PROPAGATION AND RAISING. 1^7 



&c„ is tlie same <is in the former case. This is somtimcs 

 useful when it is desirable to preserve the head, and ^-^^ 23. 

 perhaps the fruit on it, for the season ; but the ^ 

 newly-introduced part doos not often get strong || 

 during the time, and, on the whole, there is no ad- 

 vantage gained. 



Inarching is sometimes practised (fig. 29), which is 

 simply the uniting of two branches that are each 

 attached to neighboring plants. "Where it is intended ||j 

 the union is to be made, the bark, and also a thin ^ 

 slice of the wood (in thickness according to the size 

 of the intended new sort) must be cut away so that the 

 bark on both sides will unite ; the two are then brought 

 together, tied, &c., as in the former cases -p.„ 29 



of grafting. There is no advantage in this 

 plan, excepting that it may be performed 

 any time during the summer. It is, how- 

 ever, best to do it immediately after the 

 full expansion of the first leaves, which 

 better insures a perfect adhesion. When 

 it is thought that the union is complete, 

 which ought to be in five or six weeks, 

 the lower, part of the introduced branch 

 may be cut half through, and left so 

 while attached to its own parent. In another week, or 

 ten days, it may be entirely severed, and again left a sim- 

 ilar length of time, Avhen the head of the dishonored kind 

 may be cut away, and the new one allowed to take its 

 place. Without these latter precautions, there is great 

 danger of failure, as the sap may not have begun to flow 

 freely into the new parts. It should also be observed, 

 that in all the examples, when it is seen that the ligatures 

 begin to grip the bark, from the increase of growth, they 



