193 



of the top shoot which I had lately cut from below the dead 

 part. 



" With similar objects in view, a second series of ex- 

 periments have been made, at various times within the last 

 twelve years, by planting cuttings of free-growing plants 

 with their tops downwards. Placed in this way, adventitious 

 roots were protruded, and the plants grew. Cellular granu- 

 lations at first appeared on the end which was now uppermost, 

 and out of the ground, a circumstance which militates against 

 the inference drawn by some, namely, that the physical law 

 of gravitation operates in causing the sap to descend. 



" In conducting this experiment I have invariably found, 

 that no cellular callus formed at the lower extremity, as would 

 have been the case had I planted the cutting in the regular 

 way. The young roots were protruded laterally from the 

 bases of leaf-buds under ground ; when one or more of these 

 elongated, the axis made a sharp curve upwards, until it re- 

 gained its natural position. The growth and woody forma- 

 tion went on then in the usual way. In some cases the 

 portion of the cutting above ground remained alive during 

 a considerable period, though no leaf-buds grew on it. It, 

 however, soon died after the ascending shoot gained strength. 



" This experiment, in my opinion, also tends to prove, 

 that no regular return of assimilated bark-sap takes place in 

 the formation of wood ; because, if such were the fact, the 

 position of the cutting above ground would have lived, and 

 continued to receive the annual deposits, which was not the 

 case. 



" The beautiful example I have laid before the Academy, 

 of the junction of stock and graft, proves, beyond any manner 

 of doubt, that the two increase by separate growths of their 

 own wood, as thoroughly as if they still grew on separate roots. 

 I cannot, therefore, see how this fact can be got over by 

 those who hold that exogenous plants increase by annual de- 

 posits of bark-sap. It will not, however, do to draw final 



vol. v. o 



