Tree Physiology 23, 497–504
© 2003 Heron Publishing—Victoria, Canada
Climatic signal of earlywood vessels of oak on a maritime site
IGNACIO GARCÍA GONZÁLEZ1,2 and DIETER ECKSTEIN3
1
University of Santiago de Compostela, Department of Botany, EPS, E-27002 Lugo, Spain
2
Author to whom correspondence should be addressed ([email protected])
3
University of Hamburg, Institute of Wood Biology, Leuschnerstrasse 91, D-21031 Hamburg, Germany
Received April 23, 2002; accepted October 3, 2002; published online April 1, 2003
Summary Earlywood vessel lumen areas were measured in
72 consecutive tree rings in wood cores from oak (Quercus
robur L.) trees in a maritime woodland. This anatomical time
series was statistically correlated with climate data for the
same time span. There was a strong dependence of earlywood
vessel lumen area on rainfall between February and April,
which reflects the role of water availability in vessel ontogeny.
By inversion, earlywood vessel lumen areas can be used as a
proxy to reconstruct spring precipitation beyond the archived
weather records. Such information may be of value in the context of climate change.
Keywords: dendrochronology, image analysis, Quercus robur,
tree rings, vessel area.
Introduction
Information about environmental occurrences is perceived by
a tree during its entire lifetime through metabolic processes
and is encoded in the form of the width (e.g., Fritts 1976), density (e.g., Schweingruber 1988), structure (e.g., Knigge and
Schulz 1961) and stable carbon and oxygen isotopes (e.g.,
Schleser et al. 1999) of the wood that is formed annually. By
inversion, this information can be extracted from a tree and
used to reconstruct past environmental events beyond the
range of historical records, in extreme cases by hundreds and
sometimes several thousands of years. Depending on the limiting environmental factor during the growth of a tree, the climatic signal of tree ring width is mostly either summer
temperature or summer rainfall, and the climatic signal of
maximum wood density is summer temperature. The climatic
signal of the stable carbon isotope ratio in tree rings seems to
be less consistent (McCarroll and Pawellek 2001). There is
also uncertainty about the climatic significance of oxygen isotope variation in tree rings (Saurer et al. 2000).
Eckstein et al. (1977) hypothesized that size of vessels, i.e.,
water-conducting elements, of deciduous trees reflects the
availability of water during their formation. This hypothesis
has now been verified in several case studies (e.g., Woodcock
1989, Sass and Eckstein 1995, Pumijumnong and Park 1999).
However, further investigations with trees from various ecological situations are needed to assess the robustness of using
annual variability in vessel size as a proxy for dendroclimatic
reconstructions of past rainfall regimes. We report a study in
which vessel lumen areas of oak trees growing under maritime
conditions were analyzed for their climatic signal.
Materials and methods
Study site and tree sampling
The study site is at A Capela, which is located within a seminatural woodlands area of about 2300 ha in the Eume Valley in
the northwest of the Iberian Peninsula (43°24′ N, 8°03′ W).
The lower part of this valley has an unusually warm and wet
microclimate with a low frost risk, and is an important refuge
for some rare plant species (Amigo Vázquez and Norman
1995).
The study trees grow at an elevation of 200 m on a slope
with a SSW exposure. The plant community belongs to the
phytosociological association Blechno spicantis–Quercetum
roboris subassociation lauretosum nobilis (Izco et al. 1990)
and has been studied by several authors (Losa Quintana 1974,
Amigo Vázquez and Norman 1995). The tree stratum is dominated by oak (Quercus robur L.), with some scattered individuals of Castanea sativa Miller and Betula alba L. Ferns are
diverse and abundant, including species such as Blechnum
spicant L. (Roth.), Dryopteris dilatata (Hoffm.) A. Gray,
D. aemula (Aiton) O. Kuntze, Osmunda regalis L., Athyrium
filix-foemina L. (Roth.), Davallia canariensis (L.) Sm., and
several thermophile indicators (Laurus nobilis L., Arbutus
unedo L., Rubia peregrina L. and Ruscus aculeatus L.) were
also present.
Three 5-mm wood cores were taken at breast height from
bark to pith from each of 10 living oak trees. The cores were
air-dried and mounted on wooden supports in order to cut and
polish a cross-sectional surface. The cut surface was sanded
with very fine sandpaper and then rubbed with white chalk to
make the earlywood vessels clearly visible. The earlywood
and latewood ring widths were measured with a standard tree
ring measuring device. The boundary between earlywood and
latewood was distinguished by the vessel size (much larger in
earlywood than in latewood) and also by vessel distribution.
Earlywood was restricted to areas where the vessels formed a
498
GARCÍA GONZÁLEZ AND ECKSTEIN
continuous band in the tangential direction, with little surface
area occupied by other elements. Latewood and earlywood
width series were cross-dated as is usual in dendrochronology
in order to match the time series precisely on the calendar time
axis (Schweingruber 1988).
Automatic image analysis
Earlywood vessel lumen areas were measured with an automatic image analysis system (Olympus CUE-3). The system
allows automatic object recognition based on a resolution of
up to 512 × 512 pixels. Images of the wood surface were captured with a video camera attached to a binocular microscope,
using ring-lighting to obtain uniform light distribution. Images
were digitized with grey scales ranging from 0 to 255. Vessel
lumen areas were distinguished from the background tissue
based on a threshold grey level, which was adjusted during
measurement because the darkness of the wood was highly
variable.
Vessel lumen areas were measured within a frame of
4.5 mm of tangential width; its radial dimension was chosen
depending on the width of the earlywood. Frequently, features
other than vessels were “recognized” by the system as vessels.
Therefore, two filters were applied before storing the values: a
size filter and a shape filter. The minimum value of the size filter was set at 5000 µm2 (diameter = 80 µm) as this proved to be
the minimum size of an earlywood vessel; the upper value was
set at 250,000 µm2 (diameter = 565 µm) to eliminate unusually
large objects. The shape filter excluded objects of irregular
shape that were the same size as vessels. There were often unwanted objects, especially rays, that could not be excluded by
the filters. This was particularly common in heartwood, where
vesssel tyloses could not be excluded. In such cases, the image
was corrected manually.
After measuring all earlywood vessel lumina areas in each
ring of all three cores per tree, the values for the same calendar
year were pooled to obtain a time series for each tree. Thus,
about 50 vessels per year were measured. Because the vessel
time series of two of the 10 study trees showed a lack of agreement with the vessel time series of the other eight trees, they
were excluded from further analysis.
Statistical procedures
The series of the three growth variables (earlywood and latewood width and vessel lumen area) were detrended by fitting a
64-year spline function to eliminate as much non-climatically
caused variability (e.g., age-related variability (Cook et al.
1992)) as possible, using the program ARSTAN (GrissinoMayer and Fritts 1997). The same spline stiffness was applied
to all three variables to avoid differences associated with
detrending procedures. To measure the common variance in
each of the three chronologies, cross-correlations between all
possible combinations of trees (mean correlation) were calculated and averaged (Wigley et al. 1984).
The period for the final analysis was 1925–1996, which was
covered by data from most study trees. First, growth variables
were compared with each other by simple correlations and by
“Gleichläufigkeit” (Glk), a measure of the year-to-year agreement calculated as the number of times that two series show
the same upward or downward trend relative to the previous
year (Eckstein and Bauch 1969, Schweingruber 1988). Then,
the time series of the three growth variables were correlated
with contemporaneous monthly temperature and precipitation
data from the A Coruña station (60 m a.s.l.), which is located
in the same climatological sector as the study site (Martínez
Cortizas et al. 2000).
Results
The frequency distribution of all measured earlywood vessel
lumen areas (n = 35,969) was skewed to the left, with numerous vessels in the lower size classes and few in the large size
classes (Figure 1).
The time series of the three variables for the eight study
trees are shown in Figures 2a–2f before and after detrending.
In the series before detrending (Figures 2a–2c), there was a descending age trend for both width variables, whereas mean
vessel lumen area had a slightly ascending trend, at least during the first years. Only high frequency variation was retained
in the detrended series (Figures 2d–2f ), and most trees showed
parallel year-to-year variations.
Growth variables were compared with each other by simple
correlation and Glk (Table 1); in addition, the latewood width
of the previous year was also considered. The similarity between the earlywood width of the current year and the latewood width of the preceding growing season was high. On the
other hand, earlywood vessel lumen area was poorly correlated with latewood width of the previous year. Similarly, latewood width correlated only weakly with earlywood width of
the same year. The low correlation between earlywood width
and mean earlywood vessel lumen area, though statistically
significant (P < 0.001), suggests that these variables are likely
controlled by different climatic driving forces.
The common variance of the time series of each of the measured variables, which is interpreted as a measure of the common climatic signal, is given by the mean correlation (Table 2). This value was considerably higher for latewood width
and for the width of the entire tree ring than for the earlywood
variables.
To test whether the climatic signal was stronger in the large
vessels or in the small vessels, the smallest vessels from the
data pool of each tree were progressively removed and the
mean correlation calculated at each step (Figure 3). The mean
correlation remained constant until about 20% of the smallest
vessels were removed, and it decreased steadily thereafter. In
no case did the mean correlation increase in response to the removal of small vessels. Thus, most of the earlywood vessels
need to be taken into consideration to obtain the optimal common climatic signal.
Among all observed tree rings, the earlywood vessels of the
rings in 1990 are noteworthy. Most of the trees formed very
small and numerous vessels in that year (Figures 2a–2d), particularly the earlywood (Figure 4).
TREE PHYSIOLOGY VOLUME 23, 2003
CLIMATIC SIGNAL IN EARLYWOOD VESSELS OF OAK
499
Figure 1. Frequency distribution of all
measured earlywood vessel lumen areas (n = 35,969).
To establish climate–growth relationships, simple correlations were computed between the growth variables and the
meterological data for each month over the period 1925–
1996—from the previous October to the current September for
latewood width and from the previous June to the current May
for earlywood width and vessel lumen area (Table 3). Earlywood width seemed to be independent of climate, with only
precipitation in August of the previous year exhibiting a
slightly negative effect. Latewood formation, however, was favored by a moist and cool summer ((June) July to August). In
contrast, mean earlywood vessel lumen area clearly depended
on moist and cool conditions during late winter and early
spring (February to April (May)). We have been unable to
identify a biologically meaningful explanation for the statistical connection with precipitation in the previous summer (positive in July, negative in August).
Our statistical results were mirrored, and thus validated, by
a high similarity between the time series of the mean vessel lumen area and rainfall (March–April) and temperature (February–April) of the current growing season (Figure 5). We note
that rainfall and temperature are mutually dependent on each
other.
Discussion
Trees respond to environmental changes; some of these responses are transitory, whereas others are permanently expressed in the form of wood structural features and are
therefore suitable bioindicators because they can be evaluated
in retrospect. However, only those tree responses that vary
with a time resolution of one year or less are useful for dendroclimatological studies.
There are few studies on the variability in the widths of
earlywood and latewood of ring-porous trees and the driving
forces that give rise to such variability. Eckstein and Schmidt
(1974) found a stronger common climatic signal in latewood
width than in either earlywood width or in the entire tree ring
width of Q. robur. Similar findings were obtained by Zhang
(1997) for Q. robur and Q. petraea (Mattuschka) Liebl. and by
Nola (1996) for the same species and also for Q. cerris L. and
Q. pubescens Willd. Other ring-porous trees with this feature
include Fraxinus nigra (Tardif 1996), Schefflera delavayi
(Franch.) Harms (Xiong et al. 1998–1999) and Fraxinus
mandshurica Rupr. (Yasue and Funada 2001). A high positive
correlation between earlywood width and latewood width of
the preceding year, as observed by Nola (1996) and Tardif
(1996), points to a dependence of earlywood formation on the
previous year’s growing conditions. This is because ring-porous tree species initiate annual wood formation just before or
during bud break (Wareing 1951, Aloni 1995), i.e., before the
trees have started photosynthetic activity. Imagawa and Ishida
(1972), Matovic (1980) and Kitin et al. (1999) argued that
earlywood formation originates from cambial cells that divide
at the end of the previous growing season but enter winter dormancy before further differentiation. Moreover, Aloni (1991)
claims that auxin precursors, needed for the reactivation of the
cambium, accumulate during the previous growing season.
To better understand the mechanisms underlying variability
in earlywood vessel lumen area, we plotted the variability
against rainfall and temperature. The year-to-year variability
in earlywood vessel lumina area of our study oaks was significantly correlated with rainfall between February and April,
i.e., vessel cross-sectional area increased with increasing frequency of rain events and vice versa. The equally significant
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
500
GARCÍA GONZÁLEZ AND ECKSTEIN
Figure 2. Multiple plots of the three studied variables (mean vessel lumen area, earlywood and latewood width) before and after detrending. The
arrows indicate the year 1990.
negative correlation between earlywood vessel area and temperature is interpreted as an indirect influence of rainfall, because below-average temperature is often associated with
rainfall. The likelihood of rainfall influencing vessel growth
independently of temperature is discussed below.
Trees in temperate climate regions reactivate shoot growth
and cambial activity after winter dormancy, and wood formation starts only when temperature reaches a certain value in the
spring. Matovic (1980) observed that temperatures of about
7–9 °C were sufficient to initiate cambial activity in the ringporous species Fraxinus angustifolia ssp. pannonica Soó &
T. Simon. At the study site, because winters are mild (mean
Table 1. Correlation and “Gleichläufigkeit” values (Glk values; in italics) between variables.
Latewood width (previous year)
Latewood width (current year)
Earlywood width
Earlywood vessel lumen area
Latewood width
(previous year)
Latewood width
(current year)
–
38.0
66.2
56.3
–0.04
–
52.1
50.7
Earlywood width
0.67
0.25
–
62.0
TREE PHYSIOLOGY VOLUME 23, 2003
Earlywood vessel lumen area
0.22
0.02
0.38
–
CLIMATIC SIGNAL IN EARLYWOOD VESSELS OF OAK
501
Table 2. Analysis of the common signal (mean correlation between
trees) of all growth variables in the eight study trees.
Growth variable
Mean correlation between trees
Mean vessel lumen area
Earlywood width
Latewood width
Tree-ring width
0.39
0.35
0.58
0.53
monthly temperature normally not below 10 °C), cambial reactivation of oak starts around February–March, according to
phenological observations of bud break and leaf flushing.
Many ring-porous trees resume cambial activity nearly simultaneously along the whole trunk (e.g., Suzuki et al. 1996,
Schmitt et al. 2000) because of a high initial reserve of auxin
precursors distributed during the previous growing season
(Aloni 1991). However, in the ring-porous Fraxinus mandshurica, vessel formation is reported to progress down the
stem (Funada et al. 2001).
Xylem formation in ring-porous trees begins with the differentiation of earlywood vessel elements from overwintering
cambium derivatives and their subsequent expansion (e.g.,
Zasada and Zahner 1969, Imagawa and Ishida 1972, Kitin et
al. 1999). When newly formed vessels start to grow, the thin
and unlignified primary cell wall is loosened by growth regulators and stretched by turgor pressure, which in turn depends
on the uptake of water provided mostly by contemporaneous
rainfall. This expansion ceases with the development of the
secondary vessel wall and its lignification. For a single earlywood vessel, this process may take 3 to 4 weeks (authors’
unpublished observations). This is when water availability is
“recorded” by the tree. However, the onset and duration of
earlywood vessel formation varies annually, within limits, so
the exact timing of the formation of a certain row of vessels
cannot be determined. Moreover, there is a lag between the environmental stimulus, i.e., the change in water supply, and the
Figure 3. Variation of the common signal (mean correlation between
trees) following progressive removal of the smallest vessels at any
given step.
Figure 4. Core surface with tree rings of 1988 to 1991. Arrow indicates the very small earlywood vessels in 1990.
response of the tree. In a study of spruce, von Wilpert (1991)
determined a lag of 12 days between changes in soil water potential and the reaction of developing tracheids.
Although rainfall is usually abundant in March–April at our
study site (Martínez Cortizas et al. 2000), the steep slope of the
ground can result in rapid run-off, leading to a water deficit for
the oak trees. A water deficit not only directly affects vessel
size, but also acts indirectly by modifying hormonal concentrations (Kozlowski and Pallardy 1997). According to Aloni
(2001), high auxin concentrations result in narrow vessels because rapid differentiation allows only limited time for cell
growth, whereas low auxin concentrations cause slow differentiation permitting more cell expansion before secondary
wall deposition, resulting in wide vessels.
The abnormally narrow earlywood vessels in the 1990 tree
rings are probably the result of a severe winter–spring drought.
Fernández de Ana et al. (1996) found that the 1988–1992 period was characterized by an extreme reduction in rainfall and
an increase in maximum and minimum temperatures during
winter. This is especially true for January–March 1990. In
other cases, similar very small earlywood vessels have been
associated with intense frost events in winter or at the beginning of spring, as was first suspected by Fletcher (1975) for the
year 1437 A.D., and later confirmed by Leuschner and
Schweingruber (1996). Vessels with below-average size have
also been observed in roadside trees exposed to de-icing salt
sprinkled on the streets during winter (Eckstein et al. 1976).
Small vessels may also occur in years following heavy defoliation (Asshoff et al. 1998–1999). Both de-icing salt and insect
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
502
GARCÍA GONZÁLEZ AND ECKSTEIN
Table 3. Correlation of each growth variable with monthly meteorological data. Abbreviations: P = precipitation and T = temperature. Values in
bold are significant at P < 0.01.
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Vessel lumen area
1925–1974
T
P
1930–1979
T
P
1935–1984
T
P
1940–1989
T
P
1945–1994
T
P
1925–1996
T
P
–0.024
0.194
–0.095
0.215
–0.152
0.256
–0.156
0.214
–0.289
0.178
–0.206
0.172
0.065
0.195
–0.044
0.260
–0.104
0.313
–0.030
0.308
–0.165
0.351
–0.054
0.299
–0.002
–0.390
–0.002
–0.365
–0.065
–0.285
–0.111
–0.307
–0.225
–0.277
–0.102
–0.314
0.055
0.153
0.058
0.060
0.011
–0.071
0.056
–0.116
0.183
–0.057
0.118
0.084
0.191
–0.057
0.197
–0.069
0.234
–0.112
0.226
–0.108
0.186
–0.131
0.183
–0.050
0.034
0.062
0.119
0.007
–0.001
0.057
–0.081
0.099
–0.187
0.046
–0.102
0.135
–0.188
–0.197
–0.091
–0.040
–0.094
–0.109
–0.004
–0.037
–0.243
0.003
–0.259
–0.059
–0.146
–0.049
–0.136
0.021
–0.147
0.013
–0.206
–0.004
–0.224
0.040
–0.172
–0.004
–0.269
0.122
–0.287
0.184
–0.254
0.251
–0.261
0.280
–0.406
0.230
–0.345
0.184
–0.494
0.359
–0.531
0.385
–0.356
0.319
–0.362
0.266
–0.445
0.354
–0.384
0.399
–0.409
0.262
–0.454
0.283
–0.487
0.252
–0.526
0.259
–0.410
0.182
–0.336
0.225
–0.205
–0.006
–0.252
0.013
–0.195
0.008
–0.242
0.008
–0.299
–0.014
–0.226
0.033
Earlywood width
1925–1974
T
P
1930–1979
T
P
1935–1984
T
P
1940–1989
T
P
1945–1994
T
P
1925–1996
T
P
0.133
0.034
0.145
0.040
0.054
0.064
–0.062
0.149
–0.065
0.094
0.073
–0.070
–0.051
0.176
–0.040
0.069
–0.077
–0.014
–0.057
–0.062
–0.051
0.007
0.007
0.120
–0.125
–0.278
–0.065
–0.268
0.036
–0.282
–0.007
–0.314
–0.028
–0.312
–0.038
–0.328
–0.133
–0.015
0.033
–0.162
0.035
–0.075
0.155
–0.100
0.274
–0.135
0.041
–0.056
0.091
–0.086
0.012
–0.019
–0.015
0.031
–0.125
0.059
0.082
–0.078
0.159
–0.059
0.093
–0.056
0.064
0.059
–0.042
0.021
–0.111
0.091
–0.089
0.088
0.035
0.123
0.058
–0.208
0.004
–0.113
–0.050
–0.147
0.038
–0.128
0.065
–0.086
0.109
–0.081
–0.114
–0.120
0.007
–0.114
–0.051
–0.161
–0.079
–0.134
–0.119
–0.087
–0.028
–0.074
–0.307
0.131
–0.317
0.034
–0.288
0.032
–0.225
0.047
–0.196
0.023
–0.186
0.096
–0.310
0.094
–0.155
0.105
–0.137
0.127
–0.135
0.148
–0.154
0.203
–0.134
0.135
–0.418
0.191
–0.192
0.049
–0.209
–0.010
–0.099
0.012
–0.024
–0.070
–0.049
0.034
–0.326
0.101
–0.266
0.141
–0.172
0.058
–0.221
0.075
–0.088
0.001
–0.093
0.086
Latewood width
1925–1974
T
P
1930–1979
T
P
1935–1984
T
P
1940–1989
T
P
1945–1994
T
P
1925–1996
T
P
–0.119
–0.137
–0.029
–0.133
–0.028
–0.061
0.022
–0.102
0.085
–0.161
0.029
–0.152
–0.003
–0.019
0.141
–0.030
0.116
–0.034
0.096
–0.209
0.190
–0.011
0.172
0.020
0.167
0.233
0.168
0.207
0.147
0.152
0.032
0.118
–0.127
0.090
0.020
0.078
0.120
–0.022
0.096
0.169
0.033
0.120
–0.001
0.081
–0.130
0.170
–0.016
0.190
0.005
0.032
0.062
0.054
0.007
0.054
–0.116
–0.048
–0.171
0.075
–0.117
0.085
0.047
0.067
–0.042
0.114
–0.034
0.063
–0.011
0.100
–0.136
0.206
–0.059
0.233
–0.020
–0.289
0.062
–0.186
–0.005
–0.137
0.003
–0.067
–0.047
0.020
–0.004
–0.038
–0.122
0.029
–0.012
0.151
0.025
0.089
0.098
0.140
0.082
0.004
0.029
0.013
0.068
0.111
0.121
0.262
0.111
0.278
0.082
0.143
–0.019
0.094
0.042
0.016
–0.139
0.251
–0.153
0.226
–0.158
0.219
–0.182
0.241
–0.214
0.127
–0.165
0.123
–0.435
0.455
–0.407
0.354
–0.447
0.246
–0.316
0.143
–0.299
0.283
–0.246
0.322
–0.293
0.159
–0.311
0.133
–0.246
0.120
–0.249
0.111
–0.235
–0.015
–0.144
–0.037
defoliation may result in small earlywood vessels by causing a
water deficit in the cambium zone.
In conclusion, we have demonstrated that the year-to-year
variability in earlywood vessel lumen area of oak contains a
strong ecophysiological signal; however, the metabolic cause–
effect relationships are not yet fully understood. This signal reflects the water status during vessel ontogeny from the time
that the cambial derivative is determined to become a vessel up
to the time of lignification of the cell wall. This observation is
important because the climatic signal in the easy to measure
tree ring widths of temperate oak is weak and uncertain and
mostly reflects temperature rather than rainfall. Moreover,
vessel lumen area time series contain little autocorrelation and
show a greater stability through time than tree-ring width series (García González 2000). Latewood vessel lumen areas of
ring-porous trees may provide even more information than
earlywood vessels because of their greater variability, as
reported for oak by Woodcock (1989) and for teak by Pumijumnong and Park (1999). The establishment of a chronology
of vessel lumen area of several hundred years would enable the
reconstruction of a long-term water availability record for the
northwestern Iberian Peninsula, where climate data are scarce
(Martínez Cortizas et al. 2000). We conclude that vessel lumen
area chronologies could provide an important source of palaeoclimatic information of relevance to the ongoing discussion of global change.
TREE PHYSIOLOGY VOLUME 23, 2003
CLIMATIC SIGNAL IN EARLYWOOD VESSELS OF OAK
503
Figure 5. Parallel run between
mean vessel lumen area (—)
and rainfall (- - 䉬 - -) from
March to April (top panel) and
temperature from February to
April (bottom panel). The temperature curve is reversed because of negative correlation.
References
Aloni, R. 1991. Wood formation in deciduous hardwood trees. In
Physiology of Trees. Ed. A.S. Raghavendera. John Wiley & Sons,
New York, pp 175–197.
Aloni, R. 1995. The induction of vascular tissues by auxin and cytokinin. In Plant Hormones. Ed. P.J. Davis. Kluwer Academic Publishers, Dordrecht, pp 531–546.
Aloni, R. 2001. Foliar and axial aspects of vascular differentiation:
hypothesis and evidence. J. Plant Growth Regul. 20:22–34.
Amigo Vázquez, J. and G. Norman. 1995. Identification of site-types
important for rare ferns in an area of deciduous woodland in northwest Spain. Vegetatio 116:133–146.
Asshoff, R., F.H. Schweingruber and B. Wermelinger. 1998–1999.
Influence of a gypsy moth (Lymantria dispar L.) outbreak on radial
growth and wood anatomy of Spanish chestnut (Castanea sativa
Mill.) in Ticino (Switzerland). Dendrochronologia 16–17:
133–145.
Cook, E., K. Briffa, S. Shiyatov and V. Mazepa. 1992. Tree-ring standardization and growth trend estimation. In Methods of Dendrochronology. Applications in the Environmental Sciences. Eds. E.R.
Cook and L.A. Kairiukštis. International Institute for Applied Systems Analysis, Kluwer Academic Publishers, pp 104–123.
Eckstein, D. and J. Bauch. Beitrag zur Rationalisierung eines dendrochronologischen Verfahrens und zur Analyse seiner Aussagesicherheit. Forstwiss. Centralbl. 88:230–250.
Eckstein, D. and B. Schmidt. 1974. Dendroklimatologische Untersuchungen an Stieleichen aus dem maritimen Klimagebiet
Schleswig-Holsteins. Angew. Bot. 48:371–383.
Eckstein, D., W. Liese and N. Parameswaran. 1976. On the structural
changes in wood and bark of a salt-damaged horsechestnut tree.
Holzforschung 30:173–178.
Eckstein, D., E. Frisse and F. Quiehl. 1977. Holzanatomische Untersuchungen zum Nachweis anthropogener Einflüsse auf die Umweltbedingungen einer Rotbuche. Angew. Bot. 51:47–56.
Fernández de Ana, F.J., M.J. Lombardero and A. Martínez Cortizas.
1996. Nuevas aportaciones sobre el defoliador Altica quercetorum
Foudras, 1860 (col. Chrysomelidae) en Galicia. Investigación
Agraria. Sistemas y Recursos Forestales 5:275–305.
Fletcher, J.M. 1975. Relation of abnormal earlywood in oaks to dendrochronology and dendroclimatology. Nature 254:506–507.
Fritts, H.C. 1976. Tree rings and climate. Academic Press, New York,
567 p.
Funada, R., T. Iwatate-Suzuki, J. Utsumi, T. Suzuki, P. Kitin and J.
Sano. 2001. The formation of new vessel elements in the ring-porous hardwoods in springtime in relation to water transport. International Conf. Tree Rings and People, Davos, Abstracts 30.
García González, I. 2000. Estudio dendroecológico de Quercus robur
L. en el norte de Galicia. Ph.D. Diss. Univ. Santiago de Compostela, 297 p.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
504
GARCÍA GONZÁLEZ AND ECKSTEIN
Grissino-Mayer, H. and H.C. Fritts. 1997. The International TreeRing Data Bank: an enhanced global database serving the global
scientific community. Holocene 7:235–238.
Imagawa, H. and S. Ishida. 1972. Study on the wood formation in
trees. III. Occurrence of the overwintering cells in the cambial zone
in several ring-porous trees. Res. Bull. Coll. Exp. For. Hokkaido
Univ. 29:207–211.
Izco, J., J. Amigo and J. Guitián. 1990. Los robledales Galaico-Septentrionales. Acta Bot. Malacit. 15:267–276.
Kitin, P., R. Funada, Y. Sano, H. Beeckman and J. Ohtani. 1999. Variations in the lengths of fusiform cambial cells and vessel elements
in Kalopanax pictus. Ann. Bot. 84:621–632.
Knigge, W. and H. Schulz. 1961. Einfluss der Jahreswitterung 1959
auf Zellartenverteilung, Faserlänge und Gefässweite verschiedener
Holzarten. Holz Roh Werkst. 19:293–303.
Kozlowski, T.T. and S.G. Pallardy. 1997. Growth control in woody
plants. Academic Press, San Diego, 641 p.
Leuschner, H.H. and F.H. Schweingruber. 1996. Dendroökologische
Klassifizierung und Auswertung häufig auftretender intraannueller
holzanatomischer Merkmale bei Eichen und Kiefern. Dendrochronologia 14:273–285.
Losa Quintana, J.M. 1974. Vegetación y flora de la Baja Cuenca del
río Eume. Estudio botánico y fitosociológico del bosque natural de
Quercus robur L. Boll. Soc. Brot. 47:243–267.
Martínez Cortizas, A., F. Castillo Rodríguez and R. Blanco Chao.
2000. Precipitación. In Atlas Climático de Galicia. Eds. A. Martínez Cortizas and A. Pérez Alberti. Consellería de Medio
Ambiente, Xunta de Galicia, pp 75–92.
Matovic, A. 1980. Cambium activity and xylem differentiation in
Fraxinus angustifolia Vahl ssp. pannonica Soó & Simon. Drev.
Vysk. 25:25–43.
McCarroll, D. and F. Pawellek. 2001. Stable carbon isotope ratios of
Pinus sylvestris from northern Finland and the potential for extracting a climate signal from long Fennoscandian chronologies. Holocene 11:517–526.
Nola, P. 1996. Climatic signal in earlywood and latewood of deciduous oaks from northern Italy. In Tree Rings, Environment and
Humanity. Eds. J.S. Dean, D.M. Meko and T.W. Swetnam. Radiocarbon. Dept. of Geosciences, Univ. Arizona, Tucson, pp 249–258.
Pumijumnong, N. and W.-K. Park. 1999. Vessel chronologies from
teak in northern Thailand and their climatic signal. IAWA J. 20:
285–294.
Sass, U. and D. Eckstein. 1995. The variability of vessel size in beech
(Fagus sylvatica L.) and its ecophysiological interpretation. Trees
9:247–252.
Saurer, M., P. Cherubini and R. Siegwolf. 2000. Oxygen isotopes in
tree rings of Abies alba: the climatic significance of interdecadal
variations. J. Geophys. Res. 105:12,461–12,470.
Schleser, G.H., G. Helle, A. Lücke and H. Vos. 1999. Isotope signals
as climate proxies: the role of transfer functions in the study of terrestrial archives. Quat. Sci. Rev. 18:927–943.
Schmitt, U., R. Möller and D. Eckstein. 2000. Seasonal wood formation of beech (Fagus sylvatica L.) and black locust (Robinia
pseudoacacia L.) as determined by the ‘pinning’-technique. J.
Appl. Bot. 74:10–16.
Schweingruber, F.H. 1988. Tree rings: basics and applications of
dendrochronology. Kluwer Academic Publishers, Dordrecht,
Netherlands, 276 p.
Suzuki, M., K. Yoda and H. Suzuki. 1996. Phenological comparison
of the onset of vessel formation between ring-porous and diffuse-porous deciduous trees in a Japanese temperate forest. IAWA
J. 17:431–444.
Tardif, J. 1996. Earlywood, latewood and total ring width of a ringporous species (Fraxinus nigra Marsh) in relation to climatic and
hydrologic factors. In Tree Rings, Environment and Humanity.
Eds. J.S. Dean, D.M. Meko and T.W. Swetnam. Radiocarbon.
Dept. of Geosciences, Univ. Arizona, Tucson, pp 315–324.
von Wilpert, K. 1991. Intraannual variation of radial tracheid diameters as monitor of site specific water stress. Dendrochronologia 9:
95–113.
Wareing, P.F. 1951. Growth studies in woody species. IV. The initiation of cambial activity in ring-porous species. Physiol. Plant. 4:
546–562.
Wigley, T.M.L., K.R. Briffa and P.D. Jones. 1984. On the average
value of correlated time series, with applications in dendroclimatology and hydrometeorology. J. Clim. Appl. Meteorol. 23:
201–213.
Woodcock, D.W. 1989. Climate sensitivity of wood anatomical features in a ring-porous oak (Quercus macrocarpa). Can. J. For. Res.
19:639–644.
Xiong, L., N. Okada, Z. Zhangcheng and T. Fujiwara. 1998–1999.
The relationship between Schefflera delavayi growth and climate
in the Three Gorges reserve region of the Yangtze River, China.
Dendrochronologia 16/17:99–118.
Yasue, K. and R. Funada. 2001. The effects of climate on variation in
early- and latewood width of Japanese ash in northern Hokkaido,
Japan. International Conf. Tree Rings and People, Davos, Abstracts 31.
Zasada, J.C. and R. Zahner. 1969. Vessel element development in the
earlywood of red oak (Quercus rubra). Can. J. Bot. 47:1965–1971.
Zhang, S.Y. 1997. Variations and correlations of various ring width
and ring density features in European oak: implications in
dendroclimatology. Wood Sci. Technol. 31:63–72.
TREE PHYSIOLOGY VOLUME 23, 2003