20 PHYSICAL INVESTIGATION. 



to different stocks, although for these, we must add, it remains 

 as yet undecided whether they have descended from one or 

 several, and in the latter case again, whether from perfectly 

 similar, or not perfectly similar pairs. Should we be inclined to 

 assume or consider as possible, that all, or but a few species of 

 animals, have descended from several protoplasts, we are already 

 cautioned not to confound the notions of unity of species and 

 unity of descent. If, then, the notion of species and the whole 

 sphere of its applications is not, at the outset, to be placed on a 

 fluctuating basis, it will be requisite to keep the reference to 

 unity of descent separate from it, which is necessary in order 

 that every thing merely hypothetical should be excluded there- 

 from. The notion of unity of species of a number of individuals 

 rests, as we have seen, in the organic and inorganic world, 

 solely upon the similarity of their external and internal nature, 

 upon the regular coincidence of the same essential characters, 

 by which, however, nothing, either in animals or plants, is 

 yet decided as to community of origin. This community of 

 origin is merely a probable deduction from, the actual similarity 

 of their nature, because propagation and transmission seem to 

 be the only way of its preservation. It is on this ground alone, 

 that organic beings, belonging to the same species, should not 

 exhibit greater differences than such as can be traced in 

 individuals of the same stock ; and yet there remains, in spite 

 of this, a possibility that individuals whose differences do not 

 exceed the limits of variation of the same stock, are not de- 

 scended from the same parents, nor from perfectly similar 

 parents ; notwithstanding which there would be a sufficient 

 justification for including them in the same species. 



We shall, therefore, adopt the first proposition that unity of 

 species results from proved unity of origin ; but not the second, 

 which has often by zoologists been considered as inseparable 

 from it, namely, that separate descent, wherever it can be 

 traced, is a sufficient proof of difference of species. In cases of 

 the latter kind the process in modern times has usually been 

 to declare similar types, which hitherto had passed as mere 

 varieties, to be different species, if these types belonged either 

 to certain definite regions, or if apparently unsurmountable 



