THE PEAR. 91 



(D, figs. 81 and 82). The posts are thus solidly fixed 

 from the top to the bottom. Then extend from the 

 front and back of the two posts other iron wires 

 (B, figs. 81, 82, and 83), traversing a ring on the sides 

 of the posts. These wires have also each a screw 

 tightener. Finally, fix upon these last four wires (B) 

 on each side, a series of wood laths about an inch and a 

 half wide, and a third of an inch thick (E F, figs. 81, 

 82, and 83). Fasten the laths by a knot of fine wire, 

 12 inches apart upon the surface of the wires back and 

 front, alternately, as shown at fig. 82. 



Next proceed to the planting. The trees must be 

 planted back and front, a tree against each of the 

 laths to be trained as single vertical cordon espaliers 

 (fig. 80). In the front of each of the borders, a line 

 of apple trees may be planted, to be trained in low 

 horizontal espaliers (N, figs. 83 and ] 12.) 



Let us compare, for a moment, the results of this 

 new mode of training with the results of the old 

 methods. Suppose two fruit-gardens, both having 

 exactly the same extent of surface, and both equally 

 divided into borders six feet wide, separated by roads 

 40 inches in width, which is the common arrangement 

 for plantations of pyramidal trees. Let one of these 

 gardens be set apart for pyramids, and the other for 

 double contra espaliers. The total length of wood- 

 branches that we can obtain from pyramids may be set 

 down at 2,500 yards, and the contra espaliers will give 

 a length of 5,000 yards, double that of the pyramids 

 on the same extent of border. 



It must be added to this that the maximum product 



