Chap. XXVI. SUMMARY. 355 



certain parts have been greatly modified to serve some useful 

 purpose; but we almost invariably find that other parts have 

 likewise been more or less modified, without our being able to 

 discover any advantage in the change. JSTo doubt great caution 

 is necessary in coming to this conclusion, for it is difficult to 

 overrate our ignorance on the use of various parts of the organi- 

 sation ; but from what we have now seen, we may believe that 

 many modifications are of no direct service, having arisen in 

 correlation with other and useful changes. 



Homologous parts during their early development evince an 

 affinity for each other,— that is, they tend to cohere and fuse 

 together much more readily than other parts. This tendency to 

 fusion explains a multitude of normal structures. Multiple and 

 homologous organs are especiaUy liable to vary in number 

 and probably in form. As the supply of organised matter is not 

 unlimited, the principle of compensation sometimes comes into 

 action; so that, when one part is greatly developed, adjoining 

 parts or functions are apt to be reduced; but this principle is 

 probably of much less importance than the more general one 

 of the economy of growth. Through mere mechanical pressure 

 hard parts occasionally affect soft adjoining parts. With plants 

 the position of the flowers on the axis, and of the seeds in the 

 capsule, sometimes leads, through a freer flow of sap, to changes 

 of structure; but these changes are often due to reversion. 

 Modifications, m whatever manner caused, will be to a certain 

 extent regulated by that co-ordinating power or nisus formativus, 

 which is in fact a remnant of one of the forms of reproduction 

 displayed by many lowly organised beings in their power o 

 fissiparous generation and budding. Finally, the effects of the 

 laws which drrectly or indirectly govern \ri,m^t 



-ce will be favoured and ll J^ ££$%£ "* 



^TZ^t^^ r Us, offrom 

 derived from t.W ™T hh to revert to characters 



derived from their common progenitor, and, as they have much 

 m common m their comfitu^™ xi y mucn 



chaneed conditions to ' they are also Iiable ™der 



changed conditions to vary m the same manner; from these 



2 a 2 



