18 HORTICULTURAL MANUAL. 



applies to the surface-roots which are capable of develop- 

 ing adventitious buds naturally or when injured. The 

 deeper water-feeding roots do not develop buds soon 

 enough for use in propagation, hut if they come to the air 

 by soil-erosion or in other ways they soon change into 

 surface-roots with an ample su23ply of starch for the 

 development of buds and even upward growth. It is also 

 easy to graft pieces of the surface-roots successfully, but 

 in no case have we known the pieces of the deep roots to 

 unite with a scion, as they are not well stored with 

 starch. 



21, Roots as Modified by Variety of Top. — The culti- 

 vated orchard fruits vary peculiarly in the manner of root- 

 growth when grown on their own roots or grafted. With 

 a given lot of apple-seedlings, if we graft enough for a 

 nursery row of Soulard Crab and for another row of Ben 

 Davis, we find on digging when two years old that the 

 Soulard roots run down and are as difiicult to dig as pear- 

 trees. But the Ben Davis row we find easy to dig, as it 

 has a great supply of fibrous surface-roots, and the deep 

 extending ones are small but numerous. Indeed, with 

 the same seedlings we find that the roots of all the varieties 

 will maintain their usual characteristics of growth when 

 on own roots. 



Every nurseryman is familiar with the fact that when 

 one hundred varieties are root-grafted or budded on the 

 same lot of seedling stocks, each variety when dug will 

 show its characteristic roots. The Eed Astrachan roots 

 will be fibrous, branching out near the surface, with few 

 deep roots; while the Duchess, Fameuse, and Hibernal 

 will show few fibrous roots, but several pronged coarse 

 ones, one. of which, or more, runs deep into the subsoil. 

 This is also true of the pear and other fruits. If the same 

 pear-seedlings are grafted in part with Bartlett and a part 



