Tree Physiology 34, 415–425
doi:10.1093/treephys/tpu019
Research paper
Root diameter variations explained by anatomy and phylogeny
of 50 tropical and temperate tree species
Jiacun Gu1,2, Yang Xu1, Xueyun Dong3, Hongfeng Wang1 and Zhengquan Wang1,4
3School
of Forestry, Northeast Forestry University, Harbin 150040, China; 2Center for Ecological Research, Northeast Forestry University, Harbin 150040, China;
of Science, Harbin University, Harbin 150086, China; 4Corresponding author ([email protected], [email protected])
Received August 7, 2013; accepted February 14, 2014; published online April 2, 2014; handling Editor Frederick Meinzer
Root diameter, a critical indicator of root physiological function, varies greatly among tree species, but the underlying mechanism of this high variability is unclear. Here, we sampled 50 tree species across tropical and temperate zones in China, and
measured root morphological and anatomical traits along the first five branch orders in each species. Our objectives were (i)
to reveal the relationships between root diameter, cortical thickness and stele diameter among tree species in tropical and
temperate forests, and (ii) to investigate the relationship of both root morphological and anatomical traits with divergence
time during species radiation. The results showed that root diameter was strongly affected by cortical thickness but less
by stele diameter in both tropical and temperate species. Changes in cortical thickness explained over 90% of variation
in root diameter for the first order, and ~74–87% for the second and third orders. Thicker roots displayed greater cortical
thickness and more cortical cell layers than thinner roots. Phylogenetic analysis demonstrated that root diameter, cortical
thickness and number of cortical cell layers significantly correlated with divergence time at the family level, showing similar
variation trends in geological time. The results also suggested that trees tend to decrease their root cortical thickness rather
than stele diameter during species radiation. The close linkage of variations in root morphology and anatomy to phylogeny
as demonstrated by the data from the 50 tree species should provide some insights into the mechanism of root diameter
­variability among tree species.
Keywords: angiosperm species, root anatomy, root branch order, root morphology, temperate forests, tropical forests.
Introduction
Variability of root morphology among tree species is manifested by diameter, length, specific root length (SRL) and
tissue density, all of which play key roles in root functions
(Eissenstat and Yanai 1997, Pregitzer et al. 2002, Comas et al.
2012). In comparison with other root traits, diameter may be
the most important morphological feature. For example, root
absorptive efficiency mainly depends on diameter, and thinner
roots with lower tissue density and higher SRL may absorb
greater nutrients and water from the soil (Eissenstat and Yanai
1997, Eissenstat et al. 2000, Fitter 2002). In addition, thinner
roots also have shorter lifespans or faster turnover rates (Gill
and Jackson 2000, Wells et al. 2002), and significantly contribute to carbon (C) and nutrient cycles in forest ecosystems
(Norby and Jackson 2000, Joslin et al. 2006). Recent studies
found that root diameter, particularly that of root tips, varied
over 10-fold among tree species in tropical (Chang and Guo
2008) and temperate regions (Valenzuela-Estrada et al. 2008,
Comas and Eissenstat 2009, W. Shi, personal communication).
However, what caused such large variation is yet to be well
identified and explained. Other root-trait studies with multiple
species showed that diameter was closely correlated with
SRL (Comas and Eissenstat 2009), stele diameter (Guo et al.
2008) and the divergence time in evolutionary history (Comas
et al. 2012, Chen et al. 2013). Despite these previous studies,
© The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
1School
416 Gu et al.
Tree Physiology Volume 34, 2014
mentioned above, the main objectives of this study were (i) to
reveal the relationships between root diameter, cortical thickness and stele diameter among tree species in tropical and
temperate forests, and (ii) to investigate the relationship of
both root morphological traits (diameter) and anatomical traits
with divergence time during species radiation.
Materials and methods
Study site
Two study sites were chosen. Site I was located in a tropical
forest in the Jianfengling National Natural Reserve (18°23′–
18°50′N, 108°36′–109°05′E) in Hainan of southern China. This
site has a tropical climate with mean January, July and annual
temperatures of 19.4, 27.3 and 24.5 °C, respectively. The
mean annual precipitation is 2651 mm, of which ~80% occurs
between May and October. The soils are Humic Acrisol (Gong
et al. 1999) with high nutrients. Under such climate conditions,
tropical seasonal rainforest dominates the mountain areas. Site
II was located in a temperate forest at the Maoershan research
station (45°21′–45°25′N, 127°30′–127°34′E) of the Northeast
Forestry University in Heilongjiang of northeastern China. This
site has a continental temperate monsoon climate with mean
January, July and annual temperatures of −19.6, 20.9 and
2.8 °C, respectively. The mean annual precipitation is 723 mm
with 477 mm falling from June to August (Zhou 1994). The
soils are Hap-Boric Luvisols (Gong et al. 1999), well drained
with high organic matter. The site is dominated by the secondgrowth forest regenerated after the old growth of mixed Korean
pine and hardwood forest was harvested over 70 years ago.
Species selection and root sampling
We sampled a total of 50 species, 27 at tropical Site I and 23
at temperate Site II (see Table S1 available as Supplementary
Data at Tree Physiology Online). The 47 hardwoods and three
conifers belong to 19 orders and 28 families, of which four families were common at both study sites. These species included
magnoliids, rosids, malvids, asterids and campanulids for
angiosperms and pinales for gymnosperms, but did not include
basal angiosperms and monocots. The roots were sampled in
August of 2009 for tropical species (Site I) and in late July of
2007 for temperate species (Site II). At both sites, three individual trees were selected for each species, with ages ranging
from 30 to 50 years and diameters at breast height (DBHs)
from 15 to 20 cm. One root sample from each of the three
trees was taken from the top 20-cm soil layer (mineral soil
layer), following the same procedure as in Guo et al. (2008).
Root branches (including more than five branch orders) were
traced to the tree stem and cut from the main lateral woody
roots, which enabled us to identify and sort roots of target
species. Once collected, each root sample was divided into
two subsamples: one was gently washed in deionized water
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
the primary traits governing root diameter variation among tree
species are still not well established.
The cortex and stele are the two main tissues in primary roots
(Esau 1977, Peterson et al. 1999, Lux et al. 2004), and their variations will determine root diameter. However, the extent to which
these two traits influence diameter variation remains unclear. For
example, Hummel et al. (2007) reported that root diameter correlated strongly with cortical (plus rhizodermis) cross-sectional
area but only weakly with xylem diameter in 14 herbaceous
Mediterranean species. In contrast, Guo et al. (2008) showed
that root diameter was closely correlated with stele diameter
rather than cortex thickness in 23 Chinese temperate tree species. One possible reason for such a discrepancy may be that
Hummel et al. (2007) selected only the first-order roots (root
tips), while Guo et al. (2008) included all of the first five order
roots. Considering that the proportion of the cortex decreases
(or disappears) and the stele increases with ascending root
branch orders in woody plants (Guo et al. 2008, ValenzuelaEstrada et al. 2008), we propose our first hypothesis that, for
those lower-order (or absorptive) roots, root diameter is mainly
affected by cortical thickness but less by stele diameter.
Root diameter variability may also be influenced by species
evolutionary history (Fitter 2002, Kenrick 2002). However, it is
very difficult to reconstruct root evolutionary history, due to the
scarcity of root fossils in geological strata (Elick et al. 1998,
Raven and Edwards 2001, Kenrick 2002). Extant species might
provide some insights in this regard (Peterson 1992). Based on
phylogenetic analyses with woody plants, Comas and Eissenstat
(2009) and Comas et al. (2012) suggested that extant angiosperms with lineages diversifying more recently during the
Cretaceous had thinner roots with higher SRL than primitive
angiosperms. Chen et al. (2013) found that the root diameter
of extant species significantly correlated with divergence time at
the family level and decreased from 120 to 64 million years ago,
and was constant afterwards. These findings suggest that root
diameter is substantially influenced by plant phylogeny, although
further tests are needed using fossil evidence when available
(Comas et al. 2012). Anatomy is important for our understanding of plant evolutionary trends (Seago and Fernando 2013), but
it remains unclear whether evolutionary patterns of root diameter are mirrored by anatomical traits. Thus, our second hypothesis is that cortical thickness, number of cortical cell layers and
stele diameter have similar variation patterns to root diameter
during species radiation in geological time.
In the current study, we sampled 50 tree species, 27 from
tropical and 23 from temperate forests (i.e., latitude from 18°
to 45°) in China. These 50 tree species belong to 19 orders
and 28 families. Root morphological traits (diameter, SRL and
tissue density) and anatomical traits (cortical thickness, stele
diameter and number of cortical cell layers) along the first
five branch orders (distal roots numbered as first order) were
measured for each species. By testing the two hypotheses
Tree root diameter variation and phylogeny 417
and immediately fixed in formalin-aceto-alcohol (FAA) solution
(90 ml of 50% ethanol, 5 ml of 100% glacial acetic acid and
5 ml of 37% methanol); the other was immediately put on ice
and transported to the laboratory within 4 h and frozen for dissection and morphological analysis at a later date.
Root morphology and anatomy
Species phylogeny
The phylogeny of the 50 species was constructed based on
the instructions of the APG III system (APG 2009) and the
Angiosperm Phylogeny website (http://www.mobot.org/
MOBOT/Research/APweb/welcome.html) (Stevens 2001). The
free software PHYLOCOM 4.2 (Webb et al. 2008, http://phylodiversity.net/phylocom/) was used to construct the phylogenetic tree (see Figure S1 available as Supplementary Data at
Tree Physiology Online), in which the resolved PHYLOMATIC
tree (R20100701) was used as the backbone for our supertree. The Branch Length Adjuster algorithm in PHYLOCOM in
combination with estimated family ages (Wikström et al. 2001)
was used to adjust all the branch lengths in our phylogenetic
tree. In addition, we obtained divergence times of angiosperm
families from Wikström et al. (2001) based on the molecular
clock theory. We determined the geological time when a family
Data analysis
The means, median, minimum, maximum and coefficient of
variation (CV) were calculated at branch order level for five
root traits, i.e., diameter, SRL, tissue density, cortical thickness
and stele diameter, in both tropical and temperate sites. In
order to test the first hypothesis, regression analysis was used
to determine the relationships between root diameter, cortical
thickness and stele diameter for the first three branch orders
among species at each site, and the relationship between cortical thickness and number of cortical cell layers. All statistical
analysis mentioned above was performed using the SPSS software (2010, V. 19.0: SPSS, Inc., Cary, NC, USA).
Variations in first-order root morphological and anatomical
traits at the family level in geological time, and their correlations with divergence time were determined by piecewise
regression (Chen et al. 2013). Piecewise regression models
are ‘broken-stick’ models, where two or more lines are joined
at unknown points, called ‘breakpoints’ (Toms and Lesperance
2003). Breakpoints can be used as estimates of thresholds,
and were used here to determine the thresholds of divergence
time for root morphological and anatomical traits in geological time (Chen et al. 2013). With this approach, we tested the
second hypothesis in our study. In addition, we used the phylogenetically independent contrasts (PIC) method to test the
dependence of root morphological and anatomical trait variation on divergence time without the influence of phylogeny.
PIC analysis was evaluated with the ‘analysis of traits’ (AOT)
module in PHYLOCOM 4.2, which can calculate the internal
node values for continuous traits (Felsenstein 1985, Webb
et al. 2008). We did not include gymnosperms in PIC analysis
as they evolved in a different lineage and may differ markedly
in root anatomy from angiosperm species. Finally, we calculated the correlation coefficients between root diameter and
anatomical traits for 47 species with the PIC method.
Results
Variations in root morphological and anatomical traits
Root traits showed great variability in statistical characteristics between both tropical and temperate forests, in addition
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
In the laboratory, at least five intact root branches for each
sample were dissected for morphological analysis under a ×10
stereomicroscope (Motic SMZ-140, Xiamen, China), following
the procedure described in Pregitzer et al. (2002) and Wang
et al. (2006), with the distal non-woody roots regarded as firstorder roots. For morphology, roots were separated by order, with
>1000, 500, 150, 50 and 20 roots collected for the first to fifth
order per sample. After dissection, the roots of each subsample
were scanned by order with an EPSON EXPRESSION 10000XL
colored scanner (dots per inch = 400). The mean diameter, total
length and volume for each order of roots were determined with
the root system analyzer software (WinRhizo 2004b, Regent
Instruments, Inc., Québec, Canada). Then these roots were oven
dried (65 °C) to determine constant weight (nearest 0.0001 g).
The SRL for each order was calculated as the total length divided
by the corresponding dry mass (ash-free mass basis). The tissue
density for each order of roots was calculated as the dry mass
(ash-free mass basis) divided by the corresponding total volume.
For root anatomy, roots in each FAA subsample were dissected
carefully by branch order. Twenty roots from each branch order
were stained with safranine-fast green, and then dehydrated in
70, 85, 95 and 100% alcohol, embedded in paraffin; and slides
of 8-μm-thick root sections were prepared with a microtome for
determination of anatomical characteristics (Guo et al. 2008).
The slides were photographed under a compound microscope
(BX51; Olympus, Tokyo, Japan), with root diameter, cortical thickness, stele diameter and number of cortical cell layers being
measured from three cross-sections for each root segment.
and its nearest sister family diverged from their immediate
ancestor as the divergence time for this family. Since the divergence time estimated by Wikström et al. (2001) was given
for the divergence node between two adjacent related genera,
rather than between families, we chose the divergence time of
the earliest genus within a family to represent the divergence
time for the family (Chen et al. 2013). We noted that the phylogenetic tree was constructed by species at the extremes of
south and north distribution in China and did not have any species in the middle. However, the dataset covered species of
both primitive and newly divergent families at each study site.
418 Gu et al.
Effect of anatomical traits on root diameter
Multiple linear regression showed that both cortical thickness and stele diameter influenced root diameters of the first
three orders in tropical and temperate trees, but the effect
of cortical thickness was greater (see Table S2 available as
Supplementary Data at Tree Physiology Online). Root diameter was positively correlated with cortical thickness, which
explained 92 (tropical) and 93% (temperate) of the variations
in root diameter for first order, over 80% for second order and
over 70% for third order (Figure 1). Stele diameter displayed a
relatively weaker correlation, and explained 47–57% (tropical)
and 36–66% (temperate) of the variations in root diameters
among the first three orders (Figure 2).
In root cross-sections, cortical thickness positively correlated with the number of cortical cell layers across the
first three order roots in both tropical and temperate trees
(Figure 3). For example, for tropical trees, Eurya ciliata Merr.
had a first-order root cortical thickness of 55.5 ± 2.7 µm
with three cortical cell layers (Figure 4a; root diameter
of 142 ± 4 µm), while Alseodaphne hainanensis Merr. had
a cortical thickness of 301.1 ± 11.1 µm with 12 cortical
cell layers (Figure 4b; root diameter of 905.7 ± 27.9 µm).
Similarly, for temperate trees, Betula platyphylla Suk. had
a first-order root cortical thickness of 41.3 ± 2.9 µm
with four cortical cell layers (Figure 4c; root diameter of
157 ± 7.1 µm), while P. amurense had a cortical thickness
of 177 ± 6.9 µm with 10 cortical cell layers (Figure 4d; root
diameter of 522 ± 18.8 µm). Overall, 54–77% of the interspecies variation in cortical thickness could be explained by
the number of cortical cell layers (Figure 3).
Table 1. ​Descriptive statistics for five root traits of morphology and anatomy in 50 tropical and temperate tree species measured in this study.
Root trait
Diameter
(mm)
SRL (m g−1)
Tissue density
(g cm−3)
Cortical
thickness
(μm)
Stele diameter
(μm)
Root order
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
4
5
Tropical
Temperate
Minimum
Maximum
Mean
Median
0.135
0.156
0.160
0.162
0.175
3.48
3.24
2.52
1.63
1.13
0.078
0.100
0.117
0.085
0.149
25.00
16.42
0.00
38.97
68.32
81.61
123.47
147.52
1.113
1.218
1.359
1.623
1.878
288.53
220.79
163.49
135.47
26.65
0.317
0.390
0.541
0.551
0.646
362.25
432.92
415.75
261.05
321.41
365.14
1352.87
1505.84
0.422
0.475
0.552
0.678
0.843
73.94
48.09
28.67
16.74
6.76
0.162
0.197
0.242
0.254
0.299
146.47
153.52
146.23
108.38
143.89
203.86
470.11
649.73
0.356
0.408
0.478
0.614
0.822
50.23
36.48
20.02
11.41
5.58
0.150
0.191
0.225
0.246
0.264
131.05
134.55
126.71
91.03
126.41
180.16
422.06
618.52
Tree Physiology Volume 34, 2014
CV
0.607
0.581
0.547
0.504
0.470
0.968
1.011
1.129
1.489
0.770
0.376
0.331
0.387
0.366
0.336
0.615
0.653
0.760
0.528
0.458
0.414
1.804
2.063
Minimum
Maximum
Mean
Median
CV
0.168
0.164
0.212
0.293
0.394
33.70
24.30
12.70
3.77
1.309
0.136
0.136
0.155
0.220
0.183
40.91
14.94
0.00
37.45
59.24
104.62
199.05
315.23
0.520
0.566
0.806
1.530
3.209
203.82
130.22
67.87
34.21
18.96
0.407
0.363
0.524
0.554
0.519
177.74
175.64
206.11
154.55
220.78
354.49
1131.23
2567.17
0.243
0.265
0.340
0.539
0.959
99.83
84.55
45.06
18.05
6.70
0.272
0.263
0.343
0.391
0.384
79.32
71.89
53.24
70.69
110.03
194.94
425.58
758.22
0.219
0.240
0.302
0.431
0.809
95.67
76.17
45.92
16.27
5.82
0.305
0.255
0.351
0.419
0.404
73.08
72.04
25.91
54.53
103.37
178.61
411.15
721.34
0.297
0.345
0.445
0.568
0.682
0.377
0.345
0.382
0.489
0.604
0.321
0.261
0.293
0.239
0.222
0.386
0.525
1.163
0.440
0.333
0.330
1.125
1.986
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
to large differences among the first five orders (Table 1). The
average diameter was greater for tropical trees than for temperate trees. Root diameter in first order varied nearly ninefold between the thinnest (Cratoxylum cochinchinense Lour.:
135 ± 4 µm) and the thickest (Cryptocarya chinensis Hemsl.:
1113 ± 17 µm) for tropical trees, but threefold for temperate
trees (Salix koreensis Anderss.: 148 ± 7 µm vs Phellodendron
amurense Rupr.: 522 ± 19 µm). The average SRL was shorter
and tissue density was less for tropical trees than for temperate trees. The SRL varied over 80-fold between the shortest
(C. chinensis: 3.5 ± 0.3 m g−1) and the longest (C. cochinchinense: 288.5 ± 2.3 m g−1) for tropical trees, but sixfold for
temperate trees (P. amurense: 35.7 ± 0.7 m g−1 vs Euonymus
pauciflorus Sieb.: 203.8 ± 2.1 m g−1).
Cortical thickness and stele diameter were both greater for
tropical trees than for temperate trees (Table 1). In the first
three orders, the average cortical thickness for each order
ranged from 146 to 153 µm for tropical species and from 53
to 79 µm for temperate species. By contrast, stele developed
in all orders of roots, and its diameter increased with ascending root order. The average stele diameter for each order varied between 108 and 203 µm for tropical trees and between
71 and 195 µm for temperate trees.
Tree root diameter variation and phylogeny 419
Phylogenetic variations in root morphological
and anatomical traits
Phylogenetic analysis showed that angiosperms of primitive
families, such as Lauraceae and Magnoliaceae, commonly
had thicker roots in first order (Figure 5). Anatomical results
showed that those species of primitive families not only had
thicker root diameters but also had greater cortical thickness
and thicker stele, as well as a greater number of cortical cell
layers (Figures 4 and 5). Piecewise regression analysis demonstrated that both root morphological and anatomical traits
at the family level may have similar trends of variation in geological time with a consistent decline from the period of 120
to 60 million years before present (MYBP), and level off after
~60 MYBP (Figure 6). Root diameter, cortical thickness and
number of cortical cell layers, but not stele diameter, were significantly correlated with divergence time (R2 = 0.19–0.28, all
P < 0.05; Figure 6). Phylogenetically independent contrasts
also showed that the correlations between divergence time
and root diameter, cortical thickness and number of cortical
cell layers were still significant after the influence of phylogeny was removed (R2 = 0.19–0.33, all P < 0.05; Figure 6). In
addition, when the phylogenetic influence was removed, the
correlation between cortical thickness and root diameter was
strongest among the root morphological and anatomical traits
studied (see Table S3 available as Supplementary Data at Tree
Physiology Online).
Discussion
Relations of root diameter to anatomical traits
Root diameter plays a crucial role in root-driven ecophysiological processes in forest ecosystems. For example, thinner roots
of trees usually have higher mycorrhizal colonization (Reinhardt
and Miller 1990, Brundrett 2002), greater absorptive capacity (Wells and Eissenstat 2003, Dunbabin et al. 2004, Rewald
et al. 2012), shorter lifespan (Eissenstat and Yanai 1997, Kern
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
Figure 1. ​Correlations between root diameter and cortical thickness in the first three orders among 27 tropical and 23 temperate tree species in
China.
420 Gu et al.
et al. 2004, Baddeley and Watson 2005) and faster decomposition (Fehay and Hughes 1994, Schenk and Jackson 2002).
Root diameter varies greatly among tree species, even at the
same site (Guo et al. 2008, Comas and Eissenstat 2009),
but the reasons are still not well understood. Our study demonstrated that root diameter was mainly affected by cortical
thickness and less by stele diameter in a large number (50)
of tropical and temperate tree species (Figures 1 and 2, see
Table S2 available as Supplementary Data at Tree Physiology
Online). Even when the influence of phylogeny was removed,
root diameter was still strongly correlated with cortical thickness (see Table S3 available as Supplementary Data at Tree
Physiology Online). These results support our first hypothesis
that cortical thickness in absorptive roots governs the variation
in root diameter among tree species.
Several possible reasons may help to explain why cortical thickness determines root diameter. First, the root cortex
accounts for a greater proportion of the cross-sectional area of
Tree Physiology Volume 34, 2014
non-woody roots at the species level, as found by this and other
studies (Esau 1977, Lux et al. 2004, Hummel et al. 2007, Guo
et al. 2008). Second, root cortical thickness in woody plants
may affect root diameter primarily by changing the number of
cortical cell layers, as tree species with greater root cortical
thickness have more cortical cell layers (Figure 3) in this study
and others (Zadworny and Eissenstat 2011). Third, genetic
factors may also contribute to the variation in root diameter.
Recent studies of Arabidopsis roots found that two proteins,
SHR (SHORT-ROOT) and SCR (SCARECROW), were key regulators of root radial patterning, including both periclinal and
anticlinal divisions in cortical development (Di Laurenzio et al.
1996, Helariutta et al. 2000, Cui and Benfey 2009). Reports
on crops also showed that roots of different rice and common
bean genotypes had different diameters and numbers of cortical cell layers (Lynch and van Beem 1993, Lafitte et al. 2001,
Burton et al. 2013). Our results suggest that the variation in
root diameter of the 50 tree species studied should reflect the
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
Figure 2. ​Correlations between root diameter and stele diameter in the first three orders among 27 tropical and 23 temperate tree species in
China.
Tree root diameter variation and phylogeny 421
differences not only in root anatomy but also in phylogenetic
lineage (see the discussion below).
Root morphology and anatomy during evolution
Root morphology is closely linked to phylogeny (Comas and
Eissenstat 2009). Comas et al. (2012) suggested that such a
linkage should be an indication of the evolutionary adaptation
of root traits to long-term environmental changes in geological time. Previous studies with extant tree species have found
that early-divergent species on the phylogenetic tree commonly had thicker roots than late-divergent species (Pregitzer
et al. 2002, Comas and Eissenstat 2009, Comas et al. 2012).
Recently, Chen et al. (2013) reported that the root diameter of
angiosperm species at the family level significantly correlated
with divergence time during the period of 120–20 MYBP. Our
estimated pattern of variation for root diameter (Figure 6a)
was consistent with the findings of Chen et al. (2013). More
importantly, our study found that all three anatomical traits
(root cortical thickness, stele diameter and number of cortical
cell layers) had similar thresholds of divergence time to that
of diameter, which suggests that all these traits share similar
trends of variation in geological time (Figure 6b–d). This finding
was also supported by the significant correlation between root
diameter and anatomical traits when the phylogenetic influence
was removed (see Table S3 available as Supplementary Data
at Tree Physiology Online). Thus, as predicted by our second
hypothesis, the variation in tree root diameter is likely to be
associated with changes in anatomical structure during species radiation in geological time.
Patterns of evolution of root morphological traits were likely
related to global climate changes during the past 120 ­million
years, as proposed in some studies (e.g., Comas et al. 2012,
Chen et al. 2013). From the Cretaceous to the Tertiary,
angiosperm species occurred and radiated worldwide (Li and
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
Figure 3. ​Correlations between root cortical thickness and number of cortical cell layers in the first three orders among 27 tropical and 23 temperate tree species in China.
422 Gu et al.
Figure 5. ​Average diameter (a) and cortical thickness (b) of firstorder roots at the family level in phylogeny structure based on the
Angiosperm Phylogeny Group III classification.
Tree Physiology Volume 34, 2014
Johnston 2000, Crepet et al. 2004, Brodribb and Feild 2010),
while the atmospheric carbon dioxide (CO2) concentration
and temperature gradually declined (Savin 1977, Kerp 2002,
Retallack 2002). Such changes led to a relative decrease in
precipitation and an increase in the degree of drought (Tardy
et al. 1989, Beerling 1999). From the mid-Cretaceous to the
late Tertiary (~120–10 MYBP), extant species showed that
newly diverged angiosperm species decreased not only in root
diameter (Comas et al. 2012, Chen et al. 2013, this study)
but also in cortical thickness and number of cortical cell layers
(this study), which may enable roots to take up soil nutrients
and water more efficiently (Comas et al. 2012). Other evidence
also suggests that thinner roots with a thin cortex generally
have high absorptive capacity (Eissenstat and Yanai 1997,
Rieger and Litvin 1999, Wells and Eissenstat 2003, Hishi
2007). On the other hand, leaf stomatal and vein densities in
woody plants increased concomitantly with long-term gradual
declines in atmospheric CO2 to enhance their capacity for photosynthetic CO2 uptake (Royer 2001, Brodribb and Feild 2010),
which in turn may have induced the corresponding changes
in root morphology and anatomy (Comas et al. 2012, Chen
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
Figure 4. ​Typical anatomical structures of first-order roots in Eurya ciliata (a), A. hainanensis (b), B. platyphylla (c) and P. amurense (d). CO, cortex;
EP, epidermis; EX, exodermis; HS, hyphal sheath; VC, vascular cylinder; D, diameter; CT, cortical thickness; NC, number of cortical cell layers.
Tree root diameter variation and phylogeny 423
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
Figure 6. ​Relationships between first-order root traits and divergence time (MYBP) for angiosperm tree species at the family level: (a)–(d) piecewise regression between root traits and divergence time; (e)–(h) PIC correlation between root traits and divergence time.
et al. 2013). Thus, the co-evolution between leaf and root traits
could be an important evolutionary strategy for late-diverging
terrestrial species to adapt to selective pressures associated
with global environmental changes during the past 100 million
years (Raven and Edwards 2001, Westoby and Wright 2006,
Comas et al. 2012).
The results of this study showed that stele diameter was not
significantly correlated with divergence time (Figure 6c). One
possible reason may be that the root stele is entirely surrounded
by cortex. In absorptive roots, the cortex is considered to be
a ‘buffer zone’ to partially isolate the stele from environmental stresses (Lux et al. 2004) such as drought (Kondo et al.
2000), heavy-metal contamination (Reinhardt and Rost 1995)
and salinity (Enstone et al. 2003). In such cases, the stele may
be less sensitive to environmental changes and may show relatively small variation. Another reason is that the stele seems to
have been to be highly conserved in evolution (Sachs 1993,
Roth and Mosbrugger 1996, Seago and Fernando 2013). For
Tree Physiology Online at http://www.treephys.oxfordjournals.org
424 Gu et al.
Conclusion
The current study demonstrates that cortical thickness in
absorptive roots may be the most important factor influencing diameter size in both tropical and temperate trees. Cortical
thickness explained over 90% of the variation in root diameter
for the first order, and over 80 and 70% for the second and
third orders. Thicker roots generally showed greater cortical thickness and more cortical cell layers than thinner roots.
Among extant tree species, root diameter and two anatomical traits (cortical thickness and number of cortical cell layers)
showed significant correlations with divergence time, exhibiting
similar trends. Our results suggest that the variation in tree root
morphological traits should be associated with the change in
anatomical structure during species radiation in geological time.
Supplementary data
Supplementary material is available at Tree Physiology online.
Acknowledgments
The authors thank Xing Wei, Sen Song, Wei Shi, Ying Liu, Jinliang
Liu and Haibo Chen for their help with field and laboratory work,
and Dr Dongmei Jin, Jinlong Zhang and Peili Fu for their help with
statistical analysis. They also thank two anonymous reviewers for
comments and suggestions, which improved the manuscript, and
Dr Harbin Li for editing the manuscript and insightful comments.
Tree Physiology Volume 34, 2014
Conflict of interest
None declared.
Funding
This research was supported by the National Basic Research
Program of China (973 Program: 2012CB416906), the Natural
Science Foundation of China (31100470 and 30130160) and
the Program for Changjiang Scholars and Innovative Research
Team in University (IRT1054).
References
APG (Angiosperm Phylogeny Group) (2009) An update of the angiosperm phylogeny group classification for the orders and families of
flowering plants: APG III. Bot J Linn Soc 161:105–121.
Baddeley J, Watson C (2005) Influences of root diameter, tree age, soil
depth and season on fine root survivorship in Prunus avium. Plant
Soil 276:15–22.
Beerling DJ (1999) Increased terrestrial carbon storage across
the Palaeocene–Eocene boundary. Palaeogeogr Palaeoclimatol
Palaeoecol 161:395–405.
Brodribb TJ, Feild TS (2010) Leaf hydraulic evolution led a surge in
leaf photosynthetic capacity during early angiosperm diversification.
Ecol Lett 3:175–183.
Brundrett MC (2002) Coevolution of roots and mycorrhizas of land
plants. New Phytol 154:275–304.
Burton AL, Lynch JP, Brown KM (2013) Spatial distribution and phenotypic variation in root cortical aerenchyma of maize (Zea mays L.).
Plant Soil 367:263–274.
Chang WJ, Guo DL (2008) Variation in root diameter among 45 common tree species in temperate, subtropical and tropical forests in
China. J Plant Ecol (Chinese Version) 32:1248–1257.
Chen WL, Zeng H, Eissenstat DM, Guo DL (2013) Variation of firstorder root traits across climatic gradients and evolutionary trends in
geological time. Glob Ecol Biogeogr 22:846–856.
Comas LH, Eissenstat DM (2009) Patterns in root trait variation
among 25 co-existing North American forest species. New Phytol
182:919–928.
Comas LH, Mueller KE, Taylor LL, Midford PE, Callahan HS, Beerlingz
DJ (2012) Evolutionary patterns and biogeochemical significance of
angiosperm root traits. Int J Plant Sci 173:584–595.
Crepet WL, Nixon K, Gandolfo D (2004) Fossil evidence and phylogeny:
the age of major angiosperm clades based on mesofossil and macrofossil evidence from Cretaceous deposits. Am J Bot 91:1666–1682.
Cui H, Benfey PN (2009) Cortex proliferation: simple phenotype, complex regulatory mechanisms. Plant Signal Behav 4:551–553.
Di Laurenzio L, Wysocka-Diller J, Malamy JE, Pysh L, Helariutta Y, Freshour
G, Hahn MG, Feldmann KA, Benfey PN (1996) The SCARECROW gene
regulates an asymmetric cell division that is essential for generating
the radial organization of the Arabidopsis root. Cell 86:423–433.
Dunbabin VM, Rengel Z, Diggle AJ (2004) Simulating form and function of root systems: efficiency of nitrate uptake is dependent on
root system architecture and the spatial and temporal variability of
nitrate supply. Funct Ecol 18:204–211.
Eissenstat DM, Yanai RD (1997) The ecology of root lifespan. Adv Ecol
Res 27:1–60.
Eissenstat DM, Wells CE, Yanai RD, Whitbeck JL (2000) Building roots
in a changing environment: implications for root longevity. New
Phytol 147:33–42.
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
example, the average cortical thickness of species before and
after 60 MYBP decreased by 34%, while stele diameter only
decreased by 24% in our study (data not shown). Therefore,
it seems to be a root evolutionary strategy that trees may tend
to decrease their cortical thickness more strongly than stele
diameter during species radiation in geological time.
Caution is needed in the interpretation of our results on evolutionary patterns of root traits through geological time. First,
as it was limited by the number of species (families) sampled,
especially the early-divergent species, the relationship between
root diameter and divergence time established by piecewise
regression (Figure 6) may be weak. Thus, more species representing early divergence times, such as basal angiosperms
(e.g., Amborellales, Austrobaileyales and Chloranthales), should
be included in future studies, in order to improve the resolution of patterns of root-trait evolution. Second, even though our
results provided a clear comparison of the differences in root
traits between typical tropical and temperate species, species
in the transitional zone were lacking, which may lead to phylogenetic analysis overweighted by species at the extremes
of the distribution. Future studies should sample more species across widely distributed study sites with replication, and
under a broad range of environmental conditions.
Tree root diameter variation and phylogeny 425
Pregitzer KS, De Forest JL, Burton AJ, Allen MF, Ruess RW, Hendrick
RL (2002) Fine root architecture of nine North American trees. Ecol
Monogr 72:293–309.
Raven JA, Edwards D (2001) Roots: evolutionary origins and biogeochemical significance. J Exp Bot 52:381–401.
Reinhardt DH, Rost TL (1995) Salinity accelerates endodermal development and induces an exodermis in cotton seedling roots. Environ
Exp Bot 35:563–574.
Reinhardt DR, Miller RM (1990) Size classes of root diameter and
mycorrhizal fungal colonization in two temperate grassland communities. New Phytol 116:129–136.
Retallack GJ (2002) Carbon dioxide and climate over the past 300Myr.
Philos Trans R Soc Lond A 360:659–673.
Rewald B, Raveh E, Gendler T, Ephrath JE, Rachmilevitch S (2012)
Phenotypic plasticity and water flux rates of Citrus root orders under
salinity. J Exp Bot 63:2717–2727.
Rieger M, Litvin P (1999) Root system hydraulic conductivity in species with contrasting root anatomy. J Exp Bot 50:201–209.
Roth A, Mosbrugger V (1996) Numerical studies of water conduction in
land plants: evolution of early stele types. Paleobiology 22:411–421.
Royer DL (2001) Stomatal density and stomatal index as indicators
of paleoatmospheric CO2 concentration. Rev Palaeobot Palynol
114:1–28.
Sachs T (1993) A developmental perspective of the evolution of the
stele. Int J Plant Sci 154:225–227.
Savin SM (1977) The history of the earth’s surface temperature during
the past 100 million years. Ann Rev Earth Planet Sci 5:319–355.
Schenk HJ, Jackson RB (2002) The global biogeography of roots. Ecol
Monogr 72:311–328.
Seago JL Jr, Fernando DO (2013) Anatomical aspects of angiosperm
root evolution. Ann Bot 112:223–238.
Stevens PF (2001). Angiosperm phylogeny website. Version 12, July
2012. http://www.mobot.org/MOBOT/research/APweb/ (10 August
2012, date last accessed).
Tardy Y, N’Kounkou R, Probst J-L (1989) The global water cycle and
continental erosion during Phanerozoic time (570 my). Am J Sci
289:455–483.
Toms JD, Lesperance ML (2003) Piecewise regression: a tool for identifying ecological thresholds. Ecology 84:2034–2041.
Valenzuela-Estrada LR, Vera-Caraballo V, Ruth LE, Eissenstat DM
(2008) Root anatomy, morphology, and longevity among root orders
in Vaccinium corymbosum (Ericaceae). Am J Bot 95:1506–1514.
Wang ZQ, Guo DL, Wang XR, Gu JC, Mei L (2006) Fine root architecture, morphology, and biomass of different branch orders of two
Chinese temperate tree species. Plant Soil 288:155–171.
Webb CO, Ackerly DD, Kembel SW (2008) Phylocom: software for
the analysis of phylogenetic community structure and trait evolution.
Bioinformatics 24:2098–2100.
Wells CE, Eissenstat DM (2003) Beyond the roots of young seedlings: the influence of age and order on fine root physiology. J Plant
Growth Regul 21:324–334.
Wells CE, Glenn DM, Eissenstat DM (2002) Changes in the risk of
fine-root mortality with age: a case study in peach, Prunus persica
(Rosaceae). Am J Bot 89:79–87.
Westoby M, Wright IJ (2006) Land-plant ecology on the basis of functional traits. Trends Ecol Evol 21:261–268.
Wikström N, Savolainen V, Chase MW (2001) Evolution of the angiosperms: calibrating the family tree. Proc R Soc B Biol Sci 268:
2211–2220.
Zadworny M, Eissenstat DM (2011) Contrasting the morphology, anatomy and fungal colonization of new pioneer and fibrous roots. New
Phytol 190:213–221.
Zhou XF (1994) Long-term research on China’s forest ecosystems.
Northeast Forestry University Press, Harbin, China.
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from http://treephys.oxfordjournals.org/ by guest on November 7, 2016
Elick JM, Driese SG, Mora CI (1998) Very large plant and root traces
from the early to middle Devonian: implications for early terrestrial
ecosystems and atmospheric p(CO2). Geology 26:143–146.
Enstone DE, Peterson CA, Ma F (2003) Root endodermis and exodermis: structure, function, and responses to the environment. J Plant
Growth Regul 21:335–351.
Esau K (1977) Anatomy of seed plants. 2nd edn. Wiley, New York, pp
215–255.
Fehay TJ, Hughes JW (1994) Fine root dynamics in a northern hardwood forest ecosystem, Hubbard Brook Experimental Forest, NH. J
Ecol 82:533–548.
Felsenstein J (1985) Phylogenies and the comparative method. Am
Nat 125:1–15.
Fitter AH (2002) Characteristics and functions of root systems. In:
Waisel Y, Eshel E, Kafkafi U (eds) Plant roots: the hidden half. 3rd
edn. Dekker Press, New York, USA, pp 15–32.
Gill RA, Jackson RB (2000) Global patterns of root turnover for terrestrial ecosystems. New Phytol 147:13–31.
Gong ZT, Chen ZC, Luo GB, Zhang GL, Zhao WJ (1999) Soil reference
with Chinese soil taxonomy. Soils 31:57–63.
Guo DL, Xia MX, Wei X, Chang WJ, Liu Y, Wang ZQ (2008) Anatomical
traits associated with absorption and mycorrhizal colonization are
linked to root branch order in twenty-three Chinese temperate tree
species. New Phytol 180:673–683.
Helariutta Y, Fukaki H, Wysocka-Diller J, Nakajima K, Jung J, Sena G,
Hauser M-T, Benfey PN (2000) The SHORT-ROOT gene controls
radial patterning of the Arabidopsis root through radial signaling.
Cell 101:555–567.
Hishi T (2007) Heterogeneity of individual roots within the fine root
architecture: causal links between physiological and ecosystem
functions. J For Res 12:126–133.
Hummel I, Vile D, Violle C, Devaux J, Ricci B, Blanchard A, Garnier E,
Roumet C (2007) Relating root structure and anatomy to wholeplant functioning in 14 herbaceous Mediterranean species. New
Phytol 173:313–321.
Joslin JD, Gaudinski JB, Torn MS, Riley WJ, Hanson PJ (2006) Fine-root
turnover patterns and their relationship to root diameter and soil
depth in a 14C-labeled hardwood forest. New Phytol 172:523–535.
Kenrick P (2002) The origin of roots. In: Waisel Y, Eshel E, Kafkafi U
(eds) Plant roots: the hidden half. 3rd edn. Dekker Press, New York,
USA, pp 1–14.
Kern CC, Friend AL, Johnson JM-F, Coleman MD (2004) Fine root
dynamics in a developing Populus deltoids plantation. Tree Physiol
24:651–660.
Kerp JHF (2002) Comment on ‘300-million-year record of atmospheric
carbon dioxide from fossil plant cuticles’. Nature 415:38.
Kondo M, Aguilar A, Abe J, Morita S (2000) Anatomy of nodal roots in
tropical upland and lowland rice varieties. Plant Prod Sci 3:437–445.
Lafitte HR, Champoux MC, McLaren G, O’Toole JC (2001) Rice root
morphological traits are related to isozyme group and adaptation.
Field Crops Res 71:57–70.
Li P, Johnston MO (2000) Heterochrony in plant evolutionary studies
through the twentieth century. Bot Rev 66:57–88.
Lux A, Luxova M, Abe J, Morita S (2004) Root cortex: structural and
functional variability and responses to environmental stress. Root
Res 13:117–131.
Lynch J, van Beem JJ (1993) Growth and architecture of seedling roots
of common bean genotypes. Crop Sci 33:1253–1257.
Norby RJ, Jackson RB (2000) Root dynamics and global change: seeking an ecosystem perspective. New Phytol 147:3–12.
Peterson CA, Enstone DE, Taylor JH (1999) Pine root structure and
its potential significance for root function. Plant Soil 217:205–213.
Peterson RL (1992) Adaptations of root structure in relation to biotic
and abiotic factors. Can J Bot 70:661–675.