THEOET OF THE GEOWTH OF CUTTINGS. 411 



the distribution of roots, then their growth at the lower end 

 of a bramble-branch would correspond with their growth at the 

 lower end of a cutting. But further observation of the habits of 

 the Bramble make it almost certain that this cannot be the case. 

 "Where a Bramble grows on a steep bank, a large number of the 

 branches grow simply down the slope, a further number grow more 

 or less horizontally along the bank and ultimately grow down 

 hill. But a few branches will be found which grow up the slope, 

 and some of these form roots. I have observed this fact among 

 the Brambles clothing a steep bank, cut in the chalk, many parts 

 of which are inclined at 55° to 65°. 



The following are the details of some observed cases. The 

 measurements are taken from the root of the plant to which the 

 branch belongs. At the apical end of the branch the measure- 

 ments are taken to the surface of the soil where the branch enters 

 it, which is almost equivalent to the spot where the roots are pro- 

 duced. 



1. A branch grew well above the horizon for several feet, 

 then 6 inches at 10° above the horizon, ending with a single inch 

 2° or 3° below the horizon, and then produced roots. 



2. Grew for 2 feet 10 inches nearly vertically upwards (includ- 

 ing the parent branch, from which the branch which ultimately 

 took root sprung), then 8 inches at 15° below the horizon, and 

 produced roots. 



3. Grew upwards and highly inclined for 2 feet 3 inches, then 

 8-9 inches at 5° below the horizon, and produced roots. 



4. The first 4 feet were at 10° above the horizon, then came 

 1 foot 2 inches, which formed a curve (concave towards the bank) 

 whose chord was at 55°. The terminal 2 inches were at 5° below 

 the horizon ; and the extremity had taken root. 



5. The first 1 foot 6 inches were at 60° above the horizon, then 

 came 2 feet 3 inches horizontal, then 2 feet 8 inches 30° above, and 

 at last 2 or 3 inches at 5° below the horizon, then the roots at 

 the end. 



It seems hardly possible that the terminal portion (from 1 to 

 9 inches in length), being a few degrees below the horizon, can 

 determine the production of roots at the apex. "VVe must, I 

 think, conclude that when the apex of a branch can, as in these 

 cases, reach the soil without hanging downwards, it is able to 

 produce roots at the apex. This is quite inconsistent with the 

 belief that the distribution of root-development is determined 



