224 PROCESS OF DEVELOPEMENT. CHAP. IV. 



as may be seen in the case of the common Cherry 

 when the blossom becomes double, the stamens 

 being converted into petals, and the pistil not un- 

 frequently into a leaf. A similar phenomenon may 

 be sometimes observed also in the fruit of the Pear- 

 tree. In the summer of 180Q, I observed in the 

 garden of Rendlesham Parsonage two Pears, of 

 nearly the size of the rest on the same tree, having 

 each several leaves growing out of it, with the 

 insertion about half way between the base and 

 apex of the fruit : I do not know what particular 

 variety of Pear it was, but it was a large and dusky- 

 coloured winter Pear, perhaps the winter rousselet. 

 The same thing may be also observed in the fruit of 

 the Medlar ; and hence the success of Mr. Knight's 

 graft of the fruit-stalk on the leaf-stalk is the less 

 surprising. The fact then is that the new layer 

 Andde- l8 formed not absolutely from the bark, as the 

 experiments of Du Hamel might seem to indicate, 



the return- k u t from the proper juice descending through the 

 returning vessels of the leaf, leaf-stalk, and inner 

 bark, from the summit to the lower extremity of 

 the plant ; in tracing which vessels Mr. Knight 

 thought he could even discover two distinct sets, one 

 for forming the new layer of wood, and another for 

 forming the new layer of liber. This is a very nice 

 distinction indeed, and a fact, if it is the fact, diffi- 

 cult to ascertain. It is certain, however, that either 

 two layers are originally generated ; or that the ori- 



