Dust seed production and dispersal in Swedish Pyroleae species
Veronika A. Johansson , Gregor M ü ller and Ove Eriksson
V. A. Johansson (veronika.johansson@su.se) and O. Eriksson, Dept of Ecology, Environment and Plant Sciences, Stockholm Univ., SE-106 91 Stockholm, Sweden. – G. M ü ller, Dept of Biology, Univ. of Konstanz, DE-78457 Konstanz, Germany.
Dust seeds are the smallest seeds in angiosperms weighing just about a few micrograms. Th ese seeds are characteristic of most orchids, and several studies have been performed on seed features, fecundity and dispersal of orchid dust seeds. In this study we examine seed features, seed production and seed dispersal in another plant group with dust seeds, the Pyroleae (Monotropoideae, Ericaceae), focusing on six species: Pyrola chlorantha , P. minor , P. rotundifolia , Chimaphila umbellata , Moneses unifl ora and Orthilia secunda . Seed production per capsule among these species was in the range between ca 1000 and 7800, and seed production per capsule bearing shoot was in the range between ca 7000 and 60 000. Combining our results with published information on pollen-ovule ratios suggests that these Pyroleae species have a generally effi cient pollination system. Th e most fecund species was P. minor , the only species among the investigated that is probably largely self-pollinating. Th e investigated Pyroleae species have a seed production comparable to the less fecund orchid species. We studied seed dispersal in the fi eld in one of the species, P. chlorantha . Despite the extremely small and potentially buoyant seeds, the vast majority of seeds are deposited close to the seed source, within a few meters. Further studies on the recruit- ment ecology of the investigated Pyroleae species are currently under way.
‘ Dust seeds ’ are extremely small and are often produced in vast numbers. Although most common among orchids (Arditti and Ghani 2000), dust seeds occur in at least 12 diff erent families, most likely representing at least 15 inde- pendent evolutionary origins (Eriksson and Kainulainen 2011). In many of these families, dust seeds have conver- gent features such as an elongated shape and large internal air spaces, possibly benefi cial to enhance buoyancy in air or water. Furthermore, in most dust seeds the embryo is undif- ferentiated. Arditti and Ghani (2000) compiled extensive information on seed size, seed shape and seed production in orchids. Th is information is not only important as descrip- tors of orchids, but serves also as an essential background to understand their dispersal features, in turn highly relevant for studies of population and conservation biology (Diez 2007, Jacquemyn et al. 2007, Swarts et al. 2010). In this paper we present data on seed features, seed production and dispersal from another group of plants with dust seeds, the tribe Pyroleae (Monotropoideae, Ericaceae) (Pyykk ö 1968, Takahashi 1993, Kron et al. 2002). Our study concerns six of the species found in Sweden: Pyrola chlorantha Sw., P. minor L., P. rotundifolia L., Chimaphila umbellata (L.) W. P. C. Barton, Moneses unifl ora L. A. Gray and Orthilia secunda (L.) House (Fig. 1, Table 1). One of these species, C. umbellata , is presently red-listed as endangered in Sweden, and two of the other, P. chlorantha and M. unifl ora are currently declining (Jonsell 2010).
A group of plants characterized by having dust seeds is mycoheterotrophic plants (Leake 1994), i.e. plants that
lack chlorophyll and hence the capability to photosynthesize.
Instead these plants use fungi as a source of organic car- bon. Fully mycoheterotrophic plants have received much attention with particular focus on their physiology and dependence on host fungi (Bidartondo and Bruns 2001, Bidartondo 2005, Leake and Cameron 2010). For exam- ple, it has been found that species in the genus Monotropa , closely related to the Pyroleae, are extremely specialized in the use of fungal hosts (Leake et al. 2004, Bidartondo and Bruns 2005). In contrast to full mycoheterotrophs, Pyroleae species are photosynthetic as adults, except for P. aphylla (a species not included in our study) which remains fully mycoheterotrophic as adult (Hynson and Bruns 2009).
However, Pyroleae species are considered to be not fully autotrophic, but instead ‘ mixotrophic ’ (Selosse and Roy 2009), i.e. partially mycoheterotrophic, implying that they partly use fungi as a source of organic carbon (Tedersoo et al. 2007, Matsuda et al. 2012). A possible exception is C. umbellata which in one study was found to be fully auto- trophic (Hynson et al. 2012). Despite being partially myco- heterotrophic as adults, or even autotrophic, all Pyroleae species are fully mycoheterotrophic as ‘ seedlings ’ , a below-ground stage that may last for several years (Whigham et al. 2008, Eriksson and Kainulainen 2011). Moreover, in contrast to Monotropa spp., it seems as the partially mycoheterotrophic species in the Pyroleae, as adults, are generalistic concern- ing their fungal partners (Tedersoo et al. 2007, Hashimoto et al. 2012). Some evidence suggests that the host use is more restricted during juvenile stages (Hashimoto et al. 2012).
209 Erschienen in: Nordic Journal of Botany ; 32 (2014), 2. - S. 209-214
Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-256272
Th e present paper is a part of an ongoing study on recruitment biology of Pyroleae species. In Johansson and Eriksson (2013) we showed that the study species diff er with regard to what limits their recruitment. Microsite availability was the main factor limiting recruitment in P. chlorantha , C. umbellata and O. secunda , whereas seed availability was the main limiting factor in P. minor . For P. rotundifolia and M. unifl ora, seed and microsite limita- tion contributed about equally. Apart from factors that con- cern the interactions between germinating seeds and fungi, two other factors contribute to patterns of recruitment
limitation, seed production and seed dispersal ability. Seed production is an important component of the realized seed dispersal potential, since increasing number of seeds from a seed source in itself improves dispersal (Eriksson and Jakobsson 1999). In addition, the capacity of single seeds to be carried by wind determines the seed shadow around a seed source (Tackenberg 2003). Several studies have been performed on seed dispersal in orchids with dust seeds (Murren and Ellison 1998, Jacquemyn et al. 2007, Jers á kov á and Malinov á 2007), but as far as we are aware of, none have been conducted on Pyroleae species.
Figure 1. SEM-photos of seeds of six Pyroleae species. (A) Chimaphila umbellata , (B) Moneses unifl ora , (C) Orthilia secunda , (D) Pyrola chlorantha , (E) Pyrola minor and (F) Pyrola rotundifolia .
Table 1. Information on seed size, fl ower/capsule number, plant height and habitat preferences of the six studied Pyroleae species.
Species Seed length
(mm) a Seed breadth
(mm) a Mumber of
fl owers/capsules Flower
height (cm) b Habitatb Chimaphila umbellata 0.55 0.10 3 – 5 c 10 – 25 Rare in old growth boreal forests
Moneses unifl ora 0.81 0.11 1 b 5 – 15 Fairly common in mesic-moist habitats
Orthilia secunda 0.53 0.11 5 – 10 c 10 – 20 Common in mesic-moist habitats
Pyrola chlorantha 0.70 0.17 2 – 6 b 10 – 25 Fairly common in dry-mesic boreal forests
Pyrola minor 0.54 0.17 5 – 10 c 5 – 20 Common in mesic-moist habitats
Pyrola rotundifolia 0.67 0.15 6 – 15 b 15 – 30 Fairly common in mesic-moist habitats
a Mean seed size values from Johansson and Eriksson (2013).
b Occurence in Sweden, Mossberg and Stenberg (2010).
c Pers. obs.
210
Th is study has the following objectives: 1) to examine seed production in the six Pyroleae species, and 2) to exam- ine the realized seed dispersal shadow around a seed source in the fi eld. Th e second objective was focused on one of the species, P. chlorantha , but considering the similarities in seed characteristics, we assumed that the results could be genera- lized to all the six study species.
Material and methods
Whole infructescences with mature seed capsules, hereafter referred to as shoots, were collected randomly from all study species from subjectively selected fi eld sites located on Ö land (56 ° 41 ’ N, 16 ° 36 ’ E) and from the counties of Uppland (60 ° 16N ’ , 18 ° 5 ’ E) and S ö dermanland (58 ° 53 ’ N, 17 ° 1 ’ E) in Sweden. Five sites were used for each study species from which fi ve random shoots were collected resulting in a total of 25 shoots per species, except for P. minor where four sites were used, resulting in 20 shoots. We removed all seeds from each collected shoot and they were dissolved in a solution of water and detergent to remove the surface tension. Seed numbers per shoot were counted by taking ten subsamples of the seed solution for each shoot. Depending on the concen- tration of seeds in the seed solution, between 25 and 100 μ l were sampled for easier handling. Th e total amount of seeds per shoot was then calculated by dividing the total volume of the seed solution with the volume of the samples taken and multiplied by the mean number of seeds counted for the ten samples. Seed number per capsule was counted by dividing the total number of seeds found with the number of capsules per shoot. Furthermore, aborted seeds containing no embryo were counted in the same way but these are not included in the total number of seeds produced.
Comparisons between species with regard to seed num- ber/capsule and seed number/shoot were done by analysis of variance (ANOVA) followed by Tukey ’ s post hoc test to determine which species diff er from each other.
Th e fi eld study on seed dispersal was conducted on two sites on Ö land, R ä lla (56 ° 46 ’ N, 16 ° 32 ’ E) and D ö rby (56 ° 37 ’ N, 16 ° 39 ’ E) and included one of the species, P. chlo- rantha . Both sites were on sandy podsol soils dominated by old growth pine ( Pinus sylvestris L.) with undergrowth mainly consisting of Vaccinium spp., Calluna vulgaris (L.) Hull, Linnaea borealis L., various grasses and mosses and some deciduous trees like Quercus robur (L.) and Betula pendula Roth. Th e method used was to place seed traps ( ‘ sticky tape seed traps ’ ) at various distances from a seed source. Th e seed source consisted of mature capsule bear- ing shoots of P. chlorantha , with a mean of 41 shoots per site (ranging from 33 to 56). Seed traps were placed around the center of each patch of P. chlorantha . Th e patch size was ca 0.5 m 2 and the mean shoot height was 20.8 cm (rang- ing from 20 to 22 cm). Th e surroundings were checked for additional fl ower stalks and if some were found these were collected and stuck down into the soft forest fl oor together with the main patch of shoots. Each sticky tape seed trap was constructed by wrapping masking tape around a petri dish (diameter 90 mm). Th e traps were placed along six transects from each seed source. Transects were laid in
diff erent directions, at an angle of 60 ° from each other. Th e seed traps were placed along each transect at the following distances 0 cm, i.e. adjacent to the seed source (n ⫽ 1); 10 cm (n ⫽ 1), 31 cm (n ⫽ 1), 56 cm (n ⫽ 1), 100 cm (n ⫽ 1), 177 cm (n ⫽ 2), 316 cm (n ⫽ 4) and 562 cm (n ⫽ 8). Th e n-values are the number of traps at each distance in each transect, thus with an increasing number of traps at the outermost positions. Th ere were a total of 114 seed traps per seed source. In all, fi ve replications of seed sources were used (one at R ä lla and four at D ö rby; the replications at D ö rby were located at a distance of ca 300 m from each other), thus making a total of 570 seed traps. Th e traps were collected after ten days in the fi eld, which was the approxi- mate time the traps remained sticky. Th e tape was then cut off and viable seeds were counted under a dissecting micro- scope. Th e actual number of available seeds from the seed source varied, but as the analysis focused on the proportion of seeds deposited at each distance, we regard this as a minor source of error. In the fi nal analysis, the fraction of seeds deposited in relation to distance from seed source was exam- ined by fi tting both an inverse power law function, and a negative exponential function. Th e results were similar, but the inverse power law function gave slightly better fi t, and is thus shown in the following results. Statistical analyses were carried out in the program R.
Results
Th e six study species have dust seeds of similar size, but somewhat varying in shape, M. unifl ora seeds being the most elongated (Fig. 1, Table 1). Th e species diff ered with regard to the potential number of capsules per shoot (Table 1). As M. unifl ora only produces one fl ower per shoot it is con- strained to produce only one capsule per shoot. Th e other fi ve species have multi-fl owered infl orescences and are therefore able to produce several capsules per shoot. Th ere was a signifi cant diff erence between species in seed produc- tion both per capsule (Fig. 2) and per shoot (Fig. 3). Th e average number of seed per capsule ranged between 1024 ( O. secunda ) and 7882 ( C. umbellata ). Th e actual seed pro- duction potential of the Pyroleae species is however higher for those species that have several capsules per shoot (all spe- cies except M. unifl ora ) (Fig. 3). Th e average seed produc- tion per shoot ranged from 7324 ( M. unifl ora ) to 60 487 ( P. minor ). Aborted seeds, containing no embryo, were not included in the analysis but ranged between 16% for C. umbellata to 30% for P. chlorantha .
Th e results concerning dispersal ability of Pyroleae dust seeds showed that the vast majority of seeds (82.5% of the total number of seeds deposited) were deposited close to the seed source, i.e. with a dispersal distance less than 1 m, and 95.7% were deposited within ca 5 m from the seed source (Fig. 4).
Discussion
Comparing the seed production of Pyroleae species with the compilation from orchids made by Arditti and Ghani 211
(Ericaceae): a fi eld manipulation and stable isotope approach.
– Oecologia 169: 307 – 317.
Jacquemyn, H. et al. 2007. A spatially explicit analysis of seedling recruitment in the terrestrial orchid Orchis purpurea . – New Phytol. 176: 448 – 459.
Jers á kov á , J. and Malinov á , T. 2007. Spatial aspects of seed disper- sal and sedling recruitment in orchids. – New Phytol. 176:
237 – 241.
Johansson, V. A. and Eriksson, O. 2013. Recruitment limitation, germination of dust seeds, and early development of underground seedlings in six Swedish Pyroleae species.
– Botany 91: 17 – 24.
Jonsell, L. R. 2010. Th e fl ora of Uppland (in Swedish: Upplands fl ora). – SBT-f ö rlaget, Uppsala.
Knudsen, J. T. and Olesen, J. M. 1993. Buzz-pollination and patterns in sexual traits in north European Pyrolaceae.
– Am. J. Bot. 80: 900 – 913.
Kron, K. A. et al. 2002. Phylogenetic classifi cation of Ericaceae:
molecular and morphological evidence. – Bot. Rev. 68:
335 – 423.
Leake, J. R. 1994. Th e biology of myco-heterotrophic ( ‘ saprophytic ’ ) plants. – New Phytol. 127: 171 – 216.
Leake, J. R. and Cameron, D. D. 2010. Physiological ecology of mycoheterotrophy. – New Phytol. 185: 601 – 605.
Leake, J. R. et al. 2004. Symbiotic germination and development of the myco-heterotroph Monotropa hypopitys in nature and its requirement for locally distributed Tricholoma spp.
– New Phytol. 163: 405 – 423.
Matsuda, Y. et al. 2012. Seasonal and environmental chanhes of mycorrhizal associations and heterogeneity levels in mix- otrophic Pyrola japonica (Ericaceae) growing under diff erent light environments. – Am. J. Bot. 99: 1177 – 1188.
McCormick, M. K. et al. 2012. Limitations on orchid recruitment:
not a simple picture. – Mol. Ecol. 21: 1511 – 1523.
Murren, C. J. and Ellison, A. M. 1998. Seed dispersal characteris- tics of Brassavola nodosa (Orchidaceae). – Am. J. Bot. 85:
675 – 680.
Pyykk ö , M. 1968. Embryological and anatomical studies of Finnish species of the Pyrolaceae. – Ann. Bot. Fenn. 5: 153 – 165.
Selosse, M.-A. and Roy, M. 2009. Green plants that feed on fungi: facts and questions about mixotrophy. – Trends Plant Sci. 14: 64 – 70.
Swarts, N. D. et al. 2010. Ecological specialization in mycorrhizal symbiosis leads to rarity in an endangered orchid. – Mol. Ecol.
19: 3226 – 3242.
Tackenberg, O. 2003. Modeling long-distance dispersal of plant diaspores by wind. – Ecol. Monogr. 73: 173 – 189.
Takahashi, H. 1993. Seed morphology and its systematic implications in Pyroloideae (Ericaceae). – Int. J. Plant Sci.
154: 175 – 186.
Tedersoo, L. et al. 2007. Parallel evolutionary paths to mycoheter- otrophy in understorey Ericaceae and Orchidaceae: ecological evidence for mixotrophy in Pyroleae. – Oecologia 151:
206 – 217.
Willson, M. F. 1993. Dispersal mode, seed shadows and coloniza- tion patterns. – Vegetatio 107/108: 261 – 280.
Whigham, D. F. et al. 2008. Specialized seedling strategies II:
orchids, bromeliads, carnivorous plants and parasites. – In:
Leck, M. A. et al. (eds), Seedling ecology and evolution.
Cambridge Univ. Press, pp. 79 – 100.
of potentially long-dispersing seeds may nevertheless be relatively high.
In this study we have contributed information on seed production and dispersal in a group of species for which these features are relatively poorly known. In order to gain a deeper insight into the population biology of Pyroleae spe- cies one approach is to integrate studies on seed dispersal and recruitment with identifi cation and mapping of host fungi, as has been done on some orchids (McCormick et al. 2012).
Putting our results in the context of previous studies, and comparing with the best studied group of plants with these kinds of seeds, the orchids, we can conclude the following:
1) concerning seed production, the investigated Pyroleae species line up among the comparatively less fecund orchids, 2) the variation in seed production among the study species is probably related to their pollination biology but is also dependent on the infl orescence structure of the shoot, i.e.
the number of capsules, and 3) the seed dispersal capacity is similar to those orchids that have been examined in this respect, i.e. most seeds are deposited close to the source.
Acknowledgements – Th is study was supported by a grant to OE from the Swedish Research Council.
References
Arditti, J. and Ghani, A. K. A. 2000. Numerical and physical properties of orchid seeds and their biological implications.
– New Phytol. 145: 367 – 421.
Bidartondo, M. I. 2005. Th e evolutionary ecology of myco-heter- otrophy. – New Phytol. 167: 335 – 352.
Bidartondo, M. I. and Bruns, T. D. 2001. Extreme specifi city in epiparasitic Monotropoideae (Ericaceae): widespread phylo- genetic and geographical structure. – Mol. Ecol. 10:
2285 – 2295.
Bidartondo, M. I. and Bruns, T. D. 2005. On the origins of extreme mycorrhizal specifi city in the Monotropoideae (Ericaceae): performance trade-off s during seed germination and seedling development. – Mol. Ecol. 14: 1549 – 1560.
Diez, J. M. 2007. Hierarchical patterns of symbiotic orchid germi- nation linked to adult proximity and environmental gradients.
– J. Ecol. 95: 159 – 170.
Eriksson, O. and Jakobsson, A. 1999. Recruitment trade-off s and the evolution of dispersal mechanisms in plants. – Evol. Ecol.
13: 411 – 423.
Eriksson, O. and Kainulainen, K. 2011. Th e evolutionary ecology of dust seeds. – Perspect. Plant Ecol. Evol. Syst. 13: 73 – 87.
Hashimoto, Y. et al. 2012. Mycoheterotrophic germination of Pyrola asarifolia dust seeds reveals convergences with germina- tion in orchids. – New Phytol. 195: 620 – 630.
Hynson, N. A. and Bruns, T. D. 2009. Evidence for a myco- heterotroph in the plant family Ericaceae that lacks mycorrhizal specifi city. – Proc. R. Soc. B 276: 4053 – 4059.
Hynson, N. A. et al. 2012. Measuring carbon gains from fungal networks in understory plants from the tribe Pyroleae
214
