J Plant Res (2014) 127:329–338
DOI 10.1007/s10265-013-0610-2
REGULAR PAPER
Growth, allometry and shade tolerance of understory saplings
of four subalpine conifers in central Japan
Koichi Takahashi • Yoshiko Obata
Received: 25 June 2013 / Accepted: 13 October 2013 / Published online: 6 December 2013
Ó The Botanical Society of Japan and Springer Japan 2013
Abstract The conifers Abies veitchii, A. mariesii, Picea
jezoensis var. hondoensis, Tsuga diversifolia dominate in
subalpine forests in central Japan. We expected that species
differences in shade tolerance and in aboveground and
belowground architecture are important for their coexis-
tence. We examined net production and carbon allocation
of understory saplings. Although the four species allocated
similar amounts of biomass to roots at a given trunk height,
the root-zone area of T. diversifolia was greater than that of
the three other species. T. diversifolia often dominates
shallow soil sites, such as ridge and rocky slopes, and,
therefore, a wide spread of lateral roots would be an
adaptation to such edaphic conditions. Crown width and
leaf and branch mass were greatest for T. diversifolia and
A. mariesii, followed in order by A. veitchii and P. jezo-
ensis var. hondoensis. Although leaf mass of P. jezoensis
var. hondoensis was lowest among the four species, species
differences were not found in the net production per sap-
ling because net production per leaf mass was greatest for
P. jezoensis var. hondoensis. The leaf lifespan was longer
in the order A. mariesii, T. diversifolia, P. jezoensis var.
hondoensis and A. veitchii. The minimum rate of net pro-
duction per leaf mass required to maintain the current
sapling leaf mass (MRNPLM) was lowest in A. mariesii and
T. diversifolia, and increased in the order of A. veitchii and
P. jezoensis var. hondoensis. A. mariesii and T. diversifolia
may survive in shade conditions by a lower MRNPLM than
K. Takahashi (&)  Y. Obata
Department of Biology, Faculty of Science, Shinshu University,
Asahi 3-1-1, Matsumoto, Nagano 390-8621, Japan
e-mail: koichit@shinshu-u.ac.jp
K. Takahashi
Institute of Mountain Science, Shinshu University, Asahi 3-1-1,
Matsumoto, Nagano 390-8621, Japan
the two other species. Therefore, species differences in
aboveground and belowground architecture and MRNPLM
reflected their shade tolerance and regeneration strategies,
which contribute to their coexistence.
Keywords Allocation  Crown architecture 
Conifer  Net production  Regeneration
Introduction
Subalpine coniferous forests where genera Abies and Picea
coexist are widely distributed in North America (Farrar
1995), Japan (Franklin et al. 1979; Takahashi 1994) and
China (Okitsu et al. 1995; Taylor et al. 1996). Stand
structure and regeneration dynamics are somewhat similar
in Abies-Picea forests in the world. Small Abies trees have
a higher density than small Picea trees because of their
greater shade tolerance, but canopy Picea have a lower
mortality because of their greater longevity (Veblen 1986).
Some researchers suggested the importance of disturbances
on their coexistence, i.e., Picea regenerates after cata-
strophic disturbance, but sufficient time without distur-
bance would result in competitive exclusion of Picea by
the more prolific Abies (Antos and Parish 2002; Day 1972;
Peet 1981). Although coexistence mechanisms of Abies
and Picea are not fully understood, species differences in
shade tolerance and growth traits are important factors
affecting their coexistence.
Abies veitchii Lindl., Abies mariesii Mast. and Picea
jezoensis var. hondoensis Rehder grow in subalpine forests
in central Japan (Franklin et al. 1979). Tsuga diversifolia
Mast. also often grows in subalpine forests in central Japan
(Kanzaki 1984). A gap regeneration study of an Abies-
Picea-Tsuga forest found different regeneration
123
330
characteristics among the three genera: the two Abies
species grow in canopy gaps from the sapling bank that
existed before the gap formation, seedling establishment of
P. jezoensis var. hondoensis without developed seedling
and sapling banks occurs after gap formation, and the
regeneration pattern of T. diversifolia is between the two
Abies species and P. jezoensis var. hondoensis (Yamamoto
1993). A. mariesii is more shade-tolerant than A. veitchii,
but A. veitchii can grow more rapidly in high light condi-
tions than A. mariesii (Kohyama 1983). Therefore, shade
tolerance is considered to be greater in the order A. mari-
esii, A. veitchii, T. diversifolia and P. jezoensis var.
hondoensis.
Of the four species, T. diversifolia tends to occur at
shallow soil sites, such as rocky slopes and ridges (Franklin
et al. 1979; Sugita and Tani 2001). Seasonal water stress is
stronger on ridges with shallower soil than on lower slopes
(Yanagisawa and Fujita 1999). Tree species at ridges tend
to spread lateral roots to acquire soil water because deep
rooting brings few advantages of water uptake from the soil
due to total drying (Yanagisawa and Fujita 1999). Long
lateral roots are also advantageous for nutrient uptake at
infertile sites (Paz 2003), because most nutrients are con-
centrated near the soil surface where decomposition pro-
cesses principally occur (Becker and Castillo 1990).
Distribution of soil water and nutrients is spatially heter-
ogeneous (Jackson and Caldwell 1993a, b; Smithwick et al.
2005). A wide root-zone area (i.e., proliferation of lateral
roots) is suggested to be an adaptation form for Tsuga
diversifolia at shallow soil sites to explore nutrient-rich
patches. Therefore, the four coexisting species may have
different growth strategies in relation to shade tolerance
and edaphic conditions.
Plant growth is a reiterative process whereby net pro-
duction is allocated differentially to each organ (trunk,
roots, branches and leaves), and the growth of each organ
in turn influences future rates of net production (Takahashi
et al. 2001; Takahashi and Lechowicz 2008; Takahashi and
Rustandi 2006). If a tree individual allocates more carbon
to the trunk (i.e., vertical-growth species), the trunk height
growth increases, but future net production will decrease
because of less allocation to leaves. However, greater
allocation to leaves is advantageous for net production (i.e.,
lateral-growth species), but it needs much leaf-support
cost, such as branches (Kohyama 1987). Lateral-growth
species also need a thick trunk to support many leaves and
branches, which increases the biomass required for unit
trunk height growth. A large crown area increases assimi-
lation capacity because many leaves can distribute without
mutual shading. Although lateral-growth species are
advantageous for survival in dark understory conditions,
the trunk height growth rate is lower than by vertical-
growth species in high light conditions (Kohyama 1991).
123
J Plant Res (2014) 127:329–338
Therefore, a growth-survival trade-off is between vertical-
growth and lateral-growth species at the sapling stage
(Kohyama 1987).
In evergreen species with a leaf lifespan longer than
one year, senescent leaves fall and new leaves unfold every
year. If the mass of new leaves is lower than the fallen
leaves, the total leaf mass per plant decreases. Individual
trees will die if leaf mass per plant decreases every year.
Thus, evergreen species must produce, at least, the same
mass of new leaves as fallen leaves. Leaf lifespan is an
important factor affecting the annual mass of fallen leaves
because the relative amount of fallen leaves to total leaves
is larger if the leaf lifespan is shorter. The leaf lifespan of
shade-intolerant species tends to be shorter than that of
shade-tolerant species (Walters and Reich 1999). There-
fore, shade-intolerant species with shorter leaf lifespan is
suggested to require a higher net production per leaf mass
to maintain the current leaf mass per plant compared with
shade-tolerant species.
Although subalpine coniferous forests in central Japan
are dominated by the four species (A. mariesii, A. veitchii,
T. diversifolia, P. jezoensis var. hondoensis) with different
shade tolerance and suitable edaphic conditions, previous
studies did not compare aboveground and belowground
growth traits of saplings among the four species at one
time. Especially, a comparative study of roots may not
have been made. Comparison of aboveground and below-
ground growth traits would contribute to understanding of
the coexistence mechanism of the four species. Thus, the
aim of this study was to clarify the regeneration strategies
of saplings of A. mariesii, A. veitchii, T. diversifolia,
P. jezoensis var. hondoensis in central Japan by comparing
their net production and carbon allocation patterns. Espe-
cially, this study investigated the following questions.
1. Do more shade-tolerant species have a longer leaf
lifespan, wider crown area, greater sapling leaf mass,
greater net production rate per sapling and a lower
minimum rate of net production required to maintain
the current sapling leaf mass?
2. Is the root-zone area of T. diversifolia wider than for
the three other species?
Materials and methods
Study site
This study was done at 2,000 m above sea level (a.s.l.) on
the east slope of Mt. Norikura (36°060 N, 137°330 E,
3,026 m a.s.l.) in central Japan. Mean monthly tempera-
tures of the coldest month (January) and the hottest month
(August) at the study site were estimated as -9.2 and
J Plant Res (2014) 127:329–338
14.9 °C, respectively, with 2.7 °C annual mean tempera-
ture, from temperatures recorded at Nagawa Weather Sta-
tion (1,068 m a.s.l.) during a 1979–2011 period using the
standard lapse rate of -0.6 °C for each ?100 m altitude.
The study site was dominated by A. mariesii, A. veitchii
and T. diversifolia. Although P. jezoensis var. hondoensis
also grows at this study site, the density was lower than for
the three other species (Miyajima et al. 2007). Subordinate
trees were all deciduous broad-leaved trees: Betula ermanii
Cham., Sorbus commixta Hedland and Acer ukurunduense
Trantv. et Meyer. Miyajima et al. (2007) and Miyajima and
Takahashi (2007) describe in detail the species composi-
tion and forest structure along an altitudinal gradient of Mt.
Norikura.
Field methods
We investigated A. mariesii, A. veitchii, T. diversifolia and
P. jezoensis var. hondoensis. We sampled undamaged
saplings (0.1–1.4 m tall) of the four species in the under-
story between September and October in 2009. No saplings
were sampled in canopy gaps. The number of samples was
44 for A. mariesii, 58 for A. veitchii, 50 for T. diversifolia
and 28 for P. jezoensis var. hondoensis. Roots were also
dug up, except for large saplings. We sampled roots of a
total of 172 samples, 85 % of which had a trunk height
shorter than 50 cm. At the sampling, trunk height, crown
width and root-zone width of each sapling were measured.
Crown width and root-zone width in two perpendicular
directions (one was the maximum width) were measured
for each sapling. Some small lateral roots were undoubt-
edly cut when we excavated saplings. Thus, the root-zone
width might be underestimated.
We brought back the samples to the laboratory, and we
measured the trunk diameter at the 1/10 trunk height and
length of current-year leader shoot (2009) of main trunk.
We counted the number of intact needles and fallen-needle
scars on shoots of different ages along the main axis of a
branch for each sapling (Kimura 1963; Kohyama 1980).
Leaf survival at each shoot age was calculated by dividing
the number of intact needles by the total number of intact
needles and fallen-needle scars. Leaf lifespan was defined
as the year that 50 % of needles fell (Mori and Takeda
2004a). Saplings were divided into current-year shoots
(leaf and stem), leaves, branches, trunk and root. Each
organ was oven dried at 80 °C for two days, and the dry
mass was measured.
Net production rate
The cut surface of a trunk at 1/10 trunk height was sanded
to measure tree-ring width. We measured distances from
the pith to the tree ring of the current year (R09) and to the
331
tree-ring of the previous year (R08) at the 1/10 trunk height
by using a stereoscopic microscope. The relative growth
rate (RGR, year-1) of the trunk radius was calculated by
Eq. 1.
RGR ¼ ln R09  ln R08 ð1Þ
where ln is a natural logarithm. We assumed that the RGR
of trunk diameter is equal to the RGR of the trunk radius.
We estimated the trunk diameter of the previous year (D08)
by substituting D08 and D09 (trunk diameter of the current
year) for R08 and R09, respectively.
D08 ¼ expðln D09  RGRÞ ð2Þ
Net production (NP) is the current-year growth of dry
mass, i.e., total dry mass of new leaves of current-year
shoots (MNL) and non-photosynthetic organ (trunk, branch,
root) produced in the current year. Total dry mass of non-
photosynthetic organs produced in the current year was
estimated by subtracting the dry mass of the non-
photosynthetic organ of the previous year (MTBR08) from
that of the current year (MTBR09). We made an allometric
equation of MTBR09 (Y) against D209H09 (X) for each
species, where H09 is the trunk height of the current year
(Table 2). Then, we calculated MTBR08 by substituting
D208H08 for D209H09, where H08 is the trunk height of the
previous year. NP was calculated by Eq. 3.
NP ¼ MNL þ MTBR09  MTBR08 ð3Þ
Allometric analysis
Allometric relationships between parts of a plant based on
log–log linear regression were used to assess quantitative
characteristics of tree morphology for each species. Allo-
metric relationships were examined for trunk mass, leaf
mass, branch mass, root mass, crown width and root-zone
width, each plotted against trunk height. Two relationships
that include net production were also examined: net pro-
duction against sapling leaf mass and trunk height. Allo-
metric relationships were fitted using standardized major
axis regression [program (S)MATR, version 2; (Warton
et al. 2006)]. Differences in slope and intercept were
assessed by using a t test and analysis of covariance
(ANCOVA), respectively, to compare allometric regres-
sions among the four species.
Minimum rate of net production per leaf mass required
to maintain the current sapling leaf mass
We estimated the minimum rate of net production (MRNP)
required to maintain the current sapling leaf mass as a
measure of shade tolerance. Plants need to produce new
leaf mass more than the fallen leaves; otherwise plants
reduce their leaf mass with time, which may lead to death
123
332 J Plant Res (2014) 127:329–338

near future. We define MRNP as the rate of production to Table 1 Definition and unit of abbreviations
maintain constant leaf mass. To evaluate MRNP, we Symbol Definition Unit
obtained the relationship among net production rate (NP),
dry mass of current-year leaves (MNL) and dry mass of H Trunk height cm
non-photosynthetic tissues (MTBR) using multiple regres- D Trunk diameter cm
sion as follows. D2H Trunk volume expressed as the product of cm3
trunk height and trunk diameter
log NP ¼ a þ b log MNL þ c log MTBR ð4Þ
MT Trunk mass g
where log is a common logarithm (base 10), and a, b, c are ML Leaf mass g
constants. MTBR was included in Eq. 4 because production MB Branch mass g
of non-photosynthetic tissues is necessary for new leaves. MR Root mass g
MRNP was calculated by substituting the rate of leaf fall, MTBR Total mass of non-photosynthetic organ g
which is same as the dry mass of annual fallen leaves (trunk, branch, root)
(MFL), for MNL. MNL Total mass of current-year leaves g
MFL Mass of annual fallen leaves
log MRNP ¼ a þ b log MFL þ c log MTBR ð5Þ
WC Crown width cm
We applied values obtained in 2009 for MNL and MTBR. WR Root-zone width cm
NP was obtained according to Eq. 3. We calculated MFL by NP Net production per sapling (total mass of g year-1
the total leaf mass (ML) in 2009 divided by the leaf current-year needles and non-photosynthetic
organ produced in the current year)
lifespan. It would be better to use ML08 instead ML09
MRNPLM Minimum rate of net production per sapling year-1
because ML09 divided by the leaf lifespan may represent leaf mass required to maintain the current
MFL at 2010 rather than that at 2009, but we could not sapling leaf mass
reliable values of ML08. We assumed that the relationships
Values of 2008 and 2009 of a same abbreviation are distinguished by
between the production rates and sapling size do not have subscripts 08 and 09, respectively, in the main text
temporal change due to very slow growth rates. This
assumption might be valid because the relationship
between NP and plant mass at 2008 were not A. mariesii were greater than the three other species as
significantly different from that between NP and plant trunk height increased (Fig. 1c, d). Thus, the biomass
mass at 2009 for each species (ANCOVA, P [ 0.05). We required for unit height growth was least for P. jezoensis
calculated MRNP per leaf mass (MRNPLM) by dividing var. hondoensis among the four species.
MRNP by ML. MRNPLM was compared among the four Although the allometry of crown width against trunk
species by using a non-parametric Steel–Dwass multiple height had a pattern similar to the allometry of leaf mass
comparison test. and branch mass against trunk height, species differences
Frequently used symbols with definitions and respective were found only in the slope of the allometric equation
units are listed in Table 1. (Fig. 2a; Table 2). The crown width of P. jezoensis var.
hondoensis was narrowest among the four species. Species
differences were also found in the slope of the allometry of
Results root-zone width against trunk height (Fig. 2b; Table 2).
The root-zone width at a given trunk height was greatest
Species differences were not found in the allometric rela- and narrowest in T. diversifolia and P. jezoensis var.
tionship of root mass against trunk height (Fig. 1a; hondoensis, respectively, among the four species. The root-
Table 2). The mean percentage (±SD) of root dry mass to zone width of A. mariesii and A. veitchii was almost the
total sapling mass was 29.5 ± 11.4 % for the four species. same, and was between T. diversifolia and P. jezoensis var.
However, the slope and intercept of allometric relation- hondoensis.
ships of trunk mass, leaf mass and branch mass against Species differences were recognized in the intercept of
trunk height differed among the four species (Fig. 1b–d; the relationship of the net production rate against leaf mass
Table 2). The trunk mass of the two Abies species was (Fig. 3a; Table 2). The net production rate at a given leaf
greater than the two other species as trunk height increased; mass was greatest by P. jezoensis var. hondoensis among
following the trunk mass of the two Abies species, the trunk the four species, followed by A. veitchii. The net produc-
mass was greater in the order T. diversifolia, and P. jezo- tion rate per leaf mass of A. mariesii and T. diversifolia was
ensis var. hondoensis. The leaf mass and branch mass at a almost the same. No species differences were found in
given trunk height was greater in the order T. diversifolia, either slope or intercept for the relationship of net pro-
A. veitchii, P. jezoensis var. hondoensis, and those of duction rate against trunk height (Fig. 3b; Table 2).

123
J Plant Res (2014) 127:329–338
Fig. 1 Allometric relationships
of trunk height against a root
mass, b trunk mass, c leaf mass
and d branch mass for Abies
mariesii (filled square),
A. veitchii (filled circle), Picea
jezoensis var. hondoensis
(unfilled square), Tsuga
diversifolia (unfilled circle).
Table 2 shows regression
parameters of allometric
equations
The mean leaf lifespan differed significantly among the
four species (Kruskal–Wallis test, P \ 0.05), and was
greater in the order A. mariesii (6.8 ± 1.6 years), T. di-
versifolia (6.7 ± 1.2 years), P. jezoensis var. hondoensis
(6.1 ± 1.9 years), A. veitchii (6.0 ± 1.2 years). The leaf
lifespan of A. veitchii was shorter than for T. diversifolia,
but the statistical difference was marginally significant
(Steel–Dwass test, P = 0.063). The other species combi-
nations showed no statistical difference.
We calculated the minimum rate of net production per
leaf mass (MRNPLM) required to maintain the current
sapling leaf mass using a multiple regression equation for
each species (Table 3). The MRNPLM was greatest for
P. jezoensis var. hondoensis (Steel–Dwass test, P \ 0.05),
followed by A. veitchii (Fig. 4). The MRNPLM of A. mari-
esii and T. diversifolia was lower than for the other two
species (Steel–Dwass test, P \ 0.05), and showed no dif-
ference between them.
Discussion
Shade tolerance
A. mariesii and T. diversifolia had wider crowns and larger
leaf and branch mass than the two other species. However,
the net production per leaf mass was greatest by P. jezo-
ensis var. hondoensis with the least leaf mass among the
four species. The greatest net production per leaf mass of
333
P. jezoensis var. hondoensis was counterbalanced by the
smallest sapling leaf mass among the four species, which
resulted in no species difference in the net production per
sapling. However, A. mariesii and T. diversifolia with more
developed crowns maintained the current sapling leaf mass
by a smaller MRNPLM, compared with the two other spe-
cies. Therefore, species differences in crown architecture
may greatly affect the shade tolerance through the
MRNPLM.
The degree of shade tolerance of the four species was
as expected from the results of Kohyama (1983), Ya-
mamoto (1993) and Narukawa and Yamamoto (2001):
A. mariesii [ A. veitchii [ T. diversifolia [ P. jezoensis
var. hondoensis. The results of this study were consistent
with these previous studies in terms of the species order,
except T. diversifolia. The density of P. jezoensis var.
hondoensis is much lower than the three other species in
this forest (Miyajima et al. 2007) and other forests (Mori
and Takeda 2004b; Nishimura et al. 2005), which indi-
cates low survival in closed-canopy conditions due to low
shade tolerance (Mori et al. 2004). The needle shape of
A. mariesii, A. veitchii and T. diversifolia was larger and
flatter than for P. jezoensis var. hondoensis. The three
species can capture light more efficiently within their own
crown because of a larger amount of sapling leaf mass
than P. jezoensis var. hondoensis. Light intensity under
the crown is lower for shade-tolerant species than for
shade-intolerant species because more shade-tolerant
species capture much light by a larger leaf area per tree
123
334 J Plant Res (2014) 127:329–338

Table 2 Standardised major axis regressions for nine relationships

X Y Species a b R2 n t for a F for b

H MR Am 3.28 -4.16 0.75 43 2.6 1.7

H MR Av 3.40 -4.41 0.71 57
H MR Pj 3.23 -4.15 0.82 28
H MR Td 2.90 -3.76 0.79 42

H MT Am 3.33 -4.24 0.84 44 13.0** 5.9***

H MT Av 3.29 -4.21 0.80 58
H MT Pj 2.86 -3.55 0.91 28
H MT Td 2.53 -3.23 0.83 49

H ML Am 2.87 -3.61 0.69 44 7.7* 7.4***

H ML Av 2.27 -2.79 0.61 58
H ML Pj 2.33 -3.08 0.65 28
H ML Td 2.08 -2.38 0.73 50

H MB Am 3.56 -5.01 0.72 44 8.8* 11.4***

H MB Av 3.08 -4.37 0.63 58
H MB Pj 2.72 -3.95 0.69 28
H MB Td 2.57 -3.25 0.72 50

H WC Am 1.17 -0.24 0.73 44 6.3 10.3***

H WC Av 0.99 0.02 0.66 57
H WC Pj 0.94 -0.04 0.63 28
H WC Td 0.87 0.25 0.62 50

H WR Am 1.09 -0.18 0.67 43 0.8 7.6***

H WR Av 1.12 -0.25 0.60 57
H WR Pj 1.06 -0.23 0.61 28
H WR Td 1.20 -0.27 0.54 45

ML NP Am 0.98 -0.38 0.71 44 1.0 5.9***

ML NP Av 1.09 -0.37 0.65 58
ML NP Pj 1.05 -0.21 0.85 28
ML NP Td 1.07 -0.46 0.86 50

H NP Am 2.80 -3.90 0.67 44 4.8 1.0

H NP Av 2.46 -3.39 0.67 58
H NP Pj 2.44 -3.44 0.83 28
H NP Td 2.22 -3.00 0.85 50

D2H MTBR Am 0.92 0.10 0.97 42 1.4 2.5

D2H MTBR Av 0.88 0.13 0.95 57
D2H MTBR Pj 0.91 0.10 0.97 28
2
DH MTBR Td 0.89 0.19 0.95 42
Log–log regression (log Y = a log X ? b, base 10) was used. Differences in slope (a) and intercept (b) of equations among the four species were
tested by the t test and ANCOVA, respectively. The degree of freedom for t test and ANCOVA are 1 and (1, n-2), respectively, where n is the
total number of saplings of the four conifers. Statistical probability of all regressions is P \ 0.001
Am, Abies mariesii; Av, A. veitchii; Pj, Picea jezoensis var. hondoensis; Td, Tsuga diversifolia
* P \ 0.05; ** P \ 0.01; *** P \ 0.001

(Canham et al. 1994; Valladares and Niinemets 2008). By contrast, the high light transmission within a crown
Probably, the same is true for saplings of the four species of P. jezoensis var. hondoensis would result in greater
in this study. increase in photosynthetic production of P. jezoensis var.

123
J Plant Res (2014) 127:329–338 335

Fig. 2 Allometric relationships

of trunk height against a crown
width and b root-zone width for
Abies mariesii (filled square),
A. veitchii (filled circle), Picea
jezoensis var. hondoensis
(unfilled square), Tsuga
diversifolia (unfilled circle).
Table 2 shows regression
parameters of allometric
equations

Fig. 3 Relationships of net

production rate per sapling
against a sapling leaf mass and
b trunk height for Abies mariesii
(filled square), A. veitchii (filled
circle), Picea jezoensis var.
hondoensis (unfilled square),
Tsuga diversifolia (unfilled
circle). Table 2 shows
regression parameters of
equations

Table 3 Results of the multiple regression model for net production of the four species in central Japan
Species a b c F2, n-3 n R2

Am -0.1846 0.3716*** 0.5260*** 105.4*** 42 0.847

Av 0.0376 0.5118*** 0.3607*** 95.4*** 57 0.779
Pj -0.0120 0.5336*** 0.4476*** 149.5*** 28 0.927
Td -0.2188 0.3172*** 0.6027*** 180.9*** 42 0.896
Regression equation is log NP = a ? b log MNL ? c log MTBR09, where log is a common logarithm (base 10), and a, b, c are constants
Am, Abies mariesii; Av, A. veitchii; Pj, Picea jezoensis var. hondoensis; Td, Tsuga diversifolia

hondoensis in high light conditions than for the three other
species (cf. Horn 1971). The trunk height of P. jezoensis
var. hondoensis may increase faster than for the three other
species because the biomass required for unit height
growth was smaller than for the three other species.
Regeneration of P. jezoensis var. hondoensis depends on
disturbance (Mori et al. 2007; Yamamoto 1993). Therefore,
the crown architectural traits of P. jezoensis var. hondo-
ensis reflected the regeneration strategy adapted to high
light conditions. Although a trade-off between shade-tol-
erance and growth in high light conditions through carbon
allocation is often reported for deciduous and evergreen
broad-leaved species (Kohyama 1987; Takahashi et al.
2001), the same was true for the four conifers of this study.
Tree size structure reflects shade tolerance. Shade-tol-
erant and shade-intolerant species often show an L-shaped
and bell-shaped size structure, respectively (Takahashi
2010; Wright et al. 2003). T. diversifolia is considered to
regenerate after disturbance because of the bell-shaped or
discontinuous size structure (Mori et al. 2007; Nishimura
et al. 2005). Contrary to the results of these previous studies,
the size structure of T. diversifolia (taller than 1.3 m) was an
L-shaped pattern with many small trees like A. mariesii at
this study site (our unpublished data). The density of sap-
lings (0.1–1.3 m tall) of T. diversifolia was twice that of
A. veitchii. Therefore, comparing shade tolerance among
the two Abies species and T. diversifolia only from the size
structure is difficult. Seedling establishment greatly affects
size structure through the recruitment process of the popu-
lation. Seedlings of the two Abies species can establish at
various microhabitats, such as soil, fallen logs and mounds,
while most seedlings of T. diversifolia can establish only on

123
336
Fig. 4 Minimum rate of net production per leaf mass (MRNPLM)
required to maintain the current sapling leaf mass of A. mariesii (Am),
A. veitchii (Av), Picea jezoensis var. hondoensis (Pj), T. diversifolia
(Td). The mean value with positive standard deviation is shown for
each species. Values with the same letter are not significantly
different at the 0.05 level according to the non-parametric Steel–
Dwass multiple comparison test
fallen logs (Mori et al. 2004; Narukawa and Yamamoto
2001, 2002). The importance of fallen logs as regeneration
sites is widely recognized for other Picea and Tsuga species
(Anderson and Winterton 1996; Christy and Mack 1984;
Harmon and Franklin 1989; Mckee et al. 1982; Takahashi
1997; Taylor 1990). Lack of suitable microhabitats for
seedling establishment possibly limits the recruitment of
T. diversifolia. Therefore, the bell-shaped or discontinuous
size structure of T. diversifolia, observed in the previous
studies (Mori et al. 2007; Nishimura et al. 2005), may not be
explained by shade tolerance alone.
This study suggests that MRNPLM is a more important
factor for shade tolerance than the net production rates per
leaf mass and per sapling. Walters and Reich (1999)
reviewed shade tolerance. They described that growth in
shade conditions is not important for shade tolerance
because this growth induces reduced future growth by
increasing the respiration rate, and they clarified that long
leaf lifespan and low respiration rate are important for
survival in shade conditions by reducing carbon loss. Some
studies reported that more shade-tolerant species have a
longer leaf lifespan (Baltzer and Thomas 2007; Hallik et al.
2009; King 1994; Poorter and Bongers 2006). For example,
survival of understory saplings positively correlates with
leaf lifespan in a Bolivian tropical rain forest (Sterck et al.
2006). The ranking of leaf lifespan of the four species in
this study agreed with the result of Mori and Takeda
(2004a), except for T. diversifolia that they did not exam-
ine. However, the difference in mean leaf lifespan was
shorter than one year among the four species in this study,
suggesting that species differences in shade tolerance
among the four species cannot be explained much by leaf
lifespan alone. Probably, species differences in MRNPLM
123
J Plant Res (2014) 127:329–338
related to the whole-plant light compensation point (cf.
King 1994), i.e., more shade-tolerant species can maintain
a positive net production at the whole-plant level in darker
conditions. A large sapling leaf mass would compensate for
lower MRNPLM by more shade-tolerant A. mariesii and
T. diversifolia than by A. veitchii and P. jezoensis var.
hondoensis. Therefore, species differences in shade toler-
ance among the four species can be ascribed to not only
leaf lifespan, but also to MRNPLM and sapling leaf mass.
Root morphology
Root architecture is related to edaphic conditions and
absorption of soil nutrients (Fitter and Stickland 1991). The
root-zone area of T. diversifolia, which often dominates at
shallow soil sites, was greater than that of the three other
species. Although trees can avoid root competition at the
top of the soil, where root density is high, by developing
tap-roots, such tree species would suffer from nutrient
deficiency in shallow soils because of failure of root pen-
etration. By contrast, shallow soil hardly affects growth of
species with long lateral roots (Karizumi 1979). Developed
lateral roots are also effective for anchorage at rocky sites
(Kohyama and Grubb 1994; Matsuda et al. 1978; Reubens
et al. 2007). Therefore, T. diversifolia has a belowground
architecture adapted to shallow soil.
T. diversifolia was the highest shade-tolerant species, as
well as A. mariesii, among the four species, suggesting no
trade-off between shade tolerance and root morphology.
Kohyama and Grubb (1994) showed no clear relationship
between aboveground and belowground architecture in
seedlings of 14 species in a warm-temperate rain forest.
Markesteijn and Poorter (2009) also clarified no direct
trade-off between shade tolerance and drought tolerance by
62 tree species in a Bolivian tropical rain forest: shade
tolerance is due to a low growth rate and a low specific leaf
area, and drought tolerance is determined by carbon allo-
cation to roots. However, species differences were not
found in carbon allocation to roots among the four species
in this study. Therefore, root morphology rather than car-
bon allocation to roots may be more important for adap-
tation to edaphic conditions for the four species.
Although this study examined only root-zone area in
terms of belowground architecture, soil water and nutrient
uptake is also affected by other root morphological param-
eters, such as branching pattern (Nibau et al. 2008; Suther-
land and Stillman 1988), root depth (Ho et al. 2005; Poot and
Lambers 2008) and specific root length (length-to-mass
ratio) (Leuschner et al. 2004). The root turnover rate also
influences plant productivity (Keyes and Grier 1981),
although the root turnover rate is affected by soil nutrient
and climatic conditions (Aber et al. 1985; Fitter et al. 1998;
Gill and Jackson 2000). Therefore, examining these root
J Plant Res (2014) 127:329–338
morphological parameters and turnover rate is necessary to
understand species-specific strategies to acquire soil water,
nutrients and biomass production precisely.
Conclusion
This study concludes that (1) more shade-tolerant species had
a more developed crown and a lower MRNPLM among the
four species, and (2) T. diversifolia that often dominates at
shallow soil sites had a wider root-zone area than the three
other species. Therefore, the four coexisting conifer species
differed in shade tolerance and adaptation to edaphic condi-
tions through their aboveground and belowground architec-
tural traits, which results in different regeneration strategies.
Disturbances and canopy gap creation are unpredictable in
time and space. Individuals of shade-tolerant species grow
extremely slowly, but can persist for long periods in closed-
canopy conditions, which contributes to seedling and sapling
bank formation. Species differ in abundance and longevity in
seedling and sapling banks according to their shade tolerance
(Antos et al. 2005; Parish and Antos 2004). Seedling and
sapling banks largely affect future canopy composition.
Therefore, evaluation of interspecific differences in shade
tolerance is important when considering stand dynamics and
species coexistence, and MRNPLM is a good indicator to
evaluate species difference in shade tolerance. Spatial and
temporal heterogeneous disturbance and edaphic conditions
promote coexistence of the four species.
Acknowledgments We are grateful to the handling editor for his
valuable comments. This study was partially supported by grants from
the Ministry of Education, Culture, Sports, Science and Technology,
Japan.
References
Aber JD, Melillo JM, Nadelhoffer KJ, McClaugherty CA, Pastor J
(1985) Fine root turnover in forest ecosystems in relation to
quantity and form of nitrogen availability: a comparison of two
methods. Oecologia 66:317–321
Anderson LJ, Winterton A (1996) Germination as a determinant of
seedling distributions among natural substrates in Picea engel-
mannii (Pinaceae) and Abies lasiocarpa (Pinaceae). Am J Bot
83:112–117
Antos JA, Parish R (2002) Structure and dynamics of a nearly steady-
state subalpine forest in south-central British Columbia, Canada.
Oecologia 130:126–135
Antos JA, Guest HJ, Parish R (2005) The tree seedling bank in an
ancient montane forest: stress tolerators in a productive habitat.
J Ecol 93:536–543
Baltzer JL, Thomas SC (2007) Determinants of whole-plant light
requirements in Bornean rain forest tree saplings. J Ecol
95:1208–1221
Becker P, Castillo A (1990) Root architecture of shrubs and saplings
in the understory of a tropical moist forest in lowland Panama.
Biotropica 22:242–249
337
Canham CD, Finzi AC, Pacala SW, Burbank DH (1994) Causes and
consequences of resource heterogeneity in forests: interspecific
variation in light transmission by canopy trees. Can J For Res
24:337–349
Christy EJ, Mack RN (1984) Variation in demography of juvenile Tsuga
heterophylla across the substratum mosaic. J Ecol 72:75–91
Day RJ (1972) Stand structure, succession, and use of southern
Alverta’s Rocky mountain forest. Ecology 53:472–478
Farrar JL (1995) Trees in Canada. Fitzhenry and Whiteside Limited
and the Canadian Forest Service, Tronto
Fitter AH, Stickland TR (1991) Architectural analysis of plant root
systems 2. Influence of nutrient supply on architecture in
contrasting plant species. New Phytol 118:383–389
Fitter AH, Graves JD, Self GK, Brown TK, Bogie DS, Taylor K
(1998) Root production, turnover and respiration under two
grassland types along an altitudinal gradient: infuence of
temperature and solar radiation. Oecologia 114:20–30
Franklin JF, Maeda T, Ohsumi Y, Matsui M, Yagi H, Hawk GM
(1979) Subalpine coniferous forests of central Honshu, Japan.
Ecol Monogr 49:311–334
Gill RA, Jackson RB (2000) Global patterns of root turnover for
terrestrial ecosystems. New Phytol 147:13–31
Hallik L, Niinemets Ü, Wright IJ (2009) Are species shade and
drought tolerance reflected in leaf-level structural and functional
differentiation in Northern Hemisphere temperate woody flora?
New Phytol 184:257–274
Harmon ME, Franklin JF (1989) Tree seedling on logs in Picea-Tsuga
forests of Oregon and Washington. Ecology 70:48–59
Ho MD, Rosas JC, Brown KM, Lynch JP (2005) Root architectural
tradeoffs for water and phosphorus acquisition. Funct Plant Biol
32:737–748
Horn S (1971) The adaptive geometry of trees. Princeton University
Press, Princeton
Jackson RB, Caldwell MM (1993a) Geostatistical patterns of soil
heterogeneity around individual perennial plants. J Ecol
81:683–692
Jackson RB, Caldwell MM (1993b) The scale of nutrient heteroge-
neity around individual plants and its quantification with
geostatistics. Ecology 74:612–614
Kanzaki M (1984) Regeneration in subalpine coniferous forests.
I. Mosaic structure and regeneration process in a Tsuga
diversifolia forest. Bot Mag Tokyo 97:297–311
Karizumi N (1979) Illustrations of tree roots. Seibundo Shinkosha,
Tokyo (in Japanese)
Keyes MR, Grier CC (1981) Above- and below-ground net produc-
tion in 40-year-old Douglas-fir stands on low and high produc-
tivity sites. Can J For Res 11:599–605
Kimura M (1963) Dynamics of vegetation in relation to soil
development in northern Yatsugatake mountains. Jpn J Bot
18:255–287
King DA (1994) Influence of light level on the growth and
morphology of saplings in a Panamanian forest. Am J Bot
81:948–957
Kohyama T (1980) Growth pattern of Abies mariesii saplings under
conditions of open-growth and suppression. Bot Mag Tokyo
93:13–24
Kohyama T (1983) Seedling stage of two subalpine Abies species in
distinction from sapling stage: a matter-economic analysis. Bot
Mag Tokyo 96:49–65
Kohyama T (1987) Significance of architecture and allometry in
saplings. Funct Ecol 1:399–404
Kohyama T (1991) A functional model describing sapling growth
under a tropical forest canopy. Funct Ecol 5:83–90
Kohyama T, Grubb PJ (1994) Below- and above-ground allometries
of shade-tolerant seedlings in a Japanese warm-temperate rain
forest. Funct Ecol 8:229–236
123
338
Leuschner C, Hertel D, Schmid I, Koch O, Muhs A, Hölscher D
(2004) Stand fine root biomass and fine root morphology in old-
growth beech forests as a function of precipitation and soil
fertility. Plant Soil 258:43–56
Markesteijn L, Poorter L (2009) Seedling root morphology and
biomass allocation of 62 tropical tree species in relation to
drought- and shade-tolerance. J Ecol 97:311–325
Matsuda K, Haruki M, Hasegawa S, Yajima T, Sekine M, Mayama R
(1978) Studies on the Picea glehnii forest (V) Species compo-
sition and regeneration of the southern natural community of the
Picea glehnii on Mt. Hayachine. Jpn J Ecol 28:347–356 (in
Japanese)
McKee A, LaRoi G, Franklin JF (1982) Structure, composition and
reproductive behavior of terrace forests, South Fork Hoh River,
Olympic National Park. In: Starkey EE, Franklin JF, Matthews
JW (eds) Ecological research in national parks of the Pacific
Northwest. Oregon State University, Forest Research Labora-
tory, Oregon, pp 22–29
Miyajima Y, Takahashi K (2007) Changes with altitude of the stand
structure of temperate forests on Mount Norikura, central Japan.
J For Res 12:187–192
Miyajima Y, Sato T, Takahashi K (2007) Altitudinal changes in
vegetation of tree, herb and fern species on Mount Norikura,
central Japan. Veg Sci 24:29–40
Mori A, Takeda H (2004a) Functional relationships between crown
morphology and within-crown characteristics of understory
saplings of three codominant conifers in a subalpine forest in
central Japan. Tree Physiol 24:661–670
Mori A, Takeda H (2004b) Effects of undisturbed canopy structure on
population structure and species coexistence in an old-growth
subalpine forest in central Japan. For Ecol Manage 200:89–100
Mori A, Mizumachi E, Osono T (2004) Doi Y Substrate-associated
seedling recruitment and establishment of major conifer species
in an old-growth subalpine forest in central Japan. For Ecol
Manage 196:287–297
Mori A, Mizumachi E, Komiyama A (2007) Roles of disturbance and
demographic non-equilibrium in species coexistence, inferred
from 25-year dynamics of a late-successional old-growth
subalpine forest. For Ecol Manage 241:74–83
Narukawa Y, Yamamoto S (2001) Gap formation, microsite variation
and the conifer seedling occurrence in a subalpine old-growth
forest, central Japan. Ecol Res 16:617–625
Narukawa Y, Yamamoto S (2002) Effects of dwarf bamboo (Sasa sp.)
and forest floor microsites on conifer seedling recruitment in a
subalpine forest, Japan. For Ecol Manage 163:61–70
Nibau C, Gibbs DJ, Coates JC (2008) Branching out in new
directions: the control of root architecture by lateral root
formation. New Phytol 179:595–614
Nishimura N, Hara T, Kawatani M, Hoshino D, Yamamoto S (2005)
Promotion of species co-existence in old-growth coniferous
forest through interplay of life-history strategy and tree compe-
tition. J Veg Sci 16:549–558
Okitsu S, Ito K, Li CH (1995) Establishment processes and
regeneration patterns of montane virgine coniferous forest in
northern China. J Veg Sci 6:305–308
Parish R, Antos JA (2004) Structure and dynamics of an ancient
montane forest in coastal British Columbia. Oecologia
141:562–576
Paz H (2003) Root/shoot allocation and root architecture in seedlings:
variation among forest sites, microhabitats, and ecological
groups. Biotropica 35:318–332
Peet RK (1981) Forest vegetation of the Colorado Front Range.
Vegetatio 45:3–75
Poorter L, Bongers F (2006) Leaf traits are good predictors of plant
performance across 53 rain forest species. Ecology
87:1733–1743
123
J Plant Res (2014) 127:329–338
Poot P, Lambers H (2008) Shallow-soil endemics: adaptive advan-
tages and constraints of a specialized root-system morphology.
New Phytol 178:371–381
Reubens B, Poesen J, Danjon F, Geudens G, Muys B (2007) The role
of fine and coarse roots in shallow slope stability and soil erosion
control with a focus on root system architecture: a review. Trees
21:385–402
Smithwick EAH, Mack MC, Turner MG, Chapin FS III, Zhu J, Balser
TC (2005) Spatial heterogeneity and soil nitrogen dynamics in a
burned black spruce forest stand: distinct controls at different
scales. Biogeochemistry 76:517–537
Sterck FJ, Poorter L, Schieving F (2006) Leaf traits determine the
growth-survival trade-off across rain forest tree species. Am Nat
167:758–765
Sugita H, Tani M (2001) Difference in microhabitat-related regen-
eration patterns between two subalpine conifers, Tsuga diversi-
folia and Abies mariesii, on Mount Hayachine, northern Honshu,
Japan. Ecol Res 16:423–433
Sutherland WJ, Stillman RA (1988) The foraging tactics of plants.
Oikos 52:239–244
Takahashi K (1994) Effect of size structure, forest floor type and
disturbance regime on tree species composition in a coniferous
forest in Japan. J Ecol 82:769–773
Takahashi K (1997) Regeneration and coexistence of two subalpine
conifer species in relation to dwarf bamboo in the understorey.
J Veg Sci 8:529–536
Takahashi K (2010) Mid-successional stand dynamics in a cool-
temperate conifer-hardwood forest in northern Japan. Plant Ecol
211:159–169
Takahashi K, Lechowicz MJ (2008) Do interspecific differences in
sapling growth traits contribute to the codominance of Acer
saccharum and Fagus grandifolia? Ann Bot 101:103–109
Takahashi K, Rustandi A (2006) Responses of crown development to
canopy openings by saplings of eight tropical submontane forest
tree species in Indonesia: a comparison with cool temperate
trees. Ann Bot 97:559–569
Takahashi K, Seino T, Kohyama T (2001) Responses to canopy
openings in architectural development of saplings in eight
deciduous broad-leaved tree species. Can J For Res
31:1336–1347
Taylor AH (1990) Disturbance and persistence of sitka spruce (Picea
sitchensis (Bong) Carr.) in coastal forests of the Pacific
Northwest, North America. J Biogeogr 17:47–58
Taylor AH, Qin Z, Jie L (1996) Structure and dynamics of subalpine
forests in the Wang Lang Natural Researve, Sichuan, China.
Vegetatio 124:25–38
Valladares F, Niinemets Ü (2008) Shade tolerance, a key plant feature
of complex nature and consequences. Ann Rev Ecol Evol Syst
39:237–257
Veblen TT (1986) Treefalls and the coexistence of conifers in
subalpine forests of the central Rockies. Ecology 67:644–649
Walters MB, Reich PB (1999) Low-light carbon balance and shade
tolerance in the seedlings of woody plants: do winter deciduous
and broad-leaved evergreen species differ? New Phytol
143:143–154
Warton DI, Wright IJ, Falster DS, Westoby M (2006) Bivariate line-
fitting methods for allometry. Biol Rev 33:1–33
Wright JS, Muller-Landau HC, Condit R, Hubbell SP (2003) Gap-
dependent recruitment, realized vital rates, and size distributions
of tropical trees. Ecology 84:3174–3185
Yamamoto S (1993) Gap characteristics and gap regeneration in a
subalpine coniferous forest on Mt Ontake, central Honshu,
Japan. Ecol Res 8:277–285
Yanagisawa N, Fujita N (1999) Different distribution patterns of
woody species on a slope in relation to vertical root distribution
and dynamics of soil moisture profiles. Ecol Res 14:165–177