50 ANATOMY OF THE GYMNOSPERMS 



is subject to somewhat wide variation as between different spe- 

 cies and genera, and it is also found to be of a marked nature, 

 even in the same species under different conditions of growth. 

 For more extended details reference should be made to the table 

 in Appendix A, but for our present purposes a few examples may 

 be selected. Thus in all three of the investigated species of 

 Araucaria the ratio of the walls -- spring to summer wood --is 

 i : i, a relation which exactly corresponds with the absence of 

 growth rings. In Juniperus virginiana the ratio is i : 2 ; in Cryp- 

 tomeria japonica it is 1:4; in Tsuga Sieboldii it is 1:5. On 

 the other hand, the same species may exhibit two ratios as deter- 

 mined by the peculiar conditions of environment under which 

 the growth has been formed. Thus in Taxus canadensis we may 



have ' , L : 4.8 /u, ; in Larix Lyallii ., L : 8.4/u,; and in Libo- 

 4.SJ i.bj 



r 2.4 



cedrus decurrens we may get 1.4 : -I , /LA. While, therefore, it is 



[4.8 



obvious that there are certain ratio differences between the walls 

 of these two structural regions, --differences which are more or 

 less directly associated with species, --and while such differences 

 may be of some value in confirming conclusions derived from 

 other data, they are not of such a nature as to permit the formu- 

 lation of a general law applicable to species in such a way as to 

 establish a precise differentiation, since variation arises within 

 the same species as a result of different internal and external 

 conditions of growth, such as tension, soil, and climate, the lat- 

 ter being influenced by situation and exposure. In many cases, 

 where the actual thickness of the walls is the same, the apparent 

 thickness will vary considerably. This is directly attributable 

 to differences in the transverse volume of the tracheid, whereby 

 the spring tracheid, from its greater size, will have a wall rela- 

 tively thinner, and thus apparently thinner than that of the 

 summer wood. Our investigations show that for twenty genera 

 and one hundred and fourteen species the mean ratio is i : 2.12. 

 The following summary for genera may be consulted in this 

 connection : 



