Historic, archived document Do not assume content reflects current scientific knowledge, policies, or practices. April, 1924 THE NATURAL REGENERATION OF DOUGLAS FIR IN THE PACIFIC NORTHWEST _ _ By JULIUS ¥. HOFMANN, Silviculturist, Wind River Forest Experiment Station, Forest Service Introduction Distribution of Dougias Fir Climate and Site Seed Origin of Young Growth _ Migration Character of Second Growth Forests Lhe A. Methods of Study Appendix B. Botanical Characteristics Bibliograghy WASHINGTON GOVERNMENT PRINTING OFFICE - 1924 ea UNITED STATES DEPARTMENT OF AGRICULTURE Washington, D. C. April, 1924 NATURAL REGENERATION OF DOUGLAS FIR IN THE PACIFIC NORTHWEST By Junius V. Hormann, Silviculturist, Wind River Forest:Experiment Station, Forest Service. CONTENTS. Page Page Va Ces OGLE CGT Ta ee Ta asia SA ti GTO Wa abe eA ee pd 44 ‘Distribution of Douglas fir __-__-__ 2 | Silvicultural management _________ 46 Climatte and Site 220 lees pal PaEin emi egmettumt bie Oe eh 50 (2Y 29 9 path UR i PORE AR OF cir IGINLA) 8 at Ue MON aS UT eeriy 2 etal a a 54 Origin of young growth___________ 18 | Appendix A. Methods of study_____ 57 Mima timc. en ca Seg 38 | Appendix B. Botanical characteris- Character of second-growth forests_ 40 EST CS ESCADA 2 ah 2 BS Competitions. 2. ow ase oe ee AA BOS Hap yee eles ea OE a a 60 INTRODUCTION. The Douglas-fir: forests of Oregon and Washington are among the most valuable in the United States, but because the means by which they can be regenerated have not been understood or followed, about half the 3,500,000 acres? cut over in Oregon and Washington are nonproductive. To this denuded land are being added from 80,000 to 100,000 acres each year, an amount that will tend to in- _crease as other timber r regions become exhausted and the Pacific Northwest is called on for more and more of our national lumber supply. So productive are Douglas-fir forests that well-stocked stands grow at an average rate of about 600 board feet to the acre annually up to an’ age of about 200 years. The nonproductive cut-over Douglas-fir lands of Oregon and Washington could, at this rate, seventh of the total present yearly cut of this region. To trace the laws governing the establishment and growth of Douglas-fir for- 1 Douglas fir (Pseudotsuga taxifolia) (Poir. Britt.) is known by a number of common names, such as Washington fir, red fir, yellow fir, fir, Douglas spruce, spruce, Oregon pine, red pine, Puget Sound pine, or British Columbia pine. 2 Report on. Senate Resolution 311, Timber Depletion, Lumber Prices, Lumber Hxports, and Concentration of Timber Ownership. -U. S. Department of Agriculture, Forest Service, June 1, 1920. Notrre.—Grateful acknowledgment is made of the valuable assistance of C. J. Kraebel. forest examiner, and H. V. Brown, forest ranger. 60634- —24 a 2 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. ests and the methods by which new stands may be obtained is the } . purpose of this publication. | Douglas fir is so intimately related to other species in establish- ment and growth that it is necessary in a discussion of this kind to consider its chief associates, such as western red cedar (Thuja | plicata Don.), western hemlock (Ysuga heterophylla (Raf.) Sarg.). |‘ ig. 1.—Regional distribution of Douglas fir in Oregon and Washington. | noble fir (Abées nobilis Lindl.), silver fir (Abzes amabilis Forb.), | and others.® | DISTRIBUTION OF DOUGLAS FIR. The local distribution of Douglas fir in Oregon and Washingtor is definitely related with the climatic conditions prevailing in the — 3 For methods of study used in this investigation,. see Appendix A. NATURAL REGENERATION OF DOUGLAS FIR. 3 different parts of the region. These relations are brought out by Figure 1 and Table 1. The limits of the range of Douglas fir are shown in Figure 2. The climatic units into which the Douglas fir region of Oregon and Washington’ has been divided in Table 1 are based on weather records taken at the regular Weather Bureau stations. Although the weather records often are not taken at points typical of the forest regions, the averages of the records taken at all stations of each region are reliable indices of the relative differences between its climatic characteristics and those of other regions similarly determined. TABLE 1.—Vemperature and precipitation in the Douglas-fir region of Oregon and Washington based on records through periods of 5 to 64 uears with an average of about 16 years.* Mean : Mean | Highest | Lowest | temper-| . Mean | Maxi- annual ern annual mum per- | temper- ature as temper- | “stur a rare growing precipi- | annual ature. : ; Sensor tation. rainfall. Washington: FIDE oe vias WIDE Inches. | Inches. IWies OM@AScaAd CSaee ee ee aaa 50. 2 105 —6 56. 5 | 55. 56 2151. 56 Hast of Cascadeseics cess coon seats 47.7 108 — 30 61.3 14,75 8 35.77 Oregon: WiestronC@ascadess au ae ee oni tee 51.1 | 108 —16 57. 4 59. 51 4167. 29 Southwestern section............-...-. 52.7 110 —4 63. 1 31. 44 5 62. 66 HastaoniCascades eae se Gee eae 56.5 119 —27 63. 6 14. 55 6 43.65 oat Annual Latest | Earliest Precipita- Minmiam | tion durme) Ueto) Ol Cate date rainfall. | growing | C2YS1n the hs 1s Season growing frost in | frost in season. spring. | autumn. Washington: i Inches. Inches. Wiest Of; Oascad esi iis tose ys ye eyes Baca ee 715.00 18. 66 8199} Apr. 6! Oct. 30 - MAST ON CASCAMeS set sabi accesso o oe eae 5 « 93.71 2. 69 135 | May 25] Sept 30 regon: IWieS TOL @ascadesiias sae ois yu eeee 10 15, 53 17.29 197.) Apr Wi lOctymal Southwestern’section: - 352-2. 22¢ 5.5... 2. IST 99 4. 90 164 | May 7:| Oct. 13 Hastof Cascadesyorss ees gee se ee cess ee 12 4, 60 3. 21 142 | May 18] Oct. 2 1United States Department of Agriculture, 8 The length of the growing season does not coin- Weather Bureau records. For location of regions cide with the average dates of earliest and latest see figure 1. frosts due to the variation of the length of the 2 Clearwater, Jefferson County, 1899. periods through which records were taken. 3 Lyle, Klickitat County, 1894. 9 Ellensburg, Kittitas County, 1898. 4 Glenora, Tillamook County, 1896. 10 Roseburg, Douglas County, 1913. 5 West Fork, Douglas County, 1896. 11 Ashland, Jackson County, 1905. 6 The Dalles, Wasco County, 1858. 12 Umatilla, Umatilla County, 1898. 7 Port Townsend, Jefferson Veguney. 1889. In western Oregon and Washington Douglas fir is not exacting in its choice of site, except as it 1s influenced by altitude and lati- tude. This is chiefly because there is enough precipitation, especially during the long favorable growing season, to insure the establishment. of the seedling and to enable it to compete successfully with other species of trees and with shrubs. The longer growing season on the west side not only enables the Douglas fir to continue growth through a longer period, but it also insures better maturity, so that there is Tess frost injury than on the east side. Its best. development is below the altitude of 3,000 feet, although in Oregon it produces good timber trees up to about 4,300 feet. Its crowth is 4 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE always checked and sometimes inhibited by exposure to severe winds. For this reason it is not found near the open sea in exposed localities, although it occurs near by in sheltered inlets where protection is | afforded. | In eastern Oregon and Washington the Douglas fir either forms a mixed stand with species which usually occur on drier sites, sueh as western yellow pine (Pinus ponderosa Laws.) and lodgepole pine | (P. contorta Loud.), or it is found on moist slopes in mixture with species that require more moisture, such as western larch (Larix occidentalis Nutt.), white fir (Abies concolor Lindl. & Gord.) and | western hemlock. In either situation it produces inferior trees as compared with those produced in the western sections. This condi- | tion is readily explained by the limited precipitation during the P orine season, the shorter growing season, and cther climatic actors. 3 In southwestern Oregon‘ the Douglas-fir forest meets the western yellow pine, knobcone pine (Pinus attenuata Lemm.), and other species, and forms what may be called the border type, which is typical of the region. Deuglas fir covers the north slepes, and western yellow pine the south slopes, and many variations of these types also occur. CLIMATE AND SITE. Before proceeding to the main discussion of natural regeneration, it is desirable to consider briefly some of the climatic and site re- quirements of Douglas fir and its habits of seeding, because it is in these peculiarities that the solution of the problem les. These cub- jects are discussed in the next two chapters. LIGHT. The western white pine (P2nus monticola Doug].) is the only tree of the Pacific Northwest that requires more light than Douglas fir. The Douglas fir will grow in about one-fourth of the full ight in adjoining open areas, although under these conditions its develop- ment is retarded, and the more shade-enduring species, such as west- ern red cedar, western hemlock, and Sitka spruce (Picea sitchensis (Bong.) Traut & Mayer) have the advantage. The inability of Douglas fir to thrive in diffused light makes it in- capable of forming an understory. This characteristic is a disad- vantage to the tree in retaining its position in the forest, for the more shade-enduring species crowd out the Douglas fir and often completely replace it in the stand; but it is an advantage from a com- mercial standpoint, for this inability to withstand shade results in early pruning of the branches, so that a comparatively clear, straight bole is produced early in the development of the tree. On favorable sites, where the stands are dense, clear boles begin to form at 30 to 40 years of age, and at 40 to 50 years of age clear boles as high as 40 feet are often found. These characteristics illustrate the im- portance of a complete stand of young growth and the advantages’ of an even-aged stand. The relatively greater height growth of This region, bounded on-the west by the towns of Riddle, West Fork, Galice, and Mountain Ranch, on the north by the Umpqua River, and on the east by the Cascade ; Mountains, is referred to hereafter as southwestern Oregon. ' 4 : i : : aie ren ed ES poner SOURS oe ow ; ee ; : | \ \ ; ; : \ ; | ) | , t ' | i : | | ae | } tT ; . | i ‘ ) | : : | \ ; | ; } j | | | | N i i ‘ \ : | } | | | t | | | - | : | : j 7 | \ y | t 1 | : { | { | | 4 . ts | : a San | | . ‘ 7 * | \ \ was | > \ \ { | | . : cy bd : | i \ ‘ ‘ 4 } 1 \ | . \ - NATURAL REGENERATION OF DOUGLAS FIR. > Q Douglas fir during its early years helps to maintain it. Because it demands an abundance of overhead light, it produces the tallest and straightest stems in dense pure stands or in mixture with the more shade-enduring species. On the sites of poorer quality, especially in open stands, the lateral branches are persistent and form hard, dense wood. ‘These branches persist for 20 or 30 years after all foliage has died, and are embedded in a large section of the trunk. In localities where the growth is more rapid the lateral branches contain much softer wood and are not so persistent after the foliage has died. On good sites the favorable soil and abundant moisture enable Douglas-fir seedlings to endure more shade than on poorer sites. The same conditions also favor increased survival and growth of the shade-enduring western red cedar and western hemlock, with the re- sult that they often crowd out the Douglas fir. TEMPERATURE. Douglas fir is apparently adapted to severe climatic conditions in the Rocky Mountain region and on the east slopes of the Cascade Mountains of Oregon and Washington. However, the fast-growing Pacific-slope form of the species does not bear exposure to severe cold. In winter the cold, dry east winds sometimes kill the trees out- right and often kill the ‘growing tips, especially on the east side of the trees. Such conditions are particularly i injurious to young trees and either retard growth or kill the seedlings. Throughout the range of Douglas fir the seedlings are often killed on hot, exposed slopes through injury by heat to the cambium ring at the surface of the ground. It has been found that a temperature of 144° F. at the surface of the soil kills the cambium and causes girdling of the seedlings. This injury often affects seedlings or transplants in the nursery and has been described as “ stem girdle.” The cambium of older Douglas firs separates from the sapwood when it is heated above 160° F., and occasionally a scar results. If the temperature is raised to 200° F., the cambium becomes discolored and 1s permanently injured. The length of the growing season in the Douglas fir region is variable, and the seedlings have apparently not become adapted to this variability. Often late spring or early fall frosts cause exten- sive injury to young growth. If frosts occur after growth has started or before the buds mature, the buds—particularly the termi- nal buds—suffer, height growth is checked or completely stopped for one or more seasons, and the bushy seedlings, so common up to 4 or 5 years old, are formed. If the terminal buds are killed by frost, the lateral or adventitious buds develop, and it may be three or more years before a leading shoot is formed. The actual killing of young _ growth by frost is not common, but the death of seedlings as the result of heaving by frost is often extensive. The principal dis- _ advantage resulting to the tree, from its inability to withstand frost, is the loss of its place among its competitors. Western hemlock and _ western red cedar are often found uninjured by frost that has killed | the immature buds of Douglas fir. When in mixture with Douglas fir these species take advantage of the retardation of Douglas fir _and overtop it, thus eliminating it from the stands. 6 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. MOISTURE. Although young Douglas fir is found distributed on all sites throughout the region, its successful establishment and growth are largely controled by moisture. The poor growth along the Colum- bia River gorge, as compared with the adjoining side valleys, may be attributed to the drying east winds that sweep down the gorge. These winds cause excessive transpiration, which makes the buds mature early and consequently shortens the growing season. Site studies have been made in different localities, and the soil moisture correlated with the establishment and survival of Douglas- fir seedlings. The ability to extend its root system 6 or 8 inches deep during the early part of its first growing season is an important fac- tor in perpetuating Douglas fir. When it is in competition with such other species as western red cedar and western hemlock, which pro- duce shallow-rooted seedlings, the Douglas fir often is able to sur- vive where the other species. fail. Records of soil moisture taken in 1919 on the flat river bottoms and south slopes in the Cispus burn, near the Tower Rock Ranger Station north of Mount Adams, demonstrated the ability of the - Douglas fir to resist the adverse conditions of severe sites. On the | south slope the moisture in the surface soil reached.a minimum of 0.18 per cent in July and did not go above 0.85 per cent in August. At a depth of 6 inches the south slope soil contained 6.55 per cent of moisture in July and 5.50 per cent in August. The wilting co- efficient of 2-year-old Douglas fir seedlings was found to be 1.25 per cent for this soil. These data show conclusively that seedlings could not live in the surface layer of the soil, because-the available moisture was below their requirement. At a depth of 6 inches, however, there was sufficient moisture for growth throughout the season. In the flat river valley, where the soil is a silt loam, the surface soil contained 0.19 per cent of moisture in July and 0.09 per cent in August. At a depth of 6 inches the soil contained 11.45 per cent of moisture in July and 9.382 per cent in August. Obviously, then, seedlings that have root systems which penetrate to a 6-inch depth before July of their first season may become established in this region. These extremes of soil moisture are readily explained by the rec- ords of soil temperature and evaporation. The surface soil on the south slope reached a maximum temperature of 128° F. in July and 135° F. in August. From July 15 to October 1 evaporation records with the Forest Service evaporimeter showed an evaporation of 1,070 cubic centimeters on the south slope as compared with 690 cubic centimeters on the flat; that is, more than one and one-half times as much on the slope as on the fiat. With such severe conditions of temperature and moisture only those Douglas-fir seedlings that are protected by shrubs or annual plants survive the first one or two dry seasons and become established. After a forest canopy is formed and duff accumulates on the ground the site becomes more favorable; western red cedar and western hemlock are then able, gradually, to gain a foothold in the 5 Hofmann, J. V. The Establishment of a Douglas Fir Forest. Ecology, vol. 1, No. 1, January, 1920. NATURAL REGENERATION OF DOUGLAS FIR. 7 Douglas-fir forests, and so prevent Douglas fir from being the domi- nant tree for more than the first generation, except in those locali- ties which western red cedar or western hemlock do not reach by mi- gration until two or more generations of Douglas fir have occupied the ground. During the summer of 1913 meteorological readings were taken on the north slope, south slope, and flat of Warrens Gap in the Wind River Valley in southern Washington.* Some of the im- portant points shown by these readings are the maximum tempera- ture of the surface soil, soil moisture, and evaporation. The ex- tremes of all the factors mentioned occurred in July and August, which is the critical period for plants in this region. With a surface- soil temperature of 129.4° F. occurring on the south slope, while the north-slope maximum was 82.4° F., and with a surface-soil mois- ture of only 1 per cent on the south slope as compared with 6.5 per cent on the north slope, there is little need to conjecture about the failure of small, shallow-rooted seedlings on the south slope. _The comparatively short period of drought in this region, how- ever, ordinarily enables the Douglas-fir seedlings to become estab- lished.. Although the surface soil dries out to the point where seed- lings.can not survive, the water content of the soil at a depth of 6 inches usually remains above their minimum need. The soil moisture remained at 11.2 per cent on the south slope and 17.5 per cent on. the north slope at a depth of 6 inches. This amount of water pro- duces favorable conditions for growth. Although the water content of the soil is above the minimum requirement of the plant, generally _ the drier sites are exposed to higher temperatures and greater tran- spiration. The striking effect of evaporation on these sites is shown by the evaporation from open-water surfaces exposed only to vertical radiation, as, for the most part, are masses of vegetation. On the south slope 15.1 inches of water evaporated from an open water tank during the month of August, while only 1.8 inches evaporated on the north slope and 6 inches on the flat. The effect of protection from the sun’s rays on plants is shown by the survival of seedlings under the protection of shrubs on these severe sites and by their failure in the open, even where the moisture content of the soil is about the same. The greater demand for moisture by seedlings in the open because of greater transpiration can not be supplied when the mois- ture content of the soil approaches the minimum requirement. A vegetative cover of either pea vine or brush has a noticeable effect on the moisture content of the soil. To determine this effect ‘n a south exposure, an area was selected where the pea vine was very dense. One square rod was denuded of all vegetation, and the area beside the denuded plot was left intact. Readings of air tem- perature at the height of the crowns of seedlings, and of soil ‘emperature at the surface and at depths of 6 and 12 inches, were taken each week on each area. The results are given in Tables 2 and 3. : $Hofmann, J. V. ‘The Importance of Seed Characteristics in the Natural Reproduc- tion of Coniferous Forests. eindie= in the Biological Sciences, No. 2, University of | Minnesota. 1 8 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. TABLE 2.—E/ffect of vegetative cover on air and soil temperature. Average maxima—Degrees Fahrenheit. | May. June. July. | August. | September.} October. (2 et sil Air temperature:! | | | | NaturakcOver ssc. s see = See al 60.8 59.2 70.7 86.6 | 72.0 55.7 : aa Bl ow See Nhe w ce BeOS | 72.2 64.4 84.7 | 102.8 | 76.2 60.0 oil temperature: Natural cover, surface............. 55.6 56.0 62.7 73.4 | 63.5 56.2 Natural cover, 6 inches deep-..-.... 52.2 55.2 60.5 68. 4 61.0 54.5 Nature Cone. 12 inches deep..-... | es aot he oe 60.6 54.5 enu Surface £056 Ss. fos ht - ’ . 124. 89.2 (64.5 Denuded, 6 inches deep.....-..... | 528 62.0 68.0 78.7 67.0 54.2 Denuded, 12 inches deep.......... | 36.3 61.5 67.0 74.4 66.2 56.2 1 Air temp2rature taken at the crown of 1-year-old seedlings. Table 3 shows clearly the effect of evaporation from surface soil. In August the moisture in the denuded soil was 1 per cent, as com- pared with 10.5 per cent under the natural cover. The percentage of soil moisture at the 6-inch and 12-inch depths, however, was greater during the dry season on the denuded area than on the area having the natural cover of vegetation, because, where this cover was present, the moisture from the soil was absorbed by the roots of the plants. TABLE 3.—Effect of vegetative cover on soil-moisture content. Average percentages of soil moisture. May. | June. July. BERS ‘September. October. | | | Natural cover: Gariacel tit /f_ S008s) ee. TSE § 33.3 32.4 23.11 10.5 | 33.4 | 36.5 G@anchesdeen..- +2). 5.2 te 21.3 26.7 20.0 | 12.9 | 29.5 | 26.5 i2 inches deep LF 1.5. 2h. 23.9 20.5 18.6 | 15.4 | 28.4 | 27.7 Denuded: | | Stig Cas ile ted ae > RR ae Fea aia a 11.0 10.2 4.1] 1.0 | 12.7) 18.4 6 inches deep:.........-......2-2-- | 26.7 24.1 | 22: 51 17.5 | 24.2) 29.0 i 2332 25.7 | 21.1 19.8 | 27.2 | 29.2 ' j EZHRCRES ACE D ss. < Masses eo eee | The extreme maximum temperature of the surface soil of 124.2° F. in the denuded area, as compared with 73.4° F. at the same time under the natural cover, shows clearly to what rigorous conditions the seedlings are exposed when growing in the open on these severe slopes. The 12.9 per cent of soil moisture at the 6-inch depth in the natural cover, and the 17.5 per cent at the 6-inch depth in the de- nuded area, is evidence that seedlings with root systems 6 inches and deeper have moisture available even during extreme droughts. The deciding factor here, however, is the amount of drying result- ing from the exposure of the plant. A plant in the open is under a much more severe test than a plant under natural cover, for a plant exposed to the sun becomes considerably warmer than the sur- rounding air, whereas shaded plants become colder than the air. Even if the soil moisture is equal in two localities, the soil texture may have a decided influence on the availability of the moisture to the plants, as expressed by a marked difference in the wilting coefii- cients. This difference would not influence the types if the soil mois- NATURAL REGENERATION OF DOUGLAS FIR. , 9 ture were the same at all depths. But the fact that the surface soil often dries out, while the soil at a depth of 6 inches remains moist on protected slopes and dries on exposed slopes, changes the type and gives a decided advantage to the seedling with a deep root formed early in its development. In its early root growth the western yellow pine has the advantage over Douglas fir, western hemlock, and western red cedar, and it is largely on this account that the yellow pine forms the dry-slope type in the border zone of the Douglas-fir forest region. For the same reason Douglas fir is able to establish itself on the drier slopes of the Cascades, where the western red cedar and western hemlock fail. A south slope covered with Douglas fir and a north slope covered with western hemlock, western red cedar, and other species does not prove that each of these species is in its most favor- able situation, but that these are instances of successful competition and establishment. Where two types met on a ridge it was found that the south slope was seeded by the species of the north slope, and the seedlings of western hemlock and western red cedar were germinating along with those of the Douglas fir and western yellow pine in the spring. When the area was examined in the fall only seedlings of the western yellow pine and Douglas fir were left, because the small seedlings of the other species were unable to live through the dry period of the summer on account of their shorter roots and conse- gent inability to reach the moist layer of soil below the dry surface. These conditions are repeated year after year, and still the type remains the same.’ It is very noticeable that wherever a ravine or spring keeps the south slope moist the north-slope species are found. Evaporation, then, is one of the chief factors in the establishment of the seedlings, for, while the different slopes often get about the same amount of precipitation, there is such a marked difference in evaporation that the exposed slopes dry out while the north and protected slopes remain moist. SOIL. The loose volcanic-ash soil found in the Cispus region northwest of Mount Adams heats during a forest fire or when exposed to the sun’s rays. In this type of soil fire causes complete destruction of all vegetable matter and the seed stored in the forest floor is de- stroyed. The importance of seed stored in the forest floor in natural reforestation is discussed later. The hot, dry soil in these burned areas prevents establishment of seedlings after the fire. On the other hand, the heavy loam soil of the Willamette Valley and of other similar regions protects the forest floor from the heat of fires, and the greater moisture-holding capacity of this type of soil aids refor- estation. | Douglas fir grows best on sandy loam and reaches its greatest size on moist, porous, well-drained soil. It is absent from the wet bottom- land and marshy places and sphagnum bogs, on which Sitka spruce and lodgepole pine grow fairly well. It seldom occurs on light, dry, sandy soil or on heavy clay soil, and where it does occur on these soils it usually forms a light stand of poorly developed trees. ™Hofmann, J. V. Seed Vitality as a Factor in Determining Forest Types. The Ames Forester, Iowa State College, 1917. 60634 —24—_2 10 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. Soil may be largely responsible for limiting the distribution of a species, especially in local areas. Thus on the areas of serpentine in the Siskiyou Mountains of southwestern Oregon, knobcone pine produces a better growth during its seedling and pole stages than Douglas fir and takes possession of such shallow soil areas. Although the knobcone pine is only a scrub tree, the soil enables it to compete successfully with its associated timber trees, including the Douglas fir. The loose soil in the Coast and Cascade Mountain regions of Oregon and Washington affords good drainage, prevents erosion, and, with abundant rainfall, is very favorable to the growth of Douglas fir. SEED. The proposed methods of natural restocking of Douglas fir are in large part based on the characteristics of the seed and the stor- age of seed in the forest floor. In order to assure a stand of young growth after forest fires or logging, good crops of seed must be pro- duced, and the seed must be distributed and be able to retain its vitality until it has an opportunity to germinate. A study of the seed has established important facts concerning its produc- tion, distribution, germination, and viability—facts that must be clearly understood as the first step in solving the problem of natural regeneration of Douglas fir.’ SEED PRODUCTION. In the Douglas fir good seed years occur at irregular intervals, usually two or three years apart. Sometimes a fair crop of seed follows after a very heavy crop, and at other times practically no seed is produced in this region following a heavy seed year. Seed production is unquestionably influenced by the condition of the weather during the period of pollination. During seasons in which rains began at about the time the staminate flowers opened, and con- tinuous wet weather prevailed throughout the season of pollination, it was noted that the staminate flowers drooped, the pollen adhered to the stamens, and there was apparently very little distribution of the pollen. Even on trees that produced a good crop of pistillate flowers there were practically no cones. As the pistillate flowers are at the ends of the branches and the staminate flowers further down, there is very little pollination on the same branch, and the distribution of pollen to other branches is very hmited on account of its sticky condition. The wide variation of climatic conditions in mountainous regions causes a corresponding variation in the time of flowering. Although the staminate flowers on the same tree may mature before the pistillate, the trees at a higher elevation, in the same vicinity, may produce pollen at the proper time to fertilize the pistillate flowers of the trees in the valley. This cross-pollination tends to eliminate individuality among trees in the same locality, in so far as fertilization is concerned, although there are other factors affecting the development of the seed, such as soil and age of tree. The distribttion by-wind of the pollen of Douglas fir may cause cross-pollination for long distances during favorable periods. Pol- NATURAL REGENERATION OF DOUGLAS FIR, 1B len grains of forest. trees may be carried as far as 80 or 90 miles by wind.’ The method of pollination is an important factor in the seed production of trees that are left after logging or after a forest fire. Diseased trees are known to produce only about three-fifths as much good seed as sound trees.° The smaller amount of good seed pro- duced by diseased trees may be due to the infertility of the pollen. If diseased trees only were left on an area, the quantity of good seed produced would be small, and that might be a serious factor in seed production. However, if some diseased trees were left on a cutting area, and if a stand of Douglas fir remained in the surrounding ter- ritory, there would always be a chance of fertilization with pollen produced by sound trees. In that event the seed production would not be seriously handicapped. The average mature Douglas fir tree produces about 40,000 seeds per crop. There 1 1s a wide variation in seed production due to the age, size, and health of the tree, density of stand, soil, latitude, and alti- tude. Each of these factors influences seed production, although the direct effect of each individual factor can not be definitely stated. Trees 100 to 200 years old bear most prolifically. It was found that the average 15-year-old tree produced 4,000 seeds, the average 100 to 200 year old tree produced 40,000 seedsy and the average 600-year- old tree produced 7,000 seeds. At elevations of 300 to 600 feet. above - sea, level the average tree produced about 34,000 seeds per crop as compared with 4,000 seeds for the average tree at 3,000 to 4,000 feet above sea level. The effect of latitude is noticeable even within the hmits of Oregon and Washington. The average tree in central Oregon produced 35,000 seeds per tree, but a comparable average tree in northern Washington produced only 7,000 seeds. This, ap- parently, is a wide variation and may not be consistently maintained during each seed crop, but the figures indicate that latitude has a marked influence on seed production. The effect of the health of the tree is very noticeable. Diseased trees that were severely affected produced 7,700 seeds per tree, but sound trees of the same age class and same locality produced 14,200 seeds. In the check taken there was little effect in quantity production of seed that could be traced directly to soil conditions. The quality of the seed was the noticeable varying factor. Trees on poor soil produced only two- thirds as many good seeds as trees on good soil. Another factor which must not be overlooked in seed production in Oregon and Washington is the damage done by the insect Megastigmus spermotrophus Wachtl. This insect is always present and attacks some of the seed, but in a light seed year concentrates its attacks and becomes most destructive. For this reason a sparse seed crop in Douglas fir is usually equivalent to a failure. A careful examination of the cones before the seed matures during July and August will usually reveal immature stages of the seed- infesting insects. If cones of the past season are examined during the winter and spring, they will indicate whether or not the area is infested by these insects. 8 Hesselman, Henrik, Iakttagelser Over Skogstridspollens Spridningsf6rmaga (Dissemi- nation of Pollen from Forest Trees). Meddel. Statens Skogsférséksanst, 16: 27-60. ® Wil C. P.. and Hofmann, J. V. A Study of Douglas Fir Seed. Proceedings of the Society. ‘of American Foresters, Vol. X, No. 2, pp. 141-164. 1915. 12 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. The amount of seed destroyed by rodents also enters into the final result of the seed crop. Mice, chipmunks, and squirrels are very active in collecting and storing seeds and cones, their chief source of food, especially during the period of ripening. It has been noted that a white-footed mouse will eat 300 and a chipmunk 600 Douglas- — fir seeds a day when in captivity.° In regions where rodents are abundant, light crops of seeds are entirely consumed, as are also the seeds produced by scattered seed ‘trees. The average Douglas-fir tree produces about 24 bushels of cones per tree, with an average of 1,000 cones per bushel. Sometimes the number of cones produced is the direct cause of the variation in the amount of seed produced. However, diseased trees on poor soil produce almost as many cones as do sound trees and trees on good soil, and the variation in effective seed production is due entirely to the quality of seed produced. Other factors that effect the pro- duction of cones also effect the production of seed in direct propor- tion. The yield of cones per tree is greatest with medium-aged, good-sized trees that grow in open stands in warm localities. SEED DISTRIBUTION. The chief agents of distribution, or the determining factors in the migration or extension of the range of the species, are wind, animals, man, gravity, and sometimes in mountainous regions landslides and snowslides. | Wind had always been considered the prime agent in the migra- tion of conifers in the Pacific Northwest until investigations showed that its influence is not so far-reaching as was formerly believed. _An instance was noted of seed distribution from Douglas-fir trees about 125 feet tall at a time when the wind was blowing 15 to 20 miles an hour. It appeared that at least 90 per cent of the seeds fell within 1 chain (66 feet) of the parent tree, and not over 5 per cent were carried more than 2 chains. Occasionally seeds were carried great distances. Several seeds that were carried by the wind from 6 to 10 chains were gathered and found to be only empty shells, incapable of germination. This no doubt accounted for their separation from the other heavier seeds, which because of their weight continued their spiral course downward. Wind is a definite and persistent agent of distribution and occasionally car- ries seeds for long distances, but it usually is effective for dis- tances of only 3 to 5 chains from the parent tree. No dense stands of young growth, resulting from wind distribution, have been found at greater distances than these from seed trees. Rodents are among the important factors affecting the distribution of Douglas-fir seed, although in an indirect way. Rodents gather seed and store it in the forest floor for food, but do not carry it far from the seed trees. When they return, perhaps after a period of rain or snow,. they fail to find a good deal of the seed. Tests have shown that mice detect seed by scent, and that they are better able to find seed in mineral soil than in duff, especially when the duff is more than three-fourths of an inch deep. The seed buried by rodents furnishes the supply of stored seed needed for 10 Willis, C. P. The Control of Rodents in Field Sowing. Proceedings of the Society of American Foresters, Vol. IX, No. 3. 1914. i ] | NATURAL REGENERATION OF DOUGLAS FIR. Le eo | forest renewal when the forest is removed by fire or cutting. (PI. @ Lio) 1) The unintentional influence of man as a distributing agent on the | large burns or barren areas is negligible. Gravity assists distribution in mountainous or hilly regions. GERMINATION AND SEEDLING DEVELOPMENT. The factors necessary for germination are aeration, moisture, and favorable temperature. | Aeration is essential for plant growth, and the lack of it causes _ poor growth or death in growing plants and dormancy in seeds. | Even though conditions may, be favorable for the germination of | seeds, a lack of oxygen resulting from inadequate aeration will keep | them dormant. The seed that is buried out of reach of the air or is | in moist duff where the oxygen supply is small may remain dormant _ for long periods. Moisture is necessary to start germination, but too | much moisture may cause the death of the young sprout by the exelu- sion of oxygen. Although the seed may have sufficient moisture and oxygen for germination, it will remain dormant if it is kept cool. This is often the chief factor in keeping the seed dormant in the forest floor. | When the forest is removed, the temperature of the litter and soil | is raised and the seed springs to life. These same factors, with the addition of light, cause the seedlings to develop. However, a wide | variation of any one of these factors on different sites does not mean | that the widely varying factor 1s the one which determines the type, _ because other factors, varying less but approaching nearer to the _ limit of favorable conditions, may have a great influence on the estab- lishment of the seedling or on the germination of the seed. All the factors must be taken into consideration, and particularly the limits of each under which the seedlings will grow. The early development of the seedling is dependent on the food stored in the endosperm of the seed. This was demonstrated by sowing seeds of western yellow pine, Douglas fir, western hemlock, and western red cedar in sand, in soil to which nutrient solutions had been added, in potting soil made up of leaf mold and sand, and by germinating the seeds in distilled water. The following nutrient solution was used: To each liter of water were added 1 gram cal- cium nitrate, 0.25 gram potassium chloride, 0.25 gram magnesium sulphate, and 0.25 gram acid potassium phosphate. The soil was moistened with this solution and watered with it whenever necessary. The seeds germinated equally well under all of the conditions, but differences were noticeable very soon after germination. Seedlings that germinated in the sand came above the ground and appeared to be as good as those grown in the potting soil or in the nutrient solution until the seed coats were shed; then they began to fail, and apparently were unable to get any nourishment or, at least, not a sufficient amount to make growth. After the cotyledon stage these seedlings did not appear healthy; they either developed their resting buds or died. Those in the potting soil and in the nutrient solutions made a good growth and did not develop buds until they had passed through the regular growing period. Those grown in distilled water developed until the food in the seed was exhausted, and then they died. 14 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. Seeds in the forest floor are often covered with deep layers of litter and duif, and when the forest is removed the conditions may be favor- | able for the germination of the seeds. If, however, the seeds are | buried too deeply the food in: the seed is not sufficient to enable the seedling to grow to the surface. The effect of depth of cover, and the superiority of the seedling which springs from a large seed, are shown in Table 4. : TABLE 4.—Effect of depth of cover on germination. | i I | 1 Ap- H | : Depth | Gormi-| Pa Depth | _ | Ap- : ermi-| peared : Pt. | Germi-| BENE ee |nated. | above |) SE eS. anol nated. renawel | ground. |) | ground : _ Inches. | Per ct. | Per ct. | Inches.| Per ct. | Per ct Western yellow pine..... 1 82 | 82 || Western hemlock. ....... 0.25 | 96 Leds 83 | 7 BS 92 | 7 hs 71 | 2 75 86 50 | 1.0. | 64 5 “Douslaspfirs tess ose | a0 93 | 93 |, 1.25 | 42 | 0 ua 87 85 || Western red cedar....___- (12) 7 7 amelea 72 64 | 25| 64 52 [sta 67 | 50 || ee eee © 24 } 3 42 | 3 || | 15 | 25 4 4 17 0 |; 1.0 } 26 0 . 1.5 {19 | 0 Table 4 shows that seedlings will come up through a depth of soil in direct proportion to the size of the seed, and the development of the seedlings proves that they will grow to a size directly propor- tional to the size of the seed without any nourishment other than that stored in the endosperm of the seed. The fact that the seeds germi- nated even at the depth shown in this table and produced roots some- times 4 to 5 inches long, as did the western yellow pine, shows that the seedling is nourished by food stored in the seed until it can pro- duce chlorophyll bodies and manufacture its own food. If it can not reach the surface before the supply of nourishment in the seed is ex- hausted, it must die. If the seedling is able to get above the ground, even as a final effort, the cotyledons open at once and turn green, and the seedling gets a new supply of food. The loss of many of the seedlings germinating in the shade is caused by disease and is due only indirectly to shading, because the shade and moisture favor the development of the damping-off fungi. This may account for more of the seedlings being found in the open, but it does not necessarily mean that the species involved will not develop under more shade. The amount of shading that seedlings will endure largely determines the understory and, consequently, the succession of species. Douglas-fir seedlings will not survive in as dense shade as will the seedlings of western hemlock and western red cedar. For this reason the latter species eventually gain possession of an area that remains undisturbed. iat SEED VIABILITY. Viability of seed is one of the most important factors in the imme- diate reproduction of the forest as well as in the retention of a forest type. When a forest -is destroyed by fire, wind, cutting, or other NATURAL REGENERATION OF DOUGLAS FIR. 15 agencies, the immediate replacement of that forest depends upon the available supply of seed.11_ The species that is able to restock the area first may hold it through several generations or permanently, but to cdo so it must have a supply of seed at hand. There is no more imme- diate supply of seed than that stored in the forest floor, and the species that has a supply of stored seed takes possession of the area. In order to have a supply of stored seed, seed production and distri- . bution are essential, but the viability of the seed is even more i1m- portant. Seed may be stored, but if it does not retain its vitality it is of no avail to the species in replacing a forest. The condition found in the forest floor is shown in Plate I, Figure 2. RESISTANCE OF DOUGLAS FIR SEED TO HEAT. To find what temperature would kill Douglas-fir seed, lots of 200 seeds each were subjected for 10 hours to dry oven heat at tempera- tures from 100° to 300° F. Another series of tests, with temperatures varying from 100° to 240° F., was made in an oven in which the air was kept as nearly saturated as possible with water vapor. | Table 5 shows the germination from seed which was subjected to | these heating tests. TABLE 5.—LHffect of heat on germination of Douglas-fir seed. Degrees Fahrenheit of heat applied for 10 hours. not a is Mea ted 100 140 160 180 200 | 220 240 CO aouopocoboucebood soduedee 55 64 62 61.5 73 0.5 0 0 y 0 Moistiheate s2ccen 2 ee ek sok. 55 65 60 24 0 0 0 Douglas-fir seed will withstand a dry heat up to 200° F. and a moist heat. up to 160° F. for long periods. The effect of heat upon the seeds was studied with the microscope. No changes could be noted until the temperature was raised to within about 40° F. of the temperature that killed the seed. When this point was reached the endosperm began to darken, oils exuded, and the outer seed coat dried very noticeably. With the higher temperature the other seed coats also showed effects of drying. The intensity of all these changes was directly related to the degree of heat applied. In the forest floor the stored seed is, of course, surrounded by different conditions from those of the oven. In order, therefore, to ascertain the degree of protection afforded to seed during a slash fire Douglas-fir seed was artificially stored in four locations in a heavy Douglas fir-cedar-hemlock slash near the Wind River forest experiment station. The stations were selected in mineral soil in the middle of a skid road, in a rotten log, in mineral soil under duff, and in the duff. Temperature readings were taken at each station during the fire. The slash was burned on June 27, 1919, with a very hot fire, and resulted in as clean a burn as may be expected in this type of slash. (PI. IT.) 1 Hofmann, J. V. How Fires Destroy Our Forests. American Forestry, vol. 26, June, 192f, 16 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. The temperatures were recorded with a Leeds and Northrup potentiometer, which insured accurate records. The germination of seed that passed through the fire is given in Table 6. TABLE 6.—Germination from Douglas-fir seed stored in slash during fire. | | | gens Number | | Extreme ae | ofseeds (Percentage | tempera- os Goss germinated germinated.) tures | Pie }in nursery. | | during fire. ; | ite Station 1. Under 1 inch of mineral soil in skid road...... 9,104 | 2, 082 | 22.9 | 75 Station 2. 14 inches under surface of rotten log.......... 6, 751 | 1, 354 | 20.1 | 65 Station 3. In mineral soil under 1}inches of duff; direct- | | Ty 1nder ‘Statioit fost ee eer rn ee ee 9,118 | 1, 621 | 17.8 | 60 Station 4. Under 3 inch of duff; directly over station 3... 5, 235 | 1,281 | 24.5 120 Check. “Not-heated 2635 oo ee re eae pare eee ge 7, 552 2, 033 26; Sue es cee Station 5. 30 inches above ground. Directly over sta- | | | tions.) and 4.2... decel scare Seebeck coerce ARS. 8 hice anmarerd oes | arcing | 850 3 | The relatively low percentage of germination may be attributed to the poor quality of the seed used, as the check tests also show a very low germination. 3 The temperatures recorded in Table 6 explain some of the condi- tions found after slash, fires in this region. Strips of young growth commonly spring up along skid roads on certain areas, although there may be very little reproduction on the remainder of the area. The records show that any seeds buried by skidding would be pro- tected through the slash fire and would germinate along the skid reads. The same principle holds for areas lightly burned or burned when the duff is moist. When the forest floor is dry practically the only seed protected is that buried deep in the mineral soil. The moisture content of the soil and duff is a most important fac- tor in protecting the seeds during the fire, as the moisture prevents the duff from burning and reduces the temperature of the duff or soil. The moisture readings taken before and after the fire are given in Table 7. TABLE 7.—Moisture content of soil or duff where seed was stored. Percentage of mois- < ture content. | Before fire. “After fire. | Station 1. Under 1 inch of mineral soil in skid road..................- 2 ee eae |: 28. 81 11. 87 Station 2. 14 inches under surface of rotten log. .-..........--...---.-222-------- 81.71 72.78 Station 3. In mineral soil under 14 inches of duff; directly under station 4........) 9.19 12. 79 Station 4. Under 3 inch of duff; directly over station 3-.....................---- 58. 06 51.76 The fire apparently had some drying effect, as the moisture con- tent at each station, except station 8, was less after the fire. The drying at any of the stations was not pronounced enough.to- be in- jurious to the seed. : Bul. 1200, U. S. Dept. of Agriculture. Fic. 1.—A group of ninety-one 2-year-old Douglas fir seedlings that grew from a store of seeds presumably buried by rodents. Fia.2.—Cross section of typical forest floorin Douglas firtype. The following percentages of forest tree seeds were contained in each layer: (A) 80 to 85 per cent. of hemlock seed; 80 to 85 per cent of cedar seed; 40 to 50 per cent of Douglas fir seed. (B) 10 to 15 per cent of hemlock seed; 10 to 15 per cent ofcedar seed; 35 to 40 per cent of Douglas firseed. (C)0to 5 per cent ofhemlock seed; 0 to 5 percent ofcedar seed; 5 to 15 per cent of Douglas fir seed. The rule shown is divided into inches, with the top level with the surface. The kinds and amounts of seed in the layers of vegetable matter and soil explain how fire may change the proportion and kinds oftrees by burning to different depths. (See Hof- mann, J. V., Young Growth and How It Originates. West Coast Lumberman, Vol. 39, No. 463, Jan. 15, 1921.) Bul. 1200, U. S. Dept. of Agriculture. PLATE II. Fig. 1.—A heavy slash left after cutting an overmature stand of Douglas fir, western red cedar, and western hemlock. The white crosses indicate where temperature readings were taken during theslash fire. F1c.2.—After theslash fire. The pointsat which temperature readings were taken during the fire are marked with black crosses. The temperature at the station 30 inches above the ground reached 850° F.; at the point 14 inches under the vegetable material the temperature was raised only 48° F.; and at three-fourths inch under the mineral soil, only 5° F. Bul. 1200, U.S. LS PESO Saeed come fs A fe ghey PLATE III. Dept. ot Agriculture. a part of the Yacolt burn of 1902, and the area 1S vleW 1S Th wth. st fire. killed by one fore is now covered with a good stand of young gro ) lver fir, and western hemlock completely r, si A mature forest of Douglas fir, noble fi Bul. 1200, U. S. Dept. of Agriculture. PLATE IV. No seed 1 Forest. lona ia Nat se Hills, Columb i ingle fire on the Parad h sueceeded as 1¢ , wh and none were left after the fire. species ht, rees are in sig t A 20-year-old stand of Douglas fir, with western white pine and other 17 NATURAL REGENERATION OF DOUGLAS FIR. VIABILITY OF SEED IN THE FOREST FLOOR. In order to obtain records under controlled conditions of the longevity of seed in the forest floor an experiment was initiated in the summer of 1916. Thirty cages, 36 by 12 by 6 inches in size, of galvanized wire of 4-inch mesh, were used. Seed mixed with litter and duff was placed in the cages, which were then closed to exclude rodents, and placed in the forest at different localities. About 8,000 Douglas-fir seeds were stored in each cage. In the spring of 1917 three of the cages were taken up and germi- nation tests were made of their contents. A germination of about 15 per cent was secured during the summer of that year. When the cages were examined in place in the forest on July 6, 1917, it was found that rodents had dug around them in the attempt to get to the seed, and the burrows in many cases were left open. Along the edges of these burrows, where aeration and warmth had reached the seed in the cages, a vigorous germination was noted. Germina- tion tests in 1918 of this same seed resulted in very few seedlings, and no germination was secured from the stored seed in later years. This experiment proved conclusively that through the first year the seed in the forest floor remains sufficiently viable to produce good stands of reproduction and that some seeds remain viable until the second season. It is also known that seed remains viable in the forest floor much longer than two years and still produces good stands of young growth. The failure of germination in this experi- ment may be attributed to the artificial conditions of storage as com- pared with the condition of seed stored by rodents or in the cones. _Seed stored in the cones at the time of ripening or later is under more favorable storage conditions, especially in the matter of pro- tection from fire. Before the cones open, their scales afford effective protection to seed exposed to forest or slash fires. Cone scales are excellent insulation while the cones are still sealed, for a hot fire of short duration does not open the cones and the seeds escape the fire. The resistance to heat of seeds in unopened cones is shown in Table 8. TaBLeE 8.—LHffect produced on Douglas-fir seed by the heating of the cones.’ Temperature in degrees Fahrenheit inside of | Percent- cone after exposure for the indicated number | 48° of | Percent- of minutes. weight | age of Temperature outside of cone. lost in | germina- moisture | tion of | during seed. 1 3 5 6 7 10 11 15 | heating. Core CST Cape as MRP es te SU SOON a ee SHA al ay Nese ill pea WS egeeee 183 | 190 50.3 54 SE Pe eee emai te Savere apn ag seme [ Oe Sil | eee eters 20 eee NG yz |Przen ees 210 56, 1 61 SOU Aes Se Ws Ve Aue aN aD whe ee SOE Leesa) 52 See Oe eeoe LZO RE S228 230 55. 0 Ol TOO OMe ea ah Waa mene vere Oi ee as, eae TESTO VASES oo a ie ela 51.6 16 OO 2! RN ER ESL: Oe ae LOBED AC PT Pala eh be 39.0 41 DAC See ie eevee vee Ee eT eS IGS) 4 esa A eee A a Nc elt FEY 42.0 0 1 Sixty-four tests of individual cones were made, from which the series given in Table 8 is Selected to Show the effect of the range of temperature from below the ignition point of Dcuglas-fir wood to a point higher than that ordinarily reached in slash fires or fires that have air drafts mixed through flame, as do forest fires. While a crown fire is traveling the flame remains in a single crown from about 15 seconds to a minute, with a temperature of 900° to 1,100° F. 60634— 24 3 The ignition point of Douglas-fir needles is 650° F., 18 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. and at 900° F. the fire spreads rapidly through green needles. A comparison of these figures shows that mature seed would live through the fire if the seed were in cones on the trees. ORIGIN OF YOUNG GROWTH. In the Douglas-fir region of Oregon and Washington conditions are little less than ideal for the replacement of the forest by natural means, providing the harvesting of the crop and fire protection are properly managed. The Douglas fir can dominate all its competitors in this region, except under unusual conditions, and these conditions, fortunately, can be controlled by man. Fire is the most destructive agent at work in the forests, and yet it is responsible for retaining the Douglas-fir forests in some localities. The effects of fire on the Douglas-fir forests have been determined by a thorough analysis of the results of various types of forest and slash fires. Some of the most terrific fires of the Pacific Northwest have failed to wipe out the Douglas-fir forests, and splendid stands of young growth now clothe the areas that were burned. DISTRIBUTION OF YOUNG GROWTH AFTER ONE FIRE IN A MATURE FOREST. The Columbia burn (locally known as the Yacolt burn) afforded an excellent opportunity for the study of the origin and distribution of young growth following a single fire in a mature forest. The Columbia fire burned northward from:the Columbia. River in the Columbia National Forest in southern Washington over an area of about 250,000 acres in the western foothills of the Cascade Moun- tains at elevations of 500 to 4,000 feet (Pl. III). At the lower alti- tudes the forest traversed by the fire was the Douglas-fir type, which includes Douglas fir, western hemlock, western red cedar, western white pine, and grand fir (Abies grandis Lindl.). Above 1,100 feet silver fir makes its appearance, and then noble fir; and at about 3,000 or 3,500 feet the forest develops into the true fir type, com- posed almost entirely of noble and silver fir, with a shght admixture of western white pine and Douglas fir. Pacific yew (Taxus brevi- folia Nutt.) is distributed almost throughout the forest, avoiding only the subalpine summits of the higher ridges. Dwarf juniper (Juniperus communis L.), on the other hand, is restricted to the sub- alpine summits. ) The fire occurred from September 8 to 12, 1902, following an exceptionally dry season, and driven by a dry southeast wind it traveled to the northwest. So far as can be determined from local information it progressed at a maximum rate of perhaps 8 miles an hour during the time it was doing the most damage. No portion of the area studied had been burned over by a second fire. The burn was studied 11 years after the fire occurred. The main feature of interest found on the burn was the good stand of young growth which almost uniformly covered the area and consisted of the same species as those which made up the burned forest. The presence of this reproduction is obvicus to anyone passing through the area, but the reason for its appearance after so severe a fire has always been open to conjecture. The problem, then, was to determine the history of the reproduction and, so far as possible, to account for its distribution. NATURAL REGENERATION OF DOUGLAS FIR. 19 An arbitrary section, chosen to include Lookout Mountain, was studied intensively by a gridiron system of belt transects, which were run 24 chains’? apart over the entire section. Then, with this section as a hub, a township surrounding it was studied extensively. For this study eight transects were run radially from the centers of the four sides and from the four corners of the section to the cor- responding points in the township. Wherever a solid body of green timber was encountered the transect was discontinued. These beit transects served effectively to disclose the distribution of reproduc- tion over the entire township. The plan of the survey is shown in Figure 3. The lines radiating from the center section represent the transects which were run in making the study of the township. There was little young growth on the south and east slopes of Lookout Mountain, and such as did occur was confined to the draws below the barren slopes. On the north and west slopes, however, it was uniformly scattered. This distribution of the young growth is due to the local topography. The fire approached the mountain from the southeast and swept up these slopes with unusual inten- sity. After the fire the south and east slopes were hot, dry sites and were consequently unfavorable to the establishment of seed- lings. On the other hand, on the north and west slopes not only was the fire less intense, but the site was inherently more favor- able to seedling growth. | The most significant facts are found in the distribution of the age classes and their relative proportion, as shown in Table 9. This table shows that 58.9 per cent of all Douglas-fir seedlings germinated the first year after the fire, 28.7 per cent 2 to 6 years after the fire, and 12.4 per cent 7 to 11 years after the fire. Taste 9—Classification of young growth according to age classes and distri- bution on section studied in Yacolt burn.* Percentage of total number of seedlings | Percentage of all¥seed- in each age class lings found, according Percent-| found within each to age classes. age of distance. Distance toe Species. ftom seca examined| Time of germination. | Time of germination. ees. | included in each distance.| First | 2to6 |7tol1| First | 2to6 | 7tolt year | years | years ; year | years | years after aiter after after after after fire. fire. fire. fire. fire. fire. Chains. Over 10... 79.0 61.1 38. 2 0.7 | Douglas fire sees AU sees GEO Ohi 9.7 48.1 33.3 18.6 58.9 28. 7 | 12. £ Oltoras=: 10186} 67.6 14.6 17.8 c Over 10 100.0 15.8 BAD i ee eat aes Western white pine............ i COpLON a. Himes mie IRN INO Fe al lg gh LOW SW AGN eaten sa ONC ORS Meee sl fed eet aS essa Nees el RS Taee el GA UN late GUS | Over 10... 64.2 | 96.0 Ba) JC IN ODIG fir see see ae Nie OtoplOzz ae: 20.4 | 90.0 5.4 4.6 87.5 6.0 6.5 (0) 760) WE See 15.4 76.5 9.4 14.1 | 1 Over 10. 56.6 | 87.9 10.2 1.9 Silver fin ssa eee se ace i COM ONsese 24,4) 71.5 18.7 9.8 71.5 As 11.2 Opto LOFON moos 22.9 22.0 Over 10. $2.0 36.8 57.8 5.4 Western hemlock.............. 3 fe LOPS ShO Rea 76.9 23.1 18.4 67.4 14.2 50) aged is 0S ae ea al @ The total area included in transects was 18.6 acres. b One chain equals 66 feet; 10 chains equal one-eighth mile. 2 One chain equals 66 feet. 20 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. The proportion of age classes shows that about one-half of the total Douglas-fir reproduction started the first year after the fire, about one-third from 2 to 6 years, and the remainder from 7 to 11 years after the fire. The decrease in germination of seed throughout the section after the first few years subsequent to the burn and the small percentage of the young age classes found at more than 10 chains from seed trees indicated that the remaining seed trees had not been an important factor in the restocking of the area, espe- cially at distances over 10 chains from seed trees. These facts are supported by the records of the township study given in Table 10. i TABLE 10.—ClaSsification of young growth according to age classes and distribu- tion on township studied in Yacolt burn. Percentage of total number of seedlings - Percentage of all seed- in each age class lings found, according Percent- found within each to age classes. age of distance. Re thee Distance | ae PBEGES- acacia examined Time of germination. | Time of germination. : included ; ; in each | | Sees] distance.| First | 2t06 |7to11} First | 2to6 | 7toll | year | years | years ; year | years | years | after | aiter | atter after | after | after fire. fire fire. | fire. | fire. | fire. j | | : | Chains.2 | 4 lige , Over 10... 89.9 80.6 7.0 12.4 | ID QuUClAS, frees LA ere Pie ee GO S24) 4.6 68.5 14.4 LMT ee 6805 TSG 20.4 0 to 5. 5S oko | pis ae) LSEaT || | | Over 10 OBO 2heS.| Age PEs ae | Western white pine............ i told.” 120, | Oto a SOO) eo eee } BTS PWS Tee OTOH ESS} 1.0 20.0 SOSH Aes eee Fa 3 | Over 10...! 89-7 1 OE 3.0 2.9 Nobicitir ce. oo. ee ee i to 10 5.5 | 190.9 aD 6.9 84.1 ASTD ATED | WOE ESR See! 48ulias Ole 2 8.8 24.0 | Over 10 Se tain = ORY Det 1.4 | SULMershit woes eee ees see | ‘ to 10:22:22 | 55 75.0 DES 4n Sah 2 3 8258) 516: 74 15 NLO topossse= | 9.3; 80.0 20 Re Gelee | | fOver 19 78.4} 26.51 53.0} 20.5 | Western heme .--eeeeseee 8 to 10 | 953 25.0 45.8 29. 2 23:16, jen so0nk | 26.3 0 Co Sseaeeqs) 12-3 49.3 51.6 29.1 | | _ 1 The total area included in transects was 32.25 acres. 2 One chain equals 66 feet; 10 chains equal one-eighth mile. The Douglas-fir germination over the entire area in the first year after the fire was more than half of the total germination. The fact that the germination after the second year following the fire was con- fined largely to within 10 chains of seed trees indicated that the remaining seed trees were not casting seed over the burn as a whole, as germination conditions were still favorable when the examination was made 11 years after the fire. , The section and township records demonstrate that reproduction which germinated 7 to 11 years after the fire is hmited to the vicinity of seed trees or to localities which seed can reach because of favor- able ‘topography. Trees on a hillside above a canyon disseminate their seed over a wider range of territory than trees on level ground, for the seed can be blown down into the canyon and across to the opposite slope. There were no Dougilas fir seed trees on the section studied (see fig. 3), except a single broken-topped one, but there were Douglas ey =o ee (oS eee Ns ee EQ a re shown \\ ' A STUDY OF NATURAL REPRODUCTION \ DOUGLAS FIR E IN 1913 BY MAD! WIND RIVER EXPERIMENT STATION ON THE YACOLT BURN OF 1902. COLUMBIA NATL FOREST WASHINGTON Seale 4°=Imile Contour Interyal 100" Topogrephy by Wernstedt BKracbel Mapped by Gi nvut Legend ASS Ss . 1234.5 Years old [4 67,8810 Years old WYeors old =, ~ AN WY Each loasernewraicare =) [Eek laopuchicre acicara atta eee wyparatee re certs Scwesiey SS 12, 878810 %earsold = [Ul Years od se ea Sn . Over500 per Acre Over 500 per Acre ees RS Noble Fir White Pine Sees RS NV vs A Amabali H. western Hem Se ESS x BETAS ue ougia: © Wester ease STEEN ANN, ‘ x Sees Rt EEN OY HAN Sea Sen WSS SN os CY OD SS NN Ss SS SK SER SN ORO RRR Over |00yrs old TT ae ZA \ 2] oD S, ST \ K ~ ea s SS Ss =<" tx SRG Set RSA \ = ra SS, Ni Nii shown, 00634—24. (Face p, 20.) f1, 19. 3.—Map of township {n Yacolt burn, ‘Transects and reproduction after fire are > aa jot Poe ae <2 oe AE _— e . N Pts. 3 hig p of townahip ic Yea bi tp. Trangseetsa and: taprado ction * at Fy ae Pay if : Bei NATURAL REGENERATION OF DOUGLAS FIR, 2k fir seed trees within 2 or 3 chains of the northeast corner of the sec- tion on the northwest slope of Little Lookout Mountain. These trees were from 100 to 300 feet, above the areas on which germination occurred 7 to 11 years after the fire; hence it 1s entirely possible that they were responsible for the occurrence of this age class in the north- ern part of the section. The limited distribution of this same age class, with reference to seed trees and topography, was consistent throughout the section and was particularly conspicuous at several points in the township. (Vig. 3.) The transect from the west-central point of the section passed within -2 chains of green timber. The influence of this timber is shown in Figure 3 by the appearance of young growth which germi- nated 7 to 11 years after the fire for a distance of a few chains from the seed trees in Texas Gulch. The remainder of the transect had a scattered stand of reproduction of the older age classes. A very | dense stand occurred in Poison Gulch almost a mile from the nearest - seed trees. This same condition is illustrated on the southwest tran- sect, where the young growth of the older classes is very heavy along the north slope of Bear Creek Canyon at a distance of more than a mile from the nearest Douglas-fir timber. As the transect ap- proaches the timber at the top of the ridge, older seedlings are again found scattered here and there, and it is only close to the edge of the timber that the younger seedlings begin to appear at all. This pecul- iar distribution of the reproduction may be observed on all of the transects, and shows definitely that the green timber remaining after the fire has had little influence on the general occurrence of the Douglas-fir reproduction over the burn. The foregoing facts first cast a doubt upon the long-accepted theory of the restocking of large forest burns by the process of wind dissemination of seed and finally proved it untenable. As the study progressed and this fact grew steadily more convincing, the question naturally arose, ‘“ What was the source of seed for all this reproduction?” The answer to this question developed with the accumulation of evidence throughout the burn. It was found that the reproduction most often occurred, not in a solid unbroken cover, but in different-sized patches with irregular and ramifying bound- aries. Where the reproduction was lacking, the ground was coy- ered with grasses, herbaceous plants, and shrubs, evidencing an un- interrupted growth since the burn was formed. The occurrence | of these two types of cover made an interlaced pattern over the entire burn, although one type or the other often expanded solidly j over a slope or basin many acres in extent. Everywhere the feature that was most striking was the sharp line of demarcation between the reproduction and the grass areas. For all its tortuous wind- ings the boundary was always distinct. Obviously such a condition could not have resulted from natural seeding, but must rather have been produced by some force acting on the surface of the ground. The idea of ground fire suggested itself. One who has seen ground fire burning in forest duff will | remember that it burns irregularly, leaving here an island and there forming a deep bay between two points of unburned ground. When | at length the smoldering fire is stopped, the result is just such a | mosaic of burned and unburned surface corresponding to the mosaic 22 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. of reproduction and grass described above. A representative spot in the Columbia burn is illustrated by Figure 4. The most severe ground fires occur on dry sites and in localities where there is heavy litter and duff. They occur also where the fiames sweep the surface, as where a fire runs up hill. Under these conditions the soil is severely heated. The irregularity of the young growth on all sites led to the conclusion that where fire consumed iY) é Produc t/On we } AG / ; Was > 5 Lp iiegy fs, [ Area of dense//» 4 //) reproduction of? OE noble ftir about TReproduetion | (30.000 seec//ings My / per ecre . y | | ; | \ + \ VG 7: yy No ground fire | a bergie is | Y Ao Geese ‘ Yy, Ly Grass — No reproduction Pesu/t of ground fire WM, sproduGrion ey Fic. 4.—Effect ef ground fire on distribution of young growth. the duff and heated the soil no reproduction was possible except close to seed trees. With the recurrence of irregular patches of reproduction in areas where the litter and duff were not completely destroyed, and in moist places where the forest floor had not been heated, it was con- — cluded that the seed from which the young growth sprang on the Yacolt burn and on other similar burns that were studied, must have been stored in the duff or soil before the fire. (Pl. IV.) NATURAL REGENERATION OF DOUGLAS FIR. Do DISTRIBUTION OF YOUNG GROWTH AFTER ONE FIRE IN FORESTS OF DIFFERENT AGE CLASSES. The Cispus fire of 1902 afforded an opportunity to study an area in which a single fire burned in two distinct age classes of forest. This area was in the region where the Tower Rock ranger station on the Rainier National Forest now stands. One age class was a mature Douglas-fir forest, and the other a second-growth stand of pole-sized, nearly pure Douglas fir about 40 years old. The young forests that followed after burning in each of the two age classes were strikingly different. In the mature Douglas-fir forest the conditions after the burn practically duplicated the conditions found on the Columbia burn. ‘There was the same distribution of dense and sparse reproduction independent of the position of seed trees; there was essentially the same proportion of age classes in the reproduction, evidencing the greatest germination during the first two years after the fire; there was the same alternation of patches of the original brush cover and of reproduction, outlining the limits of the irregular ground fire. There was; moreover, abundant proof of the inadequacy of the sur- viving seed trees to restock a burned forest area of large extent. There were left on this burn numerous groups of living seed trees of Douglas fir and minor species ideally placed for a study of their influence in reseeding the area. In spite of the most favorable con- ditions of site for the germination and establishment of seedlings, it was invariably found that 1, 2, and 3 year old seedlings were limited to a radius not greater than 3 or 4 chains from seed trees. On the other hand, seedlings 10 and 11 years old were found everywhere, even at 12 chains, the maximum distance from seed trees attainable in the area. The destruction of a portion of the pole-sized stand of Douglas fir by the fire of 1902 produced peculiar conditions. When examined in 1914 the ground was occupied by a dense cover of shrubs and ferns which grew up through a network of dead-and-down logs of small diameter. Reproduction occurred on this area in spite of the heavy brush cover, but it was comparatively scarce. The reproduction, in its struggle against the surrounding brush, developed the tall, lank “shade form” common to any thicket or forest-grown plant. The older seedlings occurred independently of the position of seed trees, but showed a tendency to be limited to moist sites, such as occur in ravines and hollows, where the conditions were most favorable imme- diately after the fire for germination and establishment. The younger seedlings (1 to 5 years old), however, showed an increase in density in the neighborhood of seed trees. Several seedlings of this class were found dead as a result of excessive drought and heat. in open places, but they were rarely found either lving or dead at distances over 6 chains from seed trees. The occurrence of two distinct age classes of seedlings and their positions in the burn with reference to seed trees indicate again that there were two sources of seed, namely, seed trees and the duff and soil in which seeds were stored. The reasons for the failure of seed trees to restock the burn have been discussed before. The scarcity of seedlings from duff-stored seed, however, presented a problem peculiar to this burn. The dense cover of shrubby and 24 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. herbaceous growth on the area might account for a scarcity or ab- sence of seedlings from “ wind-blown” seed, but it would not ac- count for the scarcity of seedlings from stored seed. Reproduction from stored seed, as a rule, starts at the same time as the brush on the burn, and does not come in after the brush has taken possession of the ground. Hence, the reproduction has merely to hold its own in height growth with the brush, and this it has frequently proved itself able to do. The reason for the scarcity of reproduction from stored seed must. be sought in the condition of the stand. The fire burned, not in an old virgin forest, but in a young forest, which was itself successor to a burned mature forest. The earlier fire (1860) killed most of the veteran trees which were the chief source of seed then stored in the forest floor. From the stored seed of the old forest there resulted a thrifty second growth of almost pure Douglas fr. How complete this stand was could not be determined; it is likely that there were openings resulting from ground fire in parts of the original burn. In 1902, at a time when this young forest was seeding, but long befere it had shed enough litter on the ground for the storage of its seed this forest also was destroyed by fire. In the hot fire of 1902 some of the duff was undoubtedly burned and consequently some of the stored seed. The result of this combination of cir- cumstances was the occurrence of scattered, inadequate reproduc- tion amid a rank growth of brush and weeds. The reproduction was more plentiful in moist or depressed spots, where the duff had in a large measure, escaped the ground fire and provided favorable conditions for the germination and establishment of seedlings after the fire. The findings on the Cispus burn corroborated those established on the Columbia burn, and brought out additional facts regarding the effect of the original stand upon the reproduction that follows after a single fire. Where the young stand of timber was destroyed by fire, only a thin stand of reproduction occurred on the burn. The reason for the light reproduction was found in the small accumulation of duff on the floor of the-young forest, and in the smaller accumulation of seed in the duff which resulted from the limited production of seed by the young trees. EFFECT OF SOIL CONDITIONS ON DISTRIBUTION OF YOUNG GROWTH FOLLOWING ONE FOREST FIRE. Even where a mature forest of uniform type is burned by a single fire, densely stocked areas of young growth may occur alternately with sparsely stocked or barren areas. Where reproduction does not follow the first forest fire the explana- tion often lies in the condition of the soil. An extensive area on which practically no young growth followed the first fire was found in the Cispus burn of 1910 along the Johnson Creek Trail. The area was characterized by light volcanic ash soil, which, no doubt, was the chief reason for barrenness of the area after one fire. This type of soil is so dry and porous that a fire will heat it, burn all the duff, and even destroy the seed that may be buried in the mineral soil, so there is no possibility of reproduction from seed stored in the NATURAL REGENERATION OF DOUGLAS FIR. 25 duff. If a fire occurs in this type during a period when there is no mature seed in cones on the trees there is little possibility of seed be- ing left to restock the area. When the loose mineral-soil is exposed it becomes dry, and the establishment of seedlings, even in the vicinity of trees from which seed is distributed after fire, is very rare. Seedlings which do be- come established usually occur under logs or, as has often been noted, in the hollows of burned-out stumps, where the moisture is retained. | Another large burn known as the Two Lakes fire of July, 1918, in the upper Cispus region, showed a striking correlation of soil conditions and distribution of young growth. ‘The forest floor under mature timber in this upper region is quite different from the forest floor of the lower valleys. The layer of dry needles and moss was completely consumed in the drier localities and the loose, porous, ashy soil beneath was severely heated. Stored seed could not sur- vive such severe burning, so that what restocking there was had to come from seed from the remaining living trees. Although 1918 was a good seed year for all of the species of this region the condition of the area after the burn showed that the dis- tribution of the young growth was not dependent on the remaining seed trees. On the other hand, on areas where the entire forest had been killed by a crown fire, a consistent relation between the condition of the soil and the occurrence of young growth was found. The dry slopes and ridges were barren or very sparsely stocked, and the moist valleys and north slopes were well stocked. In some places there were as many as 40,000 seedlings to the acre. The relation between soil conditions and distribution of young growth is shown by a typical transect which illustrates conditions on the burn as a whole. (Table 11 and fig. 5.) | TaBLE 11.—Number of seedlings per acre found on a belt transect one year after mature timber had been killed by the Two Lakes fire of July 12-18, 1918o% Number of seedlings per acre. Chains! distant | from green tim- ber. Douglas Noble Silver Western Western |Engelmann otal fir. fir. fir. hemlock. | redcedar. | spruce.? ; 1 a pat roo A S20 uae weer see 246 CUA Fea See SOee 80 720 DES EES ARIS cay areca 80 SOR ree Mae Se SO see neal se 80 320 Bees tc ullyal eee es A A cat tA I eS a SOM tie eR a 2 80 CEES ieee eas oa ete PEMD) i Oa aad Ll gts Fe Ease ok ae 80 160 Os ce eV Ares pe seiner ea bah nk altarar cea ebaga eet aye cok aie a IO SON Seon e ues 80 160 GR Tes eRe Baa ko IG Sale a iy Me NP Pt Lt Ma eh Ga eR i Ne ota 0 a ae eee | SSS) SHPO set It BS a ONT LY Si SI by CN Ne 80 Pa NP Ea tC A eas SNS ER BRD ee SB Sep Spa yo SOF SE we eae epee 80 QU CREE HR Mn Het ACA As A er 80 LGUs eee en ees | ee ee 240 BN a aE ate Sed RTI ca | eee ey eee nc aH oy eS LENE aU STs ey LOg EN i Le Cea 0 5 Uo Spay pane Sele aS 21D RE AUR DALY Oe A OP ae Ok BPA Ve ee See ie 2 Ee a ie ae 320 Ot, Maneater gies | HAE SeSeaC DOMAIN a sah SANE AN PVN EG ea a a i a RN 240 iL SE Ri SE [ESS SC iC a nt lta is Teta 8 ch i ia ee Ni Lan 0 1A BR ae | BPEL) Petes Pa PT CESS TY RU Sa a A 320 FS SE OSES TE eI Ol eae TES ree eH es EI a ea LPN a a ee eee a ELSE Es AO aR Va a SC 0 1 (G4) 2 ate a rae Bec Hie a9. Tea a Re Open ae N eM ane aa! Sol AN. Oak ML epee gear a gat 0 TEE AE A eee R A Somes 5 Be eae Nes Sel Pepe Ane BO ern M glee eh dete emery anya me NG wie ee 0 EP REE Sa Sr mela a pay ni Ks PMN LN epeta Sal Se ec a DU T) LL eat RR GOL dee Na NM EA Oe MI ae (aft 0 LO LLANE LLL SPE ode cael ce oe ciel BS le ca df Yo Se BS aE A A PE a 0 Q0seke ssi ds ceed Petiee: = SOU SO ec EEE a Eee Mee By 3 ie Ra ese eye Ses SA ae Pes Ce 0 7 We am ie SESS Sits Sees eal lic eee a ried al eta eae aU PERC Bes Pali, AN TBSS RG UN he 80 Cpa EE Uyb ayia ehh ene Wea ti eieette UR Bt It Re Seger aan oid ris aa gt | Pe Bg ea (i 0 PRY. didi ese TSE est ep tS cls Eat se | RR a OR BR DAE ee | Lr ae AE i 0 DEE SA ADO PS ee Ma OY A ORE REY V8 TS Dane a8 QO) area hy: oe ers ann es 80 11 chain equals 66 feet; 40 chains equal one-half mile. 60634244 2 Picea engelmanni Engelm. 26 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURR. TABLE 11.—Number of seedlings per acre found on a belt transect one year after mature timber had been killed by the Two Lakes fire, etc.—Continued. Number of seedlings per acre. Chains distant | from green tim- | | ber. | Douglas Noble Silver Western | Western |Engelmann Total | fir far fir. hemlock. | red cedar. spruce. Be | | | PSY a. Dy aKa mh eel Hi eA ee ang Roh Fa esr ERS Ale exp | hh cen Lis 3) 80 PASE HAE Liege LG Oi BORON EAe UARE RN meee Ase to ee SLO INS RES AAAS aU UR Vat YAO RE aN 240 Pare NE ON Se iy 240 PLO [ee Rn OO 0 UL Ge UTA a QS SNR 6 320 DOr ee aeL) WE OM, | LG OS ae aes AA RI 0 PAO OU ee EE: aoa 400 Fo nS A Oe SU ONG | 80 SOc Son ete ANT OVI Wei ae DN ee Searle Hy Ah eso 8 800 BOER ernie cum ar GTC SL OT sept yi ea (Ba ale th 160 | SO) at Rae. 400 au 2 0 BE ea aS wpe Ce ZAO HS cei. . Ser yee aces 560 | 240 Ws, eck eee ae 1, 040 Pa SN EP as SUE NY SE SUN i HERE ACN LER ee SUMO UA gen WE 2s 2) 2s Nal Deere ee 0 Sees a ahaa 5 ed 2 LOOMS Sa ee 80 AQQU ICS (RRM SAE ee ee re 640 SU Ea Ee Rees Cena ADO) er SUE sey 240 880: 560 160 2, 240 Sa Sl als Mme ee AR (eee ste Sac, She wa Nene Monies Meee Tat Bed eo ae a 80 560 SOT Cg ns RAY 480 80 SO olin Fcc edt 3 2 Jeg, Rei ad a 160 800 Be UN oe Sinica AQQ Here tert Mite SO alee eis Gee 80 80 640 SENSU TATE. ees ZEON Ss A ES ae SS Sa DAV i SO 320 1, 040 SOR UNG HA: TUE ee. 560 80 320 SO ala ee 720 1, 760 cE eel denen a gS 800 240 80 PO4Q eet Bed etECh 400 2, 560 CO ee aN eee Loar DAO!) breenetonr He cps 80 1,120 | 320 490 2, 160 ADI ENS D ON, 5 aes SMe 240 160 240 2, 880 | 240 1, 040 4, 800 Sera ka NE TH AU AE ASME Bs ME he 560 160 | 240 1, 040 3, 120 AA la eRe ES EMIT Mia 880 1, 200 2,320 ELOUNaS sore ony ame Siete 80 80 960 2, 160 4, 800 320 8, 400 AG ASE ag see re irs 720 560 3, 200 9, 840 | 19, 200 640 34, 160 257 ll sara CAA 400 580 1, 680 1, 0640 | 10, 560 1,120 24, 960 $8 4588 ERE 8 400 160 640 880 | 240 560 2, 880 ASF (ak Nec EE ae 160 160 240 800 240 320 1,920 Oo) ea arctic del azn 1,120 89 720 880 | GOA sheen hin Sea 2, 960 Olean UB bs S80 thes Sinaia lens 400 LOO pee ee cat 80 1, 520 ES el snipe a Gea anh DAH Seen ten neres 240 160 | SUR Dies seek sean The scattered stand of young growth on this transect adjacent to the green timber, and the alternation of sparse stands and barren areas up to 25 chains from the green timber are evidence that the seed was not distributed from the remaining seed trees, and that the dense stands of reproduction that occurred farther out in the burn were a to seed that was on the area before the fire and lived through the re. From chains 38 to 51 the soil was moist, and conditions for the pro- tection of seed during the fire and for the establishment of seedlings after the fire were very favorable. These, no doubt, were the princi- pal factors responsible for the dense stands of young growth that were found in this locality. The barren areas nearer to the green timber lie on more exposed slopes, where the soil was susceptible to heating during the fire and unfavorable to the establishment of seed- lings after the fire. Often there was a pronounced line of demarca- tion between the good stands of young growth and the barren areas along moist draws or slopes, as shown in Figure 5. EFFECT OF GROUND FIRE ON CHARACTER AND DENSITY OF YOUNG GROWTH. A ground fire may be the decisive factor in the regeneration of the forest. Where it creeps through the duff and litter it destroys nearly all the seed in the forest floor and leaves a bed of deep ashes and ex- posed mineral soil. Ifa ground fire occurs in the fall when the seed is about mature or before it falls from the trees, the new crop of seed produces a dense stand of seedlings in the immediate vicinity of the seed trees. ; 7 i ; a i ; f i ie 4 \ 7 . # ‘ y, orf : i : t 4 ‘ sek ») } i 4 | 4 ( f | i * a ; A hoses - ~” est was kille CBS We She HM ccd Bn OD Z0Q0 Soi] Conditions Slope or Exposure Number of Seed lings | 720 per. Acre, in each chain distance. ts 4) Chains Distance from Green Timber 320 80 ic Ash. Loose tf - Ory 160| 0 80 Profile of Transect on which Records were Taken —__~Profile of Skyline within 10 chains of Transect ° | | pa | re Sandy Loam. Volcanic Ash.tloos r« Clay Loam va |Norith-|Moist »4 Flat}; Diry Le Northt Moist 240| O | 320] 240) © | 320} 0 | 0 | 0} O| O | O} 80} O | O | 80} 80} 240} 370] 400] 800] 400 | 1040) 0 | 640|2240 11040} 1760, 2560 2160 4800 az 3120 43 2320) 44 | Vic. 5.—Effect of soil conditions on distribution of young growth on the Cispus area where a mature forest was killed by a single fire in July, 1918. Examined September, 1919. 8400)34160| 24960 7880) r Reburned Area 1920) 2960) 1520 48 Bil |) bye 60634—24, (Face p. 26.) sApwt tdi AG e vere Te (AS te plik Ae oT1 rats Elaawls ng A catinngio\ | id Ss eqouts bao? li Be | a heduto Heit} : | a swzoqi 10 Sq cid ‘| = a | | | 1 Del Oat | Zon: ibsamtonsdmuy moro dase ni. sisA. f= aos Soke . \ NATURAL REGENERATION OF DOUGLAS FIR. 27 A fire that occurred near Guler, Wash., on September 2, 1915, ran through the crowns of medium-aged Douglas-fire trees that were heavily laden with mature cones, and killed but did not burn them severely. At the same time, a severe ground fire burned the deep layer of litter and duff and undoubtedly heated the soil sufficiently to destroy practically all of the seed stored in the forest floor. Counts of young growth two years later showed that there were 17,600 Douglas-fir seedlings per acre, and 98.8 per cent of them came up the first year after the fire. Under these conditions, sometimes found after large forest fires, the young growth may come from seed in the cones at the time of the fire. That seed will live in cones during a crown fire has been proved experimentally. (PI. V, fig. 1.) If a ground fire occurs in a mixed forest of Douglas fir, western red cedar, and western hemlock, when the mature seed is on the trees, usually the greatest part of the resulting young growth is western red cedar and western hemlock, because of the large amounts of seed pro- duced by these species as compared with Douglas fir. The dense stand of seedlings may survive through the first season, but in the succeeding years the Douglas fir is severely handicapped. If the scorched canopy of the forest remains even for a year after the fire, the shade is too dense for the Douglas fir seedlings to gain a foot- hold, and they are crowded out of the stand. The mature Douglas fir trees generally are not killed by this type of fire on account of the thick root bark at the crown of the roots, but the thin bark at the root crown of the western red cedar, western hemlock, noble fir, and silver fir leaves these species easy victims. The remaining green Douglas- fir trees may continue to produce seed, but the shade of the dead and living trees is too dense for the success of the Douglas fir seedlings. Consequently the shade-enduring species, such as western red cedar and western hemlock, have the advantage; an understory of these species develops and ultimately changes the forest to the cedar- hemlock type. (Pl. V, fig. 2.) In places where a ground fire has occurred, counts on plots have shown stands containing more than 1,000,000 seedlings, 1 and 2 years old, to the acre, with only 5 to 10 per cent of Douglas fir. The Douglas-fir seedlings were tall and weak, and seemed to have little chance of survival, and observation in later years showed that they did not survive. SEED STORED IN THE FOREST FLOOR IS THE PRINCIPAL SOURCE OF YOUNG GROWTH THAT FOLLOWS AFTER ONE FOREST FIRE. Although young growth from different sources of seed follows after various types of forest fires, the seed in the forest floor before the fire is obviously the most important factor in restocking the burned areas. As a general rule, restocking may be expected after a single burn; but such factors as soil, condition of forest floor, and type of fire may cause exceptions to the rule. The general restock- ing after one burn is shown in the summary of a representative num- ber of transects that were taken in areas burned under different con- ditions and in different places. This summary appears in Table 12. The averages are based on the number of seedlings found in each chain-length of transects that were run from the edge of green timber to a distance of 120 chains out into the burns. 28 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. TaBLe 12.—Average number of seedlings per acre in each chain distance from timber on areas where mature timber was entirely killed by one forest fire. (Ages of burns varied from 2 to 11 years, with corresponding variation in ages of seedlings.) Number of seedlings per acre. fl | Chains? distant from green timber. | = sane a | | ougias | Western estern = } | red cedar. | hemlock. | Noblesin-g ) Pewat | pe OMe A Me es lg ee | 1,341 | 210 | 713 249. | 2,506 Tite je" Lies tapenade sepia pele M ea oa a 1,256 | 313 | 1,619 259 3) 447 Sp IeE TLE erties wits. ore rf PL in | 769 | 533 | 719 254 2315 che Sad dhe ee ai a am earn | 753 72 | 679 | 305 | 1) 309 tae Uae 8 a ty Se eel ae OF Se | 678 | 60 | 456 | 314 | 1) 508 CS SY a5, Dn a ee G76 | 105 | 1, 132 260 2.173 Thea ahs pig ARE Sag A eta ep et a a f 487 38 | 1, 249 293 | 2,339 Rama A eps 222 Fes atte haces. 271 | 300 | 507 173 1,351 Do Ge eR ee ee Se oer 345 |} 230 563 183°] 3,321 Lie Wn ib UCN Lee eee eae O25 ORS Me 8g ba PR AE 290 230 1,193 | 112 | 1, 825 i ede ay Ae SANs iy bah OS 237 | 200 | 35 155 942 igre Ay Mckee ATT. eats £2. ds 261 | 180 | 232 140 863 SEL Sete aaa ee ea ott sale Rice ah 998 | 140 | 230 156 754 OS Fe aga Sout ef 333 | 130 | 177 113 753 Ce slide eee OR eee Rene aay ayn 240 | 129 | 330 | 140 830 GG eh PAD) TIMI n Vio 160 | 145 114 75 494 iy eS os ee ae arias Seer | 202 | 130 | 118 | 95 545 PRE eae eee ee ee ee ae 202 | 140 | 120 | 142] - 604 $ORON TT Oe eR OF SR ya9s ce 144 | 130 | 231 90 645 ULES Dane Te Saal Ona. RO Era | 208 | 170 302 152 832 Do Se Se CAS Si ae. Ge ee ee 245 160 267 93 765 Ti, asic Gait cts altel i eatery pedis: Be 210 | 155 | 393 | 46 804 Tsyeve Meet see his atl ere cht neg | 161 169 | 270 68 659 De ES ag Oe ane eae the ioale 288 | 155 240 | 16 699 Deeee RAs SF $s et SERRE SS SMES Pe 243 | 130 | 294 | 43 710 Dee ite ge oe Se os 157 | 140 230 20 597 eer ane ee oe ae EA NSS 222 | 129 | 253 32 687 oe Le ee Sens. ae es ee Pee eS 187 170 204 39 591 Dowie A ee ee eo cee 162 | 190 | 313 75 740 SUISSE Ah d all > Te 165 | 200 | 323 62 750 Bitte, ees ENA OR LES 183 | 380 186 | 73 822 Ons (Loa SPR et ER ar ee 217 27 84 80 651 Soe ee ae Meee ena Geeet Ok Sage 155 230 159 -47 632 oy a RE a heey s eee y Metts re ees Bee 253 575 313 | 90 1,231 SH Seca 2 ue ae Oe Oe eae 297 | 210 136 | 72 715 Spm ieee re 274 205 100. 103 682 ay Es ae Ss ey ee eee ey eed 325 200 222 96 843 Si ch ot cee Maa dipall ae bane 293 199 276 | 78 837 Speee gy Slices) enyer ete SO) 334 | 189 | 270 | 95 879 iUics 24 Sh SNS aaa a aie Snes ean: 555 170 | 543 | 172 1,440 CALL 2 Ue aca Wa de cae Sy eee, 246 315 585 | 55 1, 201 meee aes Seat NF AAAS SILL 9d 235 270 | 1, 489 | 128 2,113 RPE OF APL for ony Pkt Bry se 8 682 | 320 | 80 | 42 1,124 pe ee ree ee ee 213 305 | 440 | 57 | 1,515 pea tee err e GR ere eT Daas 199 2.555 | 1,089 | 107 | 3,932 SD. ck 17 I SESS tis Sera pee aaa 478 S330 | 4,920 | 338 | 15,566 LU Sia ot es A EE Ghee ee 315 | 5, 440 | 5,320 | 340 11,415 AG. We SRI. eT SST PAIS SS OE ey 308 | 300 | 440 | 118 , 166 CQ elit ies i eee ay | 295 | 280 413 | 181 1,099 Ga cect le RS care Re Bk wes i imams pnt 696 | 289 | 456 | 135 1, 567 | Sir ae cane i eee Say OSs pean ota 586 170 | 90 | 34 | 930 HORE Re Shae BESS. Sah eed ge OS da 240 220 | 87 | 90 637 Ca are ae Sogn Some cers ees Smee tre 237 (es arvsl 7 203 447 Ee eee ee ee OGsia ess eee. 30 | 220 513 7s be SMR) RET EEG ig eee Cd les ag ere a | 3 hs to reee Le 40. 273 586 Gite LSet We Ge | shee Sete PTD Peper WE ¢ 67 | 273 | 620 Bins APES OF ML EY TE DETAELET hartge 53 240 560 Petes PAR EY ns et eee od Dre LS We} 47 197 507 iene ated © Re, oS ch pig le) Sen tego Olea da le i a eee eee 37 | 257 531 Guatnrigers. fico oy i" alees. D7 Folie £9 oe esa ge 40 230 593 Gist TTL £0 ee aih)) ROL a sae 47 153 560 Sie eee Mh) a Ee Sa) eee 53 167 593 Gane eee St ETAT SS BEE EA gerll Sid 69 153 600 6 tL Ai er ois ae eis. | porn wats | ae pane oF 47 || 127 | 527 1 Based on 18 belt transects 8} feet wide, each 120 chains long, making a total of 27 miles ofline. Transects are taken from burned areas examined on the Columbia, Rainier, Oregon, Snoqualmie, Santiam, and Crater National Forests. 21 chain equals 66 feet; 80 chains equals 1 mile. 3 A sufficient number of areas more than 52 chains from green timber were not examined in the western red cedar region to arrive at a fair average. NATURAL REGENERATION OF DOUGLAS FIR. ® 29 TABLE 12.—Average number of scedlings per acre in each chain distance from timber on areas where mature timber was entirely killed, etc.—Continued. Chains distant from green timber. Number of seedlings per acre. Douglas Western Western - fr. red cedar. | hemlock. Noble fir. Tay ase aR Utell os a as OA ge A SOON ES Mae eee. 46 163 CGN a ae SEY Pee DR Ges weer a raed ay SAO Nes siete Uy 37 170 GEC PV Ua AN ae Le | COMA CER A rs OSH ed essa iy sie ak 27 173 ete ey MM a PO LN ASE Oe ADB ease P ne 20 o7 Ca EN AE TS Oh a as tara SS ra a ee BOTA Site Sa Me 13 47 CAO) ee ate a ut ng Mr Cent cH YB a A a it A BION aeeeak jee fk 43 CAS SIS ek GOP Sek oot ES I UIE A pe OOo ae see noe see 3 20 (PS SEER UDB Her AIO BG A ae Ns ea a A TD ee eae ee 20 23 Sater OR Rs AAS OE Ac MC Se CES cr ain eaetiests 13 43 TRESS A ah) te aN IAN ERNEST) 2 ot eA pana eA 613) E eS A 20 67 TADS BS MA te EY GAN rete A Bs Va eee a pay aN Se GGT rE ie gs 27 147 CAG) Sirs pecan oe AE AS a SS 8 6837) Bote Wey 40 167 TELS pe Mite rh NG TI OOP Ard OE A ara AY 0 9 CATS erases Site ca 37 130 TES ABN Sa UN ee 18 Fe CRA Oa ae NA age FAO Ui etre ten Ce 30 150 CA) eS at UD ia clear gl lo Ne a SS A Daggett A i, 360 SSO) RIP ga So Ro Bee oa eS 3 i BOT yon es ey kee 40 353 Eo Sa SNE RAY MORI Ss EDT A A RRM es BD CED Wee Coa 33 333 FES NBII SR aR ers Su O18 MMT 2 ee a DOTButaae see es | 33 347 CSAS lig a eas ae de hf eg A ee QFSH eee ee wa 20 303 BEE ERED Jil a Phe OR CARN Ah coals Dir iN eG Ay de acd 10 350 Bp vera e ME BS yin Ulin Main Pint Salley are rs Cae cl (SB emer ea ese 57 80 SOE AN er. A CONNER eget SRE OL Sc 2 2 GES eee LE: 37 53 Se acres tae Sth ANNIE lag WO Tae SER Aran Oa ae GOS ee aces 40 50 tt ella ete ect adhe aia sg SR I eB i ie GOBU Eee LBs ee 33 57 BORE TIER ANS ROE UN Pewter Cats 8G El 0 CUBES Ee ar 47 63 CST Oana are a A pa TO DS SEC ST a aE CTSNet SE a 40 77 SONU tS Seats, OSC AM PSR IE MEIN A eg AE AG apa S00) Zeke soe 160 107 DING: SPN ESE au ce er ns YS RE Al Bee dts RS SM CSCO eke ae Es ie a 120 107 OBER iar R EOE 4 FN GB oan Hee Hila EO! eR Ee ICO Se Zaye 133 130 OA Ely a Liki Ween Rel ra yates Meher al ea A G5Sn Meine nae 127 140 CET PG Se a are a ts Be i BOM I eh a SSR eis et 140 127 SG) haa Petes ESL SRE SP Gray On Wk BR ae Neg DOM ese ene aya 137 133 SBN i all a aR ee La A ie ee lee OST ii RIA Lo ISS 113 257 ISAS SED HOA og Me ane eS de ep IS eo ae SOS tbe eee: 133 247 EIB iata ABN Ll yc tenes OR Sneaaiec WEL MLA | ater ayaas ea ra at 227 233 TOO oc can AAU Bi eee! AM RU A Geka a A DEST SO") (ASS eae ce 207 240 TCO ey Uae eat ie NA Ni Une hed Og Ak TACOS I AS oss 213 280 TPZ ake lad SURG Oe SSN sc ares ry Cer er Na DF O23) i ieee eee 140 233 THO ys en a i API RA Soa oe ai a Br O2 Te ene eu 193 333 S(O) Nal gs IR Ae Ae SNE BOA tn ee Ba EM Bk QUAa Tas y Meee 200 400 Vay es pS ae cr SEN este cy fee oA Mal pee a AS 3GO) il Pe Uae ae 393 447 TIO) 3) ean, a DE ce ae Pee pam JA Pg 5 SAO aE cae ees 400 447 SG Sa yi a EN aula, Rg. Ae eh ga Ma FOOT) | ten eee 35 260 HOSS es ce te yan a pT ces CA NO De SiC a sity a se ea 327 477 OQ OR RENT RT ek EET URS SEE here Ato VaR AE GAT le SS gee 420 140 TCG ee 8 ip Me E27) NAN Re eee PM Ee SPAT Geal MANIA els ase 427 137 DULL: eee Abeer ga) de Danese aceite Sy ope Or BOay ieee ee 320 373 HOI GS ae ee Ne a Ries ne as Cea ts AMM OE) Weiae et csiee 260 303 IS Yes 2 OU ee ae A ee aD eee UR AGO ul ah See 230 240 Ae eeu WP Mn rot wis Nike 5.20 2 e eal te GAG 7 mere we ee eth 213 260 UTS oa AS ee A LN CE ae ga a EOE AN inher s eg saa 257 420 OOS MCs Ae ior Bie RN a Ea a a a oe VaR A Se ISON sees seen 307 367 BT eee NE popstar ge cet NEILIN ES SA LUNA, et Yet BOOS NASER EERE A 260 400 ALU SYN Pk Pac Bs ee ee a ee nr ae eerie IR eau concosece 227 407 BS) Cy A ea RSE ey a le ae a a ey Te SGOH eee ee ees 197 363 SAD Ae ess ihe ee Oe 1S AN eee 8 ap 2 AUSO RIN eos Pert ae 113 430 An area that contains 500 or more seedlings per acre requires no additional stocking, but more dense stands up to 1,000 or 2,000 to the acre are desirable in the Douglas fir region. The average stands shown in Table 12 would be satisfactory for the entire region. Actually, however, the stands range from sparse to extremely dense, and a burned region may include some barren areas here and there. No effect on density of stocking can be attributed to green trees left after a fire except within a radius of about 5 chains from the seed trees. Where sparse stands of young growth follow the first fire on 30 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. areas too distant from green timber for restocking by wind-blown seed it is extremely important to protect the young trees in order that they may serve as seed trees for the area when they reach seed- - ing age. (PI. VI, fig. 1.) : DISTRIBUTION OF YOUNG GROWTH AFTER TWO OR MORE FIRES ON THE SAME AREA. Repeated fires on large burns are a more serious menace to a forest cover than the first fire, because if a reburn occurs before the new stand has reached seeding age it may produce an area which will remain barren for an indefinite period. (Pl. VI, fig. 2.) A number of fires had occurred in the Cispus region. on the Rainier National Forest, and consequently it afforded suitable areas for the study of the effects of repeated burnings. Evidence was found of a general fire over the entire region in 1660, and of local fires in 1761, 1874, 1892, 1902, and 1910. During the studies of 1914 the types of forest following these burns were analyzed. On the areas where the forest had reached maturity or had grown to middle age (100 years or more) the same conditions prevailed that were found after one forest fire; that is, a good stand of reproduction fol- lowed the first fire, and although the reproduction varied in density, the area could be classed as restocking. In the localities where a second or third fire occurred before the stand reached seeding age, only limited areas were restocking by seeding from the remaining seed trees or stands of timber. This was definitely checked by a series of permanent plots on areas that were burned in 1902, 1915, and 1918. A good stand of young growth followed the 1902 fire where the mature forest had been completely killed. When this young growth was 13 years old, and before it had reached an effec- tive seeding age, it was destroyed by the fire of 1915. The mature Douglas fir trees in the vicinity that had escaped the 1902 fire were not injured by the 1915 fire. This same area was burned by the general fire of 1918, and again the mature trees escaped being killed. Young growth appeared near the seed trees after each fire, but only for distances of 6 chains or less, and very little at more than 4 chains from the seed trees. Another area near the source of the Cispus River demonstrated the effect of repeated burning. A burn in 1764 was followed by a sparse stand which was reburned in 1870 at the age of 106 years. The burn of 1870 was again restocked with a scattered stand, which in turn was destroyed by the fire of 1910, when the stand was 40 years old. The 1910 burn still remains a denuded area. The explanation of this process of denudation lies in the successive fires that occurred before sufficient duff had accumulated to protect seed stored in the forest floor, and in the porosity of the soil which provided such good drainage that the humus and duff were dried out. Under these con- ditions the forest floor is usually burned severely; probably all of the litter and duff are consumed, and the soil is heated deeply enough to destroy most of the seed. This process has been noted in other forest types. It occurred again in the same locality during the 1918 fire, which swept over the Cispus and Spring Creek drainages. _ The large area burned by the Cispus fire of 1902 was to a great ex- tent covered by young growth, except locally, where subsequent fires burned in 1910 and 1915. On June 27, 1918, a second big fire, which NATURAL REGENERATION OF DOUGLAS FIR. ree covered about 60,000 acres, swept over this region. While the fire was running up the main Cispus Valley and up the North Fork and Nig- gerhead Creeks, it traveled at the rate of 8 to 4 miles an hour. It swept through the patches of green timber that had been left by the 1902 fire and entirely killed many of them. The dead snags and down logs in the old burn caused a very hot fire, and when it reached the boundaries of the 1902 fire it usually killed the green timber on a strip from a few chains to over half a mile in width. This fringe of fire-killed timber and patches of timber killed by the 1918 fire that had remained after the 1902 burn provided definite checks on the source of young growth following a first or second fire. The Douglas fir seed crop of 1918, being general and unusually heavy, afforded an excellent opportunity for restocking, because the fire con- sumed the vegetation on the old burn, and where it killed the green timber left mineral soil, which was an exceptionally good seed bed for the germination of the 1918 seed crop when it matured in the fall. The fire occurred so early in the season, however, that there was no possible chance for seed to mature before the trees were killed in the burn. of 1918. Any reproduction found on the area after this fire must of necessity have been from seed that was in the forest floor from previous seed crops, or it must have come from seed blown by wind or carried by other agencies from the adjoining green timber. The reproduction following the burn of 1902 was 16 years old at the time it was killed by the 1918 fire. A few Douglas firs were found that had been producing cones for two or three years before the fire, but the amount of seed produced on these areas was very small. The mineral soil was either exposed or had a vegetative cover. In the area burned, the loose mineral soil shifted and buried the seed. This has been found to be an important factor in reproduction in some places, and is no doubt one of the underlying causes of the scattered reproduction found in the 1902 burn that was reburned in 1918. Perhaps the most striking features on this area, from the standpoint of reproduction, are the establishment of seedlings where the green timber was killed by the 1918 fire and the practically com- plete absence of reproduction where the 1902 fire was followed by the 1918 fire. The studies of 1914 and 1915 provided an accurate history of this area before the big burn of 1918. For this reason the effects of the fire of 1918 could be analyzed without any speculation as to what might have occurred before the fire. The relation of the occurrence of seedlings to the areas of timber killed by the 1918 fire was care- fully analyzed by using belt transects, which were run from pvints out in the 1902 burn to the timber killed in 1918, and through this killed timber to the present stand of green timber. These transects showed that there was practically no reproduction out in the twice-burned area, and, where the lines approached to within 1, 2, or 3 chains of the edge of the timber killed in 1918 the reproduction appeared. ‘The stands of young growth varied in den- sity from a few hundred to nearly 40,000 to the acre without relation to the remaining green timber, but with a noticeable relation to soil and topographic conditions. The principal factor in the variation in the density of reproduction apparently was the moisture of the soil. On some of the transect 32 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. lines dense stands of reproduction were encountered more than half a mile from green timber, either on moist north slopes or in ravines. As these lines crossed dry, exposed slopes or ridges the reproduction became sparse. In some localities seedlings were found either under logs, in hollows, or in other places where there was moisture. The density of reproduction at long distances from green timber was in some places greater than near it. Where the transect lines passed through the 1902 burn into a patch of timber that had been killed by the 1918 fire, and through it into the 1902 burn again, the same conditions prevailed in approaching and leaving the timber killed in 1918 as were found where the lines approached green timber on the edges of the burn; that is, the reproduction was usually found from 1 to 3 chains from the edge of the timber killed in the 1918 fire. Dense stands of young growth were found in the areas of timber Timber killed by 1902 Fire Green Timber left by Fire Timber killed by 1902 Fire Reburned July 1918 ne Reburned July 19's aN tS, x : St RAG Wuite 70 | 0} 0} O| AO] O 1180} 620;1280)11Z0 16080 %680,7740|880 | 240, 80) O | O | O ie Oe 1 | | | . “Number of Seedlings | per. Acre,in each chain a hee distance. | | | | \ es) Chains Distance | | | | trom Green Timber. lo | 9 | Bal OU L7F Bm 9st 1O Sa ee ae | | | Fic. 6.—Effect of reburning. An area in the Cispus burn of July, 1918, that was re- stocked from seed stored in the forest ficor near seed trees left by the 1902 fire. There were no seed trees within three-fourths mile after the fire of 1918. Young growth that followed the 1902: fire had not reached seeding age in 1918. Such areas were not restocked after the second fire. killed in 1918; and, as the lines left the killed timber on the opposite side, the same conditions were found to be repeated in inverse order. until practically no reproduction was found out in the 1902 burn. Rig: *6" } | These phenomena are explained by the principles already dis- cussed; that is, the reproduction in this case came entirely from seed stored in the forest floor. There was no possibility of seed coming from cones; there was no evidence of seeding from the remaining timber; consequently, the only other source of seed was the forest floor. The young growth around the edges of the timber killed in 1918 shows the zone of seeding from this green timber before the fire. Although there was no litter and duff in this area, the loose soil covered some of the seed, and large quantities were undoubtedly buried by rodents. An area in the Wind River Valley of southern Washington, where the timber had been killed by the Yacolt fire of 1902, was covered . —_—— Bul. 1200, U. S. Dept. of Agriculture. Pate V.—NILUL Ze y ty dy \ SS Ae ye Fic. 1—A mature forest completely killed by a single forest fire. Cones are left on the trees after all needles and small twigs have been burned. Cones of Douglas fir, noble fir, and white pine are shown on the trees. Germina- tion has been obtained from cones that passed through forest fires. Fig. 2.—Silver fir, western red cedar, and western hemlock trees left after logging. If such trees are killed by the slash fires, they become a fire menace; and if they are left green, they help restock the area, to the exclusion of Douglas fir. They should be cut before the first slash fire. If there was sufficient Douglas fir in the original stand, a good young growth of that species may be secured. Gs Bul. 1200, U. S. Dept. of Agriculture. PLATE VI. Fic. 1—A scattered stand of young Douglas firon a burned area. Such scattered seedlings should be protected, as they are the potential seed trees for the entire area. Fic. 2.—A good stand of young growth that followed a 1902 fire on the Oregon National Forest was partially destroyed by a 1910 fire. The portion of the stand destroyed in 1910is not restocking. Bul. 1200, U. S. Dept. of Agriculture. PLATE VII. Fic. 1.—Group of noble fir, Douglas fir, and western hemlock in unburned slash. There are 27 seedlings per square yard. Fic. 2.—An excellent stand of young growth in unburned slash after clear cutting. Thisarea was logged 15 years ago and is now covered with a young forest consisting of about 20 per cent Sitka spruce, 20 per cent western red cedar, 10 per cent Douglas fir, and 50 per cent western hemlock. PLATE VIII. Bul. 1200, U. S. Dept. of Agriculture. i) as} ive} iS Sh AA Cae ale oe So ne oo yo BS coi] ai ee ©. bf ae oe RE) Dh Ou oo So 4 & 2 & an Br ae BS as bo (Se) Bb ado qk sa i ay bo is: (= Qy an ed Ss 58 On a6 oe a8 Ere ee ao Se | & = ie oC to) jo) B o i aH 2. Gp oO qs NS no, Dy Asl eo iH as ns oS 5o) 20 Ais oie tho Aa BS =m on Fae 8 8 36 Prey Aq EBS ere aS fe Los! a Ww) 8 > Loy) f Dou rowth is 18 to 20 years A pure stand o 2. Fie. NATURAL REGENERATION OF DOUGLAS FIR. ao with a good stand of young Douglas fir and associated species. In September, 1915, this area was reburned. The second fire did not kill any of the trees that had been left alive by the 1902 fire, but it destroyed all the young growth on the area. Forty-five plots, each 1 rod square, were laid out after the fire of 1915, and definite records were kept of the germination and survival of seedlings fol- lowing the second burn. Little reproduction was found on the area at distances more than 38 chains from seed trees, although stands of Douglas fir containing 8,640 seedlings per acre followed the sec- ond burn at distances within 2 chains of groups of seed trees. A few single green trees were scattered at distances of 15 to 20 chains from the groups of seed trees and near some of the reproduction plots. The plots near these seed trees did not have more seedlings than those at greater distances, although these seed trees were known to have produced seed crops. Undoubtedly the seed crops were gathered and destroyed by rodents. On a near-by flat area, burned at the same time, scattered reproduction occurred under the remaining seed trees, and none at distances of 4 chains or more from them. On all the areas that have been studied, where two or more fires have occurred, it has been found that the young growth following the second or later fires either occurs only in the immediate vicinity of seed trees, or the seedlings are scattered over great distances but occur only occasionally. Such reproduction may come from seeds that have been transported for long distances by wind or birds, or it may have come from seeds that have passed through exceptionally long periods of dormancy. The averages of the areas studied where reburns have occurred are given in Table 13. The contrast of the stands of young growth after the first burn, with the almost complete failure of reproduction after repeated fires, is evidence that the sources of seed for the two stands could not be the same. The reproduction near the green timber in areas burned twice or more has evidently come in large part, or entirely, _from seed trees, as indicated by the rapid decrease in the number of seedlings as the distance from seed trees increases. _ TABLE 13.—Average number of Douglas fir seedlings per acre after one forest fire in which all timber was killed* and on the same areas after they were reburned 10 to 15 years after first fire.’ Afterone| After Afterone| After Distance from seed trees. forest reburn- Distance from seed trees. forest reburn- fire. ing. fire. ing. Wiehe: ye oe ee casas 7,100 The PR MALLS AeA ONSIUe Base aie een te i o¥35020 9 ZRONUALNS Se eera safe tec ante an 5,360 Dis CHALNS HS Neha Ane Seay 3,000 0 S CHARIS "452. ufed Ser tenet ks 9, 660 ABO yl lsichainsie sy mea 4S a. fk 4,175 9 ACHAMIS. 2) ie Noose eet 3,490 DAM CHAINS Maen Geiss cin wien | 770 8 SCHAMISH Iasi d Eee 1,990 PRA Uo CHatOS ty: £/338 100 SS 2 22 ar | 195 8 19 0 aE Re as See oe ae 6,310 GAO Chains sass nee: cao n | 400 7 MCNSINS SS seat be cease eee eee 3, 800 AO VEN CMAIISs Seni uewee . cee 500 0 Schaims: serie? ee ery 800 AGS HES chains: § <5 jess ae bossa | 6, 850 0 DCN aS eo ot on aha (ry see ee 2,030 | s 22M AO CHATS YW feats a since Sam | 8, 820 6 TO ichains. VOLE See 2,455 AS P20 Chaim See NR ey eee | 4,840 3 | ! Examined 5 to 12 years after fire. _? Based on 172 square-rod plots examined annually for from 2 to 5 years, and on 3.75 miles of belt transect line examined from 1 to 5 years after the second fire. Areas at more than 20 chains from seed trees that had been reburned, of which previous records were taken, were not available. Plots were arranged in Series running out from seed trees; consequently results from plots and transects could beaveraged together. 3 1 chain equals 66 feet; 10 chains equals one-eighth mile. - 34 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. YOUNG GROWTH FOLLOWING LOGGING. DISTRIBUTION AND SPECIES IN UNBURNED SLASH. Dense stands of young growth were found in unburned slash on areas where no seed trees were left. The young growth consisted largely of western red cedar and western hemlock and contained less. of the other species that were in the stand before cutting. The vege- table matter and soil of the forest floor contain the seeds of all the species that form the stand, and those seeds germinate and form the new forest. In unburned slash the new forest usually contains a larger percentage of western red cedar and western hemlock, be- ue more of the seed of these small-seeded species is near the sur- ace. On an area of unburned slash near Fairfax, Wash., where a mature stand of Douglas fir, noble fir, and western hemlock had been clear cut, permanent plots were established and were examined each year for five years. At the end of the five years there was a dense stand of reproduction, and the density did not vary in relation to the dis- tance from the edge of the green timber. Some of the densest stands were 15 chains from green timber, and some of the poorer stands were within 2 chains of green timber. The age classes were evenly distributed throughout the area. (Pl. VII, fig. 1.) The rate of germination after cutting and the density of the stand on this area are given in Table 14. : Taste 14.—-Density of stand of seedlings and rate of germination in unburned slash near Fairfax, Wash. Percentage | Total num- A eae et of total ber of seed- venttine, | germination | lings per acre yee each year. each year. 1 20 12, 000 2 58 46, 800 3 13 54, 600 4 i 58, 800 5 2 60, 009 -1The stand consisted of 7 per cent Douglas fir, 76 per cent western hemlock, and 17 per cent noble fir. In the region near Hazel, Wash., the forest is approaching the ultimate cedar-hemlock type. The remaining Douglas firs are very scattered and consist of veterans about 450 years old among western red cedar and western hemlock. Although the scil is favorable for Douglas-fir reproduction, there is generally a relatively small per- centage of Douglas fir in the young growth after cutting. This may be attributed to the scant seed from overmature Douglas fir trees that were in the stand before cutting and to the destruction of this seed by rodents. One area of unburned slash that was examined three years after logging where there were two to six western hem- lock seed trees per acre, was covered with an average of about 12,000 seedlings per acre, 16 per cent of which were Douglas fir, 28 per cent cedar, and 56 per cent hemlock. The nearest western red cedar seed trees were at least 20 chains from the area, and there were no Douglas fir seed trees within one-half mile. Many of the western red cedar seedlings were found in clumps of five to seven individuals, which - NATURAL REGENERATION OF DOUGLAS FIR. 335) looked as if they had grown from buried cones. Such clumps occurred at such distances from seed trees that the cones could not have been borne on wind-broken twigs. Another area in this region contained an average of nearly 10,000 seedlings to the acre, consisting of western red cedar, western hem- lock, and Douglas fir from 1 to 4 years old. In this area there was no preponderance of western hemlock seedlings, although seed trees of this species were present, whereas the western red cedar seed trees were 10 chains and the Douglas fir 20 chains distant, respec- tively. An area near Oso, Wash., contained from 15,000 to 20,000 seedlings per acre of western red cedar, western hemlock, and Douglas fir. There were one to three seed trees of western red cedar and western hemlock per acre. The neareast Douglas-fir seed trees were more than 10 chains distant. The larger percentage of germination oc- curred the first year after logging, germination continued for three years, and some western hemlock ger rminated the fourth year. An area of unburned slash near Pacific Beach, Wash., 15 years after logging (Pl. VII, fig. 2), in a heavy forest of western red cedar, western hemlock, and spruce, with a few veteran Douglas firs, contained a good stand of reproduction consisting of 20 per cent Sitka spruce, 10 per cent. Douglas fir, 50 per cent western hem- lock, and 20 per cent western red cedar. Unburned slash near Knappa, Oreg., 20 years after logging, con- tained a thicket of western red cedar, western hemlock, and Douglas fir, with the western red cedar and western hemlock predominating in numbers. In contrast to this, another area in the same locality, where the timber had been killed by the forest fire of September, 1918, was logged in 1919, and the slash left unburned. Two years later there was a dense cover of young growth, consisting of 85 per cent Douglas fir, 10 per cent hemlock, and 5 per cent spruce. There was no green timber within one-half mile of the area after the fire of 1918. This is an illustration of the change of forest type caused by fire. Had the fire of 1918 not occurred on this area, and had the slash been left unburned, the young growth would have been largely hemlock, but now it is almost a pure stand of Douglas fir. Young growth invariably follows cutting on areas where ‘the slash is not burned, although the proportions of species may not be the most desirable. A summary of conditions found on areas where the slash was left unburned is given in Table 15. TABLE 15.—Average number of seedlings per acre in unburned slash 2 to 15 years after cutting Chains 2 distant from seed trees. Species. 1 2 3 4 5 6 7 8 9 10 DO Me1a Sy Mra ae ea 180 60 75 172 310 215 10 305 510 18 Western red cedar. ... 230 986 385 760 | 1,620 586 160 | 12,110 | 3,405 50 Western hemlock.....| 1,460 | 10,720 | 3,680 3,140 | 5,600} 2,615 | 3,805 | 35,680) 7,680 205 Totals pees. 1,870 | 11,766 | 4,140) 4,072) 7,530) 3,416 | 3,975 | 48,095 | 11,595 273 1 Based on 3.8 acres examined by the plot and transect method. ‘1 chain equals 66 feet; 20 chains equal one-fourth mile. 36 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. TABLE 15.—Average number of seedlings per acre in unburned slash 2 to 15 years after cutting—Continued. Chains distant from seed trees. Species. | il 12 13 | 14 15 16 17 18 19 20 |__| | | Douslasnn. Mh sese: 172 65 155 78 150 35 48, 115 210 800 Western red cedar..... O| 24 36 140 | 4,500 | 6,520] 3,610; 418] 1,670] 4,860 Western hemlock. ....| 1,410 | 1,280 | 3,120 | 12,980 | 24,320 | 1,460 | 3,120) 4,200) 6,120) 8,650 Poataler cet cece! 1,582 | 1,369 | 3,311 | 13,198 | 28,970 | 8,015 | 6,778 | 4,733 | 8,000 | 14,310: DISTRIBUTION AND SPECIES OF YOUNG GROWTH FOLLOWING SLASH FIRES. The large amount of débris left after logging in the Pacific North- west makes a very dangerous fire risk and necessitates burning the slash in order to protect the surrounding forests and property. This fire risk should be removed at the earliest opportunity. It is also | necessary for good forest management that the slash be burned the first season after cutting. The great amount of slash left after cutting a mature Douglas fir forest makes broadcast burning the only practicable measure of disposal. When a broadcast slash fire runs over an area it may affect the succeeding young growth in different ways. It may burn the surface layer of litter and duff and leave conditions favorable for a good stand of young growth with Douglas fir predominating; or, all the dutf may be consumed and a scattered stand of young growth result, except in very loose soil containing a good supply of seed; or, if the fire is hot enough to heat the mineral soil below the duff, a barren area may result. The heating of the soil is the important point, and this can be con- trolled by selecting the time for slash burning. In the spring the soil and duff are wet and little heating occurs. The young growth following spring burning is evidence of the desirability of burning the slash in the spring. An area in the Puget Sound region, which was logged in the fall of 1914, and on which there was a very severe slash fire in July, 1915, afforded an opportunity to study the effect of a late spring or summer slash fire. A series of plots, each a rod square, was estab- lished immediately after the fire, and anna examinations were made for five years. (Pl. VIII, fig. 1.) Very. few seedlings fol- lowed after the July slash fire, except in the vicimity of green timber. All the seed on this area was destroyed by one fire, a pos- sibility that must not be overlooked in determining the time of slash disposal. In the same region large areas of dense stands of young growth have followed slash fires that occurred in the early Sprmae or late fale CPL’ Vill, fe 2: "Pl EX, fie. 1) The season of burning and the condition of the forest floor at the time of burning often determine whether or not young growth will foliow a slash fire. On the area burned in July several small western red cedar and western hemlock trees were left standing, but they were all killed by the slash fire. The reproduction that followed consisted of a few scattered seedlings of western red cedar, western hemlock, and Douglas fir, eenerally in rotten wood or under logs. Some Douglas fir seeds had germinated under pieces of Ss Sete NATURAL REGENERATION OF DOUGLAS FIR. OT bark and had been able to force their way to the edge of the bark before the food stored in the seed was exhausted. Some spring burns have been found where but little young growth followed the slash fire, and good stands have been found on areas burned in summer or fall; but the importance of spring burning is emphasized by the superiority of the average stands of young growth on these burns as compared with the young growth on the later burns. The average conditions of restocking are shown in Table 16. Fall burning may sometimes be done almost as safely as spring burning, but usually it is too wet to burn the slash after the forest floor has become wet enough in the fall to protect the seed. TABLE 16.—Average number of Douglas-fir seedlings per acre on areas burned over by slash fires, examined 2 to 5 years after fires. Chains? distant from seed trees. 1 2 3 4 5 6 7 8 9 10, Slash burned in spring first | season after logging.......... 690 485 | 1,125 | 4,120 570 160 218 565 | 2,714 | 12,905 Slash burned in summer first season after logging.......... 310 165 415 65 50 0 | 0 45 15 0 Areas reburned after first | Slash finde) coe ees s heel 665 | 345 | 510.) 1654) 0410 0 0 0 | 0 65 Chains? distant from seed trees. 11 12 13 14 15 16 Lar Fo ae) 20 20 Siash burned in spring first sea- son aiter logging.../........-. 5,220 | 4,130 | 9,190 | 670 | 2,115 | 4,568 1,675 | 6,460 | 5,120 | 1,365 Slash burned in summer first season aiter logging........... | 0 0 6 | 16 65 140 85 15 209 0 Areas reburned after first slash PUTO are aaa oe ey Wane so Paoy 0 0 110 35 0 | 0 0 | 0 | 0 | t} 1 Based on 7.5 acres examined by plot and transect method. Early fall burns usually occur before the duff and soil are wet; consequently the effect isthe same as that ofsummer burns. Both types are included in this summary. 21 chain equals 66 feet; 20 chains equals one-fourth mile. On areas burned more than once it is possible that a few dormant seeds still remain. The endurance of such seeds may account for some fairly good stands at considerable distances from seed trees. Likewise, the distribution of seed by wind and birds, combined with favorable conditions for germination, may result in a scattered stand in some localities, but other areas with less favorable conditions re- maim banren, “Cel YX, fig: 2.) RESTOCKING BY SEED TREES. On one area studied no seed trees of any species were left, except scattered Douglas-fir trees. The nearest stand of timber was about one-fourth mile distant. On this area, four years after the fire, an average of 3,680 seedlings per acre was found. Of these seedlings, 1 per cent was Douglas fir, 16 per cent grand fir, and 83 per cent western hemlock. The occurrence of the grand fir and hemlock seed- lings can not be attributed to seed trees, and even the origin of the 1 per cent of Douglas fir seedlings is doubtful. Although the Douglas fir trees left on the area were known to have produced crops of seed, the entire crop was evidently destroyed by rodents. This f 38 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. same condition has been noted on other areas, and it often accounts for the establishment of the cedar and-hemlock even among Douglas fir seed trees. The cedar and hemlock seeds are less sought after by _ rodents when Douglas fir seeds are available for food. | The influence of scattered seed trees of Douglas fir was found to be lhmited (Pl. X), although gradually seedlings get a foothold in the immediate vicinty of the trees, as shown by Table 17. Scattered seed trees evidently will not insure a complete stand for some time, and it is doubtful if the stand would be complete until the seedlings reach seeding age and restock the remainder of the area. Therefore, in this gradual process of seeding an area, protection should be given not only to scattered seed trees, but also to scattered young growth. (PlKeby: TABLE 17.—Average number of Douglas fir seedlings per acre on areas burned two or three times, with scattered seed trees lefts | Average || Average | Average Distant from seed | number of Distant from seed | numberof || Distant from seed | number of trees. | seedlings trees. seedings | trees. seedlings per acre. peracre. | per acre. Agchiain2e2 ai brs: 106 || 15 chains........... 0 | 99 Ghainsas eee 0 WCWANSe 5.2 5s on 20S P16 chains’. 22232 bE Q?|| 30 chains... s<25. 266 0 DICHAIH Shree nos sae oe 53a) Pe GHalnSses eee e OQ jpsl ehainsesey soe a AICHSIN SES Seas nee 60) |2t8 chainsk ss. eee (PS Chibi San ee 0 HIChaINS sees ee oo MAS et Gichainse sees ee OuiSsachaims eer. sees 0 Gichainsas=. tose. See 0:||220 eHainse see 0. |} 34 chains... 2.2.2... 0 7 clintns 2000. @) 20 chains) oa). 92 || 35 chains........-.. | 0 Bienains SO 5io Ie) Sus. O ):22 echains- 2.2 2. &: Ol) 36iehains/ 22 7s 9352 | 0 OTE RATHS ieee oh ee Ua ASiel ames ee Oe HWP BVA ANGIE Se 0 Oreharmssten: bees eA) 80)" || 24.chains: 22223225 ON pS8iehains: 22a. s eee 16 Ivehainsee os} asec ZG 2oiChain See Sense Ouli7e9ichainss sae 0 PICHON os sone cas On e2oGhains = ease 20h) a0 Chains eee seas 0 HS Chainse Aja eo eke Os 27icharmnsee~=2 oe - 0 | WA CHAINS Sia = See See te O) 228 chains. .2e. fees 0 | 1 Based on 162-rod square plots examined annually for 5 years, and on 3.8 miles of belt transect line. 2 1 chain equals 66 feet; 40 chains equal one-half mile. The foregoing facts brought out in regard to the origin of young growth in the Pacific Northwest lay the foundation for the methods required to keep the forest lands productive. They show also that present methods of slash disposal and fire control tend to devastate forest lands, and furthermore, that the methods now used can be made to conform to the requirements for securing young growth with practically no additional expense. MIGRATION. The preceding chapter emphasized the point that the greater part of the reproduction is due to seed that was within the burned or cut area before the fire or cutting and that other seed-distributing agencies are only incidental in restocking forest lands. However, denuded areas are slowly reclaimed through migration, and the species are changed within the forest through succession and compe- tition. The effect of each of these factors on the composition and establishment of the Douglas fir forests is pointed out in the follow- ing discussions. Migration of Douglas fir is a very slow process. On burned areas young growth appears in unexpected places; but, wherever it has been possible to get a complete history of the area, the explanation of such stands has usually been. that there was some source of seed within the area and that some seed had remained viable in spite of NATURAL REGENERATION OF DOUGLAS FIR. 39 fire. Large regions south of Tacoma, Wash., known as the Steila- coom Plains, have not been covered with forests for a long time, perhaps even centuries. The only apparent reason forests do not now cover this region is that young growth has been regularly destroyed by fire. The plains were probably frequently burned over by the Indians, or possibly the fires were caused by other agencies. That the forest was held back by some foreign agency is evident from the progress it 1s now making in taking possession of these plains. Where the forest is pushing its frontiers out into hitherto unoccupied regions an opportunity is afforded to study migration. (PI. XIT.) These plains are, on the whole, somewhat unfavorable for forest growth, but some seedlings become established here and there within seeding distance of the forest. The belts of successively younger trees as the distance from seed- -producing trees Increases are con- clusive evidence that the forest is moving out into the prairie by steps. Each belt, representing a forward movement of 3 to 5 and in some instances 10 chains, contains age classes from 1 to probably 20 or 25 years old. The advance seedlings beyond this strip are so rare that they can hardly be considered in the migration of the forest. It is evident that when these advance strips reach seeding age the forest 1s ready to advance another strip of about the same width. After trees become established the forest floor soon resem- bles that of a typical Douglas fir forest, and forest conditions are developed. Where forests of this type have reached maturity and have been cut or burned, the succeeding growth is the same as that elsewhere in the Douglas fir region. This migrating type of forest is in such strong contrast with the stands of young growth generally found in the Douglas fir region of the Pacific Northwest that it is evident the seed from which the stand originated had a different source. The first forest that in- vaded this prairie region consists of a very uneven-aged stand con- taining trees from 1 to 40 years of age. The second forest following in the same region after the mature forest was destroyed is the typical even-aged young Douglas fir. On areas that have been burned two or three times in other parts of the Douglas fir region, this same type of migration has been noted, but the variation in ages 1s not so distinct. Scattered seed- lings of Douglas fir are frequently found at long distances from green timber. Sometimes clumps of young trees appear, and occa- sionally slopes are covered with good stands, although no other young growth is found in the vicinity. Such young stands or seed- lings have been found to come from seed that had its source within the area. Sometimes, however, it is not possible to obtain a definite record of the conditions through the different fires. The definite migration in the prairie region also makes it seem unlikely that the young growth appearing here and there in large burned areas origi- nated from seed that came from the green timber remaining around the edges of the burn. Migration in the Steilacoom Plains is a clear demonstration of the progress of a Douglas fir forest when it is dependent upon seed- ing from adjacent timber. The aridity of these plains reduces the number of seedlings; but the proportion of age classes at different distances from the timber shows the rate of seeding, as conditions 40 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. are equally unfavorable at these different distances. In localities where moisture and soil are more favorable, as in some sections of the Willamette Valley in Oregon, the forest makes more substantial progress through migration, because the percentage of seedlings established is greater. The distance covered by each seeding genera- tion, however, is not greatly increased. CHARACTER OF SECOND-GROWTH FORESTS. Usually the same forest type that existed before the action of the disturbing or destroying agencies is reestablished. An imme- diate forest succession without the usual intermediate stages results either from dormant seed left after the removal of the previous forest or from other exceptionally favorable conditions for reseed- ing. Immediate succession is characteristic in burns in the Pacific Northwest. In successions of this character the forest which was removed by fire, cutting, or other agencies is replaced by the identical species of the original forest, although usually in different pro- ortions. Wherever reproduction is found coming in after a fire which killed the entire forest, the dead trees and snags of the same species were also found. ; The extensive, unbroken forests of Douglas fir in the Pacific Northwest have developed primarily through fires. If this region were left without fire for 600 or 700 years, the forest of Douglas fir would be greatly reduced in area, and the best Douglas-fir soils would be occupied by western red cedar and western hemlock. With- out artificial interference, such as fire or logging, Douglas fir can not compete with western red cedar and western hemlock, whose ability to endure shade permits them to form an understory which crowds out the less shade-enduring Douglas fir. (Pls. XIII and XIV.) Tf the Douglas fir trees mature or are killed by any agent other than fire, and the western red cedar and western hemlock remain uninjured, the openings made by the removal of the Douglas firs are immediately filled by the understory of western red cedar and western hemlock. For this reason a Douglas fir forest seldom main- tains a pure stand after the first generation, and sometimes is en- tirely replaced at the end of the first generation. Stands of Douglas fir 50 years old with clear, straight boles and well-developed crowns sometimes have a complete understory of western red cedar and western hemlock of practically the same age. Although the Douglas fir may be 100 to 130 feet tall, the western red cedar and western hemlock understory is only 20 to 30 feet tall. But this understory is not suppressed beyond recovery; and as soon as the forest is opened the suppressed trees renew their growth and form the forest stand. Although there may be a number of veteran Douglas firs in a forest consisting mostly of western red cedar and western hemlock, there is no chance for the replacement of Douglas fir unless the en- tire forest is removed. On the other hand, it has been found that some stands which originally contained only about 5 per cent of Douglas fir have been succeeded after a fire by young growth which contained as high as 50 per cent. Tf the enormous quantities of seed produced almost annually by the western red cedar and western hemlock had the same chances of succeeding as the Douglas fir, the entire forest would soon consist of these species. From the small seeds produced by the western Bul. 1200, U. S. Dept. of Agriculture. PEATE IDG =e sohiar Fig. 1.—A dense stand of Douglas fir young growth on an area where theslash was burned in the spring the first season after cutting. Fic. 2.—An excellent stand of young Douglas fir that followed one slash fire and was killed by a second slash fire. Thesecond burn was 5 years old at the time the photograph was taken, and no young growth has replaced the stand that was killed. eats . ‘punorssporq, a UL S001) Poos AJ SLTSNOC O18 O40) YSNOYITe ‘vole pouINqods OY UO POMOT[OJ SvY MOIS SUNOA ON “UOD'VY SVM OATOIC Of} OLOJoO SAVOA ZI] POUINGOASCM JUSTIA OY) YB COIL OY, “BULSSOT pOMOTLO] Vey} AY Seino oand Jo systsuoo 4joT Vw pueys PLATE X. Sa a lture. gricu Bul. 1209, U. S. Dept. of-A Bul. 1200, U. S. Dept. of Agriculture. PLATE Xl. Fic. 1.—A young Douglas fir bearing seed at 14 yearsofage. Douglas fir has been found producing seed when 9 years old, and effective seed crops are produced at 20 years of age. r § § r | eee i ig es oe bh: i Sf bo. Fig. 2.—A scattered stand of Douglas fir seed trees left after logging. These have not been effective in restocking the area. Bul. 1200, U. S. Dept. of Agriculture. PLATE X11. Fig. 1.—Old timber in the background; 4 to 30 years old in the middle distance; 1 to 10 years old in foreground. Seedlings are at maximum distance of 5 chains from mature timber and 2 chains irom 30-year-old class. as fir 12 to 25 years old—the sta d seedlings are coming in amon Gc > c > e before complete forest stand; the trees. an ee nee ene one ee NATURAL REGENERATION OF DOUGLAS FIR. a 4] red cedar and western hemlock very shallow rooted seedlings spring up during the first season. keeping these species from any but favorable, moist sites. This factor alone is very important in Seedlings of these species must have two or three favorable seasons in order to become established sufficiently to insure them against drought. COMPETITION. In general, Douglas fir is not fastidious in choice of soil, if certain chemical and physical extremes are excepted, such as an abundance of common salt, lime, or water,-and if it does not have too much competition. Under favorable conditions Douglas fir adapts itself to a variety of soils, and its occurrence depends more on the number and species of its competitors than on the quality of the soil. Douglas fir seedlings have a number of strong competitors, and usually the competing species have the advantage in endurance of shade or resistance to drought.1* The common and most important competitors are listed in Table 18. TABLE 18.—Important species in the understory of the Douglas-fir forest. Scientific name. ACE TICINCIN ALUM UTS Wage ps eats ees ete ere ayaa UN CETPEL Oru, OTE rage ee ee oN Ran ee Ue ipa AAceH MackophyliiumyPurshis a eee es se eee eee Alnus oregona Nutt.........-..- SSN TOR I oat Apocynum androsaemifolium L...........--..----..-- Aretostaphylosamanzanita Party... 22 042.2. 55.2522 222: Arctostaphylos uva-ursi (L.) Spreng .-.....----+---+--- Castanopsis chrysophylla (Dougl.) A.DC....... Bae Ge Ceanothus cuneatus (Hook.) Nutt..................--.- Ceanothus integerrimus Hook. & Arm............--.--- Ceanothus prostratuses empl ies Ie ss Omicini Ceanothus sanguineus Pursh.............2....1.....2- Geanothus-velutinus Dougle 4 4s .5- 3 segs eee ae Chamaenerion angustifolium (L.) Scop...........--.-- Commus) miata Adee eos oa Ce Ys OS a8 once Eriodictyon californicum (Hook. & Arn.) Torr......-- Gaultheria shallon! Prumshh yes. 2 ates ee ea Minnaeagamenicanayh OnbeShs 2240422 sae ee eee Miyricaycahitonicay Chant eee he tigisgee 2i se ee AEE Odostemon aquifolium (Pursh) Rydb................. Philadelphus gordonianus Lindl Ropulussrchocarpasetooks y) 48 oe. a nein Prunus emarginata (Dougl.) Walp !......2...0.0..522- Pteridium aquilinum pubescens Underw.............. Quercus breweri Engelm Quercus californicai@lorn)iCoope:-- 544. 2 esa. Quercusichuysolepispluiebmn ys ogee ee a ea tea Quercus densiflora ieiooke <7 Army ihe oy QuercusigarnyanaeDousil ee ee ir vars es Oa eas Quercus sadleriana R. Brown Campst...-...........-- Rhododendron californicum Hook............-..-.---- FVUDUS DAbVALOLUSWNU bE ass 2 oe) ao) a le iad Ribusspectabilisybursh eee sa. - fas eh I es SS Salix spp Sambucus callicarpa Greenes. 2. sa2-/.ssce eee boo peal glauca N wt sich APO Bi UND ay Re Pe TDM OMLeATpOsSeMOllisMNULL: 2452 eS ees ea ey Thuja DlicatarwD ons eA VAT EEE ts Bee Tsuga heterophylla (Raf.) Sargent.........-........-. Umbellularia californica (Hook. & Arn.) Nutt.......-- Vaccinium membranaceum Dougl Vaccimum) ovalifoliuim: J. H; Smiths. 2c seco see a See Figure 1. 5, southwestern Oregon. Region in nee it orms a Common name. competing ground cover.a Minermaplegec’ <2 ght Oia oa ea: 12535405. Diwierd maples. aces Oey i ae eee 125354, 5 Broadleaf maple.............-.---- 1, 2,3, 4, 5. MEV neal Cl rete Sea Oe oP TEN aera Sees 1, 2,3, 4, 5. Wosbanep Hwy hee ae eae Nor ees 1, 2,3, 4, 5. INT ZA TAG ayy uel ee ey ete ames clin a of Kinnikinnicke 1 witewes. cI. 34g Goldenleaf chinquapin........-... 1,3, 5. Wedge-leaf ceanothus.........-.-. el Owos Pl we brush sat is espe Wee Goer e. 125.3, 4:0 Mein allasim atseesc ieee sais pene IG Sie Red ceanothuspee4y: 3. ass kee 1, 2,3, 4, 5. Sel eke. CeanObaussee em ees 1, 2,3, 4, 5. ITO WCC Ge a ENN A 1, 2,3, 4, 5. Western dogwood........----...-- 13 NICE WAN SIG Aerts ooze aati sy tah Sune Sub: SEEN Me BAA a Be oe nt ee re Sate he eh a 132, 3; 4515. White hawkweed..........-....-- 1; 2,3; 4,5. American twin-flower.........---- 1, 2,3, 4,5. California bayberry..........----- 1, 2,3, 4, 5. Mall Oregon erapesss 22h 1 2535405. SRS UaVEt Ey Mae eee wis Ue Pe wal Re ree pe aay ys Black cottomwoodsn.. eke sere APY Cie) Wall cCherisy Or (Su) bo 13 Hofmann. J. V. A Forest Saved is a Forest Grown. West Coast Lumberman, vol. 39, No. 465, Feb. 15. 1921. a NATURAL REGENERATION OF DOUGLAS FIR. D1 risk from early spring to late fall, but in the open area the danger period is confined to a time in the spring before the new vegetation covers the ground and to a period in the fall after the vegetation has dried up or has been killed by frost. The inflammability of any material is in direct relation to the amount of moisture it contains, and the amount of moisture in the material is largely regulated by the condition of the air. Evapora- tion depends pr rimarily upon the relative humidity of the air. Wind velocity, temperature, vapor pressure, and barometric pressure are influencing but secondary factors. The fire risk may be about equal in a young Douglas-fir stand with snags and débris and in an open stand with ‘dry grass or weeds, so far as starting a fire is concerned, although the rate of spread and difficulty of control make the former a much ereater haz- ard. The snags and débris not only spread the fire rapidly, but also develop such a high degree of heat that all the young growth is quickly killed and dried to the point of inflammability. The snags also form a dangerous risk throughout the season. In an open Doug- las-fir stand a fire will spread rapidly if fanned by a wind, but the surrounding air is not heated so much, and consequently only a part of the young growth is killed and little of it is burned. The season of high risk is shorter, because it is confined to the periods when the grass and weeds are dry. Some of the overmature Douglas fir stands contain a large percentage of dead and decaying trees, which are a detriment to the stand from a commercial standpoint and in addition make a dangerous fire hazard. During the dry season _shags are easily ignited and burn so furiously that they spread fire rapidly and cause the flames to climb into the crowns of nearby trees. In a mature stand free of snags a crown fire is not so likely to develop from a surface fire as in a stand containing snags, and the fire is easier to control because*it is on the surface only. The needles, being the most inflammable, constitute the principal fuel for the spread of a crown fire. Douglas fir needles will ignite when they contain moisture equal to about 35 per cent of their dry weight. Consequently, any amount of moisture greater than 35 per cent must be evaporated before the needles will burn. They ignite at_a temperature of about 650° F. and burn quickly. The water content of Douglas fir needles varies greatly through the season and is, no doubt, the underlying factor in the behavior of crown fires in mature forests and in the spread of fire in young stands. During the season of 1921 the percentage of water content of needles in mature Douglas fir was as follows: 95.9 per cent in early May, 77.3 per cent in early July, and 100.4 per cent in late September. Coincidently, the respective percentages of water con- tent of the needles of well-stocked young growth 13 years old_were 120.8, 99.0, and 115.3. This means that a fire, in order to burn Doug- las fir needles and to spread, must evaporate the following amounts of water, based on the dry weight of the needles: Q 2 Early May. Early July. pase oes ff0 Bh oaks RAB Wee MLE) ogl) 0\e) ete > Snipa ee ea tap Ue TREE Se OPT Oe Vel en 65.4 In youngicromeh A058 8 Es) ee a Bae 3 Per cent. Per cent. Per cent. 60.9 42.3 85. 8 64.0 80. D2 BULLETIN 1200, U. S. DEPARTMENT OF AGRICULTURE. From these figures it is readily seen that a crown fire will spread most rapidly in summer in either mature timber or young growth, - and that young growth in this respect is not so great a fire hazard as mature timber. Young growth, during the driest season, is not as inflammable as the mature timber in the spring. The fire danger in a clean mature forest as compared with that in a young stand of Douglas fir depends upon the inflammability of the needles, which in turn is in direct relation to the amount of water contained in them. A green stand of young or mature Doug- las fir will not burn until the needles are dried to the point of in- flammability. The rate at which the needles are dried determines the rate of spread of the fire, and when the heat ahead of the fire is not intense enough to dry the needles a crown fire stops. For the same reason a crown fire will start only where there is enough in- flammable débris beneath the stand to heat the air and dry the needles before it. For these reasons a slash area adjacent to a green stand either of young growth or mature timber always constitutes a dangerous risk. (Pl. XVII.) - Experiments show that in a mature stand of Douglas fir with 32 trees to the acre, 1.500 gallons of water per acre are evaporated from the needles before the inflammability point is reached: in a stand of young growth 12 years old with 20,000 trees per acre, 2.300 gallons per acre are evaporated; and in a stand 12 years old containing 8.000 trees per acre, 3.900 gallons per acre are evaporated. Of the two types of young stands the well stocked are less inflammable than the very densely stocked. The difference in water content of the needles in different stands is due to the rate of growth of the tree and the proportion of new needles to old. Mature stands grow very slowly and retain the needles in whole or in part for 6 or 7 years, with the result that the proportion of new needles to the older ones is small. Older needles contain more resinous deposits and mineral matter and less moisture than new ones and are, therefore, more in- flammable. In a dense young stand the top shoots grow rapidly, but the side branches are crowded and make little growth. This leaves a high proportion of new needles, but the total needle growth is less than in a young stand in which each tree has growing space enough to allow all of the side branches to develop new needles. Dead twigs on the stems are very inflammable and will carry a fire through a dense young stand which would not burn if only green twigs and needles were present. Clean, weil-stocked young stands are the lowest fire risk. The de- gree of hazard in these stands depends upon the density of the stand; the denser the stand the greater the inflammability. It is an en- couraging fact in the management of Douglas fir that the most de- sirable forest—the well-stocked young stand—is at the same time the lowest fire risk. FIRE PROTECTION. The greatest factor in forest protection is the prevention of fire. An analysis of the fire risks is the greatest aid in prevention, because it makes possible the most effective use of the protective force. Fire protection may be effectively accomplished through the efforts of individual owners, although it almost invariably proves to be ex- pensive. At present a large degree of protection is afforded in the : ‘pues UIVU 914 JO Spo 10]NO oy} SUOTL OPM 709] YOT OF GL S901} JO OSUTIJ MOIIVU B ATWO poy oly oy, “WOronposdos ay svpsnoqd Ude1s vsuep Aq poddojs useq svy oY YSIS V ao PLATE XVII. a Oo tS bp J a if Kes t 3 " i: My 9 ry ' rs iy BAIR ibe o ‘ ; " ad > . ery f j a ir Rt 4 ad By 4 . ae tke ists Ae Wer aed wie veh, nob Fcc pike WULenolennae,, ft biurltls, RK RE se “Ho Frater: (fe EEL Wi 3 gitar b: hotcanal 44 aid house at stl p a aay OR Ia R: eas Ba aah perp 2. oe bly ugha Ab: ‘Lea Seah {ih boat Swen, aie eld? Be. meidclde) Teply: ain dei saad? dann tho, od: Ligoserepotge di Deets id 08 otiaimoe 7 “ HOt the ae Fig Tee foods rete BOY. hase chi emis fiona 28. SLSR Dio ae we a thal ad bbyaila a1om aoe baand wd, Clodsrtaxortctite eedloole ie ailt “PeLBe tile Oh Hokey: rp. aerieool. adh pe bornt id. od Disodhe.. eet k TH 4 ae aaa ie ded tema ease -gomboedead slates: cot eaante ie borane gly. roresl dacle, to’ pitieaud | ‘Rana + cea Aude HTS wt 2041 mee tithes: ba uruid: 200 desks anh: ‘etd Lay iperk: libgnita wi rh, Pines, a | sot te 1) ere ee hela, aes Tt ones ae igen Lae “Bites co The saat “ov [piterorea ap MOLTO. oaths. aah Hi Pid mee CECy, vio ati eb: e808 katt bar Pisratosh) sea Lg. eta nid ti it ie iether tian Deke) ssg09 ately. nh ooittbbis Rites ail, or eante ube Ra RA fit tae ah G ; : feiss it ate is ¥ EUS Shir ts phad ot ed lingotl ban b aT ee gener: 46 ebaaieial sie dy bs Myra ch (bas me soc Bhyisqeh pial inane: sont wriehiens| hailiades owiel : Bolo ofl cliche utd dd (aiee euith aoe tholepe pathess Thcagen” Tater | abe f beset ol erapn iuctaees oviaesndne bea 98 a) fe | inte ae cae VEG " ctecetto al iatasoa: IRV Aor Banh: “eileen “sft enks ake Very dos dud dasa a: we WL. Lert : Jnthy ei bon Breet fifoll been. game aN oe Pediat eiear “ATG serie. ‘bert SAO oes tele cit ee Ce oe TL1 sob cian art me ed: Tere: ENE pice: deotercerl % creer aie del farootokd DenlguoGesrlgsr oben yrolembmal odes dad pein llmrarcsy ea eit pence aeyoka ale ‘ehnatoniae yy? to saan Ble to Jasiol sina. & of ash, oldnie « motwollat: vliacd bee . Sse? arian ait PARE oe ‘githioo ber lenle TO. etiamd Baye : tit st aiorad They alt ie batata ade Rag ie sa. OE yo Toe, fet La eS uc nh Seay ie a ata sae Pu f r Q ie Dees ON APPENDIX A. METHODS OF STUDY. AREAS STUDIED. The characteristics and the reproduction of Douglas fir have been studied by the Wind River Forest Experiment Station staff for the past 10 years. Field studies have been made, and laboratory results have been verified throughout the Douglas fir region of Oregon and Washington. Not only has a general survey been made of most of the large forest burns and a number of smaller burns in the Douglas-fir region of Oregon and Wash- ington, but also a close study of about 875,000 acres located in various parts of the region, including large and small burns has been made. The effects of one forest fire and one or more reburns on the same area were analyzed as well as the effects of season and methods of slash burning and of one or more slash fires on the Same area. THE TRANSECT. In the field studies the belt transect, 83 feet wide, was generally used. On these transects the age, species, and condition of each seedling found were noted. The rod-square plot was also used to a large extent both for permanent and temporary records. For an intensive study of an area, belt transects were run 24 chains apart over the entire area, and in this way an actual examination of 5 per cent of the total area was made. For an extensive study two transects were run 24 chains apart over the parts of the area that obviously afforded the best average conditions. A total of 1,223 temporary and permanent rod-square plots and 109.9 miles of belt transect 8: feet wide formed the basis for the conclusions reached through field studies. INSTRUMENTS. All records of temperature in fires and in the heat treatment of seeds or cones were taken with the Leeds and Northrup portable potentiometer. In the field the iron-constantin lead wire was used and the iron-constantin and chromel- alumel thermocouples. The chromel-alumel and ecopper-constantin thermo- couples were used in the laboratory, but no lead wires were required. The potentiometer, thermocouples, and wire were tested and standardized by the Bureau of Standards at Washington, D. C. The copper-constantin thermo- couples were used for some of the readings in tests of cones and inflammability. For the inflammability tests the same inflammability apparatus was used that had been used in the tests at the Forest Products Laboratory of the Forest Service at Madison, Wisconsin. Standard thermometers were used for soil and air temperatures. Evaporation records were taken with the Forest Service Evaporimeter and by the open tank method. For the laboratory tests of heat resistance of seeds in cones, the inflammability point of needles and twigs, and the heat resistance of bark, the apparatus was arranged as shown in Plate XVIII. - 57 APPENDIX B. BOTANICAL CHARACTERISTICS. Only those characteristics of Douglas fir are considered that affect the distribution and regeneration of the Douglas-fir forest. FLOWERS. The pistillate flowers are borne at the ends of the branches, just’ back of the terminal buds, on distinct stalks one-eighth to one-fourth inch long. :The - staminate flowers are borne on the preceding season’s growth, farther back than the pistillate flowers, on stalks one-eighth to one-fourth inch Jong. They usually mature before the pistillate flowers. The Douglas fir is polli- nated entirely by wind, and this development and arrangement of flowers decreases the chance for self-pollination. (Pl. XIX, fig. 1.) The flowers are monoecious and diclinous. side FRUIT. The cones are pendent on stalks one-fourth to three-fourths inch long and about one-eighth inch in diameter. The cones are generally 14 to 5 inches long and about 24 to 3 inches in diameter. They are about 1% inches in diameter near the base, with a gradual taper to the rounded point. The trident bracts extend from one-fourth to one-half inch beyond the scales, with the center point about one-fourth inch longer than the side points. Two winged seeds are borne under each scale. (Pl. XIX, fig. 2.) The cones mature in one season and are ripe in late August or early September. The seeds are shed within a few weeks after maturity, but some good seeds may remain in the cones for more than a year. Seeds that do not contain endo- sperm, and are consequently infertile, are often not shed until winter, and when they leave the cones they may, because of their lightness, be widely distributed over the snow by wind. The average number of seeds per pound is about 38,000. The cones do not open appreciably until they lose about 35 per ceut of their green weight in moisture. They open best when 40 to 50 per cent has been evaporated.t When the cones are exposed to intense heat, they do not open as well as when evaporation occurs more gradually. Sometimes cones that have passed through fire remain closed. BUDS. The buds are dark russet-brown when mature. Greenish buds are subject to frost injury. LEAVES. The leaves are three-fourths to 2 inches long. They are grayish green beneath and dark green on the upper side. On older branches they are slightly flattened with rounded points; on seedlings they are slender with acute points; and on terminal shoots the leaves are flattened with short points. (PI. XG fig: 1.) The leaves are persistent for several years, some remaining for seven years in the open. and older stands, although very few remain more than three’ years in the thrifty young or dense stands. The cotyledons or seed leaves are linear with a tapering point. The number varies from six to nine. (PI. XX, fig. 2.) 1 Willis, C. P. Incidental Results of a Study of Douglas Fir Seed in the Pacific North- west. Journal of Forestry, vol. 15,, No. 8, P. 991. 1917. 58 NATURAL REGENERATION OF DOUGLAS FIR. 59 FORM. The crowns of open-grown trees with good growing space are pyramidal and become broadly rounded or flat-topped in old age. In forest stands the boles are usually clear of branches for more than one-half the height of the trees. The clearing of side branches begins at an early age in complete stands, but in open-grown trees these branches are persistent and become large. The usual diameter of mature trees is 3 to 4 feet, and the height is 175 to 250 feet. Trees even larger than this are common. The largest authentic diameter recorded is 18 feet at 6 feet above ground, and the greatest height is 325 feet. BARK, On trees 10 to 15 years old the bark is about one-fourth inch thick, smooth, lustrous, and grayish to brown. It becomes ridged on trees about 30 years of age, and on old trees it grows to be 8 to 14 inches thick, dividing into large, rounded, irregular connected ridges. Some old trees have bark 18 to 24 inches thick at the base. The bark scales off and forms mounds around the bases of the trees. These mounds make good insulation against fire. ween. The wood is light red or yellow, with heavy white Ls ol The seedlings become lignified when 30 to 40 days old and at that age begin to be resistant to “ damping-off ” disease. : LONGEVITY. ) piace Stands 300 to 450 years old are common, and some stands older still have been noted in the region studied. The best development of the tree is during the first 200 years. The oldest tree noted was 739 years of age. BIBLIOGRAPHY. Big fir trees of the Northwest. Sci. Amer., N. Y., vol. 102; p. 323, April 16, 1910. Boycs, J. S. Decay in Douglas fir. The Timberman, Vol. XXIII, No. 1, p. 129, 133. . Nov.,, 1921. Fae eo hae _ Douglas fir pitch moth, U. S. Dept. of Agriculture Bulletin : 5. CHAMBERLAIN, WILLARD JOSEPH. Bark beetles infesting the Douglas fir. Ore- gon Agricultural Experiment Station Bulletin 147, 1918. CLINE, McGArvey. Properties and uses of Douglas fir. U. S. Department of © Agriculture, Forest Service Bulletin 88. 1911, CooLtey, R. A. The Douglas spruce cone moth. Montana Agricultural Ex- periment Station, Bozeman, Bulletin 70, pp. 125-180. 1907. DETWILER, SAMUEL B. Douglas fir: Identification and characteristics. Ameri- can Forestry, Washington, D. C., vol. 22, No. 266, pp. 67-69, Feb. 1916. DOUGLAS FIR DISCOVERED one hundred and twenty-one years ago. West Coast Lum., Tacoma, Wash., vol. 25, No. 293, pp. 34-37, Dec. 15, 1913. EVENDEN, J. C. Spruce budworms in northern Idaho. Timberman, Feb., 1923. FELLING OREGON Firs. Pacific Coast Wood and Iron, vol. 35. p. 104, March, 1901. FROTHINGHAM, E. H. Douglas fir: A study of the Pacific coast and Rocky Mountain forms. U. 8S. Department of Agriculture, Forest Service Circular 150: 1909. Frye, T. C. Height and dominance of the Douglas fir. For. Quar., Wash- ington, D. C., vol. 8, pp. 465-470, Dec., 1910. Gippons, WILLIAM H. Logging in the Douglas-fir region. U. S. Dept. of Agriculture Bulletin 711. 1918. Rev. in Journal of For., vol. 17, No. 2, p. 180, Feb., 1919. GovuLtp, C. W. Commercial woods of the Pacific Coast (including the annual production of species, estimated stands, uses, properties, and physical characteristics of Douglas fir, Sitka spruce, western red cedar, red- wood, western hemlock, western white pine, Idaho white pine, Cali- fornia white and sugar pines, incense cedar, and Montana larch). The Timberman, vol. 21, pp. 3440, 63-68. Feb., 1920. GRAVES, HENRY S. The Douglas spruce for northern Oregon: The Forester, Washington, D. C., vol. 5, pp. 52-57, March, 1899. GrirFIn, A. A. Infiuence of forests upon the melting of snow in the Cascade Range. Monthly Weather Review, vol. 46, pp. 324-7. July, 1918. GROWTH OF DoucrAs, Fir. Pioneer West. Lumb., San Francisco, vol. 62, No. 10, p. 32, Nov. 15, 1914. HANZzLIK, E. J., AND Futter, F. 8.