THE USE OF FOREST TYPES IN ESTIMATING 289 



factor has an influence upon the total volume of the stand, as well as 

 its average height, though both of these are influenced even more pro- 

 foundly by differences in quality of site within the same cover type. 



These differences in type msiy be caused by altitude, slope, moist- 

 ure and depth of soil. By separating the total into sub-areas, a far 

 greater uniformity of size and density of the timber in these sub- 

 divisions may be obtained, first by securing a more uniform mixture 

 of species in the per cents of the different species represented in the 

 stand; second, by reducing differences in the density of stocking per 

 acre; third, by securing more uniform sizes both in height and diameter, 

 and a smaller range. The subdivision of an area into a number of 

 smaller units is a means of avoiding the necessity for securing a weighted 

 average of these factors in order to get the average acre. Doubling the 

 number of strips would probably secure the same result, but the expense 

 of separation of the estimate into two or more types is much less than 

 this increase in field work. 



The only increased expense of separating types consists of the 

 increase in computations required by separating the areas and the 

 precaution required in changing the tally sheet on entering the type. 

 Proper coordination between the compassman who maps the area 

 and the estimator who records the timber is necessary. 



Where areas as small as 40 acres are mapped and a large per cent 

 taken, distinctions between the two types of timber are not often made 

 by old woodsmen. The total volume of each species is obtained with- 

 out separate computations of area. 



But the principle of type separations is universally applied in sepa- 

 rating areas which do not contain merchantable timber from those which 

 do. Blank areas caused by cultivation, burns, swamps, or unmerchant- 

 able reproduction must be subtracted from the total timbered area 

 under any system which permits the completion of a cover map. The 

 arbitrary inclusion of these unstocked areas makes it practically impos- 

 sible to obtain an average stand on the remainder. In theory the same 

 law of averages applies even in this case and with a sufficient number 

 of strips which cross blank areas in such a way that a per cent of the 

 blanks is taken as the merchantable stand, no error would be incurred 

 in the average. But the extreme danger of obtaining a different per 

 cent from that on the whole tract, and the comparative simplicity 

 of mapping out these blanks to obtain net timbered area, makes this 

 method universal wherever the number of strips per forty or |-mile 

 amounts to at least two, and possible even when but one strip is run. 

 This correction requires, first, the area of the type whether timbered 

 or blank, from a map; second, the area covered by the strip in esti- 

 mating. The latter expressed in acres is computed by multiplying 



