We analysed 320 increment cores at 10 different growth ring positions for a net total of 2691 ring samples, all taken from breast height. The objective of this research was to (a) detect which wavelengths were sensitive to blue stain and (b) determine which wavelengths were sensitive to age when subjected to chemometric interpretation. It was found that wood chemistry-associated wavelengths 1115–1195 nm were most sensitive to blue stain while those at 1335–1415 nm and 1655–1685 nm were also influenced by blue stain, but to a lesser extent. When blue stained samples are not included in the calibration set, but blue stain then occurs in the field, bias in lignin prediction may be more likely, since wavelengths associated with lignin were more sensitive to blue stain. Alternatively, lignin-associated wavelengths showed clear delineation in absorbance with age while the relationship between cellulose-associated wavelengths and age was apparent, but less clear.

A theoretical model was built predicting the relationship between microfibril angle and lignin content at the Angstrom (A) level. Both theoretical and statistical examination of experimental data supports a square root transformation of lignin to predict microfibril angle. The experimental material used came from 10 longleaf pine (Pinus palustris) trees. Klason lignin (n=70), microfibril angle (n=70), and extractives (n=100) were measured and reported at different ring numbers and heights. All three traits were strongly influenced by ring age from pith while microfibril angle and extractives exhibited more of a height effect than lignin. As such, the multivariate response of the three traits were different in the axial direction than the radial direction supporting that care needs to be taken when defining juvenile wood within the tree. The root mean square error of calibration (RMSEC) for microfibril angle of the theoretical model (RMSEC 9.8) was almost as low as the least squares regression model (RMSEC 9.35). Microfibril angle calibrations were also built from NIR absorbance and showed a strong likeness to theoretical and experimental models (RMSEC = 9.0). As a result, theoretical and experimental work provided evidence that lignin content played a significant role in how NIR absorbance relates to microfibril angle. Additionally, the large variation in extractives content coupled with sampling procedure proved important when developing NIR based calibration equations for lignin and microfibril angle.

The influence of blue-stain fungi [Ophiostoma minus (Hedgcock) H. and P. Sydow and Leptographium serpens (Goid.) Siemaszko] on absorbance at the visible and near infrared wavelengths was investigated. Forty trees were sampled at breast height from longleaf pine (Pinus palustris Mill.). One half of each increment core was inoculated with one of two fungi treatments while the other half served as a control.

The variation in kink angle, number of kinks per millimeter, and curl was investigated for 10 longleaf pine (Pinus palustris) trees. Property maps indicated that kink angle and curl increased from pith to bark and from crown to tree base. The number of kinks per millimeter did not follow any particular trend in the horizontal and vertical direction. Unitless curl values were lower (0.024 to 0.067) than what would be expected after beating or refining, but perhaps they give an indication of which zones in the tree are more susceptible to curl during processing. Breast height to whole tree values confirmed that increment cores could be used to estimate whole tree values. However, individual rings at breast height were less successful in predicting rings further up the tree at similar ages from meristem production (rings from pith), suggesting that one should sample and average more than one representative ring when predicting whole tree values. Furthermore, the geographic direction of the increment core axis had an impact in predicting whole tree values, with the north having higher predictive ability than the south for all three fiber characteristics.

The prediction of tracheid length using near infrared (NIR) wavelengths can provide either useful or misleading calibrations depending on the context. This can happen since tracheid length is not directly related to the absorbance at any wavelength but is instead the result of a secondary correlation with some unknown chemical constituent. In this work, the effect of tree age and height on NIR predictability was investigated since tracheid length and chemistry may vary as a function of location within the tree. It was found that tracheid length predictability did not change with height but decreased with age. As a result, predicting tracheid length regardless of age was good ( R 2 = 0.72) while predictability holding age and height constant was mostly low to moderate with the exception of rings 1 and 4 which was quite strong.

The objective of this research was to (a) determine if blue stain in solid wood influenced calibration equations developed from a non-stained wood population, (b) assess the bias introduced when scanning was performed by the slave instrument without calibration transfer from the master instrument and (c) partition absorbance-based variation by instrument, stain and instrument × stain interaction. The results helped to determine the calibration transfer needed for this case. The dependent variables assessed from clear and stained wood were lignin, extractives, modulus of elasticity (MOE), modulus of rupture (MOR) and density When the master instrument was used for both calibration and prediction, it was found that stain-insensitive equations for the five traits could be built. However, when a slave near infrared instrument was introduced without calibration transfer, three out of five predicted traits were significantly biased by the presence of stain. Further analysis revealed an interaction between stain and instrument indicating that instrument bias was also introduced during scanning with a slave. For both multiple linear regression (MLR) and principal components regression (PCR), it was found that if a trait needed more wavelengths (or principal components) for prediction of the dependent variable, bias due to blue stain became increasingly prominent. PCR was found to perform better than MLR when stain was introduced with no calibration transfer. Such a finding alludes that PCR works better than MLR under extrapolation conditions but is not intended to support a lack of calibration transfer. Finally, the Mallows C p diagnostic proved valuable in model selection although the well-known requirement of ( C p – p ≤ 0) appeared conservative. For MLR and PCR, a C p – p ≤ 5 often yielded applicable models while C p – p > 7 was about the threshold where model performance dropped.

Process control of wood density with near infrared spectroscopy (NIR) would be useful for pulp mills that need to maximize pulp yield without compromising paper strength properties. If models developed from the absorbance at wavelengths in the NIR region could provide density histograms, fiber supply personnel could monitor chip density variation as the chips enter the mill. The objectives of this research were to a) develop density histograms from actual density versus density histograms developed through NIR modeling, and b) determine the precision of density models developed from absorbance in the NIR region with a recommendation for the sample size needed to estimate the standard deviation of density at a given precision. Models for density were developed from calibration samples (n = 170) and then validated with 93 randomly held aside samples. The samples were systematically removed from 10 longleaf pine trees of equal age, but different growth rates. The histogram patterns for actual density almost paralleled the histogram patterns developed from predictive models. Subsequently, the validation data set was randomly categorized into groups of three, and the standard deviations of density were measured. For three measurements per data point, the predicted standard deviation covaried with the actual standard deviation of density with an R 2 = 0.61 and 0.55 for the calibration and validation data set, respectively. A sample size of 30 was recommended to estimate the standard deviation of density with a precision of 0.01 g/cm 3 .

Improvement of specific gravity through tree breeding was an early choice made in the mid 20(th) century due to its ease of measurement and impact on pulp yield and lumber strength and stiffness. This was often the first, and in many cases, the only wood quality trait selected for. However, from a product standpoint, increased specific gravity has shown to lower many paper strength and stiffness properties and has been assumed to be directly attributable to increased fiber coarseness. As a result, it is currently not clear which fiber trait would best benefit a tree improvement program for paper products. This review found coarseness to be perhaps more important to paper strength and stiffness whereas tracheid length showed better promise from a breeding point of view due to its independence from specific gravity. However, both traits possessed strong heritability and influence on product performance and thus both would be beneficial to breed for depending on organizational goals and end product mix. The objective of this paper is to review and prioritize coarseness and tracheid length from both an end use and raw material perspective. To aid in prioritization, the variation, correlation, and heritability of both traits were reviewed along with significant genetic and phenotypic correlations. Variation trends within and between families as well as within a tree were reviewed.

Papers and abstracts from the 27th Southern Forest Tree Improvement Conference held at Oklahoma State University in Stillwater, Oklahoma on June 24-27, 2003.

In manufacturing, monitoring the mechanical properties of wood with near infrared spectroscopy (NIR) is an attractive alternative to more conventional methods. However, no attention has been given to see if models differ between juvenile and mature wood. Additionally, it would be convenient if multiple linear regression (MLR) could perform well in the place of more complicated multivariate models. Therefore, the purpose of this paper was to model the strength, stiffness and density of mature and juvenile longleaf pine to NIR spectra with MLR and principal component regression (PCR). MLR performed almost as well as PCR when predicting density, modulus of rupture (MOR) and modulus of elasticity (MOE). Choosing wavelengths associated with wood chemistry and developing principal components gave better predictive models (PCR2) than when all NIR wavelengths were used (PCR1). Models developed from mature wood did not predict wood properties from juvenile wood adequately, suggesting that separate models are needed. However, for density prediction, the area under the spectral curve appeared to be insensitive to mature and juvenile wood differences. Five of the six wavelengths associated with MOE were also associated with MOR, perhaps accounting for how MOE and MOR might be related. For pith wood, MOE and MOR were poorly related to NIR spectra, while density was strongly correlated. This inability to predict mechanical properties in the pith-wood zone warrants attention for those manufacturers interested in using near infrared to stress rate lumber within a mill.

‘School of Forestry, Wildlife, and Fisheries, Louisiana Agricultural Experiment Station, Louisiana State University Agricultural Center, Baton Rouge, LA 70803 2USDA Forest Service, Southern Research Station, Southern Institute of Forest Genetics, 23332 Hwy. 67, Saucier, MS 39574 31nternational Paper Inc., 719 Southlands Rd., Bainbridge, GA 31717 4Texas Forest Service, Forest Science Laboratory, Texas A & M University, College Station, Texas 77843 5USDA Forest Service, Southern Research Station, 2500 Shreveport Hwy., Pineville, LA 71360

We compared the degree of differentiation among 22 populations of white spruce (Piceaglauca (Moench) Voss), a monoecious conifer, for six nuclear allozyme loci and a paternally inherited restriction fragment polymorphism in the chloroplast DNA (cpDNA). Estimates of FST (an F-statistic for the proportion of total variability due to differences among populations) were significantly higher for the cpDNA marker (0.147) than for the nuclear allozyme loci (mean = 0.037, range = 0.017–0.053), indicating greater differentiation for the cpDNA. Both cpDNA variants were observed in every population, and there was an east–west trend in the frequencies of these variants across North America. These observations may be explained by the different modes of inheritance of the cpDNA (paternal) and nuclear loci (biparental).

Three major themes related to the improvement of salt tolerance in forest tree species are examined. First, evidence demonstrating that substantial intraspecific variation in salt tolerance exists in many species is presented. This evidence is important because it suggests that efforts to improve salt tolerance through conventional plant breeding techniques are justified. Second, the physiological and genetic mechanisms controlling salt tolerance are discussed briefly. Although salt tolerance involves the integration of numerous physiological processes, there is considerable evidence that differences in the ability to exclude Na(+) and Cl(-) from leaves are the most important factors underlying intraspecific differences in tolerance. It is also becoming apparent that, although salt tolerance is a multigenic trait, major genes play an important role. Third, progress to date in improving salt tolerance of forest tree species is assessed. Compared with agricultural crops, relatively little progress has been made with either conventional or biotechnological methods, but field trials designed to test clones identified as salt tolerant in screening trials are underway now in several countries. We conclude that there is justification for cautious optimism about the prospects for improving salt tolerance in forest tree species.

Variation in height at ages 9 and 19 years and at six polymorphic allozyme loci was examined for 22 seed sources (populations) in a range-wide white spruce (Piceaglauca (Moench) Voss) provenance test planted in Minnesota. There were strong differences among populations for height, with 48.0 and 54.1 % of the genetic variation for height at ages 9 and 19, respectively, due to differences among populations. Mean observed and expected estimates of allozyme heterozygosity were 0.306 and 0.290, respectively, with little deviation from genotype frequencies expected under a Hardy–Weinberg equilibrium. In contrast with the height data, an average of only 3.8% of this variation was due to differences among populations. Geographic trends were apparent in the height data, with northern and western sources performing the poorest. Neither univariate nor multivariate analyses revealed any geographic trends in the allozyme data. The very different distributions for height and allozyme variation suggest that evolutionary forces are acting in different ways on the genes controlling these traits, and that allozyme data will have limited value in developing sampling strategies for gene conservation programs, where the preservation of germ plasm adapted to many sites throughout a species range is important.

Considerable concern has been voiced regarding the reproducibility/transferability of RAPD markers across different genetic backgrounds in genetic mapping experiments. Therefore, separate gametic subsets (mapping populations) were used to construct individual random amplified polymorphic DNA (RAPD) linkage maps for a single longleaf pine (Pinzrs pahstris Mill.). A haploid mapping population consisting of megagametophytic DNAs from 88 wind-pollinated seeds of longleaf pine (clone 3-356), and a diploid population including 86 F, progeny from a controlled cross [longleaf pine 3-356 (0) x slash pine H-28 (O")] were employed. Seventy-one MPD primers selected for this study identified a total of 137 mapped loci in longleaf pine 3-356 based on amplification of megagametophytic DNAs. Of the 137 loci useful for comparative purposes, 62 loci (45.3%) could not be scored when DNAs of F, progeny from the controlled cross were amplified. Of the 75 loci that were storable, 26 loci (34.7%) were fixed in slash pine H-28, and 49 loci (65.3%) were segregating (42 1:1, and seven 3:l). Comparisons were made using the 49 loci common to both the haploid- and diploid-based maps. The 49 loci allowed us to determine homologous counterparts between maps. Orders were conserved for those groups containing three or more loci. Genetic distance estimates tvere found to vary considerably, but not in any systematic manner.

--Single marker regression and single marker maximum likelihood estimation were used to detect quantitative trait loci (QTLs) controlling the early height growth of longleaf pine and slash pine using a ((longleaf pine x slash pine) x slash pine) BC, population consisting of 83 progeny. Maximum likelihood estimation was found to be more power than regression and could also estimate the distance between markers and QTLs. Test statistic for the relationship between simple regression and maximum likelihood estimation is introduced. A total of four major QTLs linked to random amplified polymorphic DNA (RAPD) markers were detected explaining 19.7%, 10.7%, 12.8%, and 9.9% total variance of total height. Multiple regression analysis indicated that these four QTLs explained about 43.2% of the total variance of early height growth.