GEOGRArHICAL DISTRIBUTION 325 



In all seas there are fishes and otlicr marine animals, and each 

 creatm'e haunts the i)hices for whicli it is fitted. 



There is the closest possible aniilo«]i;y between the variations 

 of species of animals or jjlants in (HITerent districts and that of 

 words in different languages. Tlie language of any people is 

 not a unit. It is made up of words which have at various 

 times and under various conchtions come into it fr(jm the speech 

 of other people. The granunar of a language is an expression 

 of the mutual relations of these words. The word as it exists 

 in any one language represents the species. Its cognate or its 

 ancestor in any other language is a related species. The words 

 used in a given district at any one time constitute its ])liil()- 

 logical fauna. There is a struggle for existence between woids 

 as among animals. For example the words hcgiji and commence, 

 shake and agitate, work and operate (Saxon and French) are in 

 the English language constantly b.ought into competition. 

 The fittest, the one that suits English ])urposes best, will at 

 last survive. If both have elenunits of fitness, tl:e field will 

 be divided between them. The silent letters in words tell their 

 past history, as rudimentary organs tell what an animal's 

 ancestry has been. This analogy, of course, is not perfect in 

 all regards, as the passing of the words from mouth to mouth 

 is not rigidly comparable with the generation of animals. 



We may illustrate the formation of species of animals by 

 following any widely used word across Europe. Tluis the 

 Greek aster becomes in Latin and Italian stella; whence the 

 Spanish estrella and the French etoile. In Germany it becomes 

 Stern, in Danish Stjern; whence the Scottish starn and English 

 star. 



In like manner, the name cherry may be traced from country 

 to country to which it has been taken in cultivation. Its Greek 

 name, kerasos, becomes cei^asus, ceresia, ceriso, cereso, cerise, 

 among the Latin nations. This word is shortened to Kirsrh 

 and Kers with the people of the North. In England, chcrys, 

 cherry, are obviously derived from cerise. 



The study of a fauna or a flora as a whole is thus analogous 

 to tlie study of a living language. The evolution of a language 

 corresponds to the history of the life of some region. Fhilology, 

 systematic zo()logy, and botany are alike intimately related to 

 geograpliy. Tlie parallelism betw(H'ii speech districts and 

 faunal districts has been manv times notcth The spread of a 

 22 



