(2560 acres) spaced at four -mile intervals; for categories 2 and 

 3 one-section blocks (640 acres) spaced at seven-mile intervals 

 for category 2 and ten-mile intervals for category 3. The sum 

 of these area blocks constituted the sample. Each block was 

 examined in the field and the area classified and mapped by com- 

 mercial character j cover type, stand size, stocking, age, and 

 site. In the office the area of the cover classes for each 

 sample was determined by dot or square counting. The data were 

 summarized by sample categories and expanded to include the 

 total area sampled. 



VOLUME 



For the area covered by complete survey, board-foot estimates 

 for saw-timber trees in saw-timber stands were derived from 

 (1) national forest cruises, (2) other public and private 

 cruises adjusted to a common standard by factors based on check 

 cruises, and (3) ocular estimates made by field men while mapping 

 cover types. Cubic-volume estimates of saw-timber trees in saw- 

 timber stands were calculated from board-foot volumes by the use 

 of ratios based on woods-scale studies. The volume of the pole- 

 tree element in saw-timber stands was derived from mean-acre 

 volumes by types, based on check-cruise tallies. The board- and 

 cubic-foot volume of nonsaw-timber stands was calculated from 

 modified normal yield tables. Yield-table modifications were 

 based on 411 l/4-acre samples measured throughout the nonsaw- 

 timber stands at 2-mile intervals along roads and trails. All 

 board-foot volumes, initially calculated in terms of Scribner 

 Rule, were converted to International l/4-inch log rule by means 

 of factors derived from mill-scale studies. 



For the sampled area, total volume was derived from mean-acre 

 volumes and area by sample class, forest type, stand-size, and 

 stocking classes. Mean volumes were based on tree tallies on 

 2,713 l/5-aere permanent sample plots — a cluster of 2 or 3 

 plots in each sample-area segment containing commercial forest 

 land. Three-plot clusters were measured in sample class 1 and 

 two-plot clusters in sample classes 2 and 3 (see area discussion). 

 Within a cluster, plots were spaced 5 chains apart, the initial 

 plot being 2 chains in a random direction from the control point 

 of that sample-area segment, Plot centers were staked and 

 referenced to two scribed witness trees to facilitate future 

 location and remeasurement , 



36- 



