The first and best is to consult the languages derived, or which we believe to be derived, directly from the ancient tongues, as Basque for the Iberian language, Albanian for the Pelasgic, Breton, Erse, and Gaelic for Keltic. The danger lies in the possibility of mistake in the filiation of the languages, and especially in a mistaken belief in the antiquity of a plant-name which may have been introduced by another people. Thus the Basque language contains many words which seem to have been taken from the Latin at the time of the Roman rule. Berber is full of Arab words, and Persian of words of every origin, which probably did not exist in Zend. The other method consists in reconstructing a dead language which had no literature, by means of those which are derived from it; for instance, the speech of the Western Aryans, by means of the words common to several European languages which have sprung from it. Fick’s dictionary will hardly serve for the words of ancient Aryan languages, for he gives but few plant-names, and his arrangement renders it unintelligible to those who have no knowledge of Sanskrit. Adolphe Pictet’s work[24] is far more important to naturalists, and a second edition, augmented and improved, has been published since the author’s death. Plant-names and agricultural terms are explained and discussed in this work, in a manner all the more satisfactory that an accurate knowledge of botany is combined with philology. If the author attributes perhaps too much importance to doubtful etymologies, he makes up for it by other knowledge, and by his excellent method and lucidity. The plant-names of the Euskarian or Basque language have been considered from the point of view of their probable etymology by the Comte de Charencey, in _Les Actes de la Société Philologique_ (vol. i. No. 1, 1869). I shall have occasion to quote this work, of which the difficulties were great, in the absence of all literature and of all derived languages. 6. _The necessity for combining the different methods._ The various methods of which I have spoken are of unequal value. It is clear that when we have archæological records about a given species, like those of the Egyptian monuments, or of the Swiss lake-dwellings, these are facts of remarkable accuracy. Then come the data furnished by botany, especially those on the spontaneous existence of a species in a given country. These, if examined with care, may be very important. The assertions contained in the works of historians or even of naturalists respecting an epoch at which science was only beginning, have not the same value. Lastly, the common names are only an accessory means, especially in modern languages, and a means which, as we have seen, is not entirely trustworthy. So much may be said in a general way, but in each particular case one method or the other may be more or less important.