72 • Journal of the Mitchell Society [August 



ous sorts. The physical conditions, climatic and edaphic, which affect 

 the rate of growth and the composition of the forest, must be deter- 

 mined. The characteristic ecological associations and societies (for- 

 est types) must be ascertained, and the causative factors identified. A 

 knowledge of such matters as these is the foundation upon which true 

 forest research ought to rest. Unfortunately this is still a scarcely 

 touched field requiring protracted investigation, and in the meantime 

 insistent demands must be met for information which can be used at 

 once in forest management. This explains the fact that there are two 

 phases of the more pureh^ biological part of forest research now being 

 carried on at the forest experiment stations, one having to do with 

 the discovery and classification of facts and causes, the other wdth 

 results based upon somewhat empirical assumptions or only partial y 

 understood antecedents. Kesearch of the latter class is warranted, 

 first, by the great number of interacting factors and the long periods 

 of time that will be involved in their analysis; and second, by the 

 possibility of securing many reliable and directly useful results with- 

 out an appeal to underlying causes. This is particularly true of work 

 in forest measurements, planting, thinnings, and growth studies, and 

 it is much less true of studies of the succession of forests after cutting. 

 The field of forest research at the Forest Service experiment sta- 

 tions is divided into seven comprehensive groups. At the Appalachian 

 Station Avork is now being done under five of these, each represented 

 by one or more of the twelve active projects. The program ahead of 

 the station can be best described by taking up these seven groups one 

 by one. They will be given in the order Avhich nearest approaches 

 the logical sequence of the development of research, although as pre- 

 viously^ stated, other considerations make it impossible or even unde- 

 sirable to conform to this sequence in the chronological order of 

 investigations. 



Forest type studies. The unit in forest management is the stand, 

 rather than the species. The classification of forest societies (types) 

 and physical environments (sites) therefore forms the basis for all 

 other investigative work. In the Southern Appalachians the large 

 number of species and the variety of habitats make this a particularly 

 difficult matter. Very few efforts have been made to secure a logical 

 and practicable classification,* and the matter is now in the hands of 



* A comprehensive classification has been offered by W. W. Ashe. See Journal of tht 

 EUsha Mitchell Scientific Societii 37: 183198. 1922. 



