Functional Ecology 2014, 28, 108–123
doi: 10.1111/1365-2435.12150
Leaf form and photosynthetic physiology of Dryopteris
species distributed along light gradients in eastern
North America
Emily B. Sessa*,† and Thomas J. Givnish
Department of Botany, University of Wisconsin-Madison, 430 Lincoln Drive, Madison, WI 53706 USA
Summary
1. Despite the ubiquity of ferns and at least tacit recognition by botanists that their physiology is unique among land plants, most studies on fern physiology have focused on only a few,
locally distributed, usually distantly related species. No previous study has attempted to
examine physiological adaptations in a group of widespread taxa that are closely related and
whose relationships are well understood. Here we report leaf form and physiological measures
for such a group, the 11 eastern North American species of Dryopteris (Dryopteridaceae), and
examine differences in these parameters for evidence of adaptation to light availability.
2. Economic theory predicts that species from sunnier habitats should have narrower, more
steeply inclined leaves, lower specific leaf area, higher stomatal density, higher rates of maximum photosynthesis, respiration and stomatal conductance, and require more light to saturate
photosynthesis. Species should show adaptive crossover in net carbon uptake per unit leaf
mass, with a relative advantage by sun-associated species at high photon flux densities (PFDs)
and by shade-associated species at low PFDs.
3. Field studies allow us to begin characterizing the range of native light environments
occupied by members of this group, and to examine interspecific variation in several aspects of
leaf form and photosynthetic light response for evidence of adaption to light availability. We
also present a novel means for incorporating phylogeny in tests of correlated evolution in a
reticulate lineage.
4. Synthesis. Observed trends in physiology and morphology generally agree with qualitative
predictions but are often not statistically significant. We found no support for adaptive crossover in mass-based carbon uptake, and thus for light availability being the most important variable driving morphological and physiological adaptation in these ferns. We propose that
hydraulic factors related to water balance may have played a larger role in determining their
morphological and physiological variation. Allopolyploid hybrids did not show transgression
in any physiological parameter that may have allowed them to coexist regionally with their
parents. The results of our phylogenetically structured analyses highlight the importance of
incorporating phylogeny into comparative studies, particularly when hybrid or polyploid taxa
are present.
Key-words: comparative methods, ferns, gas exchange, hydraulics, light regimes, photosynthetic light response, polyploidy, transgressive trait expression
Introduction
Ferns bridge a critical functional gap in land plant evolution between nonvascular bryophytes and seed-bearing
*Correspondence author. E-mail emilysessa@ufl.edu
†
Present address. Department of Biology, University of Florida,
Gainesville, FL 32611, USA.
vascular plants. They are second only to angiosperms in
species diversity among land plants (Smith et al. 2006) and
inhabit a large range of climates, substrates and light
regimes, both in habitats where flowering plants dominate
and where few angiosperms can survive (Page 2002).
Despite their broad ecological success and key position in
land plant evolution, relatively little is known about the
physiological traits that have driven fern diversification or
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society
Leaf form and photosynthesis in Dryopteris 109
allowed them to persist despite limiting constraints of their
biology. Ferns are characterized ecologically by an independent, ephemeral gametophyte stage (Watkins, Mack &
Mulkey 2007), low rates of photosynthesis, vascular tissue
development and leaf hydraulic conductance (Brodribb &
Holbrook 2004; Brodribb, Feild & Jordan 2007; Boyce
et al. 2009; Watkins, Holbrook & Zwieniecki 2010), relatively high photosynthetic capacity in low light (Page
2002), slow plant growth rates (Page 2002) and low stomatal densities (Franks & Farquhar 2006; Kessler et al.
2007).
Biologists have traditionally viewed ferns as once-dominant organisms of ancient ecosystems that were quickly
out-competed and relegated to the shadows following the
evolution of seed plants in the Devonian, and especially
after the rise of flowering plants in the Cretaceous (Niklas,
Tiffney & Knoll 1983; Lupia, Lidgard & Crane 1999; Nagalingum et al. 2002). Recent research has challenged this
view (Rothwell 1996), and studies based on fossil and
molecular dating evidence have demonstrated that ferns
underwent their major radiation into modern lineages only
after the appearance of the angiosperms in the Cretaceous
(Schneider et al. 2004; Schuettpelz & Pryer 2009). Recent
research has identified several physiological features of
ferns that may have contributed to their success in various
terrestrial ecosystems, including cavitation-resistant, tracheid-based xylem (Watkins, Holbrook & Zwieniecki 2010;
Pittermann et al. 2011), a host of adaptations to xeric habitats (Hietz 2010), and chimeric photoreceptors that
enhance plastid movement in response to light at low fluences (Kawai et al. 2003; Suetsugu et al. 2005; Kanegae
et al. 2006). Chimeric photoreceptors combine phytochrome and phototropin functionality to allow red light,
which is typically enriched relative to blue light in forest
understoreys (Endler 1993), to trigger phototrophic
responses and chloroplast relocation, which are typically
initiated only by blue or UV light. Arens (1997) found that
several leaf characters in tree ferns are strongly correlated
with light environment; together these data suggest that
traits related to light response may have played an important role in shaping fern evolution. However, no study has
yet attempted to test the extent to which adaptation to different light environments may drive the evolution of morphological and physiological characters within a fern
lineage, or to trace the evolution of such traits in a phylogenetic framework. A high sensitivity of plastid movement
to light, for example, opens the possibility that tight correlations among the photosynthetic parameters Amax, R and
k (maximum photosynthetic rate, dark respiration rate and
the amount of light required to half-saturate photosynthesis, respectively) seen in many angiosperms (e.g. see
Bj€
orkman 1981; Givnish 1988; Walters & Reich 1996;
Givnish, Montgomery & Goldstein 2004; Chen et al. 2011)
might break down in ferns, with plastid motion affecting
Amax strongly and k and R more weakly, if at all. Alternatively, the low leaf hydraulic conductance of ferns may
make differential modulation of hydraulic traits a greater
constraint on survival and competitive success under different light regimes than variation in photosynthetic
parameters. A comparative study of the evolution of photosynthetic parameters among related fern species native
to different light regimes should thus be illuminating,
regardless of whether the traditional patterns seen in flowering plants are observed in ferns or not.
A better understanding of traits that drive adaptation to
different environments and lead to niche differentiation
may also allow us to evaluate possible mechanisms for the
persistence of hybrid and polyploid taxa, which are extremely common in ferns (Manton 1950; Wood et al. 2009).
Newly formed polyploids are subject to minority cytotype
exclusion, in which polyploid cytotypes experience a frequency-dependent mating disadvantage and waste reproductive effort in ineffective matings with the established
parental cytotype (Levin 1975). Niche differentiation
between progenitors and offspring has been suggested as a
potential mechanism allowing polyploids to persist regionally with their parental species on evolutionary time-scales
(Rodriguez 1996). In particular, ploidy increases are likely
to increase cell volume (Melaragno, Mehrota & Coleman
1993; Comai 2005), potentially increasing values for traits
such as specific leaf area (SLA, cm2 g1), as well as tracheid diameters and thus hydraulic conductance. Such
increases may have important implications for light
capture and water loss, potentially making hybrids less
susceptible to internal self-shading and drought.
The 11 eastern North American members of the woodfern genus, Dryopteris, are an ideal group for addressing
these questions. These species vary extensively in leaf morphology and habitat, from species found only on brightly
lit, well-drained cliff faces and possessing narrow, short,
erect fronds (e.g. D. fragrans), to species of heavily shaded,
mesic forest understoreys that often have relatively broad,
tall, sprawling fronds (e.g. D. goldiana; Fig. 1). Some of
these species also occur in Europe and Asia, but eight are
endemic to eastern North America. Sessa, Zimmer &
Givnish (2012a) recently demonstrated that the 11 North
American species are not each others’ closest relatives, and
that each of the six diploids in the group is more closely
related to Asian or European species than to other North
American diploid taxa. The remaining five eastern North
American species are allopolyploids, formed over the last
13 million years via hybridization and genome doubling
involving three of the extant North American diploids and
one extinct progenitor (Sessa, Zimmer & Givnish 2012b;
Table 1). Tani & Kudo (2003, 2005), and Tessier & Bornn
(2007) have examined biomass allocation and the contributions of overwintering leaves to carbon uptake in several
Dryopteris species, and R€
unk, Zobel & Zobel (2010) and
R€
unk & Zobel (2007) have compared growth and survival of
a few European species as a function of edaphic conditions
and light availability, but no rigorous study of photosynthetic adaptations to light regime has been undertaken for
the North American group, despite their strikingly different
habitats and leaf morphologies.
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
110 E. B. Sessa & T. J. Givnish
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
Fig. 1. Eleven species of eastern North
American Dryopteris included in the current study. (a) D. campyloptera, (b) D. carthusiana, (c) D. celsa, (d) D. clintoniana, (e)
D. cristata, (f) D. expansa, (g) D. fragrans,
(h) D. goldiana, (i) D. intermedia, (j)
D. ludoviciana, (k) D. marginalis.
Table 1. North American allopolyploid Dryopteris species and
their putative progenitors, with ploidy indicated. The ploidy of the
extinct progenitor “D. semicristata” is unknown, but it is hypothesized to have been diploid (Montgomery & Wagner 1993)
Allopolyploid
Putative parents
D. campyloptera (49)
D. carthusiana (49)
D. expansa (29), D. intermedia (29)
D. intermedia (29), “D. semicristata”
(29)
D. goldiana (29), D. ludoviciana
(29)
D. cristata (49), D. goldiana (29)
D. ludoviciana (29), “D. semicristata”
(29)
D. celsa (49)
D. clintoniana (69)
D. cristata (49)
In this paper, we present gas exchange data from field
studies and tests for correlated evolution of several morphological and physiological traits with light regime, which
we predict based on the following three hypotheses:
1. Species from brighter environments should have narrower, more steeply inclined leaves, lower SLA
(cm2 g1) and higher maximum photosynthetic rates
(Amax, lmol CO2 m2 per s), light compensation points
[the instantaneous light compensation point, ICP and
the leaf compensation point, LCP, the latter incorporating night-time respiration and leaf construction costs,
both measured in photon flux density (PFD, lmol photons m2 per s)], dark respiration rates (R, lmol
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 111
CO2 m2 per s), maximum stomatal conductances (gs,
mol H2O m2 per s) and stomatal densities (per mm2)
compared to species from shadier environments
(Bj€
orkman, Ludlow & Morrow 1972; Bj€
orkman 1981;
Givnish 1988; Givnish, Montgomery & Goldstein 2004);
2. Species should show adaptive crossover in net carbon
uptake per unit leaf mass as a function of PFD, with
sun-associated species outperforming others at high
PFDs and shade-associated species outperforming others at lower PFDs (Bj€
orkman, Ludlow & Morrow 1972;
Givnish 1988, 1995; Givnish, Montgomery & Goldstein
2004); and
3. Allopolyploid species should exhibit physiological,
morphological or ecological traits that fall outside the
range of variation exhibited by their parental taxa [i.e.
they should be transgressive relative to their progenitors
(Rieseberg, Archer & Wayne 1999)], and doing so may
enable them to coexist regionally with those progenitors
in the long term.
The rationale for these predictions is as follows: in
brightly lit sites with high heat loads and/or vapour pressure deficits, plants must balance increased rates of photosynthetic carbon gain due to increased light and
temperature with greater water loss and low leaf water
potentials, which can lead to desiccation. Narrower leaves
are expected in such environments in order to maintain
high rates of photosynthesis while reducing water loss and
high leaf temperatures (Givnish & Vermeij 1976; Givnish
1979), as are more steeply inclined leaves, which reduce
heat load and transpiration by decreasing the angle of incident radiation on the leaf (Givnish 1979). For deeply
divided foliage, however, such as most Dryopteris share,
the main effect of shorter, narrower fronds may be to
reduce the drop in local leaf water potential across the
length and breadth of the frond, favouring narrow, short
fronds in sunny microsites with high evaporative loads,
and broader, taller fronds in shadier microsites. Greater
stature should also increase competitive ability, especially
in sites with dense coverage by competitors (Givnish 1982,
1995). SLA is a measure of leaf area per unit dry mass;
thicker leaves or leaves with denser tissue have more dry
biomass per unit area and thus lower levels of SLA (Evans
& Poorter 2001; Wright et al. 2004; Shipley et al. 2005).
SLA is positively correlated with high photosynthetic rates
per unit leaf mass and high rates of whole-plant growth
(Shipley 1995; Kruger & Volin 2006; Shipley et al. 2006)
and is expected to decline with increasing light and
decreasing moisture availability (Givnish 1978, 1988; Ackerly et al. 2002). According to classical economic theory
(Bj€
orkman 1981; Givnish 1988), maximum photosynthetic
rates (Amax) are expected to be higher in species from
brighter environments, as are dark respiration rates (R),
the amount of light required to half-saturate photosynthesis (k) and the amount of light needed to balance leaf
respiration on an instantaneous basis (ICP, instantaneous
light compensation point; Givnish, Montgomery &
Goldstein 2004). In C3 plants, stomatal conductance (gs) –
determined by stomatal size, depth and density per unit
area (Franks & Beerling 2009b; Franks, Drake & Beerling
2009; Taylor et al. 2012) – is expected to be greater in
brighter, moister and/or more fertile microsites (Givnish &
Vermeij 1976; Cowan & Farquhar 1977; Wong, Cowan &
Farquhar 1979; Farquhar & Sharkey 1982; Givnish 1986;
Franks & Beerling 2009a; Taylor et al. 2012) in order to
maximize photosynthetic rate at a given rate or cost of
water loss, and lower in drier microsites in order to protect
the hydraulic pathway leading from the roots to the leaves
from cavitation (Brodribb & Holbrook 2004, 2006). Across
C3 plants, stomatal density and individual size are negatively correlated, and high stomatal conductance is positively correlated with stomatal density and photosynthetic
capacity (Franks & Beerling 2009a,b; Franks, Drake &
Beerling 2009; Taylor et al. 2012). In general, we might
expect photosynthesis per unit leaf mass to be higher in
species native to sunnier microsites in comparisons made
at higher PFDs, and higher in species from shadier
microsites in comparisons made at lower PFDs (Bj€
orkman,
Ludlow & Morrow 1972; Givnish 1988, 1995; Givnish,
Montgomery & Goldstein 2004).
For hybrid allopolyploid taxa, it is possible that the
expression of certain morphological, physiological or ecological traits extends significantly beyond that seen in one
or both parents. Successful polyploids are often those that
inhabit an ecological niche different from either parent
(Schwarzbach, Donovan & Rieseberg 2001; Maherali,
Walden & Husband 2009), or that display phenotypic traits
which are outside the range exhibited by the parents [i.e.
transgressive character expression (Rieseberg et al. 2003)],
especially if their distributions overlap. The conspicuous
success of allopolyploid taxa in the Dryopteris complex,
several of which occupy ranges that are transgressive relative to one or both parents (Sessa, Zimmer & Givnish
2012b), may be due to such transgressive expression, which
would have allowed them to avoid minority cytotype exclusion following polyploidization (Abbott & Lowe 2004;
Quintanilla & Escudero 2006; Jimenez et al. 2009).
Here we present data on native light regime, leaf form
and photosynthetic light response from field studies of 11
Dryopteris species in eastern North America. These data
allow us to test the hypotheses described above, in order
to determine whether leaf traits and photosynthetic parameters show the expected patterns of correlated evolution
with photosynthetic light availability in these species’
native habitats.
Materials and methods
SPECIES, STUDY SITES AND LIGHT REGIMES
For 10 species of eastern North American Dryopteris – D. campyloptera*, D. carthusiana*, D. celsa*, D. clintoniana*, D. cristata*,
D. expansa, D. fragrans, D. goldiana, D. ludoviciana, and D. marginalis – we located one field population and measured 5–10 plants
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
112 E. B. Sessa & T. J. Givnish
Table 2. Locations of field populations of Dryopteris measured in this study
Species
Location of study site
GPS coordinates
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
Bear Pen Gap, Blue Ridge Parkway, North Carolina
Huron Mountains, Big Bay, Michigan
Private property, Hertford, North Carolina
Thurber Preserve, Dryden, New York
Huron Mountains, Big Bay, Michigan
Chequamegon-Nicolet National Forest, Wisconsin
Huron Mountains, Big Bay, Michigan
Fernwood Gardens, Niles, Michigan
Huron Mountains, Big Bay, Michigan
Rocky Hock Swamp Preserve, Hertford, North Carolina
Private property, Madison, New York
35°19′16″N, 82°53′25″W
46°53′22″N, 87°53′56″W
36°13′46″N, 76°33′53″W
42°33′17″N, 76°″W
46°53′22″N, 87°53′56″W
46°17′46″N, 90°49′40″W
46°53′08″N, 87°54′22″W
41°51′55″N, 86°20′52″W
46°53′22″N, 87°53′56″W
36°11′19″N, 76°40′37″W
42°52′07″N, 75°30′57″W
campyloptera
carthusiana
celsa
clintoniana
cristata
expansa
fragrans
goldiana
intermedia
ludoviciana
marginalis
in each population (Table 2). For an eleventh species, D. intermedia, we made measurements on 10 individuals at two separate
populations in the Huron Mountains in Michigan. Allotetraploid
species are indicated by asterisks. Parental taxa are listed in
Table 1.
We assessed light regimes in the field using digital hemispheric
photographs taken with a Nikon Coolpix 5400 digital camera
(Melville, NY, USA) with a fisheye lens attachment. Photographs
were taken immediately above each plant and were analysed using
Gap Light Analyzer (Frazer et al. 1999) to estimate the percentage
of total (direct plus indirect) light transmittance through the
canopy.
GAS EXCHANGE MEASUREMENTS
We measured photosynthetic light response on attached, fully
expanded leaves using an LI-6400 portable photosynthesis system
(Li-Cor, Lincoln, NE, USA) equipped with a red/blue LED light
source and a CO2 mixing system. All measurements were made
under ambient relative humidity (c. 55–65%), leaf temperature (c.
20–26 °C) and CO2 (400 lmol per mol) between 07.00 and
13.00 h. Leaves were clamped into the measurement cuvette and
allowed to equilibrate at a moderate light intensity (200 lmol m2
per s). We then measured photosynthesis at the following PFD
values: 450, 750, 1000, 500, 260, 150, 90, 60, 40, 20, 10 and
0 lmol m2 per s and fit these data to the Michaelis–Menten
eqn (1) following Givnish, Montgomery & Goldstein (2004).
AðPFDÞ ¼ Amax PFD=ðk þ PFDÞ R
MORPHOLOGICAL AND FUNCTIONAL TRAIT
MEASUREMENTS
We measured the length and width at the widest section of each
fully expanded, mature leaf and analysed these as the ratio frond
narrowness (length/width), with larger values describing relatively
longer, narrower fronds. We used a clinometer (Suunto, Vantaa,
Finland) to measure the angle of inclination away from vertical of
each leaf tip, with vertical = 0° and horizontal = 90°. Values >90°
describe leaves whose tips drooped below horizontal. The total
width and height of each plant were also measured and analysed
as the ratio plant narrowness, similar to frond narrowness, which
effectively integrates frond angles and lengths over a whole plant.
Individuals with higher plant narrowness values tend to be taller
than wide and have leaves held at steeper angles on average, while
plants with lower values tend to be broader than tall and have
leaves that lie at an angle approaching parallel with the ground.
We used negative epidermal impressions taken from one pinna
(leaf division) per plant with clear nail polish to calculate stomatal
densities. The numbers of stomata per mm2 were counted at five
locations on each epidermal impression and averaged. We measured SLA (cm2 g1) by removing and photographing against a
cm scale a single pinna from each plant. Pinnae were then dried in
silica gel and weighed using a Sartorius electronic balance
(precision = 000001 g, Goettingen, Germany). Areas of pinnae
were calculated from photographs using IMAGEJ software (Abramoff, Magelaes & Ram 2004). Data are archived in DRYAD (doi:
10.5061/dryad.38h06).
eqn 1
where Amax – R is the asymptotic maximum rate of net photosynthesis per unit leaf area, R is the rate of dark respiration, k is the
amount of light required to half-saturate photosynthesis, and all
parameters have units of lmol m2 per s. Amax, R and k were fit
to the light response data using a three-parameter Newton–Raphson approach (Givnish, Montgomery & Goldstein 2004). R values
were normalized to 25 °C using a standard Q10 value of 2. We
calculated the ICP (lmol PAR m2 per s) by setting A(PFD) = 0
and solving for PFD in eqn (1). Leaf compensation points (LCPs,
lmol PAR m2 per s) were calculated using the same equation,
following (Givnish, Montgomery & Goldstein 2004) to adjust the
value of R to account for night-time leaf respiration and leaf construction costs. We obtained mass-based values of Amax (Amax,
2
per s) and R (Rmass nmol CO2 g2 per s) by
mass, nmol CO2 g
multiplying the area-based photosynthetic values by SLA
(cm2 g1). We also recorded maximum stomatal conductance (gs,
mol H2O m2 per s) for each individual. Data are archived in
DRYAD (doi: 10.5061/dryad.38h06).
STATISTICAL ANALYSES
We performed all statistical analyses in R (R Development Core
Team 2008). We first conducted standard linear regressions for the
means of each morphological and physiological trait per species as
a function of per cent canopy openness, and for all traits against
each other; significance at P = 005, 001 and 0001 indicated by *,
**, and ***. Data for the two populations of D. intermedia were
not statistically significantly different and so were combined prior
to regression analyses, as were data for the dimorphic sterile and
fertile leaves of D. cristata.
When making interspecies comparisons, phylogenetic nonindependence can lead to an overestimation of degrees of freedom in
statistical analyses when species are treated incorrectly as fully
independent units (Felsenstein 1985). We therefore employed the
phylogenetically independent contrast (PIC) method (Felsenstein
1985) to determine whether correlations remained significant when
phylogeny was taken into account. The PIC method incorporates
branch lengths from a phylogenetic tree to account for relation-
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 113
ships, which we derived from the multilabelled phylogeny produced by Sessa, Zimmer & Givnish (2012b) based on the biparentally inherited nuclear marker gapCp (Fig. 2). In order to
standardize branch lengths, we added a small value (00001) to
edges with negligible (≤000000001) branch lengths, so that
unequal weight would not be given to contrasts involving shorter
branches (Garland, Harvey & Ives 1992). Tetraploid and hexaploid taxa possess two and three copies, respectively, of nuclear
loci due to whole-genome duplication and are thus represented by
multiple tips on the gapCp tree. If left uncorrected, this multilabelled nature would lead to each tetraploid taxon being interpreted as two separate species and each hexaploid as three
separate species during calculation of contrasts. To avoid this, we
used a novel repetitive process to perform contrast analyses, considering each possible combination of polyploid positions on the
tree. In the first step, tips were dropped from the tree at random
such that polyploids were represented by only one tip each
(Fig. 3). Contrasts were then calculated for per cent canopy openness and each of the dependent trait variables using this random
tree, and a standard linear regression carried out on the contrasts
and forced through the origin (Garland, Harvey & Ives 1992).
P-values were stored at each step and the process repeated 10 000
times to ensure that all possible binary tree topologies (48) were
sampled in the contrast analyses. We produced histograms of
P-values for contrasts of each trait against contrasts of per cent
canopy openness in order to assess the frequency of regression
analyses that produced significant vs. nonsignificant results. Values of the correlation coefficient (r) for each of the 10 000 repetitions were also recorded, in order to determine the percentage of
correlations for each set of contrasts that were positive or negative.
Finally, we performed a one-tailed t-test to determine
whether SLA differed significantly between polyploid and diploid
species.
Results
In the field, Dryopteris occupies a range of light regimes,
with canopy transmittance varying 28-fold across species,
from 129 14% for D. goldiana to 356 19% for
D. clintoniana (Table 3). Of the 13 traits measured, most
were positively correlated with canopy transmittance
across species in nonphylogenetically corrected analyses,
but the correlation was significant only for Amax
(r = 052*) with gs (r = 058*; Fig. 4, Table 4). Several
other traits were significantly correlated with each other,
including Amax with k (r = 081**), frond narrowness
(r = 069*) with gs (r = 094***); Rmass with Amax, mass
(r = 044*); R and Rmass with each other (r = 088***) and
each with ICP (R, r = 089***; Rmass r = 074**) with LCP
(R, r = 083**; Rmass r = 072*); SLA with Amax, mass
(r = 072**), Rmass (r = 066*), frond narrowness
(r = 072**), gs (r = 054*) with average stomatal density (r = 076**); ICP with LCP (r = 098***); plant narrowness with leaf tip angle (r = 064*); and gs and k each
with frond narrowness (gs, r = 075**; k, r = 066*), and
with each other (r = 086***; Table 4). There was no evidence of adaptive crossover in static light response curves
on either a mass or area basis; species from the brightest
environments consistently outperformed species from the
shadiest environments (Fig. 5). Average SLA did not differ
significantly between polyploid (382 39 cm2 per g) and
diploid (298 49 cm2 per g) taxa (P = 023).
In the phylogenetically structured analyses, we found
a significant correlation (P ≤ 005) between trait and
canopy transmittance contrasts in a majority of the
10 000 regression analyses on binary trees for Amax
(58% with P ≤ 005), SLA (61% with P ≤ 005) and k
(61% with P ≤ 005; Table 5, Appendix S1 in Supporting Information). Contrasts for each of the other 10
traits were not significantly correlated with per cent canopy transmittance contrasts for a majority of repetitions.
D. cristata B
D. clintoniana A
D. celsa B
D. ludoviciana
D. celsa A
D. goldiana
D. clintoniana B
D. expansa
D. campyloptera B
Fig. 2. Initial phylogenetic tree used in the
phylogenetically
independent
contrast
(PIC) analyses, based on a maximum likelihood topology from Sessa, Zimmer &
Givnish (2012a). A, B, C following taxon
names denote multiple genomes present in
polyploid species. Tips were dropped randomly in the regression analyses so that
each polyploid was represented by only one
tip at a time (see Fig. 3). Branches are proportional to expected number of nucleotide
substitutions per site. Bar indicates 001
changes.
D. carthusiana A
D. campyloptera A
D. intermedia
D. carthusiana B
D. cristata A
D. clintoniana C
D. marginalis
D. fragrans
0·01
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
114 E. B. Sessa & T. J. Givnish
D. ludoviciana
D. celsa A
D. goldiana
D. clintoniana B
D. expansa
D. campyloptera B
D. carthusiana A
D. intermedia
D. cristata A
D. marginalis
D. fragrans
0·01
Fig. 3. One random binary tree (out of 48
possible trees) used in the phylogenetically
independent contrast (PIC) calculations for
the field studies. Branches are proportional
to expected number of nucleotide substitutions per site. Bar indicates 001 changes.
Table 3. Environmental, morphological, photosynthetic and functional traits measured in the field. Means standard errors are given.
No stomatal density data were collected for D. fragrans due to density of abaxial trichomes
Species
% Canopy
transmittance
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
280
205
249
356
253
174
292
129
212
229
326
campyloptera
carthusiana
celsa
clintoniana
cristata
expansa
fragrans
goldiana
intermedia
ludoviciana
marginalis
13
20
34
19
12
14
25
14
16
27
14
Amax (lmol CO2
m2 per s)
400
527
245
669
596
411
713
353
494
486
499
Rmass (nmol CO2
g2 per s)
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
campyloptera
carthusiana
celsa
clintoniana
cristata
expansa
fragrans
goldiana
intermedia
ludoviciana
marginalis
141
146
132
113
172
515
73
114
182
89
40
26
41
51
17
23
116
18
27
22
13
04
04
05
03
05
03
04
04
04
02
03
02
030
033
045
046
039
103
047
032
067
044
016
SLA (cm2 g1)
452
433
341
246
439
497
155
356
316
211
256
R (lmol CO2
m2 per s)
18
33
23
14
15
27
35
17
20
6
12
k (lmol PAR
m2 per s)
004
009
024
005
005
023
006
007
011
007
002
294
444
278
413
499
298
643
284
464
324
283
Average
stomatal
density (mm2)
327
273
369
398
349
347
–
282
442
521
544
18
13
20
14
18
19
14
17
29
38
48
58
38
38
43
46
63
43
30
29
18
Plant
narrowness
(height:width)
062
081
098
127
120
091
064
077
076
129
065
006
003
006
011
008
006
005
003
003
018
004
ICP (lmol PAR
m2 per s)
235
266
763
312
348
1012
441
296
830
311
094
05
055
506
043
051
273
06
079
186
033
011
Frond
narrowness
(length:width)
201
271
325
401
430
223
477
272
303
399
333
003
004
007
008
009
004
010
005
003
010
003
LCP (lmol
PAR m2
per s)
43
49
2208
569
634
2575
799
57
1843
577
166
093
110
1726
082
097
847
11
172
489
07
02
Leaf tip
angle (0–90°)
876
650
623
450
553
560
852
700
638
749
720
10
17
38
41
22
22
36
29
12
35
19
Amax, mass
(nmol CO2
g2 per s)
190
223
84
160
259
202
115
123
155
100
126
27
21
6
14
15
16
31
15
11
7
5
gs (mol H2O
m2 per s)
006
007
003
01
009
004
014
004
006
007
007
001
001
00
001
001
001
001
00
00
00
001
ICP, instantaneous compensation point; LCP, leaf compensation point; SLA, specific leaf area.
For all traits, regressions on binary trees produced both
positive and negative correlations between trait contrasts
and native per cent canopy transmittance contrasts
(Table 5). For most traits, the direction of the correlation (either positive or negative) for the majority of repetitions was the same as in the phylogenetically
unstructured analyses. For three traits – Amax, LCP and
ICP – exactly half of the correlations were positive and
half negative, and for one trait – k – the majority of
repetitions in the phylogenetically structured analyses
produced negative correlations with light availability
contrasts, while the unstructured analysis produced a
positive correlation (Tables 4, 5).
Discussion
CORRELATED EVOLUTION OF PHOTOSYNTHETIC
PARAMETERS AND LEAF TRAITS WITH LIGHT REGIME
Our field measurements indicate that trends in photosynthetic, morphological and ecological traits in eastern
North American Dryopteris generally accord with
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 115
1·2
1·0
0·8
0·4
0·6
R (µmol m−2 s−1)
6
4
5
(b)
D. campyloptera
D. carthusiana
D. celsa
D. clintoniana
D. cristata
D. expansa
D. fragrans
D. goldiana
D. intermedia
D. ludoviciana
D. marginalis
2
0·2
3
Amax (µmol CO2 m−2 s−1)
7
(a)
10
15
20
25
30
35
40
10
15
% Canopy transmittance
25
30
35
40
35
40
35
40
(d)
10
8
6
ICP (µmol m−2 s−1)
2
4
50
40
30
k (µmol m−2 s−1)
60
12
70
(c)
20
% Canopy transmittance
10
15
20
25
30
35
40
10
15
% Canopy transmittance
30
250
200
150
Amax mass (nmol CO2 g−2 s−1)
100
40
30
20
0
LCP (µmol m−2 s−1)
25
(f)
10
(e)
20
% Canopy transmittance
10
15
20
25
30
35
40
% Canopy transmittance
10
15
20
25
30
% Canopy transmittance
Fig. 4. Field measurements of gas exchange, morphological and functional traits plotted against per cent canopy transmittance. Species
means SE are shown, and standard linear regression lines are plotted for those traits that were significantly correlated with per cent canopy transmittance. (a) Amax; (b) R; (c) k; (d) ICP; (e) LCP; (f) Amax, mass (g) Rmass; (h) specific leaf area (SLA); (i) average stomatal density;
(j) plant narrowness; (k) frond narrowness; (l) leaf tip angle; (m) gs. Data points are coloured by species according to the legend in (a);
polyploids are shown as squares, diploids circles (For interpretation of the references to colour in this figure legend, the reader is referred
to the online version of this article.). ICP, instantaneous compensation point; LCP, leaf compensation point.
theoretical predictions based on classical economic theory
(Bj€
orkman, Ludlow & Morrow 1972; Bj€
orkman 1981;
Givnish 1988; Givnish, Montgomery & Goldstein 2004).
Species from sunnier habitats tend to have higher photosynthetic and stomatal conductance rates per unit area,
require more light to saturate photosynthesis, have lower
SLAs, higher stomatal densities and increased frond narrowness (the latter indicating that leaves were generally
longer than broad; Fig. 4). We did not find significant
statistical support for most of these results, however, as
only Amax and gs are significantly correlated with light
availability in the phylogenetically unstructured analyses,
the remaining correlations being in the expected directions but not achieving statistical significance at the 5%
level (Table 4). In the phylogenetically structured analyses, Amax is again significantly correlated with per cent
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
116 E. B. Sessa & T. J. Givnish
(g)
400
300
SLA (cm2 g−1)
40
30
10
200
20
Rmass (nmol g−2 s−1)
50
60
500
(h)
10
15
20
25
30
35
40
10
15
% Canopy transmittance
30
35
40
15
20
25
30
35
35
40
35
40
1·4
1·2
1·0
0·8
0·6
Plant narrowness (height:width)
55
50
45
40
35
10
40
10
15
20
25
30
2·0
70
60
40
2·5
50
3·0
3·5
Leaf tip angle
4·0
80
4·5
(l)
90
% Canopy transmittance
5·0
% Canopy transmittance
Frond narrowness (length:width)
25
(j)
30
Average number of stomata mm−2
(i)
(k)
20
% Canopy transmittance
10
15
20
25
30
35
40
% Canopy transmittance
10
15
20
25
30
% Canopy transmittance
0·10
0·08
0·04
0·06
gs (mol m−2 s−1)
0·12
0·14
(m)
10
15
20
25
30
35
40
% Canopy transmittance
Fig. 4. Continued.
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 117
Table 4. Correlations among all traits measured in the field
Canopy
Amax
R
k
ICP
LCP
Amax, mass
Rmass
SLA
Average
stomatal
density
Plant
narrowness
Frond
narrowness
Leaf tip
angle
gs
Amax
R
k
ICP
LCP
Amax, mass
Rmass
SLA
Average
stomatal
density
052
–
–
–
–
–
–
–
–
–
037
007
–
–
–
–
–
–
–
–
022
081
006
–
–
–
–
–
–
–
040
032
089
000
–
–
–
–
–
–
038
044
083
013
098
–
–
–
–
–
009
027
016
024
002
008
–
–
–
–
046
024
088
018
074
072
044
–
–
–
047
046
028
034
027
030
072
066
–
–
048
022
009
013
012
011
049
032
076
–
010
015
017
003
004
004
009
007
008
017
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
008
–
–
–
–
–
–
–
–
–
–
–
–
008
–
–
–
–
–
–
–
–
–
–
–
–
–
Plant
narrowness
Frond
narrowness
Leaf tip
angle
gs
047
069
018
066
022
028
021
047
072
047
001
008
038
004
033
034
033
038
024
010
058
094
014
086
031
041
011
035
054
019
043
064
003
075
ICP, instantaneous compensation point; LCP, leaf compensation point; SLA, specific leaf area.
See text and Table 3 for data and units. Significant correlations (P ≤ 005) are shown in bold.
canopy transmittance in a majority of analyses, as are
SLA and k. Amax is thus the only trait to be strongly
correlated with light availability in both the phylogenetically structured and unstructured analyses. Our measures
of plant stature (plant narrowness and leaf tip angle)
show very little or no correlation with PFD (Fig. 4,
Table 4).
The values of Amax and other photosynthetic parameters
that we report here for Dryopteris are similar to those that
have been found in ferns previously. Amax in our species
ranged from 25 03 to 71 04 lmol CO2 m2 per s,
which corresponds roughly to the middle of the range of
Amax values that have been reported for other fern genera
(Ludlow & Wolf 1975; Nobel, Calkin & Gibson 1984;
Hollinger 1987; Brodribb, Feild & Jordan 2007). Previous
estimates for Dryopteris range from 18 lmol CO2 m2
per s in Asian D. crassirhizoma (Maeda, 1970) to 40 lmol
CO2 m2 per s in D. arguta (Pittermann et al. 2011) and
90 lmol CO2 m2 per s in D. filix-mas (Bauer, Gallmetzer
& Sato 1991). Previous researchers have reported instantaneous light compensation points in ferns infrequently, but
our values (Table 3) are somewhat higher than those documented previously for Dryopteris [which ranged from
05 lmol m2 per s in the shade to 17 lmol m2 per s in
the sun (Bannister & Wildish 1982)], and are more similar
to values reported by Hollinger (1987) for two understorey
species of Blechnum and Pteridium (73 and 116 lmol m2
per s, respectively).
In addition to the relationships with PFD, several traits
are also correlated with each other as expected based on
theoretical predictions. Amax is significantly positively correlated with k, as expected based on the Michaelis–Menten
eqn (1) (Farquhar & Sharkey 1982): when maximum photosynthetic rate increases, so too should the PFD required
to half-saturate photosynthesis. R and Rmass are also both
strongly positively correlated with ICP and LCP, and the
PFDs at which carbon consumed by respiration and produced by photosynthesis are balanced. As respiration
increases (i.e. more carbon is lost to respiration), the
amount of light required to balance carbon consumption
must increase. ICPs and LCPs range from 001% to 02%
full sunlight and are each well below the PFDs necessary
to partly saturate photosynthesis. Both are negatively correlated with PFD, which is contrary to expectations, as the
amount of light required to balance respiration should
increase with daily PFD to which a species is adapted
(Givnish, Montgomery & Goldstein 2004).
Several photosynthetic parameters show greater integration with each other and with other traits when considered
on a mass rather than an area basis (Table 4), although
only area-based Amax is significantly correlated with native
PFD. Because competition should favour plants that
maximize returns on energy investment (Givnish 1988,
1995), and mass-based estimates incorporate energy investment in leaf construction, photosynthesis on a mass basis
should be more closely tied to relative performance and
competition between species than photosynthesis on an
area basis. It is therefore surprising that Amax but not
Amax, mass is significantly correlated with native PFD; in
fact, Amax, mass shows almost no relationship to native
PFD (Fig. 4f). However, ferns are known to respond conservatively to desiccation (Brodribb & Holbrook 2004),
and this sensitivity is driven at least in part by low rates of
area-based leaf hydraulic conductance (Woodhouse &
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
118 E. B. Sessa & T. J. Givnish
Table 5. Results of independent contrast analyses on binary trees
for traits measured in the field against native per cent canopy
transmittance
(a)
Amax
R
k
ICP
LCP
Amax, mass
Rmass
SLA
Average stomatal density
Plant narrowness
Frond narrowness
Leaf tip angle
gs
(b)
P ≤ 005 (%)
% Positive
58
45
61
17
17
49
39
61
34
42
39
42
42
50
42
33
50
50
33
25
8
81
55
59
33
83
ICP, instantaneous compensation point; LCP, leaf compensation
point; SLA, specific leaf area.
Percentages indicate the frequency of binary trees (out of 48 possible topologies) for which P-values were below 005 and r was positive.
Fig. 5. Static light response curves of species from the two brightest habitats (D. marginalis and D. clintoniana) and the two shadiest habitats (D. goldiana and D. expansa). (a) area-based curves
(lmol CO2 m2 per s); (b) mass-based curves (nmol CO2 g1 per
s). Points are mean SE. Curves do not show adaptive crossover,
with species from shady habitats outperforming sun species at low
photon flux densities (PFDs) and the reverse at high PFDs (For
interpretation of the references to colour in this figure legend, the
reader is referred to the online version of this article.).
Nobel 1982; Brodribb & Holbrook 2004). Our measurements of maximum stomatal conductance (Table 3) are at
the low end but well within the range of values previously
reported for Dryopteris by Pittermann et al. (2011), and
for other ferns by Hollinger (1987) and Brodribb &
Holbrook (2004); these estimates range from 001 to
023 mol H2O m2 per s. gs was significantly positively
correlated with native PFD, Amax, k and frond narrowness, and negatively with SLA, indicating that longer, narrower and thinner leaves tended to have higher maximum
stomatal conductance rates. Low conductance rates will
directly influence transpiration and photosynthetic rates,
which suggests that area rather than mass-based estimates
of these parameters may be more meaningful for ferns, in
that they incorporate the surface area from which evaporation proceeds. Maximum stomatal conductance in Dryopteris is not significantly correlated with either mass-based
photosynthetic parameter but was significantly positively
correlated with area-based Amax (Table 4).
The lack of broad statistical support for the observed
trends in trait correlation with native PFD may reflect several phenomena. First, individual species do not appear to
demonstrate the differential adaptation to light regimes we
had expected. We saw no evidence of adaptive crossover
(Givnish, Montgomery & Goldstein 2004) in static light
responses, in which species from sunnier microsites outperform species from shady habitats at high PFDs, with the
reverse at low PFDs. In contrast, species from sunnier sites
outperformed species from shadier sites at all PFDs, in
both area-based and mass-based curves (Fig. 5). As noted
above, the absence of the expected mass-based relationship
is particularly striking, as these estimates incorporate leaf
construction costs via SLA and are therefore expected to
be more tightly tied to species’ relative competitive abilities. Species from the brightest environments, including
D. clintoniana, D. fragrans and D. marginalis, rarely
occupy the extremes of any given trait, usually being outperformed by species from intermediate habitats (Fig. 4,
Table 3). Such species rarely co-occur with other Dryopteris in the field, but apparently this is not due to any advantage such ‘sun’ species have over ‘shade’ species in carbon
uptake at high PFDs per se. A likely explanation for the
somewhat lower photosynthetic rates in these species is the
much greater evaporative demand in their high-PFD environments. In ferns – as opposed to angiosperms or gymnosperms – low hydraulic conductance may strongly restrict
photosynthesis, as well as growth and survival, interfering
with the usual rise in photosynthesis with PFD. A second
explanation for the lack of significant patterns across species related to PFD may be the fact that overall, there was
relatively little variation among species in PFD, with
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 119
canopy openness varying only 28-fold across species. A
narrow range of PFD decreases by itself the likelihood of
significant relationships of individual traits to PFD across
species, even if such relationships can be seen for species
inhabiting a broader gradient. We note that many of the
expected positive relationships among plant traits and
PFD were observed but were simply not significant; perhaps a greater number of species, or occupancy of a
broader range of light regimes, might have made these
patterns significant.
Finally, and perhaps most importantly, many of the
traits we examined are likely influenced not just by light
availability but by the correlation of PFD with temperature, vapour pressure deficit, wind speed and other environmental variables. Several previous studies on ferns have
documented one or more of the expected relationships
among photosynthetic parameters that we see in Dryopteris. In a study of three species from different genera (Adiantum, Thelypteris and Woodwardia) growing in different
light regimes, Hill (1972) found that species from sunnier
habitats had higher maximum photosynthetic rates, light
compensation points and light saturation points compared
with species from a shadier habitat. Lower compensation
points and SLAs in species from shady habitats have been
documented in single species from several fern genera
(Ludlow & Wolf 1975; Bannister & Wildish 1982; Brach,
McNaughton & Raynal 1993), as have higher stomatal
densities (Ludlow & Wolf 1975) in species from sunny
habitats. Higher photosynthetic capacity with increasing
light availability has also been documented in Blechnum
(Salda~
na et al. 2009), Adiantum (Yeh & Wang 2000) and
in D. intermedia (Brach, McNaughton & Raynal 1993),
among others. However, in additional studies examining
fern traits and habitat preference, soil moisture was found
to be a primary factor influencing fern distributions in
Ohio (Greer, Lloyd & McCarthy 1997) and Quebec (Karst,
Gilbert & Lechowicz 2005), and Nobel, Calkin & Gibson
(1984), in a study of gas exchange in 14 species of ferns in
a greenhouse, found that water vapour and CO2 conductance were better predictors of maximum photosynthetic
rates than was PFD. Reudink et al. (2005) noted that
D. intermedia growing in a damp, but open site appeared
to experience greater water stress, as indicated by negative
water potential, than other species growing in the same
environment, and concluded that D. intermedia’s ability to
acclimate to high PFDs was limited, with moisture playing
a greater role in determining its distribution. R€
unk, Zobel
& Zobel (2010), in a study of three Estonian species of
Dryopteris (including D. carthusiana and D. expansa),
determined that species’ distributions and abundance were
most influenced either by edaphic factors or by ‘locale
effects’ that were not specifically related to any one variable. Taken together, these results suggest that fern distributions and adaptation of specific traits are likely shaped
by the interaction of a number of environmental factors.
That light availability is probably joined in its effects by
moisture availability and evaporative demand may account
for our general observation of expected but nonsignificant
correlations between traits and PFD, and for the lack of
expected pairwise performance comparisons at the species
level between taxa from sun and shade environments. It is
important to recognize that several traits expected to show
positive relationships to PFD based on classic economic
theory (i.e. based solely on the putative effects of PFD)
might also show such relationships based on hydraulic
considerations. As we suggested in the Introduction, a
greater sensitivity of plastid movement to PFD in ferns
might preserve a positive relationship of Amax to PFD
while reducing or eliminating that of k and R, by reducing
self-shading and maintaining a constant plastid mass.
Thicker leaves or leaves with higher mass per unit area
may be an adaptation to greater evaporative demand in
brighter sites, reducing leaf and evaporative surfaces in
order to reduce tissue desiccation and preserve xylem function (Creese, Lee & Sack 2011). Our gs values were significantly correlated with PFD (Table 4), but this correlation
could also have been driven by differences in water availability between the sites, if brighter habitats happened also
to be drier. Conductance in several fern species has been
reported to be more strongly influenced by soil water status than by other variables (Nobel 1978; Hollinger 1987),
and several recent studies have revealed that fern stomata
lack many of the active control mechanisms found in angiosperms (Doi, Wada & Shimazaki 2006; Doi & Shimazaki 2008; Brodribb et al. 2009; Brodribb & McAdam
2011), instead opening and closing passively in response to
hydraulic changes and leaf water status (Brodribb &
McAdam 2011; McAdam & Brodribb 2012). This further
suggests that hydraulic constraints, rather than responses
to PFD per se, may be primarily responsible for shaping
morphological and physiological diversity in these ferns.
Indeed, decreased conductance in brighter environments
could countervail the general tendency towards higher
Amax, k and R and lower SLAs in higher-PFD environments.
TRANSGRESSIVE TRAIT EXPRESSION IN POLYPLOIDS
The eastern North American Dryopteris complex includes
five allopolyploids – tetraploids D. carthusiana, D. cristata,
D. campyloptera, D. celsa and hexaploid D. clintoniana –
the latter three of which have both parental species extant
(Sessa, Zimmer & Givnish 2012b). We expected to see evidence of transgressive expression in these species relative
to their parents in at least some of the traits measured.
Novel or extreme trait expression can allow hybrids to
escape competition and the effects of minority cytotype
exclusion and will tend to favour their establishment and
long-term success (Abbott 1992; Rieseberg 1997; Rieseberg, Archer & Wayne 1999; Buerkle et al. 2000; Rieseberg
et al. 2003; Quintanilla & Escudero 2006). Two recent
studies have also shown that polyploids tend to occupy
larger geographic ranges than their progenitor taxa
(Parisod, Holderegger & Brochmann 2010; Parisod 2012).
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
120 E. B. Sessa & T. J. Givnish
This is true of D. campyloptera, D. celsa and D. clintoniana (Sessa, Zimmer & Givnish 2012b) and suggests that in
some traits they must be transgressive relative to their
parents. Sessa, Zimmer & Givnish (2012b) also found
that the allotetraploid hybrids have resulted from crosses
between diploid parents that display intermediate levels
of genetic divergence relative to all possible progenitor
pairs. Such intermediate levels of parental divergence
have been found to be linked to transgressive trait
expression in hybrid angiosperms (Stelkens & Seehausen
2009).
Contrary to our expectations, we found little evidence
for transgressive trait expression in any of the three allopolyploids we examined. In almost all traits, measurements
in the polyploids were intermediate relative to their parents
and frequently overlapped significantly with one or both
parents (Fig. 4). SLA was also not significantly different
between polyploids and diploids. Transgressive expression
of photosynthetic parameters is therefore evidently not the
means by which these allopolyploid taxa are being maintained. However, the native PFDs of two of the three
allopolyploids (D. campyloptera and D. clintoniana) were
transgressive relative to their parental species (Fig. 4), with
each allopolyploid occupying a much brighter habitat
than its parents. Coupled with the transgressive ranges of
these species relative to their parents, this points to the
existence of physiological differences that provided the
allopolyploid taxa an advantage in geographic range that
apparently allowed them to coexist with their progenitors
at a regional scale. As discussed above, water balance
mediated by hydraulics is the most likely driving factor.
Although we found no significant difference in SLA
between polyploids and diploids, this does not rule out the
possibility of an increase in tracheid size due to polyploidy
that could affect species’ hydraulic conductance and thus
their ability to tolerate brighter habitats. Further studies
of the allopolyploids and their parents, including sampling
from multiple populations of each in the field, will be
needed in order to determine the relative roles of light and
water availability in influencing these species’ evolution.
Conclusions
The current study is the first to document physiological
and morphological traits in a widespread, closely related
group of ferns whose phylogeny is well understood, and to
attempt to determine whether photosynthetic PFD has
been a primary factor driving adaptation in fern leaf form
and physiology. Although we found little evidence to support PFD as the main determinant of diversification in the
traits measured, our results suggest several avenues for
future studies. Common garden studies, in which species
are grown together and environmental variables can be
carefully controlled, will be particularly important, and
will allow us to more rigorously examine the extent to
which species may be adapted to their native light regimes.
Such studies will allow both ‘soft’ tests of adaptation, in
which traits can be compared among species grown within
the same PFD and related to the native PFDs of each
species in the field (Givnish, Montgomery & Goldstein
2004), and ‘hard’ tests, which will allow us to ask whether
species showing the highest realized rates of photosynthesis
shift with light treatment, and if they are the species
that occupy habitats with similar native light regimes. Our
results also indicate that hydraulic demands may have
played a key role in the physiological and morphological
evolution of these ferns, supporting results from several
recent studies that have found significant differences
between ferns and other land plants in traits related to
water balance, including hydraulic conductance, vein
density and the nature of fern stomatal responses
(Brodribb & Holbrook 2004; Brodribb, Feild & Jordan
2007; Boyce et al. 2009; Watkins, Holbrook & Zwieniecki
2010; Brodribb & McAdam 2011; McAdam & Brodribb
2012).
This study is the first to propose a statistical framework
for incorporating phylogeny into comparative analyses
when the subjects include species with reticulate evolutionary histories. Phylogenetic relationships must be considered in comparative analyses due to the danger of inflating
type I error rates when species are erroneously considered
to be independent samples (Felsenstein 1985). The method
we present allows for the incorporation of phylogeny even
in the case of hybridization or polyploidy, by accounting
for all possible sets of relationships (trees) for a given set
of taxa. The results of the phylogenetically structured and
unstructured analyses presented here are largely congruent
with each other, and this agreement increases our confidence in the results. However, the distribution of negative
and positive r values (Table 5) obtained from repetitive
PIC analyses utilizing all possible trees indicates that the
placement of hybrid species in the tree can greatly
influence the results of such analyses, and speaks to the
importance of incorporating phylogeny in comparative
studies and considering carefully the placement of reticulate taxa. If we had selected only a single tree, for example,
our results for any given trait would have been either
100% positive or negative, when in fact there is a distribution between negative and positive correlations depending
on which tree is utilized in a given repetition of the analysis. Many researchers now incorporate phylogeny into
their comparative studies of ecological and physiological
traits, and we hope that the method presented here will be
useful in such studies involving hybrid and/or polyploid
organisms.
Acknowledgements
The authors thank A. Jandl, T. Goforth, P. Gonsiska, C. Line, J. Perry
and K. Woods for help in the field, and the Finger Lakes Land Trust, Nature Conservancy of North Carolina, Huron Mountain Club, Blue Ridge
Parkway, Fernwood Botanical Gardens, and J.E. Watkins, Jr. and J. Perry
for allowing us access to field populations. Many thanks to E. Kruger for
advice on this manuscript, and to C. Ane for assistance with developing the
statistical tools used in this study, particularly for the phylogenetically
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 121
structured analyses of polyploids. Financial support for this research was
provided by the Huron Mountain Wildlife Foundation (EBS, TJG), a microMORPH award from the National Science Foundation (EBS), and
graduate research awards from the Botanical Society of America, the
American Society of Plant Taxonomists and the Torrey Botanical Society
(EBS). The authors declare no conflict of interest.
Data Accessibility
Physiological data are archived in DRYAD (doi: 10.5061/
dryad.38h06).
References
Abbott, R.J. (1992) Plant invasions, interspecific hybridization and the evolution of new plant taxa. Trends in Ecology and Evolution, 7, 401–404.
Abbott, R.J. & Lowe, A.J. (2004) Origins, establishment and evolution of
new polyploid species: Senecio cambrensis and S. eboracensis in the British Isles. Biological Journal of the Linnean Society, 82, 467–474.
Abramoff, M.D., Magelaes, P.J. & Ram, S.J. (2004) Image processing with
ImageJ. Biophotonics International, 11, 36–42.
Ackerly, D.D., Knight, C.A., Weiss, S.B., Barton, K. & Starmer, K.P.
(2002) Leaf size, specific leaf area and microhabitat distribution of chaparral woody plants: contrasting patterns in species level and community
level analyses. Oecologia, 130, 449–457.
Arens, N.C. (1997) Responses of leaf anatomy to light environment in the
tree fern Cyathea caracasana (Cyatheaceae) and its application to some
ancient seed ferns. Palaios, 12, 84–94.
Bannister, P. & Wildish, K.L. (1982) Light compensation points and specific leaf areas in some New Zealand ferns. New Zealand Journal of Botany, 20, 421–424.
Bauer, H., Gallmetzer, C. & Sato, T. (1991) Phenology and photosynthetic
activity in sterile and fertile sporophytes of Dryopteris filix-mas. Oecologia, 86, 159–162.
Bj€
orkman, O. (1981) Responses to different quantum flux densities. Physiological Plant Ecology. I. Encyclopedia of Plant Physiology (eds O.L.
Lange, P.S. Nobel, C.B. Osmond & H. Ziegler), pp. 57–107. SpringerVerlag, New York, New York, USA.
Bj€
orkman, O., Ludlow, M.M. & Morrow, P.A. (1972) Photosynthetic performance of two rainforest species in their native habitat and analysis of their
gas exchange. Carnegie Institute of Washington Year Book, 71, 94–102.
Boyce, C.K., Brodribb, T.J., Feild, T.S. & Zwieniecki, M.A. (2009) Angiosperm leaf vein evolution was physiologically and environmentally transformative. Proceedings of the Royal Society. B, Biological Sciences, 276,
1771–1776.
Brach, A.R., McNaughton, S.J. & Raynal, D.J. (1993) Photosynthetic
adaptability of two fern species of a northern hardwood forest. American
Fern Journal, 83, 47–53.
Brodribb, T.J., Feild, T.S. & Jordan, G.J. (2007) Leaf maximum photosynthetic rate and venation are linked by hydraulics. Plant Physiology, 144,
1890–1898.
Brodribb, T.J. & Holbrook, N.M. (2004) Stomatal protection against
hydraulic failure: a comparison of coexisting ferns and angiosperms.
New Phytologist, 162, 663–670.
Brodribb, T.J. & Holbrook, N.M. (2006) Declining hydraulic efficiency as
transpiring leaves desiccate: two types of response. Plant, Cell and Environment, 29, 2205–2215.
Brodribb, T.J. & McAdam, S.A.M. (2011) Passive origins of stomatal control in vascular plants. Science, 331, 582–585.
Brodribb, T.J., McAdam, S.A.M., Jordan, G.J. & Feild, T.S. (2009) Evolution of stomatal responsiveness to CO2 and optimization of water-use
efficiency among land plants. New Phytologist, 183, 839–847.
Buerkle, C.A., Morris, R.J., Asmussen, M.A. & Rieseberg, L.H. (2000) The
likelihood of homoploid hybrid speciation. Heredity, 84, 441–451.
Chen, Z.Y., Peng, Z.S., Yang, J., Chen, W.Y. & Ou-Yang, Z.M. (2011) A
mathematical model for describing light-response curves in Nicotiana
tabacum L. Photosynthetica, 49, 467–471.
Comai, L. (2005) The advantages and disadvantages of being polyploid.
Nature Reviews Genetics, 6, 836–846.
Cowan, I.R. & Farquhar, G.D. (1977) Stomatal function in relation to leaf
metabolism and environment. Symposium of the Society for Experimental
Biology, 31, 471–505.
Creese, C., Lee, A. & Sack, L. (2011) Drivers of morphological diversity and
distribution in the Hawaiian fern flora: trait associations with size, growth
form, and environment. American Journal of Botany, 98, 956–966.
Doi, M. & Shimazaki, K.-I. (2008) The stomata of the fern Adiantum capillus-veneris do not respond to CO2 in the dark and open by photosynthesis in guard cells. Plant Physiology, 147, 922–930.
Doi, M., Wada, M. & Shimazaki, K.-I. (2006) The fern Adiantum capillusveneris lacks stomatal responses to blue light. Plant and Cell Physiology,
47, 748–755.
Endler, J.A. (1993) The color of light in forests and its implications. Ecological Monographs, 63, 1–27.
Evans, J.R. & Poorter, H. (2001) Photosynthetic acclimation of plants to
growth irradiance: the relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain. Plant, Cell and Environment,
24, 755–767.
Farquhar, G.D. & Sharkey, T.D. (1982) Stomatal conductance and photosynthesis. Annual Review of Plant Physiology, 33, 317–345.
Felsenstein, J. (1985) Phylogenies and the comparative method. The American Naturalist, 125, 1–15.
Franks, P.J. & Beerling, D.J. (2009a) CO2-forced evolution of plant gas
exchange capacity and water-use efficiency over the Phanerozoic. Geobiology, 7, 227–236.
Franks, P.J. & Beerling, D.J. (2009b) Maximum leaf conductance driven by
CO2 effects on stomatal size and density over geologic time. Proceedings
of the National Academy of Sciences, USA, 106, 10343–10347.
Franks, P.J., Drake, P.L. & Beerling, D.J. (2009) Plasticity in maximum
stomatal conductance constrained by negative correlation between stomatal size and density: an analysis using Eucalyptus globulus. Plant, Cell
and Environment, 32, 1737–1748.
Franks, P.J. & Farquhar, G.D. (2006) The mechanical diversity of stomata
and its significance in gas-exchange control. Plant Physiology, 143,
78–87.
Frazer, G.W., Canham, C.D., Sallaway, P. & Marinakis, D. (1999) Gap
Light Analyzer. Simon Fraser University, Burnaby, British Columbia,
Canada, and Institute for Ecosystem Studies, Millbrook, New York,
USA.
Garland, T., Harvey, P.H. & Ives, A.R. (1992) Procedures for the analysis
of comparative data using phylogenetically independent contrasts. Systematic Biology, 41, 18–32.
Givnish, T.J. (1978) Ecological aspects of plant morphology: leaf form in
relation to environment. Acta Biotheoretica, 11, 83–142.
Givnish, T.J. (1979) On the adaptive significance of leaf form. Topics in
Plant Population Biology (ed. O. Solbrig), pp. 375–407. Columbia University Press, New York.
Givnish, T.J. (1982) On the adaptive significance of leaf height in forest
herbs. The American Naturalist, 120, 353–381.
Givnish, T.J. (1986) Optimal stomatal conductance, allocation of energy
between leaves and roots, and the marginal cost of transpiration. On the
Economy of Plant Form and Function (ed. T.J. Givnish), pp. 171–213.
Cambridge University Press, Cambridge, UK.
Givnish, T.J. (1988) Adaptation to sun and shade: a whole-plant perspective. Australian Journal of Plant Physiology, 15, 63–92.
Givnish, T.J. (1995) Plant stems: biomechanical adaptation for energy capture and influence on species distributions. Plant stems: Physiology and
Functional Morphology (ed. B.L. Gartner), pp. 3–49. Chapman and Hall,
New York, New York, USA.
Givnish, T.J., Montgomery, R.A. & Goldstein, G. (2004) Adaptive radiation of photosynthetic physiology in the Hawaiian lobeliads: light
regimes, static light responses, and whole-plant compensation points.
American Journal of Botany, 91, 228–246.
Givnish, T.J. & Vermeij, G.T. (1976) Sizes and shapes of liane leaves. The
American Naturalist, 110, 743–778.
Greer, G.K., Lloyd, R.M. & McCarthy, B.C. (1997) Factors influencing the
distribution of pteridophytes in a southeastern Ohio hardwood forest.
Journal of the Torrey Botanical Society, 124, 11–21.
Hietz, P. (2010) Fern adaptations to xeric environments. Fern Ecology (eds
K. Mehltreter, L.R. Walker & J.M. Sharpe), pp. 140–176. Cambridge
University Press, Cambridge.
Hill, R.H. (1972) Comparative patterns of CO2 exchange of three fern species of southeastern Michigan. The Michigan Academician, 4, 303–310.
Hollinger, D.Y. (1987) Photosynthesis and stomatal conductance patterns
of two fern species from different forest understoreys. Journal of Ecology, 75, 925–935.
Jimenez, A., Quintanilla, L.G., Pajaron, S. & Pangua, E. (2009) Genetic
variation in the allotetraploid Dryopteris corleyi (Dryopteridaceae) and
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
122 E. B. Sessa & T. J. Givnish
its diploid parental species in the Iberian Peninsula. American Journal of
Botany, 96, 1880–1886.
Kanegae, T., Hayashida, E., Kuramoto, C. & Wada, M. (2006) A single
chromoprotein with triple chromophores acts as both a phytochrome
and a phototropin. Proceedings of the National Academy of Sciences,
103, 17997–18001.
Karst, J., Gilbert, B. & Lechowicz, M.J. (2005) Fern community assembly:
the roles of chance and the environment at local and intermediate scales.
Ecology, 86, 2473–2486.
Kawai, H., Kanegae, T., Christensen, S., Kiyosue, T., Sato, Y., Imaizumi,
T. et al. (2003) Responses of ferns to red light are mediated by an
unconventional photoreceptor. Nature, 421, 287–290.
Kessler, M., Siorak, Y., Wunderlich, M. & Wegner, C. (2007) Patterns of
morphological leaf traits among pteridophytes along humidity and temperature gradients in the Bolivian Andes. Functional Plant Biology, 34, 963.
Kruger, E.L. & Volin, J.C. (2006) Reexamining the empirical relation
between plant growth and leaf photosynthesis. Functional Plant Biology,
33, 421–429.
Levin, D.A. (1975) Minority cytotype exclusion in local plant populations.
Taxon, 24, 35–43.
Ludlow, C.J. & Wolf, F.T. (1975) Photosynthesis and respiration rates of
ferns. American Fern Journal, 65, 43–48.
Lupia, R., Lidgard, S. & Crane, P.R. (1999) Comparing palynological
abundance and diversity: implications for biotic replacement during the
Cretaceous Angiosperm radiation. Paleobiology, 25, 305–340.
Maherali, H., Walden, A.E. & Husband, B.C. (2009) Genome duplication
and the evolution of physiological responses to water stress. New Phytologist, 184, 721–731.
Manton, I. (1950) Problems of Cytology and Evolution in the Pteridophyta.
Cambridge University Press, Cambridge.
McAdam, S.A.M. & Brodribb, T.J. (2012) Stomatal innovation and the rise
of seed plants. Ecology Letters, 15, 1–8.
Melaragno, J.E., Mehrota, B. & Coleman, A.W. (1993) Relationship
between endopolyploidy and cell size in epidermal tissue of Arabidopsis.
The Plant Cell, 5, 1661–1668.
Montgomery, J. & Wagner, W.H. (1993) Dryopteris in Flora of North
America North of Mexico. Oxford University Press, New York, New
York, USA.
Nagalingum, N.S., Drinnan, A.N., Lupia, R. & McLoughlin, S. (2002)
Fern spore diversity and abundance in Australia during the Cretaceous.
Review of Palaeobotany and Palynology, 119, 69–92.
Niklas, K.J., Tiffney, B.H. & Knoll, A.H. (1983) Patterns in vascular land
plant diversification. Nature, 303, 614–616.
Nobel, P.S. (1978) Microhabitat, water relations, and photosynthesis of a
desert fern, Notholaena parryi. Oecologia, 31, 293–309.
Nobel, P.S., Calkin, H.W. & Gibson, A.C. (1984) Influences of PAR, temperature and water vapor concentration on gas exchange by ferns. Physiologia Plantarum, 62, 527–534.
Page, C. (2002) Ecological strategies in fern evolution: a neopteridological
overview. Review of Palaeobotany and Palynology, 119, 1–33.
Parisod, C. (2012) Polyploids integrate genomic changes and ecological
shifts. New Phytologist, 193, 297–300.
Parisod, C., Holderegger, R. & Brochmann, C. (2010) Evolutionary consequences of autopolyploidy. New Phytologist, 186, 5–17.
Pittermann, J., Limm, E., Rico, C. & Christman, M.A. (2011) Structurefunction constraints of tracheid-based xylem: a comparison of conifers
and ferns. New Phytologist, 192, 449–461.
Quintanilla, L.G. & Escudero, A. (2006) Spore fitness components do not
differ between diploid and allotetraploid species of Dryopteris (Dryopteridaceae). Annals of Botany, 98, 609–618.
R Development Core Team (2008) R: A Language and Environment for
Statistical Computing. R Development Core Team, R Foundation for
Statistical Computing, Vienna, Austria, http://www.R-project.org.
Reudink, M.W., Snyder, J.P., Xu, B., Cunkelman, A. & Balsamo, R.A.
(2005) A Comparison of physiological and morphological properties of
deciduous and wintergreen ferns in southeastern Pennsylvania. American
Fern Journal, 95, 45–56.
Rieseberg, L.H. (1997) Hybrid origins of plant species. Annual Review of
Ecology and Systematics, 28, 359–389.
Rieseberg, L.H., Archer, M.A. & Wayne, R.K. (1999) Transgressive segregation, adaptation and speciation. Heredity, 83, 363–372.
Rieseberg, L., Raymond, O., Rosenthal, D., Lai, Z., Livingstone, K., Nakazato, T. et al. (2003) Major ecological transitions in wild sunflowers
facilitated by hybridization. Science, 301, 1211–1216.
Rodriguez, D.J. (1996) A model for the establishment of polyploidy in
plants. The American Naturalist, 147, 33–46.
Rothwell, G.W. (1996) Pteridophytic evolution: an often underappreciated
phytological success story. Review of Palaeobotany and Palynology, 90,
209–222.
R€
unk, K. & Zobel, K. (2007) Phenotypic plasticity and biomass allocation
pattern in three Dryopteris (Dryopteridaceae) species on an experimental
light-availability gradient. Plant Ecology, 193, 85–99.
R€
unk, K., Zobel, M. & Zobel, K. (2010) Different factors govern the performance of three closely related and ecologically similar Dryopteris species with contrastingly different abundance in a transplant experiment.
Botany, 88, 961–969.
Salda~
na, A.O., Hernandez, C., Coopman, R.E., Bravo, L.A. & Corcuera,
L.J. (2009) Differences in light usage among three fern species of genus
Blechnum of contrasting ecological breadth in a forest light gradient.
Ecological Research, 25, 273–281.
Schneider, H., Schuettpelz, E., Pryer, K., Cranfill, R., Magall
on, S.A. &
Lupia, R. (2004) Ferns diversified in the shadow of angiosperms. Nature,
428, 554–557.
Schuettpelz, E. & Pryer, K.M. (2009) Evidence for a Cenozoic radiation of
ferns in an angiosperm-dominated canopy. Proceedings of the National
Academy of Sciences, 106, 11200–11205.
Schwarzbach, A., Donovan, L.A. & Rieseberg, L. (2001) Transgressive
character expression in a hybrid sunflower species. American Journal of
Botany, 88, 270–277.
Sessa, E.B., Zimmer, E.A. & Givnish, T.J. (2012a) Phylogeny, divergence
times, and historical biogeography of New World Dryopteris (Dryopteridaceae). American Journal of Botany, 99, 730–750.
Sessa, E.B., Zimmer, E.A. & Givnish, T.J. (2012b) Unraveling reticulate
evolution in North American Dryopteris (Dryopteridaceae). BMC Evolutionary Biology, 12, 104.
Shipley, B. (1995) Structured interspecific determinants of specific leaf
area in 34 species of herbaceous angiosperms. Functional Ecology, 9,
312–319.
Shipley, B., Vile, D., Garnier, E., Wright, I.J. & Poorter, H. (2005)
Functional linkages between leaf traits and net photosynthetic rate: reconciling empirical and mechanistic models. Functional Ecology, 19,
602–615.
Shipley, B., Lechowicz, M.J., Wright, I. & Reich, P.B. (2006) Fundamental
trade-offs generating the worldwide leaf economics spectrum. Ecology,
87, 535–541.
Smith, A., Pryer, K., Schuettpelz, E., Korall, P., Schneider, H. & Wolf, P.
(2006) A classification for extant ferns. Taxon, 55, 705–731.
Stelkens, R. & Seehausen, O. (2009) Genetic distance between species predicts novel trait expression in their hybrids. Evolution, 63, 884–897.
Suetsugu, N., Mittmann, F., Wagner, G., Hughes, J. & Wada, M. (2005) A
chimeric photoreceptor gene, NEOCHROME, has arisen twice during
plant evolution. Proceedings of the National Academy of Sciences, USA,
102, 13705–13709.
Tani, T. & Kudo, G. (2003) Storage ability of overwintering leaves and rhizomes in a semi-evergreen fern, Dryopteris crassirhizoma. Ecological
Research, 18, 15–24.
Tani, T. & Kudo, G. (2005) Overwintering leaves of a forest-floor fern,
Dryopteris crassirhizoma (Dryopteridaceae): a small contribution to the
resource storage and photosynthetic carbon gain. Annals of Botany, 95,
263–270.
Taylor, S.H., Franks, P.J., Hulme, S.P., Spriggs, E., Christin, P.A.,
Edwards, E.J. et al. (2012) Photosynthetic pathway and ecological adaptation explain stomatal trait diversity amongst grasses. New Phytologist,
193, 387–396.
Tessier, J.T. & Bornn, M.P. (2007) Old fronds serve as a vernal carbon
source in the wintergreen fern Dryopteris intermedia. American Journal
of Botany, 94, 25–28.
Walters, M.B. & Reich, P.B. (1996) Are shade tolerance, survival, and
growth linked? Low light and nitrogen effects on hardwood seedlings.
Ecology, 77, 841–853.
Watkins, J.E., Holbrook, N.M. & Zwieniecki, M.A. (2010) Hydraulic properties of fern sporophytes: consequences for ecological and evolutionary
diversification. American Journal of Botany, 97, 2007–2019.
Watkins, J.E., Mack, M.K. & Mulkey, S.S. (2007) Gametophyte ecology
and demography of epiphytic and terrestrial tropical ferns. American
Journal of Botany, 94, 701–708.
Wong, S.C., Cowan, I.R. & Farquhar, G.D. (1979) Stomatal conductance
correlates with photosynthetic capacity. Nature, 282, 424–426.
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123
Leaf form and photosynthesis in Dryopteris 123
Wood, T.E., Takebayashi, N., Barker, M.S., Mayrose, I., Greenspoon, P.B.
& Rieseberg, L.H. (2009) The frequency of polyploid speciation in vascular plants. Proceedings of the National Academy of Sciences, 106,
13875–13879.
Woodhouse, R.M. & Nobel, P.S. (1982) Stipe anatomy, water potentials,
and xylem conductances in seven species of ferns. American Journal of
Botany, 69, 135–140.
Wright, I.J., Westoby, P.B.R., Ackerly, D.D., Baruch, Z., Bongers, F.,
Cavendar-Bares, J. et al. (2004) The worldwide leaf economics spectrum.
Nature, 428, 821–827.
Yeh, D.M. & Wang, H.M. (2000) Effects of irradiance on growth, net
photosynthesis and indoor performance of the shade-adapted plant,
maidenhair fern. Journal of Horticultural Science and Biotechnology, 75,
293–298.
Received 9 September 2012; accepted 5 June 2013
Handling Editor: Mark Tjoelker
Supporting Information
Additional Supporting Information may be found in the online
version of this article:
Appendix S1. Results of phylogenetically structured analyses of
field data.
© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 28, 108–123