BOTANICAL NOMENCLATURE. 116 remain at a standstill. From time to time the revision of a large group has to be undertaken from a uniform and comparative point of view. It then often occurs that new genera are seen to have been too hastily founded on insufficient grounds, and must therefore be merged in others. This may involve the creation of a large number of new names, the old ones becoming henceforth a burden to literature as synonyms. It is usual in such cases to retain the specific portion of the original name, if possible. If it is, however, already preoccupied in the genus to which the transference is made, a new one must be devised. Many modern systematists have, however, set up the doctrine that a specific epithet once given is indelible, and whatever the taxonomic wanderings of the organism to which it was once assigned, it must always accompany it. This, however, would not have met with much sympathy from LinnsBus, who attached no importance to the specific epithet at all : " Nomen specificum sine generico est quasi pistillum sine campana." "^^ Lin- naeus always had a solid reason for everything he did or said, and it is worth while considering in this case what it was. Before his time the practice of associating plants in genera had made some progress in the hands of Tournefort and others, but specific names were still cumbrous and practically unusable. Genera were often distinguished by a single word ; and it was the great reform accomplished by Linnaeus to adopt the binominal principle for species. But there is this difference. Generic names are unique, and must not be applied to more than one distinct group. Specific names might have been constituted on the same basis ; the specific name in that case would then have never been used to designate more than one plant, and would have been suffi- cient to indicate it. We should have lost, it is true, the useful information which we get from our present practice in learning the genus to which the species belongs ; but theoretically a nomencla- ture could have been established on the one-name principle. The thing, however, is impossible now, even if it were desirable. A specific epithet like vulgaris may belong to hundreds of different species belonging to as many different genera, and taken alone is meaningless. A Linnean name, then, though it consists of two parts, must be treated as a whole. Nomen omne plantarum con- stabit nomine generico et specifico." t A fragment can have no vitality of its own. Consequently, if superseded, it may be replaced by another which may be perfectly independent. | It constantly happens that the same species is named and described by more than one writer, or different views are taken of specific differences by various writers ; the species of one are there- fore '4umped" by another. In such cases, where there is a choice
- Phil, 219. t Phil., 212.
I As Alphonse de Candolle points out in a letter published in the Bull, de la Soc. Bot. de France (xxxix.), " the real merit of Linnaeus has been to combine, for all plants, the generic name with the specific epithet." It is important to remember that in a logical sense the "name" of a species consists, as Linnseus himself insisted, in the combination, not in the specific epithet, which is a mere fragment of the name, and meaningless when taken by itself. i2
