212 PROPAGATION. 



The crooked, dangling shoots of a jargonelle pear, for 

 instance, should not be chosen for grafts ; nor should 

 those of any other tree be selected which do not 

 show the desired properties as well in habit of growth 

 as in those of perfect health and maturity. 



That the character of the stock affects the growth 

 of the graft is well known. Rampant growing, and 

 consequently barren fruit-trees, by being worked on 

 dwarf-growing stocks, are greatly improved for the 

 purposes of the fruit-grower ; and the reverse of this 

 practice, i. e. placing weakly growing kinds on robust 

 growing stocks effects a similar improvement. 



These counter dispositions affect only the growth 

 of the sorts united to each other by grafting ; but 

 such unions produce other remarkable consequences. 

 One ot the most curious is the well-known circum- 

 stance of the variegation of the colour of the graft 

 appearing on suckers which rise from the roots of 

 the stock. 



This circumstance shows that there is some inter- 

 communication between the head and root, and must 

 be, it is supposed, caused either by a descent of some 

 member of the graft, or of its sap. No detachments of 

 the wood or bark can possibly be prolonged down- 

 ward ; because, as soon as these members are formed, 

 they remain ever after unchanged longitudinally. The 

 vital body is only capable of being so extended ; but 

 whether as fibres, or in any other way from the graft 

 or bud, it is difficult to conceive. If such a process 

 obtains in the jasmine, we may expect that some- 



