Frequencies of Sexual Morphs in Sagittaria latifolia Populations.
|
by: Sebastian Molnar Abstract Comparisons of ephemeral and stable populations of Sagittaria latifolia are presented. Flower number, leaf area, and flower diameter were recorded per plant, and with these variables, populations were analyzed using t-tests, Pearson correlations, and regression ANOVAs. It was found that flower number and flower diameter can, in many cases, increase with leaf area. Therefore, leaf area can be used as an indicator of resource availability. Means of the three measured variables were used in comparisons of males, females, and hermaphrodites, both within and between populations. Overall, hermaphrodites in S. latifolia populations might have lower fitness levels than males or females. Whether the observed hermaphrodites were “inconstant males” could not yet be determined (as they would need to be followed in subsequent growing seasons). Sagittaria latifolia represents a subdioecious (or trioecious) species. Subdioecy may be a useful strategy for the colonization of new areas. Introduction The plant kingdom contains a variety of reproductive systems, with hermaphroditism being the most common (Dellaporta and Calderon-Urerea, 1993). One breeding system of interest is that of dioecy. A dioecious species is one where male and female reproductive organs (e.g. male and female flowers) are present in separate individuals. The question is, how did dioecy evolve? Various models have been put forth to explain this phenomenon (e.g. Charlesworth and Charlesworth, 1978; Maurice et al., 1994). In general, it has been argued that dioecy has evolved many times independently in different species and along either of two pathways (Freeman et al., 1997): 1) hermaphroditism --> monoecy --> dioecy, or 2) hermaphroditism --> gynodioecy --> dioecy. In both cases, at least two mutations would be required to generate dioecy -- one affecting the female function (i.e. ovules) and the other affecting the male function (i.e. pollen) (Charlesworth and Charlesworth, 1978). Monoecy occurs where male and female (i.e. unisexual) flowers are present on the same individual plant. Unisexual flowers could have arisen either through a homeotic gene function controlling the development of a sex organ, or by sex-determination genes acting independently of homeotic genes (Dellaporta and Calderon-Urrea, 1993). Sexual specialization, leading to greater efficiency in reproduction, has been argued to be the driving force behind the evolution from monoecy to dioecy (Freeman et al. 1993). For the second pathway, a mutation in the hermaphrodite ancestor causing male-sterility would produce a population of females and hermaphrodites (i.e. a gynodioecious population). A subsequent mutation in the hermaphrodites leading to female-sterility (and therefore, generating males) would allow for the gradual emergence of a dioecious population. Alternatively, an initial mutation causing female-sterility could generate an androdioecious population (i.e. males and hermaphrodites), and with male-sterility mutations second, leading to dioecy. Androdioecy is, however, considered to be rare in plants (Charlesworth and Charlesworth, 1978; Dellaporta and Calderon-Urrea, 1993). The evolution of dioecy from gynodioecy could be viewed as an “inbreeding avoidance” mechanism to reduce deleterious effects of inbreeding depression (Freeman et al., 1997). Whether inbreeding avoidance or sexual specialization is the reason for the evolution of dioecy, it may really depend on the particular species under study (reviewed in Freeman et al., 1993). Dellaporta and Calderon-Urrea (1993) have listed the various modes of sexuality at different levels. For example, flowers can either be hermaphroditic (i.e. bisexual) or unisexual (i.e. either male or female). At the individual level, plants can be hermaphrodites, monoecious, dioecious, etcetera. Plant populations can be hermaphroditic, monoecious, dioecious, gynodioecious, androdioecious, or trioecious. Trioecy is rare, yet it may still be useful in understanding other reproductive systems, such as dioecy (Fleming et al. 1994). Trioecious populations contain male, female, and hermaphrodite plants. Trioecy has also been referred to as “subdioecy”, “leaky dioecy”, or “incomplete dioecy” for those dioecious species capable of generating either hermaphroditic or “opposite sex” flowers (Fleming et al. 1994; Humeau et al., 1999). For example, hermaphrodites can be thought of as “inconstant males” -- male plants that occasionally produce female or hermaphrodite flowers. Thus, subdioecious populations could represent intermediate stages between hermaphroditism and dioecy. Some species, however, may truly exist in a trioecious state (i.e. they have three distinct sexual forms, with stable frequencies) rather than in a subdioecious state (Fleming et al. 1994). Dellaporta and Calderon-Urrea (1993) use the term "hermaphrodite" in a restricted sense to include plants which only have hermaphrodite flowers. Herein, I will use “hermaphrodite” in the more general sense (as used by Fleming et al., 1994) to mean “any plant that can reproduce through both male and female function”. This usage includes those individuals that would otherwise be classified as “monoecious”, “inconstant males”, as well as individuals with bisexual flowers. In this paper, the sex frequencies of an aquatic plant, Sagittaria latifolia, at the population level will be studied with respect to the evolution of dioecy. As will be mentioned in more detail below, populations of S. latifolia were found to contain various proportions of males, females, and hermaphrodites. An important point to determine is whether the hermaphrodites in this species are stable entities, or whether they are really “inconstant males”. If the hermaphrodite frequencies are found to be stable, then S. latifolia should be considered as a trioecious species. If unstable, it would be more appropriate to view the species as “subdioecious” (i.e. the hermaphrodites are really “inconstant males”). It will be recommended that S. latifolia is subdioecious. This study will focus on comparisons between stable and ephemeral populations. Materials and Methods Sagittaria latifolia (Alismataceae) is an aquatic plant that lives in the shallow waters of marshes, swamps, and streams. This species has arrow-shaped leaves, is insect pollinated, and has unisexual flowers (each with three white petals) that are typically arranged in whorls of three along an aerial stalk. Whorl variation on some individual plants, however, was observed. In some cases, a secondary stalk replaced one flower in a whorl -- this secondary stalk often had fewer flowers than the primary stalk. For the purposes of this study, only total flower number was considered (see below). The unisexual flowers may contain sterile reproductive organs of the opposite sex -- e.g. a staminate (male) flower may have sterile carpels (Judd et al., 1999).
The present study was carried out over a period of three days in mid-August of 2000. The five populations of Sagittaria latifolia were located at various distances from the Queens Biological Field Station (QUBS) in Kingston (Ontario). A summary of the general features for these populations is shown in table-1. Environmental factors of the populations were noted, such as the water depth, density of surrounding flora, and sunny/shady location. Not all plants in each population had flowered at the time samples were taken, and only plants that had flowers, fruit, or flower buds were used in this study. The total number of flowers were counted (and identified as male or female) for each plant. Flower buds were counted per plant and identified as male or female, based on the hardness of the bud through squeezing (i.e. male buds are soft, and female buds are hard). Where applicable, flower diameter was measured, but for only one flower per plant. To have a relative measurement of resource availability, the area of one leaf per plant was calculated from measurements of length (i.e. tip-to-tip) and width (i.e. edge-to-edge at the approximate mid-point of the leaf). Flowers of random individual plants from populations 3a and 4 were collected and weighed in microcentrifuge tubes, in order to determine if flower mass differed between males, females, and hermaphrodites. Statistical analyses were carried out using SPSS 10.0. For convenience, total numbers counted (per plant) of flowers, flower buds, and fruit, were all included under the heading “flower number” with the assumption that all fruits had once been flowers and that buds would eventually develop into flowers. In addition to the above-mentioned characterization of Sagittaria latifolia populations, pollinator observations and dye experiments were done to determine insect behaviour. Pollinator observations were carried out over periods of 30 minutes at a time, for a particular group of plants (e.g. 2 to 5 plants) in a population (Table-7). Fluorescent powders were used to dye male flowers. Nearest neighhbouring female flowers were collected at the end of the day and stored in microcentrifuge tubes. The collected female flowers were placed under UV light to determine whether dye transfer had occurred (Table-6), which is an indicator for pollen transfer (Campbell, 1989).
Results The means and standard errors for flower number, leaf area, and flower diameter are shown in Table-2 for the five populations. Maps of the populations are shown in Fig.-1(A-E) -- these maps are not drawn to scale, and only show the approximate locations of plants. Table-3 summarizes Pearson correlations, and Table-4 summarizes regression ANOVAs, carried out for each population. In each regression analysis, it was assumed a priori that either flower number or flower diameter is dependent on leaf area (i.e. leaf area was thought to be an indicator for resource availability). Observational and statistical descriptions of each population is given below, as well as the data from pollinator watches and dye experiments. Population 1 -- The sample taken from population 1 consisted entirely of male plants (n=10). This population was small (probably no more than 20 plants) and ephemeral (Fig-1A). Scatterplots for population 1 are shown in Fig-2A and -2B. There was a correlation between flower number and leaf area (r=0.654, df=8, P=0.020), and between flower diameter and leaf area (r=0.746, df=5, P=0.027). A regression ANOVA for leaf area (independent variable) and flower number (dependent variable) showed that flower number increases positively with leaf area (B=0.166; Table-4). The same procedure with flower diameter as the dependent variable failed to reject the null hypothesis (i.e. u1 = u2) when comparing F-tabulated values to F-calculated values (Table-4). Thus, there is no evidence that flower diameter increases with leaf area in this sample. Population 2 -- The sample from population 2 contained 14 males, 7 hermaphrodites, and zero females. Population 2 was a short distance away from population 1 (~0.1 km), but on an adjacent smaller lake, and it was surrounded by a dense growth of trees. With respect to environmental conditions, this population differed from the other four in that it was in the shade all day long (Fig-1B). Note also, that this local population is small and possibly ephemeral (Table-1). Scatterplots for population 2 are shown in fig-3A and -3B. Pearson correlations failed to detect any correlations for leaf area and flower number, or for leaf area and flower diameter, for either the total population or just males or just hermaphrodites (Table-3). Regression ANOVAs detected negative slopes for all of the above mentioned parameters (Table-4, only data for the ìtotalî population 2 are shown). Means for flower number, leaf area, and flower diameter were compared between males and hermaphrodites using t-tests. Mean leaf area was found to be different between males and hermaphrodites (t a(2)=0.05, df=19 = -2.348, P=0.03). On the other hand, mean flower number (t a(2)=0.05, df=19 = 1.924, P=0.069) and mean flower diameter (t a(2)=0.05, df=8 = 1.276, P=0.238) were found to be statistically the same between males and hermaphrodites. Population 3 -- The largest sampling came from population 3 (Fig-1C), which was affectionately named “Leech Hell”. This population had existed for at least 10 years (J. Shore, personal communication), and is therefore considered to be stable (vs. ephemeral). Population 3 had been demarcated into two sectors, 3a and 3b. These sectors were located on opposite sides of the same swamp. Sector 3a was located on the south side of the swamp and was near an outflowing drainage pipe, while sector 3b on the north side, was near an inflowing creek. A two-sample t-test for averages of flower number (t a(2)=0.05, df=163 =1.361, P=0.175 ) and flower diameter (t a(2)=0.05, df=94 = 1.911, P=0.059) from the two sectors showed that 3a and 3b were statistically similar. A comparison of leaf area means between the two sectors did reveal some difference (t a(2)=0.05, df=162 = 2.102, P=0.37). This difference is probably due to the locations of the two sectors -- i.e. there could be a difference in nutrient availability as a consequence of water-flow patterns in the swamp. There was, however, no major barrier for pollinators (save for an approximate 70-100 m distance) between 3a and 3b. Thus, the two sectors can be considered to be part of the same population, rather than in separate populations. Samples from 3a and 3b gave a total of 81 males, 84 females, and 1 hermaphrodite. Data for the one hermphrodite (flower number = 25; leaf area = 175.5 cm2; flower diameter = 3.5 cm) was excluded from the statistical analyses below, as only males and females were compared. Scatterplots of data from male plants are shown in Fig-4A and -4B, and that from female plants in Fig-5A and -5B. A two-sample t-test showed that mean flower number did not differ statistically (t a(2)=0.05, df=163 = 0.5275), whereas both mean leaf area (t a(2)=0.05, df=162 = -2.912) and mean flower diameter (t a(2)=0.05, df=94 = 3.143) did differ, for males and females. The mean leaf area for females (59.3466 ± 6.2535 cm2) was larger than for males (38.9448 ± 3.0411 cm2), while the mean flower diameter for females (2.1654 ± 0.0084 cm) was smaller than for males (2.5900 ± 0.0076 cm). Scatterplots of population 3 are shown in Fig-6A and -6B. Pearson correlations showed that there is a relation between leaf area and flower number, and between leaf area and flower diameter (Table-3). Results from regression ANOVAs suggest that both flower number and flower diameter increase positively with leaf area (Table-4). Population 4 -- The sample from population 4 consisted of 1 male, 5 females, and 5 hermaphrodites. I have listed this population as “gynodioecious” (Table-1), even though it had one male plant. Scatterplots for this population are shown in Fig-7A and -7B. A Pearson correlation showed that there is a relation between leaf area and flower n number (Table-3), however, comparison of tabulated F-values with calculated F-values failed to reject the null hypothesis for a regression ANOVA with leaf area (independent variable) and flower number (dependent variable). Therefore there is no evidence for an increase of flower number with leaf area in this sample. Population 5 -- The sample from population 5 contained 45 males, 13 females, and 4 hermaphrodites. This population, located next to QUBS and referred to as “Cow Island Marsh”, was considered to be stable, as it had existed for a number of years prior to this study (J. Shore, personal communication). Scatterplots of the data for this population are shown in Fig-8A and -8B. Pearson correlation revealed a relation between leaf area and flower number, and between leaf area and flower diameter. Regression ANOVAs suggest a dependency of flower number, and of flower diameter, on leaf area (Table-4). Flower Mass -- Flowers were collected in microcentrifuge tubes from random plants in population 3 and population 4 and subsequently weighed. That is, all flowers from a single plant were collected and then averaged for each plant. Next, mean flower masses were averaged over all male, female, or hermaphrodite individuals from each population. These values are given in Table-5. Mean flower masses of females are very similar for populations 3 and 4 (i.e. 0.144g and 0.146g, respectively) and are larger than values from males and hermaphrodites. Male flowers from hermaphrodites were smallest in mass (0.094g) compared to male flowers from male plants (0.122g) and female flowers from female plants (0.145g). It should be noted that female flowers were not taken from hermaphrodite plants, and therefore no flower-mass comparisons can be made directly between females and hermaphrodites. Pollinators -- Pollinator observations were carried out for populations 2, 3a, and 4. Overall, a wide variety of insect species (e.g. belonging to such families as Halictidae, Coccinellidae, and Cephidae -- identified by M. Sniatenchuk) were observed to visit flowers on S. latifolia plants. The most common insects observed to visit were flower-weavils and ladybugs -- these, however, were probably not pollinators. Ants and unidentified species of flies were found to visit a flower, and stay for minutes at a time. It was found that both male and female flowers produced nectar. Thus, not all insects that visit flowers on Sagittaria latifolia are pollinators, and they simply feed on the nectar. There were, however, a number of species observed to visit flowers that were considered to be pollinators. These included sweat bees, syrphid flies, and skipper moths. Dye experiments showed that dye placed on male flowers can be easily transferred to female flowers through pollinator behaviour. In one case, dye was placed on a male flower, and almost immediately a pollinator (a bee) visited the flower, and eventually flew to a female plant -- the female flowers were collected and the transferred dye could be visualized under UV light. In other cases, male flowers were labeled with dye, and these were left for a few hours. Later, flowers on nearest neighbouring female plants were collected and again, dye that had been transferred could be seen under UV. It should be mentioned that not all collected flowers from a single female plant showed dye transfer (see Table-6). This is to be expected as the amount of dye transferred should decrease as more flowers are visited. Pollinators seem to follow a “nearest neighbour” approach to visiting flowers. For example, a pollinator visits the flowers of a given plant, and tends to move to the flowers on the closest neighbouring plant. Sometimes, pollinators were observed to move far from a plant (rather than just the nearest neighbour). Table-7 shows the average number of visits to occur in 30 min, as well as average length of time for visiting a given flower. Female plants seemed to have the lowest visitation times per flower, males were intermediate, and hermaphrodites had the longest times. This could be due to the sampling effects with a small sample size or the different times of day when the watches were carried out for each population. A more rigorous or consistent method should be done to determine typical pollinator behavior towards Sagittaria latifolia. In any case, pollinator visitation times is of less significance than evidence of actual pollen transfer -- i.e. a pollinator could have a very low visitation time, and yet still transfer enough pollen required for setting seeds. Discussion Flower number and flower diameter, do seem to increase with leaf area. This effect may have to do with nutrient quality and availability in the environment. One way to test this would be to take soil samples, and determine the soil quality at different sites in the populations, and then see whether there are any correlations or regressions to leaf area (or plant size). Other factors may also be involved, such as intra- and inter-specific competition. When comparing the means for flower number, flower diameter, and leaf area across the different populations, there did not appear to be any obvious differences between males, females, and hermaphrodites. The larger the plant, the more likely it will produce a larger number of flowers, regardless of itís gender. The five Sagittaria latifolia populations that were studied show a diversity in gender frequencies (see table-2). With the exception of population 1, it can be seen that hermaphrodites are present at higher frequencies relative to males and/or females, in smaller populations than in large populations. For example, population 4 had equal numbers of females and hermaphrodites; population 2 had half the number of hermaphrodites as males. Population 5, on the other hand, had 4 hermaphrodites out of a sample size of 61. Population 3 had only one hermaphrodite in a sample size of 166 and nearly equal frequencies of males and females. This trend would suggest that hermaphrodites in Sagittaria latifolia populations have lower fitness levels than males and females. Thus, hermaphrodites may be unstable entities in these populations. As such, Sagittaria latifolia is likely to be a subdioecious species. The floral arrangement on the hermaphrodite plants did not seem to occur in a regular pattern. For example, sometimes a male flower was present in a whorl with two female flowers; sometimes, a whorl had three female flowers and another whorl on the same stalk had three male flowers. This could mean that these hermaphrodites are really “inconstant males” (or, quite possibly, “inconstant females”). With respect to evidence from the flower mass data, hermaphrodite male flowers were found to be lowest in mass (Table-5). This would be expected if resources are shunted (in hermaphrodites) into producing female flowers. A comparison (which was not done in this study) of female flowers from females and hermaphrodites would be useful in determining whether or not there are fitness differences between genders. Population 1 is interesting in that it is composed entirely of males. If hermaphrodites really are ìinconstant malesî, then it might be expected that a few of the males in population 2 could produce female flowers at some point. On the other hand, this population appears to be in an area of low nutrient availability (e.g. there was ìsparseî vegetation in the area). Humeau et al. (1999) has indicated that “leaky dioecy” may be favored in small populations and that environmental variation may influence the breeding system. To test this idea in Sagittaria latifolia, some of the male plants in population 1 could be transplanted to another area -- an area with higher nutrient availability -- and then these individuals could be followed for one or more seasons to see if they do show signs of being “inconstant males”. As noted by Fleming et al. (1994), many factors are involved when unisexual morphs (i.e. males or females) invade hermaphrodite populations. In particular, the fertility of the unisexual must be higher than that of the hermaphrodites. As demonstrated by Charlesworth and Charlesworth (1978), an established unisexual morph in the population enables the inclusion of the second unisexual morph. Maurice et al. (1994) have shown that males can be selected for and that dioecy can evolve when sex-determination is nuclear-cytoplasmic (vs. nuclear). In the sample from population 4, the number of hermaphrodites equaled the number of females (i.e. 5 and 5), and there was one male. Early on, this small population might have consisted of only females and hermaphrodites. The presence of the one male plant could represent the initial stages of a shift towards a more dioecious population (i.e. one that has predominantly males and females). It would be necessary to follow this population in subsequent years to determine whether or not a shift in gender frequencies can take place in this species. It would also be useful to determine whether sex-determination in Sagittaria latifolia is nuclear or nuclear-cytoplasmic. In summary, Sagittaria latifolia is a subdioecious species. Different populations were found to vary with respect to gender frequencies, which may be an effect of population size. This subdioecy, however, may not necessarily be headed towards complete dioecy. It has been found that some species contain intrinsic factors that can maintain intermediate stages between hermaphroditism and dioecy (Ashman, 1999). Subdioecy appears to be a strategy used to colonize new areas effectively, and for the maintenance of established populations.
References:
|
Get your own Free Home Page