EVOLUTION AS A FACT 479 



but this is by no means universally the case. Indeed, it is by 

 no means easy to exactly state what a species is. Some such 

 definition may be given, for example, as the one by Swainson: — 

 "A species, in the usual acceptation of the term, is an animal 

 which, in a state of nature, is distinguished by certain peculiarities 

 of form, size, colour, or other circumstances, from another animal. 

 It propagates, ' after its kind ', individuals perfectly resembling 

 the parent; its peculiarities, therefore, are permanent." But un- 

 fortunately there are such things as varieties or races, which are 

 subdivisions of species, and might be taken for such if seen in 

 a museum. In the case of the Field Snail {Hehx horiensts), for 

 example, there are many such races, distinguished by variously 

 coloured and banded shells. But in cases like this we usually 

 find that the different varieties, when crossed, produce offspring 

 (mongrels) which are perfectly fertile as regards one another 

 and the parent varieties. On the other hand, the offspring 

 (hybrids) produced by crossing two undoubted species are usually, 

 but not always, infertile. A notable instance is seen in mules, 

 which are obtained by crossing horses i^Equus caba/his) with 

 asses [E, asimis). We further find that two or more apparently 

 distinct species may be connected by a series of intermediate 

 varieties. This is beautifully seen in some of the extinct Lamp- 

 Shells and Snails, while the early turns of the spiral in some 

 Ammonites (see p. 465) may resemble one adult species, though 

 the later turns may correspond to another adult species. Facts 

 of the kind cited, while only susceptible of interpretation in a 

 mystical manner by the doctrine of special creation, harmonize 

 very well with the evolution theory, according to which organ- 

 isms are constantly being adapted to changing surroundings, and 

 new specializations are continually coming into existence. On this 

 hypothesis we may regard varieties as species "in the making". 



Species are aggregated into larger groups known as genera, 

 these into families, and so on, to orders, classes, and phyla or 

 sub-kingdoms, respectively marked out by agreements and differ- 

 ences of increasingly broader and more general kind. If these 

 various groups of, e.g., Backboned Animals, are diagrammatically 

 arranged so as to best express their mutual arrangements, a 

 tree-like arrangement results (see vol. i, p. iii), the phylum 

 corresponding to a main branch. This was perceived in pre- 

 evolutionary days, and the only rational explanation so far given 



