48 



REPORT ON THE 



apart, with the exception of some where the buds intended to 

 produce branches did not break, as is often the case ; then a 

 shoot, <z, is trained up, and, when growing in summer, a small slice 

 is taken off near its extremity, and a corresponding extent of sur- 

 face immediately below the inner bark of the stem is exposed ; 

 the two are joined together, and the point of the shoot a is in- 

 clined in the direction to form the branch c. 



The most remarkable feature in the trees at Corbeil was the 

 uniformity of vigour in the respective branches. It appeared as 

 if the supplied branches, ccc, had been allowed to grow in con- 

 nection both with the stem at b b, and the branch from which they 

 originated at a a a, till their length and thickness corresponded 

 sufficiently with that of the branches above and below them. This 

 is a great advantage which the mode possesses over budding or 

 side-grafting. At the distance of a foot apart for the horizontal 

 branches, it takes as many years to cover the wall as the latter is 

 feet in height ; for although the leading shoot may grow three or 

 four feet in length in a season, yet by shortening it to two feet, 

 although the branches dd would be produced, the buds at b b, to 

 furnish the intermediate stage, most probably would not. In 

 fact, the attempt to form two tiers of horizontals in one season is 

 generally followed by more or less disappointment. The inter- 

 mediate stage might, however, be readily supplied by the method 

 above detailed ; and a wall twelve feet high might be covered 

 as well in six years as it otherwise would be in twelve. 



We had not time to ascertain the name of the place where 

 these Pear-trees are growing, nor that of the gentleman who 

 owns the property. He is said to be enterprising, and carries 

 on some manufactory closely adjoining. The place is situated 

 some distance south from Corbeil. Thither, in all haste, we re- 



