The earliest account of the _Shilha_ language of which I am at present aware, is that given by Jezreel Jones, in a Latin epistle published at the end of Chamberlayne’s Oratio Dominica, in 1715. He says, “Lingua Shilhensis vel _Tamazeght_, præter planities _Messæ_, Hahhæ et provinciam Daræ vel Drâ, in plus viginti viget provinciis regni Sûs in Barbaria Meridionali. Diversæ linguæ hujus dantur dialecti in Barbaria, quæ ante Arabicam, primariam Mauritaniæ Tingitanæ et Cæsariensis provinciarum linguam ibi obtinuêre, et hodiernum inter Atlanticorum Sûs Dara et Reephean Montium incolas solum exercentur.” A specimen is added, consisting of about one hundred words. In the excellent account of Morocco published in Danish, by George Höst, in 1779, there is also a short vocabulary of this language, where the words are given with apparent accuracy, in the Arabic character. Several years ago you were so kind as to transmit for me to Mr. Matra, His Majesty’s Consul at Morocco, (a gentleman whose exertions for the advancement of useful knowledge, and particularly of that which is the object of the African Association, deserve the highest praise,) a copy of an extensive alphabetical list of English words, which I printed and distributed with the view of facilitating the attainment of languages not to be met with in dictionaries, and in consequence of which I received from him, through your hands, a very valuable communication. “It is not, (he says in a letter dated in 1791, that accompanied it,) the printed copy Mr. Marsden sent me, but an exact duplicate. His copy, with the words translated into Arabic, is sent to _Tombuctoo_, I fear but with little chance of its returning.” That copy never did return; but the transcript I received contains a version of all the words into the Mauritanian dialect of Arabic, for the purpose of enabling a _talb_, or priest, from the _Shilha_ country, to write opposite to each, in the same character, the corresponding terms in his language. I have been used to consider this as a very curious document, even when I believed it to apply only to the western coast of Africa, but its importance will be much increased, if we should find, as there is ground to presume, that the _Shilha_ or _Berber_ extends across the whole continent, in a direction between the Negro dialects on the southern side, and the Moorish or Arabic of the Mediterranean coasts, and that it was the general language of all Northern Africa before the period of the Mahometan conquests. Independently of the Arabic terms, which must ever accompany the progress of that religion, I think it exhibits some strong marks of affinity to that class of Oriental languages which the German writers have distinguished by the name of Shemitic; and if this should be established, (contrary, however, to the opinion of Höst,) it will not be unreasonable to suppose it the ancient Punic, corrupted by the influx of words successively