Holzforschung, Vol. 59, pp. 669–674, 2005 • Copyright
by Walter de Gruyter • Berlin • New York. DOI 10.1515/HF.2005.
107
Comparison of morphological and chemical properties
between juvenile wood and compression wood of loblolly
pine
Ting-Feng Yeh1, Barry Goldfarb2, Hou-min
Chang1,*, Ilona Peszlen1, Jennifer L. Braun3 and
John F. Kadla3,*
1
Department of Wood and Paper Science, North
Carolina State University, Raleigh, NC, USA
2
Department of Wood Sciences, North Carolina State
University, Raleigh, NC, USA
3
Forest Science Center, University of British Columbia,
Vancouver, BC, Canada
*Corresponding authors.
Department of Wood and Paper Science, North Carolina State
University, Raleigh, NC 27695, USA
4034 Forest Science Center, University of British Columbia,
Vancouver, BC, V6T 1Z4, Canada
E-mail: hchang@unity.ncsu.edu; john.kadla@ubc.ca
Abstract
In conifers, juvenile wood (JW) is always associated with
compression wood (CW). Due to their similar properties,
there is a common belief that JW is the same as CW. To
resolve whether JW is identical to CW, 24 rooted cuttings
of one loblolly pine clone were planted in growth cham-
bers under normal, artificial bending, and windy environ-
ments. The results show that the morphology of JW is
significantly different from CW. Furthermore, chemical
analyses revealed that JW and CW are significantly dif-
ferent in chemical composition. Our results indicate that
JW is different from CW, and the wood formed under a
controlled windy environment is a mild type of compres-
sion wood.
Keywords: C9 formula; compression wood; fiber quality
analysis (FQA); juvenile wood; light microscopy; loblolly
pine (Pinus taeda); nitrobenzene oxidation; ozonation;
sugar analysis.
Introduction
In the southeast US the demand for timber and pulp-
wood has increased greatly. However, due to environ-
mental concerns, the land available for wood production
has decreased. To meet future wood demand in the US,
more wood with target characteristics must be produced
from the limited forestland by intensive management
practices and genetic improvements (Bendtsen 1978;
Zobel and Van Buijtenen 1989). Forest product manufac-
turing is shifting raw materials from mature trees to short-
rotation plantation or juvenile stock.
Juvenile wood, sometimes referred to as core wood,
is formed by a young vascular cambium, and is the wood
formed near the stem center (Zobel and Sprague 1998).
Generally, a 15-year-old loblolly pine contains as much
as 85% juvenile wood by volume, whereas a 40-year-old
tree contains only 20% juvenile wood by volume (Zobel
1975). Compared to mature wood, juvenile wood has dif-
ferent properties, such as lower wood density, shorter
fiber length, higher lignin content, and higher compres-
sion wood content (Zobel 1981; Timell 1986).
Due to environmental disturbances, such as prevailing
winds or gravitational responses, compression wood is
formed under leaning stems and branches and manifests
its effect by slowly ‘‘pushing’’ the stems or branches
back to the original orientation (Scurfield 1973). For juve-
nile pine, the percentage of compression wood is on
average approximately 18% (Bendtsen 1978), but can be
as high as 44% (Zobel and McElwee 1958). Compression
wood is generally considered to be inferior for both pulp
and solid wood products.
When compared to mature normal wood, both juvenile
and compression wood have higher lignin content and
lower cellulose content, resulting in higher chemical con-
sumption during pulping and lower pulp yield. Subse-
quently, it is generally considered that juvenile wood is
the same as compression wood (Lee 1961; Watanabe
et al. 1963; Zobel 1984; Zobel and Sprague 1998). As a
result, the increased use of juvenile wood (compression
wood) from fast-growing plantation forests can signifi-
cantly impact industrial production costs and product
quality. However, systematic comparisons of the similar-
ity between juvenile wood and compression wood in
conifers are limited. Our objective in this research was to
identify whether juvenile wood is identical to compres-
sion wood in loblolly pine.
Materials and methods
Materials
A total of 24 rooted cuttings of one loblolly pine (Pinus taeda)
clone were transplanted from Ray Leach Supercells (164 ml) into
9-l pots. After conditioning inside a greenhouse for 4 weeks,
they were moved into growth chambers (Phytotron, North Car-
olina State University) for a 9-month growth period. The cham-
ber photoperiod and temperature were 18 h at 288C under
incandescent light, and 6 h at 208C in the dark. The trees were
randomly assigned to three environmental conditions, i.e., con-
trol, bent, and augmented wind.
The control and bent trees (eight trees each) were in a reduced
wind chamber; the average wind speed was 0.26 m sy1. The
eight augmented-wind trees were in a growth chamber with two
oscillating fans, and an average wind speed of 1.43 m sy1 (range
0.5–4.0 m sy1). The tree sway frequency was approximately 5
times miny1.
670 T.-F. Yeh et al.

The bent trees were rendered at 458 to the perpendicular

stems. When the new growth stem reached approximately
30 cm, another bend was applied. The trees were bent a total
of three or four times during the 9-month period. The initial aver-
age height of the trees was ;20 cm, and reached ;150 cm
(control and augmented wind trees) or ;100 cm (bent trees) at
harvest. All wood samples were collected from the same relative
height of each stem. By visual inspection, they were separated
into five groups: juvenile control normal wood (NW); bent oppo-
site wood (BOW); bent compression wood (BCW); wind opposite
wood (WOW); and wind compression wood (WCW).
The extractive-free wood sawdust (y35q60 mesh) was milled
in a planetary micro mill (Pulverisette 7, Fritsch, Germany) fol-
lowed by dioxane (96%, v/v) extraction. Milled wood lignin
(MWL) was prepared according to the method of Bjorkman
(1956). The yield of purified MWL was approximately 20% based
on the total lignin weight in the OD wood. The MWLs were acet-
ylated according to Adler et al. (1987).

Microscopic analysis

Wood blocks, collected from the same relative height of each

stem, were softened in water and sliced in the radial direction
Figure 1 Cross-sections of loblolly pine wood from different
using a microtome (Spencer Lens Co., Buffalo, NY, USA). The
groups: (a) control normal wood; (b) bent opposite wood; (c)
thickness was 10 mm. Microtome slices were stained with 1%
bent compression wood; (d) wind compression wood; and (e)
aqueous Safranin O, dried, and fixed on microscope slides with
wind opposite wood.
Permount (Fisher Science). A Nikon light microscope (Eclipse
E200) was used to photograph (40=10) the anatomical struc-
tent analyses were performed according to the procedure of
tures of the wood slices.
Goto et al. (2001).

Wide-angle X-ray diffraction (WAXD)

Quantitative 13C-NMR spectroscopy
Wide-angle X-ray diffraction profiles were recorded at the Uni-
All NMR spectra were recorded on a Bruker Avance 300-MHz
versity of British Columbia on a Bruker D8 Discover X-ray dif-
spectrometer at 300 K using DMSO-d6 as the solvent. The con-
fractometer with an area detector. Scans were collected from
centration of lignin was approximately 20%, and chromium(III)
wood samples at 40 kV and 20 mA in the range 48-2u-408
acetylacetonate (0.01 M) was added to lignin solutions as the
using CuKa radiation. The spectra were corrected for incoherent
relaxation agent (Capanema et al. 2004). A total of 20 000 scans
scattering (Krassig 1993) and then the crystallinity fraction was
were collected. A pulse width of 908, acquisition time of 1.2 s
calculated using the method of Vonk (1973).
and relaxation delay of 1.7 s were used (Capanema et al. 2004).

Basic chemical composition and fiber analysis

Wood extractives were removed according to Tappi standard
T264 om-88. The total lignin content was determined by the Kla-
son lignin method (Dence 1992; Yeh et al. 2005), combining the
Klason lignin and acid-soluble lignin. Holocellulose content was
determined according to Yokoyama et al. (2002) and used for
fiber quality analysis (FQA). Fiber length, width, coarseness, kink
and curl indexes were recorded using a fiber quality analyzer
(FQA; Op Test Equipment, Hawkesbury, ON, Canada) (Yokoya-
ma et al. 2002).
Monomeric carbohydrate determination
Monomeric sugar analysis was carried out according to the pro-
cedures of Coimbra et al. (1996).
Results and discussion
Compression wood has long been noted for its distinct
morphological structures (Wardrop and Davies 1964;
Shelbourne and Ritchie 1968; Scurfield 1973; Timell
1986; Kwon et al. 2001). For example, compression
wood tracheid lengths are approximately 10–40% short-
er than normal wood tracheids (Shelbourne and Ritchie
1968). They are rounder in shape, with thicker walls and
larger intercellular spaces, and they lack the S3 layer
(Scurfield 1973). The simplest comparison of juvenile
wood and compression wood tracheids can be made
using light microscopy.

Nitrobenzene oxidation and ozonation Tracheid morphology and fiber qualities

Nitrobenzene oxidation was performed according to the method
of Chen (1992). The stereochemistry (erythro and threo forms)
of the arylglycerol-b-aryl ether linkages of the lignin was deter-
mined by ozonation analysis according to the method of Akiya-
ma et al. (2002).
Element and methoxy content analyses
Elemental analyses (C, H, and N) were performed by Complete
Analysis Laboratories, Inc., Parsippany, NJ, USA. Methoxy con-
The tracheid morphology of the five different loblolly pine
wood samples is shown in Figure 1. Tracheids of the bent
compression wood (BCW, Figure 1c) were round in shape
with thick cell walls and large intercellular spaces, where-
as those of the juvenile control normal wood (NW, Figure
1a) were rectangular in shape without any intercellular
space. The bent opposite wood (BOW, Figure 1b) and
wind opposite wood (WOW, Figure 1e) showed similar
characteristics to those of the normal wood. Tracheids of
 Comparison of juvenile wood and compression wood 671

Table 1 Fiber properties of the various wood specimens.

Group Length Width Coarseness Curl Kink Crystallinity

(mm) (mm) (mg my1) index index fraction

Normal wood 1.221 A 29.1 A 0.128 A 0.0179 A 0.03 A 0.58 A

Wind opposite wood 1.317 A 29.8 A 0.142 A 0.0195 B 0.04 A 0.58 A
Wind compression wood 1.043 B 24.8 B 0.210 B 0.0283 C 0.28 B 0.59 A
Bent opposite wood 1.088 B 23.5 B 0.134 A 0.0183 BC 0.03 A 0.57 A
Bent compression wood 0.852 C 19.4 C 0.200 B 0.0276 C 0.35 C 0.54 A
Values are expressed as means and length is given as weighted length. The letters (A,B,C) indicate significant differences using the
Tukey-Kramer HSD test at as0.05.

the wind compression wood (WCW, Figure 1d) showed
a similar pattern to the bent compression wood (Figure
1c). The compression wood content of each of the trees
was estimated microscopically. It was found that the bent
trees and wind trees had approximately 50% and 30%
(by volume) compression wood, respectively, whereas
the control trees on average had less than 5% compres-
sion wood.
The mean values obtained from the fiber quality anal-
yses are listed in Table 1. The mean tracheid length of
the bent compression wood (0.852 mm) was approxi-
mately 40% shorter than that of control normal wood
(1.221 mm). The mean tracheid length of wind compres-
sion wood (1.043 mm) was also significantly shorter than
normal wood (1.221 mm) or wind opposite wood
(1.317 mm). Both the bent and wind compression wood
had a smaller tracheid width compared to opposite or
normal wood.
The coarseness of the compression wood (;0.2
mg my1) was significantly higher than that of the oppo-
site or normal wood (;0.13 mg my1). Since compression
wood had a smaller mean tracheid width than the oppo-
site wood and normal wood, the higher coarseness is
obviously due to thicker cell walls in the compression
woods (Figure 1). These results are consistent with data
reported by others (Shelbourne and Ritchie 1968; Timell
1986) concerning the higher density and shorter fiber
length of compression wood.
Fiber curl and kink are important fiber characteristics
known to impact paper properties. An increase in fiber
curl and kink (all other factors held constant) has a pos-
itive impact on out-of-plane tear, bulk, wet web strength,
etc., and a negative impact on tensile strength, burst,
and bending stiffness. The compression wood samples
(BCW and WCW) showed a higher tendency to curl and
kink in comparison with the opposite and normal wood.
Obviously, compression wood is different from juvenile
normal wood in tracheid fiber morphology and
properties.
The results of wood cellulose crystallinity are also listed
in Table 1. Although the average crystallinity value of the
bent compression wood was lower than that of normal
or bent opposite wood, however due to the large error
associated with separating the crystalline and amor-
phous moieties there is no statistical difference. These
results are in line with those reported by Tanaka et al.
(1981), who reported that the degree of crystallinity in
compression wood cellulose (45–50%) was lower than
that in normal wood (approx. 50%) and in opposite wood
(50–60%).
Basic chemical properties
The lignin and sugar contents of the five different groups
of wood samples are presented in Table 2. The cell wall
of compression wood had lower glucose, mannose,
xylose and arabinose contents, and higher galactose and
lignin than normal wood. The lignin content of normal
wood (NW) was 29.4% and was similar in magnitude to
that of opposite wood (BOW and WOW), at 29.1% and
28.9%, respectively. However, the lignin content in the
compression wood specimens (BCW and WCW) was
rather high at ;36%. Furthermore, sugar analysis shows
that compression wood had significant amounts of
galactose and relatively low amounts of the other sugar
units (Table 2). Thus, the juvenile normal wood was quite
different from the juvenile compression wood in basic
chemical composition, but similar to opposite wood.
These results are in agreement with other reports in the
literature (Bland 1958; Timell 1986; Lohrasebi et al. 1999).
Lignin heterogeneity
Compression wood is reported to be rich in p-hydroxy-
phenylpropane (H-type) lignin (Yasuda and Sakakibara
1975; Timell 1986; Fukushima et al. 1997; Önnerud and
Gellerstedt 2003). As shown in Figure 2, the nitrobenzene
oxidation products obtained from the different wood

Table 2 Lignin content and carbohydrate composition of different samples.

Group Mean composition (%)

Lignin Arabinose Xylose Mannose Galactose Glucose

Normal wood 29.4 A 4.0 A 14.4 A 16.2 A 3.3 A 62.1 A

Wind opposite wood 28.9 A 3.9 A 14.0 A 16.1 A 3.4 A 62.5 A
Wind compression wood 36.0 B 3.1 B 11.1 B 12.9 B 16.6 B 56.2 B
Bent opposite wood 29.1 A 4.0 A 13.8 A 16.4 C 3.5 A 62.3 A
Bent compression wood 36.6 B 2.8 B 10.2 B 11.4 D 21.1 C 54.5 C
Percentage composition is based on OD wood weight for lignin or total carbohydrates for the individual carbohydrates. The letters
(A,B,C) indicate significant differences using the Tukey-Kramer HSD test at as0.05.
672 T.-F. Yeh et al.
Figure 2 Nitrobenzene oxidation products from different
groups: NW, normal wood; WOW, wind opposite wood; WCW,
wind compression wood; BOW, bent opposite wood; BCW, bent
compression wood; H, p-hydroxybenzaldehyde; V, vanillin; and
VA, vanillic acid.
specimens indicate that the bent and wind compression
wood samples had relatively high amounts of p-hydroxy-
benzaldehyde units compared to the control or opposite
wood (BOW and WOW). Nitrobenzene oxidation yields
oxidation products from uncondensed lignin structures
(Chang and Allan 1971; Chen 1992) and therefore the
yield of nitrobenzene oxidation products can be an indi-
cator of how condensed a lignin structure is. The total
oxidation product yield of the bent and wind compres-
sion wood was lower than that of the normal or opposite
wood. This implies that compression wood lignin is likely
more condensed than juvenile normal wood.
Statistically, the yields of nitrobenzene oxidation prod-
ucts for bent compression wood were significantly dif-
ferent from the wind compression wood. Therefore, the
wind compression wood produced under our experimen-
tal conditions can be considered as a mild type of com-
pression wood.
Ozonation is a quantitative tool used to analyze the
stereochemistry werythro (E) and threo (T) isomersx of
arylglycerol-b-aryl ether linkages (Akiyama et al. 2002).
Ozonation results for the five different wood specimens
are shown in Figure 3. The total mol.% yield of the two
isomers (EqT) decreased from an average of 22.0% for
normal, bent opposite and wind opposite wood to an
average of 18.5% for the two compression wood sam-
ples (WCW and BCW). These results suggest an approx-
imate 15% decrease in arylglycerol-b-aryl ether linkages
in compression wood lignin. The E/T ratio of juvenile nor-
mal wood was 1.08, with no statistically significant dif-
ference from the bent opposite (1.07) or wind opposite
(1.09) wood. However, the E/T ratios of wind (1.15) and
bent (1.23) compression wood were significantly higher
than those of normal wood or opposite wood, suggesting
that the decrease in arylglycerol-b-aryl ether linkages in
compression wood lignin results more decrease from the
threo isomer (T) than the erythro isomer (E).
Normally, the ratio of erythro to threo forms in gym-
nosperms is approximately 1.0 (Lundquist 1980; Matsu-
moto et al. 1986; Akiyama et al. 2003). The results here
show that in compression wood lignin the erythro form
is more predominant. This is similar to reports on tension
wood of hardwood lignin (Akiyama et al. 2003). Thus, it
Figure 3 Erythro/threo (E/T) ratios and total yield (EqT) of dif-
ferent wood samples: NW, normal wood; WOW, wind opposite
wood; WCW, wind compression wood; BOW, bent opposite
wood; and BCW, bent compression wood.
seems that the erythro isomer is predominant in reaction
wood lignin.
The empirical C9 formulas of the different MWLs
obtained from elemental and methoxyl content analyses
are listed in Table 3. The methoxyl content of the wind
compression wood and bent compression wood was
approximately 0.15 and 0.26 per C9-unit lower than that
of juvenile normal wood, respectively. These data are
consistent with the results obtained from nitrobenzene
oxidation (Figure 2) and the results of Yasuda and Saka-
kibara (1975) who reported a 0.2 per C9-unit reduction in
methoxyl content between compression wood and nor-
mal wood MWL from larch (Larix leptolepis).
Quantitative MWL structure characterization was per-
formed by 13C NMR. The major differences in functional
groups and inter-unit linkages between the five MWLs are
listed in Table 4. The total H-unit content (p-hydroxyphe-
nyl moieties) was higher in compression wood (WCW and
BCW) than in juvenile normal (NW) and opposite wood
(WOW and BOW). The major difference in H-units is
attributed to non-conjugated p-hydroxyphenyl moieties,
as shown in Table 4. The amount of total hydroxyl groups
was slightly higher in the compression wood samples
than the juvenile normal wood or opposite wood. Both
phenolic and aliphatic hydroxyl group contents were
slightly higher in the compression wood samples. These
results are similar to a recent report by Onnerud and Gel-
lerstedt (2003), wherein the authors reported higher
amounts of phenolic units per 100 phenylpropane units
in the compression wood of spruce compared to normal
wood. Similarly, the amount of 5-59 etherified moieties
was higher in compression wood samples compared to
Table 3 Empirical formulas of different MWLs.
Group C9 empirical formula
Normal wood C9H8.32O3.01(OCH3)0.92
Wind opposite wood C9H8.57O3.19(OCH3)0.88
Wind compression wood C9H8.39O3.18(OCH3)0.77
Bent opposite wood C9H8.28O3.10(OCH3)0.92
Bent compression wood C9H8.76O3.26(OCH3)0.66
The C9 formula did not correct for sugar content.

Table 4 Major differences in functional groups and inter-unit linkages between five MWLs as obtained by
Structure
NW WNW
Total H units 0.05 0.04
Conjugated 0.01 0.01
Non-conjugated 0.04 0.03
Total OH 1.42 1.43
Phenolic OH 0.28 0.28
Aliphatic OH 1.14 1.15
5-59 etherified 0.10 0.12
The peak assignments and calculations are according to Capanema et al. (2004) and Balakshin et al. (2005), and references therein.
NW, normal wood; WOW, wind opposite wood; WCW, wind compression wood; BOW, bent opposite wood; BCW, bent compression
wood.
normal and opposite wood. Based on these results, it is
apparent that the lignin structure in MWL isolated from
compression wood is different from that obtained from
juvenile normal wood.
Conclusion
To resolve the question as to whether juvenile wood is
identical to compression wood, loblolly pine ramets were
planted in three different environments, an unaltered con-
trol, constrained bending and excess wind, yielding five
different wood specimens. Our results show that juvenile
wood is different from compression wood in morphology
and chemical characteristics, such as tracheid shape and
length, sugar composition, and lignin structure. Com-
pared to the compression wood produced by artificial
bending, wood grown in the windy environment can be
classified as a type of compression wood submitted to
mild mechanical stress.
Acknowledgements
We thank the NCSU Phytotron Center for the use of growth
chambers and Dr. Robert J. Downs for wind speed measure-
ments. We also thank Drs. Yuji Matsumoto, Takuya Akiyama,
Fikret Isik, and Mayuree Doungpet for their suggestions to the
manuscript. We gratefully acknowledge USDA-CSREES (IFAFS
grant 2001-52104-11224) for financial support of this research.
References
Adler, E., Brunow, G., Lundquist, K. (1987) Investigation of the
acid-catalysed alkylation of lignins by means of NMR spec-
troscopic methods. Holzforschung 41:199–207.
Akiyama, T., Sugimoto, T., Matsumoto, Y., Meshitsuka, G. (2002)
Erythro/threo ratio of b-O-4 structures as an important struc-
tural characteristic of lignin. I: Improvement of ozonation
method for the quantitative analysis of lignin side-chain
structure. J. Wood Sci. 48:210–215.
Akiyama, T., Matsumoto, Y., Okuyama, T., Meshitsuka, G. (2003)
Ratio of erythro and threo form of b-O-4 structures in tension
wood lignin. Phytochemistry 64:1157–1162.
Balakshin, M.Y., Capanema, E., Goldfarb, B., Frampton, J., Kad-
la, J.F. (2005) NMR studies on Fraser fir Abies fraseri (Pursh)
Poir lignins. Holzforschung 59:488–496.
Bendtsen, B.A. (1978) Production of wood from improved and
intensively managed trees. For. Prod. J. 28:61–72.
Comparison of juvenile wood and compression wood 673
13
C NMR spectroscopy.
Number per aromatic ring
WCW BOW BCW
0.11 0.04 0.12
0.02 0.01 0.02
0.09 0.03 0.10
1.47 1.40 1.48
0.31 0.28 0.31
1.16 1.12 1.17
0.15 0.09 0.14
Bjorkman, A. (1956) Studies on finely divided wood. Part I.
Extraction of lignin with neutral solvents. Sven. Papperstidn.
59:477–485.
Bland, D.E. (1958) The chemistry of reaction wood. Part I. the
lignin of Eucalyptus goniocalyx and Pinus radiata. Holzfor-
schung 12:36–43.
Capanema, E.A., Balakshin, M.Y., Kadla, J.F. (2004) A compre-
hensive approach for quantitative lignin characterization by
NMR spectroscopy. J. Agric. Food Chem. 52:1850–1860.
Chang, H.M., Allan, G.G. (1971). Oxidation. In: Lignin: Occur-
rence, Formation, Structure and Reactions. Eds. Sakanen,
K.V., Ludwig, C.H. Wiley-Interscience, New York. pp.
433–485.
Chen, C.L. (1992) Nitrobenzene and cupric oxide oxidation. In:
Methods in Lignin Chemistry. Eds. Lin, S.Y., Dence, C.W.
Springer-Verlag, Berlin. pp. 301–321.
Coimbra, M.A., Delgadillo, I., Waldron, K.W., Selvendran, R.
(1996) Isolation and analysis of cell wall polymers from olive
pulp. In: Modern Methods of Plant Analysis, Vol. 17. Eds.
Linskens, H.F., Jackson, J.F. Springer-Verlag, Berlin. pp.
33–44.
Dence, C.W. (1992). The determination of lignin. In: Methods in
Lignin Chemistry. Eds. Lin, S.Y., Dence, C.W. Springer-Ver-
lag, Berlin. pp. 34–35.
Fukushima, K., Taguchi, S., Matsui, N., Yasuda, S. (1997) Distri-
bution and seasonal changes of monolignol glucosides in
Pinus thunbergii. Mokuzai Gakkaishi 43:254–259.
Goto, H., Koda, K., Matsumoto, Y., Meshitsuka, G. (2001). Pre-
cise determination of methoxy content as an important indi-
cation of the extent of lignin oxidation remaining in bleached
pulp. In: Proceedings of the 11th JSWPC, Nice, France, June
11–14, 2001, v. III. pp. 417–420.
Krassig, H.A. (1993). Cellulose: Structure, Accessibility, and
Reactivity. Gordon and Breach Science, Switzerland. pp.
91–92.
Kwon, M., Bedgar, D.L., Piastuch, W., Davin, L.B., Lewis, N.G.
(2001) Induced compression wood formation in Douglas fir
(Pseudotuga menziesii) in microgravity. Phytochemistry
57:847–857.
Lee, C.L. (1961) Crystallinity of wood cellulose fiber. For. Prod.
J. 11:108–112.
Lohrasebi, H., Mabee, W.E., Roy, D.N. 1999. Chemistry and
pulping feasibility of compression wood in black spruce.
J. Wood Chem. Technol. 19:13–25.
Lundquist, K. (1980) NMR studies of lignins. 4. Investigation of
spruce lignin by 1H NMR spectroscopy. Acta Chem. Scand.
Ser. B 34:21–26.
Matsumoto, Y., Ishizu, A., Nakano, J. (1986). Studies of chemical
structure of lignin by ozonation. Holzforschung 40(Suppl.):
81–85.
Önnerud, H., Gellerstedt, G. (2003). Inhomogeneities in the
chemical structure of spruce lignin. Holzforschung 57:
165–170.
Scurfield, G. (1973) Reaction wood: its structure and function.
Science 179:647–655.
674 T.-F. Yeh et al.
Shelbourne, C.J.A., Ritchie, K.S. (1968) Relationships between
degree of compression wood development and specific
gravity and tracheid characteristics in loblolly pine. Holzfors-
chung 22:185–190.
Tanaka, F., Koshijima, T., Okuyama, T. (1981) Characterization of
cellulose in compression wood and opposite woods of a
Pinus densiflora tree grown under the influence of strong
wind. Wood Sci. Technol. 15:265–273.
Tappi standard (1988) T264om-88. Preparation of wood for
chemical analysis.
Timell, T.E. (1986) Compression Wood in Gymnosperms. Spring-
er-Verlag, Berlin.
Vonk, C.G. (1973) Computerization of Ruland’s X-ray method for
determination of the crystallinity in polymers. J. Appl. Crys-
tallogr. 6:148–152.
Wardrop, A.B., Davies, G.W. (1964). The nature of reaction
wood. VIII. The structure and differentiation of compression
wood. Aust. J. Bot. 12:24–38.
Watanabe, H., Tsutsumi, J., Kojima, K. (1963). Studies on juve-
nile wood. I. Experiments on stems of Sugi trees (Crypto-
meria japonica D. Don). Mokuzai Gakkaishi 9:225–230.
Yasuda, S., Sakakibara, A. (1975). The chemical composition of
lignin from compression wood. Mokuzai Gakkaishi 21:
363–369.
Yeh, T.F., Yamada, T., Capanema, E., Chang, H.M., Chiang, V.,
Kadla, J.F. (2005) Rapid screening of wood chemical com-
ponent variations using transmittance near-infrared spectro-
scopy. J. Agric. Food Chem. 53:3328–3332.
Yokoyama, T., Kadla, J.F., Chang, H.M. (2002) Microanalytical
method for the characterization of fiber components
and morphology of woody plants. J. Agric. Food Chem. 50:
1040–1044.
Zobel, B.J. (1975). Using the juvenile wood concept in the south-
ern pines. South. Pulp Pap. Manuf. 38:14–16.
Zobel, B.J. (1981). Wood quality from fast-grown plantation.
Tappi J. 64:71–74.
Zobel, B.J. (1984). The changing quality of the world wood sup-
ply. Wood Sci. Technol. 18:1–17.
Zobel, B.J., McElwee, R.L. (1958) Variation of cellulose in loblolly
pine. Tappi J. 41:167–170.
Zobel, B.J., Sprague, J.R. (1998) Juvenile Wood in Forest Trees.
Springer-Verlag, Berlin.
Zobel, B.J., Van Buijtenen, J.P. (1989) Wood Variation: Its Causes
and Control. Springer-Verlag, Berlin.
Received June 19, 2005. Accepted July 8, 2005.