1321
Leaf characteristics reflect growth rates of
2-year-old Populus trees
Constance A. Harrington, M.A. Radwan, and Dean S. DeBell
Abstract: We examined the relationships between biomass or growth rates and leaf characteristics of 2-year-old trees of
two clones of Populus. Leaf characteristics were total plant leaf area or leaf weight, mean size (or weight) of fully
expanded terminal leaves, and foliar concentrations and contents of N, P, K, Ca, Mg, total chlorophyll, and total
available carbohydrates. Sample trees (n = 156) were chosen from two irrigation regimes and several fertilization
treatments to provide a wide range of environmental conditions and growth rates for each clone. Total plant leaf area or
weight was strongly correlated with total aboveground biomass (r = 0.98–0.99); however, mean size (area or weight) of
the fully expanded terminal leaves was also quite strongly correlated with biomass (r = 0.64-0.72), height growth (r =
0.54–0.72), and diameter growth (r = 0.53–0.73). With one exception (correlation between foliar K concentration and
height growth of one clone, r = 0.67), leaf size characteristics were more strongly correlated with biomass or growth than
were concentrations or contents of foliar chemicals. Since size of the terminal leaves is easy to measure, it may be useful
as a simple indicator of potential productivity.
Resumé: Nous avons examiné les relations entre la biomasse ou le taux de croissance et les caractéristiques des feuilles
chez des arbres de deux ans appartenant à deux clones de Populus. Les caractéristiques des feuilles comprenaient la
surface foliaire ou le poids des feuilles des plants au complet, la dimension moyenne (ou le poids) des feuilles terminales
pleinement déployées ainsi que la concentration et le contenu en N, P, K, Ca, Mg, la chlorophylle totale et les hydrates
de carbone totaux disponibles. Les arbres échantillonnés (n = 156) avaient été soumis à deux régimes d’irrigation et à
plusieurs traitements de fertilisation de telle sorte qu’ils représentaient tine vaste gamme de conditions
environnementales et de taux de croissance dans chaque clone. Le poids ou la surface foliaire des plants entiers était
fortement correlé avec la biomasse épigée totale (r = 0,98–0,99). Cependant, la dimension moyenne (en surface ou en
poids) des feuilles terminales pleinement déployées était également fortement corrélée avec la biomasse (r = 0,64-0,72),
la croissance en hauteur (r = 0,54-0,72) et la croissance en diamètre (r = 0,53–0,73). À l’ exception de la corrélation
entre la concentration de K foliaire et la croissance en hauteur chez un clone (r = 0,67), les caractéristiques
dimensionnelles des feuilles étaient plus fortement corrélées avec la biomasse ou la croissance que la concentration ou le
contenu en éléments minéraux foliaires. Étant donné que la dimension des feuilles terminales est facile à mesurer, elle
petit être utile comme indicateur simple de la productivité potentielle. [Traduit par la Rédaction]
Introduction
It has long been recognized that plants growing
under substantial soil moisture or nutrient stress
have smaller leaves and lower growth rates than
plants of the same genotype growing under more
favorable conditions. In addition, rapid produc­
tion of leaf area appears to be an important
attribute of fast-growing plants. Previous work
on Populus has shown that (1) mean leaf size per
clone and clonal performance are correlated
(Ridge et al. 1986; Isebrands et al. 1988;
Ceulemans 1990), (2) mean leaf size per clone
increases as ortet location becomes more mesic
and the expression of this relationship is greater
at a more mesic test site (Dunlap et al. 1995),
Received October 18, 1996. Accepted March 31, 1997.
C.A. Harrington,1 M.A. Radwan (retired),
and D.S. DeBell. Pacific Northwest Research
Station, Forestry Sciences Laboratory, 3625 93rd
Ave SW, Olympia, WA 98512-9193, U.S.A.
1
and (3) mean leaf size and leaf growth rates were
greater in irrigated than in nonirrigated trees
(Roden et al. 1990). This past work generally
involved relatively few samples (two to five
trees) per clone and did not examine withinclone variation in leaf size and productivity.
In this study, we examined the relationships
between growth rates or attained size and leaf
characteristics of 2-year-old trees of two
Populus clones. We concentrated on evaluating
the size, weight, and selected chemical
characteristics of the fully expanded leaves
produced on the current terminal shoot. In
young, fast-growing, short-rotation, intensively
cultured plantings, leaves on the current terminal
are considered to be the most important suppliers
of photosynthate for height and diameter growth
(Isebrands and Nelson 1983). In addition, we
examined the relationship of total plant leaf area
or weight to total aboveground biomass,
diameter growth, and height growth.
Author to whom all correspondence should be addressed.
© 1997 NRC Canada
1322
Materials and methods
Unrooted cuttings of two Populus clones were
planted at 2 x 2 m spacing in alternate rows in a
0.7-ha block near Olympia, Wash., U.S.A. The
clones were 11-11, a Populus deltoides Bartr. ex
Marsh × Populus trichocarpa Torr. & Gray
hybrid, and 7-75, a Populus trichocarpa selection
from a natural stand near Orting, Wash.
(approximately 40 km from the study area). Both
clones were developed (or selected) by the
University
of
Washington/Washington
State.University Poplar Research Program
(Quinsey et al. 1991). The study area is relatively
flat, elevation is 50 m, and the soil is Nisqually
loamy fine sand (sandy, mixed, mesic Pachic
Xerumbrepts), a very deep, somewhat excessively
drained soil that formed in sandy glacial out-wash
(USDA Soil Conservation Service 1990). Four
plots in each of three contiguous blocks were
randomly assigned to an irrigation regime (low or
high) and four-tree subplots were assigned to
fertilization treatments that applied varying
amounts of N (0–500 kg NAha-1 as ammonium
nitrate), P (0–1000 kg PAha-1 as triple
superphosphate), K (0–1000 kg KAha-1 as muriate
of potash), and lime (0–10 Mg limeAha-1). The
fertilization subplot treatments were laid out in a
continuous function design (Shoulders and Tiarks
1983). Irrigation was applied with a drip system;
during the second growing season (May 1
through August 30), the low regime received 14
cm and the high regime 51 cm of rainfall plus
irrigation.
Samples for this study were collected from
selected fertilizer sub-plot treatments to represent
the range of nutrient environments established in
the plantation. Equal numbers of trees per clone
and irrigation regime were sampled in each
treatment to provide a balanced data set (n = 156;
39 for each clone and irrigation regime). The data
set also encompassed the range of tree sizes and
current growth rates occurring in the plantation.
Fully expanded terminal leaves (leaf plastochron
index ≥ 6 as per Larson and Isebrands 1971) were
collected from 2-year-old trees during the last
week of August. Most of the annual growth and
nutrient uptake had occurred and the foliage had
not yet deteriorated. Sampling was done in early
Can. J. For Res. Vol. 27, 1997
morning and consisted of three or four leaves per
sample tree.
Immediately after harvest, fresh weight, leaf
area, and number of leaves per sample tree were
determined. Chlorophyll a and b were extracted
from the blade portions of a subsample of fresh
leaves by maceration in 80% acetone, optical
densities measured spectrophotometrically, and
contents computed according to Amon (1949).
Remaining leaf samples (blades and petioles)
were dried to constant weight at 65°C, ground to
40 mesh, and then analyzed for total N (including
nitrate) by the micro-Kjeldahl procedure
(Bremner and Mulvaney 1982), P by the
molybdenum blue technique (Chapman and Pratt
1961), and K, Ca, and Mg by atomic absorption
(Perkin-Elmer Corporation 1976). Determination
of total available carbohydrates was done by
extraction and hydrolysis in sulfuric acid (Smith
et al. 1964). Concentrations of total available
carbohydrate were calculated as percent glucose
in ovendried leaf tissue.
Tree height and basal diameter (0.15 m) were
measured at the end of the first growing season
and in the second growing season when the leaves
were sampled. The plantation was thinned the
same week the terminal leaves were collected and
total aboveground biomass was determined for
each tree by weighing the entire aboveground
portion of the plant. In addition, half of these
thinned trees were partitioned into stem,
branches, and leaves and component weights and
leaf areas were determined. Subsamples were
taken to determine moisture content and fresh
weight/dry weight relationships. For the trees that
were not partitioned, total plant leaf area and leaf
weight were predicted from total plant weight
using regression equations developed for each
clone and irrigation regime.
Data analysis
Analyses were done to explore relationships
among leaf characteristics and tree productivity
rather than to test specific models or determine
significant differences among discrete fertilizer
treatments. Lateral root development was rapid
and roots of trees planted in one subplot treatment
commonly extended into adjacent fertilizer and
irrigation treatments. Thus, individual tree
characteristics did not reflect responses to distinct
© 1997 NRC Canada
treatments; however, the treatments provided an
extensive variety of growing conditions that
resulted in a wide range of tree sizes, growth
rates, and leaf characteristics.
Plottings of tree biomass or growth (height and
diameter growth during the second growing
season) versus leaf characteristics were examined.
Nonlinear relationships between variables were
not apparent and data transformations were
assumed unnecessary for subsequent analyses.
Simple correlation coefficients were determined
between the biomass or growth variables and the
leaf characteristics for each clone and irrigation
regime. If differences in the relationships
observed between irrigation regimes or clones
were nonsignificant or minimal, the data were
pooled and the correlations determined for the
combined observations. Only correlations for
variable combinations for which at least one of
the clones had an r ≥ 0.60 are presented. Simple
and multiple regression equations using leaf size
and chemical characteristics were examined for
their ability to predict second-year height growth.
The biomass and growth variables were also
examined using analysis of variance with clone,
irrigation, and clone by irrigation as the model
sources of variation.
Results
Mean growth of both clones in the study
plantation was good, but due to the imposed
range in nutrient conditions, plant biomass and
growth rates varied substantially within each
clone and irrigation regime (Table 1). Mean total
aboveground biomass, height growth, and
diameter growth were significantly greater for
clone 11-11 than for clone 7-75 and greater for
the high-versus the low-irrigation regime.
Clone 11-11 had greater mean area and weight
of terminal leaves and greater total plant leaf area
(or weight) than clone 7-75 (Table 2). Leaf length
was similar, but shape of the leaves differed
between clones, with 11-11 having a broader leaf
base and a more deltoid shape. Clone 7-75 had
higher mean concentrations of N, P, K, and
chlorophyll and lower concentrations of Ca than
clone 11-11, but concentrations of Mg and total
available carbohydrate were similar in both
clones. On average, high irrigation increased
mean area and weight of terminal leaves, total
plant leaf area, and P concentrations but
decreased concentrations of most other nutrients
and chlorophyll. Total available carbohydrates
were essentially unaffected by irrigation regime.
K levels in clone 11-11 were also unaffected by
irrigation; in clone 7-75, however, K levels were
substantially greater in the high-irrigation regime.
Mean area and weight of terminal leaves were
positively correlated with total aboveground
biomass, height growth, and diameter growth of
both clones (Table 3). The correlations between
mean weight of terminal leaves and total biomass
or growth variables were similar to those for
mean terminal leaf area because area and weight
of terminal leaves were very strongly correlated
(r = 0.97 for clone 11-11; r = 0.89 for clone 7­
75). Correlations between total plant leaf area or
leaf weight and total biomass were stronger than
those of terminal leaf traits with total biomass or
diameter; however, correlations of total plant leaf
traits with height growth were weaker than those
of terminal leaf traits with height growth.
Foliar K concentrations were correlated with
biomass and growth of clone 7-75, but not with
biomass or growth of clone 11-11 (Table 3).
Foliar concentrations or contents (i.e., con­
centration multiplied by mean leaf weight) of
most nutrients, chlorophyll, and available
carbohydrates were significantly correlated (P <
0.05) with total biomass and growth of both clones.
In all cases except K concentration and height growth
of clone 7-75, however, correlations between
biomass or growth and mean area and weight of
terminal leaves or total leaves were substantially
higher than correlations with concentrations or
contents of nutrients and chlorophyll. For clone 7-75,
the multiple regression equation predicting height
growth that included K concentration and mean
terminal leaf area (R2 = 0.53) accounted
© 1997 NRC Canada
Notes
1323
© 1997 NRC C
1324
Fig. 1. Relationship between mean area of fully
expanded terminal leaves and second-year height
growth for clones 11-11 and 7-75. For clone 11­
11, one regression relationship was sufficient for
both low- and high-irrigation regimes (y = 0.0043x
+ 0.8). For clone 7-75, the low-irrigation regime
could be described by the same equation as for
clone 11-11, but the high-irrigation regime was
best described by y = 0.0014x + 2.1.
for substantially more variation than the equation
using only mean terminal leaf area (R2 = 0.39).
For clone 11-11, none of the multiple regression
equations that included additional foliar variables
increased R2 values more than 0.02 over the
equation with mean terminal leaf area as the
independent variable.
Relationships between mean terminal leaf area
and height growth were similar for high- and lowirrigation regimes of clone 11-11, despite the fact
that trees in the high-irrigation regime generally
had much larger leaves (Fig. 1). Moreover, the
relationship between leaf area and height growth
for trees of clone 7-75 in the low-irrigation regime
(Fig. 1) was essentially identical to the
relationship for clone 11-11. The relationship for
trees of clone 7-75 in the high-irrigation regime,
however, was not as strong (i.e., the slope of the
line was not as steep). Maximum area per terminal
leaf was clearly lower for clone 7-75 than for
clone 11-11; none of the clone 7-75 trees had
Can. J. For. Res. Vol. 27, 1997
mean leaf areas >500 cm2 whereas 14% of clone
11-11 had leaf areas exceeding that size.
Discussion
A strong correlation between total leaf area or
total leaf weight and tree productivity (size or
growth) has been reported previously (Larson and
Isebrands 1971; Larson et al. 1976) as have high
positive correlations between mean size of mature
leaves and relative growth of various Populus
clones (Ridge et al. 1986; Isebrands et al. 1988;
Ceulemans 1990). The pre-sent study expands on
these general relationships to demonstrate a very
simple, yet strong correlation between mean size
of the fully expanded terminal leaves and
productivity of two Populus clones. Our findings
apply to within-clonal differences in productivity
(size and growth) “created” by manipulating
several growth factors: N, P, K, and lime
amendments and water availability.
Despite the range of nutritional status affecting
tree size and growth and significant correlations
between production variables and chemical
concentrations or contents, mean size of terminal
leaves (area or weight) was more strongly
correlated with productivity than concentration or
content of any single chemical with the one
exception of K concentration and height growth in
clone 7-75. Thus, this simple, easy-to-measure
characteristic may be a very useful indicator of
potential productivity or future growth. It merits
further testing as a possible tool to aid in site
selection, match clones to sites, monitor tree
response to cultural treatments or provide an early
indicator of the relative performance of various
clones.
Acknowledgments
This work was supported and coordinated by the
Short Rotation Woody Crops Program (now
Biofuels Feedstock Development Program of the
U.S. Department of Energy through Interagency
Agreement DE-A105-810R20914.
References
Arnon, D.I. 1949. Copper enzymes in isolated
chloroplasts. Polyphenoloxidase in Beta
vulgaris. Plant Physiol. 24: 1-15.
Bremner, J.M., and Mulvaney, C.S. 1982.
Nitrogen—total. In Methods of soil analysis,
part 2. Agronomy (Madison), 9: 595-624.
Ceulemans, R. 1990. Genetic variation in
© 1997 NRC Canada
Notes
functional
and
structural
productivity
determinants in poplar. Thesis Publishers,
Amsterdam, The Netherlands.
Chapman, H.D., and Pratt, P.F. 1961. Methods of
analysis for soils, plants, and waters. Division of
Agricultural Science, University of California,
Berkeley, Calif.
Dunlap, J.M., Heilman, P.E., and Stettler, R.F.
1995. Genetic variation and productivity of
Populus trichocarpa and its hybrids. VIII. Leaf
and crown morphology of native P. trichocarpa
clones from four river valleys in Washington.
Can. J. For. Res. 25: 1710-1724.
Isebrands, J.G., and Nelson, N.D. 1983.
Distribution of 14C-labeled photosynthates
within intensively cultured Populus clones
during the establishment year. Physiol. Plant.
59: 9-18.
Isebrands, J.G., Ceulemans, R., and Wiard, B.
1988. Genetic variation in photosynthetic traits
among Populus clones in relation to yield. Plant
Physiol. Biochem. 26: 427-437.
Larson, P.R., and Isebrands, J.G. 1971. The
plastochron index as applied to developmental
studies in cottonwood. Can. J. For. Res. 1: 1-11.
Larson, P.R., Dickson, R.E., and Isebrands, J.G.
1976. Some physiological applications for
intensive culture. In Intensive plantation culture:
five years research. USDA For. Serv. Gen.
Tech. Rep. NC-21. pp. 10-18.
Perkin-Elmer Corporation. 1976. Analytical
methods
of
atomic
absorption
spectrophotometry. Perkin-Elmer Corporation,
1325
Norwalk, Conn.
Quinsey, S., Stettler, R., Heilman, P., Delany, D.,
Fenn, R., Ager, A., and Glackin, P. 1991. Clone
register. University of Washington/Washington
State University Poplar Research Program.
College of Forest Resources, University of
Washington, Seattle, Wash.
Ridge, C.R., Hinckley, T.M., Stettler, R.F., and
Van Volkenburgh, E. 1986. Leaf growth
characteristics of fast-growing poplar hybrids
Populus trichocarpa × P. deltoides. Tree
Physiol. 1: 209-216.
Roden, J., Van Volkenburgh, E., and Hinckley,
T.M. 1990. Cellular basis for limitation of
poplar leaf growth by water deficit. Tree
Physiol. 6: 211-219.
Shoulders, E
., and Tiarks, A.E. 1983. A
continuous function design for fertilizer rate
trials in young pine plantations. In Proceedings,
Second Biennial Southern Silvicultural
Research Conference. Edited by E.P. Jones, Jr.
U.S. Dep. Agric. For. Serv. Gen. Tech. Rep.
SE-24. Southeastern Forest Experiment Station,
Atlanta, Ga. pp. 252-256.
Smith, D., Paulsen, G.M., and Raguse, C.A. 1964.
Extraction of total available carbohydrates from
grass and legume tissue. Plant Physiol. 39: 960­
962.
USDA Soil Conservation Service. 1990. Soil
survey of Thurston County, Washington.
Washington State Department of Natural
Resources,
Olympia,
Wash.
© 1997 NRC Canada