. Circular No. 97 .. .. , V May 1950 SURVIVAL and GROWTH of PLANTED SLASH and LONGLEAF PINES Results of 12-Year Experiment with Effects of Fire and Site Preparation on Plantings of the Two Species in the Gulf Coast Region of Alabama I AGRICULTURAL EXPERIMENT STATION jt'e ALABAMA POLYTECHNIC INSTITUTE M. J. Funchess, Director Auburn, Alabama CONTENTS Page REVIEW OF LITERATURE 3 DESCRIPTION OF EXPERIMENTAL AREA 4 4 6 EXPERIMENTAL DESIGN AND STATISTICAL ANALYSIS FIELD WORK DISCUSSION OF RESULTS EFFECTS OF FIRE ON SLASH PINE--------------------EFFECTS OF SITE PREPARATION ON SLASH PINE EFFECTS OF FIRE ON LONGLEAF PINE 6 6 10 11 13i EFFECTS OF SITE PREPARATION ON LONGLEAF PINE --SUMMARY AND CONCLUSIONS 13 15 LITERATURE ------------- FIRST PRINTING 5M SURVIVAL and GROWTH of PLANTED SLASH and LONGLEAF PINES Results of 12-Year Experiment with Effects of Fire and Site Preparation on Plantings of the Two Species in the Gulf Coast Region of Alabama G. I. GARIN, Associate Forester K. W. LIVINGSTON, Assistant Forester MANY INTERRELATED FACTORS are responsible for the failure of southern forests to grow as many products as is believed possible under good forestry practices. One of these contributing factors is the lack of information on basic phases of forest management. An experiment with two common but commercially important species of southern pines, slash pine, Pinus caribaea, Morelet, and longleaf pine, Pinus palustris, Miller, was begun in 1986 at the Gulf Coast Substation of the Alabama Agricultural Experiment Station. The purpose of the study was to determine the effects of burning and of site preparation by furrowing on the survival and growth of the two species in plantations. REVIEW OF LITERATURE In his monograph, "Longleaf Pine," W. G. Wahlenberg (11) discusses in detail the problems of longleaf pine regeneration and the role of fire in establishment of stands of this species. Judicious use of fire is considered helpful and at times even necessary for successful culture of longleaf pine (2, 3). Longleaf pine is considered to have inherent adaptations that make it fire hardy. Fire prevents smothering of seedlings by dead grass, and it will retard the spread of the brown spot disease (8). Osborne and Harper (7) point out that disturbance of the sod by site preparation increases survival of seedlings. The reduction of competition from grass and weeds is beneficial for initial establishment. After seedlings emerge from the grass stage, application of fire is no longer necessary, and longleaf pine saplings will thrive *Valued assistance and cooperation of Otto Brown, superintendent of the Gulf Coast Substation, is acknowledged by the authors. [1] under the condition of fire exclusion. Later in the life of a stand, moderate controlled burning practiced to reduce wild-fire hazard or for other reasons causes no appreciable injury (3, 5, 6, 9). A comprehensive treatise on slash pine is not available, and less is known of the behavior of this species. It is recognized that slash pine is not a fire-hardy species. Fire has a lethal effect on slash pine seedlings, and it will frequently kill the saplings. Site preparation may be of some benefit in early survival and growth, but its practice is relatively unimportant in slash pine plantations because of the rapid growth of the seedlings during their first few years (7). DESCRIPTION OF EXPERIMENTAL AREA The 20 acres on which this experiment was conducted is a part of the Gulf Coast Substation near Fairhope, Alabama. The site's terrain is rolling, with small drainages grading into ravines. The soils, developed from unconsolidated sands and clays, are of the Norfolk family. Topsoils are mostly sandy loams, and the subsoils are very deep. Under natural conditions, a thick grass cover, or rough, will develop within a few years. A longleaf pine stand had existed on the experimental area for many years. It had been worked for turpentine in the past, and had been repeatedly cut and burned. Records show that all trees suitable for pulpwood were cut in 1932. While a number of longleaf pine saplings were left, they were much too scattered to be considered a stand. EXPERIMENTAL DESIGN AND STATISTICAL ANALYSIS The three types of planting used were pure slash pine, pure longleaf pine, and a mixture of the two species in alternate double rows. Three replications of each type of planting were established on nine main plots. Each row of three plots extending north and south constituted a block. (See Figure 1.) For the application of fire treatments, each block was divided into three north-south strips of equal width. One strip was burned annually for 3 years; another one was burned once prior to planting; and the third one was completely protected from fire. Within each block, the fire treatments to strips were assigned at random. A further subdivision of plots was made by establishing eastwest lines that divided each tier of plots into halves. Furrows, in which the trees were to be planted, were plowed 8 feet apart [4] N -I-~ I~ Burned prior to planting and in the two succeeding winters Burned once prior to planting FIGURE 1. Diagram showing layout of experimental plots. Site prepared by furrowing on half of each tier. On the other half, trees were planted without site preparation. All of the 18 possible combinations of treatments were present in each block. Three replicate blocks had a total of 54 sub-plots. These sub-plots were one-third of an acre in size, and 208 seedlings were planted on each. A modification of the method of Cochran and Cox (4) for analysis of variance when sub-unit treatments are applied in strips was employed for testing each of the several sets of data. All figures that were expressed as percentages were first transformed into corresponding angles by the use of Bliss' (1) tables. The plots planted to a mixture of the two species of pines were treated in the first analyses as if planted to a third species. Since it was evident that mixed planting had not affected the response of the individual species to sub-plot treatments in the early growth of the plantation, a second series of analyses was made. In this series, each species on the plots of mixed slash and longleaf pine was treated as if it were on a separate plot. This, in effect, doubled the number of replications for comparing slash pine with longleaf pine and for testing their individual reactions to fire and cultural treatments. To place all data on an equal basis, [51 those for the original plots with pure plantings were reduced by one-half for these analyses. FIELD WORK The entire experimental area was cleared. All hardwoods and small pine poles and saplings left from the 1932 cutting were cut close to the ground. In December 1936, fire lanes 12 feet wide were constructed, dividing the experimental area into nine strips. The first burnings were done in January, 1937, and seedlings were planted that same month, using a spacing of 8 by 8 feet. Since in the judgment of those then in charge of the experiment it was desirable to establish as uniform a stand as possible, all seedlings found to be dead or missing were replaced the following January. The strips designated for annual fires were reburned in the winter of 1938 and again in 1939. Thereafter, shortage of personnel and restricted transportation during World War II prevented continuance of the work. Hence, these strips actually received a total of three burns, at one year intervals, beginning just before the seedlings were planted. The first field measurements were made in October, 1937, when all living seedlings were tallied. Survival counts were repeated in October, 1938, and in November, 1939. In 1939 and 1943, samples were taken to estimate the proportion of longleaf pines that had emerged from the grass stage. In the summer of 1948, the diameters at breast height of all trees more than 41/2 feet tall were measured. Samples of heights of both species were also taken. One dominant tree was selected at random from each row, and its height was measured. Longleaf pines still in the grass stage were not included in average diameter or height estimates. They were included in the survival count of trees. A relatively small number of such trees was present in 1948. DIscussION OF RESULTS EFFECTS OF FIRE ON SLASH PINE. Before the seedlings were planted in 1937, two strips in each block were burned as previously outlined. On these strips first-year survival of slash pine seedlings was significantly lower than on the remaining strips where grass was left unburned. Burning took place in winter when the grass was dry. The roots of the grasses were little affected by the burning of the dry stems above the ground. It has [6] been observed that winter burning of a rough is frequently followed by a vigorous summer growth of the grass. Root competition from this vigorous growth of new grass and the absence of tall, dry grass stems, which would have offered some shade to the small seedlings, may account for the high mortality of pine seedlings on areas subjected to fire just before planting. In January 1938, all dead seedlings were replaced. After two growing seasons on strips not burned after planting, the number of living slash pine seedlings was slightly lower than it had been on the same strips the year previous. Mortality of the seedlings in their second year, those planted in 1937, added to the loss of the seedlings replanted in 1938 was high enough to produce this slight reduction in survival. During the following year, a noticeable reduction in the number of living slash pine seedlings occurred (Table 1). When these strips were examined in 1948, TABLE 1. SURVIVAL REPLANTED IN OF SLASH AND LONGLEAF PINE TREES PLANTED IN 1987 AND 1988, GULF COAST SUBSTATION' Year Species Type of planting Pure Mixed Both Pure Mixed Both Pure Mixed Both Pure Mixed Both Pure Mixed Both Pure Mixed Both Pure Mixed Both Pure Mixed Both Protected from fire Pct. 86.9 86.1 86.6 64.4 61.2 63.3 85.7 80.4 83.9 67.1 75.1 69.8 77.9 73.2 76.3 70.02 71.8 70.6 74.6 75.2 74.8 43.2 28.0 38.1 Burned Burned Sod Sod once for three undis- furrowed before successive turbed in planting winters strips Pct. 85.5 81.5 84.2 55.5 52.9 54.6 85.4 79.3 83.4 68.3 68.0 68.2 80.2 68.7 76.4 68.9 69.0 68.9 79.4 67.4 75.4 34.3 25.2 31.3 Pct. 85.1 79.6 83.3 58:2 61.7 59.4 27.6 26.1 27.1 64.4 62.6 63.8 8.2 9.5 8.6 69.2 59.6 60.0 3.4 2.5 3.1 57.2 55.2 56.5 Pct. 85.0 82.2 84.1 51.8 49.6 51.1 62.0 57.6 60.5 63.4 62.5 63.1 52.7 47.5 51.0 63.3 62.6 63.1 51.9 49.3 51.0 44.6 35.9 41.7 Pct. 86.6 82.5 85.2 66.9 67.5 67.1 70.4 66.2 69.0 69.7 74.6 71.3 58.1 53.4 56.5 69.3 71.0 69.9 52.9 47.3 51.0 45.1 36.3 42.2 Average all plots Pct. 85.8 82.4 84.7 59.4 58.6 59.1 66.2 61.9 64.8 66.6 68.6 67.3 55.4 50.5 53.8 66.3 66.8 66.5 52.4 48.3 51.0 44.9 36.1 42.0 1987 Slash Longleaf 1938 Slash Longleaf 1939 Slash Longleaf 1948 Slash Longleaf 1 Data from 54 sub-plots 0.3 acre in size with 208 seedlings planted 8 by 8 feet. 2 Apparently a few of the living seedlings were overlooked on several plots when the counts were made in 1938 and 1939. [7] slash piniwXXas presenlt inl numblers comlparale wXith thlose foiud in the fall of 19.39. Silicc theC seedlings5 replanted in 19.38 had been'1 thiroughl two grt~owXing sCasons1 Im the falI of 1939. it is cX i(ICIt that c Cf little nmortali tX occu red 2 x Cars after theC scedI liogs had beCCn traunsplan~ted. Thiis ini lcates that, althouigh slash pilc sC ed~linlgs con1tinuedC to (lie at at consp5icuoIu s ratC for the first ph ntiuig. tlhe 51rX iXal ini laterX I 5cr hadI~ beCn nIlaintaiiicd at a high IcX (1. The) sameD jatti'rI . X ith no si gi ilant differces (CS.Cistcdl on all protected strip~s, XXhether tl)cX XXCrC not 2 rearsI after burned or XXcIC holrned on1ce before plan tjing. Onl strips burne ill 19:38 and 19,39, the' mortal it, of slash pinC ld great, indicating that slash 1 )ine is rIot at fir-rsistai specis. OnkX small iinumb)rs of slash pineI rem1aine d in 19039. XX as XeCI 1Few ot these Xwere 1fo111d to bli ing in 1948, Purir 2. HeR- XXaS resp~onsilel for a IlighlI significant reductionl in the numbell)r of liX 11(4 slash p)inlC 5CCelihl(45 in the 2 X cars ilimIII- 1)CatCel 1)11ilig (liatclX following1( plitihng. ICsIrX iXat perOn strips protected from fires after planitinig. the seedC~lin gs after the first X car XXas (rcat('t cCI tag4c of slash p~in than that of the) lon~gleaf pinC sCedlinlgs. The dw(iffrece Xas ('C hi (4ly sig4 i [kat uit. Vs hCr- fire wC r c CIlue after pliantinlg, as XXC pie wer plne1natraed u l o s ftr1 e ro l n ls iei FliGUe t 2. ea burcned rosIins the lrgtre iscndyheaferud ogefpn is thriving at the extreme right and left. TABLE 2. GROWTH DATA OF SLASH AND LONGLEAF PINE TREES PLANTED IN 1937 AND REPLANTED IN 1938, GULF COAST SUBSTATION' Type of planting Percentages Protected from fire Burned once before planting Burned for three successive winters So Year Species So undisturbed Sod furrowed in strips Average all plots of longleaf pine seedlings emerging from grass stage 2.0 16.8 1.1 8.2 1.4 34.2 1939 1943 Lco Longleaf Longleaf SPure and Smixed 1.7 20.1 1.3 19.3 27.3 15.6 17.1 13.9 16.0 2.8 1.5 2.4 1.5 19.7 25.0 16.8 15.4 13.6 14.8 3.4 1.8 2.6 Average height 94 1948 of dominant trees in feet " Slash Longleaf 5Pure and Smixed 30.1 18.6 20.5 21.9 21.0 0.8 .8 .8 31.5 16.2 25.2 18.5 23.0 0.9 .8 .9 13.4 15.8 0.4 4.2 3.9 4.1 .4 .4 22.7 18.1 13.6 13.3 13.5 3.9 2.1 3.3 Average basal area in square feet 1948 1948 Slash Longleaf Pure Mixed Both Pure Mixed Both 'Data from 54 sub-plots one-third acre in size with 208 seedlings planted 8 by 8 feet. j' P FIGURE 3. Area completely protected from fire 12 years after slash and longleaf pines were planted in alternate double rows. In center two rows, a few longleof pines are living (left-center row). Slash pine is thriving at extreme right and left. wxas the case oni two strips in each b lock. th e si irx ix al o{ slash ( wXas (01 isistett greater for the (Ilratioli of' the (pe(ilinlt. i( It is appar ('lt that it xxas easier to e'stablishi a \\ ('1 stockedI slash plie stand( 1)\ planting than it xxas to establlisli a lonlcaf pik stand. prox ided fires xxere not allotted to xxipe out the slash pine seeclliligs. Fgnc3 Ini adi(ltion to killing~ iiost ofi the slash pine sedliligs. repfeated fires adlx rscl\ affected height groxxti i. Ini 1945, the di ff eren ce lbetwxeen ax (rage hecights of dominan t slash p~ine trees on bi rned and iinbiiriid strips xxas lhighix significant (Table 2_ The basal area o)f slash pine groinig on strip~s that xxere buniied for 2 si cccssix e c ars after plantin g xxas S cV\ sli oni the subtal all p)lots. This wxas a normal result of a x (IX lotsx suir ix al. TIhe differen ce in b~asal area of slash pine xxas inot signiiificanut 1 etxxeen the siub -plo ts on strips protected frotnfire and those 1 iniecd on1ce ( befoire phlli ig. p)in e I:r S ot. uit Pnrv uI'xn \TIOu ON SIAS PsIIINI. 't'hre xxas no0 (differelice in first-x ear sn rx ixal betxx ee the slash pine seelin gs planited ini furrowxs alil onies set ini so~d. This continiued to be the case fo r the 2 followin11g xears oni strip~s p~rotected Iron t rln - ['(ii ing. The fact that slash pine seedlings begin their height growth soon after planting might partly explain the absence of a significant response to site preparation. Slash pine seedlings planted in the rough showed the ability to withstand competition of grass. Planting of seedlings in furrows had a significant effect only on strips that were burned for 2 years after planting. On such strips, survival percentage of slash pine after the first year was significantly higher than on sub-plots where planting was done without site preparation. The reduction in volume of grass near the planted rows apparently moderated the lethal effect of fire and resulted in improved seedling survival. Slash pine planted in furrows had better height growth than that planted without the site preparation. In 1948, the difference between the average heights of dominant trees on the two groups of sub-plots was highly significant. The basal area of slash pine growing on sub-plots with seedlings planted in furrows was greater than that on the sub-plots without site preparation. This difference closely approached the significant value. EFFECTS OF FIRE ON LONGLEAF PINE. Longleaf pine seedlings planted on strips not burned before planting had a first-year survival significantly higher than had the ones on burned strips. Since the same was true of the slash pine seedlings on strips not burned, it should be noted that preparatory burning had a detrimental effect on the initial survival of planted seedlings of both species. The effect was more pronounced with respect to the longleaf pine seedlings. This effect should not be confused with that of burning for seedbed preparation to encourage natural regeneration. Such a practice is often desirable, since the resulting exposure of mineral soil is a necessity for a high rate of germination of longleaf pine seeds (7). A favorable seedbed increases germination of seeds very appreciably; and, even if a somewhat higher mortality of seedlings may result, there is a considerable net gain in the number of living seedlings on the areas where preparatory burning was used. In the fall of 1938, because of the replanting done earlier the same year, larger numbers of living longleaf pine seedlings were present than were found the previous year. During the next growing season, there was no decline in survival on the strips protected from fires. Slightly increased survival figures for some plots in 1939 are probably due to a miscount in 1988. During [ 11] the following 9 years, there were large losses on strips burned only once before planting and on strips not burned at all. Protection from fires was definitely detrimental to the survival of longleaf pine. In 1948, only about half as many longleaf pine trees were present on the strips protected from fires as there were in 1938, Figure 3. Replanting increased the number of living longleaf pine seedlings on strips burned in 1988 and 1939. In the fall of 1938, there was no significant difference in survival of-longleaf pine due to the fire. In 1939, due to a small decline in survival on burned strips, there was a significant difference in the number of living longleaf pines between burned and protected strips. This trend was reversed sometime during the following 9 years. In 1948 the stand of longleaf pine on strips that were burned in 1938 and 1939 showed a high survival in comparison with the protected strips, Figure 2. This difference in survival was highly significant, Table 1. In 1943, a sample was taken to estimate the percentage of living longleaf pine seedlings that had begun height growth. A significantly larger proportion had emerged from the grass stage on the strips burned twice after planting than on either of the other groups of strips (Table 2). Fires are considered to be beneficial to the longleaf pine seedlings in the grass stage (9). Such fires help to control the spread of brown spot disease. The volume of dry grass is reduced and seedlings are not smothered by it. Fire may result in an increase in the amount of certain nutrients in the surface layer of the soil, or even indirectly act as a stimulant in starting the height growth of young longleaf pines (10). The relatively high mortality rate of longleaf pine seedlings on strips protected from fires was perhaps due to smothering by grass and to a heavy toll by brown spot disease. In this experiment, the beneficial effects of fires on early survival and growth of longleaf pine have been marked. In 1948, no significant difference was observed between the heights of average dominant longleaf pines due to fire after planting. At the early ages, longleaf pine is considered to be a slower growing species than slash pine. In no small degree, this is due to the peculiar habit of longleaf pine seedlings to remain in the grass stage for several years. In this experiment, after 12 growing seasons, the height growth of the longleaf pine was significantly lower than that of the slash pine. [12] The basal area of longleaf pine on strips burned twice after planting was numerically almost five times as great as that on comparable strips protected from fire. Statistically, this difference was highly significant. EFFECTS OF SITE PREPARATION ON LONGLEAF PINE. Where sod was disturbed by furrowing, initial survival of longleaf pine seedlings was considerably higher than on sub-plots where seedlings were planted without site preparation. This difference was highly significant. Reducing competition from grass roots in furrows apparently proved to be an effective measure in bringing seedlings through their first growing season. This is in agreement with previous findings (7) that the survival of longleaf pine seedlings is favored by site preparation. As was the case in the first year, there was a highly significant increase in survival of longleaf pine in 1938 and 1939 planted in furrows over that on the strips without site preparation. By 1948, however, this difference had reduced itself to only a fraction of one per cent. When, in 1943, a sample was taken to estimate the percentage of living longleaf pine seedlings that had begun height growth, there was no difference in emergence from the grass stage between longleaf pine seedlings planted in furrows and those on the sub-plots without site preparation. Longleaf pine on sub-plots where seedlingswere planted in sod had a significantly greater height growth in 1948 than on sub-plots where they were planted in furrows. Site preparation at the time of planting had, after 12 growing seasons, opposite effects on the height growth of longleaf pine and slash pine. Site preparation at the time of planting did not affect significantly the basal area of longleaf pine. However, opposite to slash pine, longleaf pine on sub-plots with undisturbed sod had a greater basal area than did longleaf pine on sub-plots with seedlings planted in furrows. This opposite trend between the two species was significant. SUMMARY AND CONCLUSIONS An experiment to test the effects of fire and of the practice of planting in furrows on survival and growth of slash pine and longleaf pine planted in the Gulf Coast Region of Alabama was begun in 1936. Pine seedlings of these species were planted 8 by 8 feet on 54 sub-plots, one-third acre in size. A group of 18 sub-plots constituted a block. The experimental area was [13] first planted in January, 1937. In 1938, dead seedlings were replaced with new seedlings. One strip in each of the three blocks was protected from fires. Another strip was burned once before the seedlings were planted. The last strip was burned before planting, 1 year after planting, and again 2 years after planting. On one-half of the sub-plots, furrows were plowed in which seedlings were set. The ground was left undisturbed on the other, half. On strips that were burned before planting, first-year survival of slash pine seedlings was significantly lower than on those strips on which grass was left unburned. Slash pine seedlings on strips protected from fires had relatively high mortality for the first 2 years after planting. The mortality in later years was very small. On strips that were burned in 1938 and 1939, the mortality of slash pine was very great. Repeated burning practically wiped out the slash pine stands. In addition to killing most of the slash pine seedlings, repeated fires adversely affected height growth. On strips protected from fires, there was no difference in survival between slash pine seedlings planted in furrows and the ones set in sod. Seedlings planted in furrows showed improved survival only on strips that were burned for 2 years after planting. In 1948, slash pine on sub-plots with seedlings planted in furrows had better height growth and greater basal area than on the subplots where seedlings were planted in sod. It was found that after 12 years the slash pine had higher survival, greater height growth, and a larger basal area than did the longleaf pine, except on strips subjected to repeated fires. It can be concluded that it is easier to establish a well stocked slash pine stand by planting than it is to establish a longleaf pine stand, provided fires are kept out of the area. Preparatory burning decreased initial survival of longleaf pine. However, the difference between seedling survival on the areas burned once and on those not burned disappeared after the first year, and no significant difference was later found in height or basal area. Although application of fire in the second and third years decreased survival in those years, it had caused after 12 years a marked increase in survival over that on the strips protected from fires. In 1943, a larger proportion of longleaf pine seedlings had emerged from the grass stage on strips burned for 2 years after planting than on the other strips. The basal area of longleaf pine [14] was greatest on strips burned for 2 years after planting. The height of trees, however, was not significantly affected by fire. For the first 3 years, longleaf pine seedlings planted in furrows had a higher survival percentage than did those established without cultivation. In the 12th year, however, there was no difference in survival due to the site preparation. The height of longleaf pine trees planted in furrows was lower than that of those planted in sod. The basal area of longleaf pine trees planted in furrows and in sod showed a trend opposite to that of slash pine. It was adversely affected by site preparation at the time of planting. Planted longleaf pine seedlings were greatly benefited by burning in the first 2 years after planting. Good survival and growth resulted from this fire treatment. Longleaf pine should not only be recommended for planting ip preference to slash pine where there is a fire risk, but in its culture it requires judicious use offire as an aid to its better survival and growth in the early stages of stand establishment. LITERATURE (1) (2) (3) (4) (5) BLIss, C. I. Plant Protection, No. 12, Leningrad. 1937. CHAPMAN, H. H. Some further relations of fire to longleaf pine. Jour. Forestry 30: 602-604. 1932. . Effect of fire in preparation of seedbed for longleaf pine seedlings. Jour. Forestry 34: 852-854. 1936. COCHRAN, W. G. and Cox, G. M. Experimental design. John Wiley and Sons, New York. 1950. (6) (7) DEMMON, E. L. The silvicultural aspects of the forest-fire problem in the longleaf pine region. Jour. Forestry 33: 328-331. 1985. MACKINNEY, A. L. Some effects of three annual fires on growth of longleaf pine. Jour. Forestry 32: 879-881. 1984. OSBORNE, J. G., and HARPER, V. L. The effect of seedbed preparation on first-year establishment of lorigleaf and slash pine. Jour. Forestry 85: 68-68. 1937. SIGGERS, P. V. The brown spot needle blight of longleaf pine seedlings. Jour. Forestry 30: 579-593. 1932. SIMERLY, N. G. T. Controlled burning in longleaf pine second growth timber.. Jour. Forestry 34: 671-673. 1936. WAHLENBERG, W. G. Effect of fire and grazing on soil properties and the natural reproduction of longleaf pine. Jour. Forestry 33: 331-3338. 1935. (8) (9) (10) (11) . Longleaf pine. Charles Lathrop Pack Forestry Foundation. Washington, D. C. 1946. [15]