In the above I is the length of the log, a^, a. and Of the sec> 

 tional area of the log at the base, middle and top, resiectively. 



Both formulae contain an error, the extent of which is propor- 

 tionate to the amoant of difference between the diameters at the 

 top and base, respectively, of the log, that is to say, to its degree of 

 taper; and this error increases as the square of that difference. 

 Haber's formala always gives too small, and Smalian's too great, a 

 resnlt, the error of defect in the one case being one-half the error 

 of excess in the other. 



Haber's formnla has also another advantage, for which it is to 

 be preferred : the modes of measarement and calcnlati'n adopted 

 in France and Germany give, as a rnle, too high a figure for the 

 sectional area concerned in each case. This excess is partly com- 

 pensated for by the employment of Haber's formala, whereas the 

 other would only exaggerate it. 



In order still further^ diminish error, long logs should be mea- 

 sured in two or more sections, the number of the sections increas- 

 ing, i.e., their length diminishing, with the taper of each log. 

 The contents of those of regular shape and not exceeding 20 feet in 

 length may, however, be deduced from their sectional area in the 

 middle. Longer logs, even if of regular shape, should be cubed 

 in two or three sections. 



All large round logs should be measured singly. 



If the logs are stacked so that they cannot be conveniently 

 measured in the middle, the mean of the sectional areas at the base 

 and at the top must be taken. The mean sectional area should 

 never, under any circumstances, be deduced from the mean of the 

 two girths at the two extremities respectively, or an error of 

 from 10 to 15 per cent, may result. 



Poles are seldom cubed singly ; nearly always in stacks, built 

 up of poles of one and the same length, and of approximately one 

 and the same girth. Their solid contents are generally ascertained 

 by inspection from special tables. 



Straight and regular^haped branches are measured in the same 

 way as logs. 



2. Measurement of Square-cut Timber. 



Such timber must of course be cubed by the formnLi, length 

 X width X thickness. 



