788 A TEXTBOOK OF THEORETICAL BOTANY 



this point of view, for although endogenous branching is usually reckoned to 

 be one of the fundamental characters of roots it is, in fact, by no means 

 without exceptions. Adventitious roots in particular often originate 

 exogenously, though their subsequent branching is endogenous. 



Attempts have been made to distinguish between different parts of the 

 root system as functionally either anchoring or absorbing systems respec- 

 tively. Research into the root systems of fruit trees has shown that, at 

 least in young trees, it is possible to distinguish two separate systems. Firstly, 

 an absorptive system of fibrous roots which remain relatively near the 

 surface of the ground and do not grow out beyond the spread of the aerial 

 branches, and, secondly, stout anchorage roots which are few in number 

 and are comparatively little branched. These latter, after running more or 

 less horizontally just below the soil for a distance of several feet around the 

 tree, turn and dive vertically downwards to a considerable depth. These 

 anchorage roots apparently arise only during the early growth of the tree 

 and cannot usually be replaced if they are once damaged. It is a familiar 

 sight to see fruit trees in commercial orchards leaning at various angles and 

 only a few standing vertically. This is probably due to the ploughing of the 

 ground among the young trees to keep down weeds. In so doing some of 

 the anchorage roots may be cut and, since they cannot be replaced by the 

 tree, it loses support and can be readily blown sideways by the prevailing 

 winds. 



In general, however, the functional distinction between different parts of 

 the root system is seldom complete. The principal distinction which actually 

 exists is in respect to the reaction to gravity. The main root is the only one 

 which is positively geotropic. Each branch of the main root, i.e., branches 

 of the second order, sets itself at a characteristic angle to the vertical, which 

 is smallest in those nearest to the apex and gradually increases up to a right 

 angle in the branches nearest the surface of the soil, which therefore grow 

 horizontally. 



Branches of the third order have no geotropic reaction and simply grow 

 at right angles to their parent roots. These variations in the direction of 

 growth naturally lead to the spreading of the branches away from each other, 

 so that they are well spaced out and permeate the soil most effectively. 



Variations of angle and of the rate of growth in different parts of the system 

 naturally cause a good deal of variation in the form of the system and perennial 

 plants can be roughly classified, for instance, into surface-rooters and deep- 

 rooters. Xerophytes, like the Cacti, often show both tendencies in an extreme 

 state, their deep roots penetrating downwards to the permanent water table, 

 which may be 30 ft. down, while their surface roots grow horizontally for 

 many yards, just below the surface, where they may pick up the moisture 

 from every light shower. 



Root branches do not appear on the parent root with any regularity, 

 and there is no transverse division of a root into joints as there is in a stem. 

 The branches do, however, form a definite number of longitudinal rows, 

 corresponding to the number of xylem strands in the parent root, except in 



