6 Forestry Quarterly. 



the results. For example, in the determination of the 'Modulus 

 of Rupture' and 'Crushing Strength' of timber, values are read not 

 infrequently to units or decimals. These figures are the averages 

 of a large number of individual tests. Unfortunately, wood, like 

 most organic matter, is an extremely variable material and dif- 

 fers to a marked degree in different representatives of the 

 same species and in different portions of the same tree. Fur- 

 thermore the same piece of timber will vary greatly with changes' 

 in its environment. Thus the 'Modulus of Rupture' or 'Crush- 

 ing Strength' of a given kind of wood represents an average of 

 many widely fluctuating values. As yet no satisfactory method 

 has been discovered for determining, a priori, the variation of a 

 given piece of timber from the normal or average strength values 

 of its class or grade. Since a structure is in most cases de- 

 pendent upon the strength of its weakest member the engineer 

 and architect must allow for these fluctuations by the use of 

 the so-called 'factor of safet}',' an approximate and arbitrary fig- 

 ure. Therefore, elaborate methods of testing timbers, and values 

 which record units or decimals are refinements that are not justi- 

 fied except perhaps in theoretical researches when the investigator 

 endeavors' by means of carefully selected small specimens to ana- 

 lyze certain factors that produce variation in the strength of 

 wood. 



In a similar manner the contents of logs and trees fluctuate 

 greatly with variations in certain natural and economic factors' 

 of which the most significant are form, defect, and methods ot 

 logging, milling and utilization. Thus, log scales and volume 

 tables, compilations of averages, are inherently inaccurate except 

 when applied to more than a limited number of logs or trees. 

 Furthermore, they are untrustworthy unless the natural and 

 economic factors to which they are standardized are homologous 

 with those which prevail in regions where they are applied, or 

 unless accurate converting factors are available. The problem 

 of successfully standardizing tables to given conditions and of 

 using in each step of the process a justifiable degree of accuracy 

 is a difficult undertaking. For, even in the case of homogeneous 

 bodies of coniferous timber, volume is subject to considerable 

 variation due to differences in method of logging, milling, and 

 utilization; a point that has not always been sufficiently em- 

 phasized in the construction and use of volume tables. 



