80 PROPAGATION BY DIVISION. [CHAP. IV. 



desirable, not only that the kinds to be united 

 should be of the same genus, or at least of the 

 same natural family, but that they should agree 

 as closely as possible in their time of leafing, 

 in the duration of their leaves, and in their 

 habits of growth. This is conformable to com- 

 mon sense; as it is quite obvious that unless 

 the root send up a supply of sap at the time 

 the leaves want it, and only then, the s;raft 

 must suffer either from famine or repletion. 

 For this reason, a deciduous plant cannot be 

 grafted on an evergreen, and the reverse. The 

 necessity of a conformity in the habit of growth 

 is strikingly displayed in Mr. Loudon's Arbo- 

 retum Britarwicum, in a wood engraving of a 

 flowering ash grafted on a common ash, and 

 growing at Leyden ; by which it is shown, that 

 an architectural column, with its plinth and 

 capital, may be formed in a living tree, where 

 there is a decided difference in the growth of 

 the stock and the scion. 



These examples show that no intimate union 

 takes place between the scion and the stock; 

 and the fact is, that, though they grow together 

 and draw their nourishment from the same 

 root, they are in every other respect perfectly 

 distinct. The stock will bear its own leaves, 

 flowers, and fruit, on the part below the graft ; 

 while the scion is bearing its leaves, flowers, 

 and fruit, which are widely different, on the 

 part above the graft. Nay, five or six grafts 

 of different species on the same tree will each 

 bear a different kind of fruit at the same time. 

 This want of amalgamation between the scion 

 and the stock is particularly visible in cases of 



