Life forms and life strategies in Nanocyperion communities from the Netherlands Frisian Islands

Nanocyperion communities (s.l.) are considered here as “warp-and-woof” communities; the Nanocyperion components are described separately as synusiae. On the Netherlands Frisian Islands, four main synusiae have been recognized. Raunkiaer life form spectra show few differences between the communities. Life strategy spectra of the Nanocyperion synusiae, based on systems for phanerogams (modified after Bakker 1966) and bryophytes, yield the clearest patterns. A comparison of the ecology of the communities and an interpretation of the spectra in terms of avoidance of stress or competition suggest that inundations and standing crop of the communities are the main factors determining the distribution of the synusiae. Winter inundations overrule the influence of differences in productivity level, which becomes prominent in drier situations.


INTRODUCTION
In The Netherlands the Nanocyperion has been studied by Diemont et al. (1940). Since then, these communities have become very rare, like elsewhere in W. and Central Europe, though on the Netherlands Frisian Islands several of them are still fairly well represented.
The communitiesare found at the margins of lakes and ponds, on sand banks in rivulets, in dune valleys, in ditches in moist arable fields. They also occur in situations with a rather high level of "disturbance" such as drinking places of cattle, car tracks, paths, and grazed parts of salt-marshes. They belong to the "warp-and-woof communities" sensu Tuxen & Lohmeyer 1962;in such com-munities the warp component is outcompeted quickly or destroyed annually, but it maintains itself by shifting to other, regularly occurring, suitable microsites in the phytocoenosis (shuttle succession ("Pendelsukzession") sensu Tuxen & The communities of the alliance Nanocyperion flavescentis are characterized by a group of consistently very small, hapaxanthous plants and a considerable number of bryophytes.
Life form and life strategy systems are meant to reflect certain adaptations to the habitat of the plants (e.g., Raunkiaer 1904, 1934,Iversen 1936, MacArthur& Wilson 1967, Grime 1979. The choice of the system used determines to a large extent, which information is yielded by spectra showing the proportion of life form or life strategy categories in communities. The aim of this paper is to show that insight in the ecology of Nanocyperion communities may be gained by turning to life strategies focusing on life cycle period, regeneration and colonisation.
In the well-known system of Raunkiaer (1904Raunkiaer ( , 1934 life forms are distinguished mainly on the place of the meristematic tissues relative to the soil surface (Whittaker 1962). In this system, most of the Nanocyperion species are Therophytes, while the biennials belong to the Hemicryptophytes. In the r-K continuum of Mac Arthur & Wilson (1967), the hapaxanthous species all more or less have an r-strategy (emphasis on reproduction); an assessment of the reproductive effort per time unit might be worthwhile, but this has not yet been done. In the system of Grime (1974Grime ( , 1979 they all belong to the stress-tolerant ruderals. I prefer to use a refined subdivision of the hapaxanths made by Bakker (1966), which is based mainly on shortest life span, stage of development during winter, occurrence of seed dormancy, and vernalisation requirements. The last two characteristics are not known for all species involved and winter annuals s.l.
hardly occur in the communities studied here.
Bryophytes are not easily accomodated in any life form system presented so far. In most systems they all fall into one or two categories. In the Raunkiaer system as elaborated by Ellenberg & Müller-Dombois( 1967) for example, the bryophytes are divided into bryo-therophytes and bryo-chamaephytes. Recently During (1979) tried to distinguish some categories of bryophyte life strategies based on the colonizing species groups of the Lauwerszee area in The Netherlands (Joenje & During 1977).

MATERIAL AND METHODS
Also general ecological data were estimated by eye, such as a description of the surroundings, grazing and trampling intensity, salt influence, and soil profile, including (in 1971 only) pH at c. 1 cm and c. 8 cm depth.
The relevés were classified by hand sorting of the relevé table.
Here only a synoptical table is presented, giving frequency values and median cover values of the species for each community ; ifa species was foundin only one relevé of a community, the frequency class is left out. Species occurring in three relevés or less are not listed. On the basis of the relevés made all over the atlantic dune area (During 1973) and literature data (e.g., Philippi 1968, Pietsch 1973 (1967) in the bryo-chamaephytes, since the plants remain green and alive throughout the resting season. In the system of During (1979), the category of the colonists   Br y u m spec.

Communities
The releves of the Frisian Islands can be classified into four main communities, each of which is subdivided (table 3). Since a full treatment of the Nanocyperion communities of the dunes is not yet available, I will use here provisional names only. For each synusia, the subdivisions and ecology are briefly mentioned.
A. Centaurio-Saginetum moniliformis Ecology: on the upper parts of (frequently grazed) salt-marshes and in drier parts of young dune valleys; on sandy soil, layer of humus rather thin or absent, pH c.

6.
Subdivision: on the basis of the available material the synusia itself is not easily subdivided. However, the companion species clearly fall into two groups, by which the releves from the salt-marshes (Al) are readily separated from those of the dune valleys (A2). Nowadays virtually confined to anthropogeneous habitats such as skating lanes in the dunes or places in wet dune valleys where sods have been cut.
Subdivision: Within the synusia, a rather species-rich form (Cl) may be recognized which occurs on places with slightly higher pH and probably soil nutrient status (often with more human influence), and a species-poor form of pling, or cutting of sods. Salt influence is rare. The pH is slightly higher than in the foregoing subgroups. D4 and D5 are restricted to the uppermostparts of saltmarshes, on heavily grazed or trampled sites. In D4, Centaurium littorale is frequent, but flowers seldom. Ecologically, this subgroup is related to B2. D5 is an impoverished form of the synusia at slightly more salt-influenced, trodden places with higher pH.

Spectra
The Raunkiaer life form spectra (table 4) of the communities comprising bryophytes and phanerogams are rather similar, in spite of the large floristic differences between them. The hemicryptophytes are consistently dominant, and the bryo-chamaephytes are prominent, too. All communities show a rather large proportion of therophytes compared to other dune vegetations (cf. Westhoff 1947); in community C, this proportion of therophytes is greatest. C and D are characterized by a (small) contributionof bryo-therophytes. The "drier" communities A and D have slightly more geophytes and chamaephytes.
The phanerogam life strategy spectra (table 5) show slightly more differentiation. Community C stands out with a high proportion of ephemerals. The Table 4. Raunkiaer life form spectra of the communities, based on all species. Table 5. Phanerogam life strategy spectra of the communities, based on all phanerogam species. Table 6. Life strategy spectra of the bryophytes in the communities, based on all bryophyte species. All Nanocyperion hapaxanths remain very small and grow slowly. In view of their longer growing season, the annuals probably grow at an even slower rate than the ephemerals. The biennials and pauciennials should produce enough organic matter during the first growing season to survive at least one season of stress, and in the subsequent year produce a considerable amount of seeds (cf. Hart 1977). Besides, they generally grow to a slightly larger size in the first year than the shorted-lived species. Consequently, both ephemerals and biennials will grow faster than the annuals treated here. It may be expected, therefore, that annuals may occur in even less productive communities than the categories. In the present study, nutrient status of the soil and productivity of the communities have not been determined, but pH-values and comparison with literature data 494 H. J. DURING (Keener 1972, Jefferies 1977, Joenje& Wolff 1979) are circumstantialevidence.
When considering the combined strategy spectra of the Nanocyperion synusiae from these view-points, the following tendencies emerge.
Centaurio-Saginetum: in this synusia, the group of biennials and pauciennials dominates strongly. This is in accordance to its habitat: young, moderately productive, seldom inundated habitats in dune valleys or potentially highly productive salt-marshes (Ketner 1972) which are grazed or disturbed by winter storms; the high pH also points in this direction.
Bryum marratii synusia: characterized by a combination of short-lived shuttle species, ephemerals, summer annuals, and (seldom flowering) biennials. This diversity in life strategies is the more remarkable, since both the synusia and the community as a whole are rather poor in species. It is found in productive sites which are grazed intensively and may be inundated sometimes in winter. High pH and soil moisture availability also indicate a high productivity level.
Juncus mutabilis -Cicendia synusia: characterized by a high proportion of annuals and ephemerals. The sites are inundated for a long time during winter.
The fertility of the wet to moist substrate, slightly enriched by the inundations, is apparently high enough to allow the ephemerals to complete their life cycle. In the species-poor subgroup C2, with very low pH and little or no disturbance, annuals become more prominent.
Radiola linoides synusia: here, the annuals (Al, Ab, Cv partly) reach their highest values, though ephemerals still abound. The synusia is met with on drier, acid, soils with high humus contents and low soil fertility. The subgroup D4 resembles B2 both in habitat and strategy spectrum. The ecological differences between these two communities remain as yet unclear.
The general tendency is, that in moist to wet situations stress-avoiding strategies dominate (E2, A1), and B and Sb (if present) rarely reach the flowering c.q. sporulating stage. The main seasonal stress is here constituted by winterly inundations resulting in a short growing season, lower soil temperatures in spring, and bad aeration of the soil. In drier sites, the strategies Sb, Cv, B, and Pa become more prominent and standing crop differences are reflected more strongly in the spectra. Here the main seasonal stress is winter cold.
In conclusion, it may be stated that a comparison of plant communitieson the basis of life strategies of the species may lead to the generation of hypotheses on the environmental aspects and the functioning of these communities as well as on the main limiting factors determining these ecosystems. In a warp-and-woof community, it is likely that the "shuttle-synusia" and the perennials are determined by different ecological factors due to differences in rooting depth and responses to seasonally induced stresses (cf. Werger 1978). This is clearly reflected in their strategy spectra.
It may be worth while to establish a combined system in which colonizing strategies as well as life cycle period and avoidance type are reflected. The ecological value of the system would be optimized especially, if the other criteria used by Barker (1966) could be incorporated for both plant groups. These criteria, i.e. occurrence of seed dormancy and vernalisation requirements are 495 ON NANOCYPERION COMMUNITIES important adaptations to seasonal stress and do occur in some bryophyte groups (During 1979). Their relative importance and distribution over the bryophyte taxa remains as yet unknown.
Several of the communities treated here belong to the most species-rich vegetations of the dunes. This high diversity cannot be explained satisfactorily by the corridor model of Grime (1979), since the stands often adjoin vegetations with approximately the same productivity level but a much lower diversity. In both this corridor model and the hypothesis of Grubb (1977) concerning the partitioning of the regeneration niche, a suppression of dominancein the vegetation is important for the existence of a high diversity in certain vegetations. This point is elaborated in a more general hypothesis of species diversity presented by Huston (1979), which culminates in a two-dimensional model relating species diversity to frequency of population reduction and rate of competitive displacement. Beside the frequency of the population reduction, its distribution over the seasons and its predictability will also have influence on diversity (cf. Grubb 1977). Animal effects are very important in this respect (cf. Whittaker 1977).
Though it is not yet possible to place exactly specific communitieson the axes in the model, the general trends indicate that the Nanocyperion communitiesare in good accordance to Huston's predictions; the highest diversity occurs at rather low levels of growth rates and population reduction frequencies, and at higher growth rate levels a higher frequency of population reduction through grazing and seasonal climatic effects is necessary for the establishment of the Nanocyperion shuttle community which is partly responsible for the local raise in diversity (cf. tables 3, 7).
Careful field observations may yield more information on the applicability of Huston's model; a rigid testing of his hypothesis will require field experiments in which the effects are studied of harvesting of the production and trampling, both at different frequencies and intensities, upon diversity of species and of life strategies in the community.