Widths of 22,734 bands of earlywood and latewood were measured systematically along the average radii at the centers of the annual height increments of 18 Douglas fir trees. Shapes of the annual growth layers of earlywood and of latewood, respectively, formed during an accrued growth period of 589 years were investigated: (i) using relative measures embodied in diagramatic computer outputs; (ii) using absolute measures by statistical and graphical techniques; (iii) by computing Hohenadl's form factor (lambda 0.9) for each year's growth of; a) imaginary "earlywood stems" consisting of layers of earlywood; b) imaginary "latewood stems" consisting of layers of latewood; c) actual stems consisting of total annual layers. The form of earlywood layers differed markedly and consistently from that of latewood layers. The maximum width of earlywood layers in individual trees occurred within a zone located in the upper portion of the live crown; in the stand it was within a zone parallel with the surface of crown canopy. Width of earlywood was at its minimum at some distance above the stem base. This distance increased with tree age. Latewood layers were usually widest along the basal portion of the stem. As a result, the form factors of "earlywood stems" were considerably higher than those of "latewood stems". The shapes of the growth layers, and hence the form of stems consisting of these layers, could not be reconciled satisfactorily with the tenets of Schwendener - Metzger's mechanistic, or Hartigfs nutritional, or Jaccard's water conductive, or hormonal theories of stem formation. Therefore, a new scheme was proposed using the following concepts: (1) Heating of stems by solar energy constitutes a purely physical process; the rate of energy transfer between a tree and its environment determines the temperature of its cambial tissues. (2.) Because trees are not homoiothermous organisms, at a given time various parts of the cambial cylinder may possess different temperatures even in an isothermal environment. (3) A pronounced stratification of the environment due to gradients in air temperature or in length of time of positive net flux of energy has been observed in forests throughout the world. (4) Radial growth may proceed at varying rates for unequal periods of time within the different parts of the cambial cylinder, depending largely on the levels of subcortical temperatures. (5) Consequently, the shapes of the annual layers of earlywood and latewood may reflect the respective spring and summer environmental energy gradients. (6) Then the average form of trees from forests of the temperate latitudes, which is that of a quadratic paraboloid, may reflect the average microclimatic structure prevailing in these forests during the growing season. Form of open-grown trees, eccentricity of stems, roots and branches, and other so far unexplained anomalies in radial growth may be clarified similarly. Indirect and some preliminary direct evidence supporting the proposed conceptual scheme of stem formation was presented. In addition, influence of some selected factors of macroclimate on the amount of radial growth expressed in terms of the average widths of growth layers was analysed. Individual trees have been used as sampling units. The trends in the growth series were removed by analysis of covariance: average layer width indices were derived by calculating deviations from the straight lines fitted by least squares to the adjusted mean layer widths. The degree of autocorrelation of both growth and weather series was largely nonsignificant. Correlations between the growth indices of earlywood and latewood were nonsignificant or low. In the individual trees, six weather variables accounted for from 10 to 48 per cent of the total variability observed in the radial growth of latewood. Temperatures of the previous summer could not be related to the amount of radial growth of earlywood of the current year. Since the approximate minimum true correlation in the universe was zero the general influence of macroclimate was nonsignificant. It appears that other studies have neglected the influence on growth of microenvironmental factors and that there is need for research on the means by which distribution and amount of radial growth are controlled by the net flow of energy. / Forestry, Faculty of / Graduate

Variability in the relationship between tree leaf area (TLA) and selected stem measurements was examined in three Douglas-fir stands (Pseudotsuga menziesii (Mirb.) Franco, var. menziesii) that were less than 50 years-old, spaced to approximately 550 to 650 stems/ha, and differed in soil moisture and nutrients. Attention was given to the effect of mean annual ring width (MARW), cross-sectional area of the live bark (ALB - a surrogate measure of relative nutrient storage in the stem), and cross-sectional area of the most recent annual rings equal in number to the number of whorls in the live crown (ALC), on variability in the relationship between TLA and cross-sectional area of sapwood (ASW). At breast height, basal area, ASW, and cross-sectional area of sapwood plus live bark (ASWLB) were not linearly related to TLA, and linear regression equations using log transformed variables varied significantly between sites. Nonlinear regression equation for ASW at breast height was: TLA = 0.064ASẆ¹•³³ (I² = 0.856). Including D (the distance between breast height and the center of the live crown) in the nonlinear equation, did not significantly improve the regression. Tree leaf area prediction models using stem measures from the base of live crown (blc) had higher adjusted R² values than models using stem measures from breast height. At the blc, basal area, ASW, and ASWLB were linearly related to TLA (adjusted R² = 0.926, 0.908, and 0.934, respectively). Multiplying ASW by MARW did not improve the fit of the regression models. Multiplying ASW by ALB improved the linearity of the relationship of ASW at breast height to TLA. The best fitting TLA model overall used the product of ASW at blc and ALB at blc as the independent variable (adjusted R² = 0.967). The results indicate that research into the allometric relationship of TLA to stem measures should give consideration to more than hydraulic measures and include measures of bark function. At breast height and the blc, the independent variable ALC was linearly related to tree leaf area and had higher adjusted R² values than did ASW. In most trees the ALC stem measure was found to include a portion of heartwood area. The strong relationship between TLA and ALC suggests that a given transpiring leaf mass or area is related to a proportional amount of conducting stemwood and physical support stemwood. A quick alternative approach for estimating individual tree leaf area using photographs taken at fixed distance and angle from the target tree did not result in a reliable tree leaf area prediction technique. The difficulty of obtaining views of the tree crown which were not obstructed by adjacent tree crowns was the major obstacle. Using a fixed distance and camera angle was a problem because of variable tree heights. However, altering these fixed positions introduced additional variation into the tree leaf area estimation. Mean specific leaf area (SLA) varied significantly by site, needle age class, and crown position. Mean SLA per needle age class per branch can be predicted with 95% confidence and a 10% allowable error using six 10-needle samples. / Forestry, Faculty of / Graduate

Aphids are a widespread family of plant pests, whose abilities to suppress shoot and root growth are well documented for many terrestrial plants. Only a few studies have been conducted on conifer aphids of the genus Cinara. Cinara pseudotsugae are known to attack Douglas-fir seedlings, an important crop in the Pacific Northwest. Douglas-fir are most susceptible to aphid damage as seedlings, especially in nurseries where conditions favor aphid outbreaks. A parasitoid wasp (Pauesia sp.) attacks C. pseudotsugae, and may be useful as a biological control agent. Studies of its natural history and host interactions are needed to assess its potential as a bio-control agent. This study examined the effects of an experimental range of aphid densities on the growth of total shoot and root volume and biomass and shoot morphology of Douglas-fir seedlings. Eighteen-week tests explored short term effects of different aphid feeding intensities in both the greenhouse and field. Long term effects were tested by exposing greenhouse seedlings to 16 months of aphid feeding. The ability of the plants to recover was tested by allowing one set of seedlings to grow aphid free for one year, after being exposed to aphid feeding for 18 weeks. The success rate of parasitoids over an 18 week period was compared to aphid density in both greenhouse and field tests. Increasing aphid destiny was significantly related to decreasing root and shoot dry weights in greenhouse tests. Growth suppression increased slightly during the second year of testing, regardless of whether or not aphid feeding continued. The results for root and shoot volumes were highly variable. However, root tissue density was significantly reduced after the second year of testing. Few shoot characteristics showed consistently significant aphid effects among the trials. Stem diameter and height decreased and needle density of new buds increased significantly with aphid feeding in most tests. Root and shoot growth of field plants did not show any significant aphid effects. Percentage of parasitoid success was independent of aphid density except at the lowest aphid densities. There was a block effect on parasitoid success in the field test, that may have been a result of varying environmental conditions. These results indicate that even short term aphid feeding can have long lasting effects on plant growth and structure. The effect on shoot and root growth was small, but there were no signs of recovery. The long term effects of the reduced root tissue density on Douglas-fir is unknown. / Graduation date: 1998

Measurements of the energy balances and net photosynthesis rates of two low productivity coniferous forest canopies (12 and 22 years old), were made successfully during both wet and dry growing seasons, using a modified Bowen Ratio method. Canopy conductances (gc) were calculated from canopy evaporation rates (E) using the Penman-Monteith equation. A model was developed to predict canopy growth and evaporation rates from basic soil and weather data, and compared with the measured data. The photosynthesis model was physiologically based, derived from recent work of Farquhar and coworkers. The canopy conductance model used an empirical approach, based on simple relationships with recorded environmental variables, while canopy E was predicted from the Penman-Monteith equation. Findings were: (1) Daytime E and canopy net photosynthesis rates (Fc) were generally lower in the younger canopy. (2) In the old canopy, E was more strongly decoupled from net irradiance (Rn) and more dependent on the atmospheric vapour pressure deficit (D) in accordance with the predictions of McNaughton and Jarvis (1983). (3) In the old canopy, Fc was significantly reduced by low soil water potential (Ψs) within the range of soil water storages at which measurements were made, while gc was less dependent on Ψs. From consideration of changes in intercellular C0₂ concentration, gc was not found normally limiting to Fc. (4) No simple relationship was apparent between solar irradiance (S) and F at the canopy level. However highest Fc and canopy water use efficiency ratios occurred on cloudy days with low air temperature and low D. (5) Night-time Fc measurements indicated that canopy respiration rates are generally very high and hence air temperature was a major factor limiting overall forest productivity. (6) The computer model could predict gc from four variables (D, S, root-zone soil water storage, W and time since dawn, t) with reasonable 2 success (r² 0.75). However, on days when gc was low, due to high D, E was occasionally significantly in error, because the Penman-Monteith equation is more sensitive to gc when the latter is low. Best agreement between measured and modelled E occurred on cloudy days when D was low and gc consequently high. (7) Values for the maximum rates of carboxylation, as limited by foliar carboxylase activity and electron transport rate, were set at one third of those reported by Farquhar and coworkers, in order to obtain best overall agreement between measured and modelled data. This requirement indicated that poor nutrition was also limiting to stand productivity. (8) Model prediction of canopy net photosynthesis was not satisfactory (r² 0.50), attributed mainly to using too simple an approach to estimating irradiance at the individual leaf level, and partly to unexplained variation in the measurements of Fc. In spite of its limitations, the model was found to respond realistically to changes in weather and Ψs, suggesting the approach was valid, and might be more successful with further development. / Forestry, Faculty of / Graduate

The purpose of this study was to find out if ecological races with different growth rates existed in douglas fir. Seeds were collected from six areas by the Forest Service and the author. A study was made of stem and tap root growth utilizing three photoperiods of varying length. Rate of tap root growth was significant for ecotype suggesting selection for tap root length in the plants from utah and Arizona. Each geographical area represents a growth rate ecotype with a wide break in the data on stem growth.