GENERATIVE REPRODUCTION IN RUMEX CONFERTUS WILLD. POPULATIONS

Tomasz Stosik
Department of Botany and Ecology, University of Technology and Life Sciences, Bydgoszcz, Poland

ABSTRACT

Generative reproduction is crucial in fast and effective dispersal of synanthropic species. Ecological success of these species depends to a high degree on their fertility, seed dispersal adaptations, an ability to sprout over a long period of time and a high germination rate. Russian Dock (Rumex confertus) has a high seed production potential. Its diaspores are adapted to wind dispersal, have a high germination rate, and exhibit an ability to sprout over a long period of time. These features make the species a successful, invasive anthropophyte.

Key words: Rumex confertus, Russian Dock population, generative reproduction, biomass allocation, seed bank, seed mass, germination.

INTRODUCTION

Seed dispersal allows plants to invade new areas, even beyond their natural distribution range. High seed production increases the chances of successful development in a new place. Other essential factors are seed adaptation to long-distance dispersal, an ability to sprout over a long period of time and a high germination rate. These features enable Russian Dock, an expansive anthropophyte, to invade new areas.

The reproductive potential of an individual depends on its biomass [7], therefore due to its large size the species is privileged, all the more so because the biomass allocated to the generative reproduction is usually significantly correlated with the total biomass [11].

Usually, to illustrate the volume of generative reproduction, the potential fertility is determined, i.e. the maximum number of individuals that might be produced if the environment resistance was zero [16]. In higher plant populations the number of diaspores produced by a population is often used as a criterion for biotic potential, however, even a very high reproductive potential can be sometimes not fully exploited in local conditions [5].

Seeds produced by a plant may be accumulated in persistent seed bank, may be eaten or loose their viability. Sometimes only one seedling develops into adulthood from thousands or even hundreds of thousands seeds [13]. Composition of a seed bank varies within a growing season and during subsequent years due to germination, seed mortality and influx of new diaspores [8,10].

Only a small part of seeds of meadow perennials remains in soil after germination time; the remaining ones exhibit a low germination capacity that diminishes every year. That is why the influence of persistent seed bank on the population size of perennials is questionable [4,8,13].

Characterizing the generative reproduction of Rumex confertus could be crucial for understanding the invasiveness of the species. Therefore the aim of this work is determining the reproductive effort, fertility of female individuals, germination capacity and the amount and viability of diaspores remaining in the soil seed bank.

MATERIALS AND METHODS

The studies were performed on Rumex confertus populations developed in various plant communities of Fordon Valley.

One of the population analyzed is an extensively used, periodically flooded hay-growing meadow (population A). Two other populations represent ruderal communities: the first one is developed close to the river-bed (population B), the second one occurs on a side-space of an unpaved road surrounded by plough-lands (population C).

In these populations, in the years 2000-2002, overground parts of multi-module female genets were collected. Additionally, each September shoots bearing fruit from 15 clumps chosen at random were gathered. The seed sample collected in such a way made it possible to evaluate the germination capacity.

In order to characterize the soil seed bank of Rumex confertus, after the growing season in autumn 2001 and in spring 2002 thirty soil samples were collected in each object studied using a cylinder of diameter 20 mm. The samples were then split into two depth layers, 0-4 cm and 4-8 cm, according to commonly used methods [17,18,19,20]. The soil was sifted through a 0.25 mm sieve to recover the nuts. Fruits obtained in such a way were germinated to determine their viability.

The number of fruits per area unit was determined for each sample plot based on the density of shoots bearing fruit and fruit production by an average female shoot.

These data were used to assess the effectiveness of generative reproduction on permanent plots and in a separate experiment in which 1000 seeds of Rumex confertus were sown on 1 m² of uncovered soil.

Thousand seed mass and the so called actual fertility were calculated for each population studied. Additionally, reproductive effort [6,7,9,12] defined as a percentage of dry biomass taking part in reproductive processes was assessed as follows:

Germination rate, according to suggestions given in Ellis et al. [3] and Jehlik et al. [14], was estimated under laboratory conditions for three subsequent years. Whole nuts with perianth fragments were germinated. Experiments were conducted twice: immediately after harvest (in October) and in spring next year. As fruits were harvested earlier from meadow population, in this case the germination experiment was carried out in July, too. Fruits were not sterilized. Four samples of 50 nuts with perianth from each population were placed on sterilized filter paper in Petri dishes supplemented with 3 ml of distilled water. Germination experiment was run at 20°C and 12000 lx illumination. Every other day, germinated seeds were counted and removed, and water was replenished.

Analysis of variance was applied to assess if the differences between populations studied were significant.

RESULTS

Potential reproductive effort in particular populations ranged from 34.0 to 54,7% within the research period. It reached the highest value for population C in the year 2001. In turn, the highest value of actual reproductive effort (29.7%) was recorded for population A in the year 2000.

Loss in actual reproductive effort values compared with the potential reproductive effort was proportional in all populations and was, on the average, 41.8% for population A, 45% for population B and 43.6 for population C (Fig. 1).

Characteristics of an average individual can be complemented with the thousand seed mass. In this case whole fruits (nuts with perianth fragments) were analyzed. The index depended markedly on the average mass of generative female ramets and reached its lowest values for shoots from population A: 6.289 g in the year 2000, 4.799 g in 2001 and 6.915 g in 2002. For population B the thousand seed mass in three subsequent years was: 6.770 g, 5.812 g and 7.650 g. The seeds from population C were the most shapely and, on the average, reached the mass of 7.270 in the year 2000, 6.234 g in 2001 and 7.713 g in 2002 (Table 1).

The differences in the thousand seed mass between populations were statistically significant except for those between populations B and C in the year 2002 (Table 1). The lowest values within the research period were recorded in the year 2001. This is probably due to the weather in that year, especially a lower temperature during fruit development.

For all three cases, the highest number of fruits per m² was recorded in the year 2002. The number of fruits per m² in that year in populations A, B and C was: 3151 (the highest value for the year), 3046 and 2434, respectively.

In the next years distinct differences were noticed. They resulted from differentiation in the mean number of fruits per shoot and changes in density of generative female shoots. The number of fruits per shoot ranged from 669 in population A in 2001 to 1590 in population C in 2002. The density was up to 3 shoots per m² in population A, though sometimes, as in population C in the year 2002, the density of generative shoots was lower than 1 shoot per m² (Table 2).

The richest autumn seed bank was observed in population C, where in the 0-8 cm soil layer was 1592 seeds per m², on the average. However, the winter loss was also the highest in this population (-67%) and in spring the seed number per m² dropped to 530. Slightly lower loss was observed in population B, where the number of seeds per m² dropped from 742 in autumn to 424 in spring. The lowest winter loss (33.5 %) was reported for population C, where 636 fruits out of 955 survived the winter.

Diaspores were located mainly in the so called active seed bank, i.e. in 0-4 cm soil layer. In population A all fruits were found in this layer; no seeds were found in deeper layers, either in autumn, or in spring. In the case of seed bank in population B the situation was more complex. Almost 30% of diaspores were found below 4 cm depth. Analogously, in population C more than 13% of fruits were found below the active seed bank layer.

The winter loss was recorded mainly for the upper layer of the seed bank. The seed bank loss was 33.5% in population A and 20% in population B. The loss in population C was the highest (60.3%). The loss calculated by reference to overall number of seeds found in autumn for the 4-8 cm soil layer was 8.6% in population B, and 6.7% in population C (this does not apply to population A, where no seeds were found). However, the loss related to the number of seeds found in that layer in autumn ranged from 50 to 100% (Table 3).

As it turned out, quite a number of seeds were able to germinate. It is noteworthy that the viability of autumn seed bank was high: 67% of seeds started sprouting in sample plots A and C. The figure was slightly smaller for population B, where 43% of alive seeds were found. In the spring seed bank 50% of seeds retained their viability only in population A; in the other plots no alive seeds were found (Fig. 2).

Favorable conditions of the year 2002 promoted the generative development of Russian Dock. Generative female shoots produced more fruits that in the previous two years. The number of seeds that entered the seed bank varied among populations. In population C over 65% of seeds entered the seed bank analyzed. The figures for the other populations (A and B) were about half smaller: 30.3% and 24.4%, respectively. In sample plots, because of dense vegetation cover, seedlings rarely appeared, in spite of quite high fruit production. In natural conditions, in population A only 17 seedlings (0.13 · m^-2) were found in the year 2003. No seedlings at all were observed in other areas studied. Therefore, studies of seed bank were conducted also in areas with vegetation cover removed where 1000 seeds were sown additionally. The existing seed bank enriched that way with additional seeds still produced very low number of seedlings. This shows that the efficiency of generative reproduction is very low, at least under conditions in the phytocoenoses studied.

The most diversified and, at the same time, the lowest values of germination dynamics were found for population A. In this case most seed sprouted in autumn: 50%, on the average. In May next year the figure was 43%, and it was only 12% immediately after harvest. Seeds from ruderal populations exhibited much higher germination capacity. In population B in autumn and spring the figure was ca. 80%. Slightly less seeds sprouted in population C: 69% on the average in autumn and spring (Fig. 3).

DISCUSSION

Potential reproductive effort of Rumex confertus during the study period ranged from 34 to 51%. Actual reproductive effort reached its maximum value of 29.7% in population A in the year 2000.

Klimeš [15] determined the actual reproductive effort of Rumex obtusifolius L. according to the density of shoots. It amounted to 48.7% for a low density and decreased with the increase of density to 27.5% for 121 generative shoots per m². These figures are similar, but quite high considering the ecological forms of the two species.

The nut size of Rumex confertus also depended on habitat conditions. Thousand seed (nut) mass varied among populations and years. The density, habitat fertility, weather conditions and anthropogenic influences could be the differentiating factors here.

It is sometimes assumed that the average seed weight is more or less constant within a species and does not fluctuate among individuals from extremely different habitats. There are also reports that seed size varies among populations of a species, which can be related to the density, habitat variability or water accessibility [13]. Populations of Rumex confertus differ in many respects which results in different seed mass, among others.

During the study period the number of diaspores per m² in particular populations of Rumex confertus ranged from 905 to 3151. The highest fruit production was observed in population A in the year 2002. This explicitly resulted from a high density, as the average number of seeds per generative shoot in meadow population (A) did not exceed 1047. In population C (on a side-space of an unpaved road) the figure was 1590, and in population B (close to the river-bed) it was 1725. Almazova and Rabotnov [1] reported at least 1550 fruits per shoot of Rumex confertus populations within its natural distribution range.

According to the results of this work, from 700 to 1500 (and more) nutlets contributed to the seed bank. The nutlets are placed mainly in the upper soil layer (up to 4 cm). By winter the number of nutlets decreased by 30-70%, and the germination capacity dropped drastically (up to zero). However, Almazova and Rabotnov [1] showed that Rumex confertus generative propagules constituted only a little percentage of the whole seed bank. Other authors [4,8,13] seem to share this view. They claim that only a small amount of seeds of meadow species remains in the soil after germination, and germination capacity of the seeds remaining in the soil decreases each year. Therefore the influence of permanent seed bank on the population size of perennials is questionable. A relatively rich seed bank was described in this work, which may be ascribed to the early sample acquisition (before germination).

Seed parameters for Rumex confertus populations analyzed in this work were comparable to those obtained by other investigators. The highest germination capacity (80%) exhibited seeds from population B. The figure was slightly lower for population C (69%). The lowest germination capacity was recorded for meadow population: the figure ranged from 12% immediately after harvest to 50% in autumn. For all cases, diaspores started sprouting on the fourth day of the experiment. Germination dynamics was slightly different. Populations B and C displayed a high germination rate up to eight days, while in population A the process of germination lasted up to 12 days.

Evaluation of the germination capacity under laboratory conditions revealed that the diaspores of Rumex confertus could sprout in light and in darkness. However, light enhances the germination process significantly. According to Jehlik et al. [14] the germination capacity determined for Rumex confertus populations growing in the Czech Republic was 80%. A slightly higher figure is reported by Almazova and Rabotnov [1] who studied seeds of this species growing in Oka Valley (Russia). They reported that 90% of diaspores could sprout in one to three years after harvest. They also determined the germination capacity according to harvest time. It turned out that the germination rate of fruits harvested at the beginning of July was only 4%, but the figure increased with time to 69%. Germination rate determined for seeds from Rumex confertus populations growing in the vicinity of Siedlce was around 85% on the soil and up to 100% on filter paper [2].

Although seed germination rate under near-optimal conditions is high, it does not hold true for plant communities. Seedlings at younger development stages cannot be easily observed in nature, as they probably require a special mosaic of habitat conditions [2].

Russian Dock is a species that allocates a considerable amount of its biomass to generative reproduction, though its vegetative propagation is effective, too. As the seed productivity of Rumex confertus is high, generative reproduction of the species is important. In meadow the seeds, though not uniform in size, have high germination capacity, even if they are not fully mature by hay harvest time. Nuts are enclosed in dried perianth segments that facilitate long-distance dispersal. They germinate under natural conditions as soon as the habitat factors permit, therefore they do not form a persistent seed bank.

CONCLUSIONS

The expansion potential of Rumex confertus is affected by:

• effective vegetative reproduction.

• high production of seeds of high germinability

• capacity of diasporas for effective mobility.

REFERENCES

1. Almazova D. I., Rabotnov T. A., 1953. K biologii shchavelya konskogo (Rumex confertus Willd.). I. Semennoe razmnozhenie shchavelya konskogo [Biology of the Russian Dock (Rumex confertus Willd.) I. Generative reproduction of the Russian Dock]. Byulleten΄ M. O-va Isp. Prirody. Otd. Biologii 58(6), 47-54 [in Russian].

2. Ćwikliński E. 1990., Rumex confertus Willd. na terenach kolejowych województw siedleckiego i białopodlaskiego. [Rumex confertus Willd. on by railway grounds of the Siedlce and Biała Podlaska Districts (Central-Eastern Poland)]. Zeszyty Naukowe Wyższej Szkoły Rolniczo-Pedagogicznej w Siedlcach. Nauki Przyrodnicze 24, 187-199 [in Polish].

3. Ellis R. H., Hong T. D., Roberts E. H., 1985. Handbook of seed technology for genebanks. Compendium of specific germination information and test recommendations. International Board for Plant Genetic Resources. Rome. volume II, 546-548.

4. Eriksson Å., Eriksson O. 1997. Seedling recruitment in seminatural pastures: the efects of disturbance, seed size, phenology and seed bank. Nordic J. Bot. 17, 469-482.

5. Falińska K., 1977. Strategia i taktyka reprodukcyjna populacji roœlinnych. [Reproduction strategy and tactics of plant populations]. Wiad. ekol. 23(3), 230-257 [in Polish].

6. Falińska K., 1981. Eksperymentalne badania biologii populacji wieloletnich roslin zielnych [Experimental studies of perennial plants biology]. Wiad. bot. 25(3), 209-230 [in Polish].

7. Falińska K., 1990. Osobnik, populacja, fitocenoza [Individual, population, phytocoenosis]. PWN. Warszawa, 310.

8. Falińska K., 1994. Bank nasion w glebie a dynamika roslinnosci [The soil seed bank in vegetation dynamics]. Wiad. bot. 38(1-2), 35-46 [in Polish].

9. Falińska K., 1996. Ekologia roslin. Podstawy teoretyczne, populacja, zbiorowisko, procesy [Plant Ecology. Theoretical foundations, population, community, processes]. PWN. Warszawa, 453 [in Polish].

10. Falińska K., 1998. Seed bank patterns. (in:) Plant population biology and vegetation processes. Falińska K. (ed.) Polish Academy of Sciences. Kraków, 230-243.

11. Hara T., Kawano S., Nagai Y. 1988. Optimal Reproductive Strategy of Plant with Special Reference to the Modes of Reproductive Resource Allocation. Pl. Sp. Bol. 3, 43-59.

12. Harper J. L., Ogden J. 1970. The reproductive strategy of higher plants. I. The concept of strategy with special reference to Senecio vulgaris L. J. Ecol. 58, 681-698.

13. Jankowska-Błaszczuk M., 1996. Ekologiczne znaczenie wielkoœci nasion [Ecological aspect of seed size]. Wiad. bot. 40(2), 21-28 [in Polish].

14. Jehlik V., Sádlo J., Dostálek J., Jarolimova V., Klimeš L. 2001. Chorology and ecology of Rumex confertus Willd. in the Czech Republik. Botanica Lithuanica. 7(3), 235-244.

15. Klimeš L. 1996. Population ecology of Rumex obtusifolius. (in:) Prach K., Jenik J., Large A.R.G. (eds.) Floodplain Ecology and Management. Amsterdam. 155-179.

16. Odum E. P., 1982. Podstawy ekologii [Bases of ecology]. PWRiL. Warszawa, 661 [in Polish].

17. Pirożnikow E., 1983. Seed bank in the soil of stabilized ecosystem of a deciduous forest (Tilio-Carpinetum) in the Białowieża National Park. Ekol. Pol. 31(l), 145-172.

18. Thompson K., Grime J. P., 1979. Seasonal variation in the seed banks of herbaceous in ten contrasting habitats. J. Ecol. 67, 893-921.

19. Wesołowski M., 1979. Skład gatunkowy i liczba nasion chwastów w glebach południowo-wschodniej Polski. I. Gleby terenów nizinnych [Type composition and number of weed seeds in the soils of the South-East Poland. Part I. Soils of the Lowland Region]. Annls Univ. Mariae Curie-Skłodowska. E. 34(3), 23-37 [in Polish].

20. Wesołowski M., 1986. Zapas nasion chwastów w niektórych glebach południowo-wschodniej i œrodkowej Polski. III Czarnoziemy, mady i rędziny [The reserve of weed seeds in certain soils of South-Eastern and Central Poland. Part III. Chernozem, Silt and Rendzina]. Annls. Univ. Mariae Curie-Skłodowska. E. 41(5), 45-58. [in Polish].

21. Wesołowski M., Jędruszczak M., 1987. Nasiona chwastów w kompleksach przydatnosci rolniczej gleb w makroregionach południowo-wschodniej i srodkowej Polski [Weed seeds in the soil complexes of South-Eastern and Central macroregions of Poland]. Annls. Univ. Mariae Curie-Skłodowka. E. 42(3), 73-81. [in Polish].

Accepted for print: 24.10.2007

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.