enough to use, no attempt was made to collect private 
cruises. 
Private timberland owners contributed materially to the 
success of this undertaking. With very few exceptions and 
reservations they tendered the use of their cruise data, 
which in the aggregate are estimated to have cost them 
more than a million dollars. Each cooperator was assured 
that private cruises given would be kept in strict confidence 
and that cruise data from private sources would be made 
public only in such combinations as would safeguard their 
confidential character. It was emphasized also that the 
type maps would not indicate density of stand for the mature 
types. 
Field Procedure 
The intensive reconnaissance method applied on national 
forests consists in mapping areas that are uniform as to type 
conditions and estimating the average volume per acre for 
each of these type areas. Type boundaries were deter- 
mined by working along trails, roads, and ridges, by using 
high points for lookouts, and by running random strips, 
and were placed directly on base maps. For each saw- 
timber-type area, estimates of average volume per acre 
were made ocularly and were checked by use of data taken 
on a number of well-distributed sample plots. Sample 
plots were either quarter-acre circles (58.9-feet radius) or 
1-acre strips (66 by 660 feet). For all trees of saw-timber 
size on these plots species, height, and diameter were 
recorded and volume was computed. For second-growth 
areas, that is, areas occupied by stands less than about 
150 years of age, averge age of timber and average stocking 
were recorded. At frequent intervals site observations were 
taken and recorded; the age of the stand was determined 
with an increment borer, average height of the dominant 
and codominant trees was determined with an Abney 
level, and site values were read from a curve of height over 
age. 
ured where the Douglas-fir classification was used, and only 
ponderosa pines were measured where the ponderosa-pine 
classification was used. 
In the Cascade Range, the Siskiyou Mountains, and the 
Olympic Mountains, because of the distinctness of the 
topography it was possible to determine type boundaries 
largely by observation from vantage points. On the Sius- 
law National Forest, in the Coast Range, mapping was 
seriously impeded by numerous small canyons and short 
ridges with no definite topographical pattern, by luxuriant 
brush and tree cover, and by poor weather. Here aerial 
photography was used as an adjunct to ground work. 
Oblique rather than vertical pictures were taken, because 
In determining site only Douglas-fir trees were meas- 
of lower cost and greater ease of orientation, 
For areas outside national-forest boundaries the first step 
in preparing type maps was to record the collected type 
data on transparent vellum plats fitted over base maps. 
Each type mapper visited the county seat in search of addi- 
tional information and familiarized himself with the county 
in a general way by driving over the roads. Having 
147 
selected an area on which to begin work, he mapped as 
much as he could from the roads and trails. Picking points 
that would give the best view of the country and using as a 
control the roads, streams, and other features on the base 
map and the type areas already entered on the vellum 
from office records, he oriented himself with a compass and 
mapped all that could be seen. Each type area was 
viewed from several vantage points to determine its exterior 
boundaries. In this region of dense cover and irregular, 
often rugged topography, once under forest cover it is 
difficult to see out, and great care was necessary to avoid 
overlooking any small farms, pasture lands, burns, or small 
second-growth areas. On areas of mixed types it was cus- 
tomary to map the smaller type areas first, thus fixing the 
boundaries of the larger types. 
For areas not covered by existing data, which were princi- 
pally second-growth areas, land cut over prior to 1920, 
burns, woodland areas, farm woods, agricultural lands, 
grassland, brush areas, and barrens, the field examiner 
located, and sketched on the map, the boundaries of each 
type. 
age class of the timber, its species composition, and the 
degree of stocking. For areas occupied by merchantable 
timber he estimated the board-foot content of the stand 
by species. For all coniferous types except lodgepole pine, 
noncommercial rocky, and subalpine, he made site deter- 
minations at frequent intervals. 
Several large agricultural areas contain scattered forests 
and woods that are too small to be shown on the type map 
but that in the aggregate constitute a forest resource too 
large to be ignored. These agricultural areas are fairly 
well defined; an example is the Willamette Valley of 
Oregon. To get a statistical expression of the extent and 
character of their forest stands, they were covered by a 
linear survey. Type and volume data were taken on tran- 
For second-growth areas, he determined also the 
sects at intervals of 3 miles or less. 
Lands shown by county and private records to have been 
clear cut since the beginning of 1920 (type 36) were not 
classified in the field but were examined in the field to verify 
that they had been clear cut. These areas cannot satis- 
factorily be classified as to restocking, because of the 
periodicity of adequate seed crops, the practice of slash 
burning, high fire hazard, and the nature of logging practice 
in the region. However, a statistical expression was ob- 
tained of the condition of those logged prior to the period of 
general seed-crop failure that began in 1924. A linear sur- 
vey was made of the areas logged in 1920-23, transects 
being spaced 2 miles apart or at the rate of 1 mile of strip 
for every 1,280 acres. 
four 13.2-foot quadrats were examined, and each of these 
was Classified as stocked or nonstocked according to whether 
it contained one well-established seedling (the stocked- 
quadrat method). 
The site map was made on a skeleton vellum overlay map 
of the county, scale one-half inch to the mile. All the site 
determinations made were plotted on the map and by inter- 
polation site-class boundary lines were sketched in. This of 
course, gave only a generalized picture, but by referring to 
At 1-chain intervals on these transects 
