Abstract: Four clones of Populus were planted in replicated monoclonal plots near Olympia, Washington, to evaluate their suitability for use in short-rotation culture. All clones were easily established and had minimal problems from damaging agents during the first five years. Observed differences among clones in pattern and amount of growth appeared to be associated with differences in number and density of buds, sylleptic branching, and phenology. In addition, differences in drought tolerance and stockability may also have influenced clonal differences in annual growth and stand productivity. Individual tree growth was limited by the dense 1.0-m spacing, but the best-growing clones averaged 13 to 16 m tall, 7 to 9 cm in breast-high diameter (1.3 m), and produced stand basal areas exceeding 38 m2 ha-1 at 8 years. Mortality was negligible for 7 years, after which various combined effects of competition, stem borer damage (Cryptorhyncus lapathi), and a severe windstorm caused mortality ranging from 18 to 36% in the three fastest growing clones.

Keywords: Tree improvement, short rotation, intensive culture, biomass, productivity, stockability

Interest in short-rotation production of native cottonwoods and various Populus L. hybrids for fiber and energy has accelerated greatly during the past decade (Ranney and others 1987; Miner 1990; Heilman and others 1991). Several large forest products corporations have made or are considering substantial investments in poplar tree farms in the Pacific Northwest. Farmers and other land owners are assessing opportunities for producing and marketing cottonwood from their smaller acreage (Associated Forest Products Consultants, Inc. 1990). Productivity can be very high, provided that suitable clones are planted on proper sites and appropriate cultural techniques are applied. Many clones have been selected for planting based on performance in small evaluation plots, but few comparative data have been published on growth and yield in larger monoclonal plots where competition from neighboring clones had little or no influence on absolute or relative form, structure, and growth of the individual clones. This report describes survival, growth, stem characteristics, and stand development to age 8 of four Populus clones planted at 1.0-m spacing in monoclonal plots having sufficient size and borders to limit effects from trees growing in adjacent plots. To aid in interpreting clonal differences in stand productivity, detailed measurements were taken in some years to assess leaf area, bud populations, branching characteristics, and the phenology of height and diameter growth.

Application

Productivity was high and differed substantially among four Populus clones, ranging from 11 to 18 Mg ha-1yr-1. Findings of interest to managers of clonal plantations include:

• Clonal rankings in height, diameter, and stand productivity were firmly established by age 3,
• Stem form differed markedly among clones and was correlated with differences in bud and branch characteristics,
• Differences in tree growth and stand productivity among clones were much less in our large monoclonal plots than had been previously reported from smaller plots, and
• Clones appeared to differ in stockability (tolerance to crowding or competition). This implies that optimal management regimes may differ among clones with respect to one or more of the following: target diameter, stand density, or rotation age.

Methods

The study was established in cooperation with the Washington State Department of Natural Resources at the Meridian Seed Orchard located 12 km east (lat. 47º 0' N, long. 122º 45' W) of Olympia, Washington. The low-elevation site (50 m) has a mild climate with an average growing season of 190 frost-free days and a mean July temperature of 16º C. Annual precipitation is approximately 1290 mm, falling mostly as rain from October through May; summers are periodically dry. The soil is Nisqually loamy fine sand (a sandy, mixed, mesic Pachic Xerumbrept); it is a deep, somewhat excessively drained, medium acid (pH 5.6) soil formed in glacial outwash, and would not be considered suitable for Populus growth without irrigation. Nearby unmanaged land is occupied by either prairie vegetation or Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) and several hardwood shrubs, depending on fire history. The land is level and was previously farmed for strawberry and hay crops. It was prepared for planting by plowing and disking in winter 1985-86.

The study compared four Populus clones planted in monoclonal plots at 1.0-m square spacing; all four clones were under consideration for use in biomass plantations. Two clones were collected by Stettler and his colleagues (Weber and others 1985; Quinsey and others 1991) in wild stands of native black cottonwood (P. trichocarpa Torr. & Gray); clone 7-75 was collected at 91-m elevation along the Puyallup River near Orting, Washington, and clone 8-81 was obtained at 7-m elevation in the Nisqually River Delta which is near the study site. Early growth of the Orting and Nisqually clones was above average for black cottonwood clones in initial screening trials (Heilman and Stettler 1985). The other two clones were Populus hybrids; clone 11-11 was a P. trichocarpa × P. deltoides Bartr. ex Marsh. var. deltoides hybrid developed and tested by the University of Washington/Washington State University Poplar Research Program (Heilman and Stettler 1985; Quinsey and others 1991); clone D-01 was a Populus hybrid (taxonomic identity unknown, but suspected to be either P. trichocarpa × P. nigra L. or P. trichocarpa × P. angustifolia James) developed originally at the University of Idaho and subsequently selected from a Canadian planting by Dula's Nursery of Canby, Oregon (Dula 1984).

Each clone was tested in three plots of 100 measurement trees planted at 1.0-m × 1.0-m spacing, surrounded on all sides by at least four rows of similarly spaced trees. Plots of the two hybrid clones (11-11 and D-01) were randomly located within three blocks of a larger study that compared “woodgrass” (i.e., << 0.3-m spacing, coppiced annually) and wider- spaced systems (square spacings of 0.5-, 1.0-, and 2.0-m) for biomass production (DeBell and others 1993). Plots of the two native black cottonwood clones (7-75 and 8-81) were randomly positioned within a satellite block immediately adjacent to the larger study. An additional 20 cuttings of clone 11-11 also were planted in the satellite block. Conditions of the soil and other environmental characteristics, however, were very uniform among all blocks and plots.

The clones were established by hand planting unrooted, hardwood cuttings in late April 1986. All cuttings were 30-cm long and had a minimum upper diameter of 1 cm; they were planted 20-cm deep with at least two healthy axillary buds remaining above ground. Previous experience indicated establishment success (i.e., survival and early growth) was poor if cuttings did not have at least one healthy bud above ground (Radwan and others 1987). Requiring two above-ground buds resulted in a small additional increase in establishment success but also increased the prevalence of multiple stems. At the end of the first growing season, all multiple-stemmed plants were pruned to one stem and any positions occupied by dead trees were replanted with unrooted cuttings; this resulted in stands composed solely of single-stemmed trees and with all planting spots occupied.

Supplemental nutrients and water were provided uniformly in plots of all treatments to enhance survival and growth. A preplanting application of mixed or complete fertilizer (16-16-16) provided the equivalent of 100 kg ha-1 nitrogen, 43 kg ha-1 phosphorus, and 83 kg ha-1 potassium. Additional nitrogen fertilizer (ammonium nitrate) was applied at 100 kg N ha-1 in May 1988 and again in March 1992. Plots were irrigated throughout each summer by a drip system; amounts applied averaged 400-600 mm per growing season but varied from 260 to 1200 mm; yearly differences were associated more with scheduling conflicts and malfunctioning of irrigation pumps than with differences in potential evapotranspiration. All plots were kept free of weeds during the first year by tilling and hoeing; in the second and third year, sporadic weed patches were controlled by spot applications of herbicides and hoeing.

Tree survival, total height, and stem diameter were evaluated at the end of each of 8 growing seasons on the central 100 trees in each plot, and any tree damage or unusual conditions were noted. Height and diameter were also measured periodically on 25 trees per plot to assess seasonal growth patterns. Information on basal diameter (measured at stem height of 0.3m) was obtained for all years, and breast-high diameter (dbh) was also measured after mid-season in the second year. Figures presenting periodic growth utilize basal diameter as it is available for all periods. Cumulative basal area over time and diameter at age 8 are presented based on measurement at breast height (1.3 m) to facilitate comparisons with standard forestry measurements. Breast-high diameters at the end of the first year were estimated from height measurements and dbh-height relationships derived from data collected in June of the second year.

Indices for lower-stem taper were calculated from basal and breast-high diameters (basal/ dbh) and slenderness was characterized with the ratio of height/basal diameter. After the first, second, and third seasons, numbers of buds and branches were counted for each annual height increment. Such data provided information on initial bud populations, development of sylleptic and proleptic branches, and branch longevity. Observations were recorded on leaf phenology, insect and disease problems, and wind damage.

Yields for ages 1 through 5 were estimated from oven-dry biomass component equations applied to basal diameter and height measurements of individual trees. The equations were developed via destructive sampling of trees at the end of years 1 through 5; selected trees were from buffer rows and were representative of the range of sizes for each clone. Equations of the form, Ln(Y)=f(basal diameter, height, and age), were fit independently for each clone. Dry weights of stems and branches were estimated by separate equations. Coefficients of determination (R2) between predicted and measured values ranged from 0.906 for branch weight of clone 8-81 to 0.999 for stem weight of clone 7-75. Stem, branch, and leaf weights were estimated by separate equations; stem and branch weights were summed to provide above-ground dry woody biomass. Above-ground woody biomass estimates for all trees on each plot were summed, and the resulting plot sums were expanded to woody biomass yield per hectare. Destructive sampling of all clones did not occur after year 5; thus, stand growth parameters beyond age 5 are limited to height, diameter, and basal area.

Leaf area index was estimated by determining the projected area of a sub-sample of leaves from each biomass sample tree using an area meter (LICOR 3100). Leaf area per tree was then estimated via area/leaf weight relationships; it was expanded to a per unit area basis by summing estimates for all trees on the plot; total leaf area (m2) was then divided by ground surface area (m2) to provide the dimensionless leaf area index (LAI).

Annual diameter growth (per tree) and basal area growth (per hectare) were plotted against a competition index calculated by summing the product of basal diameter (m2) × height (m) (both measured at the beginning of the growth year) for all trees on the plot and expanding it to a per hectare value. This allowed comparison of clonal growth rates at similar levels of competition, even when the same competition level was reached in different years.

Statistical comparisons required special consideration because the four clones had been assigned to plots randomly as components of two adjacent experimental plantings rather than as one experiment. We first examined the data for differences in growth performance among blocks and concluded that there were none. Two approaches were used to arrive at that determination: (1) analyses of variance for height and diameter of clones 11-11 and D-01 indicated that block effects were non-significant for both height (p=0.66) and diameter (p=0.53); and (2) graphic comparisons revealed that early growth of 20 trees of clone 11-11 in the satellite block (composed primarily of replicated plots of clones 7-75 and 8-81) was essentially identical to mean growth of clone 11-11 in the other three blocks. Given the lack of evidence for environmental or growth differences among blocks, we examined mean tree and plot data via standard analysis of variance procedures assuming a completely randomized design (i.e., block was not considered as a source of variation). Where appropriate, clonal treatment means were compared by the Ryan-Einot- Gabriel-Welsch multiple-stage test (Ramsey 1978) (using p £ 0.05 to judge significance) and standard deviations were calculated. These data are displayed in tables or figures to illustrate trends in development of the clonal plantings.

Results and Discussion

Survival and growth during the establishment year

Cuttings of all four clones produced roots and shoots very well, and survival at the end of the first growing season and prior to replacement planting was 96% or higher. The two black cottonwood clones (7-75 and 8-81) averaged 2.4 m in height and 2.2 cm in basal diameter; hybrid clone 11-11 was 2.3-m tall and 2.2 cm in basal diameter, but hybrid clone D-01 was considerably shorter in height (1.7 m) and smaller in basal diameter (1.3 cm). Above-ground dry woody biomass averaged 2.9 Mg ha-1 for the two black cottonwood clones and hybrid 11-11; whereas, clone D-01 produced only 1.2 Mg ha-1.

Patterns of height and diameter growth

Annual increments of height and diameter growth are shown by clone in Figure 1 for all trees that survived through age 8. By the end of the second year, height and diameter of hybrid clone 11-11 were greater than either of the black cottonwood clones, which in turn were larger than hybrid clone D-01. Ranking by size thereafter remained similar through age 8 years, except that clone 7-75 was surpassed eventually by clone 8-81 in both height and diameter and was equaled by clone D-01 in diameter. Annual growth of all clones declined as trees became older and larger and inter-tree competition became more intense, but growth of the black cottonwood clones accelerated somewhat for diameter during the seventh year and for height in the eighth year, presumably in response to increased crown differentiation within the stands. Relative growth of the different clones differed among years and over time as follows. During the third through the seventh years, annual growth of hybrid clone D-01 was equal to or greater than either black cottonwood clone (7-75 or 8-81) in both height and diameter, and its growth during the fifth year (1990) was slightly greater than that of clone 11-11. Diameter growth of clone 8-81 was equal to or greater than clone 11-11 in the first, third, seventh, and eighth years.

Figure 1. Eight-year cumulative growth in height and basal diameter of trees surviving to year 8 by clone.

Differences among clones in magnitude and trends in annual growth can be attributed in part to differences in leaf and growth phenology and prevailing weather conditions. Clones D-01 and 7-75 commonly broke bud in early March; clone 8-81, by mid-March; and clone 11-11, at the end of March or in early April. Terminal bud-set occurred in early to mid- September in clone D-01, and foliage soon yellowed and dropped. Terminal growth continued into October for the two black cottonwood clones and hybrid 11-11, and green foliage sometimes remained on the latter clone into early November. Such differences in shoot and leaf development are reflected in seasonal patterns of periodic daily height and diameter growth of dominant and co-dominant trees as shown in Figure 2 for the third growing season — a year of average temperature and moisture conditions. By mid-May, height growth of all clones had accelerated; all attained maximum height growth of 2 to 3 mm per day in July. Height growth of clone D-01 then declined rapidly whereas growth of the two black cottonwood clones and hybrid clone 11-11 continued at a near maximum rate into September. In general, diameter growth of the Populus clones peaked earlier than height growth. Although this finding may run counter to phenological trends commonly found for other conifers and hardwoods in temperate zones, it is consistent with the pattern for red alder (Alnus rubra Bong.), another species with a “sustained growth” habit (Zimmerman and Brown 1971), growing at the study site and other locations (DeBell and Giordano 1994).

Figure 2. Phenology of height and basal diameter growth of four Populus clones planted at 1.0-m spacing.

There were also differences in diameter growth patterns among clones. Diameter growth of clone D-01 peaked earlier than that of the two black cottonwood clones. Diameter growth of the two black cottonwood clones and, to a lesser extent, hybrid clone 11-11 remained at a high level later into the season whereas clone D-01 declined rapidly after its peak in mid-May. When moisture was abundant and temperatures were warm early in the growing season and conditions later in the season were unfavorable (i.e., dry or unusually cool), growth of clone D-01 was relatively better than growth of the other clones. Conversely, adverse growing conditions during the early season followed by more favorable late season temperature and moisture conditions favored growth of the other clones. Similar though less pronounced differences existed between the two black cottonwood clones and hybrid clone 11-11, but relative growth performance probably is more specifically influenced by the duration of drought stress as the hybrids of P. trichocarpa and P. deltoides have greater stomatal control and drought resistance than black cottonwood (P. trichocarpa) clones (Schulte and others 1987).

Although general declines were expected with age, growth was less than expected in the fourth (1989) and sixth (1991) years (Fig. 1a and b), with the apparent reduction being greater for the two black cottonwood clones and clone D-01 than for clone 11-11. Weather and irrigation records show that during the March-June period in both 1989 and 1991, precipitation was below normal and irrigation inputs were very low because of equipment problems. Total input of water during the early growing season of 1989 and 1991 was at least 50% less than that occurring during the first three years of growth (1986-1988) and in the fifth (1990) and seventh (1992) years.

Development of buds, branches, and leaf area

Height and diameter growth of the clones prior to the onset of intertree competition were strongly influenced by the dynamics of bud and branch populations and associated effects on the development of leaf area (Table 1). Although the clones produced similar numbers of buds on the main stem during the establishment year, substantial differences occurred in the extent of sylleptic branching and development of leaf area. A branch is classified as sylleptic if it develops and elongates during the same growing season the bud is formed. If the bud does not develop into a branch until the following growing season, the branch is classified as proleptic. The two black cottonwood clones and clone 11-11 had many more sylleptic branches and at least three times the leaf area of clone D-01 by the end of the first growing season. The total number of branches per tree produced by clone D-01 during the first two years was comparable to those produced by the other clones; however, since almost all of its branches were proleptic, they did not contribute to the development of leaf area until the second year. In addition, proleptic branches have been shown to export a lower percentage of photosynthate than sylleptic branches (Scarascia-Mugnozza 1991; Hinckley and others 1992). The combination of the differences in pattern of leaf area development and the carbon export behavior of the two branch types has been suggested as conferring a potential productivity advantage to clones with greater production of sylleptic branches (Hinckley and others 1992). Although the degree of syllepsis may not be the only contributing factor, differences in second-year sylleptic branching and leaf area were consistent with subsequent growth rankings among the four clones (Table 1, Fig. 1). For example, clone 11-11 produced more buds and more sylleptic branches than any other clone; it produced 39 to 124% more leaf area and maintained its size ranking. A shift in rank occurred between the two native clones. Clone 7-75 outgrew clone 8-81 during the first two seasons, but by the end of the second year the greater sylleptic branching of clone 8-81 resulted in greater leaf area. By the end of the third season, these two black cottonwood clones were very similar in height but clone 8-81 was slightly larger in diameter. By the fifth season, clone 8-81 had surpassed clone 7-75 in both height and diameter. Number of branches retained on the first year's height increment was greater for clone D-01 than for the other three clones. After the third growing season, little branch mortality had occurred on the lower portion (first year's growth) of clone D-01 whereas, essentially all branches had died on the other clones.

Stem form

The clones also differed in slenderness ratio (height/basal diameter) and lower-stem taper (basal diameter/dbh) (Table 2). At age 2, clones 7-75, 8-81, and 11-11 had smaller values for lower stem taper and higher slenderness ratios than D-01. The values for lower stem taper at age 2 were negatively correlated (r=-0.92; p=0.07) to the percentage of buds producing sylleptic branches. Thus, the higher the percentage of buds becoming sylleptic branches, the less tapered the tree bole. This effect is consistent with the previously mentioned observation that sylleptic branches export a higher percentage of photosynthate than proleptic branches (Scarascia-Mugnozza 1991; Hinckley and others 1992). Such differences among clones were reduced over time as the live crowns lifted and lower stem growth was influenced by more complex interactions of branch type, age, and crown position, including clonal differences in branch retention. Slenderness ratios have increased over time as would be expected at this dense spacing. Clone 8-81 grew proportionately more in diameter than in height between years 2 and 8 than 11-11 and thus shifted in rank from first to third in relative slenderness.

Leaf area indices and accumulation of woody biomass and basal area

Leaf area expanded rapidly in the second and third growing seasons, nearing or reaching a peak in all clones during the third year (Fig. 3a). Leaf area index attained maxima of 7.7 and 7.8 for hybrid clone 11-11 at age 3 and 4 and then declined to 6.4 at age 5. Maximum leaf area indices for the other three clones ranged from 5.0 to 5.7. Leaf area indices for clones 11-11, 7-75, and 8-81 had declined by age 5, averaging about 20% lower than the maxima. Leaf area index of clone D-01, however, changed relatively little after attaining a plateau of about 5.0 at age 3.

Above-ground woody biomass production accelerated markedly during the third season, and rates were more or less maintained through the fifth year (Fig. 3b). At 5 years, total production was 56 Mg ha-1 for D-01, 63 and 66 Mg ha-1 for 7-75 and 8-81, and 91 Mg ha- 1 for 11-11. Mean annual increments through age 5 were 11 Mg ha-1 for D-01, 13 Mg ha-1 for 7-75 and 8-81, and 18 Mg ha-1 for 11-11. Subsequent evaluations of biomass production of clones 11-11 and D-01 through year 7 in adjacent spacing trials revealed that current annual biomass increment had peaked in the 1.0-m spacings in the 4th year for clone 11-11 and in the 5th year for clone D-01. The decline was so great in clone D-01 that mean annual increment also culminated in the 5th year; mean annual biomass increment did not peak in clone 11-11, however, until the 6th year. Because height and diameter increment patterns of the two black cottonwood clones were intermediate between those of the two hybrids, we suspect that similar biomass growth trends occurred and that mean annual biomass increments at 5 years represented a near maxima for all four clones in the study environment. Moreover, our measured rates of production for clone 11-11 are very similar to those obtained in 7-year-old research trials (17 Mg ha-1 yr-1) and operational plantings (12 to 17 Mg ha-1 yr-1) by James River Corporation in the lower Columbia River Valley [pers. comm. from William Schuette of James River Corporation, Camas, Washington (January 4, 1995)].

The clones also differed in the amount and proportion of branches in the woody biomass yields. Clone D-01 had the most branches (~10 Mg ha-1), and they comprised about 18% of the woody biomass. Branch weights for other clones were 5 to 7 Mg ha-1 and accounted for only 7 to 10% of the woody biomass.

Basal area growth accelerated during the second and third season (Fig. 3c). Rate of accumulation declined slightly thereafter, but basal area of all clones continued to increase through the seventh season. Accumulated basal area of two clones, however, was lower for different reasons at the end of the eighth season. Losses in basal area of hybrid clone 11-11 were caused by combined effects of borer damage and a severe windstorm; losses for clone 7-75 were associated primarily with competition-related mortality. Highest basal areas per hectare attained at age 7 for clones 11-11 and 7-75 were 48 m2 and 31 m2, respectively, whereas highest basal areas attained at year 8 for clones 8-81 and D-01 were 38 m2 and 33 m2.

Figure 3. Accumulation of leaf area (a), above-ground woody biomass (b), and basal area (c) in stands of four Populus clones (legend as in Fig. 2b). Basal area based on diameter at 1.3 m. Standard errors shown as vertical lines.

Relative tolerance to intertree competition

Mean tree growth in height and diameter peaked in the second or third year, and, in general, annual basal area and, presumably, woody biomass growth per hectare had begun to decline by the fifth year for all clones planted at the 1.0-m spacing. Such declines in mean tree and per hectare growth in these young plantings are largely due to the development of stress associated with intense intertree competition. In order to determine whether differences existed among clones in tolerance to competition or “stockability” (DeBell and others 1989a), measures of annual mean tree (diameter) and per hectare (basal area) growth were plotted against a stand competition index (Figure 4a and 4b). With the exception of clone 8-81 in year 7 or 8, growth of hybrid clone 11-11 exceeded that of other clones at similar levels of competition. For other clones, annual basal diameter growth per tree began to decline at lower levels of competition index, and the decline was more abrupt than for hybrid 11-11 (Fig. 4a); the same is true of basal area growth per hectare (Fig. 4b). The sharp decline in apparent basal area growth during the last period (year 8) for clones 11-11 and 7-75 is due to mortality discussed in the previous section on basal area accumulation. Thus, clone 11-11 not only grows more rapidly than other clones, it also has the capacity to tolerate and grow more rapidly at higher levels of stand density. Both traits — individual tree growth and stockability — are important to stand productivity.

Figure 4. Current annual diameter growth per tree (a) and basal area growth per hectare (b) in relation to an index of stand competition (Sum D²H in m³ha-¹) at the beginning of the growing season for four Populus clones (legend as Fig. 2b). Standard errors shown as vertical lines.

Tree and stand characteristics at 8 years

Current annual increment in basal area and above-ground woody biomass peaked in all clones, and total accumulation in basal area (and presumably biomass) has declined in two clones. Tree survival at the end of the eighth year ranged from 68 to 95% (Table 3). Little mortality had occurred in any clone through age 7 years, but the two native clones and hybrid clone 11-11 suffered considerable mortality during the eighth growing season. Clone 11-11 was clearly superior to all other clones in height and diameter. The sizable differences among clones in mean basal area at age 8 were not statistically significant (Table 3), primarily because the values are strongly influenced by recent mortality and this was not distributed uniformly among the replicate plots of individual clones.

Damaging agents

During the first eight growing seasons, insect and disease problems were generally minimal. Aphids appeared on all clones but soon were controlled by an expanding population of ladybugs. Melampsora rust developed during late summer on the two native clones (7-75 and 8-81) and may partially account for their poorer subsequent growth in comparison to clone 11-11 which remained essentially free of rust. Some lateral buds were destroyed on all clones by eriophyd mites (Eriophyes parapopuli). This damage occurred primarily on the ends of lower lateral branches and was roughly proportional to the timing of budset (damage greatest on D-01, intermediate on 7-75 and 8-81, and least on 11-11).

Heavy infestations of the poplar-and-willow borers (Cryptorhynchus lapathi) were observed after the fifth year, especially in clone 11-11. Observations made in adjoining plots of other spacings suggested that infestations were greater in stands having higher density or more intense competition. By the end of the seventh growing season, 27 and 16% of the trees in clones 11-11 and 7-75 had been attacked, respectively, whereas only 8 and 5% of clones D-01 and 8-81 were affected.

After trees were measured at the end of the seventh growing season, a severe windstorm occurred in the area and resulting damage varied significantly among clones. Hybrid clone D-01 and the two native clones (7-75 and 8-81) were basically unaffected; although some trees of these clones had broken branches or tops, none were windthrown or seriously broken. Clone 11-11, however, suffered substantial damage; some stems were windthrown or “lodged”, but most damaged stems were broken at 0.2 to 1.0 m above ground. About 7% of all measurement trees of clone 11-11 in the 1.0-m spacing suffered such breakage, which occurred primarily in stem segments with evidence of considerable borer activity, and thus considerably weakened structure.

Conclusions and Implications

All four clones were easy to establish with unrooted hardwood cuttings at the experimental site under the imposed management regime and survival was excellent. Growth differences, however, were manifested early and clonal rankings were established by age 3. Bud and branching characteristics were closely related to leaf area development, tree growth rates, and stem form of the four clones tested.

Competition in these 1.0-m spaced plantings was such that growth in height and diameter of all clones declined substantially by the fourth growing season. Subsequent work in adjacent spacing trials involving clones 11-11 and D-01 only indicated that mean annual increment peaked during the sixth season for clone 11-11 and during the fifth year for clone D-01. Given that declines in height and diameter increm'ent of the two black cottonwood clones were intermediate between clones 11-11 and D-01, it seems reasonable to rank the clones based on productivity attained through year 5. Thus, clone 11-11 with 18 Mg ha-1 yr -1 was about 40% more productive than the two black cottonwood clones and more than 60% more productive than clone D-01.

Other research (DeBell and others 1996a) and operational experience [pers. comm. from William Schuette of James River Corporation, Camas, Washington (January 4, 1995)] suggest that these productivity values are indicative of the levels of mean annual production that may be obtained at wider spacings with slightly longer rotation ages. Our productivity values, however, are lower and differences among clones are less than values reported previously. Mean annual production values determined at age 5 by Heilman and Stettler (1985) for clone 11-11 and the population sources from which clones 7-75 and 8- 81 were derived were about 28 and 16 Mg ha-1, respectively. These authors appropriately acknowledged that clonal interactions among adjacent 9-tree plots exaggerated differences between high and low producers. Although the small plots established in most genetic trials are necessary (and useful) for screening large numbers of clones, they do not provide adequate information on per hectare productivity or on relative productivity differences among clones selected as appropriate for operational use. More accurate definition of such differences may be important when other factors, such as susceptibility to damaging agents or wood quality, may be significant considerations.

Plottings of annual growth in relation to stand competition indices suggested that stockability differences (tolerance of crowding or competition) exist among the clones. The potential importance of such differences to stand productivity merits additional investigation over longer time periods and with additional clones. Any finding of substantial differences in stockability implies that optimal management regimes may differ among clones with respect to one or more of the following: target diameter, stand density, or rotation age. Moreover, given the same management regime, clones having higher stockability offer greater flexibility for extending rotation length because yield reductions due to mortality or growth deceleration will be lower than for clones having lower stockability.

Pests and weather conditions had differential influences on the four clones despite the short time involved in our evaluation. The most productive clone (11-11) was resistant to the native Melampsora rust and least affected by drought, but was the most susceptible to borer infestation and wind damage. Such differences could attain considerable importance in some locations, particularly with longer rotations.