^ TIMBER ESTIMATING. 77 



will fit a coat pocket but not a hip pocket. The books have stiff 

 board covers, which do away with the old cumbersome and 

 insecure tally-sheet holders. 



The trees are tallied by dots and lines, in blocks of ten, as indi- 

 cated in the following table, which shows the marks corresponding 

 to different numbers: 



' •••:::r:nnnizi 



This method is economical of space and enables the recording of 

 a large number of trees on a single sheet. 



NUMBER OF STRIP SURVEYS REQUIRED. 



Usually the sample strips should comprise from 5 to 10 per cent 

 of the total area. Sometimes it is possible to include 20 or 30 per 

 cent, but on large tracts from 5 to 10 per cent is considered suffi- 

 cient. On very large areas of 100,000 or 200,000 acres the strips 

 cover 2 to 3 per cent. In recent work on the National Forests the 

 strips have been run one-quarter or one-half of a mile apart, and 

 thus take 5 per cent or 2 J per cent of the total area. 



COMPUTATION OF RESULTS. 



After the measurements are secured, the average yield per acre 

 :||xay be computed in two ways: 



(1) By computing separately the yield of each acre, and averag- 

 ing all together; or (2) by constructing a model acre through the 

 adding together of the number of trees of each diameter which 

 occur on all the sample plots and dividing the result by the num- 

 ber of plots, which gives the average number per acre of trees of 

 each diameter. For example, to construct a model acre the aver- 

 age number of 6-inch trees on all the acres measured is calcu- 

 lated; then the average number of 7-inch trees; then of 8-inch 

 trees, of 9-inch trees, etc. The result is a model acre having the 

 average number of trees of each inch diameter. Only one com- 

 putation of yield is then required, and this will represent the 

 ^Iverage of all the sample acres. 



