INTEODUCTION. XXV 



earlier naturalists, to furnish sufficient characters 

 whereby to distinguish the primary divisions of 

 the animal world. It was afterwards ascertained 

 that these tests, however suitable they might at 

 first sight appear, were applicable only within cer- 

 tain limits. For it is easy to select two animals, 

 in all essential features strikingly dissimilar, 

 though upon examination it will be found impos- 

 sible to decide whether the organisation of one be 

 superior, as a whole, to that of the other. Hence, 

 all arrangements which exhibit the animal king- 

 dom as a continuous series, leading step by step 

 from the humblest of all animal forms to those in 

 which the specialisation of functions is most strongly 

 marked, are evidently unnatural.'^ For as there 

 are two distinct points of view from which all 

 animals may be considered, so also one animal 

 may differ from another either as to the greater 

 or less complexity of its structure, or the general 

 ^lan upon which its structure has been framed. 

 Every animal, as Huxley has well observed, may 

 be regarded as the resultant of two tendencies, the 

 one physiological, the other morphological.^ 



By applying a process of generalisation to the 

 numerous facts which the study of morphology 

 has disclosed, we are led to the conclusion that 

 five ultimate plans of structure may be traced 

 among animal forms. And accordingly the entire 

 animal kingdom may be divided into the same 

 number of primary groups, technically known 



