Annals of Botany 95: 263–270, 2005
doi:10.1093/aob/mci021, available online at www.aob.oupjournals.org
Overwintering Leaves of a Forest-floor Fern, Dryopteris crassirhizoma
(Dryopteridaceae): a Small Contribution to the Resource Storage
and Photosynthetic Carbon Gain
T O M O K A Z U T A N I 1,2,* and G A K U K U D O 1
1
Graduate School of Environmental Earth Science, Hokkaido University, Sapporo 060-0810, Japan and
2
Center for Far Eastern Studies, Toyama University, Toyama 930-8555, Japan
Received: 10 April 2004 Returned for revision: 26 July 2004 Accepted: 8 September 2004 Published electronically: 16 November 2004
Background and Aims Dryopteris crassirhizoma is a semi-evergreen fern growing on the floor of deciduous forests.
The present study aimed to clarify the photosynthetic and storage functions of overwintering leaves in this species.
Methods A 2-year experiment with defoliation and shading of overwintering leaves was conducted. Photosynthetic
light response was measured in early spring (for overwintering leaves) and summer (for current-year leaves).
Key Results No nitrogen limitation of growth was detected in plants subjected to defoliation. The number of leaves,
their size, reproductive activity (production of sori) and total leaf mass were not affected by the treatment. The
defoliation of overwintering leaves significantly reduced the bulk density of rhizomes and the root weight. The
carbohydrates consumed by the rhizomes were assumed to be translocated for leaf production. Photosynthetic
products of overwintering leaves were estimated to be small.
Conclusion Overwintering leaves served very little as nutrient-storage and photosynthetic organs. They partly
functioned as a carbon-storage organ but by contrast to previous studies, their physiological contribution to growth
was found to be modest, probably because this species has a large rhizome system. The small contribution of
overwintering leaves during the short-term period of this study may be explained by the significant storage ability
of rhizomes in this long-living species. Other ecological functions of overwintering leaves, such as suppression
of neighbouring plants in spring, are suggested.
ª 2004 Annals of Botany Company
Key words: Defoliation, Dryopteris crassirhizoma, Dryopteridaceae, field experiment, forest floors, leaf shading,
overwintering leaves, photosynthesis, semi-evergreen fern.
INTRODUCTION
The significance of leaf longevity has been argued mainly
from the viewpoint of physiological functions (see Chabot
and Hicks, 1982). A prolonged photosynthetic period
enables evergreen and semi-evergreen plants to maximize
carbon gain during a short growing season or seasonally
fluctuating light conditions in ecosystems having clear
seasonality (Karlsson, 1985; Shaver and Kummerow,
1992; Kikuzawa and Kudo, 1995; Kudo et al., 2001).
Increased nutrient storage is another advantage of extended
leaf longevity (Reader, 1978; Shaver, 1981; Karlsson,
1994). By storing nutrients in old leaves, plants with overwintering leaves can reduce the loss of nutrients, which
enables them to grow at nutrient-poor sites (Small, 1972;
Chapin, 1980; Aerts and Van Der Peijl, 1993). Furthermore,
overwintering leaves sometimes function as a carbonstorage organ for new growth (Kimura, 1970a; Jonasson,
1989). There is some debate as to the interpretation of
physiological functions even in the same species (e.g.
Karlsson, 1994, 1995; Jonasson, 1995a, b), with some
researchers suggesting multiple functions of overwintering
leaves in a plant.
Different functions of overwintering leaves have been
observed among/within the species of forest-floor ferns.
For example, Minoletti and Boerner (1993) observed
nitrogen and phosphorus resorption in the semi-evergreen
* For correspondence. E-mail [email protected]
fern Polystichum acrostichoides (Dryopteridaceae) in
spring, suggesting that overwintering leaves may have a
role in nutrient storage. In other studies, positive photosynthetic rates of overwintering leaves were observed in
P. acrostichoides and Dryopteris filix-mas (Dryopteridaceae) (Bauer et al., 1991; Noodén and Wagner, 1997).
Recently, Tani and Kudo (2003) measured seasonal
changes of dry weight and nitrogen allocation in a rhizomatous semi-evergreen fern, Dryopteris crassirhizoma, and
suggested that overwintering leaves may function as a
carbon-storage organ rather than a nutrient-storage organ.
However, the relative importance of storage ability of overwintering leaves was assumed to be much smaller than that
of the rhizome. Overwintering leaves of this species may
have functions other than that of a storage organ. Light
conditions on the floor of deciduous forests are favourable
during the leafless season of canopy trees (Anderson, 1964).
Because high photosynthetic production is expected on the
forest floor in early spring (Kikuzawa, 1984; Yoshie et al.,
1990), photosynthetic production of overwintering leaves
may be beneficial in this species.
The purpose of this study was to compare the relative
contributions of overwintering leaves to photosynthetic and
carbon storage functions in D. crassirhizoma. Because the
defoliation treatment alone cannot help distinguish between
the photosynthetic function and the resource-storage function of overwintering leaves, the physiological function
of overwintering leaves was evaluated by conducting an
Annals of Botany 95/2 ª Annals of Botany Company 2004; all rights reserved
264
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
experiment that combined a shading treatment and a defoliation treatment for whole plants. Based on the assumption
that shading treatment prevents assimilation in overwintering leaves, the following hypotheses were tested: (a) if the
photosynthetic function of overwintering leaves contributes
to plant growth, shading of overwintering leaves should lead
to a decrease in the subsequent growth; and (b) if the storage
function of overwintering leaves contributes to plant
growth, shading will not affect plant growth but defoliation
will result in carbon limitation. Because the effects of such
manipulations may affect the growth of below-ground as
well as above-ground organs, the responses for all organs
of a plant were examined. Because the rhizomes in
D. crassirhizoma have a large storage capacity (Tani and
Kudo, 2003) measuring the responses of rhizomes to manipulation is important for determining the actual contribution
of overwintering leaves to growth. Leaf nitrogen content
was also compared in plants subjected to the two types of
treatment to investigate whether overwintering leaves
served as a source of stored nutrients, as was predicted
by some previous results (Tani and Kudo, 2003). Finally,
CO2 gas exchange rates were measured in early spring and
summer to directly determine the photosynthetic function of
overwintering leaves. Net production was estimated from
the CO2 exchange rates before and after overwintering.
MATERIALS AND METHODS
Overwintering
leaves
B. Summer
Rhizome
Leaf primordia
C. Late autumn
D. Experiment
Control
n = 20
Study site and plant material
This study was conducted at the Tomakomai Experimental
Forest of Hokkaido University, Hokkaido, northern Japan
(42 400 N, 141 360 E). The experimental plot (approx. 20 ·
50 m) was in a secondary deciduous forest dominated by
Quercus crispula Blume and Acer amoenum Carr. This site
is usually covered with snow from early December to late
March (the maximum snow depth is approx. 06 m).
Dryopteris crassirhizoma Nakai is a common semievergreen perennial fern on the floors of deciduous forests
in northern Japan. The height of adult plants is approx. 80 cm
and the leaves can be as long as 100 cm. At our site, all
leaves begin to expand simultaneously in early May
(Fig. 1A), reach complete expansion by late May or early
June, and together form a funnel shape during the summer
(Fig. 1B). The leaves recline on the ground in a radial
pattern in mid-November, around the time of leaf fall for
canopy trees (Fig. 1C). Although the tips of the leaves begin
to wither in early winter, most parts of the leaves survive
during the winter under snow. Intensive leaf senescence
occurs in early May, when new leaf growth starts.
The body of D. crassirhizoma consists of leaves, a rhizome, leaf primordia and roots. Each leaf has many pinnae
and a leaf axis that is subdivided into a midrib and a petiole.
In fertile plants, sori are produced on the abaxial side of
the leaves in summer. The rhizome comprises many old
trophopods, which are the petiole bases of previous leaves
(Wagner and Johnson, 1983). Leaf primordia are discernible
for at least 2 years before leaf expansion. New root formation begins around October. Because roots account for only
5 % of the total dry weight of the plants, the storage function
New leaf expansion
A. Spring
Defoliation
n = 20
Shading
n = 20
F I G . 1. Leaf phenology of Dryopteris crassirhizoma. (A) New leaf
expansion occurs concurrently with senescence of overwintering leaves in
spring. (B) Leaves are arranged in a funnel shape in summer. (C) Leaves
recline in a radial pattern on the ground in late autumn, just around the leaffall season of canopy trees. (D) A schematic drawing of the field experiment,
which included a control, a shading and a defoliation group. The experiment
began in late autumn 1997; there were 20 replications per group. In the
shading-treatment group, only overwintering leaves were covered with
meshed black cloths in which holes were cut exactly to the size of the
leaf primordium area to allow for new leaf expansion in spring. The
cloths were fixed on the ground using spikes.
of the root system is small (Tani and Kudo, 2003). Vegetative reproduction does not occur in this species.
Experiment
In mid-August 1997, 60 fertile plants of D. crassirhizoma
were selected within the plot and randomly assigned to three
groups (20 plants per group): intact control, shading treatment and defoliation treatment (Fig. 1D). All the leaves of
the plants in the shading-treatment group were shaded by
meshed black cloths from late November to the following
May until all the leaves withered. The cloths were laid on
the ground directly over the plants and fixed with spikes.
Light irradiation which was measured beneath the cloth for
each plant was reduced by 742 6 35 % (means 6 s.d.) in
this treatment group. All the leaves in the defoliationtreatment group were artificially defoliated after the
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
growing season in mid-November; trophopods were not
removed. The same treatment was repeated again for
each plant the following year during the period from
early December 1998 to May 1999.
The number of leaves, their size and reproductive effort
were measured for each plant in all the groups before the
experiment, in mid-August 1997. The leaf size was defined
as the sum of all leaf lengths per plant; reproductive effort
was defined as the total length of the soriferous region per
plant, i.e. the combined length of the fertile segments. The
length of the soriferous region from top to bottom on the
abaxial side of the leaves strongly correlated with the number of sori (R2 = 085, n = 43). The mean number of leaves
and the standard deviations were 114 6 18 for the controlgroup plants, 108 6 20 for the shading-treatment group
and 119 6 19 for the defoliation-treatment group. There
was no significant difference among the three groups
(F2,57 = 155, P = 022, one-factor-ANOVA). The means
and standard deviations for the size of the leaves and reproductive effort observed in 1997 were 1013 6 230 cm and
243 6 93 cm, respectively, in the control plants, 982 6
224 cm and 194 6 125 cm in the shading-treatment plants
and 1051 6 203 cm and 253 6 116 cm in the defoliationtreatment plants. There was no significant difference in the
size of the leaves (F2,57 = 048, P = 062) and reproductive
effort (F2,57 = 157, P = 022) among the three groups. Thus,
the size of the leaves and their reproductive effort were
comparable among the three groups before the experiment.
During the experiment, the number of leaves, their size
and reproductive effort per plant were measured in midAugust of 1998 and 1999 when the leaf expansion was
complete. To calculate the initial height growth rate,
plant height was measured from the soil surface to the
top of the leaves immediately after leaf expansion and
then again 2 weeks after the leaf expansion. In the experimental plot, leaf expansion started between 8 and 14 May
in 1998 and 10 and 16 May in 1999. The first measurement
of plant height was conducted on 14 May 1998 and 16 May
1999, and the second one on 26 May 1998 and 29 May 1999.
Because the height growth rate can be influenced by
plant size, the changes in plant height were divided by
the plant height observed during the previous summer for
every plant to obtain the relative height growth rate
(RHGR) as follows:
RHGR ¼ ðH2 H1 Þ=Hlast
ð1Þ
where H1 and H2 indicate the plant height at the first and
second measurements, respectively. The plant height
observed in the previous summer (Hlast) was measured on
23 September 1997 and 13 August 1998.
Leaf dry mass per unit leaf area (LMA) and leaf nitrogen
content per unit leaf area (leaf N per area) were measured
in mid-September in 1998 and early August in 1999. As a
sampling unit, pinnae from one leaf of each plant were used.
Nine to 12 pinnae were sampled from each leaf. Because the
area of the sampled pinnae was only 1–3 % of the whole leaf
area of a plant, the sampling effect on the following year’s
plant performance was assumed to be negligible. In the
laboratory, the area of the sampled pinnae was measured
265
using a scanner connected to a PC (NIH Image ver. 155,
Wayne Rasband, National Institutes of Health, Bethesda,
MD, USA). The dry weight of each pinna was measured
after desiccation at 80 C for 3 d. The nitrogen content of
each sample was measured using a C-H-N analyser (Vario
EL, Elementar Analysensysteme, GmbH, Hanau, Germany).
At the end of the 2-year experiment, 12 plants from each
of the three groups were harvested from early to midOctober 1999. These harvested plants were washed and
divided into four parts (roots, rhizomes, developed leaf
primordia and leaves); then each organ was weighed after
desiccation at 80 C for 3–4 d. The dead tissue of the rhizomes was carefully removed and only the living tissue was
used in the measurement. To assess the movement of carbohydrates, the bulk density of each rhizome (g mL1) was
calculated from the weight and volume of the living tissue.
To measure their volume, the rhizomes were lowered into
water after desiccation and the increase in the water level
calculated. The number of leaf primordia was also counted.
Measurement of photosynthesis
Photosynthetic CO2 gas exchange rates were measured
using a LI-6400 portable photosynthesis system (Li-Cor,
Lincoln, NE, USA) for 12 fertile plants randomly selected
in the neighbourhood of the experimental plot. The measurements were carried out on 22 Apr. 2003, for overwintering leaves and on 15 July 2003 for current-year leaves. For
each measurement, an intact pinna was chosen from the
middle part of one leaf from each plant, and it was put
into a chamber cell of the LI-6400 system. When the
area of pinna enclosed by the chamber cell was smaller
than the size of the chamber cell, the pinna was sampled
after the photosynthetic measurements, its area obtained
using PC software (NIH Image ver. 155, Wayne Rasband,
National Institutes of Health), and then the photosynthetic
rates were collected as a proportion of the pinna area.
Light-response curves of photosynthetic rates were
obtained on both measurement days under regulated temperature (20 C). Photosynthetically active radiation (PAR)
was administered at nine intensities (1500, 1000, 500, 250,
100, 50, 20, 10, 0 mmol m2 s1) using a red-blue LED light
source. The ambient CO2 concentration was held constant
at 350 ml l1 and the humidity of the incoming air was
controlled at 1.0 VPD (vapour pressure deficit). The
measurement data were fitted to a non-rectangular hyperbola eqn (2) to estimate the photosynthetic parameters using
a freely available PC software program, R (http://www.
r-project.org/)
P = f f I + Pmax ½ð f I + Pmax Þ2 4f IqPmax 05 g=2q R
ð2Þ
where P, Pmax, f, I, q and R are, respectively, the net photosynthetic rate (mmol m2 s1), the light-saturated rate
of photosynthesis (mmol m2 s1), the initial slope of the
light-response curve, the intensity of PAR (mmol m2 s1),
the curvature of the line, and the dark respiration rate
(mmol m2 s1) (Lambers et al., 1998).
The PAR was measured at around noon on a sunny and
a bright–cloudy day on 19 and 18 Apr. 1999, respectively.
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
266
The measurements were carried out in 15-s intervals for a
total duration of 150 s at five randomly selected points
within the experimental plot by using a LI-189 quantum
sensor (Li-Cor). The mean PAR value obtained during the
150-s period was calculated for each point, and the values
for the five points were then averaged. PAR was also measured on 22 July (on a sunny day) and on 14 August (on a
bright–cloudy day) 1998, using the same procedure as
described above.
Based on the light-response curves and PAR values, daily
net production (Pday) in the field was estimated for overwintering leaves (late April) and current-year leaves (midJuly) assuming a 12-h day as follows:
Pday = ðPi RÞ · 12 h · 3600 s
ð3Þ
2
where Pi is the net photosynthetic rate (mmol m
when PAR is i.
1
s )
Statistical analyses
All variables measured in the experiment had similar
variance in all three groups (P > 005; Bartlett test).
Repeated-measures analyses of variance (ANOVAs)
(Underwood, 1997) were performed on RHGR, number of
leaves, leaf size, reproductive effort, LMA and leaf nitrogen
per area to determine the effect of group and sampling year.
When there was a significant interaction between year and
group, the data for each year were then separately analysed
by one-factor ANOVA. The potential for increased Type I
error due to pseudoreplication was addressed by using the
Bonferroni adjustment (see Underwood, 1997). When significant differences were observed, Fisher’s protected least
significant difference method (PLSD) was used as a multiple comparison test with the alpha level set at P = 005. Leaf
weight and all below-ground parts were compared by using
one-factor ANOVA and Fisher’s PLSD method (P = 005)
when there was a significant difference among the groups.
During the winter, falling branches of canopy trees sometimes hit some of the plants in the study, and wild deer
occasionally trampled over new leaf primordia. Such
damaged plants were therefore excluded from the statistical
analyses. All statistical analyses were performed by Stat
View 5.0 (SAS Institute Inc., Cary, NC, USA).
RESULTS
Field experiment
The performance of the above-ground parts was little
affected by the experimental manipulations (Table 1).
The relative height growth rate (RHGR) differed significantly between the two years, especially for the defoliation-treatment group (a 35 % reduction in RHGR was
observed in the second year; Fig. 2A) but there was no
significant effect of treatment. The number of leaves produced by each plant was stable during the 2-year experiment; there was no significant difference among the groups
and between the years (Fig. 2B). There was a significant
difference for leaf size between the two years: it increased
T A B L E 1. Repeated-measures analyses of variance
(ANOVAs) for relative height growth rate, number of leaves,
leaf size, reproductive effort, leaf dry mass per unit leaf area
and leaf nitrogen content per unit leaf area
Variable and source
d.f.
MS
F
Relative height growth rate (RHGR)
Treatment
45
0.05
1.01
Year
1
0.31
7.26
Treatment · year
2
0.12
2.91
Number of leaves
Treatment
45
9.06
1.42
Year
1
3.58
1.98
Treatment · year
2
1.01
0.56
Leaf size
0.84
Treatment
45
7.50 · 104
8.45
Year
1
9.77 · 104
Treatment · year
2
2.36 · 104
2.04
Reproductive effort
2.78
Treatment
45
4.78 · 104
1.16
Year
1
3.32 · 104
7.74
Treatment · year
2
2.21 · 104
Leaf dry mass per unit leaf area (LMA)
Treatment
45
1.56
12.0
Year
1
0.32
3.32
Treatment · year
2
0.15
1.56
Leaf nitrogen content per unit leaf area (leaf N per area)
1.93
Treatment
45
2.63 · 104
4.27
Year
1
4.96 · 104
0.46
Treatment · year
2
5.33 · 104
P
0.37
<0.001
0.07
0.25
0.17
0.58
0.44
0.006
0.14
0.07
0.29
0.001
<0.001
0.08
0.22
0.16
0.04
0.64
more in control plants than in plants in the shading- and
defoliation-treatment groups in both 1998 and 1999
(Fig. 2C). However, there was no significant effect of treatment. There was a significant year · treatment interaction
for reproductive effort; thus, the values for each year were
compared separately. There was no significant difference
for reproductive effort among the groups for 1998 (F2,45 =
043, P = 065). By contrast, for 1999, there was a significant
difference among the groups (F2,45 = 722, P = 0002);
in that year, the shaded plants produced significantly
fewer sori per plant than the control and defoliated plants
(Fig. 2D).
Leaf dry mass per unit leaf area (LMA) differed significantly among the groups, but the between-year difference
was not statistically significant (Table 1). The shading- and
defoliation-treatment groups showed similar significant
reductions in LMA, in contrast to the control plants
(Fig. 2E). Although the leaf nitrogen per area increased
during the period from 1998 to 1999, the effect of treatment
on leaf nitrogen per area was not significant (Fig. 2F).
The total leaf mass at the end of the experiment was
smaller in the defoliated plants than in the plants in
the other two groups (Table 2) although the difference
among the three groups was not statistically significant
(F2,33 = 252, P = 007).
In the below-ground parts, significant differences were
observed in the bulk density of the rhizomes (F2,33 = 726,
P = 0002), developed leaf primordia weight (F2,33 = 114,
P < 0001) and root weight (F2,33 = 344, P = 004) among
the groups (Table 2). However, the number of developed
leaf primordia did not differ significantly among the groups
267
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
1·0
D 500
0·8
400
Reproductive effort (cm)
A
RHGR
0·6
0·4
0·2
0·0
B
a
b
b
300
200
100
0
E
15
5
a
LMA (mg cm−2)
Number of leaves
4
10
5
b
S
D
a
b
b
S
D
3
2
1
0
C 1400
F 0.12
1200
0·10
Leaf N per area (mg cm−2)
0
1000
Leaf size (cm)
b
800
600
400
200
0
0·08
0·06
0·04
0·02
0·00
C
S
1998
D
C
S
D
1999
C
1998
C
1999
F I G . 2. Changes in (A) relative growth rate (RHGR), (B) the number of leaves per plant, (C) leaf size defined as the sum of leaf lengths per plant, (D)
reproductive effort represented by the combined length of fertile segments per plant, (E) leaf dry mass per unit leaf area (LMA) and (F) leaf nitrogen content
per unit leaf area (leaf N per area) between 1998 and 1999. Open columns represent control plants (C), hatched columns represent plants subjected to shading
treatment (S), and filled columns represent plants subjected to defoliation treatment (D). Vertical bars show the standard deviation. n = 15 in the control group,
17 in the shading-treatment group, 16 in the defoliation-treatment group. Different lowercase letters above the columns indicate significant differences
among the groups within each year (P < 0.05, Fisher’s PLSD test).
(F2,33 = 060, P = 056). The bulk density of the rhizomes
was significantly lower (12–15 %) in the defoliationtreatment group than in the other groups. The developed
leaf primordia weight decreased by 22 % in the shadingtreatment plants and by 39 % in the defoliation-treatment
plants compared with the control plants. These decreases
were statistically significant. The root weight decreased
by 23 % in the shading-treatment group and by 45 % in
the defoliation-treatment group compared with that in the
control plants. Only the difference between the
control and defoliation-treatment groups was statistically
significant.
Photosynthetic ability of overwintering leaves
The photosynthetic rates of overwintering leaves in the
spring were much smaller than those of current-year leaves
in the summer (Fig. 3). The maximum photosynthetic rate
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
268
T A B L E 2. Comparison of total mass and below-ground parameters among the groups at the end of the 2-year experiment
Treatment
Parameter
Control
Total leaf mass (g)
46.7
0.74
Bulk density of
rhizome (g ml1)
No. of developed leaf
11.7
primordia
Leaf primordia weight (g) 15.2
Root weight (g)
11.6
Shading*
Defoliation*
6 14.8a 46.3 6 10.7a
6 0.05a 0.72 6 0.06a
42.7 6 11.2a
0.65 6 0.08b
6 2.31a 10.8 6 1.95a
11.6 6 1.88a
6 4.08a 11.9 6 2.26b 9.35 6 2.37c
6 5.55a 8.91 6 5.83ab 6.30 6 2.90b
* Of overwintering leaves.
Values are means 6 s.d.; n = 12 in each group.
Superscript letters indicate results of a multiple comparison test
(Fisher’s PLSD test); significant differences are indicated by different
letters (P < 005).
5
Net photosynthetic rate (µmol m–2 s–1)
4
3
2
1
0
–1
0
500
1000
PAR (µmol m–2 s–1)
1500
F I G . 3. Light-response curves of photosynthetic rates of overwintering
leaves obtained on 22 Apr. (crosses and the solid line) and of currentyear leaves obtained on 15 July (circles and the broken line) in 2003.
Photosynthetic rates were measured in 12 plants, at nine intensity levels
of photosynthetically active radiation (PAR) on both days. Means and
standard errors for each PAR measurement are shown. Long and short
upward open arrows indicate the values of PAR for control plants
obtained, respectively, under sunny and bright–cloudy conditions in late
April. Long and short filled arrows indicate PAR values for shaded plants
obtained, respectively, under sunny and bright–cloudy conditions in late
April. Long and short downward open arrows indicate PAR values for
control plants obtained, respectively, under sunny (late July) and bright–
cloudy conditions (mid-August) in the summer.
(Pmax) was 212 mmol m2 s1 in the overwintering leaves
and 543 mmol m2 s1 in the current-year leaves. The dark
respiration rate was larger in the overwintering leaves
(026 mmol m2 s1) than in the current-year leaves
(012 mmol m2 s1). The initial slope of the light-response
curves was steeper for the current-year leaves (0063) than
for the overwintering leaves (0023).
The PAR under sunny and bright–cloudy conditions
obtained in the spring was 1103 mmol m2 s1 and
294 mmol m2 s1, respectively (Fig. 3). Given that 26 %
of total radiance can still penetrate a meshed black cloth,
the PAR in the early spring under the shading treatment was
287 and 76 mmol m2 s1 (filled arrows in Fig. 3) under
sunny and bright–cloudy conditions, respectively. Based on
the PAR values and light-response curves of photosynthetic
rates, it was assumed that the overwintering leaves had
similar photosynthetic rates to those of the control plants
(107 mmol m2 s1) and shaded plants (099 mmol m2 s1)
under direct solar radiation. Under bright–cloudy conditions, the photosynthetic rate of the control plants was
098 mmol m2 s1 and that of the shaded plants was
056 mmol m2 s1. Thus, the shading-treatment in this
study moderately reduced the photosynthetic carbon gain
by the overwintering leaves.
On sunny and bright–cloudy days in the summer, the
PAR was 224 and 53 mmol m2 s1 within the experimental
plot (Fig. 3). Based on the light-response curve measured
in the summer, the photosynthetic rate was estimated
to be 367 mmol m2 s1 under sunny conditions and
205 mmol m2 s1 under bright–cloudy conditions.
The estimated daily production by overwintering leaves
was 3491 mmol m2 on sunny days and 3145 mmol m2
on bright–cloudy days. The estimated daily production of
current-year leaves in the summer was 15336 mmol m2 on
sunny days and 8338 mmol m2 on bright–cloudy days.
Therefore, the potential photosynthetic ability of overwintering leaves was 23–38 % that of the current-year leaves.
The period from the completion of leaf expansion to the
beginning of snow accumulation was 168 d and the period
from the snow melting to leaf senescence was 56 d in 1998–
1999. On snow-melting days in 1999, the living area of the
overwintering leaves was, on average, 72 % of the total leaf
area in the control plants. Based on these findings, the total
contribution of overwintering leaves to a photosynthetic
function was estimated to be 5–9 % of that of currentyear leaves.
DISCUSSION
Contrary to expectations, neither the shading nor the defoliation treatment significantly influenced the number of
leaves, leaf size and total leaf mass of D. crassirhizoma
during the two-year experiment. The reproductive effort of
the shaded plants was significantly lower in the second year
compared with that of the other plants, but this may not
necessarily reflect the effect of the manipulation because
it is more difficult to explain a strong resource limitation
by shading than by defoliation. Defoliation of overwintering
leaves usually leads to a quick reduction in vegetative
growth or reproductive output in both angiosperms and
ferns (Reader, 1978; Karlsson, 1994; Jonasson, 1995b;
Van Buskirk and Edwards, 1995; Eckstein et al., 1998). For
example, in another Dryopteris species, D. intermedia,
defoliation of overwintering leaves caused a 60 % reduction
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
in plant size within 1 year (Van Buskirk and Edwards,
1995). By contrast, the physiological contribution of overwintering leaves in D. crassirhizoma was much smaller.
Uemura (1994) speculated that the rapid expansion of
new leaves in D. crassirhizoma might be due to both the
stored resources in overwintering leaves and their assimilated products. However, because the RHGR showed no
effect of the defoliation or shading treatments, overwintering leaves did not appear to contribute to the initial leaf
growth.
A significant effect of treatment was observed in LMA,
which decreased in both the shading- and defoliation-treatment plants. However, the reduction in LMA is unlikely to
be a consequence of carbon deficiency because there was
no significant difference in the final leaf weight among the
groups. The leaf nitrogen per area did not change among
the three groups during the experiment, indicating that no
nitrogen limitation was imposed by the defoliation and that
the overwintering leaves did not serve as a nitrogen storage
organ. This result agrees with previous results (Tani and
Kudo, 2003). Several studies have demonstrated that overwintering leaves serve as a nutrient storage organ (Reader,
1978; Chapin et al., 1980; Moore, 1980; Chabot and Hicks,
1982; Karlsson, 1994), and most of these studies were conducted in nutrient-poor habitats. In contrast, the habitats
of D. crassirhizoma are nutrient-rich understoreys of deciduous forests where plants may easily absorb nutrients
from the soil independent of nutrient storage within the
plant body.
By contrast to the response patterns of the above-ground
organs, the effect of treatment on the below-ground parts
was clear. The bulk density of the rhizomes showed a significant reduction only in the defoliation-treatment group.
This suggests that consumption of carbohydrates does occur
in the rhizomes when overwintering leaves are removed.
Because some carbohydrates were translocated from the
rhizomes to support new leaf development in this species
(Tani and Kudo, 2003), it is reasonable to conclude that
carbohydrates in the rhizomes of defoliated plants are translocated for leaf development. Because of the compensatory
movement of carbohydrates from the rhizomes, total leaf
mass in defoliated plants may be maintained at a constant
level as in shaded plants.
Although a significant difference in the root mass was
detected only between the control and defoliation-treatment
groups, the root mass of the shaded plants showed intermediate values between the control and defoliated plants.
This indicates that root production may be partly supported
by carbon storage and the photosynthetic function of overwintering leaves. It has been reported that plants under
shade conditions tend to show a lower allocation to roots
(reviewed in Ackerly, 1997). Although the root mass of the
defoliated plants decreased by 45 % compared with that
of the control plants, the nitrogen content in the leaves
was similar among the groups. The reduction of root
mass in addition to the removal of overwintering leaves
did not result in nitrogen limitation in this species, at
least during the short-term period of this study.
The number of leaf primordia did not change during
the experiment. This was predictable as D. crassirhizoma
269
begins to form leaf primordia at least 2 years in advance of
their functioning as photosynthetic organs (Tani and Kudo,
2003). A significant reduction in leaf primordium weight
was observed in the shading- and defoliation-treatment
groups, which may be indirectly caused by the decrease
in root mass as described above. The mass difference in
leaf primordia may affect the leaf size in subsequent years,
but a longer experiment is needed to detect the treatment
effect on the above-ground mass in D. crassirhizoma.
Although the defoliation treatment caused a greater reduction in size in leaf primordia than did the shading treatment,
it is difficult to determine whether the photosynthetic function alone affected the formation of leaf primordia due to
a partial cut-off of radiance in the shading treatment or
whether the carbon storage function also played a role in
this process.
The impact of shading in this study may be modest
because the effectiveness of the shading treatment was
quite different on sunny and bright–cloudy days. On cloudy
days, the carbon gain by overwintering leaves of the shaded
plants was estimated at 57 % that of the control plants.
On sunny days, however, there was only a 7 % reduction
in carbon gain. Nevertheless, according to the authors’ calculations, spring net production by overwintering leaves
was <10 % that in the summer, irrespective of weather
conditions. Previous studies have shown that overwintering
leaves of forest-floor herbs can assimilate a substantial
amount of dry matter during a short period in spring
(Kimura, 1970a, b; Yamamura, 1984; Yoshie et al.,
1990). By contrast, there can be little contribution of
overwintering leaves to the photosynthetic function in
D. crassirhizoma.
Overwintering leaves of D. crassirhizoma probably contribute to a carbon storage organ, but the storage ability of
the rhizome is likely to be adequate to maintain the aboveground plant size. The physiological function of overwintering leaves in this species was modest contrary to
expectations and it was different from that in other angiosperms or ferns, at least in this short-term observation. In
addition to the physiological function, other explanations
can be suggested for the maintenance of overwintering
leaves. Shading of neighbouring plants by overwintering
leaves during spring (the most light-rich season on the forest
floor) may allow the fern to monopolize the space for new
leaf development in canopy forests because most understorey plants start to grow before canopy closure. In the
experimental site, D. crassirhizoma occupied 431 % of
the total above-ground biomass of the forest-floor vegetation, and the density of this species was 12 m2 (T. Tani,
unpubl. res.). Such predominance of this species may be
related to the unique leaf behaviour resulting from an ecological function, i.e. suppression of neighbouring plants.
ACKNOWLEDGEMENTS
The authors are grateful to T. Kohyama for valuable suggestions, to N. Wada for kind support, to T. Seino and
A. Uraguchi for assistance, and to all the staff of
Tomakomai Experimental Forest for support.
270
Tani and Kudo — Functions of Overwintering Leaves in Dryopteris
LITERATURE CITED
Ackerly D. 1997. Allocation, leaf display, and growth in fluctuating light
environments. In: Bazzaz F, Grace J, eds. Plant resource allocation.
San Diego: Academic Press, 231–264.
Aerts R, Van Der Peijl MJ. 1993. A simple model to explain the dominance of low-productive perennials in nutrient-poor habitats. Oikos 66:
144–147.
Anderson MC. 1964. Studies of the woodland light climate. II. Seasonal
variation in the light climate. Journal of Ecology 52: 643–663.
Bauer H, Gallmetzer Ch, Sato T. 1991. Phenology and photosynthetic
activity in sterile and fertile sporophytes of Dryopteris filix-mas (L.)
Schott. Oecologia 86: 159–162.
Chabot BF, Hicks DJ. 1982. The ecology of leaf life spans. Annual Review
of Ecology and Systematics 13: 229–259.
Chapin FS III. 1980. The mineral nutrition of wild plants. Annual Review of
Ecology and Systematics 11: 233–260.
Chapin FS III, Johnson DA, McKendrick JD. 1980. Seasonal movement
of nutrients in plants of differing growth form in an Alaskan
tundra ecosystem: implications for herbivory. Journal of Ecology
68: 189–209.
Eckstein RL, Karlsson PS, Weih M. 1998. The significance of resorption
of leaf resources for shoot growth in evergreen and deciduous woody
plants from a subarctic environment. Oikos 81: 567–575.
Jonasson S. 1989. Implications of leaf longevity, leaf nutrient re-absorption
and translocation for the resource economy of five evergreen plant
species. Oikos 56: 121–131.
Jonasson S. 1995a. Nutrient limitation in Rhododendron lapponicum, fact or
artefact? Oikos 73: 269–271.
Jonasson S. 1995b. Resource allocation in relation to leaf retention time of
the wintergreen Rhododendron lapponicum. Ecology 76: 475–485.
Karlsson PS. 1985. Photosynthetic characteristics and leaf carbon economy
of a deciduous and an evergreen dwarf shrub: Vaccinium uliginosum L.
and V. vitis-idaea L. Holarctic Ecology 8: 9–17.
Karlsson PS. 1994. The significance of internal nutrient cycling in branches
for growth and reproduction of Rhododendron lapponicum. Oikos
70: 191–200.
Karlsson PS. 1995. Nutrient or carbon limitation of shoot growth in
Rhododendron lapponicum—both or neither! A reply to Jonasson.
Oikos 73: 272–273.
Kikuzawa K. 1984. Leaf survival of woody plants in deciduous broadleaved forests. 2. Small trees and shrubs. Canadian Journal of Botany
62: 2551–2556.
Kikuzawa K, Kudo G. 1995. Effects of the length of the snow-free period
on leaf longevity in alpine shrubs: a cost-benefit model. Oikos 73:
214–220.
Kimura M. 1970a. Analysis of production processes of an undergrowth
of subalpine Abies forest, Pteridophyllum racemosum population. 1.
Growth, carbohydrate economy and net production. Botanical
Magazine, Tokyo 83: 99–108.
Kimura M. 1970b. Analysis of production processes of an undergrowth
of subalpine Abies forest, Pteridophyllum racemosum population.
2. Respiration, gross production and economy of dry matter. Botanical
Magazine, Tokyo 83: 304–311.
Kudo G, Molau U, Wada N. 2001. Leaf-trait variation of tundra plants
along a climatic gradient: an integration of responses in evergreen
and deciduous species. Arctic, Antarctic and Alpine Research 33:
181–190.
Lambers H, Chapin FSI, Pons TL. 1998. Plant physiological ecology.
New York: Springer, 10–95.
Minoletti ML, Boerner REJ. 1993. Seasonal photosynthesis, nitrogen
and phosphorus dynamics, and resorption in the wintergreen fern
Polystichum acrostichoides (Michx.) Schott. Bulletin of the Torrey
Botanical Club 120: 397–404.
Moore P. 1980. The advantages of being evergreen. Nature 285: 535.
Noodén LD, Wagner WH Jr. 1997. Photosynthetic capacity and leaf
reorientation in two wintergreen ferns, Polystichum acrostichoides
and Dryopteris intermedia. American Fern Journal 87: 143–149.
Reader RJ. 1978. Contribution of overwintering leaves to the growth of
three broad-leaved, evergreen shrubs belonging to the Ericaceae
family. Canadian Journal of Botany 56: 1248–1261.
Shaver GR. 1981. Mineral nutrition and leaf longevity in an evergreen
shrub, Ledum palustre ssp. decumbens. Oecologia 49: 362–365.
Shaver GR, Kummerow J. 1992. Phenology, resource allocation, and
growth of arctic vascular plants. In: Chapin FS III, Jefferies RL,
Reynolds JF, Shaver GR, Svoboda J, eds. Arctic ecosystems in a
changing climate. San Diego: Academic Press, 193–211.
Small E. 1972. Photosynthetic rates in relation to nitrogen recycling as an
adaptation to nutrient deficiency in peat bog plants. Canadian Journal
of Botany 50: 2227–2233.
Tani T, Kudo G. 2003. Storage ability of overwintering leaves and rhizomes
in a semi-evergreen fern, Dryopteris crassirhizoma (Dryopteridaceae).
Ecological Research 18: 15–24.
Uemura S. 1994. Patterns of leaf phenology in forest understory. Canadian
Journal of Botany 72: 409–414.
Underwood AJ. 1997. Experiments in ecology. Cambridge: Cambridge
University Press, 403–408.
Van Buskirk J, Edwards J. 1995. Contribution of wintergreen leaves to
early spring growth in the wood fern Dryopteris intermedia. American
Fern Journal 85: 54–57.
Wagner WH Jr, Johnson DM. 1983. Trophopod, a commonly overlooked storage structure of potential systematic value in ferns. Taxon
32: 268–269.
Yamamura Y. 1984. Matter production processes of Reineckia carnea
Kunth, an evergreen forest floor herb in the warm-temperate region
of Japan. Botanical Magazine, Tokyo 97: 179–191.
Yoshie F, Arai H, Nakashima H, Kawano S. 1990. Seasonal changes in
nitrogen fractions of Pachysandra terminalis, a forest evergreen
chamaephyte. Physiologia Plantarum 79: 7–14.