilization. Where these names correspond in different countries, we may confidently infer, that the one has been peopled from the same stock with the other. Thus, the names of the parts of the human body, of the relations of consanguinity, and of the animals most familiar to man, constitute a class of words without which we cannot suppose mankind to exist, even in the rudest state of society. To adopt the hypothesis of the learned Bayer, we must suppose the inhabitants of Hindostan to have waited till Alexander the Great conquered Bactria, in order to obtain appellations for the most endearing ties of nature, and to enable them to express the venerable relations of father and mother. The words we propose to exhibit consist and that their authority extended not only up the Ganges, but through the whole peninsula of solely of the class we have described." Besides the two hundred examples above mentioned, there are a great number of valuable observations in the Review, which have a greater or less bearing on the subject now in question; but I must refer for more to the book itself; the inference I contend for being simply this; that, if in the oldest times, of which we have any account, a close connexion subsisted between the British Islands aud India, and if it is fairly presumable that in the construction of an artificial language, as the Latin, by the Europeans, the English would have had a share in the work, it should not appear surprizing that a great many words from the Sanscrit and Hindoo tongues should be found in such artificial language, agreeable to the fact which the reviewer records. But notwithstanding the loug citation already given from this. Review, I hope to be excused for extracting from it the following passage: "We have now said enough, we trust, to shew the analogy between the languages of the East and those of the West. All the distinctive characters which discriminate ancient from |