, Volume 187, Issue 4, pp 967–975 | Cite as

Leaf stable isotopes suggest shared ancestry is an important driver of functional diversity

  • Ellie M. Goud
  • Jed P. Sparks
Special Topic


Plant physiological strategies of carbon (C) and nitrogen (N) uptake and metabolism are often regarded as outcomes of environmental selection. This is likely true, but the role of evolutionary history may also be important in shaping patterns of functional diversity. Here, we used leaf C and N stable isotope ratios (δ13C, δ15N) as integrators of physiological processes to assess the relative roles of phylogenetic history and environment in a diverse group of Ericaceae species native to North America. We found strong phylogenetic signal in both leaf δ13C and δ15N, suggesting that close relatives have similar physiological strategies. The signal of phylogeny was generally stronger than that of the local environment. However, within some specialized environments (e.g., wetlands, sandy soils), we found environmental effects and/or niche conservatism. Phylogenetic signal in δ13C appears to be most closely related to the constraints on metabolic demand and supply of C, and δ15N appears to be most strongly related to mycorrhizal associations within the family.


δ13Ericaceae Mycorrhizae δ15Phylogenetic signal Physiological strategies 


Determining the mechanisms that shape the diversity of physiological and morphological features of organisms is a central question in ecology. However, in any given environment, it is challenging to partition the factors that may generate the observed adaptations. The physical environment directly drives adaptation through selection, but an underappreciated mechanism is the constraint put on an organism by its phylogenetic position. Ecologically important characters have traditionally been regarded as labile and physiological changes were assumed to be important responses to environmental variation. However, recent phylogenetic studies have revealed that many lineages of closely related species have maintained ecological and phenotypic similarities through evolutionary time (Wiens et al. 2010). Rather than selection alone driving the relationship between environment and physiology, shared ancestry may play an important role in maintaining groupings of traits important for success in a given environment.

Examining the relative influence of shared ancestry and environment on physiology is nothing new; this has been explored several times in plants by tracing the phylogenetic signal of anatomical characters (Savage and Cavender Bares 2012; Yang et al. 2014). Previous studies have focused on ecologically relevant characters or ‘functional traits’ that hopefully represent more direct measures of physiological processes. Examples include leaf nitrogen (N) content as a proxy for photosynthetic rate, and leaf lifespan, size and leaf mass per area (or its inverse, specific leaf area) as proxies for rates of resource acquisition and conservation (Wright et al. 2004; Osnas et al. 2013). One challenge with this approach is that trait-environment relationships and phylogenetic signal are often inconsistent among taxonomic groups and environments (Kerkhoff et al. 2006; Yang et al. 2014; Flores et al. 2014; Forrestel et al. 2015; Bhaskar et al. 2016), making interpretation and generalization difficult. One potential reason for these inconsistencies is that traits may correlate with multiple processes that have contrasting relationships with the environment and/or different rates of evolution. Because physiological responses often involve suites of inter-related traits, single traits like leaf N content or specific leaf area may only describe a part of a physiological process or even function differently across species. Instead of looking at individual functional traits, the alternative approach we use in this work is to use a plant measurement that potentially integrates a physiological process. Stable isotope ratios of leaf C and N are good candidates because they integrate physiological processes and often vary in relation to known ecological variables.

Leaf carbon isotope (δ13C) values are directly related to the integrated strategy a plant uses for C acquisition. δ13C reflects the long term balance of leaf internal (ci) to external (ca) carbon dioxide (CO2) concentrations, which is dependent upon the resistances to CO2 entry into the leaf and the rate of CO2 consumption during photosynthesis (Farquhar et al. 1982). Any factor that increases the difference between ci and ca leads to δ13C enrichment, such as photosynthetic capacity and investment in photosynthetic machinery, the size of the boundary layer of still air around the leaf, the length of the diffusion pathway between the stoma and the site of carboxylation within the chloroplast, and stomatal conductance. Morphological characteristics that increase the size of the leaf boundary layer include increasing leaf size, and surface characters such as hair, waxes and trichomes (Ehleringer et al. 1976; Evans and Loreto 2000). Stomatal conductance and behavior (i.e., how often and for how long the stomata are open relative to closed) are sensitive to changes in temperature and water availability, which would also be reflected in the δ13C value (Ehleringer et al. 1992).

Leaf nitrogen isotope ratios (δ15N) integrate variation in plant metabolism, differences in ecosystem N cycling and mutualist associations that facilitate biotic N uptake such as mycorrhizal fungi and N-fixing bacteria (Hobbie et al. 1999). Plant metabolic processes that fractionate 15N include different pathways of N assimilation (Yoneyama et al. 1991) and internal recycling of N in the plant (Kolb and Evans 2002). Leaf δ15N values are also affected by changes to the baseline value of δ15N present in the soil solution that is available for plant uptake. Any factor that changes the rate of soil nutrient turnover (e.g., temperature, water availability) via mineralization, nitrification or denitrification can potentially change the baseline δ15N of the soil solution and consequently in the leaves (Amundson et al. 2003). Additionally, leaf δ15N values can vary substantially among species with belowground mutualist associations because different microorganisms access different soil N-pools which consequently affects 15N fractionation (Hobbie et al. 1999; Hobbie and Colpaert 2004; Hobbie and Hobbie 2006). Although not as clearly associated to a single physiological process as δ13C, δ15N can also integrate physiological interactions with the environment.

Determining phylogenetic signal in leaf δ13C and δ15N can shed light on the relative strength of shared ancestry and environment in structuring functional diversity. While δ13C and δ15N can vary among closely related species, they have not been evaluated from a phylogenetic perspective because it is generally thought that variation in stable isotope values are predominantly driven by environmental factors, not evolutionary history. However, if metabolic rates are phylogenetically constrained based on similar groupings of physiological traits (Liu et al. 2015), then there may be phylogenetic signal in leaf δ13C and δ15N. Moreover, if mutualist associations that are important for resource uptake, such as mycorrhizal fungi, are phylogenetically conserved, then the physiology that depends on these mutualisms would also have phylogenetic signal.

The heath family (Ericaceae) is a diverse and geographically widespread plant group that has radiated into a variety of environments, including arctic and alpine tundra, wetlands, boreal and hardwood forests, scrublands and chaparral. 212 species from 46 genera are native to North America and display extensive variation in anatomy, physiology and ecology (Tucker 2009). Phylogenetic and ecological diversity within Ericaceae makes this plant family an excellent study system for investigating leaf δ13C and δ15N within a phylogenetic context. Foliar δ13C and δ15N values are expected to vary among Ericaceae taxa based on ecological differences such as climate (e.g., temperature, precipitation), latitude and elevation (Friend et al. 1989). Most Ericaceae are evergreen shrubs (e.g., Rhododendron, Vaccinium), but deciduous shrubs and trees, herbaceous perennials, and parasitic species that lack chlorophyll are also common in North America (Tucker 2009). As such, δ13C values could also vary with plant growth form (e.g., trees, shrubs, herbs), leaf persistence (e.g., evergreen, deciduous), and metabolism (e.g., fast versus slow growth rates). Moreover, Ericaceae species are often associated with acidic and nutrient poor soils, and their success is largely attributed to mutualisms with three unique types of mycorrhizal fungi: arbutoid, ericoid and monotropoid mycorrhizae (Lallemand et al. 2016). These mycorrhizal types each fractionate N in different ways, which would result in different foliar δ15N values among host plants (Zimmer et al. 2007).

The objectives of this study were to (1) determine the extent of phylogenetic signal in leaf δ13C and δ15N and (2) to assess how phylogeny and environmental conditions relate to δ13C and δ15N variation. If shared ancestry is an important driver of physiological diversity, then leaf δ13C and δ15N values will have phylogenetic signal and similarity in δ13C and δ15N values among species will correlate more strongly with phylogeny than environment. To test these predictions, we compared relationships between δ13C and δ15N, phylogeny and environment and evaluated the strength of phylogenetic signal from a molecular phylogeny of 57 Ericaceae species native to North America. We then examined factors that could potentially contribute to phylogenetic signal in leaf stable isotopes, including different mycorrhizal types and other leaf traits important to C and N acquisition and metabolism.

Materials and methods

To determine the phylogenetic relationships among species, we constructed a phylogeny using maximum likelihood analyses in PAUP (Swofford 2002) using sequence data for all Ericaceae species native to North America that had available matK, nrITS and rbcL gene regions from GenBank (107 species total). The tree was rooted on two outgroups: Enkianthus chinensis and E. campanulatus and the final tree was topologically congruent with published phylogenies of the Ericaceae as a whole (Kron et al. 2002). The tree was then pruned to 57 species from 12 genera that represent the range of phylogenetic, environmental and physiological diversity of the group.

To generate trait data, we collected leaves of the 57 Ericaceae species represented on the pruned phylogeny from herbarium specimens at the Liberty Hyde Bailey Hortorium (Cornell University) in August, 2015. The 57 species were geographically widespread across North America (Fig. 1). Leaf material (5–50 leaves, depending on leaf size) was sampled from four vouchers per species from a range of habitats and geographic locations. Each leaf was assessed for leaf longevity (deciduous, evergreen), leaf surface characteristics (glabrous, hairy), leaf size, percent element (%C, %N), C:N, and isotopic ratio (δ13C, δ15N). Isotope ratios and percent element of all samples were measured using a continuous flow isotope ratio mass spectrometer (Thermo Finnigan Environmental Delta V) coupled to an elemental analyzer (Thermo Finnigan Carlo Erba NC2500). Isotope ratios are expressed as δ values (per mil):
$$\delta^{15} {\text{N }}\text{or} \, \delta^{13} \text{C} = \left( {R_{\text{sample}} /R_{{\text{standard}}} - 1} \right) \times 1000 ( {\permille})$$
where Rsample and Rstandard are the ratios of heavy to light isotope of the sample relative to the international standards for C and N, Vienna-Pee-Dee Belemnite and atmospheric N2, respectively. To account for atmospheric CO2 source changes in δ13C over the sampling time period (1895–2015), we adjusted leaf δ13C values accordingly. Atmospheric CO2–δ13C isotope values from 1890 to 1970 were taken from Zachos (2007) and values from 1970 to 2015 were obtained online from the Scripps CO2 Program (, Mauna Loa Observatory, Hawaii). Using − 8‰ as a baseline, we calculated the difference between the baseline and atmospheric CO2–δ13C during the collection year for each herbarium specimen.
Fig. 1

Geographic locations of herbarium specimen collection sites across North America for 57 Ericaceae species (n = 225). The legend shows phylogenetic relationships among the 12 genera represented in this study. The phylogeny was constructed by maximum likelihood analyses using sequence data from matK, nrITS and rbcL gene regions from GenBank. This figure is available in color in the online version

We classified each species by mycorrhizal type using information from published floras and primary literature. Each species is reported to identify with one of three mycorrhizal types: arbutoid, ericoid or monotropoid mycorrhizae (Lallemand et al. 2016). Arbutoid and monotropoid mycorrhizal fungi are similar to ectomycorrhizal fungi (Smith and Read 2008a), while ericoid mycorrhizal fungi are similar to arbuscular/endomycorrhizal fungi (Smith and Read 2008b).

We used vapor pressure deficit (VPD) as a general description of a plant’s average environment in terms of temperature and water availability. VPD determines the difference between the amount of moisture in the air and the amount of moisture that the air can hold, which in turn drives water loss from plant leaves. This difference is temperature-dependent, as warmer air can hold more water than colder air. Thus, variation in VPD affects stomatal behavior and leaf gas exchange (Oren et al. 1999), and potentially C isotope discrimination (Bowling et al. 2002). Given that leaf δ15N values vary with mean annual precipitation and temperature (Amundson et al. 2003), VPD also, at least indirectly, influences leaf δ15N values (Craine et al. 2015). We calculated the average July VPD for each origin environment, as indicated from specimen voucher labels. We generated this dataset using long-term (1974–2012) July average air temperature and dew point values obtained from weather stations closest to the collection locations (National Weather Service,

In addition to the average climate environment represented by VPD, we also independently examined species within five specialized environments where either soil type (e.g., sand) or water inundation (e.g., wetlands) was likely to uncouple physiology from the local VPD environment. Using habitat descriptions from herbarium voucher labels in combination with ecological information from published floras, we identified species in our dataset that occupy peatlands, riparian zones, rock barrens, sandy soils, and swamps. Peatlands are acidic, nutrient poor wetlands common in arctic, boreal and temperate regions. Swamps and riparian zones are distributed throughout North America, and many of the species in our dataset were from swamps and riparian zones in the southern United States. Rock barrens are nutrient poor environments characterized by open rock overlain by patches of shallow organic soil and are restricted to artic and alpine tundra, particularly in coastal regions of eastern Canada. Ericaceae in sandy soils are from California chaparral, sand scrub of the southeastern US, and pine barrens of the Atlantic coast and northern Rocky Mountains (Tucker 2009).

To assess the degree to which closely related species are more similar to one another than expected by chance, we calculated phylogenetic signal using Pagel’s λ for continuous (δ15N, δ13C, %C, %N, C:N, leaf size, VPD) and categorical variables (longevity, surface characteristics, mycorrhizal type, specialized environments). Pagel’s λ is a branch scaling parameter, where λ = 1 indicates that the trait distribution scales with tree topology in accordance with Brownian motion. λ = 0 indicates that the tree topology does not structure trait variation (Pagel 1999; Münkemüller et al. 2012). We tested whether λ was > 0 by comparing the log-likelihood of the fitted λ with that of λ = 0 using a log-likelihood ratio test using the ‘phylosig’ function in the phytools R package (Revell 2012). Variables with λ > 0.5 (at α = 0.05) have phylogenetic signal (i.e., relatives are more similar to each other than expected by random chance) (Pagel 1999; Münkemüller et al. 2012).

To compare isotopic, phylogenetic, and environmental (VPD) similarity among species, we calculated the absolute difference in isotope values and VPD and the phylogenetic distance among all pair-wise species comparisons. Absolute differences were converted into similarities by subtracting absolute differences from one (S = 1 − D). We evaluated relationships between phylogenetic, isotopic and environmental (VPD) similarity using linear regression. All analyses were performed in R3.2.4 (R Core Team 2016).


Phylogenetic signal was observed in δ15N and δ13C, %C, leaf longevity (deciduous, evergreen), hairy leaves and mycorrhizal type, but not in %N, C:N, VPD, leaf size, or glabrous leaves (Table 1). Similarity in δ15N and δ13C increased with phylogenetic similarity and explained 32 and 20% of the total variation in these traits, respectively (P < 0.0001, Fig. 1a, b). Similarity in δ13C and δ15N was also related to similarity in VPD, but these relationships explained less than 1% of the total variation (P < 0.02, Fig. 1c, d).
Table 1

Phylogenetic signal using Pagel’s λ of 10 leaf characteristics, average July vapor pressure deficits and mycorrhizal type (Arbutoid, Ericoid, Monotropoid) for 57 Ericaceae species (n = 225)


Pagel’s λ








< 0.01

C:N ratio

< 0.01

Leaf size

< 0.01

Leaf longevity


Hairy leaves


Glabrous leaves

< 0.01

Mycorrhizal type


Vapor pressure deficit

< 0.01

Bold values indicate significant phylogenetic signal

Phylogenetic signal is significant at Pagel’s λ > 0.5 (α = 0.05), indicating variables that are more similar among close relatives than expected by chance

Species within each specialized environment were isotopically similar to other species from that environment (Table 2). Species from rock barrens and sandy soils tended to be more closely related to each other (λ = 0.77 and 0.88, respectively) while species from peatlands, riparian zones and swamps were more distantly related (λ = 0.38, < 0.01 and < 0.01, respectively).
Table 2

Phylogenetic signal and isotopic similarity among Ericaceae plants from specialized environments


Pagel’s λ

δ13C similarity

δ15N similarity

Peatlands (n = 26)

< 0.01



Riparian (n = 26)




Rock barrens (n = 51)




Sandy soils (n = 40)




Swamps (n = 15)

< 0.01



Bold values indicate significant phylogenetic signal

Pagel’s λ > 0.5 indicates phylogenetic signal. Isotopic similarity ranges from 0 to 1, with 1 being completely similar and 0 being completely different. The number of individuals from each environment is indicated in parentheses

δ13C and δ15N varied in relation to the type of mycorrhizal association (Fig. 2). Individuals with Monotropoid mycorrhizae displayed the most enriched δ13C and δ15N values, followed by Arbutoid and Ericoid.
Fig. 2

Relationships between (a) phylogenetic similarity and similarity in leaf δ15N (r2 = 0.32, P < 0.0001), (b) phylogenetic similarity and similarity in leaf δ13C (r2 = 0.20, P < 0.0001), (c) environmental similarity and similarity in leaf δ15N (r2 = 0.002, P = 0.02), and (d) environmental similarity and similarity in leaf δ13C (r2 = 0.01, P < 0.0001) for 57 Ericaceae species (n = 225). Phylogenetic similarity is derived from species pair-wise comparisons of phylogenetic branch lengths. Environmental similarity is derived from species pair-wise comparisons of average July vapor pressure deficits. Similarities range from 0 to 1, with 1 being completely similar and 0 being completely different


We present evidence that physiological strategies related to C and N acquisition in the North American Ericaceae are strongly influenced by evolutionary history. As integrators of physiological processes, we suggest that isotope ratios may be a more robust predictor relative to commonly measured anatomical leaf characteristics. In support of this hypothesis, leaf δ13C and δ15N values exhibited strong phylogenetic signal (Table 1), suggesting that close relatives have similar physiological strategies with respect to C and N acquisition and metabolism.

Observing strong phylogenetic signal in isotope ratios does not necessarily preclude a strong relationship between isotope ratios and environment. To this end, we examined the relationship between foliar δ13C and δ15N over a range of environmental vapor pressure deficits. Representing a plant’s average environment with vapor pressure deficits is a coarse measure of environment, but it is one that has long been shown to drive global patterns of plant diversity (Hutchinson 1918; Sexton et al. 2009). If species living in similar environments have comparable physiology, then similarity in δ13C and potentially in δ15N should correlate with similarity in vapor pressure deficits. However, the relationships we observed between δ13C, δ15N and vapor pressure deficits explained very little variation (Fig. 2c, d), suggesting that species with similar physiology do not necessarily occupy similar average environments. Surprisingly, the signal of phylogeny on physiological variation among species was much stronger than that of the average environment (Fig. 2a, b). This combination of strong phylogenetic signal and a weak relationship with environment suggests that closely related species have similar physiological strategies across different environments.

While a coarse-scale environmental variable, VPD, explained a very small amount of variation in physiology, finer-scale microsite conditions could explain a larger portion of the variation in these traits and processes. For example, a set of species could be pre-adapted to certain conditions and track favorable microclimates across different vapor pressure deficit environments. In this case, close relatives maintain some common suite of characteristics that together contribute to variation and phylogenetic signal in foliar δ13C and δ15N. To explore this possibility, we independently examined several environments where, because of soil type or constant soil water availability, physiological patterns were less likely to be tightly coupled to the vapor pressure deficit environment. Peatlands, riparian zones, rock barrens, sandy soils and swamps are geographically widespread across North America and Ericaceae species within these environments all had similar δ13C and δ15N values relative to each other (Table 2), providing some evidence that similar physiological strategies in these species have arisen in response to strong environmental selection. However, the degree of phylogenetic relatedness was not constant across these specialized environments (Table 2). Individuals from rock barrens and sandy soils were more closely related than expected by chance (Pagel’s λ > 0.75; Table 2). Rock barrens and sandy soils are water-limited environments and may represent examples of ‘phylogenetic niche conservatism’, whereby species maintain the phenotypes and environmental requirements of their most recent common ancestor (Wiens et al. 2010). In contrast to rock barrens and sandy soils, individuals from peatlands, riparian zones and swamps were physiologically similar but more distantly related than expected by chance (Pagel’s λ < 0.5; Table 2); this suggests that plants from these wet habitats are potentially converging onto similar strategies regardless of their phylogenic history.

These patterns diverge from other studies that report phylogenetic similarity among wetland species (Savage and Cavender Bares 2012) and phylogenetic divergence (species less related) among species from drier environments (Savage and Cavender Bares 2012; Jara Arancio et al. 2014). A possible explanation is that the direction of environmental shift is influencing phylogenetic and phenotypic patterns. Rather than the environmental conditions per se (i.e., arid versus mesic) driving convergence or divergence, the degree of deviation from ancestral conditions may be more important (Klak et al. 2004; Kraft et al. 2007). For example, shifting into a new environment that shares features in common with the ancestral state may not require substantial adjustments or adaptations. However, a new environment that is considerably different from the ancestral state in climate or soil conditions may require novel adaptations and phenotypic divergence over time. The ancestral Ericaceae environment was likely arid and water-limited, so a shift into other water-limited environments, such as rock barrens, may not have required novel adaptations. On the other hand, key innovations for tolerance to water-logged soils may have been necessary to successfully radiate into wetlands. Convergent adaptations may have arisen independently in multiple taxa, resulting in a pattern of phylogenetic diversity and phenotypic similarity in wetlands.

Given that δ13C integrates multiple aspects of C acquisition in plants, our observation of phylogenetic signal might be driven by similar groupings of traits defined by phylogenetic history. We examined the strength of phylogenetic signal for traits that impact the balance of leaf external and internal CO2 concentrations (ci/ca) and C gain through their effects on CO2 diffusive resistance or carboxylation. Leaf hair, leaf longevity and leaf C content (%) all had phylogenetic signal (Pagel’s λ > 0.5; Table 1) and, therefore, may be contributing to phylogenetic signal in δ13C. The presence of leaf hair affects the size of the leaf boundary layer and contributes to CO2 diffusive resistance (Ehleringer et al. 1976; Ehleringer and Mooney 1978; Meinzer and Goldstein 1985), while leaf longevity and C content relate to structural investment and stomatal control (Evans and Loreto 2000). We might also expect leaf size and N content to have an impact on C gain and potentially contribute to phylogenetic signal because of their effect on photosynthetic efficiency and carboxylation capacity (Wright et al. 2004), but these traits did not have phylogenetic signal (Pagel’s λ < 0.01; Table 1). Despite these limitations, integrative measures such as δ13C appear to have more power than individual anatomical traits in this context because they are integrating multiple factors that influence C gain, including characters that are not often measured.

Much of the variation and phylogenetic signal in leaf δ15N appears to be related to belowground mutualisms with different mycorrhizal fungi. Mycorrhizal fungi influence the leaf δ15N values of their host plant both in the amount of N that they supply to the plant and in the isotopic composition of the N itself. The three mycorrhizal types that associate with Ericaceae species are more host specific than typical arbuscular and ectomycorrhizal fungi (Smith and Read 1997; Lallemand et al. 2016). Species with monotropoid mycorrhizae are ‘mycoheterotrophic’, meaning they are parasitic, do not photosynthesize (e.g., Monotropa, Allotropa) and are completely dependent on mycorrhizal fungi for both C and N (Smith and Read 2008a). Mycoheterotrophic species were considerably more enriched in δ13C and δ15N values (Fig. 3). Strong enrichment in leaf δ13C and δ15N in mycoheterotrophic plants relative to other autotrophs is attributed to their dependence on isotopically enriched C and N sources that have been metabolized by mycorrhizal fungi (Tedersoo et al. 2006; Zimmer et al. 2007). Similarly, plants with arbutoid mycorrhizae can be either autotrophic (obtaining C from photosynthesis) or ‘mixotrophic’, meaning that they photosynthesize but also rely heavily on isotopically enriched C from mycorrhizal fungi sources (e.g., Pyrola, Chimaphila). While ericoid and arbutoid mycorrhizae deliver isotopically depleted N from the soil to their host plant, arbutoid mycorrhizae can also exchange isotopically enriched N that is internal to the fungi rather than from the soil (Tedersoo et al. 2006). Consistent with this, species with arbutoid mycorrhizae in our study were relatively enriched in δ15N, although δ13C values did not differ between arbutoid and ericoid mycorrhizal types (Fig. 3). The large spread in δ15N values for arbutoid mycorrhizae is likely driven by opposing values from those species that use enriched fungal N and other species that use depleted soil N, yet both are classified as associating with arbutoid mycorrhizae (Zimmer et al. 2007).
Fig. 3

Results of a one-way analysis of variance between leaf (a) δ13C and (b) δ15N and three mycorrhizal associations (Arbutoid, Ericoid, Monotropoid) for 57 Ericaceae species (n = 225). Asterisks indicate significant differences among mycorrhizal types (P < 0.0001)

These large mycorrhizal effects are operating across ecosystems with differential baselines in soil 15N. Variation in the baseline 15N present in the soil solution is influenced in part by climate (e.g., annual precipitation and temperature) and local conditions (e.g., water availability) such that cold and/or wet sites are generally more depleted in δ15N relative to warm and/or dry sites. These ecosystem differences would be reflected in the δ15N of plant leaves (Hobbie et al. 2000; Kahmen et al. 2008; Craine et al. 2009, 2015). The baseline soil 15N, and consequently leaf δ15N, are also influenced by soil N turnover rates, which can depend on rates of root N uptake, leaf N turnover and litter decomposition. Phylogenetic signal in leaf δ15N may, therefore, be related to groupings of traits associated with root N uptake and leaf N turnover. The effects of root N uptake rates in Ericaceae are likely to be predominantly reflected in the mycorrhizal effects discussed above. Leaf traits related to leaf N turnover include leaf longevity and C:N. Evergreen leaves and leaves with large C content and/or large C:N generally have slower rates of nutrient turnover and can be relatively depleted in δ15N values compared to deciduous leaves and leaves with less C content (Amundson et al. 2003). We found phylogenetic signal in leaf longevity and C content but not in C:N. Although leaf longevity and C content may contribute to δ15N phylogenetic signal, their influence is smaller than the primary mycorrhizal effects.

This is the first study to assess patterns of leaf C and N stable isotopes from a phylogenetic perspective. We found significant phylogenetic signal in leaf δ13C and δ15N, which shows that shared ancestry is an important driver of functional diversity in the North American Ericaceae. Moreover, we demonstrate that the signal of phylogeny is stronger than that of the average, coarse-scale environment, although environmental selection and phylogenetic niche conservatism may be important at finer spatial scales. We recommend the use of stable isotopes in future studies to better represent physiology. Defining and measuring the physical environment in a more consistent way across systems will aid in identifying the underlying mechanisms that structure functional diversity.



We would like to thank Kevin Nixon, Anna Stalter and the Cornell Liberty Hyde Bailey Hortorium for granting access to herbarium vouchers. We are grateful to Kim Sparks and John Pollack for technical assistance and Elizabeth Murray for statistical support.

Author contribution statement

EMG and JPS conceived and designed the project. EMG collected and analyzed the data. EMG and JPS wrote the manuscript.


This study was funded by a Kieckhefer Adirondack Research Grant to EMG.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

442_2018_4186_MOESM1_ESM.xlsx (58 kb)
Supplementary material 1 (XLSX 57 kb)


  1. Amundson R, Austin AT, Schuur EAG et al (2003) Global patterns of the isotopic composition of soil and plant nitrogen. Glob Biogeochem Cycles. CrossRefGoogle Scholar
  2. Bhaskar R, Porder S, Balvanera P, Edwards EJ (2016) Ecological and evolutionary variation in community nitrogen use traits during tropical dry forest secondary succession. Ecology 97:1194–1206. CrossRefPubMedGoogle Scholar
  3. Bowling DR, McDowell NG, Bond BJ et al (2002) 13C content of ecosystem respiration is linked to precipitation and vapor pressure deficit. Oecologia 131:113–124. CrossRefPubMedGoogle Scholar
  4. Craine JM, Elmore AJ, Aidar MPM et al (2009) Global patterns of foliar nitrogen isotopes and their relationships with climate, mycorrhizal fungi, foliar nutrient concentrations, and nitrogen availability. New Phytol 183:980–992. CrossRefPubMedGoogle Scholar
  5. Craine JM, Elmore AJ, Wang L et al (2015) Convergence of soil nitrogen isotopes across global climate gradients. Nat Publ Group 5:889. CrossRefGoogle Scholar
  6. Ehleringer JR, Mooney HA (1978) Leaf hairs: effects on physiological activity and adaptive value to a desert shrub. Oecologia 37:183–200. CrossRefPubMedGoogle Scholar
  7. Ehleringer J, Bjorkman O, Mooney HA (1976) Leaf pubescence: effects on absorptance and photosynthesis in a desert shrub. Science 192:376–377. CrossRefPubMedGoogle Scholar
  8. Ehleringer JR, Phillips SL, Comstock JP (1992) Seasonal variation in the carbon isotopic composition of desert plants. Funct Ecol 6:396. CrossRefGoogle Scholar
  9. Evans JR, Loreto F (2000) Acquisition and diffusion of CO2 in higher plant leaves. Photosynthesis. Springer, Dordrecht, pp 321–351CrossRefGoogle Scholar
  10. Farquhar GD, O’Leary MHO, Berry JA (1982) On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust J Plant Physiol 9:121–137. CrossRefGoogle Scholar
  11. Flores O, Garnier E, Wright IJ et al (2014) An evolutionary perspective on leaf economics: phylogenetics of leaf mass per area in vascular plants. Ecol Evol 4:2799–2811. CrossRefPubMedPubMedCentralGoogle Scholar
  12. Forrestel EJ, Ackerly DD, Emery NC (2015) The joint evolution of traits and habitat: ontogenetic shifts in leaf morphology and wetland specialization in Lasthenia. New Phytol 208:949–959. CrossRefPubMedGoogle Scholar
  13. Friend AD, Woodward FI, Switsur VR (1989) Field measurements of photosynthesis, stomatal conductance, leaf nitrogen and δ 13 C along altitudinal gradients in Scotland. Funct Ecol 3:117. CrossRefGoogle Scholar
  14. Hobbie EA, Colpaert JV (2004) Nitrogen availability and mycorrhizal colonization influence water use efficiency and carbon isotope patterns in Pinus sylvestris. New Phytol 164:515–525. CrossRefGoogle Scholar
  15. Hobbie JE, Hobbie EA (2006) 15 N in symbiotic fungi and plants estimates in nitrogen and carbon flux rates in arctic tundra. Ecology 87:816–822.[816:NISFAP]2.0.CO;2 CrossRefPubMedGoogle Scholar
  16. Hobbie EA, Macko SA, Shugart HH (1999) Insights into nitrogen and carbon dynamics of ectomycorrhizal and saprotrophic fungi from isotopic evidence. Oecologia 118:353–360. CrossRefPubMedGoogle Scholar
  17. Hobbie EA, Macko SA, Williams M (2000) Correlations between foliar δ15 N and nitrogen concentrations may indicate plant–mycorrhizal interactions. Oecologia 122:273–283. CrossRefPubMedGoogle Scholar
  18. Hutchinson AH (1918) Limiting factors in relation to specific ranges of tolerance of forest trees. Bot Gaz 66:465–493. CrossRefGoogle Scholar
  19. Jara Arancio P, Arroyo MTK, Guerrero PC et al (2014) Phylogenetic perspectives on biome shifts in Leucocoryne (Alliaceae) in relation to climatic niche evolution in western South America. J Biogeogr 41:328–338. CrossRefGoogle Scholar
  20. Kahmen A, Wanek W, Buchmann N (2008) Foliar δ15 N values characterize soil N cycling and reflect nitrate or ammonium preference of plants along a temperate grassland gradient. Oecologia 156:861–870. CrossRefPubMedPubMedCentralGoogle Scholar
  21. Kerkhoff AJ, Fagan WF, Elser JJ, Enquist BJ (2006) Phylogenetic and growth form variation in the scaling of nitrogen and phosphorus in the seed plants. Am Nat. PubMedCrossRefGoogle Scholar
  22. Klak C, Reeves G, Hedderson T (2004) Unmatched tempo of evolution in Southern African semi-desert ice plants. Nature 427:63–65. CrossRefPubMedGoogle Scholar
  23. Kolb KJ, Evans RD (2002) Implications of leaf nitrogen recycling on the nitrogen isotope composition of deciduous plant tissues. New Phytol 156:57–64. CrossRefGoogle Scholar
  24. Kraft NJB, Cornwell WK, Webb CO, Ackerly DD (2007) Trait evolution, community assembly, and the phylogenetic structure of ecological communities. Am Nat 170:271–283. CrossRefPubMedGoogle Scholar
  25. Kron KA, Judd WS, Stevens PF, Crayn DM (2002) Phylogenetic classification of Ericaceae: molecular and morphological evidence. Botanical 68:335–423CrossRefGoogle Scholar
  26. Lallemand F, Gaudeul M, Lambourdière J et al (2016) The elusive predisposition to mycoheterotrophy in Ericaceae. New Phytol 212:314–319. CrossRefPubMedGoogle Scholar
  27. Liu H, Xu Q, He P et al (2015) Strong phylogenetic signals and phylogenetic niche conservatism in ecophysiological traits across divergent lineages of Magnoliaceae. Nat Publ Group. CrossRefGoogle Scholar
  28. Meinzer F, Goldstein G (1985) Some consequences of leaf pubescence in the andean giant rosette plant Espeletia timotensis. Ecology 66:512–520. CrossRefGoogle Scholar
  29. Münkemüller T, Lavergne S, Bzeznik B et al (2012) How to measure and test phylogenetic signal. Methods Ecol Evol 3:743–756. CrossRefGoogle Scholar
  30. Oren R, Sperry JS, Katul GG et al (1999) Survey and synthesis of intra- and interspecific variation in stomatal sensitivity to vapour pressure deficit. Plant Cell Environ 22:1515–1526. CrossRefGoogle Scholar
  31. Osnas JLD, Lichstein JW, Reich PB, Pacala SW (2013) Global leaf trait relationships: mass, area, and the leaf economics spectrum. Science 340:741–744. CrossRefPubMedGoogle Scholar
  32. Pagel M (1999) Inferring the historical patterns of biological evolution. Nature 401:877–884. CrossRefPubMedGoogle Scholar
  33. Revell LJ (2012) phytools: an R package for phylogenetic comparative biology (and other things). Methods Ecol Evol 3:217–223. CrossRefGoogle Scholar
  34. R Core Team (2016) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
  35. Savage JA, Cavender Bares J (2012) Habitat specialization and the role of trait lability in structuring diverse willow (genus Salix) communitiesGoogle Scholar
  36. Sexton JP, McIntyre PJ, Angert AL, Rice KJ (2009) Evolution and ecology of species range limits. Annu Rev Ecol Evol Syst 40:415–436. CrossRefGoogle Scholar
  37. Smith SE, Read DJ (1997) Mycorrhizal symbiosis. Mycorrhizal symbiosisGoogle Scholar
  38. Smith SE, Read D (2008a) Mycorrhizas in achlorophyllous plants (mycoheterotrophs). Mycorrhizal symbiosis, 3rd edn. Academic Press, London, p 458CrossRefGoogle Scholar
  39. Smith SE, Read D (2008b) Ericoid mycorrhizas. Mycorrhizal symbiosis, 3rd edn. Academic Press, London, pp 389–418CrossRefGoogle Scholar
  40. Swofford DL (2002) PAUP*. Phylogenetic analysis using parsimony (*and other methods). Sinauer Associates, SunderlandGoogle Scholar
  41. Tedersoo L, Pellet P, Kõljalg U, Selosse M-A (2006) Parallel evolutionary paths to mycoheterotrophy in understorey Ericaceae and Orchidaceae: ecological evidence for mixotrophy in Pyroleae. Oecologia 151:206–217. CrossRefPubMedGoogle Scholar
  42. Tucker GC (2009) Ericaceae. In: Flora of North America North of Mexico. New York and Oxford, pp 364–504Google Scholar
  43. Wiens JJ, Ackerly DD, Allen AP et al (2010) Niche conservatism as an emerging principle in ecology and conservation biology. Ecol Lett 13:1310–1324. CrossRefPubMedGoogle Scholar
  44. Wright IJ, Reich PB, Westoby M et al (2004) The worldwide leaf economics spectrum. Nature 428:821–827. CrossRefPubMedGoogle Scholar
  45. Yang J, Zhang G, Ci X et al (2014) Functional and phylogenetic assembly in a Chinese tropical tree community across size classes, spatial scales and habitats. Funct Ecol 28:520–529. CrossRefGoogle Scholar
  46. Yoneyama T, Omata T, Nakata S, Yazaki J (1991) Fractionation of nitrogen isotopes during the uptake and assimilation of ammonia by plants. Plant Cell Physiol 32:1211–1217. CrossRefGoogle Scholar
  47. Zachos JC, Bohaty SM, John CM, McCarren H, Kelly DC, and Nielsen T (2007) The Palaeocene–Eocene carbon isotope excursion: constraints from individual shell planktonic foraminifer records. Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences 365. The Royal Society: 1829–1842.
  48. Zimmer K, Hynson NA, Gebauer G et al (2007) Wide geographical and ecological distribution of nitrogen and carbon gains from fungi in pyroloids and monotropoids (Ericaceae) and in orchids. New Phytol 175:166–175. CrossRefPubMedGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Ecology and Evolutionary BiologyCornell UniversityIthacaUSA

Personalised recommendations