Am. J. Bot. Visit Plant Cell Online
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stenström, A.
Right arrow Articles by Augner, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Stenström, A.
Right arrow Articles by Augner, M.
Agricola
Right arrow Articles by Stenström, A.
Right arrow Articles by Augner, M.
(American Journal of Botany. 2002;89:1410-1421.)
© 2002 Botanical Society of America, Inc.

Ecology

Genetic and environmental effects on morphology in clonal sedges in the Eurasian Arctic1

Anna Stenström2, Ingibjörg S. Jónsdóttir2,3,5 and Magnus Augner4

2Botanical Institute, Göteborg University, Box 461, SE-405 30 Göteborg, Sweden; 3The University Courses on Svalbard, UNIS, P.O. Box 156, N-9171 Longyearbyen, Norway; 4Swedish Polar Research Secretariat, P.O. Box 50005, SE-104 05 Stockholm, Sweden

Received for publication January 3, 2002. Accepted for publication April 19, 2002.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
We studied the variation in morphological characters of importance for resource acquisition and storage in 21 populations of four clonal sedge taxa in arctic Eurasia, Carex bigelowii, C. ensifolia subsp. arctisibirica, C. lugens, and C. stans, and the response to transplantation to a common garden in Tromsø, Norway. The morphology of C. stans was distinct from the other three taxa, all of which belong to the C. bigelowii species complex. However, differences among populations within taxa were even greater than differences among taxa, and environmental variables explained 40–50% of the among-population variation in the morphological characters. Stomatal size decreased with temperature while stomatal density increased. Shoot height and leaf width were smaller at peak lemming population phase, while rhizome length was shorter at higher longitudes. Transplantation to a common garden affected stomatal density in all taxa, stomatal size and shoot height only in some taxa, while leaf width was not affected. We found a weak, but highly significant correlation between geographical, morphological, and genetic distances. We concluded that although genotypic differentiation in arctic rhizomatous Carex species is reflected in their morphology, they are also capable of plastic morphological responses to the environment and that these responses are specific for each taxon.

Key Words: Arctic • Carex • climate • genetic distance • grazing • leaf width • rhizome length • stomata


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
An important consequence of the sedentary lifestyle of plants is that they cannot escape from the environment in which they grow or from any changes in this environment. To cope with this, many plants are able to alter one or more morphological characters in response to both abiotic (e.g., climate and weather) and biotic (e.g., grazing and competition) factors of the environment with a potential effect on resource acquisition. For example, leaf size and leaf area of many alpine plants change with altitude (Meinzer, Goldstein, and Rundel, 1985 ; Körner et al., 1989 ), and some arctic plants may produce more or larger leaves during warmer summers than during colder ones (Havström et al., 1995 ; Stenström and Jónsdóttir, 1997 ). Grazing and competition can also change plant morphology to a great extent (Lubchenco and Cubit, 1980 ; Turkington, 1983 ; Pollard, 1986 ). Furthermore, these environmental factors can interact. Douglas (1981) showed that competition reduced the size of Mimulus primuloides plants at low altitudes, while low temperatures reduced size at higher altitudes and consequently the largest plants were found at intermediate altitudes.

Morphological variation within and among populations can either be due to genotypic differentiation or to phenotypic plasticity. Genotypic differentiation among plant populations is common (for reviews, see Heslop-Harrison, 1964 ; Langlet, 1971 ) and has been shown to occur on spatial scales as small as a couple of meters or even decimeters (Jain and Bradshaw, 1966 ; Antonovics, 1971 ; Shaver, Chapin, and Billings, 1979 ; McGraw and Antonovics, 1983 ). Although genotypic differentiation is widely reported, large morphological differences among populations in different environments may also be due to phenotypic plasticity (Heathcote, Davies, and Etherington, 1987 ; Williams and Black, 1993 ). Plasticity should be selected for in heterogenous environments, rather than specialized phenotypes, given that there is no added cost of plasticity (Sultan, 1992 ). In stressful environments of low productivity, plants are usually slow growing and thus likely to show a physiologically rather than morphologically plastic response to heterogeneous environment (Grime, 1979 ; Hutchings and de Kroon, 1994 ; Jónsdóttir and Watson, 1997 ). Although arctic environments can be characterized as stressful, i.e., with low temperatures, short growing seasons, and low nutrient availability, arctic plant populations often express considerable morphological plasticity (Havström et al., 1995 ; Molau, 1997 ; Stenström and Jónsdóttir, 1997 ; Welker et al., 1997 ; Stenström, 2000 ).

The degree of genotypic differentiation generally increases with time of isolation, and, in widely distributed species, we would therefore expect to find genotypic differentiation among populations. Arctic plants in general have large distributional areas (Bay, 1992 ), and many arctic populations studied are indeed genotypically different (Shaver, Chapin, and Billings, 1979 ; Chapin and Chapin, 1981 ; McGraw, 1987 ; Fetcher and Shaver, 1990 ). During the Pleistocene, glaciers periodically advanced over large areas in the Eurasian Arctic, forcing plants to retreat, while other areas, like Eastern Siberia, were seldom glaciated and could thus function as arctic refugia (Forman et al., 1999 ; Svendsen et al., 1999 ). In a previous study of clonal Carex populations we showed that the difference among areas in glaciation history is related to the degree of genetic diversity based on isozyme polymorphism (Stenström et al., 2001 ). Most of the genetic diversity was found within populations in these widely distributed taxa and the low among-population diversity indicated a low degree of population differentiation. However, while isozyme polymorphism is assumed to be selectively neutral, we would expect adaptive characters to show somewhat higher degree of population differentiation.

Knowledge of how ecologically important morphological characters vary within the distributional range of plant species, as well as the underlying control mechanisms for such variation, is essential to understand how the plants may respond to environmental change. Rhizome length, shoot height, leaf width, and the density and size of stomata are all important characters for resource acquisition and storage of rhizomatous graminoids. In this paper we study the variation in these five characters among 21 populations of arctic rhizomatous sedges, representing four closely related Carex taxa, and how this variation relates to both abiotic and biotic environmental factors. We further studied the presence of genetic variation in these characters by transplanting plants to a common garden in the Botanical Garden in Tromsø, Norway, and by comparing the distance in morphology among populations with their geographic distances and genetic distances, which were obtained in a previous study (Stenström et al., 2001 ). The populations sampled are widely distributed over the Eurasian Arctic, i.e., over more than 160° in longitude and 9° in latitude covering a substantial proportion of the environmental variation within the distributional limits of the represented taxa. Most of the populations were sampled during the Swedish-Russian Tundra Ecology Expedition 1994 (Hedberg, Hjort, and Sonesson, 1999 ).

All four Carex taxa studied here have rather wide geographical distributions. We, therefore, expected the populations to be genotypically different and to show relatively little morphological response to transplantation. We also expected a strong correlation between the morphological distances among populations and their geographical and genetic distances. The goals of our study were (1) to assess how much of the variation in morphology among populations could be explained by abiotic (climate and weather) and biotic (grazing) factors; (2) to assess whether this variation was due to genotypic differentiation or phenotypic plasticity; and (3) to test whether morphological, geographical, and genetic distances among populations were correlated.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
The taxa
The taxa studied are Carex bigelowii Torr. ex Schwein, Carex ensifolia Krecz. subsp. arctisibirica Yurtsev, Carex lugens Holm, and Carex stans Drej, all within the Carex section Phacocystis Dumort. The first three taxa are members of the C. bigelowii complex, which is a circumpolar species complex. Taxonomically they have either been treated as components of C. bigelowii sensu lato (s.l.) (Murray, 1994 ), as different species (Egorova et al., 1966 ), or with C. ensifolia as a subspecies to C. bigelowii (Fedorov, 1976 ). Carex stans has been treated either at the species level (Egorova et al., 1966 ), subspecies level (to C. aquatilis, Polunin, 1959 ), or as a hybrid between C. aquatilis and C. bigelowii (Nilsson, 1986 ). Usually the name C. stans has been used in the High Arctic and C. aquatilis in the Low Arctic (Murray, 1994 ). The four taxa will here be referred to as C. bigelowii, C. ensifolia, C. lugens, and C. stans, respectively.

Carex bigelowii, C. ensifolia, and C. lugens have partly overlapping geographical distributions, while C. stans is circumpolar (Fig. 1). All four taxa reproduce both sexually and vegetatively, with large variation in genet size (Jonsson, 1995 ; Jonsson, Jónsdóttir, and Cronberg, 1996 ; Jónsdóttir et al., 2000 ; Stenström et al., 2001 ). The individual ramets in C. bigelowii may flower after 2–4 yr (Carlsson et al., 1990 ), and 27 or more ramet generations may stay interconnected (Kershaw, 1962 ; I. S. Jónsdóttir, unpublished data). A ramet dies after flowering as the apical meristem is then used up (Carlsson et al., 1990 ). The flowering frequency is highly variable both among sites (Stenström, 1999 ; I. S. Jónsdóttir et al., unpublished data) and among years (Carlsson and Callaghan, 1994 ; Stenström, 1999 ). All taxa are rhizomatous, with large variations in rhizome length both within and among taxa. Shorter rhizomes and the absence of spreading rhizomes have been used to separate C. lugens from C. ensifolia (Egorova et al., 1966) and from C. bigelowii (including C. ensifolia; Polunin, 1959 ). Carex stans usually has two different types of ramets, with either long spreading rhizomes or shorter rhizomes (Shaver, Chapin, and Billings, 1979 ; Jónsdóttir et al., 2000 ). Ramet differentiation has also been reported for some populations of C. bigelowii (Carlsson and Callaghan, 1990 ), while there was no such differentiation in C. ensifolia (Jónsdóttir et al., 2000 ). Carex aquatilis (including C. stans) is reported to be amphistomatous, while C. bigelowii is supposed to be hypostomatous (Standley, 1986 ).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 1. Geographic range of Carex stans, C. bigelowii, C. ensifolia subsp. arctisibirica, and C. lugens in Eurasia and the studied sites. Plant distributions are drawn after Egorova et al. (1966) and Mossberg, Stenberg, and Ericsson (1992) . For site names and taxa studied at respective site, see Table 1

 
Study sites and sampling
The plant material was collected between 6 July and 30 August 1994 from 15 sites along the northern coast of Russia during the Swedish-Russian Tundra Ecology Expedition 1994 (Hedberg, Hjort, and Sonesson, 1999 ; Jónsdóttir, Virtanen, and Kärnefelt, 1999 ; Fig. 1). The sites were visited in the following order: 14, 13:1, 15, 16, 17, 13:2, 12, 10, 9, 8, 6, 5, 4, 3, 2 (for the site names, see Table 1). To minimize within-site variation we chose an area with mesic, levelled ground or gently south-facing slope at all sites with one or more of the target species present. Since C. stans is mainly confined to wet habitats, this habitat restriction reduced the number of sites sampled for this taxon. At two sites two taxa were sampled: at site 12, Yana Delta, C. ensifolia and C. stans were sampled, and at site 16, Ayon Island, C. lugens and C. stans were sampled. When a homogenous stand of the target taxon was found, a 20-m transect was set out and ten vegetative ramets were chosen at random along the transect. These ramets and at least three previous ramet generations still attached to them were excavated. At site 16 and in the C. stans population at site 12, only six clone fragments were sampled for logistic reasons. The clone fragments were pressed dry. Along the same transect, 20 plants, also chosen at random, were dug up alive and transferred to Tromsø Botanical Garden, Norway (69°34.4' N, 19°10' E). Tromsø has higher precipitation (1031 mm) and a warmer climate than all the original sites, both when measured as yearly mean temperature (+2.5°C) and July mean temperature (+11.8°C). Because the climatic differences between Tromsø and the various original sites were not equal, we cannot expect an equal amount of plasticity to be expressed in all populations in response to transplantation. In July 1997, two additional sites were sampled to extend the geographical gradient further west. Carex bigelowii and C. stans were collected at Latnjajaure, north Lapland, Sweden, and C. bigelowii was collected at two sites (100 and 700 m above sea level [asl]) on Mount Aikuaivenchor, Kola peninsula, Russia (Fig. 1, Table 1). The clone fragments at these additional sites were sampled in the same way as in 1994.


View this table:
[in this window]
[in a new window]
  		Table 1. Environmental variables for the sampled populations. The environmental variables for the populations of the Carex bigelowii-complex were used as X variables in the partial least squares (PLS) analysis

 
Voucher herbarium specimens were collected in all populations and are placed in Göteborg Herbarium, Sweden (GB) and University of Alaska, Fairbanks, USA (ALA; Holmgren, Holmgren, and Barnett, 1990 ).

Measurements and analyses
Shoot height and rhizome length were measured on all vegetative ramets of the clone fragments. Shoot height was measured from the basal meristem to the tip of the longest leaf, while rhizome length was measured from the base of one shoot to the base of the next shoot in the clone fragment. In C. stans the frequency of short and spreading rhizomes was recorded, with rhizomes longer than 2 cm counted as spreading rhizomes. However, as there was no clear differentiation in many of the populations, the population means for all rhizomes were used in the analyses. For measurements of leaf characters, the two vegetative ramets with the tallest shoots were chosen from each clone fragment. From these two ramets a 5-mm piece of their longest leaf was cut, one-third up the length of the leaf. The leaf pieces were boiled in water for 1 min, put in FPA-50 (50 parts formalin, 50 parts propionic acid, and 900 parts 50% ethanol) for 24 h and then transferred to 70% alcohol until analysis. At the time of analysis the leaf pieces were put in Herr's clearing fluid (Herr, 1971 ) for 24 h and the number of adaxial stomata was counted in two adjacent fields of view (each 0.15 mm2) using a Wild light microscope. The lengths of ten adaxial stomata were measured in each field of view, and the widths of the leaf pieces were measured. The mean value for each clone fragment was used in all statistical analyses, exept when testing frequencies of short and spreading rhizomes.

By August 1997, the plants transferred to Tromsø Botanical Garden had been growing there for 3 yr. We assumed that all differences caused by unequal sizes of transplanted clone fragments and other effects from the original site had been eliminated by that time. Not all the transplants survived, which left nine of the populations with six or more surviving plants. This we used as a criterion for including a population in the transplant sampling. Aboveground parts (including all of the shoots) were sampled 1 August 1997 and analyzed in the same way as the original populations, but, to avoid killing whole genets, belowground parts were not sampled.

Data analysis
An overview of the morphological variation among populations before and after transplantation was obtained by using principal components analysis (PCA). Differences in morphology among taxa and populations nested within taxa were tested for by Multivariate Analysis of Variance (MANOVA; Johnson, 1998 ) corrected for unequal sample sizes with Tukey HSD (honestly significant difference) post hoc test for unequal sample sizes. To avoid false positives showing up by chance, the significance of all measured morphological variables were tested together first with an overall MANOVA, and if this proved significant, the next step was a performance of ANOVA of each morphological variable. Differences among populations in the frequency of short and spreading rhizomes were tested with a {chi}2 test. Changes in morphology after transplantation were determined with a factorial MANOVA with sampling locality and taxa as factors, likewise corrected for unequal sample sizes. To obtain normal distribution and homoscedasticity, shoot height and rhizome length were log transformed and stomata density was arcsine transformed.

It was not possible to use multiple regression when looking for relationships between morphology and environmental factors due to high colinearity among the environmental variables. Instead, projection to latent structures by means of partial least squares analysis (PLS) was used. This multivariate statistical method finds a linear relationship between the matrix of dependent variables (Y values) and the matrix of predictor variables (X values) and can handle colinear and incomplete data sets. The X and Y matrices can be seen as points in two spaces and PLS modelling consists of simultaneous projections of both the X and Y spaces on low dimensional hyper planes (Eriksson et al., 1999 ). Shoot height, rhizome length, and leaf width were used as dependent variables in three separate PLS analyses because they were expected to respond differently to different environmental variables. There was a strong negative correlation between stomata density and stomata size (correlation coefficient = –0.76, P < 0.001), and these two variables were therefore used as dependent variables in one PLS analysis. Population means of the morphological variables were used as dependent variables. The environmental factors used are listed in Table 1, where lemming index is the relative lemming population size (Angerbjörn, Tannerfeldt, and Erlinge, 1999 ) and genetic variation was measured as genet-level gene diversity (Stenström et al., 2001 ). Deviation from mean July temperature the year before sampling has been shown to be important for tillering (Carlsson and Callaghan, 1994 ; I. S. Jónsdóttir et al., unpublished data). Precipitation, lemming index, and ramet density were log transformed and genetic variation was arcsine transformed. In Eurasia, lemming populations show large fluctuations in population size that are cyclic, and four population phases can be identified: increase, peak, decrease, and low (Erlinge et al., 1999 ). Lemming population phases were introduced into the PLS model as three "dummy variables" with 1 value for presence and 0 value for absence.

To test for correlations between morphological, geographical, and genetic distances among populations, Mantel tests were performed (cf. Mantel, 1967 ; Sokal, 1979 ; Douglas and Endler, 1982 ) using Spearman rank correlation coefficient. Morphological distances were obtained by calculating Euclidean distances on standardized values ([x – ]/SD) on all the five morphological variables from the plants collected in the field. Geographic distances were calculated by the "geod" program from the U.S. Geological Survey (http://www.indo.com/distance/). Genetic distances among populations were calculated as Roger's genetic distance (Roger, 1972 ) using data from Stenström et al. (2001) . Mantel tests were performed with 1000 permutations, and when geographic distances were used, they were log transformed. As the populations of C. stans were assumed to be both genetically and morphologically more distant to the populations of the C. bigelowii species complex, the Mantel tests were made both on all populations and including only populations of the C. bigelowii species complex. For the Mantel test between morphological and genetic distances only populations in common for this study and Stenström et al. (2001) could be used, which excluded populations B1, 10, and 13:1.

The statistical package Statistica 4.5 (StatSoft, Tulsa, Oklahoma, USA) was used for MANOVA and calculations of Euclidean distances, while PCA and PLS analyses were made by the program Simca P 8.0 (Umetri AB, Umeå, Sweden). The {chi}2 test was performed using Statview 5.01 (SAS Institute, Cary, North Carolina, USA). The Mantel test was made using the program Isolde on GenePop, available on the web (http://www.biomed.curtin.edu.au/genepop/genepop_op6.html).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Population differences
There were significant differences in morphology among populations within each taxon (overall MANOVA: C. bigelowii: Wilks' lambda10,44 = 0.102, P < 0.001; C. ensifolia: Wilks' lambda40,334 = 0.046, P < 0.001; C. lugens: Wilks' lambda10,38 = 0.102, P = 0.004; C. stans: Wilks' lambda25,158 = 0.082, P < 0.001; Table 2). In C. bigelowii, shoot height, rhizome length and the stomata measurements differed among the three populations of that taxon, whereas there were no significant differences in leaf width (Table 2). In C. ensifolia, all measured characters differed among the eight populations, but for leaf width it was only the population from Kanin Peninsula that had broader leaves than the other seven populations (Table 2). The three populations of C. lugens showed no significant differences in leaf width, stomata size, or stomata density. The population from northeast of Kolyma Delta, however, had longer rhizomes than the other two populations and the population from Wrangel Island had shorter shoots than the other populations (Table 2). In C. stans, there were differences among populations in all measured characters, but in shoot height, leaf width, and stomata size it was only the population from Faddeyevskiy Island that differed from the other five populations (Fig. 2, Table 2). The populations from Ayon Island and Latnjajaure had higher stomata densities compared to the other populations (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Morphological measurements of Eurasian Carex populations and values for each taxa (mean ± 1 SE). Population values with different letters are significantly different within a taxon and morphological variable. Mean values for taxa with different letters are significantly different within a morphological variable. Carex bigelowii: dfpopulation = 10, dferror = 44; C. ensifolia: dfpopulation = 40, dferror = 334; C. lugens: dfpopulation = 10, dferror = 38; C. stans: dfpopulation = 25, dferror = 158

 


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 2. Principal components analysis (PCA) plot made on five morphological measurements (rhizome length, shoot height, leaf width, stomata size, and stomata density) on Eurasian Carex populations sampled in the field. See Table 1 for explanation of populations numbers. Population means ±1 SE are plotted for clarity

 
Taxon differences
The overall MANOVA for morphology among taxa was significant (Wilks' lambda15,524 = 0.287, P < 0.001). There was no difference among taxa in shoot height, but all taxa differed in rhizome length with Carex stans having the longest rhizomes followed by C. bigelowii, C. ensifolia, and C. lugens in decreasing order (Table 2). Carex bigelowii and C. stans had broader leaves than C. ensifolia and C. lugens. The size of stomata did not differ among taxa. However, stomata density was significantly higher in C. bigelowii and lower in both C. lugens and C. stans than in C. ensifolia (Table 2). The first two axes of the PCA plot explained 35 and 30% of the variation, respectively, and had eigenvalues of 1.74 and 1.51 (Fig. 2). The PCA plot points to a large variation in morphology of C. bigelowii, C. ensifolia, and C. lugens (Fig. 2), and some factors other than taxa specific differences are clearly needed to explain the variation. However, the morphology of all the C. stans populations was rather similar, with one exception, i.e., the Faddeyevskiy Island population. This population had shorter shoots, broader leaves, and larger, but fewer stomata than the other C. stans populations (Table 2).

The degree of ramet differentiation into short and spreading ramets differed among populations of C. stans, ranging from a continuous distribution of rhizome lengths at the Latnjajaure population to a clearly bimodal distribution of rhizome length at the Faddeyevskiy population (Fig. 3). Using a limit of 2 cm to separate short from spreading rhizomes, the frequency of the two ramet types also differed among populations ({chi}2 = 12.83, df = 5, P = 0.025; Fig. 3). There was no clear ramet differentiation due to rhizome length in any of the other taxa.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 3. Distribution of rhizome lengths for the six populations of Carex stans. The populations show a decreasing degree of ramet differentiation from the uppermost to the lowest part of the diagram. Rhizomes <2 cm long were classified as short rhizomes and longer rhizomes as spreading rhizomes

 
Morphology and environmental factors
The morphology of Carex stans was distinct compared to the other taxa (Fig. 2), and only populations from the C. bigelowii complex were therefore included in the PLS analysis. The PLS analyses in this study were all significant, as the Q2 statistics was larger than the limit (limit = 0.097) in each of the analyses (Eriksson et al., 1999 ; Table 3A). The environmental variables used explained a rather large part of the among-population variation in the morphological variables (R2Y = 40–50%; Table 3A). Variables with "variable influence on the projection parameter" (VIP) larger than one are those most influential for the model (Eriksson et al., 1999 ). Shoot height was mainly influenced by the phases of the lemming population cycle and by latitude, but there was also a small influence of the deviation from mean July temperature and lemming index. Shoots were lowest during peak lemming phases, at high latitudes and high lemming indices, but relatively high during increased lemming phases and when July temperature was higher than normal (Table 3B). Rhizome length decreased with increasing lemming index and longitude and increased with increasing precipitation and increasing yearly mean temperature (Table 3B). The leaves were narrow at peak lemming phase, high lemming indices, and high longitudes while they were broader at low lemming phase, high precipitation, and temperature. For stomata size and density, the environmental variables temperature and precipitation were most important (Table 3B). Stomata density increased with temperature and precipitation, but stomata size showed the opposite response. Large lemming indices had a small negative influence on stomata density and thereby a small positive influence on stomata size (Table 3B).


View this table:
[in this window]
[in a new window]
  		Table 3. Results from the partial least squares (PLS) analysis on Carex bigelowii, C. ensifolia, and C. lugens with (A) R2 = the cumulative fraction of the variable(s) explained, eigenvalues, and Q2 = the cumulative predicted fraction (after cross-validation) of the variation in Y. (B) Variable influence on the projection parameter (VIP) and regression coefficients. The PLS analyses are significant when Q2 > limit, the limit being 0.097 in these analyses. Variables with VIP > 1 are most influential for the model (Eriksson et al., 1999)

 
Transplant experiment
Transplantation to Tromsø Botanical Garden changed the morphology of all taxa (Fig. 4, Table 4). The morphological range of the populations became smaller when growing in the common garden environment as compared to the range in their original habitats. This is reflected by the more compact clustering of the transplant populations in the PCA plot (Fig. 4). The first two axes of the PCA plot explained 38 and 31% of the variation and had eigenvalues 1.54 and 1.24, respectively. The overall MANOVA was significant for both the transplant effect (Wilks' lambda4,158 = 0.621, P < 0.001) and for taxa (Wilks' lambda8,316 = 0.436, P < 0.001). There was a trend for interaction between these factors (Wilks' lambda8,316 = 0.911, P = 0.063), i.e., the response of the measured variables to transplantation was not the same in all taxa. Carex ensifolia developed taller shoots after transplantation (P < 0.001), but there was no significant change in shoot height in either C. lugens (P = 0.161) or C. stans (P = 0.547). There was no significant change in leaf width in any taxa (P = 0.860–0.978). The stomata became more numerous in all taxa in Tromsø (P = 0.001 < 0.001), and they became smaller in C. ensifolia (P = 0.004) and C. lugens (P = 0.013), while the stomata size was not affected by transplantation in C. stans (P = 0.381).



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 4. Principal components analysis (PCA) plot made on four morphological measurements (shoot height, leaf width, stomata size, and stomata density) on populations sampled before and after transplantation to Tromsø Botanical Garden. Figure Abbreviations: E, Carex ensifolia; L, C. lugens; S, C. stans. See Table 1 for explanation of population numbers. Population means ± 1 SE are plotted for clarity

 

View this table:
[in this window]
[in a new window]
  		Table 4. Morphological measurements of the Carex transplants in Tromsø Botanical Garden (69°34.4' N, 19°10' E), after 3 yr of transplantation. Values for populations and taxa (mean ± 1 SE) are given for each variable. Population values with different letters are significantly different within each measured variable, across taxa. Taxa values with different letters are significantly different within each measured variable. Dfpopulation = 8, dferror = 68. For values before transplantation see Table 2

 
There was still a significant difference among the transplanted populations after 3 yr in a common environment (the over-all MANOVA was significant; Wilks' lambda32,241 = 0.242, P < 0.001), but this difference was much less than among the original populations (Wilks' lambda32,289 = 0.049, P < 0.001). Before transplantation, the C. stans populations and the C. ensifolia population from Kanin peninsula were taller than the other populations, while after transplantation the only significant differences found were the ones between the two C. stans populations and the C. ensifolia population from Northern Yamal (P = 0.023–0.038; Table 4). Before transplantation, it was only the C. ensifolia population from Kanin that differed in leaf width and this difference disappeared after transplantation (P = 0.533–0.999). There were four C. ensifolia populations that differed in stomata size before transplantation, but these differences disappeared, as well, in the common environment (P = 0.563–1.000; Table 4). All C. stans and C. lugens populations and the C. ensifolia population from northwest (NW) Taymyr differed in stomata density from all the other C. ensifolia populations. After transplantation the C. ensifolia population from NW Taymyr still differed from the Kolguyev population (P = 0.002) and almost so from two of the other C. ensifolia populations (P = 0.063–0.072; Table 4). There were small taxon differences after transplantation to the common garden. Carex stans had taller shoots than C. ensifolia and C. lugens (P = 0.027–0.053), while C. ensifolia had higher stomata density than C. stans (P = 0.005; Table 4).

Correlations between morphological, geographical, and genetic distances
There were positive correlations between the morphological, geographical, and genetic distances of populations from all taxa (Fig. 5). However, although highly significant, the correlations between geographic distance and the other two distances were weak and explained much smaller part of the variation than did the correlation between morphological and genetic distances (morphology–geography: R = 0.14, P = 0.031; morphology–genetics: R = 0.44, P < 0.001; geography–genetics: R = 0.24, P = 0.002; Fig. 5). These correlations were still evident when only including populations of the C. bigelowii complex (morphology–geography: R = 0.26, P = 0.004; morphology–genetics: R = 0.44, P = 0.011; geography–genetics: R = 0.14, P = 0.031).



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 5. Correlations between (a) geographic distances and morphological distances, (b) geographic distances and genetic distances, and (c) morphological distances and genetic distances among Eurasian Carex populations. P values from Mantel tests of correlation between distance matrices

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
This study shows that although genetic differentiation in arctic rhizomatous Carex species is reflected in their morphology, there are considerable plastic morphological responses to the environment, and these responses are specific for each taxon. To tell whether the plasticity of the individual characters is itself adaptive, or simply a change induced by environmental stress, would need further experimentation. The large variation among populations that we found here is congruent with other studies of Carex that have shown variation in morphology, both due to ecotypic differentiation (Shaver, Chapin, and Billings, 1979 ; Chapin and Chapin, 1981 ; Standley, 1986 ) and phenotypic responses to the environment (Clausen, Keck, and Hiesey, 1940 ; Kislyuk et al., 1983 ; Smythe and Hutchinson, 1989 ; Stenström and Jónsdóttir, 1997 ). There were differences among taxa in most of the measured variables, but the differences among populations within taxa were usually larger than any taxon differences, especially when comparing populations within the C. bigelowii complex. The correlation between morphological and genetic distances suggests a significant genetic effect. The genetic distances were based on isozyme polymorphism that is generally regarded as selectively neutral in contrast to the ecologically important characters measured here. Therefore, the weak correlation between the two distance measures, morphological and genetic, does not have to imply weak genotypic differentiation with respect to morphology. Furthermore, the environmental variables included in our analyses suggests also a strong environmental influence, which may be manifested genetically (Table 3A). However, the transplantation experiment indicated a large potential for plastic phenotypic responses to the environment. This high potential placticity suggest that plasticity in ecologically important characters may be adaptive also in the relatively stressful environments of the Arctic.

Comparison of character response to environmental factors that vary on different time scales may give some indication of how plastic they are. Factors that vary on a relatively short time scale (temporally variable factors), such as cyclic herbivore populations and weather, explained a large part of the variation in shoot height and leaf width. In contrast, factors that vary on a larger time scale or not at all (spatially variable factors), such as longitude and climate, explained most of the variation in rhizome length and stomata density and size, suggesting that rhizomes and stomata are less plastic than shoot height and leaf width. However, the transplant experiment revealed that stomata characters as well are plastic in some populations. Although the design of the transplant experiment did not allow us to quantify either the degree of morphological plasticity (i.e., the reaction norms) or the heritability of the character, it showed that transplants from all populations responded in one or more of the measured characters, further indicating morphological plasticity.

It is possible that the deviation of the C. stans population from Faddeyevskiy Island from all other populations is due to character fixation through genetic drift. This population was at the fringe of the distribution limits for C. stans and was only represented by a few isolated genets (Jónsdóttir et al., 2000 ). Unfortunately, the Faddeyevskiy population could not be included in the analyses of the transplant effects.

Shoot height
Photosynthetic area is related to shoot height, and plasticity in this character may improve competitive ability for light in closed stands. Although shoot height was mainly influenced by temporally variable factors, spatial factors also seem to have some effects. Shoot height was negatively correlated to latitude, which was the second most important factor in the PLS analysis. Many plants are known to be shorter at higher latitudes (Chapin and Chapin, 1981 ; Billings, 1987 ; Farmer, 1993 ; Farmer, O'Reilly, and Shaotang, 1993 ), due to lower temperatures and increased wind speed (Woodward, 1983 ; Fitter and Hay, 1987 ). Other factors may also be involved as shown in C. aquatilis subsp. aquatilis in which shoots were found to be shorter at higher latitudes, but latitude was correlated to both lower temperatures and lower phosphate availability (Chapin, 1981 ). The lower plant stature of high latitudes or altitudes has been found to be genetically determined in many tundra plants (Turesson, 1922 ; Clausen, Keck, and Hiesey, 1940 ; Callaghan, 1974 ), including some populations of Carex aquatilis (Chapin and Chapin, 1981 ). However, these C. aquatilis populations were also strongly affected by the environment. In the present study the influence of temporally variable factors also provides strong indications of plastic responses to the environment. Accordingly, shoot height changed (increased) in response to transplantation to a common environment in Tromsø. Interestingly, in the C. bigelowii complex, pretransplant differences among populations in this character disappeared. This is the same result as found in different studies of boreal and temperate graminoids, where the variation in shoot height was mainly due to plasticity, and none or only a small part of the variation was due to genetic differentiation (Wu and Jain, 1978 ; Rapson and Wilson, 1988 ; Smythe and Hutchinson, 1989 ).

The phenological date of sampling seems not to have influenced the shoot height, probably because the shoots reached their full length early in the sampling period. Carex stans did not have significantly taller shoots than the other taxa despite a 50% larger mean. This was probably a combined effect of high within-taxon variation and low statistical power due to uneven sample sizes. However, after 3 yr of transplantation to a common garden, C. stans had taller shoots than C. ensifolia and C. lugens.

Shoot height appears to be responsive to temperature. The weather variable, deviation in temperature from the mean July temperature, was positively correlated with shoot height. This agrees with previous results in C. bigelowii from Latnjajaure (Stenström and Jónsdóttir, 1997 ) and C. lugens from Wrangel Island (Kislyuk et al., 1983 ), where both species increased their leaf length when exposed to higher temperatures. Shoot height also seems to respond to other temporally variable factors, such as grazing. Shoots were shortest at peak lemming population phase and were negatively correlated to lemming index. Carex bigelowii is extensively grazed by some species of lemmings (Moen, 1990 ; Moen, Lundberg, and Oksanen, 1993 ; Virtanen, Henttonen, and Laine, 1997 ), reindeer (Warenberg, 1982 ), and sheep (Thorsteinsson, 1980 ). It showed similar responses to sheep grazing on Iceland, where shoot size was smaller in grazed areas compared to ungrazed areas (Jónsdóttir, 1991 ). Tolvanen and Henry (2000) showed that leaves of C. stans, as well as other sedges, were significantly shorter in vegetation grazed by muskoxen compared to sedges of ungrazed vegetation.

Rhizome length
The lateral movement of clonal plants by rhizomes may improve their nutrient acquisition by increasing the volume of soils that can be exploited (Hutchings and deKroon, 1994 ; Jónsdóttir, Callaghan, and Headley, 1996 ). Rhizome length showed the strongest relationship with spatial variables, i.e., longitude and climate. Rhizomes became shorter with increasing longitude and with variables strongly correlated with longitude (i.e., precipitation and yearly mean temperature). This may partly reflect the fact that C. lugens, with its shorter rhizomes, is better represented at higher longitudes, i.e., in the eastern part of the sampled area. However, a comparable pattern was found among three populations of C. bigelowii at a single site that differed in soil humidity where the wettest population had longest rhizomes (Kibe and Masuzawa, 1994 ). We also found a small negative effect of lemming index on rhizome length. The same factors (latitude and lemming index) were found to have significant but positive effect on ramet density of the same populations (I. S. Jónsdóttir et al., unpublished data), but ramet density may be a function of rhizome length (among other factors) where shorter rhizomes may result in higher densities. Similar responses in rhizome lengths to grazing have been found in other studies: Jónsdóttir (1991) found higher ramet densities and a trend for shorter rhizomes in C. bigelowii populations grazed by sheep compared to ungrazed populations; Tolvanen and Henry (2000) found the same pattern for ungrazed and muskox-grazed C. stans populations in the High Arctic; and Moen, Ingvarsson, and Walton (1999) found shorter rhizomes in reindeer-grazed plots of the subantarctic Acaena magellanica compared to ungrazed plots. Rhizome length was not measured in the transplant experiment in this study, but Chapin and Chapin (1981) found rhizome mass in Carex aquatilis to be influenced both by original population, environment, and the interaction in a reciprocal transplant experiment.

The degree of ramet differentiation with short and long rhizomes is another aspect of the foraging strategy of the plants, the long ones searching for good patches and the short ones exploiting them (Callaghan, 1977 ). The degree of differentiation in C. stans differed among populations and in most populations there was no clear differentiation. In the Latnjajaure population there was a continuous distribution of rhizome lengths, similar to the one found in C. bigelowii at the same site. This might be due to hybridization between these two populations, for which we have some genetic indications (Stenström et al., 2001 ).

Leaf width
Leaf width is another character related to photosynthetic area. Judging from the transplant experiment, leaf width seems to be the least plastic character of those we measured. However, it is possible that this is due to lack of variation among the initial populations rather than lack of potential plasticity: all except one of the populations that were included in the transplantation experiment had the same initial leaf width and they all became equal in the common environment. This is further supported by the PLS analysis where leaf width correlated best with lemming population cyclicity and lemming index, which vary on a relatively short time scale. Furthermore, in a study by Stenström (2000) , leaf width in C. bigelowii differed among years and showed a plastic response to experimentally enhanced temperature, i.e., leaf width increased by 0.1–0.2 mm. However, this character may also be genetically differentiated to some extent. For example, in a reciprocal transplant experiment C. aquatilis subsp. aquatilis showed differences due to population, but not due to environment in leaf width, although there was an interaction between population and environment (Chapin and Chapin, 1981 ).

In this study, leaves were narrower at peak lemming phase and at high lemming indices. As shoot height responded in similar way to these factors, our results indicate that these plants respond to grazing with reduced leaf area through both a reduced leaf width and a reduced shoot height. In this study leaf width was also influenced by longitude and factors correlated to this variable (yearly mean temperature and precipitation). The reduction in leaf width with increasing longitude (eastwards) might reflect taxonomic differences as C. lugens, which had the narrowest leaves, was more common in the east. However, leaf width is more highly correlated to precipitation than longitude, indicating that narrower leaves might be an adaptation to drier environment in the eastern part of the study area.

Stomata size and density
Stomata density is presumably a plastic character that increases in warmer and more humid habitats, as most of the differences among populations disappeared in the common environment, with one exception though. The C. ensifolia population from NW Taymyr seems to be fixed for a lower number of stomata and although the density increased in the common environment it did not reach the levels comparable to the other populations of the C. bigelowii complex. The plasticity of stomata density is congruent to a previous study of C. lugens and Arctagrostis arundinaceae at Wrangel Island. Those species changed their stomata density within a single growing season when transplanted to warmed chambers (Kislyuk et al., 1983 ). However, arctic and alpine populations of Oxyria digyna and boreal populations of Carex aquatilis differed in stomata density when grown in a common garden, indicating ecotypical differentiation in stomata densities in these species (Au, 1969 ; Standley, 1986 ). Stomata size decreased by transplantation in C. ensifolia and C. lugens, showing that in these taxa it is a plastic character, even though we found no significant differences among taxa based on the original populations (Table 2).

Stomata density was positively correlated to variables related to different measures of temperature and precipitation and negatively correlated with latitude and longitude, but there was also a small negative correlation with lemming grazing. Stomata size had the opposite directions of the correlations and was slightly less influenced by the environmental factors (Table 3B). However, the negative correlation between stomata density and stomata size makes it difficult to tell on which character selection might be working. The literature reports on opposite trends in stomatal characters in different species in response to latitude and environmental factors correlated with latitude. A decrease in stomata density with latitude was also found in Oxyria digyna, a common arctic-alpine forb (Au, 1969 ). However, in C. lugens at Wrangel Island stomata density decreased when exposed to higher temperatures (Kislyuk et al., 1983 ). When comparing six different species occurring both in the Russian Arctic and boreal areas, the arctic populations had higher stomata density in all species, except in Vaccinium vitis-idaea where the arctic populations had lower stomata density (Miroslavov, Voznesenskaya, and Koteyeva, 1998 ). In dry environments selection is expected to optimize the water use efficiency by, e.g., reducing stomata density (Fitter and Hay, 1987 ). This is congruent with our results where stomata densities decreased with decreasing precipitation and increased after growing in the wet Tromsø climate.

Conclusions
The variation in morphological characters among populations of arctic rhizomatous Carex species is to some extent due to genotypic differentiation. However, the populations are also capable of considerable plastic morphological responses to the environment, suggesting that plasticity in ecologically important characters may also be adaptive in the relatively stressful environments of the Arctic. These responses are specific for each taxon and the degree of plasticity differs among populations and taxa. The among-population differences might either be due to differences in the adaptive value of a character or its plasticity or that characters are fixed due to genetic drift in populations at the edge of the distribution. In general, the taxa within the Carex bigelowii complex are more variable than C. stans. Therefore, the C. bigelowii complex may be expected to have greater potential for adjustment to environmental change, at least in the short term. The relationships we found among the individual characters and environmental factors suggest that in response to climate warming we might expect increased shoot height, longer rhizomes, broader leaves, and more numerous but smaller stomata. Given that other community components do not change this might lead to greater plant productivity. In response to increased grazing pressure, however, we might expect smaller plants with narrower leaves and fewer but larger stomata and shorter rhizomes, consequently, slower growing plants.


   FOOTNOTES
 
1 The authors thank the Swedish Polar Research Secretariat for providing the opportunity to go to Siberia with the Tundra Ecology Expedition 1994; Komarov Botanical Institute for hospitality during herbarium studies; Margit Fredriksson for help with the measurements; Olga Khitun for sampling in Kola Peninsula and translation of Russian scientific papers; Torbjörn Fagerström for help in the field and for comments on earlier versions of this manuscript; Torsten Engelskjøn for taking care of the plants in Tromsø Botanical Garden; Kari Anne Bråthen for hospitality and help when sampling in Tromsø; Susanna Andersson, Jette Knudsen, and Björn Nordén for help with the statistical analyses. Financial support was provided to A. Stenström from Rådman och fru Ernst Collianders Stiftelse för välgörande ändamål, Helge Ax:son Johnsons stiftelse, Vilhelm och Martina Lundgrens Vetenskapsfond, Adlerbertska forskningsfonden and Uddenberg-Nordingska stiftelsen, and to I. S. Jónsdóttir from the Swedish Natural Science Research Council. Back

5 Author for reprint requests (isj{at}unis.no ) Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Angerbjörn A. M. Tannerfeldt S. Erlinge 1999 Predator-prey relationships: arctic foxes and lemmings. Journal of Animal Ecology 68: 34-49

Antonovics J. 1971 The effects of a heterogenous environment on the genetics of natural populations. American Scientist 59: 593-599[ISI][Medline]

Au S. 1969 Internal leaf surface and stomatal abundance in arctic and alpine populations of Oxyria digyna. Ecology 50: 131-134[CrossRef][ISI]

Bay C. 1992 A phytogeographical study of the vascular plants of northern Greenland—north of the 74° northern latitude. Meddelelser om Grønland, Bioscience 36: 1-102

Billings W. D. 1987 Constraints to plant growth, reproduction, and establishment in arctic environments. Arctic and Alpine Research 19: 357-365

Callaghan T. V. 1974 Intraspecific variation in Phleum alpinum L. with specific reference to polar populations. Arctic and Alpine Research 6: 361-401[CrossRef]

Callaghan T. V. 1977 Adaptive strategies in the life cycles of South Georgian graminoid species. In G. A. Llano [ed.], Adaptations within Antarctic ecosystems, 891–1002. Proceedings of the 3rd Scientific Committee on Antarctic Research symposium on Antarctic biology, Smithsonian Institution, Washington, D.C., USA

Carlsson B. Å. T. V. Callaghan 1990 Programmed tiller differentiation, intraclonal density regulation and nutrient dynamics in Carex bigelowii. Oikos 58: 219-230[CrossRef][ISI]

Carlsson B. Å. T. V. Callaghan 1994 Impact of climate change factors on the clonal sedge Carex bigelowii: implications for population growth and vegetative spread. Ecography 17: 321-330[CrossRef][ISI]

Carlsson B. Å. I. S. Jónsdóttir B. M. Svensson T. V. Callaghan 1990 Aspect of clonality in the Arctic: a comparison between Lycopodium annotinum and Carex bigelowii. In J. van Groenendael and H. de Kroon [eds.], Clonal growth in plants: regulation and function, 131–151. SPB Academic Publishing, The Hague, The Netherlands

Chapin F. S., III. 1981 Field measurements of growth and phosphate absorption in Carex aquatilis along a latitudinal gradient. Arctic and Alpine Research 13: 83-94

Chapin F. S., III M. C. Chapin 1981 Ecotypic differentiation of growth processes in Carex aquatilis along latitudinal and local gradients. Ecology 62: 1000-1009[CrossRef][ISI]

Clausen J. D. D. Keck W. M. Hiesey 1940 Experimental studies on the nature of species. 1. Effect of varied environments on western North American plants,. Carnegie Institution of Washington Publication Number 520, Wachington, D.C., USA

Douglas D. A. 1981 The balance between vegetative and sexual reproduction of Mimulus primuloides (Scropholariaceae) at different altitudes in California. Journal of Ecology 69: 295-310[CrossRef]

Douglas M. E. J. E. Endler 1982 Quantitative matrix comparisons in ecological and evolutionary investigations. Journal of Theoretical Biology 99: 777-795[CrossRef]

Egorova T. V. B. A. Jurtzev V. V. Petrovskij A. I. Tolmatchev 1966 Flora Arctica URSS Fasc. III Cyperaceae. Nauka, Leningrad, Soviet Union

Eriksson L. E. Johansson N. Kettaneh-Wold S. Wold 1999 Introduction to multi and megavariate data analysis using projection methods (PCA & PLS). Umetrics, Umeå, Sweden

Erlinge S. K. Danell P. Frodin D. Hasselquist P. Nilsson E. Olofsson M. Svensson 1999 Asynchronous population dynamics of Siberian lemmings across the Palearctic tundra. Oecologia 119: 493-500[CrossRef][ISI]

Farmer R. E., Jr. 1993 Latitudinal variation in height and phenology of balsam poplar. Silvae Genetica 42: 148-153[ISI]

Farmer R. E., Jr. G. O'Reilly D. Shaotang 1993 Genetic variation in juvenile growth of tamarack (Larix laricina) in northwestern Ontario. Canadian Journal of Forest Research 23: 1852-1862[CrossRef]

Fedorov A. 1976 Flora Partis Europaeae URSS Tomus II. Nauka, Leningrad, Soviet Union

Fetcher N. R. G. Shaver 1990 Environmental sensitivity of ecotypes as a potential influence on primary productivity. American Naturalist 136: 126-131[CrossRef][ISI]

Fitter A. H. R. K. M. Hay 1987 Environmental physiology of plants, 2nd ed. Academic Press, London, UK

Forman S. L. Ó. Ingólfson V. Gataullin W. F. Manley H. Lokrantz 1999 Late quaternary stratigraphy of western Yamal Peninsula, Russia: new constraints on the configuration of the Eurasian ice sheet. Geology 27: 807-810[Abstract/Free Full Text]

Goryachkin S. V. R. I. Zlotin G. M. Tertitsky 1994 Diversity of natural ecosystems in the Russian Arctic: a guidebook. Reprocentralen, Lund, Sweden

Grime J. P. 1979 Plant strategies and vegetation processes. John Wiley & Sons, Chichester, UK

Havström M. T. V. Callaghan S. Jonasson J. Svoboda 1995 Little ice age temperature estimated by growth and flowering differences between subfossil and extant shoots of Cassiope tetragona, an arctic heather. Functional Ecology 9: 650-654[CrossRef][ISI]

Heathcote C. A. M. S. Davies J. R. Etherington 1987 Phenotypic flexibility of Carex flacca Schreb. Tolerance of soil flooding by populations from contrasting habitats. New Phytologist 105: 381-391[CrossRef][ISI]

Hedberg D. C. Hjort M. Sonesson 1999 The Northeast Passage: an ecological approach. Ambio 28: 210-211[ISI]

Herr J. M., Jr. 1971 A new clearing squash technique for the study of ovule development in angiosperms. American Journal of Botany 58: 785-790[CrossRef][ISI]

Heslop-Harrison J. 1964 Forty years of genecology. Advances of Ecological Research 2: 159-247

Holmgren P. K. N. K. Holmgren L. C. Barnett 1990 Index Herbariorum, Part I; The herbaria of the world. New York Botanical Garden, New York, New York, USA

Hutchings M. J. H. de Kroon 1994 Foraging in plants: the role of morphological plasticity in resource acquisition. Advances in Ecological Research 25: 159-238[ISI]

Jain S. K. A. D. Bradshaw 1966 Evolutionary divergence among adjacent plant populations. Heredity 21: 407-441[ISI]

Johnson D. E. 1998 Applied multivariate methods for data analysts. Duxbury Press, Pasific Grove, California, USA

Jónsdóttir I. S. 1991 Effects of grazing on tiller size and population dynamics in a clonal sedge (Carex bigelowii). Oikos 62: 177-188[CrossRef][ISI]

Jónsdóttir I. S. M. Augner T. Fagerström H. Persson A. Stenström 2000 Genet age in marginal populations of two clonal Carex species in the Siberian Arctic. Ecography 23: 402-412[CrossRef][ISI]

Jónsdóttir I. S. T. V. Callaghan A. Headley 1996 Resource dynamics in arctic clonal plants. Ecological Bulletins 43: 53-64

Jónsdóttir I. S. R. Virtanen I. Kärnefelt 1999 Large-scale differentiations and dynamics in tundra plant populations and vegetation. Ambio 28: 230-238[ISI]

Jónsdóttir I. S. M. A. Watson 1997 Extensive physiological integration: an adaptive trait in resource-poor environments?. In H. de Kroon and J. van Groenendael [eds.], The ecology and evolution of clonal plants, 109–136. Backhuys, Leiden, The Netherlands

Jonsson B. O. 1995 Old populations of the rhizomatous sedge Carex bigelowii show little intermingling of clones (genets). Abstracta Botanica 19: 105-113

Jonsson B. O. I. S. Jónsdóttir N. Cronberg 1996 Clonal diversity and allozyme variation in populations of the arctic sedge Carex bigelowii (Cyperaceae). Journal of Ecology 84: 449-459[CrossRef]

Kershaw K. A. 1962 Quantitative ecological studies from Landmannahellir, Iceland. III. Variation in performance in Carex bigelowii. Journal of Ecology 50: 393-399[CrossRef]

Kibe T. T. Masuzawa 1994 Growth form of Carex bigelowii growing in south central Alaska. Proceedings of the National Institute for Polar Research Symposium Polar Biology (extended abstract) 7: 305-307

King L. 1986 Zonation and ecology of high mountain permafrost in Scandinavia. Geografiska Annaler 68A: 131-139[CrossRef]

Kislyuk I. M. M. D. Vaskovskii L. S. Bubolo T. V. Paleeva 1983 The effect of temperature on the leaf structure and photosynthesis in Carex lugens (Cyperaceae) and Arctagrostis arundinaceae (Poaceae). Botanicheskii Zhurnal 68: 1325-1333

Körner C. M. Neumayer S. P. Menendez-Riedl A. Smeets-Scheel 1989 Functional morphology of mountain plants. Flora 182: 353-383[ISI]

Langlet O. 1971 Two hundred years genecology. Taxon 20: 653-722[CrossRef]

Lubchenco J. J. Cubit 1980 Heteromorphic life histories of certain marine algae as adaptation to variations in herbivory. Ecology 61: 676-687[CrossRef][ISI]

Mantel N. 1967 The detection of disease clustering and a generalized regression approach.