64 



[»(';iI(mIIv (Iciiioiisl i;iI(mI in tlir ((tiirsc oC ('.\|K'i-iiii('iil;i I ^\(»l•|<, (luit 

 arc A\"()i-lli\' of (•((iisi(i('i-al ion al lliis (iiiic. 1 ^\an( (o iiiciitiou 

 and very liricHy discuss six of these: 



(Ij. Lateral or obli(j[ue coiHluction. 



There seems to be a rather widespread (l»nl erroneous) idea 

 that the crude and ehiborated saj) of a tree can pass up and 

 doAVU the trunk or branch only in a longitudinal direction, that 

 is, leuiithAvise of the fibres or "lirain" of wood or bark, or at most 

 with but sliiiht (h'viation from this i-oute. The fact that it is 

 transferred almost entirely in a ionjuitudinal direction in a 

 healthy uninjured tree may be true enoutih under noniial c<)n- 

 ditions, but it is far from true in trees that have been injured 

 in certain ways, and, as alT students of plant physioloij;T know, 

 not strictly true under perfectly normal conditions. 



It is a fact of common knowl(Ml<;e that a tree will (»r<liiiarily 

 cover or ^row over, an area of bare wood wliere the l)ark has 

 been removed. It is conuiiou knowiedj^c to all observant persons 

 that these scars heal over mainly from the sides. In all proba- 

 bility this is largely because thej'^ adjoin the uninjured vessels 

 through wliich sap is being conducted in t he normal longitudinal 

 direction, but doubtless in part also to other causes to which I 

 shall allude directly. If a partially or entirely healed over scar 

 slionld be dissected, it will be found that in the layers of wood 

 formed immediately after the injury the fi))res are curved out- 

 AN'ard around the injury, and continue in a nearly hnigihidinal 

 direction both above and below the scar. When the scar is par- 

 tially covered, the newly formed fibres are straighter, and finally 

 after the scar is entirely covered, the youngest lil)res will be 

 found to have assumed their normal longitudinal direction, or 

 very nearly so. 



If it were not for this possibility of oblique conduction, a tree 

 that had a large lesion extending lialf way around the trunk 

 on the north side, for instance, and an equally large one on the 

 soulh side, either above or below the other, would, to all intents 

 and purposes, be girdled. 



In the chestnut tree, the angle from the perpendicular to 

 which these fibres can be made to curve, as a result of experimen- 

 tal cuttings, may seem surprisingly great. In one instance the 



