Abstract

In this experiment, phenotypic variation in 3 species of North American Solidago species (S. canadensis, S. rugosa, and S. gigantea) was assessed. Individuals were randomly sampled from roadside populations in Centennial Woods, Burlington, VT. Plant height and diameter were measured in order to determine (1) whether height and diameter varied significantly between the 3 species and (2) whether height and diameter were correlated within the species. Height and diameter were found to be significantly correlated with species (p = 4.43e-08 for height and p = 2.76e-10 for diameter). Height and diameter were found to be highly correlated, particularly in S. gigantea (r = 0.60 for S. canadensis, r= 0.57 for S. rugosa, and r = 0.91 for S. gigantea).

Introduction

A species is generally defined as a distinct, reproductively isolated group of organisms which shares a set of phenotypic and genotypic characteristics. However, in nature, intermediate cases can exist, particularly among species that can hybridize. Species arise by the gradual differentiation of populations by evolution.

Evolution is the change in allele frequency over the course of generations. Evolution can occur as a result of a number of processes, including natural selection, sexual selection, and genetic drift. In order for natural selection and evolution to occur, heritable phenotypic variation with consequences to individual fitness must exist within the population. Phenotypic variation is variation in the observable characteristics of an organism, including anatomy, physiology, and, in animals, behavior. The adaptiveness of a given phenotype is dependent upon the organism’s environment. Niche differentiation, the process of adaptation to more and more specialized niches in the environment, can lead to speciation.

Phenotype is produced by a complex interaction between genetics and environmental factors, such as nutrient availability, developmental conditions, and climate. Consequently, changes in phenotype are not always due to evolution; in some species, individuals are able to alter their phenotype in response to immediate conditions such as predation or competition. In these cases, evolution has not occurred because allele frequency within the population has not been changed—rather, a pre-existing genetic potential has been realized.

Solidago is a genus in the Aster family, encompassing 100 or more species of primarily open-habitat dwelling, herbaceous perennial plants, 45 of which occur in the Northeast. Members of the genus frequently hybridize and are notoriously difficult to differentiate. Previous studies have shown a high degree of infraspecific phenotypic variation in Solidago species, often as a function of environmental parameters such as elevation (Hirano et. al, 2017). Weber and Schmid (2008) showed that S. gigantea, a relatively recently introduced species in Europe, varies phenotypically across its introduced range. It can be difficult, however, to tease apart what variation can be attributed to plasticity versus evolution. Environmental selection pressures can lead to long-term evolutionary changes and the formation of different ecotypes—phenotypic variants occupying particular niches—within a population. Both processes-phenotypic plasticity and evolution-can contribute to phenotypic variation within and between populations. Both processes-phenotypic plasticity and evolution-can contribute to phenotypic variation within and between populations.

In this study, variation in plant height and diameter was compared between three Solidago species (S. canadensis, S. rugosa, and S. gigantea) found in Centennial Woods in Burlington, Vermont in order to assess phenotypic variation both within and between species. Specifically, we ask (1) do height and diameter vary between the three species? and (2) are height and stem diameter correlated? Previous studies have shown phenotypic variation in Solidago over a broad scale. However, the range of phenotypic variation in Solidago over a small scale (i.e. within small, adjacent populations) has not been previously assessed, to our knowledge.

We hypothesize that there will be consistent phenotypic variation between the three species, and that plant height and stem diameter will be positively correlated. Due to niche differentiation, species height should vary as a function of the different niches they occupy. Furthermore, in order to support greater height, thicker stems are necessary, both to keep the plant upright and to accommodate additional vascular tissue for effective water and nutrient transport.

Methods

Data Collection Three Solidago species—S. canadensis (SC), S. rugosa (SR), and S. gigantea (SG) were randomly selected from roadside populations in Centennial Woods, Burlington, VT. Stem height (in cm) was measured using a meterstick, and stem diameter (in mm) was recorded 25cm from the base of the stem using calipers. 10 of each species were surveyed for a total of 30 plants. Data from other groups were compiled for use in analysis for a total sample size of 180.

Data Analysis R was used to analyze the data. Histograms of height and diameter were created to show trait distribution across the species. Scatterplots comparing height to diameter were created for each species to determine whether the two variables were correlated. A correlation coefficient was calculated for each. Two boxplots were created—one depicting variation in height and one depicting variation in diameter. Finally, an ANOVA (analysis of variance test) was run to determine whether height and diameter varied significantly between the three species.

Results

Histograms showing the variation in height and diameter of the 3 species were created. Mean heights and diameters varied between the species: SC was taller, with a mean of 142.9cm, while SR and SG were shorter, with similar mean heights of 112.01cm and 112.5cm. Mean diameter for SC was 0.69mm, and, again, SR and SG had similar, smaller means of 0.50mm and 0.51mm. Height variation values were 460.23, 786.97, and 1770.74, and diameter variation values were 0.022, 0.023, and 0.037 for SC, SR, and SG, respectively.

Figure 1 Variation in height (cm) of S. canadensis (SC), S. rugosa (SR), and S. gigantea (SG) from 180 individuals.

Figure 2 Variation in stem diameter (mm) of S. canadensis (SC), S. rugosa (SR), and S. gigantea (SG) from 180 individuals.

## [1] 0.6054339
## [1] 0.575363
## [1] 0.9124772

## Warning in par(opar): graphical parameter "cin" cannot be set
## Warning in par(opar): graphical parameter "cra" cannot be set
## Warning in par(opar): graphical parameter "csi" cannot be set
## Warning in par(opar): graphical parameter "cxy" cannot be set
## Warning in par(opar): graphical parameter "din" cannot be set
## Warning in par(opar): graphical parameter "page" cannot be set

Scatterplots of height (cm) vs. diameter (mm) showed highest correlation between the two variables in SG (r=0.91) and lowest correlation in SR (r=0.57). For SC, a correlation coefficient of 0.60 was found.

Figure 3 Scatterplots showing correlation between height (cm) and diameter (mm) in S. canadensis (SC), S. rugosa (SR), and S. gigantea (SG) from 180 individuals. Pearson’s correlation coefficient (r) is 0.60 for SC, 0.57 for SR, and 0.91 for SG.

Figure 4 Height variation for S. canadensis (SC), S. gigantea (SG), and S. rugosa (SR).

Figure 5 Stem diameter variation (mm) for S. canadensis (SC), S. gigantea (SG), and S. rugosa (SR).

An analysis of variance test was run to determine whether height and diameter vared significantly between the three species. The p-value for height was 4.43e-08, and the p-value for diameter was 2.76e-10.

Discussion

In this study, our objectives were to establish whether the three Solidago species varied significantly in height and diameter and to determine whether or not height and diameter were correlated. A significant degree of phenotypic variation was found among the populations sampled (p=4.43e-08 for height and p=2.76e-10 for diameter), allowing for the rejection of the null hypothesis (that height and diameter do not vary between the species). Scatterplots of height vs. diameter for each species yielded correlation coefficients of 0.60 for SC, 0.57 for SR, and 0.91 for SG. These results show that diameter and height are positively correlated and thus support our hypothesis. However, in both SC and SR, there were several extreme outliers that may have contributed to their relatively lower correlation coefficients. Correlation between height and diameter was both much higher and clustered around a height of ~125cm in SG. This suggests either a lack of phenotypic plasticity in that species or selection for a particular ratio between height and diameter.

A previous study by Abrahamson and colleagues (2005) found that differences between sympatrically occurring North American goldenrold species helps to reduce competition for resources. In particular, they found that S. rugosa preferred acidic soils containing little clay, while S. gigantea was most abundant on neutral soils and preferred stable soil moisture conditions. The plants also varied in life-history traits. S. gigantea, of the plants studied, allocated the least resources to roots and the most resources to leaves. S. rugosa, by comparison, invested more heavily in roots and reproduction. If S. gigantea invests less in roots relative to other species of goldenrod, this may account for the high correlation (r=0.91) seen between height and diameter for that species. Without a solid anchor in the ground, structural stability above ground may matter more. These results, taken with previous findings, offer tentative support to the hypothesis that a high, positive correlation between height and diameter lends the plant greater stability.

There are several sources of error in this study. First, the sampling was performed by separate groups in the same vicinity, so some plants may have been measured multiple times. This would have led to a reduction in the dataset’s variance. Measurement errors or incorrect species identifications could also have contributed to error. Additionally, we did not control for plant maturity, which certainly would have contributed to the dataset’s variance. Finally, there may have been bias in our choice of plants towards extreme or non-extreme phenotypes.

Future studies could further elucidate phenotypic variation in Solidago by integrating phenotypic data with ecological data. For example, competition, water and light availability, or soil qualities could all mediate phenotypic plasticity or impact plant growth and thus contribute to variation. Genetic data could then verify whether that variation is due to niche differentiation or phenotypic plasticity. Additionally, these results, taken with a previous study by Abrahamson et. al (2005), hint at a relationship between the extent of a plant’s root system and correlation between height and diameter, potentially as a way of adding structural stability. A follow-up study could compare the correlation coefficient for height vs. diameter between plants that allocate varying amounts of resources to their root systems.

Literature Cited

Abrahamson, W. G., K. B. Dobley, H. R. Houseknecht, and C. A. Pecone. 2005. Ecological divergence among five co-occuring species of old-field goldenrods. Plant Ecology 177: 43-56.

Hirano M., S. Sakaguchi, and K. Takahashi. 2017. Phenotypic differentiation of the Solidago virgaurea complex along an elevational gradient. Ecology and Evolution 7: 6949– 6962.

Weber, E. 2008. Phenotypic variation of the introduced perennial Solidago gigantea in Europe. Nordic Journal of Botany 17: 631-638.