172 



THE GARDENER'S ASSISTANT. 



season, the spaces between d and /, also between 

 p and n, are widened so as to leave room for 

 another branch on each side, e o. All young 

 growths are shortened back to firm wood as 

 before, and shoots are left when disbudding to 

 form leaders, for bearing wood wherever there 

 is room, for the two new branches, e and o, 

 and for four more in the centre, hi hi. Thus 

 at the end of the season there will be eighteen 

 principal branches, which are to be trained as 

 the dotted lines show. These will be quite 

 sufficient to furnish the whole tree with bearing 

 wood. 



Many other modes of training are practised, 

 but most of them are open to the objection of 

 encouraging one or two leaders on each side 

 which serve as bases for the other branches. 

 In a properly -trained fan -shaped tree all the 

 branches should be of equal strength, and dis- 

 posed so as to divide the sap equally over all 

 portions of the tree. They are also con- 

 veniently placed for removal should a branch 

 die, as the others can be lowered to fill its 

 place, and the centre refilled with young 

 branches. 



In training the Peach an equality of vigour 

 amongst all the branches should be aimed at. 

 The branches should be frequently examined in 

 detail; a comparative inspection ought to be 

 made of every three, of the lowest three with 

 the next three, and so on ; and again, every three 

 on the one side with the three opposite on the 

 other side. If one side of the tree is found to 

 be weaker than the other, the branches of the 

 weaker side should in general be elevated above 

 their assigned position, and, on the contrary, 

 three of the stronger side should be depressed. 

 The latter ought also to be disbudded, and their 

 summer-shoots nailed in before those on the 

 weaker side of the tree; over-vigorous shoots 

 should be checked at an early stage of their 

 growth. All superfluous shoots ought to be 

 cut away. Whilst a sufficient supply of shoots 

 for succession must be encouraged, none beyond 

 these should be allowed to exist, with the ex- 

 ception of those necessary for leaders. If, in 

 the early part of the season, it is seen that a 

 succession shoot will unquestionably become too 

 strong, it should be stopped at an early stage of 

 its growth. 



With regard to the weak side, an opposite 

 mode of proceeding should be adopted. Shoots 

 not absolutely required for succession may very 

 properly be allowed to grew. The succession 

 shoots generally should be trained at full length, 

 and where there is space, the shoots at the extre- 



mities of the bearing shoots may also be left 

 unshortened. If a few over-strong shoots should 

 start on the weak side, they had better be 

 checked; but all others on that side ought to 

 be encouraged. 



If the leaves on one side of the tree were 

 equal in number to those on the other, then 

 the amount of vigour would also be equal, pro- 

 vided the leaves on both sides were of the same 

 average size and equally healthy. This is a 

 fact which should be borne in mind; but in 

 stating it we do not mean to imply that the 

 leaves should be counted, or their dimensions 

 calculated. In practice, a sufficiently correct 

 estimate of the relative amount of foliage on 

 both sides can be formed by looking at the 

 foliage on branch a, fig. 975, then at that on 

 the opposite branch s, and so on. Presuming 

 that a, b, c and d, e, f are, according to their re- 

 spective lengths, equally furnished with leaves, 

 then the quantity of foliage borne by the first 

 three, as compared with the three above them, 

 will be nearly as 7 to 9 ; that being about the 

 proportion which the united length of a, b, c 

 bears to that of d, e, f. In that proportion, 

 therefore, the increase of the base of the 

 branches d e will exceed that of the base of 

 a, b, c; and the consequence will be that the 

 former must every year become so much 

 stronger than the latter, instead of which it 

 would be desirable that they should be of equal 

 vigour and thickness. This condition would 

 be obviated by not allowing the branches to 

 extend beyond the semicircle a s, but then, on 

 referring to the figure, it will be seen that a 

 large portion of wall is left uncovered. 



Supposing the trees to be planted 20 feet 

 apart, and the wall to be 12 feet high, the space 

 which a tree like that presented in the figure 

 might occupy would be 240 square feet; but if 

 limited within the semicircle, it would only 

 cover 166 square feet, leaving 74 square feet, 

 or nearly one-third of the space, unoccupied. 

 Rather than this should be the case, it would 

 be almost better to put up with the loss of the 

 three lower branches on each side. But, in 

 order to guard against this, something may be 

 done; allowing the radius of the semicircle to 

 be 10 feet, then the distance between the 

 branches where they intersect the dotted arc 

 line — that is to say, when they have extended 

 10 feet from the centre — will be very nearly 

 21 inches, which would be wider than necessary. 

 To fill that space branches may be originated 

 from the upper side of a, b, c, at about 5 feet 

 from the stem. There will then be six branches 



