k^K Division of Ag u r a I S c UNIVERSITY OF CALIFORN ^♦h&f CaVrfornte f k¥ y rf SARIATI WOOD fetasKCi OF INCEI xr pAGRICULTUREI LIBRARY g^CT 10 1967 CEDA > * nr or caufosnu BERKasr •/ *< we; wo ■ CALIFORNIA AGRICULTURAL EXPERIMENT STATION BULLETIN 833 this bulletin reviews growth characteristics of incense cedar; discusses patterns of variation in specific gravity, heartwood percentage, ring width, summerwood percentage, and shrinkage; and analyzes the correlation of these wood quality characteristics with each other as well as with position within trees, tree classes, tree height and diameter, soil composition, and weather conditions. Fig. 1. Sampled locations in Blodgett Forest, indicated by circled numbers. SEPTEMBER, 1967 THE AUTHORS: Helmuth Resch is Assistant Professor of Forestry and Assistant Wood Tech- nologist; Shui-Mu Huang is Junior Specialist; both at the Forest Products Lab- oratory, Richmond Field Station. [2] VARIATION IN WOOD QUALITY OF INCENSE CEDAR TREES 1 Incense cedar (Libocedrus decurrens Torr.) grows in an area ranging from the Sierra de San Pedro Martir in southern California to the slopes of Mount Hood in Oregon, and from the Coast Ranges of California in the west over the Cascades, Siskiyou Mountains and the Sierra Nevada to Lake Tahoe in Nevada in the east (30, 41). The net volume of live incense cedar sawtimber stocking on commercial forest land in this area is estimated at 13.3 bil- lion board feet, 73 per cent of it in Cali- fornia (49). Because the trees are native in a climate of dry summers with little or no percipita- tion and with temperatures above 100°F, and of winters with extremes below 10°F, their growth often appears to be limited by the moisture available. The lowest an- nual precipitation in the area is about 15 inches (46). On north-south mountain ranges, incense cedar is more common on western than on eastern slopes. Depending on the available moisture the trees occur between 2,000 and 8,000 feet elevation (21,30,41). The growth of incense cedar is sup- ported by a variety of soils developed from a large number of rock substrata, i.e., diorite, andesite, rhyolite, granite and its metamorphic equivalents, sandstone, shale, basalt, and serpentine (31). These soils range in texture from coarse to very fine and in acidity from nearly neutral to strongly acid. Optimum growth, however, is found on deep, well drained, sandy loam soils originated from granite and sand- stone; on deep clay loams developed on basalt and rhyolite; and on deep, coarse- textured, well drained soils developed on pumice. Incense cedar is usually a minor com- ponent of mixed forest stands; it may, however, under favorable conditions form up to 50 per cent of some stands. On rare occasions it occurs pure over relatively small areas (14, 47). It is found most fre- quently in the mixed conifer, i.e., the pon- derosa pine-sugar pine-fir type (47). The trees are intermediate in shade tolerance between the less tolerant pines and Doug- las fir and the more shade-enduring true firs (6); they require more sunlight to de- velop fully after reaching the sapling stage than in the seedling stage (21). In- cense cedar will always remain a secondary species, regardless of its good response to release, because its height growth is slower than that of associated species (12). The fastest growth in height was re- ported on trees between 50 and 75 years of age and on trees up to 150 years if stocked on good sites (21). Tree heights of 150 feet, with diameters at breast height of about 7 feet, are frequent on the best stands in the central Sierra-Nevada in Cali- fornia on soils of the Holland series origi- nated on granite and on those of the Olympic series developed from basalt and andesite (21, 41, 47, 48). In Southern Cali- fornia and in the Coast Range the trees rarely exceed 80 feet in height and 4 feet in diameter. Growth in basal area also is usually slower than that of associated spe- cies, except on some poorer sites (12). Data on seasonal growth in height and diameter are limited to measurements made in the Stanislaus Experimental For- est (13). According to these observations, the height growth started during the last third of May lasting for an average of 91 days, and was preceded by the diameter growth which began in mid-April and con- tinued for 136 days. Incense cedar possesses relatively few natural enemies and therefore reaches a high age of up to 540 years (21). A small number of insects cause serious damage (17). Although incense cedar is noted for the durability of its heartwood in timber products, standing overmature trees are attacked by the pocket rot fungus (Poly- 1 Received for publication, February 10,1967. [3] porus amarus Hedgcock). Trees under 150 years of age are relatively free of decay (53, 7). The fungus attacks the heartwood only, which it frequently reaches through open fire wounds and large open knots, causing the formation of numerous cavities from 14 inch to 1 inch in diameter and from 1 inch to several feet in length. Aver- age cull, based on gross volume, was 4 to 6 per cent for immature dominant trees, 20 per cent for mature dominants, and 68 to 77 per cent for overmature dominants (7, 53, 18). The resistance against insects and fungus attack is due to the presence of extraneous material in various amounts in the bark and in the wood. The composi- tion of these substances has been studied in great detail (1, 2, 9, 36, 50, 57, 58). The sapwood of incense cedar is white to cream colored; the heartwood is brown, often with a tinge of red or purple, and possesses a markedly aromatic odor. The border between heartwood and sapwood is well denned and the growth rings are dis- tinct. The transition from early- to late- wood is gradual. The physical and me- chanical properties of the wood have been determined only on specimens obtained from four known and some unknown num- ber of other trees (19). According to these observations, the wood is low in specific gravity, moderately weak and soft, low in shock resistance, and lacks stiffness. Its shrinkage values are relatively low. The initial moisture content of the sapwood sometimes may exceed 250 per cent, that of the heartwood is generally slightly above the fiber saturation point. The wood can be dried fairly easily, except the darker, more extractive rich heartwood which dries very slowly and tends to col- lapse at elevated temperatures (34, 35, 27, 29). The wood splits readily and evenly, and is easy to work with tools. It takes a good polish and is classed with the woods that hold paint longest and suffer least when protection against weathering becomes in- adequate. Incense cedar heartwood is highly durable, ranking with cypress, red- wood, black locust, and other cedars of high durability. In 1956, the production of incense cedar in California was 104 million board feet — about 2 per cent of the total production of California softwood species (20). The trees are mainly converted to lumber and fence posts. Nearly all of the high-grade lumber is remanufactured into pencil slats which account for almost 98 per cent of the total amount of wood used for pencil making in the United States (23). The wood has also been used for rough con- struction work, siding, general carpentry, interior finish of dwellings, furniture, and shingles. Research relating tree growth and wood properties, as reported on many other tree species, has not been conducted for in- cense cedar. Variations in the rate of di- ameter growth of trees, specific gravity, per- centage of summerwood, and amounts of sapwood and heartwood, are factors which should be considered in the utilization of this species. Wood density, for instance, is a useful index for evaluating mechanical and physical properties of wood as well as yields of pulp produced from it. Heart- wood percentage in trees and the width of their growth rings affect the manufacture of pencil slats and other products. OBJECTIVES The main purpose of this study was to de- termine patterns of variation of certain wood quality characteristics. Specifically it was desired to: • Find the range of specific gravity of wood within trees and how it may relate to the specific gravity as determined from increment cores taken at breast height. • Determine the radial and tangential shrinkage of sapwood and heartwood and whether it can be estimated from specific gravity values. • Measure the relative amounts of sap- wood and heartwood and their change with diameter and height of the tree cross section. • Determine whether variations of sea- sonal distribution, amount of rainfall, and of temperature correlate with variations in annual growth ring width. [4] • Find the percentage of summerwood within growth rings. • Determine which of these wood quality characteristics are correlated with each other and how they vary with posi- tion within the trees, tree class, height, diameter, and the particular soils on which the trees grow. MATERIAL AND PROCEDURE Twelve sample trees were cut in 1965, on the University of California's Blodgett Forest. This forest is located on the west- ern slope of the Sierra-Nevada northeast of Placerville in El Dorado County. Most of the land is between 4,100 and 4,600 feet elevation. From 1900 to about 1913 various parts of the forest were extensively logged, and today the bulk of the timber is second growth that became established after logging, with some old trees inter- spersed. Blodgett Forest is situated in an area of high productivity for the Sierra mixed conifer forest type. Any species may be prevalent over extensive areas within the forest. However, because incense cedar and white fir were not harvested during the early 1900's, these two species abound, followed by Douglas fir, ponderosa pine, and sugar pine (32). The climate of Blodgett Forest can probably best be described by interpo- lating long term average precipitation and temperature data from nearby weather sta- tions, because the climatological station in the forest was not established until 1961. The outstanding features of the general area from which the sample trees were taken are the dry, warm summer with an average maximum temperature of about 80 °F, and the mild winters with an aver- age minimum temperature of about 30°F (42). Four locations were selected as sample plots, each representing a different soil composition; namely, the soil series Windy (on andesite parent material), Holland, Chawanakee and Musick (all on granodio- rite parent material) (48). The location of these sample plots are marked in figure 1 (inside cover) and their description given in table 1. Three trees, each representing a dif- ferent tree class, were selected from each of the four locations. A series of cross sec- Table 1. DESCRIPTION OF SAMPLED LOCATIONS Item Location 1 2 3 4 Soil type Windy Holland Chawanakee Musick Site quality class 6 6-5 5 6-5.5 6-5.5 Soil moisture moderate good moderate good-very good Topography level undulating undulating undulating Elevation (ft) 4,320 4,250 4,200 4,200 Associated* species present Y,W,I,B Y,W,I,S,B,D,T W,I,S,Y,B Y,W,I,S,B,D,T Stand composition Location 1, 2, and 4: Immature and mature (20 to 50 per cent) trees covering 50 to 80 per cent of the ground surface. Location 3: Less than 20 per cent mature trees covering over 80 per cent of the ground surface. *Y = Ponderosa pine W = White fir I = Incense cedar S = Sugar pine B = Black oak D = Douglas fir T = Tanoak [5] tional discs was cut from each sample tree at regular height intervals of 20 feet, the lowest being 2 feet above the ground. A standard increment core borer was used for taking increment cores from each tree at the height of 4.5 feet above the ground. If the tree had a diameter greater than 24 inches, the increment core was taken to the full length of the borer, i.e., 12 inches. In the laboratory, discs of i/2-inch thick- ness were cut from those originally col- lected in the forest. After sanding these discs, the ring width and the summerwood percentage was determined along a radial line at magnifications of 45 times for nar- row rings and of 19.5 times for wide rings. On the same discs the widths of sapwood were measured visually with a scale and the heartwood percentage was calculated from these measurements. The samples were cut from the discs for the determina- tion of specific gravity. Their size was \/ 2 inch along the grain, 2 inches in width (tangentially) and 20 annual rings in a radial direction. The specific gravity of both increment cores and test specimens was determined by using the maximum moisture content method (33). The specific gravity of sam- ples taken from the tree sections was also determined by the water immersion method, and the values obtained by both methods were compared to each other. The specific gravity of individual trees was calculated according to a method sug- gested by the U. S. Forest Products Labora- tory (54) from the volumes of individual bolts among the sample discs on which the specific gravity had been determined. Sixty-eight shrinkage specimens were cut from sapwood and 61 from heartwood. Their dimension was 1x1x4 inches whereby the measured length of 4 inches extended on 41 specimens in the radial and on 88 specimens in the tangential di- rection. The radial tangential shrinkage from the green to the oven-dry dimension was determined according to the ASTM standard (5). Because rainfall and temperature rec- ords have been kept only for a few years at Blodgett Forest, monthly averages of pre- cipitation obtained at the U. S. Weather Bureau gauging station at Georgetown and of temperatures from Grass Valley (52) were tested for their correlation with growth rings widths of individual trees at 2 feet height above the ground. The growth rates for the life of trees or the last 64 years, whatever was longer, were studied. Because Georgetown, the closest gauging station, is located 13 miles west and about 1,500 feet lower in elevation, and Grass Valley, the closest station with temperature records, is situated 32 miles northwest and also 1,500 feet lower in elevation than the sample location, a cer- tain error was inherent in this comparison. However, rainfall and temperature in these locations were proportional (42). All statistical calculations were carried out on an IBM 7094 computer. Duncan's multiple range test (11, 10) was employed to test differences among mean values. A multiple regression and correlation pro- gram (3) was used for determining the ex- istence of relationships among the vari- ables studied. RESULTS AND DISCUSSION The trees selected from the four sites ex- tended over a wide range of diameter, height, and age (table 2). Nevertheless, they represented only a small sample of the total population of incense cedar. The detail with which their wood-quality char- acteristics was studied, however, should make the findings valuable for recognizing general trends and, if necessary, for design- ing a broader, less detailed investigation. Specific Gravity A statistical comparison between specific gravity values obtained by the maximum moisture content and the water immersion methods from 343 specimens showed no statistical difference at the 95 per cent con- fidence level. For this reason, and to save time, the maximum moisture content method was adopted for all further mea- surements of specific gravity. [6] T3 O O 5 E E as c_> csicdcm r~- r-~ — i o oi ^ — ,_ CM— 'CM CMCM0O CM — O a> Q. - CO CM O CL a coco ^j- co ^j- oo <* -a- r-^oo^r cd o in ^r co in r-- -^ co co co o o TO a> c O in — i CM O CO ^3" O ID N in OO CO 09 CO >^ '> CO oo o O Q. GO CD o E CD <_> o- oo co ai id in r-- oo r-. o cm ^ COO ooo ooo ooo ten a> r^o> in in n co — i co r--. — > cm o o ooo ooo ooo CD CD £ TO i- OOltO ^-«S-0 OOTLf) OO Ps (O co to n m cmocm i — tn io r-*^j-^o OJ^H— <— • i— i — c CM CM CM >1 a) "tr £ co XJ go m "O -o — (/> -o ._ Coir ^j UO LO cd o to OO CM I — r--. CO CD CO CM O m ■*— cmi — m co c^ en — « t ^r cr> co *3- * 3: CD Q •S in m O OOO OOO OOO "S — i CT> *3- »J- M CO — ■ c CO — — . — i ^ CM U->—< CM ~ ~ 1 — 3 -h CM CO >el- m CO tv OO OI O — 'CM a. >> o 00 a> CO >n = g o T3 CO S ' — J 's - 1 S i o S I 2 J CN C CD ^_ ^ CD of CD c CO E o cd -js -^ a- a> o §■<= a in CO *- II 3.5? .E 5 3 II - E 42*-- co ^ e '^ _oj -S5 . o oE« <_, O CD E «2 CO II CD CD ._ qo> o-o CO *3- a. °'e un lo r^ oo *i !t jp o o o o Z3 E S3£ 03 CO 5 CL CD # d * * CO co ch r^ * n GO 3 CO CM T CO TO in oi r-. n O •— • "1 do CO 3 U " 5"^ > TO OO E o c O O E r< O) OO IS T3 o O cu CD a. Q CO CD "o a> o cu o CJ ob 1o CD CO E — 1 CM CO ^3" E 3 CD o £= cu -a 2 c >•- >N to «3" CD O O >- > CD OO LT> CO -£= cu o o o o oo o o o o CD ■= CO c to £ TO jfT 1 CO CD TO Jr £= -° TO ^ ?.i '> TO CM CO CO CO CD f> .£; co CD CD ob ^ O O CO CD CM OO CO ci "-. > bfl ^^ OO TO ob cu ~- o CD CO O O CO CO CO O o £ CD a. CO CD CD 'o £= .„ o a. cu TO J* E cu CO OO — i OO CO co cm co r-. ^H _ O O a c <= c OO OO Op o CO 00 CO z * * * ui whether the tree specific gravity might be influenced by tree class, i.e., the position a tree held in its association with others, or by soil composition (table 1). In our sample, no effect of tree class or soil com- position was found by an analysis of vari- ance. If tree age, a variable confounded in the present analysis, could be factored out, a different result might be obtained. Such an attempt is difficult to undertake because no pure, even-aged stands of incense cedar exist. The four sampled locations had been classified as having approximately the same site quality index, although their soil com- position was different. The influence of different site qualities and other soil com- positions remains to be observed in a more extensive investigation. The variation of specific gravity within the sampled trees, as well as tree classes, was not consistently correlated with a number of other tree characteristics (ap- pendix A). An increase in the percentage of summerwood was in most cases, and es- pecially in dominant trees, associated with an increase in specific gravity. The width of growth rings, however, was almost never correlated with density. Both of these re- sults are in agreement with observations on a number of other coniferous species (25, 26, 55). Specific gravity in respect to position within trees was much more contradictory. In a number of stems the specific gravity decreased with height above the ground, but was not significantly correlated with vertical position in others. As a group, only the dominant trees exhibited this de- creasing trend. Specific gravity varied with the radial position (measured in number of growth rings from the bark) — it in- creased in some trees and decreased in others. A decrease in density toward the bark was noticeable in intermediate and suppressed trees taken as groups. When the dominant or all 12 trees were pooled, no significant correlation was found in this respect. This finding is interesting because wood density tends to increase toward the bark in a number of tree species (25, 24, 16, 15). This seemingly contradictory trend in incense cedar may be explained by the difference in the specific gravity between heartwood and sap wood (table 3) where the higher values for heartwood could com- [9] pensate for any increase from pith to bark that might otherwise exist. Clear-cut trends could not be detected for the correlation of density to tree height, tree age, and diameter at breast height. In dominant trees, the greater the height and the age, and the lower the d. b. h., the denser was the wood. This trend could not be recognized or was the opposite for trees having had a different position in the stand. After recognizing the diversity of asso- ciations with wood density, it was no sur- prise to obtain a wide distribution of co- efficients of multiple determination which for some individual trees were much higher than for tree groups. This coefficient was very low for all trees pooled together. A knowledge of the volume of trees and their contents of fiber mass is often desired when planning their utilization. For this reason the volume and the amount of oven-dry wood substance contained in each bolt and in each tree were calculated (table 5). These data combined with the values of d. b. h. and tree height reflect greatly the taper observed in the sampled trees. Radial and Tangential Shrinkage The shrinkage from the water-saturated to the oven-dry condition of 129 samples taken from all trees, except for tree num- ber 9, was determined (table 6) and com- pared to average values listed in the Wood Handbook (45). The observed values, aver- ages of sapwood and heartwood, were slightly lower than those listed in the liter- ature, i.e., 3.15 per cent vs. 3.3 per cent for radial and 5.00 per cent vs. 5.2 per cent for tangential shrinkage. Heartwood samples always shrank less than sapwood specimens; this difference was statistically significant for tangential but not for radial shrinkage. (The sample size may have been too small for statistical proof in the latter case.) This difference between the shrinkage of sapwood and heartwood may be best explained by the greater amounts of extraneous materials in the heartwood and bulking effect hind- ering the shrinkage of the cell walls (40). This explanation is further supported by the good correlation of radial and tan- gential shrinkage of sapwood samples to their specific gravity that could not be ob- Table 5. VOLUME AND CONTENTS OF WOOD SUBSTANCE OF INDIVIDUAL TREES* Tree no. Bolt height above ground feet Bolt volume cubic feet Oven-dry wood substance pounds 1 2 - 22 22 - 42 42 - 62 62 - 82 82 - 102 71.984 46 488 34.420 21.144 8.840 4490.554 2900.851 2147.808 1319.386 551.616 Total bole 182 856 11410.214 2 2 - 22 22 - 42 42 - 82 51.054 13.012 5.846 3185.770 811.949 384.970 Total bole 69.912 4362.509 3 2 - 22 22 - 42 11.436 4.840 716.726 302.016 Total bole 16.326 1018.742 4 2 - 22 22 - 42 12.448 5.016 776.755 312.998 Total bole 17.664 1089.753 5 2 - 23 23 - 42 42 - 62 11.622 5.236 3.042 725 213 389.126 189.821 Total bole 20.900 1304.160 6 2 - 22 22 - 42 42 - 61 61 - 82 50.914 33.038 19.362 7.054 2177.034 2061.571 1208.189 440.169 Total bole 110.368 6886.963 7 7 - 22 22 - 42 42 - 62 62 - 82 82 - 102 102 - 122 105.900 102.976 84.562 62.848 35.824 13.290 6608.160 6425.702 5276.669 3921.715 2235.418 829.296 Total bole 405.400 25296.960 8 2 - 24 14.216 887.078 9 2 - 22 4.588 286.291 10 2 - 22 22 - 42 42 - 62 57.668 33.040 18.000 3598.483 2061.696 1123.200 Total bole 116.416 7264.358 11 2 - 22 22 - 42 14.472 5.846 903.053 364.790 Total bole 20.318 1267.843 12 2 - 22 8.376 522.662 * Bolt v olume inside bark , in green cond tion [10] -x c c: CD DO H > cd -o -D c to oo r-» o 1 CO r-» . O O | to CD oo I < per cent 5.46 4.54 5.00 CSI LO No. of specimens ■<*■ *3- 1 OO lO CO | OO 10 to 1 To a: 09 T3 -6 to 00 O OO 1 CO TJ" O O | =3 > CD OO 1 , per cent 3.25 3.04 3.15 CO CO Q. H(M | ^ CD O. -o o o Sapwood Heartwood Average o o -O -o c to "O O O DO > < OT3 5 ° %9 served for heartwood pieces (table 7). It has been substantiated that heavy woods shrink more than light woods over the same range of moisture content and the volumetric shrinkage of most softwoods and many hardwoods is approximately proportional to the change in moisture content below the fiber saturation point (8). The notable exceptions are woods in which either the cell lumen changes ap- preciably in size during shrinkage or, as most probably in incense cedar heartwood, the shrinkage of the cell wall is decreased because of the bulking effect that extrane- ous substances exhibit. Percentage of Heartwood The amount of heartwood contained in incense cedar trees has significance for the [ii ^ * o to co +-. <_- °j 5 <->° Lg; 15; C f= = C c_> CJ *■ DO CO CO E E '= iu DO k. 00 high high non< non; t= -^ o c ^.S O "* CD *3- co to m o) CD CO CD CO I s ! OO "tf- ^ ^ o o O O O O o « > CD CO CO OO CO •ci O O O O ^ TO O O O O ■> 00 >- to 1— DO > £ < CD O O O O CL o LU Q_ OO Q z. CD -o o r- oo o UJ co to t rs ■D TO o o o o !£ 03 oo Z to OC n: CO 00 ■z. CD LU OO 0j to CD it q- LU o m f o m 5= CD |j co to co «a- h- < Q, LU OO z O h- c DO = wo to to C3 I— ^£ — C C .ES ■^ CD "^ CD 1- ^ C '"O C to to to ro ce H- q: j_ CD Q. >^ -o O o 5 ■c ■G Til? o o 5 5 s 5 -r r D. Q. CO TO TO TO CD CD 00 O" I I producer of lumber products as well as for the manufacturer of pencil slats. In the first case, it is sometimes important to know how much of the product can be sold as decay-resistant material; in the sec- ond case, drying and treating times can be greatly affected by the heartwood-sap- wood mix. The amount of heartwood contained in trees generally varies from species to spe- cies and with a number of interrelated fac- tors such as age of the tree, its diameter, crown class, or crown width (15, 37, 44). In an attempt to find an obvious indicator for a simple estimation of the heartwood per- centage in incense cedar trees, the height above the ground and the diameter of the sample discs were arbitrarily chosen (table 8). When the data of all 12 trees were pooled and examined by means of a re- gression and correlation analysis, only the diameter of the discs, on which the heart- wood percentage was measured, turned out to be a significant indicator. Percentage of heartwood decreased with disc diameter, the influence of sample height above the ground was nonsignificant. A significant effect was not found even when the sample height was converted from its absolute value (in feet) to a percentage of total tree height, and the analysis repeated. The resulting prediction equation, a straight line, least square fit of heartwood percentage versus disc diameter, reads as follows: Heartwood percentage = (2.230) (diameter, in inches) - (3.405). The coefficient of determination was cal- culated to be 0.75, the average heartwood percentage was equal to 29.12, and the average disc diameter 14.59 inches. Because of the small sample size, the above equation may not be the precise for- mulation for all incense cedar trees; how- ever, it should indicate the approximate relationship between heartwood percent- age and the diameter of tree cross sections. Table 8. HEARTWOOD PERCENTAGE AT VARIOUS HEIGHTS Tree no. Height of disc Diameter of disc Heartwood Tree no. Height of disc Diameter of disc Heartwood feet 2 22 42 82 82 102 2 22 42 62 2 22 42 2 22 42 2 22 42 2 22 42 61 82 inches 29.64 21.72 19.56 15.84 11.88 6.12 17.76 12.00 9.72 4 80 12.12 8.14 4.92 12.95 8.40 5 16 12.60 8.04 6 96 24.36 18.84 15.96 10 80 5 24 per cent 53.11 43.18 30.48 25.76 12 50 0.61 45 65 33.64 12.72 0.56 55.14 32.66 23.80 36.22 7.37 00 23.59 8 04 0.00 49.62 38.96 30.13 11 86 00 7 feet 2 10 22 42 62 82 102 122 2 24 2 22 2 10 22 42 62 77 84 2 22 42 2 22 inches 40.56 35.40 31.92 29 52 26 28 21.72 14.52 7 68 14 68 7.32 8 56 4.32 29 04 22.32 19.68 15.24 10 44 6 96 4.80 13.68 9.48 5.16 11.40 6 00 per cent 67 65 7... 72.39 7 7 7. 65 94 55 95 50.74 7. 53.18 7 49.35 2 7 29 90 2 8 2 22.60 2 8 3 25 2 9 9 17.85 3 00 3 10 83.40 4 10 65.04 4 10. 45.81 4 10 10 42.71 45.99 5 10. 4.27 5 10. 6.25 5 11 12 93 6 11. . 4.63 6. . 11 12 0.00 6 6 21.45 6. 12 0.00 [12] Growth Ring Width that in the early years tree growth was . strongly influenced by harvesting opera- Special attention was given to the widths tiom which ended amund m ^ and of growth rings because their variation not forest fires of ^.^ ^ lagt seve de only reflects changes in the growth condi- stIuctive one occurred in 1919. In general, tions during which the wood cells were ,* ■,. . .-, i . ^ 6 , , . the diameter growth appeared to be more formed but also in some cases, because it influenced b variations in temp erature has a substantial bearing on the utilization ^ , d itation . of the tree. (Ring width is one of the grad- XT . / co » , ar . . f .* v . & , . • , , , Ninety, 68, 78, and 69 per cent of the ing and sorting characteristics developed . . . . '. ,, . 4 Q . ^ .,,<*,! , ,1 variation in growth ring width in trees 3, tor pencil slats.) Although an appreciable , , n .. , ° i • i 1 / . . . y . & . _ , _ rr 4, 8, and 9, respectively, was explained by variation in the ring width of trees was , « , \ ~, u . ./ u . ^ . & _ . the weather data. The ring width in trees found among tree classes and sou composi- , , , ^ , °, . , . . . °. . . „ * . 3, 4, and 8 seemed to have been increased tion {table 2), a two-way analysis of van- . , , , • 1 • a 1 \. . 7 , ; . . , ; . mainly by higher temperatures in April ance did not show a statistical significance j # ■»* j ■ * q /i ^ n u r . ..„ & and /or May, and in trees 3, 4, and 9 by ot these differences. , . / . . . . ^., higher spring precipitation. The pattern Within individual trees and within tree of weather influe nce in the fall and winter classes, the greatest association existed be- months preceding the growth season was tween ring width and radial distance of erradc wkh some o£ ^ trees ^ posi . these rings from the bark (appendix B). 2 tiydy influenced by higher temperatures Growth rings were wider in the vicinity of and preci p itationj others negatively or not the pith and gradually decreased in width ^ aU These resuks are in faif agreement toward the periphery as is usual in nor- wkh those obtained in a study on the in . mally grown conifers (39, 44, 59). Although fluence of precipitation on the diame ter changes in ring width with tree height had h of ponder0 sa and sugar pines (32). also been reported in a number of species (44, 15, 16, 26) no such definite variation Percentage of Summerwood pattern could be detected by the linear A two-way analysis of variance showed that regression technique used. This does not the differences in the amount of summer- imply that such a pattern may not exist wood tna t had been developed in the ob- but rather that it did not follow a linear serv ed trees existed among tree classes as pattern. Other researchers found that two we H as S oil compositions (appendix C). maxima of ring width exist in a number of Dominant trees contained the highest, in- trees, one close to the base and one toward termediate trees the lowest percentage of the crown, and that these maxima gener- summerwood. ally change with age (4, 15). Such a trend Summerwood was not immediately obvious in the sam- Tree class: percentage pled trees. Intermediate 15 The dominant and especially the inter- Suppressed 19 mediate trees as groups exhibited growth Dominant 23 rings of decreasing width with increasing Soil composition: age. Musick 13 The combined influence of temperature Chawanakee 14 and precipitation on variations in growth Windy 21 ring width was statistically significant only Holland 28 for three suppressed and one intermediate A multiple range test showed that all tree. Larger multiple correlation coeffi- values w ithi n both tree class and soil com- cients were obtained in the statistical anal- ition were statistically different from yses of weather data and diameter growth „. other when the time period from 1920 to 1964 As has been inted om a good cor . rather than from 1901 to 1964 was con- relation existed in most trees betw een the sidered. Thi s may be explained by the fact percentage of summer wood and specific 2 The correlation analysis with the weather gravity, where an increase in one was as- information required the annual-ring data to sociated with an increase in the other (ap- be entered on a bark-to-pith basis. pendix A). The fact that specific gravity [13] did not vary consistently with summer- wood percentage had been reported pre- viously (44). Only trees 5 and 6, growing on Holland soil, did not show that trend. The summerwood percentage decreased with height in the tree and from bark to pith only in a number of trees — trends that have been reported for other species (15, 28, 56). A correlation to ring width did not exist for trees 1 and 6. This is also in general agreement with other findings (25, 55). When the data of all trees were pooled, slight trends of increasing amounts of summerwood with decreasing diameters at breast height and increasing tree heights and tree age were recorded. Summerwood percentage for all sampled trees could be estimated only with a multiple coefficient of determination of 0.30. However, this coefficient was much higher for the groups of intermediate and suppressed trees alone, i.e., 0.73 and 0.81 respectively. CONCLUSIONS Relationships between the growth condi- tions of trees and the anatomy as well as the physical and mechanical properties of their wood, have made it worthwhile in the past, to search for certain characteristics which may be used as indicators for the evaluation of wood and the application of silvicultural treatments. The reported study does this on a limited scale for in- cense cedar. Because the results were ob- tained from a very small sample of trees grown on four soils of different composi- tion — all within a small area — they only provide an indication of trends that may exist in terms of interrelationships be- tween examined growth and wood charac- teristics. They further should be helpful in designing a study that would include a number of sample plots distributed over the entire geographical range of incense cedar. The specific gravity of the wood of this species, one of the most important physi- cal properties, is slightly higher in heart- wood than in sapwood, the contents of extraneous substances being the likely cause for the difference. The method of estimating the average specific gravity of trees from single incre- ment cores by a simple regression equation can be improved by including the easily obtainable measurements of tree height and diameter at breast height in the anal- ysis. Within dominant and codominant trees the specific gravity decreases from stump to crown while there is no or very little increase from pith to bark. Among trees it decreases with increasing d. b. h., and in- creases with age and tree height. In inter- mediate and suppressed trees, tree age and height do not affect the density, which gen- erally decreased from pith to bark. In al- most all trees a close relationship exists between specific gravity and percentage of summerwood; however, an effect of growth rate on density is rarely found. There is no significant difference in den- sity between tree classes, nor does differ- ence in soil composition affect the tree density. The shrinkage from the water-saturated to the oven-dry condition tends to be lower for the heartwood than for the sapwood in both transverse directions. The interde- pendency between specific gravity and transverse shrinkage, found in a great number of woods, exists for incense cedar sapwood only, but not for the heartwood. Ring width, as a consequence of growth, depends on the productive capacity of the tree, which is influenced by a number of factors, especially site quality and stand density. The inability to correlate growth rate with tree class, in this study, stems from the unequal age of the sampled trees, and because ring width was found to de- crease with age as well as distance from pith to bark. Differences in the composi- tion of the four soils do not account for variation in the growth rate. Strong evi- dence exists that some of the variation in annual diameter growth of some sup- pressed and intermediate trees is caused by fluctuations in temperature and precipita- tion while others with perhaps less compe- tition in the stand are insignificantly af- fected by climatic differences from year to [ 14 year. Affected trees respond to higher April increase in section diameter. Therefore, and /or May temperatures and greater stronger — that is, generally older trees — ■ spring rain fall with wider growth rings. contain a greater percentage of heartwood, Tree height d. b. h. have little or no rela- but the influence of taper must not be tion to ring width. overlooked. Differences among tree classes and soils In a more extensive study of relation- are reflected in the amount of summer- ships between growth conditions of trees wood produced in individual trees. The and the anatomy and physical properties average summerwood percentage decreased of the wood the following should be con- in the sequences: dominant, suppressed sidered: extension of sampling over the and intermediate tree classes, and Holland, geographical distribution of incense cedar Windy, Chawanakee, and Musick soils. Be- and the most important soils; inclusion of cause of these influences, relationships of tree characteristics such as stem form summerwood percentage to height, age, (taper) and crown class (mass of foliage) and d. b. h. of trees, as well as with posi- and wood characteristics such as extractive tion within these trees, are not clear. and moisture content, and the occurrence The heartwood percentage within a of pocket rot. horizontal tree section increases with an LITERATURE CITED 1. Anderson, A. B. and T. C. Scheffer. 1958. Chemistry of decay of heartwood on ageing in incense cedar (Librocedrus decurrens, Torr.). Nature 194, 4828: 410. 2. Anderson, A. B., T. C. Scheffer and C. C. Duncan, 1962. On the chemistry of heart- wood decay on ageing in incense cedar (Libocedrus decurrens, Torrey). Chem. & Ind. (28): 1289-1290. 3. Anon. 1961. BMD series of computer programs. Div. of Biostatistics, School of Medi- cine, Univ. of California, Los Angeles. 4. Assman, E. 1961. Waldertragskunde. BLV Verlagsges. Munchen, Bonn, Wien. 5. ASTM Standards. 1966. American Soc. for Testing Materials. Philadelphia, Pa. 6. Baker, F. S. 1949. A revised tolerance table. Jour. For. 47:179. 7. Boyce, J. S. 1920. The dry rot of incense cedar. USDA Bull. 871. 8. Brown, H. P., A. V. Panshin, and C. C. Forsaith. 1952. Textbook of wood technology, Vol. II. New York: McGraw-Hill. 9. Bynum, H. H. 1965. Effect of incense cedar heartwood extract on growth of Polype rus amarus. Mycologia 57 (4): 642. 10. Dixon, W. J. (editor). 1965. BMD biomedical computer programs. Health Science Computing Facility, Univ. of California, Los Angeles. 11. Duncan, D. B. 1955. Multiple range and multiple F tests. Biometrics 2 (1):42. 12. Dunning, D. 1923. Some results of cutting in the Sierra forests of California. USDA Bull. 1176. 13. Fowells, H. A. 1941. The period of seasonal growth of ponderosa pine and associated species. Journ. For. 39:601. 14. Harlow, W. M. and E. S. Harrar. 1950. Textbook of dendrology, 3rd edition. New York, Toronto, London: McGraw-Hill. 15. Knigge, W. and H. Schulz. 1966. Grundrisse der Forstbenutzung. Hamburg, Berlin: Parey Verlag. 16. Kollmann, F. 1951. Technologie des Holzes und der Holzwerkstoffe, Bd. I. Berlin: Springer Verlag. 17. Keen, F. P. 1952. Insect enemies of western forests. USDA Agr. Misc. Publ. 273. 18. Kimmey, J. W. 1954. Cull and breakage factors for pines and incense cedar in the Sierra Nevada. California For. and Range Exp. Sta. For. Res. Note 90. 19. Markwardt, L. J. and T. R. C. Wilson. 1935. Strength and related properties of woods grown in the United States. USDA Tech. Bull. 479. [15] 20. May, R. H. a. H. L. Baker. 1957. Lumber production in California, 1956. California For. and Range Exp. Sta. For. Survey Release 30. 21. Mitchell, A. J. 1918. Incense cedar. USDA Bull. 604. 22. Mitchell, H. L. 1964. Patterns of variation in specific gravity of southern pines and other conifer species. Tappi 47(5): 276. 23. Panshin, A. J., E. S. Harrar, J. S. Bethel and W. J. Baker. 1962. Forest products, 2nd edition. New York: McGraw-Hill. 24. Paul, B. H. 1957. Juvenile wood in conifers. U. S. For. Prod. Lab. Rep. 2094. 25. 1963. The application of silviculture in controlling the specific gravity of wood. USDA Tech. Bull. 1288. 26. Ralston, R. A. and E. A. McGinnes, Jr. 1964. Short leaf pine wood density unaffected by ring growth. Lumberman 208(2592): 17-19. 27. Rasmussen, E. F. 1961. Dry-kiln operator's manual. USDA Agric. Handbook 188. 28. Rendle, B. J. 1958. Note in scientific reviews. News Bull. Intern. Assoc, of Wood Anatomists. No. 2. 29. Resch, H., E. T. Choong and H. H. Smith. 1967. Segregating incense cedar lumber for drying. For. Prod. Jour., in press. 30. Sargent, C. S. 1922. Manual of the trees of North America, 2nd ed. Boston: Houghton Mifflin. 31. Schubert, G. H. 1957. Silvicultural characteristics of incense cedar. California Forest and Range Exp. Sta. Techn. Paper 18. 32. Skolmen, R. G. 1958. Growth fluctuations of ponderosa and sugar pine as related to the amount and seasonal distribution of rainfall. MS thesis in forestry Univ. of California, Berkeley. 33. Smith, D. M. 1954. Maximum moisture content method for determining specific gravity of small wood samples. U. S. For. Prod. Lab. Rep. 2014. 34. Smith, H. H. and C. P. Berolzheimer. 1957. Air drying of incense cedar: tests under summer conditions in California. California For. and Range Exp. Sta. For. Res. Note 123. 35. . 1959. Air drying of incense cedar: tests under winter conditions in California. Pacific Southwest For. and Range Exp. Sta. Techn. Paper 38. 36. Smith, J. E. and E. F. Kurth. 1953. The chemical nature of cedar barks. TAPPI 36:71. 37. Smith, J. H. G., J. Walters, and R. W. Wellwood. 1966. Variation in sapwood thick- ness of Douglas fir in relation to tree and section characteristics. For. Science 10(1):97. 38. Society of American Foresters, Committee of Forest Types. 1954. Forest cover types of North America (exclusive of Mexico). Soc. Am. For., Washington, D.C. 39. Spurr, S. H. and W. Y. Hsiung. 1954. Growth rate and specific gravity in conifers. Jour, of For. 52(3): 191. 40. Stamm, A. J. 1964. Wood and cellulose science, New York: Ronald Press. 41. Sudworth, G.B. 1908. Forest trees of the Pacific slope. USDA For. Service, Washing- ton, D.C. 42. Tappeimer, J. C. 1966. Natural regeneration of Douglas fir on Blodgett Forest in the mixed conifer type in the Sierra Nevada of California. Ph.D. thesis in forestry, Univ. of California, Berkeley. 43. Taras, M. A. and H. E. Wahlgren. 1963. A comparison of increment core sampling methods for estimating tree specific gravity. USDA For. Ser. Res. Paper SE-7. 44. Trendelenburg, R. and H. Mayer-Wegelin. 1955. Das Holz als Rohstoff. Miinchen: C. Hanser Verl. 45. U. S. Department of Agriculture. 1955. Wood handbook. USDA Handbook 72. 46. 1941. Climate and man. Agricultural Yearbook 1941. 47. U. S. Forest Service. 1907. Incense cedar. USDA Silvical Leaf. 9. 48. 1954. Field manual for soil-vegetation surveys in California. California For. and Range Exp. Sta., Berkeley. [16] 49. 1955. Timber Resources for America's Future, For. Resource Rep. 14. USD A. Wash., D.C. Govt. Print. Off. 50. 1964. A clue to heart rot resistance in incense cedar? P. S. W. For. and Range Exp. Sta. Rep. 17. 51. 1965. Western wood density survey, Report I. U. S. For. Prod. Lab. FPL-27. 52. U. S. Weather Bureau. Summary of climatological data for the United States by section. Central California. U. S. Government Printing Office, Washington, D.C. 53. Wagner, W. W. and R. V. Bega. 1958. Heart rots of incense cedar, USDA. For. Pest Leaf. 30. 54. Wahlgren, H. E. and D. L. Fassnacht. 1959. Estimating tree specific gravity from a single increment core. U. S. For. Lab. Rep. 2146. 55. Wellwood, R. W. 1960. Specific gravity and tracheid length variations in second growth western hemlock. Jour, of For. 58(5): 361. 56. Young, H. E. 1952. Differential time of change from early-wood to late-wood along the bole of young loblolly pine trees. Jour, of For. 50(8):614. 57. Zavarin, Eugene. 1958. Extractive components of incense cedar heartwood (Heyderia decurreus Torr.) VII. On the occurrence of Heyderial. Jour of Org. Chem. 23(9): 1264. 58. . 1958. The extraneous substances of incense cedar. California For. and For. Prod. 5. 59. Zobel, B. J. and R. L. McElwee. 1958. Natural variation in wood specific gravity of lobolly pine and an analysis of contributing factors. TAPPI 41(4): 158. ACKNOWLEDGMENT The authors wish to extend their appreciation to the Pacific Southwest Forest and Range Experiment Station, U. S. Forest Service for sponsorship, in part, in the form of a grant-in-aid. Acknowledgment is also made to Professor H. C. Sampert and his staff for assistance in harvesting the trees. Appendix tables follow [17] > ■*= s 4 UL. TO j o m in Q. cm in *3- co *; E cor~.'»3-r-. c :cD»3-cT5 fo co co CM 3 <5 O O O O O O O O O O O E "S *o Q 00 c o CD cz o o in co O co co oo m cm r-. cm OO to oo r~- «a- o . oo co oo ro co n *— ' r — r-~ CD oo in co in co O o OO a> cooocooo ^cncooi in co oo * £ O O O O O O O CO O O o O * -c # * * * oo CD E . co • in . oo CO g oo oo co o- oo co oo o- oo o 00 00 E c c ^ = c « er d = « = ^ cz o o o o bs 1 1 ' ■a * * 1 * * * * * * * * * * * * o * * * * * * * * # * 3 cz r-. cni cm co «a- cv O OO CO CO CD LD . O) CD >J . CM CZ oo oo in CO CD E C_> co w «* co m oo cm r- co in cd CO «3- ,J cm co en ,: CO cc CM CO CO CM E a> a. O O O O O C o o o O zs 00 .2 w * * * * * .-tz'-Q * * * * * CO CO CO CO co r-~ o E Q. o . CD in CO . ■q- . m co oo cm r-- CD oo co "> v 00 CD CD £=' **>. °? ^ =' °» = c c - in co E H?* - o o o o o o T3 QO TO c ac- _cu * I i * * 11 * * * * * * * * CO o <=> cc in in o co ~i ^ .CM . IT CM o CD «3- CO OO 0- 00 CO 00 c- CM 00 00 a> > >£ " * " c d 'O c « co c - E CM o CJ -E o o o o o c o o er O0.Q 1 1 1 1 1 1 o QJ TO X to o CJ * TO 00 * * 0O rt CO JC CO CO in TO J3 r~- o oo CO a. o —. °° = ci c o o 1 o * * * TO to TO CU CO o CO 00 oo 00 E e o E * j= * * OO co r-~ co CD . O «a- -EZ CD 00 CO CD a> *~! t= "^ cu o o O 1 ■ cu : d oo ; "S "S CZ TO = ^ & " CU « 2 I s 0) a>-o cu ^TO cc c= CM IT o- — i_ Q. E a> cl o "£ o Q _ 00 < .O JD J3 O O O If 00 5 .2 cuO •^ -E. < I cz * * * * * * * o * * * * * * ■* * -*ocoor-~oooco ci rv to >»N|NinrHO(flCVJ r-» O co CD _a> ooootnO'tusmM O) IV UJ oo g ivin^-Nrvooicrv oooooooo CO 0O CO O O o CO o = E Q o co c= c_> c inoi O) (O CO £ oooooooo o o o o O * I * * * & CD .co . . in . oo CO (/> CO CO 00 CO co co MOM o c OJ <= c: °°. =' r= e= « ss •™J •j3 o o o o a> | I 1 Q. oo ■a o * * o 5 1= OO 00 co oo CM E <-> CDCNJ00COCOCOCOC CO CO CO "i "> c c t= e «=" c cz' cz cz "~^ E a. o o o 3 1 1 00 .2 w * * * * * * * * * * * * * * * * * * * * * * * * .■^-o * * * CD 0O CD * 0O CD CO CM CM oc c E Q. O LT3 .CD . CO CD CD ^ 00 CO co oo cr> co — i « « »j oo c\ o- O r-~ CD 0O d o v- 00 03 CO CO CO CD CO ~ - CO C CO — ■ CO CO CO ° »» 03 c c= cz ^ c °° d c ~! "> e £ l| O o I o o I o 1 o 03 CO o X CO CO a. * * CO J= 03 . o -Q ci cz O CZ * * 03 1 * 1 M CD CM TO TO u-> cd CD 00 CO ° o 1 a> 03 o o 1 1 J= * m -tZ CO CO CO 03