CONCERNING SITE 387 



noting, however, and is beautifully exemplified in the table presented, 

 that those who advocate height on age or ultimate height as the cri- 

 terion of site are up against exactly the same difficulty in the matter 

 of the influence of stocking as is experienced in attempting to deter- 

 mine the normal increment; and what is worse, under optimum 

 moisture conditions "crowding stimulates height growth" according 

 to the old proverb, while it is plain from the table that under other 

 conditions crowding retards height growth. 



And here I wish to bring forth an idea which is somewliat relevant 

 to this discussion and which, I think, is worthy of some consideration. 

 Height is solely controlled by the moisture of the soil. Height can- 

 not, therefore, be an index to the productive capacity of the site, the 

 latter being quite as much dependent on insolation and heat of other 

 sources. Both heat and moisture are necessary to produce starch and, 

 ultimately, wood. 



Pearson has recently informally announced the discover}- in Ari- 

 zona that the annual height accretion is controlled by the precii)itation 

 during the short period of early summer when height growth is made. 

 But the control of height by moisture goes much further than this. 

 It goes to the point of ultimate height. Just as a thoroughly efficient 

 vacuum pump will lift water only about 30 feet and mercury only about 

 30 inches, so a tree having a given osmotic pressure may lift water of a 

 given density (of salts in solution) only so far. Thus the height of the 

 tree is limited on the one hand by the density of the soil solution, 

 whicli is, of course, least when the moisture is greatest; on the other 

 hand, by the density and osmotic pressure of the solution in the upper- 

 most extremity of the tree, which in turn is limited by hygroscopic 

 qualities of the specific protoplasm. It is a pure question of physics 

 that there must be a definite gradient in sap density from the root hairs 

 to the to)) of the tree, to maintain osmotic transfer of water at a definite 

 rate. Height growth, then, ceases when the minimum gradient com- 

 mensurate with the demands of the tree and the maximum density 

 which can be tolerated by the protoplasm of the topmost cells have 

 been reached. When the tree can no hjiiger grow upward, it "flattens 

 out," as we say. Every one knows that beyond this stage it may still 

 be capable of its best volume increment. 



A young tree grows rapidly because the density gradient may be very 

 high. Similarly, a crowded tree grows rapidly upward bccau.sc it has a 

 minimum of lateral branches to supply with water, and the existing 

 gradient is, lluri-fure, more than adecjuate to maintain the necosary 



