Available via license: CC BY 4.0
Content may be subject to copyright.
Submitted 6 November 2019
Accepted 25 March 2021
Published 16 April 2021
Corresponding author
Piotr Kiełtyk, kieltyk@wp.pl
Academic editor
Gabriele Casazza
Additional Information and
Declarations can be found on
page 11
DOI 10.7717/peerj.11286
Copyright
2021 Kiełtyk
Distributed under
Creative Commons CC-BY 4.0
OPEN ACCESS
Intraspecific morphological variation of
Bellidiastrum michelii (Asteraceae) along
a 1,155 m elevation gradient in the Tatra
Mountains
Piotr Kiełtyk
Institute of Biological Sciences, Cardinal Stefan Wyszynski University in Warsaw, Warszawa, Poland
ABSTRACT
Plant species that inhabit large elevation gradients in mountain regions are exposed
to different environmental conditions. These different conditions may influence
plant morphology via plastic responses and/or via genetic adaptation to the local
environment. In this study, morphological variation was examined for Bellidiastrum
michelii Cass. (Asteraceae) plants growing along a 1,155 m elevation gradient in the
Tatra Mountains in Central Europe. The aim was to contribute to gaining a better
understanding of within-species morphological variation in a mountain species across
elevation gradients. Twelve morphological traits, which were measured for 340 plants
collected from 34 sites, were plotted against elevation using Generalised Additive
Models. Significant variation in B. michelii morphology was found across the elevation
gradient. Plant size, in the form of plant height, total aboveground mass and total
leaf mass, decreased significantly with increasing elevation. Similarly, floral traits, such
as flower head mass, total flower mass, individual flower mass, flower head diameter
and ligulate and tubular flower length, also decreased significantly with increasing
elevation. However, the changes in these floral traits were not as large as those observed
for plant size traits. Interestingly, the number of flowers produced by the plant, both
ligulate and tubular, did not change across the studied elevation gradient. In this study,
elevation was found to be an important gradient across which significant intraspecific
morphological variation occurred in a mountain plant. These morphological changes
may have occurred in response to various abiotic and biotic factors that change along
elevation gradients.
Subjects Ecology, Plant Science
Keywords Altitudinal gradient, Biomass, Floral characteristic, Within-species variation
INTRODUCTION
Plants that grow across mountain slopes experience different stresses caused by the variable
environmental conditions found along elevation gradients. The most notable changes
associated with increasing elevation in mountain regions include decreases in the mean air
temperature, shortening of the growing season, increases in light intensity, UV radiation
and wind velocity, reductions in carbon dioxide and oxygen concentrations, and decreases
in nutrient availability (Billings, 1974;Körner, 2003;Nagy & Grabherr, 2009). Soil microbial
activity and resource competition intensity also decrease at higher elevations (Körner, 2003).
How to cite this article Kiełtyk P. 2021. Intraspecific morphological variation of Bellidiastrum michelii (Asteraceae) along a 1,155 m ele-
vation gradient in the Tatra Mountains. PeerJ 9:e11286 http://doi.org/10.7717/peerj.11286
Meanwhile, competition for pollinator services increases with elevation as the number of
pollinators declines at higher elevations (Maad, Armbruster & Fenster, 2013;Zhao & Wang,
2015;Arroyo, Pacheco & Dudley, 2017). Elevation therefore encompasses many abiotic and
biotic components. Thus, elevation likely acts as a key summary variable that relates to the
phenotypic variation observed in plants growing on mountains (Kelly, 1998;Hautier et al.,
2009;Scheepens & Stöcklin, 2013;He et al., 2017;Miljković et al., 2019;Paudel et al., 2019).
Plants growing along large elevational ranges can adapt to the local conditions at the
different elevations by adjusting their morphology and allocation of biomass to different
vegetative and reproductive structures (Guo et al., 2012;Takahashi & Matsuki, 2016). For
example, plant height is a trait that is related to fitness through vegetative competitiveness
and ability to use available resources. Plant height has commonly been found to decrease
within species with increasing elevation (Nishizawa et al., 2001;Alexander et al., 2009;Zhu
et al., 2010;Maad, Armbruster & Fenster, 2013;He et al., 2017;Paudel et al., 2019), possibly
as a response to limiting climatic conditions (Körner, 2003). Moreover, flowers, which are
organs that are relatively invariant within species, have been found to vary quantitatively
in plants of the same species across mountain elevation gradients (Herrera, 2005;Maad,
Armbruster & Fenster, 2013;Seguí et al., 2018). Intraspecific flower size can gradually
increase (Kudo & Molau, 1999;Herrera, 2005;Maad, Armbruster & Fenster, 2013;He et al.,
2017) or steadily decrease (Totland, 2001;Zhao & Wang, 2015;Hattori et al., 2016) with
increasing elevation. Increased intraspecific flower size at high elevations, as observed in
some entomophilous species, has been explained by pollinator selection for larger flowers at
high elevations. This is because, at high elevations, pollinators are rare but generally have a
greater size (Malo & Baonza, 2002;Maad, Armbruster & Fenster, 2013). At high elevations,
the reproductive success of outcrossing insect-pollinated plants is expected to be limited by
pollination (Zhao & Wang, 2015;Theobald, Gabrielyan & HilleRisLambers, 2016;Arroyo,
Pacheco & Dudley, 2017). An increased flower size corresponds to an increased insect
visitation rate (Totland, 2004), which, in turn, increases the chances of producing viable
seeds and achieving reproductive success (Arroyo, Primack & Armesto, 1982;Ohara &
Higashi, 1994;Bingham & Orthner, 1998). Conversely, reductions in flower size at high
elevations may result from individual plastic responses to climatic constraints (Zhao &
Wang, 2015). Flower size reductions may also result from local adaptation via abiotic
selection for smaller flowers in unfavourable environmental conditions. This selection
occurs due to the lower development and maintenance costs associated with smaller
flowers (Herrera, 2009).
Different selective abiotic and biotic pressures occurring at different elevations (Frei et
al., 2014) may further generate non-linear responses in plant morphological traits along
elevation gradients (Malo & Baonza, 2002). Such non-linear responses have been reported
for floral traits; a unimodal relationship was found between flower size and elevation,
with the maximum flower size of Cytisus scoparius (L.) Link (Malo & Baonza, 2002),
Solidago minuta L. (Kiełtyk, 2018) and Viola maculata Cav. (Seguí et al., 2018) occurring
in the middle of the elevational range. Such unimodal relationships may result from the
trade-off between pollinator selection for larger flowers (Totland, 2001;Totland, 2004;
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 2/15
Malo & Baonza, 2002;Maad, Armbruster & Fenster, 2013) and climatic selection for flower
miniaturisation at higher elevations (Herrera, 2009;Zhao & Wang, 2015).
Intraspecific morphological variation in plants growing across elevation gradients in
mountain regions is gaining increasing research attention (e.g., Takahashi & Matsuki,
2016;He et al., 2017;Olejniczak et al., 2018;Seguí et al., 2018). Knowledge of this variation
can help us to understand how plants have adapted to different environmental conditions
found across steep mountain gradients. In turn, this allows us to predict possible plant
responses to climatic change, particularly in cold mountain environments (Theurillat
& Guisan, 2001;Frei, Bodin & Walther, 2010). In this study, phenotypic variation was
investigated in the high-mountain plant, Bellidiastrum michelii Cass. (Asteraceae), across
a 1,155 m elevation gradient in the Tatra Mountains, Central Europe. The variation in
B. michelii traits with elevation were assessed using Generalized Additive Models (GAMs).
GAMs provide flexible, nonlinear functions that can account for both linear and non-linear
phenotypic responses without the need for a priori selection of candidate models. The aim
of this study was to contribute to obtaining a better understanding of morphological
variation in a single mountain species across elevation gradient. Specifically, the following
questions were addressed: do morphological traits of B. michelii significantly change across
the elevation gradient? And if so, what is it the variation pattern of these traits in relationship
to the elevation gradient?
MATERIAL AND METHODS
Species and study area
Bellidiastrum michelii Cass. (Asteraceae, syn. Arnica bellidiastrum All., Aster bellidiastrum
(L.) Scop., Doronicum bellidiastrum L.) is a perennial plant growing in the mountains
of Central and Southern Europe, from the Western Carpathians, Alps and Jura to the
Apennines and the western part of the Balkan Peninsula (Merxmüller & Schreiber, 1976;
Aeschimann et al., 2004). The scape of B. michelii is erect, 10–35 cm long, not leafy, with
one flower head at the top (Fig. 1). Leaves are suborbicular, spatulate, obovate or elliptical
in shape and gathered in a basal rosette. Flowers, gathered in the flower head, are insect-
pollinated, with outer female white or pink ligulate flowers and inner hermaphrodite yellow
flowers. The species blooms from May to August (Rostański, 1971;Merxmüller & Schreiber,
1976). B. michelii grows predominantly in grasslands, pastures, on rocks and along springs
on calcareous substrate (Rostański, 1971;Aeschimann et al., 2004).
The research was conducted along an elevation gradient in the Tatra Mountains in
southern Poland (Fig. 2), within the protected area of the Tatra National Park (study
permission of the Tatra National Park: Bot/380 DBN.503/63/16). The Tatras have the
highest altitudinal range among Carpathian Mountains (the highest peak, Gerlach, is at
2,655 m a.s.l. in the Slovak part of the mountains), with a well-developed subnival zone at
the highest elevations. The general elevational range in the Polish Tatras extends from ca.
900 m to 2,500 m a.s.l; the lower montane belt extends from an elevation of 600 m a.s.l. in
the lower regions of the Carpathian Mountains and reaches an elevation of 1,250 m a.s.l.
in the Tatra Mountains. The upper montane belt extends from ca. 1,250–1,500 (1,550)
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 3/15
Figure 1 Bellidiastrum michelii plants (A) and the species habitat (B) with white flowering B. michelii
plants in the subalpine belt at an elevation of 1,700 m a.s.l. in the Tatra Mountains.
Full-size DOI: 10.7717/peerj.11286/fig-1
m a.s.l., the subalpine belt from ca. 1,500 (1,550)–1,800 m a.s.l., the alpine belt from ca.
1,800–2,300 m a.s.l. and the subnival belt above 2,300 m a.s.l. (Mirek & Piekoś-Mirkowa,
1992). In the literature, B. michelii has been reported in the Polish Tatras at elevations
ranging from ca. 800 m a.s.l. in the Tatra foothills to 2,150 m a.s.l. (Rostański, 1971).
Field sampling and morphological measurements
In 2016, from the end of May to the beginning of July, the flowering plants of B. michelii
were collected from 34 sites ranging from 920 m a.s.l. to 2,075 m a.s.l. (Fig. 2,Table 1).
An attempt was made to ensure that the sampled sites were distributed approximately
evenly along the elevational range of the species (Table 1). The elevation at each site was
determined in the field using two devices: a GPS receiver with built-in barometric altimeter
(Garmin GPS MAP 76s, Olathe, USA) and a wrist barometric altimeter (Suunto Core
All Black, Vantaa, Finland); obtained elevation records were also consulted. To assure
recording accuracy, the altimeters of both devices were frequently calibrated at points
of known elevation, as read from topographic maps and field elevation markers. The
geographic coordinates and elevations of the study sites are presented in Table 1.
At each site, 10 plants with well-developed flower heads (in blossom peak) were sampled,
and their aboveground parts (i.e., flowering scapes with leaf rosettes) were collected. Each of
the sampled plants was sufficiently distant (>ca. 2 m) from other sampled plants to ensure
they represented different genetic individuals. Sampled individuals were well-developed
and did not show restriction in growth and reproductive function. In this study, 12
morphological traits were analysed; details on their measurements are presented in Table 2.
The first measurement was the flower head diameter expressed as the maximum distance
between the tips of two opposed petals. This measurement was made with a digital calliper
(Powerfix Profi, Neckarlsum, Germany) immediately after a plant was collected, when
the flowers were fresh and the flower head not deformed. The plants were then placed
in a botanical press, between drying paper and jute fabric, and preserved for further
analyses at the Plant Biology Laboratory of the Cardinal Stefan Wyszynski University in
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 4/15
state border
national park border
roads
lakes
sampling sites
Tatra National Park
N
S
WE
GERMANY
CZ EC H RE PU BL IC
SLOVAKIA
UKRAINE
POLAND
04km
1
23
4
5
6
78
9
10
11 12
13
14 15 16
17
18 19
SLOVAKIA
POLAND
20
21
22 23
24
25
26
27
28
29
31
32
33
34
30
Figure 2 Location of the 34 elevational sites sampled for Bellidiastrum michelii in the Polish Tatra
Mountains. For geographic coordinates and elevations, see Table 1.
Full-size DOI: 10.7717/peerj.11286/fig-2
Warsaw. Plant height was assessed by measuring the length from the plant base to the top
of the flower head. The plants were further separated into three fractions (scape, leaves
and flower head) and dried for 48 h at 80 ◦C in a laboratory drying oven with natural
air circulation (Pol-Eko-Aparatura SLN 240, Wodzisław Śląski, Poland) to obtain the
dry matter content (Pérez-Harguindeguy et al., 2013) by weighing on an analytical balance
(Radwag AS 60/220.X2 PLUS, Radom, Poland). The flower heads were then soaked in water
in a Petri dish for 1 h and then separated into ligulate and tubular flowers. This rehydration
of the flower heads and flowers was done to enable the separation of the flower head without
causing damage to the dry (and hence brittle) flowers and to ease straightening of the flower
corollas for subsequent measurements. The flowers were counted and the width of the
ligule and lengths of the ligulate and tubular flowers were measured using a stereoscopic
microscope (Delta Optical SZ-450T, Mińsk Mazowiecki, Poland). The number of ligulate
and tubular flowers was counted directly for each flower head, whereas the sizes of ligulate
and tubular flowers were measured for one randomly sampled flower from the head. After
the flowers were measured, they were stored in paper envelopes and dried for 48 h at 80
◦C in a drying oven with forced air circulation (Binder FD 115, Tuttlingen, Germany)
to obtain dry matter content, followed by weighing on the analytical balance. All weight
measurements were carried out immediately after the samples were removed from the
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 5/15
Table 1 Study sites of Bellidiastrum michelii in the Polish Tatra Mountains. Geographic coordinates
were determined with a WGS84 geodetic system.
Site Elevation (m a.s.l.) Latitude (N) Longitude (E) Date
1 920 49◦16053.800 19◦59053.300 2016-05-30
2 955 49◦16023.100 19◦55001.500 2016-06-17
3 970 49◦16008.800 19◦55050.500 2016-06-19
4 1040 49◦16036.000 19◦49031.200 2016-05-28
5 1090 49◦15043.600 19◦59005.600 2016-06-17
6 1135 49◦15029.900 19◦59025.900 2016-06-20
7 1162 49◦15028.800 19◦59032.300 2016-06-17
8 1225 49◦15025.500 19◦59037.000 2016-06-20
9 1253 49◦15012.700 19◦52052.500 2016-06-16
10 1290 49◦16000.300 19◦53023.400 2016-05-27
11 1305 49◦15016.100 19◦59040.500 2016-06-20
12 1330 49◦15041.900 19◦59042.000 2016-06-20
13 1370 49◦15015.300 19◦59047.400 2016-06-20
14 1385 49◦15002.400 19◦53005.900 2016-06-16
15 1400 49◦15009.100 19◦59052.600 2016-05-29
16 1430 49◦15017.400 19◦59057.700 2016-06-20
17 1470 49◦14055.000 19◦53012.000 2016-06-16
18 1480 49◦15015.900 20◦00008.300 2016-06-20
19 1505 49◦15016.300 20◦00013.200 2016-06-20
20 1531 49◦15021.000 20◦00016.900 2016-06-20
21 1570 49◦14056.600 19◦59049.200 2016-06-19
22 1600 49◦14041.500 19◦53015.800 2016-06-21
23 1620 49◦14052.800 19◦59049.500 2016-06-19
24 1660 49◦13036.600 19◦53052.600 2016-06-18
25 1695 49◦14034.000 19◦53025.200 2016-06-21
26 1707 49◦14038.700 19◦54042.200 2016-07-04
27 1755 49◦14026.000 19◦53028.200 2016-06-21
28 1790 49◦14020.100 19◦53032.100 2016-06-21
29 1818 49◦14041.400 19◦54048.500 2016-07-04
30 1843 49◦14018.700 19◦53037.700 2016-06-21
31 1925 49◦13038.400 19◦54014.900 2016-06-18
32 1977 49◦13042.000 19◦54014.300 2016-06-18
33 2030 49◦13046.900 19◦54013.100 2016-06-21
34 2075 49◦13049.600 19◦54012.600 2016-06-21
oven to prevent humidity absorption from air in the laboratory, which may influence the
weight measurements.
Statistical analyses
Statistical analyses were based on measurements of 12 morphological traits for 340 plants
collected from 34 sites (10 plants from each site) distributed continuously along the
1,155 m elevation gradient. All statistical analyses were performed using R version 3.6.1
(R Core Team, 2019). Prior to the analyses, the normality of distribution was checked
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 6/15
Table 2 Bellidiastrum michelii traits used in the study.
Trait Measurement details Accuracy/significant
digits
Plant height—measured from the scape base to the top of
the flower head.
Measured with a ruler on herbarium specimens. 1 mm
Plant mass—dry mass of the aboveground plant parts. Weighed on an analytical balance after drying for 48 h at 80
◦C.
0.0001 g
Total leaf mass—total dry mass of all plant leaves. As above. 0.0001 g
Flower head mass—dry mass of the flower head. As above. 0.0001 g
Total flower mass—total dry mass of ligulate and tubular
flowers in the flower head.
As above. 0.00001 g
Individual flower mass—calculated by dividing total flower
mass in the flower head by total number of flowers in the
head.
– 0.00001 g
Flower head diameter—the maximum distance between
the tips of two opposed petals.
Measured with a digital calliper on fresh plants in the field
immediately after plant collection.
0.1 mm
Number of ligulate flowers in the flower head. Counted by stereomicroscopy at ×10 magnification after
separation of the flower head into ligulate and tubular
flowers.
–
Number of tubular flowers in the flower head. As above. –
Ligulate flower length—corolla length measured for one
randomly chosen ligulate flower from the head.
Measured by stereomicroscopy at ×10 magnification after 1
h rehydration in water.
0.1 mm
Ligulate flower width—ligule width of the chosen flower. As above. 0.1 mm
Tubular flower length—corolla length measured for one
randomly chosen tubular flower from the head.
As above. 0.1 mm
for the all studied variables by comparing the observed variable distribution with the
theoretical normal distribution using the ggqqplot() function from the ggpubr package
(Kassambara, 2019) and by conducting the Shapiro–Wilk test for normality with the
shapiro.test() function of the stats package (R Core Team, 2019). None of the variables,
with the exception of the number of ligulate flowers and the ligulate flower length, were
normally distributed; therefore, to meet normal distribution, they were transformed with
the Box–Cox method using the boxcox() function of the bestNormalize package (Peterson
& Cavanaugh, 2019) and were standardised (mean =0 and standard deviation =1) for
subsequent analyses. The performance of the various methods of data transformation
available in the bestNormalize package was assessed and the Box–Cox method was found
to produce the best data normalisation in terms of agreement with the theoretical normal
distribution. However, for three variables (total flower mass in the head, tubular flower
length and ligulate flower width), a normal distribution could not be met by any method
of data transformation. Therefore, when assessing results of subsequent analyses for these
three variables, caution is required. The shape of an elevational variation in morphological
traits was assessed using Generalised Additive Models (GAMs). The GAMs were run on
the transformed and standardised trait values, and the elevation variable was expressed in
kilometres to avoid numerical estimation problems (Zuur, 2012). In the GAM models the
elevation was treated as a fixed effect and sites of samples collection were set as a random
model component. The GAMs were fitted with the Restricted Maximum Likelihood
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 7/15
(REML) method. The analyses were accomplished with the gamm4() function available in
the gamm4 package (Wood & Scheipl, 2017). In the GAMs summary the effective degrees
of freedom (edf) represent the complexity of the smoothing. An adf of 1 is equivalent to a
straight line, an edf of 2 is equivalent to a quadratic curve, etc., with higher edfs describing
more wiggly curves. The F-statistics and P-values are test statistics used in ANOVA to test
the significance of the GAM models.
RESULTS
The GAMs of the relationships between plant traits and elevation revealed that 9 of
the 12 studied traits changed significantly with elevation (Table 3). Plant size decreased
considerably with increasing elevation. The plant height, plant mass and total leaf mass
decreased by 69%, 64% and 66%, respectively (Figs. 3A–3C), across the 1,155 m elevation
gradient. Hereafter, the percentage changes in trait values will refer to the 1,155 m elevation
gradient, unless stated otherwise. Flower head mass and total flower mass showed slightly
hump-shaped patterns (Figs. 3D and 3E). Initially, from 920 m a.s.l. to 1,225 m a.s.l., the
flower head mass increased by 3%, and then from 1,225 m a.s.l. to 2,075 m a.s.l, decreased
by 22%. The total flower mass initially increased by 0.5% from 920 m a.s.l. to 1,040 m a.s.l.,
and then decreased by 23% from 1,040 m a.s.l. to 2,075 m a.s.l. However, it is important
to note that these initial slight increases in flower head mass and total flower mass at low
elevations may not be statistically significant. This is because the 95% confidence intervals
for the fitted model lines were wider than the observed hump-shaped patterns in the fitted
lines at low elevations (Figs. 3D and 3E). Meanwhile, the individual flower mass decreased
by 20% (Fig. 3F) and the flower head diameter decreased by 25% (Fig. 3G) from the
lowest to highest elevation. The number of flowers in the flower head, both ligulate and
tubular, showed no elevational tendency. These traits were very variable at all sites across
the elevational species range and did not exhibit any elevational trend (Figs. 3H and 3I).
Ligulate flower length decreased by 23% between the lowest and highest elevation (Fig. 3J).
Ligulate flower width showed high variation at all sites across the elevation gradient and
did not exhibit any significant relationship with elevation (Fig. 3K). Tubular flower length
decreased by 7% from the lowest to the highest elevation (Fig. 3L).
DISCUSSION
Most of the morphological traits of B. michelii varied significantly across the elevation
gradient. The plant height, total aboveground mass and total leaf mass decreased
considerably from low to high elevations. Similarly, floral traits, such as flower head
mass, total flower mass, individual flower mass, flower head diameter, and ligulate and
tubular flower length, also decreased with increases in elevation. However, the reductions
in these floral traits were not as great as those observed for plant size traits. Meanwhile, the
number of ligulate and tubular flowers produced by the plant did not change across the
studied elevation gradient. The intraspecific variation observed in this study suggests that
variable stress factors correlated with elevation impact plant morphotype. The elevational
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 8/15
Table 3 Generalised Additive Model (GAM) summaries for fitting Bellidiastrum michelii traits to elevation. The significance level =0.05; ns,
non significant models. The effective degrees of freedom (edf) represent the complexity of the smoothing. An adf of 1 is equivalent to a straight line,
an edf of 2 is equivalent to a quadratic curve, etc., with higher edfs describing more wiggly curves.
Trait edf F P R-sq (adj.) Fitted value at
920 m a.s.l.
Fitted value at
2075 m a.s.l.
Change 920–2075
m a.s.l. (%)
Plant height (mm) 1.639 34.24 0.000 0.528 278.5 86.3 –69
Plant mass (g) 1.709 21.97 0.000 0.314 0.3152 0.1122 –64
Total leaf mass (g) 1.000 30.74 0.000 0.242 0.2040 0.0700 –66
Flower head mass (g) 2.198 4.93 0.006 0.091 0.0349 0.0279 –20*
Total flower mass (g) 1.870 5.54 0.017 0.088 0.0203 0.0157 –23**
Individual flower mass (g) 1.640 6.70 0.013 0.098 0.00015 0.00012 –20
Flower head diameter (mm) 2.256 27.21 0.000 0.386 29.93 22.56 –25
No. ligulate flowers 1.000 0.00 0.953 ns 0.003 – – –
No. tubular flowers 1.108 0.90 0.386 ns 0.000 – – –
Ligulate flower length (mm) 2.290 19.25 0.000 0.325 11.95 9.14 –23
Ligulate flower width (mm) 1.000 0.48 0.487 ns 0.001 – – –
Tubular flower length (mm) 1.447 6.31 0.015 0.073 3.84 3.57 –7
Notes.
*Flower head mass had maximum fitted value of 0.0359 g at elevation 1,225 m a.s.l.
**Total flower mass had maximum fitted value of 0.0204 g at elevation 1,040 m a.s.l.
variation in morphology of B. michelii may result partially from phenotypic plasticity and
partially from local adaptations at different elevations.
The considerable decrease in intraspecific plant size with increasing elevation observed
in this study is in line with the results of previous studies (e.g., Nishizawa et al., 2001;
Alexander et al., 2009;Zhu et al., 2010;Maad, Armbruster & Fenster, 2013;Paudel et al.,
2019). These within-species size reductions may result from restricted growth caused by
resource limitations, such as low air temperature, short growing seasons, strong winds,
and shallow soil, that occur at high altitudes (Körner, 2003;Nagy & Grabherr, 2009).
Moreover, these intraspecific plant size reductions may also partly result from local
adaptation to high elevation environments via selection. Smaller phenotypes are thought
to be more advantageous in limiting environmental conditions due to their lower resource
requirements, and thus may be favoured by selection (Herrera, 2005;Zhao & Wang, 2015).
In this study, significant elevational variations were also found in several floral traits. In
general, flowers are relatively invariant within a singular species, as this is important for
securing the compatibility of the species mating system. Therefore, intraspecific variations
in floral traits are often of adaptive significance (Nosil, 2012). Increasing flower size with
increasing elevation has been found in several mountain species (e.g., Kudo & Molau, 1999;
Maad, Armbruster & Fenster, 2013;He et al., 2017). This increase in flower size is thought
to represent an adaptation that enhances flower attractiveness to pollinators (Herrera, 2005;
Maad, Armbruster & Fenster, 2013). However, in B. michelii, neither flower nor flower head
size increased with increasing elevation. On the contrary, the sizes and masses of the flowers
and flower heads of B. michelii decreased as elevation increased. These observed reductions
in the floral traits of B. michelii at high elevations are in line with the results of other studies
that found that within-species flower size decreased with increasing elevation (e.g., Totland,
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 9/15
Elevation (m a.s.l.)
Plant mass
Flower head diameter
No. tubular flowers
No. ligulate flowers
Ligulate flower width
Ligulate flower length
Tubular flower length
Plant height
Total leaf mass
Flower head mass
Total flower mass
Individual flower mass
a) b) c)
e) f)
j) k) l)
-3
-2
-1
0
1
2
3R =0.528, P=0.000
sq
-3
-2
-1
0
1
2
3R =0.314, P=0.000
sq
-3
-2
-1
0
1
2
3R =0.242, P=0.000
sq
-3
-2
-1
0
1
2
3R =0.091, P=0.006
sq
-3
-2
-1
0
1
2
3R =0.088, P=0.017
sq
-3
-2
-1
0
1
2
3R =0.098, P=0.013
sq
-3
-2
-1
0
1
2
3R =0.386, P=0.000
sq
-3
-2
-1
0
1
2
3R =0.003, P=0.9
sq 53 ns
-3
-2
-1
0
1
2
3R =0.000, P=0.38
sq 6 ns
g) h) i)
d)
-3
-2
-1
0
1
2
3
1000
1200
1400
1600
1800
2000
-3
-2
-1
0
1
2
3
1000
1200
1400
1600
1800
2000
-3
-2
-1
0
1
2
3
1000
1200
1400
1600
1800
2000
R =0.325, P=0.000
sq R =0.001, P=0.487
sq ns R =0.073, P=0.015
sq
Figure 3 (A–I) Elevational variation in Bellidiastrum michelii traits fitted by Generalised Additive
Models (GAMs). All traits are standardised (mean =0 and standard deviation =1); the grey band repre-
sents a 95% confidence interval for the mean shape of the effect (smoother). P –p-value of model signifi-
cance test at 0.05 significance level; ns, non significant models. Summary of the GAMs are presented in Ta-
ble 3.
Full-size DOI: 10.7717/peerj.11286/fig-3
2001;Zhao & Wang, 2015;Hattori et al., 2016). Reductions in flower size and mass with
increasing elevation can be linked to the more limiting conditions caused by increases in
climate severity at high altitudes. Intraspecific flower miniaturisation has been suggested to
be advantageous for plants growing under the resource-limited environmental conditions
of high mountain elevations, due to the lower cost of flower development and maintenance
(Herrera, 2009;Zhao & Wang, 2015).
In conclusion, significant intraspecific morphological variations were found in B.
michelii across the elevation gradient. The results of this study suggest that important
ecological factors associated with elevation gradients drive intraspecific morphological
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 10/15
variation in mountain plants growing across large elevational ranges. It will be possible
to identify these drivers by thoroughly analysing local abiotic (such as climatic) and
biotic (such as pollinator services) ecological factors across elevation gradients. Notably,
the observed elevational variation in B. michelii probably results partially from plastic
responses to abiotic (mainly climatic) conditions and partially from genetic adaptation
to locally prevailing conditions via selection. However, further study will be required to
disentangle the contributions of genetic differences among locally adapted populations,
and of plastic responses, to morphological variation across elevation gradients. Such further
study will need to be based on experiments involving reciprocal plant transplantation to
different mountain elevations (e.g., Gonzalo-Turpin & Hazard, 2009;Hautier et al., 2009;
Scheepens, Frei & Stöcklin, 2010;Hamann et al., 2016).
ACKNOWLEDGEMENTS
I am grateful to Wiesława Kiełtyk for help in collecting plants in the field. I thank the
Research Department of the Tatra National Park for providing kind support. Three
anonymous reviewers are acknowledged for their constructive comments and suggestions
on the earlier version of manuscript.
ADDITIONAL INFORMATION AND DECLARATIONS
Funding
This work was supported by the Cardinal Stefan Wyszyński University in Warsaw. The
funders had no role in study design, data collection and analysis, decision to publish, or
preparation of the manuscript.
Grant Disclosures
The following grant information was disclosed by the author:
Cardinal Stefan Wyszyński University in Warsaw.
Competing Interests
The author declares no conflict of interest.
Author Contributions
•Piotr Kiełtyk conceived and designed the experiments, performed the experiments,
analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the
paper, and approved the final draft.
Field Study Permissions
The following information was supplied relating to field study approvals (i.e., approving
body and any reference numbers):
Field experiments were approved by the Tatra National Park (permission Bot/380,
DBN.503/63/16).
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 11/15
Data Availability
The following information was supplied regarding data availability:
The raw measurements are available as a Supplemental File.
Supplemental Information
Supplemental information for this article can be found online at http://dx.doi.org/10.7717/
peerj.11286#supplemental-information.
REFERENCES
Aeschimann D, Lauber K, Moser DM, Theurillat J-P. 2004. Flora alpina. Bern: Haupt
Verlag.
Alexander JM, Edwards PJ, Poll M, Parks CG, Dietz H. 2009. Establishment of
parallel clines in traits of native and introduced forbs. Ecology 90:612–622
DOI 10.1890/08-0453.1.
Arroyo MTK, Pacheco DA, Dudley LS. 2017. Functional role of long-lived flowers in
preventing pollen limitation in a high elevation outcrossing species. AoB Plants
9:plx050 DOI 10.1093/aobpla/plx050.
Arroyo MTK, Primack R, Armesto J. 1982. Community studies in pollination ecology in
the high temperate Andes of central Chile. I. Pollination mechanisms and altitudinal
variation. American Journal of Botany 69:82–97
DOI 10.1002/j.1537-2197.1982.tb13237.x.
Billings WD. 1974. Arctic and alpine vegetation: plant adaptations to cold summer
climates. In: Ives JD, Barry RG, eds. Arctic and alpine environments. London:
Methuen, 403–443.
Bingham RA, Orthner AR. 1998. Efficient pollination of alpine plants. Nature
391:238–239 DOI 10.1038/34564.
Frei E, Bodin J, Walther GR. 2010. Plant species’ range shifts in mountainous areas—all
uphill from here? Botanica Helvetica 120:117–128
DOI 10.1007/s00035-010-0076-y.
Frei ER, Hahn T, Ghazoul J, Pluess AR. 2014. Divergent selection in low and high eleva-
tion populations of a perennial herb in the Swiss Alps. Alpine Botany 124:131–142
DOI 10.1007/s00035-014-0131-1.
Gonzalo-Turpin H, Hazard L. 2009. Local adaptation occurs along altitudinal gradient
despite the existence of gene flow in the alpine plant species Festuca eskia.Journal of
Ecology 97:742–751 DOI 10.1111/j.1365-2745.2009.01509.x.
Guo H, Weiner J, Mazer SJ, Zhao Z, Du G, Li B. 2012. Reproductive allometry in
Pedicularis species changes with elevation. Journal of Ecology 100:452–458
DOI 10.1111/j.1365-2745.2011.01884.x.
Hamann E, Kesselring H, Armbruster GFJ, Scheepens JF, Stöcklin J. 2016. Evidence of
local adaptation to fine- and coarse-grained environmental variability in Poa alpina
in the Swiss Alps. Journal of Ecology 104:1627–1637
DOI 10.1111/1365-2745.12628.
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 12/15
Hattori M, Nagano Y, Shinohara Y, Itino T. 2016. Pattern of flower size variation along
an altitudinal gradient differs between Impatiens textori and Impatiens noli-tangere.
Journal of Plant Interactions 11:152–157 DOI 10.1080/17429145.2016.1226437.
Hautier Y, Randin CHF, Stöcklin J, Guisan A. 2009. Changes in reproductive in-
vestment with altitude in an alpine plant. Journal of Plant Ecology 2:125–134
DOI 10.1093/jpe/rtp01.
He JD, Xue JY, Gao J, Wang J, Wu Y. 2017. Adaptations of the floral characteristics
and biomass allocation patterns of Gentiana hexaphylla to the altitudinal gradient
of the eastern Qinghai-Tibet Plateau. Journal of Mountain Science 14:1563–1576
DOI 10.1007/s11629-017-4424-x.
Herrera J. 2005. Flower size variation in Rosmarinus officinalis: individuals, populations
and habitats. Annals of Botany 95:431–437 DOI 10.1093/aob/mci041.
Herrera J. 2009. Visibility vs. biomass in flowers: exploring corolla allocation
in Mediterranean entomophilous plants. Annals of Botany 103:1119–1127
DOI 10.1093/aob/mcp046.
Kassambara A. 2019. ggpubr: ‘ggplot2’ based publication ready plots. R package version
0.2.4.999. Available at https:// rpkgs.datanovia.com/ ggpubr/ .
Kelly CA. 1998. Effects of variable life history and insect herbivores on reproduction in
Solidago macrophylla (Asteraceae) on an elevational gradient. The American Midland
Naturalist 139:243–254
DOI 10.1674/0003-0031(1998)139[0243:EOVLHA]2.0.CO;2.
Kiełtyk P. 2018. Variation of vegetative and floral traits in the alpine plant Solidago
minuta: evidence for local optimum along an elevational gradient. Alpine Botany
128:47–57 DOI 10.1007/s00035-017-0197-7.
Körner CH. 2003. Alpine plant life. Functional plant ecology of high mountain ecosystems.
2ed. Berlin: Springer-Verlag DOI 10.1007/978-3-642-18970-8.
Kudo G, Molau U. 1999. Variations in reproductive traits at inflorescence and flower
levels of an arctic legume, Astragalus alpinus L.: comparisons between a subalpine
and an alpine population. Plant Species Biology 14:181–191
DOI 10.1046/j.1442-1984.1999.00012.x.
Maad J, Armbruster WS, Fenster CB. 2013. Floral size variation in Campanula rotun-
difolia (Campanulaceae) along altitudinal gradients: patterns and possible selective
mechanisms. Nordic Journal of Botany 31:361–371
DOI 10.1111/j.1756-1051.2013.01766.x.
Malo JE, Baonza J. 2002. Are there predictable clines in plant–pollinator interactions
along altitudinal gradients? The example of Cytisus scoparius (L.) Link in the
Sierra de Guadarrama (Central Spain). Diversity and Distributions 8:365–371
DOI 10.1046/j.1472-4642.2002.00161.x.
Merxmüller H, Schreiber A. 1976. Aster L. In: Tutin TG, Heywood VH, Burges NA,
Moore DM, Valentine DH, Walters SM, Webb DA, eds. Flora Europaea 4. London:
Cambridge University Press, 112–116.
Miljković D, Stefanović M, Orlović S, Stanković Nedić M, Kesić L, Stojnić S. 2019. Wild
cherry (Prunus avium (L.)L.) leaf shapes and leaf variations in natural populations at
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 13/15
different elevations. Alpine Botany 129:163–174
DOI 10.1007/s00035-019-00227-1.
Mirek Z, Piękoś-Mirkowa H. 1992. Plant cover of the Polish Tatra Mountains (S.
Poland). Veröff. Geobot. Inst. ETH, Stiftung Rübel, Zürich 107:177–199.
Nagy L, Grabherr G. 2009. The biology of alpine habitats. Oxford: Oxford University
Press.
Nishizawa T, Kinoshita E, Yakura K, Shimizu T. 2001. Morphological variation of
the head characters in Solidago virgaurea L. inhabiting three mountains in central
Honshu. Journal of Phytogeography and Taxonomy 49:117–127.
Nosil P. 2012. Ecological speciation. Oxford: Oxford University Press.
Ohara M, Higashi S. 1994. Effects of inflorescence size on visits from pollina-
tors and seed set of Corydalis ambigua (Papaveraceae). Oecologia 98:25–30
DOI 10.1007/BF00326086.
Olejniczak P, Czarnoleski M, Delimat A, Majcher BM, Szczepka K. 2018. Seed size in
mountain herbaceous plant changes with elevation in species-specific manner. PLOS
ONE 13(6):e0199224 DOI 10.1371/journal.pone.0199224.
Paudel BR, Dyer AG, Garcia JE, Shresta M. 2019. The effect of elevational gradient on
alpine gingers (Roscoea alpina and R. purpurea) in the Himalayas. PeerJ 7:e7503
DOI 10.7717/peerj.7503.
Pérez-Harguindeguy N, Díaz S, Garnier E, Lavorel S, Poorter H, Jaureguiberry P, Bret-
Harte MS, Cornwell WK, Craine JM, Gurvich DE, Urcelay C, Veneklaas EJ, Reich
PB, Poorter L, Wright IJ, Ray P, Enrico L, Pausas JG, De Vos AC, Buchmann N,
Funes G, Quétier F, Hodgson JG, Thompson K, Morgan HD, Ter Steege H, Vander
Heijden MGA, Sack L, Blonder B, Poschlod P, Vaieretti MV, Conti G, Staver AC,
Aquino S, Cornelissen JHC. 2013. New handbook for standardised measurement
of plant functional traits worldwide. Australian Journal of Botany 61:167–234
DOI 10.1071/BT12225.
Peterson RA, Cavanaugh JE. 2019. Ordered quantile normalization: a semiparametric
transformation built for the cross-validation era. Journal of Applied Statistics
47:2312–2327 DOI 10.1080/02664763.2019.1630372.
R Core Team. 2019. R: A language and environment for statistical computing. Vienna: R
Foundation for Statistical Computing. Available at https:// www.R-project.org/ .
Rostański K. 1971. Bellidiastrum Cass. In: Pawłowski B, Jasiewicz A, eds. Flora Polska 12.
Warszawa: Państwowe Wydawnictwo Naukowe, 135–136.
Scheepens JF, Frei ES, Stöcklin J. 2010. Genotypic and environmental variation in
specific leaf area in a widespread Alpine plant after transplantation to different
altitudes. Oecologia 164:141–150 DOI 10.1007/s00442-010-1650-0.
Scheepens JF, Stöcklin J. 2013. Flowering phenology and reproductive fitness along a
mountain slope: maladaptive responses to transplantation to a warmer climate in
Campanula thyrsoides.Oecologia 171:679–691
DOI 10.1007/s00442-012-2582-7.
Seguí J, Lázaro A, Traveset A, Salgado-Luarte C, Gianoli E. 2018. Phenotypic and
reproductive responses of an Andean violet to environmental variation across an
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 14/15
elevational gradient. Alpine Botany 128:59–69
DOI 10.1007/s00035-017-0195-9.
Takahashi K, Matsuki S. 2016. Morphological variations of the Solidago virgaurea L.
complex along an elevational gradient on Mt Norikura, central Japan. Plant Species
Biology 32:238–246 DOI 10.1111/1442-1984.12148.
Theobald EJ, Gabrielyan H, HilleRisLambers J. 2016. Lilies at the limit: variation in
plant–pollinator interactions across an elevational range. American Journal of Botany
103:189–197 DOI 10.3732/ajb.1500416.
Theurillat J-P, Guisan A. 2001. Potential impact of climate change on vegetation in the
European Alps: a review. Climatic Change 50:77–109
DOI 10.1023/A:1010632015572.
Totland Ø. 2001. Environment-dependent pollen limitation and selection on floral traits
in an alpine species. Ecology 82:2233–2244
DOI 10.1890/0012-9658(2001)082[2233:EDPLAS]2.0.CO;2.
Totland Ø. 2004. No evidence for a role of pollinator discrimination in causing
selection on flower size through female reproduction. Oikos 106:558–564
DOI 10.1111/j.0030-1299.2004.12991.x.
Wood S, Scheipl F. 2017. gamm4: Generalized Additive Mixed Models using ‘mgcv’ and
‘lme4’. R package version 0.2-5. Available at https:// CRAN.R-project.org/ package=
gamm4.
Zhao Z-G, Wang Y-K. 2015. Selection by pollinators on floral traits in generalized
Trollius ranunculoides (Ranunculaceae) along altitudinal gradients. PLOS ONE
10(2):e0118299 DOI 10.1371/journal.pone.0118299.
Zhu Y, Siegeolf RTW, Durka W, Körner CH. 2010. Phylogenetically balanced evidence
for structural and carbon isotope responses in plants along elevational gradients.
Oecologia 162:853–863 DOI 10.1007/s00442-009-1515-6.
Zuur AF. 2012. A beginner’s guide to generalized additive models with R. Newburgh:
Highland Statistics Ltd.
Kiełtyk (2021), PeerJ, DOI 10.7717/peerj.11286 15/15
