368 Scientific Intelligence. 



paration is made before-hand for the developement of the roots, and 

 whence originate those which are developed along the branches of trees, 

 whether in the air, in the water, or when buried in the earth. 



%dly, The young root communicates with the ligneous body of the 

 branch by its axis, which visibly proceeds from it ; and its bark appears 

 also to be a continuation of that of the branch. 



3dly, It pierces the epidermis of the branch in its developement, and 

 carries with it some portion of the cellular covering. 



\>thly, The root increases only at its extremity, which part alone, of the 

 whole root, is capable of becoming green by the action of the light. 



6thly, The developement of the roots is generally more rapid in the 

 dark than in the light, although experiments produce very varied results. 

 6thly, Branches immersed in water do not sensibly imbibe the fluid by 

 the bark, but by the exposed parts of the woody body, whether in conse- 

 quence of a transverse section, or upon the external surface simply deprived 

 of the bark. 



Ithly, Water absorbed by the woody body has a tendency naturally up- 

 wards, towards the upper parts from which it has entered ; because the 

 leaf-buds, when they are excited by heat, attract towards them the water 

 absorbed by the roots or the exposed wood, and it is by a similar process 

 that the trees are enabled to produce leaves in spring. 



Sthly, Water penetrates more slowly in branches which are inverted, 

 than in those in a natural position. 



9ihly, Water absorbed by the cut base of a branch, unites with the co- 

 louring matters with which it may be charged in the branches which are 

 developed by this action. 



10th, The roots which are put forth in coloured water, imbibe from it 

 the colouring matter, which they transmit to those roots which are above 

 them, but which are themselves not immersed in the coloured liquid. 



11th, The length and ttye form even of the roots may be much modified 

 by the nature of the medium in which they grow. 



47. New plan of grafting- Pear trees. — A very interesting paper on the 

 cultivation of an early and a late variety of the pear on the same wall-tree, 

 by Mr D. Montgomery, gardener to the Duke of Montrose, is printed in 

 the Transactions of the Horticultural Society of London, vol. vi. p. 367. 

 He proposes to graft the half of the late pear-trees with the early sorts, and 

 half the early trees with the late sorts; for example, every alternate branch 

 of the crassane with the jargonelle, and of the jargonelle with one of the 

 best late pears. In this way there are two chances of success. Should the 

 jargonelle, which is very early in blossom, fail, from unfavourable weather, 

 the late sort, which flowers at another time, may succeed. Another 

 advantage arises from the crop coming at different times. The jar- 

 gonelle ripens off before much effort is required from the tree to support 

 the late sorts, so that the tree is more capable of supplying nourishment to 

 half a crop of jargonelles, than if the crop were all of that sort ; and as the 

 early pears are all gathered before the late sort begins to swell to size, the 

 tree is at once relieved from half its crop, and is better able to mature in 



