Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2017 Dec 29;121(2):377–383. doi: 10.1093/aob/mcx163

The enigma of sex allocation in Selaginella

Kurt B Petersen 1, Martin Burd 1,✉
PMCID: PMC5808784  PMID: 29300810

Abstract

Background and Aims

The division of resource investment between male and female functions is poorly known for land plants other than angiosperms. The ancient lycophyte genus Selaginella is similar in some ways to angiosperms (in heterospory and in having sex allocation occur in the sporophyte generation, for example) but lacks the post-fertilization maternal investments that angiosperms make via fruit and seed tissues. One would therefore expect Selaginella to have sex allocation values less female-biased than in flowering plants and closer to the theoretical prediction of equal investment in male and female functions. Nothing is currently known of sex allocation in the genus, so even the simplest predictions have not been tested.

Methods

Volumetric measurements of microsporangial and megasporangial investment were made in 14 species of Selaginella from four continents. In five of these species the length of the main above-ground axis of each plant was measured to determine whether sex allocation is related to plant size.

Key Results

Of the 14 species, 13 showed male-biased allocations, often extreme, in population means and among the great majority of individual plants. There was some indication from the five species with axis length measurements that relative male allocation might be related to the release height of spores, but this evidence is preliminary.

Conclusions

Sex allocation in Selaginella provides a phylogenetic touchstone showing how the innovations of fruit and seed investment in the angiosperm life cycle lead to typically female-biased allocations in that lineage. Moreover, the male bias we found in Selaginella requires an evolutionary explanation. The bias was often greater than what would occur from the mere absence of seed and fruit investments, and thus poses a challenge to sex allocation theory. It is possible that differences between microspores and megaspores in their dispersal ecology create selective effects that favour male-biased sexual allocation. This hypothesis remains tentative.

Keywords: Heterospory, lycophyte, male bias, Selaginella, sex allocation, wind dispersal

INTRODUCTION

In contrast to our reasonably ample knowledge of sex allocation in angiosperms (Goldman and Willson, 1986; Campbell, 2000; de Jong and Klinkhamer, 2005), we know little about the sexual division of reproductive investment in free-sporing land plants: liverworts, mosses, hornworts, the early diverging vascular lineage of lycophytes, and the ferns and related lineages (monilophytes) (Judd et al., 2008). Life histories in these lineages, simpler in many ways than in flowering plants, make them potentially interesting models for probing the theory of sex allocation (Charnov, 1982; West, 2009).

Sex allocation in land plants is closely tied to the alternation of generations, a fundamental innovation underlying the ecological success of land plants in the terrestrial environment (Niklas and Kutschera, 2009). In some plant life histories, sex allocation occurs in the gametophyte generation. This is true of about half the species in the non-vascular lineages – liverworts, mosses, hornworts – and in the vast majority of monilophytes (Tryon and Tryon, 1982; Wyatt and Anderson, 1984). Sporophytes of these species produce spores of a small, unimodal size (isospores) that give rise to bisexual gametophytes. Such spores have no unique sexual identity. Only in the development of the bisexual gametophytes does the opportunity arise to invest resources in specifically male or female functions (in antheridia and archegonia).

In contrast, sex allocation resides in the sporophyte generation of those species in which obligately unisexual gametophytes arise from separate male and female spores. In the non-vascular lineages (McDaniel et al., 2012), these spores are morphologically isospores [or with slight size difference (Vitt, 1968)] but there is a chromosomal basis to sexual identity (Immler and Otto, 2015), and spore mother cells necessarily give rise to two female and two male spores in a meiotic tetrad (Allen, 1917; McLetchie, 1992). Gametophytes grown from isolated spores produce nearly equal sex ratios in some but not all species (Newton, 1972; Stark et al., 2010). Sexual differences in spore abortion (Longton and Greene, 1979; McLetchie 1992) may alter sporophytic sex allocation from strict equality between male and female investment.

Among vascular plants, unisexual spores are associated with heterospory, a key innovation in the evolutionary history of land plants (Petersen and Burd, 2017). Small, male microspores and large, female megaspores differ greatly in size and are produced in separate sporangia in the heterosporous vascular plants (Petersen and Burd, 2017). Angiosperms, for example, determine sex allocation by the production of microsporangia (anthers), megasporangia (ovules), and the associated floral and extra-floral structures that assist male and female function. One important prediction of sex allocation theory has been abundantly demonstrated in angiosperms. Self-pollination results in a highly structured mating arena of competition among sibling pollen grains that creates selection for decreased allocation to male function (Charlesworth and Charlesworth, 1981; de Jong et al., 1999). Shifts towards greater relative female allocation with increasing rates of self-fertilization are, indeed, among the best documented aspects of sexual allocation in flowering plants (Goldman and Willson, 1986; Campbell, 2000; de Jong and Klinkhamer, 2005).

Outside the seed plants, the only heterosporous vascular lineages are two closely related families of leptosporangiate water ferns, Marsileaceae and Salviniaceae, and two closely related monotypic families within the lycophytes, Selaginellaceae and Isoetaceae (Tryon and Tryon, 1982). As in flowering plants, sporophytes of these lineages invest in the production of microspores with specifically male function, and megaspores with female function. They can adjust sexual allocation through the size and number of microsporangia and megasporangia produced. In contrast to angiosperms, however, megaspores are not retained on the parent plant in these free-sporing lineages, and so their life cycle provides no opportunity for post-fertilization maternal investments in secondary structures like seeds and fruits. This simple but important life-history difference makes the free-sporing heterosporous plants useful contrasts to the angiosperms for exploring sex allocation. Here, we focus our attention on Selaginella.

Seed and fruit maturation consumes much of a flowering plant’s maternal investment (Burd and Head, 1992; Day and Aarssen, 1997; Baker et al., 2005) and so estimates of sex allocation at the end of a reproductive season tend to be more female-biased than the corresponding allocations measured at anthesis (Cruden and Lyon, 1985; Goldman and Willson, 1986; Ågren and Schemske, 1995). Female-biased allocation at the fruiting stage is common among flowering plants (Campbell, 1992; Klinkhamer et al., 1997; Baker et al., 2005). For example, six outcrossing species examined by Cruden and Lyon (1985) devoted 81–93 % of their total reproductive biomass to female function.

In Selaginella, paternal and maternal sexual investments yield microspores and megaspores, which are dispersed from the sporophyte. Gametophyte development, gamete production and mating then occur in the environment away from the parent sporophyte. The absence of megaspore retention precludes any sporophytic maternal investments beyond the reserves originally placed in the megaspores. In the absence of seed and fruit investment, we would expect Selaginella to show less female bias in sexual investment than is typically the case in angiosperms. Indeed, spore dispersal and external mating should create fairly homogeneous mating opportunities throughout populations of Selaginella, so that they conform to the touchstone prediction of sex allocation theory that hermaphrodites should allocate half their reproductive investment to each sex function in a well-mixed mating arena (Charnov, 1982; West, 2009). Confirmation of these simple predictions would reveal something of the reproductive ecology of free-sporing plants and at the same time provide a phylogenetic reference point for understanding the sex allocation patterns of angiosperms.

Sex allocation appears not to have been measured previously in Selaginella. We examined 14 species of Selaginella with a diverse array of growth habits native to four continents. In five of these species we also measured the main axis length of individual plants to see whether sex allocation varied with plant size. Of the 14 species, a pronounced male-biased allocation occurred in 13 of them. While this result confirms our expectation that the evolutionary innovations associated with seeds and fruits tend to give angiosperms their characteristic female-biased investment, it also presents the issue of explaining the pervasive and sometimes extreme male-biased investment in Selaginella. Wind dispersal differences between megaspores and microspores may be implicated, although our evidence of a plant size effect on sex allocation is equivocal on this point. We consider other factors that may play a role in Selaginella sex allocation that should also be relevant for sex allocation theory in general.

MATERIALS AND METHODS

Species and locations

We sampled 14 Selaginella species with a variety of growth forms from a variety of habitats, including temperate eucalypt forest in south-eastern Australia, rainforest in north-eastern Australia, wet dipterocarp forest in Peninsular Malaysia and Borneo and wet lowland forest in Central America, along with species of various tropical and sub-tropical origins occurring in the Singapore Botanic Gardens (Table 1). Sampling took place between November 2010 and August 2015. The species we examined represent a phylogenetically as well as geographically broad sample. The majority-rule consensus tree from a Bayesian analysis of nuclear and plastid genes by Weststrand and Korall (2016) had a basally diverging lineage of two species that is sister to the remainder of the genus, itself divided into two diverse clades labelled A and B. In a Bayesian analysis of rbcL sequences from additional species, including 12 of the 14 species examined here, we recovered this basic topology (K. B. Petersen and M. Burd, unpubl. res.). Within clade A, the Australian S. uliginosa occurs in the subclade ericitorum, and the neotropical S. arthritica and South African S. kraussiana in subclade gymnogynum. Four subclades within clade B are represented: the south-east Asian species S. willdenowii, S. mayeri and S. plana in one, S. longipinna and S. frondosa in a second, and the South American species S. erythropus and S. haematodes in the third, with S. intermedia in a fourth. We have no information on the phylogenetic position of S. padangensis and S. brisbanensis. In the absence of complete phylogenetic information for all 14 species, we present our results here without a phylogenetic comparative framework, and defer such an analysis for the future and a larger data set.

Table 1.

Species descriptions, sample sizes and sampling locations

Species Habit and typical maximum size N Sample sites
S. uliginosa Erect and delicate, lightly branched stems ~20 cm high 50 Victoria, Australia
S. kraussiana Prostrate and heavily branched, 150 Victoria, Australia
S. willdenowii Scrambling climber, main axis up to 700 cm 30 Ampang, Selangor, Malaysia
S. intermedia Low, suberect to ~25 cm high, densely foliaged 30 Gunung Mulu, Sarawak, Malaysia
S. plana Frondose, heavily branched, to ~70 cm high 30 Gunung Mulu, Sarawak, Malaysia
S. haematodes Frondose and densely foliaged 30 Barro Colorado Island, Panama
S. arthritica Frondose 30 Barro Colorado Island, Panama
S. australiensis Prostrate, creeping, sometimes growing on trees 30 Wooroonooran and Barron Gorge, Queensland, Australia
S. brisbanensis Prostrate, delicate, heavily branched 30 Wooroonooran, Queensland, Australia
S. longipinna Frondose 60 Wooroonooran, Queensland, Australia
S. padangensis Somewhat frondose, heavily branched 30 Singapore Botanic Gardens, Singapore
S. erythropus Somewhat frondose, ventral surface distinctly red 30 Singapore Botanic Gardens, Singapore
S. mayeri Prostrate, spreading 30 Singapore Botanic Gardens, Singapore
S. frondosa Frondose, basal stem reddish 30 Singapore Botanic Gardens, Singapore
Open in a new tab

Measurements

We estimated sex allocation for individual plants from two components, a count of the number of microsporangia and megasporangia in a sample of strobili from a plant, and measurements of the volume of megasporangia and microsporangia. Sporangium contents are completely converted to spores (Gifford and Foster, 1989; Morbelli and Rowley, 1993), so the volume of a spore sac will reflect resource investment by the sporophyte. In 2011, we checked the biochemical composition of mature spores of S. uliginosa through Fourier transform infrared spectroscopy. Spectra for microspores and megaspores were nearly identical (unpubl. res.), implying equivalent material composition and thus equal costs per unit volume for each sex function. We assume similar equivalence of the micro- and megaspore contents in the other species.

For most species we sampled 30 plants per species, more from three Australian species for which we could sample multiple populations (Table 1). Single plants may produce hundreds of strobili, each containing dozens of sporangia. On each plant we removed ten randomly selected strobili (five for S. uliginosa and S. kraussiana) and counted the number of megasporangia and microsporangia on each under a dissecting microscope. The type of spore sac is readily distinguished by the presence of a single large tetrad of spores within megasporangia, and if there was any doubt we simply crushed the sac to determine whether it contained four large spores or many tiny ones. Even immature spore sacs could be distinguished in this way.

Sporangial volumes were calculated from linear measurements made directly with an adjustable ocular micrometer on a dissecting microscope or from calibrated digital micrographs. Volume calculation required different approaches for micro- and megasporangia, and occasionally idiosyncratic variations for particular species depending on the shape of the sporangia. (1) Microsporangia are ellipsoid to reniform, depending on species. For ellipsoid sporangia, we measured the major axis (length) and minor axis (height) of the transverse profile of the sporangium, and calculated the volume as an ellipsoid of rotation about the major axis. That is, for a microsporangium with length 2w and height 2x, we took its volume to be V0 = 4πwx2/3. For reniform shapes, we made micrographs of both transverse and tangential profiles. We divided the transverse face into ten sections of equal width along the major axis and determined the area of each section. We multiplied this area by the depth of the section measured on the tangential face to obtain a volume for the section. The summed volumes of the ten sections gave the whole microsporangium volume. (2) Megasporangia are broadly tetragonal, following the arrangement of the four meiotic products, and megaspores are approximately spherical. It was simpler to measure individual spore volumes and then multiply by 4 to yield an estimate of megasporangium volume. Individual megaspore volumes were calculated as an ellipsoid of rotation of the spore profile about its major axis (major and minor axes were usually nearly equal, so megaspores differed only marginally from spheres). That is, a megaspore with a major diameter 2y and minor diameter 2z would have its volume calculated as Vs = 4πyz2/3, and the megasporangium as V1 = 4Vs.

Sporangium volumes V0 and V1 were calculated in this way for 30−50 mature sporangia of each type per species, from which we determined mean volumes for the species. We took the product of the mean sporangial volume and the count of the sporangia of that type on an individual plant to obtain a volumetric estimate of resource allocation to male or female function for the plant. Individual sex allocation was quantified as the proportion of microsporangial volume relative to total volume of all sporangia. That is, a plant that produced N0 microsporangia and N1 megasporangia in its sample of strobili would have a sex allocation N0V0/(N0V0 + N1V1), i.e. M/(M + F) for resource investment in male, M, and female, F, functions. Thus, equal allocation corresponds to a value of 0.5.

We attempted to detect any dependence of sex allocation on vegetative size for a subset of the species in our sample. As an estimate of plant size, we measured the total length of the main axis of S. uliginosa, S. kraussiana, S. willdenowii, S. intermedia and S. plana. Main axis length is only an approximate basis for comparison among these five species, given substantial differences in growth habit among them (Table 1). We use axis length as an indicative measure here, but better study of the effect of vegetative size or resource status on sex allocation is needed for Selaginella, as for most plant species.

Analysis

The main objective of the analysis was to test whether sex allocation deviated significantly from equal male and female investment for individual plants and for population means. Because sex allocation was determined for individual plants from a subsample of strobili, we calculated Studentized 99 % confidence intervals (CI) for the sex allocation of each plant from 1000 bootstrap samples of its strobili. A 99 % CI that did not include 0.5 was considered to indicate significant departure from equal allocation. At the population level, we calculated the mean sex allocation among the plants in each sample, and determined the Studentized 99 % CI for population means from 1000 bootstrap samples of the individual plants. Bootstrapping was carried out with the R package ‘boot’ (Canty, 2002).

To assess the relationship of sex allocation to plant vegetative size, we first normalized the axis length measurements to each species’ maximum, because the range of this metric was vastly incommensurate among species, and then tested whether there were homogeneous slopes for this relation using analysis of covariance with a plant size × species interaction term. This interaction was significant, indicating heterogeneity in the sex allocation–plant size relationship among species. We therefore conducted a separate regression analysis for each species. The analysis of covariance and regression analyses were calculated using the R base package, ver. 3.3.1 (R Core Team, 2016).

RESULTS

Nearly every species had a mean population sex allocation that was significantly male-biased (Table 2), as did most individual plants (Fig. 1). Only S. longipinna, a frondose species of tropical northern Australia, had nearly equal mean sex allocation and a substantial number of individual plants with female biased allocation. Among the remaining species, male bias was sometimes extreme. Six species invested ≥80 % of their sporangial investment in microspores, and it was not uncommon to find individual plants that produced no megaspores among the strobili we sampled. An occasional plant with significantly female-biased allocation occurred in S. padangensis, S. erythropus and S. haematodes, but these exceptions did not obscure the evident pattern of male bias in these species, and in the sample at large (Fig. 1).

Table 2.

Numbers and sizes of megasporangia and microsporangia, and population sex allocation for the study species. N0, N1, number of microsporangia and megasporangia, respectively, per plant; V0, V1, volume of microsporangium and megasporangium, respectively. Values are means ± 1 s.d. Sex allocation was measured as the proportion of the total volume of sporangia used for male function (i.e. microsporangial volume). Thus, values >0.5 represent male-biased allocation. Confidence intervals were obtained by bootstrapping (see Materials and methods)

Species N 0 N 1 V 0 (mm3) V 1 (mm3) Sex allocation
Mean 99 % CI
S. uliginosa 57.1 ± 23.5 12.7 ± 7.6 0.113 ± 0.046 0.142 ± 0.064 0.78 (0.67, 0.89)
S. kraussiana 74.1 ± 11.3 5 ± 0.0 0.136 ± 0.033 0.802 ± 0.227 0.70 (0.65, 0.75)
S. willdenowii 357.7 ± 102.1 15.0 ± 8.0 0.106 ± 0.028 0.369 ± 0.062 0.87 (0.78, 0.94)
S. intermedia 572.2 ± 194.2 166.3 ± 89.7 0.128 ± 0.011 0.056 ± 0.02 0.87 (0.82, 0.94)
S. plana 868.0 ± 169.2 61.3 ± 58.2 0.071 ± 0.013 0.125 ± 0.049 0.89 (0.82, 0.94)
S. haematodes 302.1 ± 87.6 71.6 ± 50.1 0.018 ± 0.005 0.024 ± 0.005 0.77 (0.71, 0.84)
S. arthritica 480.5 ± 181.7 10 ± 0.0 0.029 ± 0.017 0.089 ± 0.04 0.93 (0.91, 0.94)
S. australiensis 563.4 ± 156.7 9.8 ± 0.4 0.017 ± 0.006 0.233 ± 0.113 0.80 (0.73, 0.84)
S. brisbanensis 116.2 ± 22.8 8.8 ± 1.9 0.018 ± 0.004 0.132 ± 0.045 0.65 (0.56, 0.77)
S. longipinna 270.3 ± 95.3 99.1 ± 45.8 0.024 ± 0.008 0.058 ± 0.012 0.51 (0.42, 0.66)
S. padangensis 611.0 ± 267.0 159.9 ± 106.4 0.057 ± 0.015 0.106 ± 0.027 0.68 (0.60, 0.77)
S. erythropus 314.7 ± 140.7 97.5 ± 97.6 0.023 ± 0.007 0.039 ± 0.012 0.71 (0.62, 0.81)
S. mayeri 452.1 ± 174.9 108.0 ± 51.6 0.057 ± 0.007 0.086 ± 0.03 0.74 (0.64, 0.86)
S. frondosa 572.6 ± 208.0 51.4 ± 68.1 0.019 ± 0.004 0.041 ± 0.008 0.85 (0.78, 0.94)
Open in a new tab

Fig. 1.

Fig. 1.

Open in a new tab

Sex allocation (proportion of male investment measured as microsporangial fraction of total sporangial volume) in 14 species of Selaginella. Values >0.5 indicate an investment bias towards male function and <0.5 an investment bias towards female function. Each point represents an individual plant. Plants with 99 % CIs for sex allocation that include 0.5 are indicated by open circles; plants with significantly sex-biased allocations are indicated by filled circles. Geographically distinct populations of S. longipinna, S. kraussiana and S. uliginosa are shown on separate lines.

The differences among species in sex allocation were due to variation in both the sizes and numbers of male and female sporangia they produced. There was, nonetheless, greater variation in numbers than sizes. Megasporangia ranged from less than half the size of microsporangia (S. intermedia) to 13.7 times the size of microsporangia (S. australiensis), while the number of microsporangia ranged from ~3 times (S. brisbanensis) to >57 times (S. padangensis) the number of megasporangia (Table 2). Because both contribute to species differences in sex allocation, neither the size nor the number ratio was systematically related to sex allocation (Fig. 2).

Fig. 2.

Fig. 2.

Open in a new tab

Relationship of species mean sex allocation to sporangial size ratio (mean megasporangium volume relative to mean microsporangium volume, V1/V0) (filled symbols) and sporangial number ratio (mean number of microsporangia relative to mean number of megasporangia, N0/N1) (open symbols).

Sex allocation bore little relation to vegetative size in the species we measured. Homogeneity of slopes among species was strongly rejected (species × normalized axis length interaction: F4,260 = 7.63, P = 0.000008), requiring separate regression for each species. The slope of the relationship between sex allocation and axis length was significantly different from zero in two species, S. kraussiana and S. plana, but the relationships were not strong, accounting for only ~6 % of the variance in sex allocation in S. kraussiana and ~30 % in S. plana (Table 3). In both cases, male allocation increased with increasing axis length. Selaginella kraussiana and S. plana differ in plant habit and typical height, so the connection between male allocation and plant size in these species does not appear to follow from a distinctive morphology relative to the other species.

Table 3.

Regression relationships between sex allocation and vegetative size (relativized main axis length) of individual plants

Species Slope F d.f. P R 2
S. uliginosa –0.00030 1.70 1, 48 0.198 0.034
S. kraussiana 0.00013 7.60 1, 128 0.007 0.056
S. willdenowii –0.00001 0.43 1, 28 0.517 0.015
S. intermedia –0.00024 3.26 1, 28 0.081 0.104
S. plana 0.00038 12.10 1, 28 0.002 0.302
Open in a new tab

DISCUSSION

In flowering plants, maternal tissues such as fruits and arils or maternally supported tissue such as endosperm are a principal source of female-biased sex allocation (Burd and Head, 1992; Day and Aarssen, 1997; Baker et al., 2005). Because the Selaginella life cycle lacks these features, we expected to find a more balanced division of resource allocation between male and female function. Indeed, no species in our sample had female-biased allocation. On the contrary, 13 of the 14 species showed a statistically significant and substantial male bias, with male investment often exceeding female by a factor of ≥2 (Table 2). This bias is more extreme than can be accounted for merely by the absence of fruit and seed investment (Goldman and Willson, 1986; Ågren and Schemske, 1995; Baker et al., 2005). For example, only six of 13 xenogamous angiosperm species examined by Cruden and Lyon (1985) had a mean male allocation based on stamen and pistil biomass that exceeded 0.67 at the floral stage, while this was true of 12 of the 14 Selaginella species we examined. Thus, our results confirm one simple expectation but raise the question of what accounts for the male bias among Selaginella species.

Although it is rare in angiosperms, male-biased sex allocation has been reported in Andropogon gerardii and Sorghastrum nutans (Poaceae), which had 60–90 % male allocation for most investment currencies, including biomass (McKone et al., 1998). The male investment bias in these two species arises, in all likelihood, from wind pollination and passive dispersal of seeds. When one sex function has more restricted dispersal than the other, selection favours investment in the better-dispersing sex (Bulmer and Taylor, 1980; West, 2009; Fromhage and Kokko, 2010). Wind pollination and passive seed dispersal are likely to afford better dispersal opportunity to male than to female function (Fromhage and Kokko, 2010; Pickup and Barrett, 2012). Accordingly, male allocation increases with plant height in several wind-pollinated species because pollen dispersal is greatly enhanced by elevated release height (McKone and Tonkyn, 1986; Burd and Allen, 1988; Solomon, 1989; Bickel and Freeman, 1993; Fox, 1993).

Does this effect occur in Selaginella? In principle it could. Microspores of Selaginella are typically 18–60 μm in diameter, while megaspores range from 200 to 1033 μm in diameter (Tryon and Lugardon, 1991). The smaller, lighter microspores generally disperse further than megaspores (Filippini-DiGiorgi et al., 1997), setting the stage for selection favouring allocation to male function. However, we found only modest evidence of an effect of plant height on sex allocation within species (Table 3). Selaginella kraussiana and S. intermedia have prostrate growth, so that the length of the main plant axis would not be related to release height of spores. Curiously, male allocation increased with increasing axis length in S. kraussiana, but the relationship was weak (Table 3). Selaginella willdenowii has a main axis that can reach several metres in length, but it is a scrambling climber on surrounding plants, and its axis length may also be poorly related to spore release height. Only S. plana and S. uliginosa have erect growth forms in which the main axis length would reflect height above ground, but S. uliginosa is diminutive and may have too little height variation to reveal the expected effect of spore dispersal on sex allocation. In S. plana we found a statistically significant positive relationship between main axis length and male allocation (Table 3), in support of the hypothesized effect of wind dispersal on sex allocation. Thus, the evidence from intraspecific variation suggests, but only tentatively, that male–female dispersal differences affect sex allocation in at least some Selaginella species.

Dispersal of microspores and megaspores may differ generally within the genus whether or not individual plant size has a strong effect, of course. It remains possible, therefore, that differences in spore dispersal explain the apparently widespread male allocation bias in Selaginella, but this is an open hypothesis. The apparent extent of male-biased allocation within the genus is consistent with such an effect, but much additional investigation would be needed to substantiate it.

We have some qualitative indication that the male bias we found for most of the species in our sample is widespread among Selaginella species generally. As part of our continuing study of Selaginella ecology, we examined specimens of over 100 species at the US National Herbarium of the Smithsonian Institution in order to extract and photomicrograph microspores and megaspores. As we searched specimens for sporangia that had retained spores, we noticed that it was very frequently difficult to find megasporangia, although microsporangia were abundant, implying that these specimens had male-biased sexual allocation. This pattern was not universal, however. Specimens from occasional species did bear abundant megasporangia. The enigmatic variation in sex allocation within and among Selaginella species, and differences from the angiosperms, are interesting aspects of land plant biology in their own right, but will also provide useful empirical models for probing the theory of sex allocation.

ACKNOWLEDGEMENTS

We thank the following institutions and individuals for providing permissions or assisting us with collection of specimens in the field: Parks Victoria and the Department of Sustainability and Environment, Victoria, and the Department of Environment and Heritage Protection, Queensland, for permission to collect native Australian flora; Cathy Yule of the Monash University Malaysia campus, the Sarawak Forestry Department, and the staff of Gunung Mulu National Park for facilitating the field work in Malaysia; the Singapore Botanic Gardens for access to their facilities and plant collections. We are especially grateful to Jennifer Grundy, Kirsten Barks and Shennai Palermo for the many hours they spent collecting and measuring Selaginella sporangia. This work was supported by an Australian Government Research Training Program (RTP) Scholarship to K.B.P. and Australian Research Council grant DP140103196 to M.B.

LITERATURE CITED

  1. Ågren J, Schemske DW. 1995. Sex allocation in the monoecious herb Begonia semiovata. Evolution 49: 121–130. [DOI] [PubMed] [Google Scholar]
  2. Allen CE. 1917. A chromosome difference correlated with sex differences in Sphaerocarpos. Science 46: 466–467. [DOI] [PubMed] [Google Scholar]
  3. Baker AM, Burd M, Climie KM. 2005. Flowering phenology and sexual allocation in single-mutation lineages of Arabidopsis thaliana. Evolution 59: 970–978. [PubMed] [Google Scholar]
  4. Bickel AM, Freeman DC. 1993. Effects of pollen vector and plant geometry on floral sex ratio in monoecious plants. American Midland Naturalist 130: 239–247. [Google Scholar]
  5. Bulmer MG, Taylor PD. 1980. Dispersal and the sex ratio. Nature 284: 448–449. [Google Scholar]
  6. Burd M, Allen TFH. 1988. Sex allocation in wind-pollinated plants. Evolution 42: 403–407. [DOI] [PubMed] [Google Scholar]
  7. Burd M, Head G. 1992. Phenological aspects of male and female function in hermaphroditic plants. American Naturalist 140: 305–324. [DOI] [PubMed] [Google Scholar]
  8. Campbell DR. 1992. Variation in sex allocation and floral morphology in Ipomopsis aggregata (Polemoniaceae). American Journal of Botany 79: 516–521. [Google Scholar]
  9. Campbell DR. 2000. Experimental tests of sex allocation theory in plants. Trends in Ecology and Evolution 15: 227–231. [DOI] [PubMed] [Google Scholar]
  10. Canty AJ. 2002. Resampling methods in R: the boot package. R News 2: 2–7. cran.r-project.org/doc/Rnews/. [Google Scholar]
  11. Charlesworth D, Charlesworth B. 1981. Allocation of resources to male and female function in hermaphrodites. Biological Journal of the Linnean Society 15: 57–74. [Google Scholar]
  12. Charnov EL. 1982. Sex allocation. Princeton: Princeton University Press. [Google Scholar]
  13. Cruden RW, Lyon DL. 1985. Patterns of biomass allocation to male and female functions in plants with different mating systems. Oecologia 66: 299–306. [DOI] [PubMed] [Google Scholar]
  14. Day T, Aarssen LW. 1997. A time commitment hypothesis for size-dependent gender allocation. Evolution 51: 988–993. [DOI] [PubMed] [Google Scholar]
  15. Filippini-De Giorgi A, Holderegger R, Schneller JJ. 1997. Aspects of spore dispersal in Selaginella. American Fern Journal 87: 93–103. [Google Scholar]
  16. Fox JF. 1993. Size and sex allocation in monoecious woody plants. Oecologia 94: 110–113. [DOI] [PubMed] [Google Scholar]
  17. Fromhage L, Kokko H. 2010. Spatial seed and pollen games: dispersal, sex allocation, and the evolution of dioecy. Journal of Evolutionary Biology 23: 1947−1956. [DOI] [PubMed] [Google Scholar]
  18. Gifford EM, Foster AS. 1989. Morphology and evolution of vascular plants, 3rd edn. New York: Freeman. [Google Scholar]
  19. Goldman DA, Willson MF. 1986. Sex allocation in functionally hermaphroditic plants: a review and critique. Botanical Review 52: 157–194. [Google Scholar]
  20. Immler S, Otto SP. 2015. The evolution of sex chromosomes in organisms with separate haploid sexes. Evolution 69: 694–708. [DOI] [PubMed] [Google Scholar]
  21. de Jong TJ, Klinkhamer PGL. 2005. Evolutionary ecology of plant reproductive systems. Cambridge: Cambridge University Press. [Google Scholar]
  22. de Jong TJ, Klinkhamer PGL, Rademaker MCJ. 1999. How geitonogamous selfing affects sex allocation in hermaphroditic plants. Journal of Evolutionary Biology 12: 166–176. [Google Scholar]
  23. Judd WS, Campbell CS, Kellog EA, Stevens PF, Donoghue MJ. 2008. Plant systematics: a phylogenetic approach, 3rd edn Sunderland, MA: Sinauer. [Google Scholar]
  24. Klinkhamer PGL, de Jong TJ, Metz H. 1997. Sex and size in cosexual plants. Trends in Ecology and Evolution 12: 260–265. [DOI] [PubMed] [Google Scholar]
  25. Longton RE, Greene SW. 1979. Experimental studies of growth and reproduction in the moss Pleurozium schreberi (Brid.) Mitt. Journal of Bryology 10: 321–338. [Google Scholar]
  26. McDaniel SF, Atwood J, Burleigh JG. 2012. Recurrent evolution of dioecy in bryophytes. Evolution 67: 567–572. [DOI] [PubMed] [Google Scholar]
  27. McKone MJ, Tonkyn DW. 1986. Intrapopulation gender variation in common ragweed (Asteraceae: Ambrosia artemisiifolia L.), a monoecious, annual herb. Oecologia 70: 63–67. [DOI] [PubMed] [Google Scholar]
  28. McKone MJ, Lund C, O’Brien J. 1998. Reproductive biology of two dominant prairie grasses (Andropogon gerardii and Sorghastrum nutans, Poaceae): male-biased sex allocation in wind-pollinated plants?American Journal of Botany 85: 776−783. [PubMed] [Google Scholar]
  29. McLetchie DN. 1992. Sex ratio from germination through maturity and its reproductive consequences in the liverwort Sphaerocarpos texanus. Oecologia 92: 273–278. [DOI] [PubMed] [Google Scholar]
  30. Morbelli MA, Rowley JR. 1993. Megaspore development in Selaginella. I. “Wicks”, their presence, ultrastructure, and presumed function. Sexual Plant Reproduction 6: 98–107. [Google Scholar]
  31. Newton ME. 1972. Sex ratio differences in Mnium hornum Wedw. and M. undulatum Sw. in relation to spore germination and vegetative regeneration. Annals of Botany 36: 163–178. [Google Scholar]
  32. Niklas KJ, Kutschera U. 2009. The evolution of the land plant life cycle. New Phytologist 185: 27–41. [DOI] [PubMed] [Google Scholar]
  33. Petersen KB, Burd M. 2017. Why did heterospory evolve?Biological Reviews 92: 1739–1754. [DOI] [PubMed] [Google Scholar]
  34. Pickup M, Barrett SCH. 2012. Reversal of height dimorphism promotes pollen and seed dispersal in a wind-pollinated dioecious plant. Biology Letters 8: 245–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. R Core Team 2016. R: A language and environment for statistical computing.Vienna: R Foundation for Statistical Computing; https://www.R-project.org/. [Google Scholar]
  36. Solomon BP. 1989. Size-dependent sex ratios in the monoecious, wind-pollinated annual, Xanthium strumarium. American Midland Naturalist 121: 209–218. [Google Scholar]
  37. Stark LR, McLetchie DN, Eppley SM. 2010. Sex ratios and the shy male hypothesis in the moss Bryum argenteum (Bryaceae). Bryologist 113: 788–797. [Google Scholar]
  38. Tryon AF, Lugardon B. 1991. Spores of the Pteridophyta: surface, wall structure, and diversity based on electron microscope studies.New York: Springer. [Google Scholar]
  39. Tryon RM, Tryon AF. 1982. Ferns and allied plants with special reference to tropical America.New York: Springer. [Google Scholar]
  40. Vitt DH. 1968. Sex determination in mosses. Michigan Botanist 7: 195−203. [Google Scholar]
  41. West SA. 2009. Sex allocation.Princeton: Princeton University Press. [Google Scholar]
  42. Weststrand S, Korall P. 2016. Phylogeny of Selaginellaceae: there is value in morphology after all!American Journal of Botany 103: 2136–2159. [DOI] [PubMed] [Google Scholar]
  43. Wyatt R, Anderson LE. 1984. Breeding systems in bryophytes. In: Dyer AF, Duckett JG eds. The experimental biology of bryophytes. London: Academic Press, 39–64. [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES