758 JOURNAL OF FORESTRY 



growth as a site-index, however, which must be met. While not insur- 

 mountable, they point to precautions and modifications in the simple 

 height-growth classification which must be accepted if the method is to 

 receive serious consideration. The most important of the objections 

 are: (i) That height growth is too sensitive to incidents in the history 

 of the stand, such as origin (sprout or seed), changes in density, and 

 interferences in the normal growth of the stand, such as culling, fire, 

 grazing, etc.; (2) that the mere determination of site on the basis of 

 height growth tells nothing about the factors which produced it, nor 

 what the same site would produce if other species are grown, nor 

 whether even the same species would grow equally in height on the 

 same site a second time under the same or a different kind of treat- 

 ment,^^ and (3 ) that a given species may exhibit the same height growth 

 on widely different sites, as in swamps and on dry uplands. 



With respect, first, to the infinite possibilities in the history of the 

 stand which may affect height growth, we must admit the justice of 

 this objection, at least in so far as it concerns the ultimate, precise 

 classification of local sites which it is probable the future will bring. 

 As to the larger determinations in the w'ild woods, culled or virgin, it is 

 necessary to assume, until the contrary is proved, that the effect on 

 height growth of the physical factors is far more potent than the effects 

 of all interfering agencies which are not so plain to the eye that they 

 would be discounted or excluded as a matter of course. There is a 

 certain consistency in the variation both in crown height and merchant- 

 able length^- of our native forests, in spite of their checkered careers. 

 In the mature forest, therefore, the height growth having practically 

 ceased, the average total height of dominant trees of each species can 

 be quite safely regarded as reflecting the relative possibilities of the 

 site. This has been pointed out by Roth.^" In the juvenile stage height 

 growth would certainly be misleading, as it is known to be more or less 

 independent of site.^* Between these two extremes the effects on 

 height growth of past changes in density, of fire. etc.. must be avoided 

 by careful choice of specimens for measurement — healthy trees which 

 have alwavs been dominant. 



"These points are discussed by Schiffel (Ueber Bestandeserzichung. Cent, 

 f. d. g. Forstwesen, Aug.-Oct., 1906, pp. 333-5. 405-425- Reviewed in For. Quart., 

 IV, 4, pp. 326-336) and Vater (Ueber die Anstellung waldbaulicher Versuche 

 und ueber die Klassen der forstlichen Ertragstafeln. Tharandter forstliches 

 Jahrbuch. 1912, pp. 252-264. Reviewed in For. Quart., XIII. 3, pp. 391-4)- 



'■ Merchantable length is, in fact, the present basis of site distinction used by 

 the Forest Service in the Southern Appalachians. 



" Op. cit. 



" Mayr, Heinrich : Waldbau, p. 132, 1909. 



