86 



find the limits of species and genera, &c., in the wider spaces, and of 

 less important distinctions in the narrower spaces ; or if these spaces 

 are small and more nearly equal, it becomes difficult to decide what 

 shall be species, and what shall be varieties, what shall be genera and 

 what shall be only sections of genera. Now it is well known that 

 some classes and families in the animal kingdom have the spaces on 

 this scale of differences more unequal than others have, so that most of 

 their species and even of their more comprehensive groups appear to 

 be well defined by the broad spaces. But other families, as the Naia- 

 dce and Melanidas of the United States, and the Helicidce and Cyclos- 

 tomidaj and especially the Geomelanise of Jamaica, have these spaces 

 small and nearly equal. In such families there is therefore an intrin- 

 sic difficulty in the w^ay of describing good species. The preconceived 

 notions that many persons entertain, and with which probably most of 

 our working naturalists began the study of nature, lead them to expect 

 in all cases very distinct natural limits of species. Hence they are 

 liable to ascribe a want of discrimination to the naturalist, who shall 

 have undertaken to describe such groups. 



The progress of discovery tends, more or less rapidly in different 

 classes, to fill up the wider spaces existing between known types. 



The laws of hybridity do not conflict with these views of the na- 

 ture of species. Hitherto observations relating to them have been 

 chiefly directed to species which are clearly separated from their con- 

 geners. So far as observations extend, the fiicility of hybridity ap- 

 pears to be inversely as the space between the types. If this could 

 be proved to be a strict law, obviously hybridity would furnish a 

 measure of these interspaces, or conversely might be predicted from 

 them. 



If these views are correct, we cannot retain the prevalent theory, 

 that species are natural gi'oups, but that genera and other groups are 

 artificial. All gi'oups are natural, so far as they are founded on actual 

 types. Any groups are artificial, or to speak more correctly and in- 

 telligibly, are arbitrary, in proportion as the differences between indi- 

 viduals, varieties, species, and the sucessively more comprehensive 

 groups, present a series of an indefinite number of nearly equal de- 

 grees. In proportion to the equality of the degrees, are the points, 

 at which we mark off species and genera &c., arbitrary. 



It must be acknowledged that these views are not essentially differ- 

 ent from the theory of spontaneous generation, when that; theory is dis- 

 encumbered of speculations on the nature or conditions of a supposed 



