‘One may be sceptical of the idealisation of the natural state. In the forest as elsewhere, virginity is not accompanied by excessive virtue. The harmony-of-nature concept romanticises a status-quo in which man plays no part, and doubtless it holds more charm for those who lack untouched vegetation than for those of us who can go out and see (frequently with dismay) what theory assures us ought to be a beautifully balanced condition, healthy, harmonious and in every way ideal.’

— J. S. Rowe (1961)

Introduction

The ecological literature, in general, assures that as the conceptualised state definable as climax, mature, stable or steady-state is attained, there is progressive increase in community structure and diversity, massiveness or stature, productivity, soil maturity, relative stability, maintenance and regulations of populations etc. This review examines the multitude of terms and concepts pertaining to this state.

Stability

In ecological dynamics stability has both a very broad general meaning and various specific meanings. In a general sense, stability is used adjectivally (‘stable’) to describe any state of equilibrium without specifically defining its dynamic nature. As relative concepts, stability and instability form some of the basic elements of ecological dynamics (van Leewen 1966) but lack the precision to describe the nature of a dynamic equilibrium.

Stability means quite different things in different disciplines and it is essential for an ecologist to define his terms (Preston 1969, Whittaker 1957). Temporary stability is all that biological systems may ever attain. Absolute stability would require cessation of astronomical or cosmic causes of fluctuation (Preston). To define any state of stability in the vegetation/soil system we must qualify this stability in relation to the functional relationships of the environmental factors. Any property of the ecosystem is a product of all factors operating together. Stability cannot be based on time alone (Greig-Smith 1964) any more than it can be based on other single factors of vegetation/soil system formation. If stability is defined in terms of the human time scale, it can become very abstract in terms of actual change within the vegetation/soil system, particularly in page 133 forest. This same state of stability may in terms of geological time, be a fleeting part of a succession (Jenny 1941).

Any definition of stability in terms of species populations is very dependent on their growth-form and the physiognomy of the vegetation to which they contribute. Forest vegetation which remains the same for only one generation is ecologically no more stable than ephemeral vegetation which changes after one generation (Greig-Smith 1964). Time-lag after change involving environmental stress, for example rapid temperature decline, will vary according to the growth-form of the dominants; the period between initial climatically induced non-regeneration and death of all emergent rimu in a tall rimu-rata/tawa forest will be in the order of hundreds of years (Wardle 1964). In herbfield or grassland, the same degree of climatic change will produce considerable changes in canopy composition in only a few years. Ease of entry of new species and re-attainment of a self-replacing state will vary accordingly. Holloway (1954) has described the situation in the lowland forests of the South Island where the inability of apparently long-established stable vegetation to regenerate is thought to be a consequence of recent climatic change. The climatic change has, according to Holloway, induced widespread physiological maladjustment in podocarp populations. Conditions within the vegetation/soil system produce a greater degree of stability than those external to the system (Rode 1947, Greig-Smith 1964). Any external change will be lessened by feedback effects within the vegetation/soil system maintaining stability, for example, forest regeneration to the same species and structure after a windfall.

In a very specific definition of stability, Preston (1969) considered a species as stable in time if, over a large number of years, its fluctuations in numbers correspond to a lognormal distribution. A stable ecosystem, he defines as one in which all niches are fully occupied by appropriate species. Lewontin (1969) approached ecological stability from mathematical theory using the concept of the vector field in n-dimensional-space. Lewontin compared the change through time of a population or community with the changes in the position of a particle in this n-dimensional-space, and considered deviations from equilibrium as a measure of stability. The position of the particle in Lewontin's definition of stability is never stationary, because of perturbations. Goodall (1962) and Whittaker (1969) discussed a similar analogy. Whittaker considered a species as a moving point in the n-dimensional-space. All points together at any one time form a community.

Lewontin (1969) also discussed stability in terms of energy dissipation. In a physical system, the greatest stability is achieved at the state of minimal potential energy. A similar solution for an ecological system has not yet been formulated. To begin to understand stability it is essential to work with systems that are in relative dis-equilibrium (Lewontin 1969).

The stability principle of Odum (1959) referred specifically to dispersal of energy in a closed natural system, such as a tarn. In such a system, a state of stability is reached and maintained by self-regulatory mechanisms such as mortality-increase consequent on density-increase of organisms. Sanders (1969) formed the stability-time hypothesis to explain the progressively changing patterns of diversity under increasing physical stress.

The limitations of the concept of stability in ecological dynamics were recognised by Bray (1958) and Margalef (1969). Margalef stated that a natural system is stable only if, when changed from a steady-state, it develops forces that tend to restore it to its original condition.

‘If the capacity to return from a different (unstable) state is never needed or realised, the system cannot prove itself stable, but neither is it possible to declare it unstable.’

With this reasoning, Margalef concluded that the whole notion of stability appears hopelessly confused.

Climax

The literature on the climax concept was extensively reviewed by Whittaker (1953, 1957) and Selleck (1960). To interpret vegetation development one has, in the words of Bukovsky (1935) ‘to overcome both the static fossilized conception of European phytosociologists and the somewhat fatalistic theory of succession of American scientists’.

In a review of the literature, two main approaches to climax definition can be distinguished. Many of the theoretical and descriptive considerations of climax vegetation have been based on the floristics and physiognomy of vegetation and its relation to climate (Whittaker 1953). The other approach centres on the problems of population and productivity in relation to all environmental factors. Both approaches are largely restricted to vegetation.

The basic vegetational criteria of the climax concept preclude its relevance to other components of the whole ecosystem.

The climax concept, whilst burdened by vagueness and variety of usage, is useful in that basically it contrasts stages of relative stability with stages of relative instability (succession) (Braun-Blanquet 1932, Clements 1936, Weaver and Clements 1938); but implicit in its definition is the notion of terminality and ultimate control by climate (Oosting 1953, Odum 1959). Clements (1936) and most phytosociologists considered the climax community to be a highly organised association.

Data on vegetation pattern (Kershaw 1957) in relation to the climax state in a variety of ecosystems, are strongly contrary to the Clementsian principle that a climax community is a complexly organised association. Rather, the individualistic concept (Gleason page 135 1939) is a more satisfactory alternative. Greig-Smith (1952, 1961) in analyses of pattern in regenerating tropical forest and stabilising sand-dunes, and Kershaw (1962) in similar work in colonising water-communities, demonstrated decreasing species-association toward the most stable stands of vegetation.

Whittaker (1953) compromised between the individualistic ideas of continua in all dimensions; the basis of gradient analysis (Whittaker 1951, 1967), and the classical organismal usage of Clements (1936) whilst retaining the term ‘climax’. He formulated the ‘climax pattern’ hypothesis:—

‘Climax vegetation is a pattern of populations corresponding to the pattern of environmental gradients and more or less diverse according to the diversity of environments and kinds of population in the pattern.’

From this Whittaker created the more abstract prevailing climax:—

‘The prevailing climax is the average population of self-maintaining stands on the type of site as defined and limited.’ This term was designed as a practical working definition in the same sense as the ‘monoclimax’ (Clements 1904, 1916, 1936) and the ‘polyclimax’ (Tansley 1911).

Whittaker (1953) critically assessed these three main climax hypotheses in relation to the field situation. The ‘monoclimax’ is the most abstract, in that it is largely speculative, describing all vegetation patterns in terms of associations leading to an ultimate climatic climax. The ‘polyclimax’ achieves greater effectiveness in relating vegetation to environment, though retaining the static assumptions of the community unit and terminality. The more sophisticated ‘prevailing climax’ avoids these assumptions, and allows for continuity of spatial change in all vegetation components and the environment.

Whittaker's climax pattern and prevailing climax hypotheses were criticised by Selleck (1960) who reasoned that a comparatively minor constituent of the community may be sufficiently common in the under-storey to become a dominant in the next generation. In effect, this is similar to the criticism that has been levelled at the ‘gradient’ approach to vegetation (Whittaker 1967) by Daubenmire (in Goff 1968). He felt that gradient analysis does not emphasise the dynamic relations of vegetation. Very recent work has been done in this aspect of vegetation dynamics by Goff (1968) and Goff and Zedler (1968). For example, size class data collected from one stand, at one point in time, can reveal much about the history, rapidity of change and probable future of the vegetation/soil system. (Kittredge 1938, Stearns 1950, Goff 1964, 1968.)

The climax concept must be discussed in the light of its development which was mainly by the different phytosociological schools of thought. These schools emerged from a floristic framework, with page 136 varying emphasis on the recognition of vegetation continuity and discontinuity (Becking 1968).

In general, the phytosociological approach has made too little reference to the holistic nature (Bray 1958) of an ecosystem, or even its vegetation component in the formation of the climax concept (Whittaker 1953). The productivity, biomass and structural parameters of terrestrial ecosystems has been only infrequently studied in relation to the climax concept. This is, to some extent, attributable to a lack of field theory (Bray 1958). Sampling and analysis is usually restricted to single factor environmental spatial gradients (Whittaker 1967) with a consequent neglect of the time-gradient as well as vegetation structure, productivity etc.

Frequently, when these parameters have been applied to climax theory in vegetation studies, the assumption has been that the stand of greatest productivity, or greatest stature is the most stable, i.e. it is then referred to as the climax (Del Villar 1929). Scott (1969) suggested that productivity and structure are quite unrelated. He considered productivity to be a function of dominance while stability is a function of diversity.

Most phytosociological schools have paid relatively little attention to the dynamic equilibrium concept and temporal change in vegetation (Becking 1968). Their techniques were developed for the description, analysis and synthesis of ‘static’ communities. This static descriptive approach to climax theory was a function of descriptive and quantitative floristic data being collected at only one point in time, and automatically adapted, a priori to the preconceived notions (Rowe 1961) of terminality, stability, finality which defined the climax.

In terms of species populations, the climax, like the phytosociological ‘association’, is a concept which does not stand up to critical examination (Whittaker 1953). Whittaker himself (1953) stated that his definition of climax vegetation, as a self-maintaining system of interacting populations was far from absolute. His definition emerged from his ideas on species individuality and time-space continua, but neglects the whole ecosystem characteristics of nutrient availability, productivity, structure etc., that should be the basis for any concept of ecological dynamics. Whittaker (1953) suggested that the climax condition may be better defined by the term ‘steady-state’ defined as ‘vegetation of changing and developing character, incompletely stabilised, with its balances gradually shifting’. Within this definition the usual distinctions between climax and succession based on relative stability and directional change, break down.

Atkinson et al (1968) following Whittaker (1953) also defined the climax in terms of attainment of a steady-state condition. They considered that true equilibrium is not, in fact, reached as in a closed chemical system. Vegetation is part of the ecosystem which is an open system constantly subject to gains and losses.

Although his reasoning was tenuous Becking (1968) has been one of the few ecologists to conceptualise the climax state as a basic feature of the whole ecosystem. Becking equated the dynamic ecosystem with a cybernetic system. He suggested that climax communities have a maximum of entropy, but was using ‘entropy’ incorrectly, to mean stored energy. Whittaker (pers. com. to Becking) strongly disagreed with the latter's use of entropy. ‘Entropy’ (Barrow 1966) is a state of disorder in the energy component of a system. The most stable state is one of lowest entropy, or energy loss. In Whittaker's terms, climax communities probably represent maximum negentropy. ‘Negentropy’ is free and available energy such as that generated by photosynthesis. At any stage subsequent to a state of stability in a system, there is an increase of entropy through dissipation of energy and loss of organisation.

Whittaker also disagreed with Becking's (1968) contention that without frequent periods of disturbance, a climax community cannot maintain itself and will ‘degrage’. Becking stated that is is necessary to counter energy accumulations in the ecosystem by processes generating disorder and that this must be considered in the conservation of climax communities.

Maturity

Whereas the concept of ‘climax’ pertains, in the main, to vegetation, ‘maturity’ has a similar meaning, but pertains to the soil part of the ecosystem.

The concept of soil maturity has developed from being strictly morphological, referring to descriptive field criteria, to being dynamic and based on soil forming processes (Jenny 1941). A mature soil is normally defined as being in equilibrium with the environment (Jenny 1941, Lutz and Chandler 1947, Joffe 1949) although it is often difficult to establish just what is meant by the environment of a soil. Early pedologists denoted differences in degree of profile development as young, immature, mature and senile. These terms are still in use, but only for descriptive and comparative purposes (Taylor and Pohlen 1962). Morphological and dynamic maturity may or may not be coincident (Jenny 1941). In terms of a soil property — time function, soil maturity (Jenny 1941) is reached when the curve becomes and remains flat, indicating zero change. Jenny, discussing the concept of soil maturity, stated that not all soil components approach maturity at the same rate or simultaneously. This was endorsed by Stevens 1963, 1968) and other chronosequence workers. According to Jenny, at the particular conditions of dynamic equilibrium: steady-state, the change (ds) in some soil property during a time interval (dt) will be represented by— ds/dt = 0 in which case time can be neglected since it has no further effect. page 138 Jenny commented that very slow rates of change may be mistaken on the human time scale for apparent steady-state.

Lutz and Chandler (1949) believed that a soil can attain a state of ‘near equilibrium’ with its environment. An important consideration is the initial state of the substratum.

Wilde (1946) and Joffe (1949) very inadequately denned a mature soil as having a characteristic profile reflecting the influences of environmental factors. Wilde considered soil to be a continually changing medium, and vegetation and soil as integrated parts of the same system, but believed that this system attains a climax state. Dansereau (1957) equated soil maturity with zonality and used the term pedoclimax. Maturity, he said, is attained under the joint impress of climate and vegetation, and mature soils will also support a mature vegetation. This is the traditional Russian viewpoint established by the School of Dokuchaev (Rode 1947). It is essentially similar to the ecological viewpoints of Clements (1936) and Braun-Blanquet (1932).

Rode (1947) in his monograph on soil evolution did not accept the widely held Russian concept of soil maturity nor the analagous concept of climax as presented by Clements. He objected to the way in which these concepts denote a terminal stable state. He did, however, accept their terminology denoting a relatively stable state, following a state of continuous instability, in the vegetation and soil.

Webster (1968) rigorously defended the idea that soil is a polythetic (Sokal and Sneath 1965) system in which any parameter is, and is part of a continuum. Webster stated that soils, as such, are very rarely in equilibrium with the environment. Hence they cannot be classified by any ‘natural’ or genetic system of classification (e.g., Taylor and Pohlen 1962, U.S.D.A. 1960) which relies on selected ‘genetically significant properties’.

Margalef (1969) used maturity in quite a different sense to define the ratio of biomass to productivity in an ecosystem. This was a zoo-ecological definition. Whittaker (1962, 1963) and Whittaker and Woodwell (1968) discussed similar biomass/productivity relationships in a structural analysis of forest but did not employ the term ‘maturity’.

Steady-State

In studies involving vegetation/soil system development, the parameter of time and the associated concept of adjustment become chief considerations. The concept of adjustment is termed ‘dynamic equilibrium’ or quasi-equilibrium (Leopold and Langbein (1962). This concept includes the condition of ‘steady-state’ which refers to the tendency for a constant state to develop (Abrahams 1968).

‘Steady state’ is a concept of General Systems Theory. The term has been used in thermo-dynamics (Defay 1929), biology (Bertalanffy 1932), and pedology (Nikiforoff 1959) to define certain conditions page 139 of an open system, where the emphasis is on the continuous interaction of process and the system components.

Bertalanffy (1950) defined steady state as when —

‘…the system remains constant as a whole and in its phases; though there is a continuous flow of the component materials’.

Nikiforoff (1959) equated steady-state with dynamic equilibrium, distinguishing the latter from what he termed ‘static equilibrium’. Static equilibrium it is presumed, would occur when all systems come to a standstill. This is the equivalent of the ‘climax’ in the sense of Clements (1936).

In vegetation, and soils (Lavkulich 1969) there is no static equilibrium. Lavkulich was of the opinion that all ‘natural’ processes are more or less irreversible. In this sense, he considered that use of the concept of dynamic equilibrium or any kind of equilibrium was ‘erroneous’ but that ‘steady state’ was a suitable term. ‘Steady state’ itself, he defined as a state attained when the properties of the system do not change with time, but when there is an irreversible flow of materials and energy through the system (Barrow 1966). However, in an open system such as an ecosystem, the concept of dynamic equilibrium, signifying adjustment (Abrahams 1968, Margalef 1969) does not infer reversibility back to a prior state.

Ecologically, ‘steady state’ can be defined as a temporary state of dynamic equilibrium in an open system. Whittaker (1953) attributed its original usage to Russian ecologists, and used the term himself as a substitute for ‘stability’ with reference to species populations and structural-functional variables such as productivity. Whittaker referred to steady-state as the levelling-off of the time-distribution of any parameter such as productivity.

The distinction from ‘climax’ can be immediately seen. The vegetation component of the ecosystem considered to be climax in the sense that it is self-maintaining (Whittaker 1953, 1957) may appear to be constant with time. However, within the whole ecosystem, there is simultaneously a continual imbalance between input and output of materials (Ovington 1962, Miller 1963, Lavkulich 1969). This imbalance is a function of loss of nutrients from the soil by leaching in excess of nutrient release by weathering (Viro 1953, Miller 1963) particularly in podzols (Ponomoreva et al 1968), loss by erosion, lateral and surface runoff of water, uptake by plants, input by rainfall and other atmospheric processes (Miller 1963) all of which contribute to the open nature of the ecosystem. Bray (1958) considered that the attainment of steady-state is characterised by increasingly closed nutrient cycles, that are somewhat independent of matter input. For example, in a steady state forest, the amount of CO² removed from the air by photosynthesis is equal to the amount released by respiration (c.f. Denbigh 1951). At the level of the individual organism, however, the nutrient cycles must remain open to matter and energy exchange to maintain life. This is the page 140 basis of the individualistic concept of vegetation (Gleason 1939) and species tolerance with time (Becking 1968).

Odum (1959) was one of the few authors who specifically defined the steady-state for an ecological system. According to Odum, the steady-state is always temporary, passing to another steady-state because of environmental changes. The steady-state can, in fact, be altered by conditions initiated by both internal processes surpassing critical levels and the operation of external stresses. It is maintained only so long as external conditions remain unchanged (Abrahams 1968).

Bray (1958) and Whittaker (pers. comm. to Becking 1968) stated that a steady-state ecosystem will normally have —

Bray stated that in the development of a ‘community’ to steady-state, there is an overall increase in order. This is possible because the simultaneous negative change in ‘entropy’ is greater than the positive change occurring through irreversible processes, such as soil leaching. At steady-state, entropy is at a minimum. According to Bray (1958) the mechanism of community order-increase does not depend solely on an increased amount of photosynthesis. Rather, as the community reaches steady-state, it is based on the capability of re-utilisation efficiency to reduce the relative loss through production of non-available energy forms. Sears (1959) considered that the efficiency is a function of the number and kinds of organisms and their ability to create new energy niches and raise the number and complexity of energy exchange pathways (cf. Preston 1969). It is possible (Bray 1958) that in some steady-state ecosystems this increased utilisation efficiency more than compensates for a slightly decreased photosynthetic productivity as noted by del Villar (1953) and Whittaker (1953). Bray emphasises that any increase in order is dependent on the ability of individual species to enter the ecosystem.

Inherent in the steady-state concept is considerable scope for fluctuation. The degrees to which fluctuations can occur in a steady-state system are indicated in a recent study (Rabotnov 1965) of structural dynamics in polydominant meadow communities. As a result of the distinctive ecological tolerances of each species (see Becking 1968), the several dominants in this vegetation display wide fluctuations in compositional combinations from year to year. This type of situation was also discussed by Becking (1968). He emphasised the importance of structural dynamics in assessing such communities.

Rode (1947) discussed the concept of dynamic equilibrium (= steady-state) in soil evolution, insisting that the self-development of an ecosystem (biogeocenose) could never reach dynamic equilibrium. Rode's inference was that rates of change during a stage of apparent steady-state could become so slow as to be undetectable. From a pedological point of view, Rode (1947), Nikiforoff (1959), Lavkulich (1969), considered that a steady-state must be ‘completely reversible’ in the sense that it can be lost. Equilibrium can be lost either as the result of a change in an external factor or because of the progressive internal development of soils.

With reference to steady-state and subsequent changes it is relevant to quote Holling (1969) who considered thresholds in ecological systems. He said, ‘there are finite limits to the resources and the responses of all organisms to these resources. Together with historical and spatial effects, these thresholds introduce a fundamental nonlinear character into interacting ecological processes.’

The exceeding of a threshold in any vegetation/soil system gives rise to ‘post steady-state’ changes.

Conclusion

In perspective it can be concluded that the concept of ‘climax’ has —

An alternative to the untenable concepts of climax, maturity and stability is the concept of ‘steady-state’. Steady-state adequately defines a temporary state of dynamic equilibrium in an open system, such as the vegetation/soil system. It employs ‘dynamic equilibrium’ as denoting adjustment in a non-reversible sense to a state of minimal change-with-time. Fundamental to the meaning of steady-state is a minimum of continuous variation within and between all parts of the ecosystem; that every variable is, and is part of, a continuum (Webster 1968). Steady-state is a state of comparative stability that may be preceded and succeeded by states of comparative instability — ‘pre-steady state’ and ‘post-steady state’ respectively.