58i 



lining?. The swelling can become so great that even the outermost mem- 

 brane tears (p) and this tearing, as a rule, extends to several adjacent cells, 

 so that the changed secondary membrane with the swollen intercellular 

 substances, coalesces into a uniform yellowish to brown stripe in which 

 are recognizable parallelly deposited remnants of the primary membrane 



(St). 



It has thereby been proved experimentally that processes of loosening 

 are initiated in the cell membranes by frost. These become apparent in 

 the so-called "frost lines." 



Internal Splitting of the Trunk and Branches. 



In the section on frost blisters, the disturbances were considered which 

 take place in smooth barked branches and trunks without any external 

 injury being noticeable at first. Not until the year following the produc- 

 tion of the blister do the primary bark layers, which cover the frost blister, 

 rupture because this blister has gradually enlarged. These torn bark 

 layers surround the protruding new structure as dried edges. The cause, 

 however, is to be seen only in the raising up of the bark layers, without 

 any splitting of the wood. 



If, however, the occurrence out of doors in so-called frost holes, i. e., 

 places where late frost occurs almost annually and very extensively, is 

 more closely examined, blister-like protuberances will be found on the 

 branches and trunks which internally show repeated splittings of the annual 

 ring. 



I have accidentally succeeded in producing such blisters artificially bv 

 exposing to sharp brief frost action branches in which the wood ring of the 

 current year had attained a considerable thickness. The subjoined Fig. 130 

 represents a healed inner wound, due to the splitting of a cherry branch. 

 The frost wound has been produced by a one-sided raising of the bark from 

 the young wood, a is the old wood of the previous year; b, the spring 

 wood of the current year, formed before June; g is the sapwood region 

 with the normal cambial zone. About this time the branch vv'as placed in 

 a freezing cylinder. It was found, in the subsequent investigation, that the 

 bark had been split off from the sapwood in a wide curve (s p) and that 

 the young wood (b) seemed split radially. The splitting extended along 

 the medullary rays which more rarely were torn apart than loosened at one 

 side from the prosenchymatous cells and ducts and then partially dried up. 

 A radial enlargement of the holes represented at in the drawing takes 

 place in many cases because of the extensive drying of the prosenchy- 

 matous cambial elements which are still partly thin-walled. In general the 

 radial clefts in the wood remain slender and only the walls of the elements 

 which drew apart from one another turn a deep brown. 



Near the breaking buds in which a medullary bridge traverses the 

 whole wood body from the pith to the bark in all trees, the tissue is more 

 tender, the number of thick-walled cells is less; the elements lying next to 



