An Ecological Approach to Biosystem Thermodynamics
10201 Wildflower PlaceSidney, B.C. V8L 3R3Canada
ABSTRACT: The general attributes of ecosystems are examined and a naturallyoccurring "reference ecosystem" is established, comparable with the "isolated" system ofclassical thermodynamics. Such an autonomous system with a stable, periodic input ofenergy is shown to assume certain structural characteristics that have an identifiablethermodynamic basis. Individual species tend to assume a state of "least dissipation"; thisis most clearly evident in the dominant species (the species with the best integration ofenergy acquisition and conservation). It is concluded that ecosystem structure resultsfrom the antagonistic interaction of two nearly equal forces. These forces have theirorigin in the Principle of Most Action ("least dissipation" or "least entropy production")and the universal Principle of Least Action. "Most action" is contingent on the equipar-titioning of the energy available, through uniform interaction of similar individuals. Thetrend to "Least action" is contingent on increased dissipation attained through increasingdiversity and increasing complexity. These principles exhibit a basic asymmetry. Giventhe operation of these opposing principles over evolutionary time, it is argued thatecosystems originated in the vicinity of thermodynamic equilibrium through the resonantamplification of reversible fluctuations. On account of the basic asymmetry the systemwas able to evolve away from thermodynamic equilibrium provided that it remainedwithin the vicinity of "ergodynamic equilibrium" (equilibrium maintained by internalwork, where the opposing forces are equal and opposite).
At the highest level of generalization there appear to be three principles operating: i)maximum association of free-energy and materials; ii) energy conservation (decelerationof the energy flow) through symmetric interaction and increased homogeneity; and iii)the principle of least action which induces acceleration of the energy flow throughasymmetrical interaction. The opposition and asymmetry of the two forces give rise tonatural selection and evolution.
KEY WORDS: Action principles, ecosystem structure, evolution, information, naturalselection, non-equilibrium thermodynamics, teleology.
But ask now the beasts, and they shall teach thee,and the fowls of the air, and they shall tell thee,Or speak to the earth, and it shall teach thee,and the fishes of the sea shall declare unto thee,
Job vii-viii. 12
If science is dependent on progress in subsuming a class of phenomena under
Biology and Philosophy 7: 35-60, 1992. 1992 Kluwer Academic Publishers. Printed in the Netherlands.
laws of a higher order of generality, then the unity of all knowledge is surelydependent on a thermodynamic interpretation of biological phenomena(Landsberg 1987). Nevertheless, the establishment of a coherent thermodynamicbiological paradigm has proved to be extremely difficult on account of theapparent contravention of the Second Law of Thermodynamics by biologicalprocesses. There is no conflict at the level of the universe, organisms exploit anenergy gradient by delaying the energy transformation involved and utilizing thedelay in energy flux for the performance of work. However, at the local level ofthe Earth the tantalizing question remains as to the mechanism whereby orderand organization increase with time and give rise to systems having a very lowprobability of occurrence.
Despite Watt's (1971) vivid warning that "If we do not develop a strongtheoretical core that will bring all parts of ecology back together we shall all bewashed out to sea on an immense tide of unrelated information" only littleprogress toward an accepted synthesis has been achieved in the 20 years thathave elapsed since it was first voiced. This lack of a generally accepted theoryunderpinning biology has had an inhibiting effect on the development of thescience and allowed the development of "invisible colleges" (McIntosh 1980),each following its own suite of ideas.
The problems of unresolved controversy in ecology are compounded by thoseof semantics. One is inhibited from using words such as "stable", "climax","dominance", or "diversity", without several pages of qualification, becausethey have either been appropriated to a specific meaning by individualauthorities, or the basic concept behind the word has been questioned. I believewe cannot allocate these words to specific usage until we have determined thephysical basis on which each expressed concept actually rests.
Less controversial, but of equal difficulty, is that in biology we must talk intrends, presumably caused by underlying forces, without reference to fixedpositions. Biological trends are therefore like vectors, forces having magnitudeand direction but without co-ordinates. Only when these vectors manifestthemselves in a particular ecosystem do they form part of a co-ordinated system,but even here, both magnitude and co-ordinates are virtually impossible toevaluate and only occasionally do we get a firm indication of direction.
Despite these difficulties, there seems little doubt that a ring is closing arounda thermodynamic solution that will remove many of the problems encountered(Weber et al. 1989; Weber et al. 1988; Brooks et al. 1989). Notwithstanding,without major input from both ecologists and more formal thermodynamicistsand a concerted effort on both sides to understand each other, the struggle toresolve the question is likely to be unnecessarily protracted. I would even gofurther and state that unless thermodynamic concepts are now complete andincapable of modification, biology must lead the physics, not the reverse, for theobservations of ecology are usually of great complexity, and are the ones insearch of explanation.
While I believe that the approach of Weber et al. (1989) is the correct one, Ithink that their ecological sources have led them astray. They overlook certain
factors when detailing the general characteristics of ecosystems that are vital toeffecting a satisfactory synthesis. Without a detailed accounting of the"emergent properties" of ecosystems it is unlikely that a satisfactory ther-modynamic solution will be attained.
A major road-block in isolating biological generalities can be traced toLotka's (1922) statement:
In every instance considered, natural selection will so operate as to increase the totalmass of the organic system, to increase the rate of circulation of matter through thesystem, and to increase the total energy flux through the system so long as there ispresented an unutilized residue of matter and available energy.
This statement, while difficult to refute in general terms, conceals rather thanilluminates the problem, for the trends identified are not uni-directional asindicated. The views of Odum (1969) and Schneider (1988), which in turn formthe basis of Weber et al.'s (1989) analysis, are consistent with those of Lotka.The properties of successional or maturing communities, Weber et al. state, canbe divided into five categories:
i. increased energy flow;ii. greater variety of species;iii. more narrow trophic specialization by their members;iv. enhanced amount of cycling;v. longer retention of media in the system.
While it is undeniable that there is evidence for all these statements, none ofthem can be taken to the limit in the manner implied, for trends observed duringearly succession, may be reversed in the terminal stages.
From the ecological evidence, I believe there are essentially two antagonisticforces operating, which, in accordance with Le Chatelier's Law, tend to reachequality at a "climax". I hope to show that these two forces are nearly equal,each dominating system behaviour within a different time-frame and eachhaving an identifiable physical origin.
THE EMERGENT PROPERTIES OF ECOSYSTEMS
The proposition that a mature ecosystem has a greater variety of species thanone at a less mature stage is not universal. For example, McIntosh (1980) statesthat Clements (1905) ("to his eternal credit") was the first ecologist to formulatethe general proposition that over the course of succession:
the number of species is small in the initial stages, it attains its maximum in inter-mediate stages: and again decreases in the ultimate formation, on account of thedominance of a few species. (Emphasis added.)
Over the course of succession species diversity and biomass increase more orless hand-in-hand although alternating stages of dominance and diversity arefrequently evident (Harper 1969). This continues until the final stages which seethe dominant species gaining ascendancy and suppressing certain species. Manyexamples of this decline in species diversity in the terminal stages of successionare listed by Whittaker (1969). This increase and subsequent decline in thenumber of species is compatible with the general pattern of net and grossproduction described by Odum (Figure 1, in Odum 1969).
Similarly, Connell (1978), examining the role of disturbance in tropicalforests through indigenous man's activity, concluded that continuous, relativelyminor disturbance of the climax, leads to an increase in diversity.
However, all ecosystems do not behave in this manner. For example, Paine(1966) in his classic work on Pacific coast marine ecosystems showed thatremoval of the starfish Pisaster ochraceus, the terminal predator, caused areduction in diversity due to the assumption of dominance by the mussel Mytiliscalifornianus. Paine states:
The removal of Pisaster has resulted in a pronounced decrease in diversity asmeasured simply by counting species inhabiting this area, whether consumed byPisaster or not, from a 15 to an eight-species system. The standing crop has beenincreased by this removal and should continue until the Mytilus achieve theirmaximum size.
The role of starfish in this situation appears to be equivalent to that of man inmaintaining higher diversity in tropical forest regions as noted above (Connell1978).
From such evidence Stanley (1973) developed a "cropping principle" toaccount for the sudden flowering of many new multicellular species in the LatePrecambrian. Nevertheless, an increase in species richness continues only up toa certain level of cropping intensity; beyond this point "overgrazing" reducesdiversity. Slobodkin et al. (1967) believe that the overgrazed condition occursonly through the effects of man-protected herbivores or introduced species. Forexample, when rabbits (Oryctolagus cuniculus) (a species introduced intoEngland in the 12th century) were eliminated from English downland by thedisease Myxomatosis, a number of species previously thought to have beeneliminated from the region suddenly re-appeared (Harper 1969).
The above observations can be interpreted most readily on the basis of twoantagonistic forces driving the ecosystem in different directions, with somewhatdifferent end-points being reached in different environmental settings: one forceinduces increased diversity, the other increased biomass. Over most of the serethese two processes go hand-in-hand with continuous jockeying for dominanceby one or the other, with immediate success depending on the physical environ-ment and the nature of species present. Only at the theoretical end point or"climax" does the system attain a degree of stability when the two underlyingforces approach equality.
In many cases it is evident that, in the terminal stages, the system behaviour is
dominated by the force tending to reduce diversity. This reduction in diversity isusually small relative to the maximum diversity attained in the penultimatestage. In the case of the starfish-dominated community studied by Paine, it maybe presumed that the climax is reached when a balanced state is attained close tothat of maximum diversity. This would be consistent with the fact that there hasbeen a much longer period for evolution in the marine environment than interrestrial ecosystems.
Similarly with biomass, it is evident that Paine's removal of the starfishpermitted an increase in biomass, yet conversely it is equally apparent thatremoval of the dominant trees in a forest will reduce biomass. In general, it maybe said that removal of a dominant predator will tend to increase biomass,whereas removal of dominant vegetation will reduce biomass. Removal of apredator eliminates one energy transformation stage. Because each energytransformation has a very high energy demand, biomass can increase eventhough total energy input remains constant. This implies that the specific energyflux (the energy to support unit biomass) will decline as biomass increasesbecause of longer retention time within the system of each unit of energyassimilated.
It seems impossible to define the "climax" or the "mature" ecosystem in preciseterms of diversity or biomass, even in relatively stable systems. Plant-dominatedand animal-dominated systems have similarities and they have differences. Inthe animal-dominated system the primary producer initiates a food-chain, orfood-web, which is terminated by the dominant animal species, whereas in theforest system plants are both primary producer and dominant. In either case,only in autonomous systems with a constant environmental cycle, will theclimax state be attained.
Irrespective of its successional stage, any perturbation of a system will set itback to a new position, from which it will begin, once more, to move toward theclimax, possibly by a different route since conditions rarely repeated themselvesprecisely. If extermination of part of the species complement occurs, resultingfrom the evolution or immigration of a new species, or a major environmentalcatastrophe, the system, over the course of time, either through further immigra-tion or evolution, will again proceed toward increasing diversity. Thus, theclimax is rather an elusive state, somewhat akin to a perspective point on thehorizon toward which a system converges, but one that changes and recedes asthe immediate conditions or species change. Ecosystems may therefore beregarded as dynamic entities whose major characteristic is stability-seekingrather than stability.
"More narrow specialization by members of a mature ecosystem" is subject tosimilar restrictions. In general, as Whittaker (1965) states, species tend to bepartial rather than direct competitors, tending
...to evolve also toward habitat differentiation, toward scattering of their centers ofmaximum population density in relation to environmental gradients so that fewspecies are competing with one another.
It is evident that over successional time, species will so distribute themselves asto ensure a more specialized division of the total "ecological space," but thistrend may be reversed when a forest tree or large animal assumes dominance. Infact, as McNaughton and Wolf (1970) point out, the dominant species has thebroadest "niche width." Characteristic of most dominant species is their largesize and great mean age relative to other species in the system. To achieve thisstatus they must arrogate to themselves resources that otherwise would be usedby additional species. Over evolutionary time it seems that there have been twodivergent trends, one toward specialization and the other toward generalization.The second trend is clearly evident with the arrival of man, the supremegeneralist.
Retention of MediaSimilar considerations again apply, but it is extremely difficult to reach generalconclusions without more detailed knowledge of the forces operating.
It may be concluded that these patterns result from the interaction of twoantagonistic trends leading, at the highest level of generality, to homogeneityand heterogeneity (Johnson 1981). Homogeneity implies symmetric interaction,a reduction in energy flux, accumulation of biomass and the dominance of oneor a few species. Heterogeneity implies asymmetric interaction, increaseddiversity, increased energy uptake, and increased energy flux.
The "system" characteristic is toward increased heterogeneity; individualspecies in the system tends to oppose this trend by accumulating biomass andreducing diversity. For such a system to continue in existence over ecologicaltime, the sum of the individual trends to increased biomass through conservationmust exceed or equal the system trend to increased energy flux. This slightasymmetry is the basis of the homeorhetic mechanism (the stabilization offlows) (Waddington 1968) operating over successional time. Because the trendsare nearly equal in magnitude but opposite in direction, a "compromise"situation results in which the forces assume equality at an idealized staterecognized as the climax.
This is in conformity with the conclusion of Brooks et al. (1989) that thereare:
two classes of energetic transformations allow this: heat-generating transformationsresulting in a net loss of energy from the system, and conservative transformationschanging unusable energy into states that can be stored and used subsequently.
However, Brooks et al. do not set these transformations in direct opposition toeach other.
AN ECOLOGICAL REFERENCE SYSTEM
Work on arctic lakes in Victoria Island and on the Kent Peninsula, in theCanadian Northwest Territories, has provided interesting insight into ecosystemstructure. The virtue of many of these lakes is that they may be regarded asdynamic reference systems, comparable with the imaginary "isolated" system ofclassical (or static) thermodynamics. Many are uniform limnologically; they aresmall and highly autonomous with relatively simple ecosystems and they haveremained virtually undisturbed since their formation during the late Pleistocene.With the recession of the ice, the lands on which these lakes now occur emergedfrom the sea owing to isostatic rebound. Some of the lakes have river outflowsmaintaining a navigable connection to the sea for anadromous fish species whileothers are now isolated from the sea by impassable falls. Sequestered beneath athick layer of ice for 9 to 10 months of the year, these ecosystems closelyapproach a "closed" thermodynamic condition in which only energy and entropyare exchanged with the external world.
Several salient characteristics have an important bearing on the ther-modynamics underlying ecosystem structure: the successional changes duringtheir formation, their manifest stability, and the structure of the stable stateattained.
During succession, the changes that have occurred since they were isolatedfrom the sea, have invariably involved a decline in species richness. Themajority of the larger species inhabiting the lakes is of marine origin, or hasgained access via coastal movement. In those lakes that support anadromouspopulations of Arctic charr (and therefore still have access to the sea) thereexists a set of fish and invertebrate species from which all sub-sets have beenderived. These are: i) the freshwater fishes: Arctic charr (Salvelinus alpinus),lake charr (lake trout) (Salvelinus namaycush), lake whitefish (Coregonusclupeaformis), Arctic cisco (C. automnalis), least cisco (C. sardinella) and nine-pine sticklebacks (Pungitius pungitius); ii) a rather heterogeneous collection ofspecies, collectively referred to as "marine-glacial relicts." They are species ofmarine origin that have adapted to freshwater conditions in circumpolar lakes(Johnson 1964): the sculpin (a fish) Myoxocephalus quadricornis thompsonii;the amphipod: Gammaracanthus aestuariorum, the isopod, Mesidotea entomon,the mysid, Mysis relicta and the copepod Limnocalanus macrurus; iii) theubiquitous amphipod Gammarus lacustris and a large number of chironomidspecies. The chironomids have not been studied in sufficient detail, nor are theyof marine origin, so cannot be included in the analysis. Each of the lakes thatbecame isolated from the sea developed a unique climax condition, in whichmost possible sub-sets of the above-listed species have been found. Presumably,each lake developed a biota commensurate with its environmental characteris-
tics. No correlation has been found between species complement and time sinceemergence (height above sea level), or other specific limnological characteristic.Thus, all the lakes may be considered to have had the same initial complementof species, but each has subsequently developed independently.
The subsequent course of succession is characterized by a decline in speciesrichness, as each lake could support only some of the species initially available.In most cases either lake charr or Arctic charr assumed dominance, eliminatingthe alternate species. Where Arctic charr have assumed dominance all other fishspecies (with the exception of sticklebacks) are frequently eliminated; in thesecircumstances Arctic charr is a highly generalized feeder, utilizing all availablefood resources.
The stability of the dominant fish populations, with respect to the conditionsthey have experienced over their recent history, is expressed in their populationstructure, which invariably consists of large, old individuals with few juveniles.Secondly, if the population is severely disturbed it returns to its originalconfiguration in a well-damped manner (Johnson 1972, 1975, 1976, 1983, 1985;Vanriel 1989). Thirdly, stability is confirmed by the direct observation of anumber of lakes over a period of 30 years. It is evident that the ecosystems inthese lakes attain a state very close to their theoretical climax.
The dominant may be defined as the species with the "best" integration ofenergy acquisition and convertion; it is the species showing "least dissipation"or "least entropy production" relative to the biomass accumulated, within theconstraints of the environmental conditions and its genetic make-up. It is notnecessarily the species with the greatest biomass (Vanriel 1989; Vanriel andJohnson, in prep.). Least dissipation is a property of the species population or"ecodeme". The characteristics of least dissipation are: 1) unimodal lengthfrequency distribution indicating uniformity in size; 2) large size; 3) great meanage; 4) great variation in age at modal length; 5) a dearth of juvenile replace-ment stock; and 6) indeterminate age at death (Johnson 1976, 1983, in press). Along life-span indicates a long turnover time and a low requirement for juvenilereplacement stock to maintain numerical constancy. Specific metabolic ratedeclines with increasing mass (Kleiber 1961; Lindstedt and Calder 1981),therefore the larger the individual size the greater biomass per unit energy input,or the lower the specific energy flux.
Finally, it has been shown that a single species, Arctic charr, within a singlelake, has the capacity to partition the total energy available to it and formvarious sub-groups or "morphs". Each of these morphs function as a cohesivegroup, as indicated by a distinctive size mode, with each mode containingreproductive individuals. This characteristic is regarded as heterochrony (deBeer 1958; Gould 1977; McNamara 1988) at the ecological level. Heterochronyinvolves a redistribution of the factors of energy and time (Parker and Johnson1991).
These characteristics of the dominant species are not confined to fishpopulations in Arctic lakes. They have been shown to exist in many undisturbed,autonomous, ecosystems over a wide variation in latitude and in a wide range of
conditions both aquatic and terrestrial (Johnson 1985, in press). Jones (1945),reports that in the virgin forests of Europe many apparently even-aged stands oftimber, are in reality, widely ranging in age, a conclusion supported by McIn-tosh (1985). Almost identical with some northern fish populations are the ageand size distributions of the much studied giant land tortoises (Geochelonegigantea) of Aldabra Atoll in the Indian Ocean (Gaymer 1968; Grubb 1971;Swingland and Coe 1978, 1979). Similar findings are reported for the dominantfish species in the pelagic waters of Lake Tanganyika (Coulter 1970, 1976), andfrom the deep seas for molluscs (Hutchings and Haedrich 1984) and fish(Gauldie, West and Davies 1989). Such is the range and variety of ecosystemconsidered that these characteristics may be considered universal (Johnson, inpress) given the basic conditions of autonomy and regularity of energy input.
It is evident that uniformity in size, despite variation in age, in an organism ofindeterminant growth pattern necessitates interaction and the exchange ofinformation between individuals as implicit in homeokinesis (Soodak and Iberall1978). Similar evidence for cohesion in plant communities exists (Harper 1967,1969; Jones 1945; Tamm 1948, 1956; Werner and Caswell 1977).
Given that the dominant species tends toward a state of least dissipation, it isreasonable to assume that the system as a whole also tends to least dissipation.As Frank (1968) states, the stability characteristics of a community are largelythose of one or a few dominant species. A state of least dissipation is, in fact,equivalent to stating that the mature ecosystem assumes the least attainablevalue of the Production/Biomass (PIB) ratio (Margalef 1968; Odum 1969).
The major characteristics of evolution are an increase in diversity and anincrease in average complexity (Simpson 1953), but over the course of evolutionthe increase in diversity has not been monotonic. Numerous significantdecreases in diversity are evident in the fossil record (Raup and Sepkoski 1982),notably in the later stages of the Permian era and more recently at theCretaceous-Tertiary boundary. However, after each event the trend to increasingdiversity was resumed, resulting in an overall increase with time.
It is postulated that increasing diversity and increasing complexity arecontingent on an increase in the internal work performed, thereby necessitatingan increase in specific energy flux.
The two antagonistic trends can now be defined as trends toward decelerationand acceleration of the energy flow. The overall trend to deceleration isexpressed in the successional process over ecological time, whereas accelerationis evident in the increase in complexity over evolutionary time. The interactionbetween the two trends results in the changes recognized as evolution. Wherethere is acceleration there is a force, therefore the important question is what arethe forces that initiate these trends?
Uniting these observations is the fundamental physical concept of "action."Action in this context has little connection with the vernacular usage of the term:it is used here in the strictly physical meaning of the product of energy and time.The concept is most frequently encountered in the Principle of Least Action. Theprinciple of least action states that the trajectory of the system is that path whichmakes the value of S (the action) stationary relative to nearby paths between thesame configurations and for which the energy has the same constant value. Theprinciple is really misnamed as only the stationary property is required. Actionis expressed in terms of energy x time (i.e., in joule-seconds in the S.I. system ofnotation).
According to Feynman (1963) and Watson (1986), the fundamental equationsof motion can be put in the form of the principle of least action and Zee (1986)states that some physicists even go so far as "to believe that the UltimateDesigner thinks in terms of action." Certainly, it seems action was uppermost inHis mind when plans for the biological world were being formulated.
The action (S) of a body in motion is computed from the difference betweenits potential (PE) and kinetic (KE) energies at each instant in time, integratedover that portion of the trajectory under consideration (Feynman 1963):
Action = S = (KE-PE)dt. (1)t1
In most mechanical systems this value is a minimum thus the trajectory of thesystem follows the "path of least action". Feynman states that the system "sniffsout" the path of least action from a large number of possible paths. The impor-tant path becomes the one for which there are many nearby paths which give thesame phase." He continues: "What we really mean by 'least' is that the first-order change in the value of S, when you change that path is zero. It is notnecessarily a 'minimum' ". That is, the action may also be a "maximum". Thepath of least action, therefore, is that path, small deviations from which, do notincrease the elapsed time taken for the event under consideration.
Organisms may be regarded as products of energy-time and materials.In the biological world, it is postulated that a Principle of Most Action
functions in opposition to the Principle of Least Action, with the systemseventually settling down in a compromise situation in the vicinity of a"minimax" or climax state. At the climax, action is thus a maximum(accumulation of maximum energy for the longest time period possible) withthis state being refined to a minimum under the envelope of constraints imposedby the principle of least action. Provided the trend to most action exceeds orequals the trend to least action, this "minimax" condition implies homeorhesis,with flows being stabilised around a condition of least work.
In biological systems the action of an organism, ecodeme or ecosystem may
be regarded as the difference between the kinetic energy, (energy assimilatedover the course of a single annual cycle and equivalent to the production energy(P)) and the potential energy (the energy in the biomass) (B).
Unless a fundamental change is postulated, it must be assumed that the ther-modynamic forces operating initially were the same as those of today. It is,therefore, most probable that organisms originated close to thermodynamicequilibrium by the resonant amplification of reversible processes. Thisresonance may still be seen in the endogenous rhythms of plants and animals.An atom to which an electron is transferred, experiences a short period ofexcitation before returning to the ground state with the emission of a photon. Itis hypothesized that a periodic energy input from the sun amplified the time-delay, allowing temporary energy storage and a more prolonged period ofexcitation. In the excited state the atom became more reactive allowing theformation more complex entities and more complex cycles. The uniformlycharged particles thus formed (thermodynamic engines, indivisible atomisms, orproto-organisms) interacted to maintain uniformity in accord with the doctrineof homeokinesis, energy being equipartitioned by "internally time-delaying,processing and transforming collisional inputs, generally using many fluid-likemobile steps" (Soodak and Iberall 1978).
Two energy transport processes, one cyclic (energy conservation) and theother rectilinear (dissipation of radiant energy), were now in conflict. Accordingto Onsager's reciprocal relations (Onsager 1931), when two energy transportprocesses (e.g., diffusion and conduction) interfere with each other the systemtends to assume a state of minimum entropy production, or least dissipation.Least dissipation is equivalent to most action. Rigorously, Onsager's relationsare applicable only to isothermal systems near equilibrium, where the energygradients are weak.
Least dissipation implies deceleration of the energy flux and, therefore, in asuitable medium, may give rise to the formation of "dissipative structures"(Prigogine and Wiame 1946; Prigogine 1978; Prigogine and Stengers 1984).Dissipative, or perhaps preferably, ergodynamic structures, are patterns in theenergy flow maintained by internal work done through the dissipation of storedenergy. According to Prigogine and Stengers (1984) dissipative structuresrepresent:
A new type of order .... We can speak of a new coherence, of a mechanism of"communication" among molecules. But this type of communication can arise only infar-from-equilibrium conditions. It is quite interesting that such communication seemsto be the rule in the world of biology. It may in fact be taken as as the very basis of thedefinition of a biological system.
However, Prigogine and Stengers (1984, p. 140) state further that:
Far-from-equilibrium, the system may still evolve to some steady state, but in generalthis state can no longer be characterized in terms of some suitably chosen potential(such as entropy production for near-equilibrium states).
This is contrary to the ecological evidence: biological systems, as ergodynamicstructures, proceed toward a state of least dissipation in accordance withOnsager's relations. Thus, it must be assumed, they originated close to thethermodynamic equilibrium, subsequently evolving away from that positionowing to the fundamental asymmetry. At the same time the forces wereamplified while maintaining their original, antagonistic, near-symmetricrelationship.
The primitive system may be envisaged as being composed of uniformlycharged particles interacting in an indeterminant manner. Such a system exhibitsthe properties of a tabula rasa, or a clean sheet, essential for identifying andencoding environmental signals. As stated by Polanyi (1968),
It is the physical indeterminacy of the sequence that produces the improbability ofoccurrence of any particular sequence and thereby enables it to have a meaning - ameaning with a mathematically determinate information content equal to thenumerical improbability of the arrangement.
Scheer (1970) stresses the need for indeterminacy of the symbols for transmis-sion of information and adds:
the production of order, or a decrease of entropy, in any system requires an input ofinformation as well as of energy to the system.
Initially, information contained in the environmental signals to which the systemwas exposed, was identified and "coded" by the tabula rasa; energy for thenecessary work was provided by the time-delayed dissipation of the energyassimilated. An environmental signal may be defined as any sustained periodicfluctuation in the energy environment.
The system, being subject to errors in replication, contained a degree ofmalleability. This malleability allowed specific signals to be identified. Anincrease in the number of environmental signals identified demands an increasein the work done (Pierce 1961) which, in turn, necessitates an increase the rateof energy dissipation. Identification of signals and exploration of the total signalspace was thus stimulated by the principle of least action. Energy "entrapped" incyclic pathways could obey the principle of least action only by doing morework.
This work is manifest in the support of greater complexity and greaterdiversity. Greater complexity also permitted greater acquisitive capacity andmore complex storage cycles. On the other hand, the trend to most action tendedto reduce entropy production, increase energy storage and eliminate fluctuation.Given these conflicting trends, continued existence of a specific configuration ofan ecosystem depends on the short-term domination of system behaviour by thetrend to most action. Any change must lead either to an increase or a decrease inaction; to be viable a change must be such as to allow the system to assume anew stationary state. Over evolutionary time, changes resulting in both an
increase and a decrease in action evidently occurred, but over the long term, thetrend has been slightly in favour of decrease.
A self-regulating ratchet mechanism was thus created which graduallywinched-up the system to greater diversity, and greater complexity by increasingboth energy accumulation and rate of dissipation (Figure 1). Given the long-termtrend to least action, it must be assumed that the probability of any specificchange leading to less action is somewhat greater than that of one leading togreater action. Interestingly, entropy production, not entropy, always "wins" inthe end.
Succession: Decreasing Action ---- >
(Action = Energy x time)
(The sum of the trends (Interactions between species +of individual species increase in individual complexity= most action) = least action)Decreased energy flux Increased energy fluxper unit biomass per unit biomassLess power More power
System moves to the left
Climax (= a minimax):
System remains stationary
Increased diversityIncreased average complexitySystem moves to the right
Out of these counter trends emerges evolution
Fig. 1. Diagram of relationship of forces operating during Succession, the Climax andAbscession and their relationship to Evolution.
These changes are possible only if the system remains in a state homologouswith thermodynamic equilibrium where the opposing forces are weak and smallin magnitude. That is, the system can evolve away from thermodynamicequilibrium provided that the system remains in the vicinity of ergodynamicequilibrium (Gnaiger 1986). Ergodynamic equilibrium is an equilibrium statemaintained by work done. It represents an amplification of the forces operatingat thermodynamic equilibrium. It may be envisaged as being on the "far-side" ofthermodynamic equilibrium. Initially changes must have been very small, butwith increasing time and stronger forces, greater changes could be absorbed.
The trend to most action leads to succession, whereas the trend to increasingdissipation may be termed bscession (= a moving away). Out of the conflictbetween them emerges evolution. Evolution, therefore, is not unidirectional, it isthe resultant of the two slightly asymmetric forces working on a malleablemedium. The trend to symmetry in an inherently asymmetric world provides theevolutionary driving mechanism.
Following partitioning of the tabula rasa and the identification of more than asingle signal, a hierarchy was formed, based on the relative capacities of theindividual groups of atomisms (now species) to acquire and conserve energy andperpetuate their existence. As Bastin (1968) concluded:
the only logical way in which it is possible to discriminate a number of activities intoa hierarchy is by considering their reaction times, a higher-level in the hierarchyalways having a much longer reaction time than a level classified as lower.
This gradient is readily recognized in relationships along a simple food-chain:there is an increase in size (Dickie, Kerr, and Boudreau 1987), an increase inlife-span and a decline in the value of the P/B ratio (Brylinsky 1980; Mann andBrylinsky 1975; Saunders et al. 1980; Winberg 1972; Winberg et al. 1972), andan increase in energy density (Hirata and Fukao 1977; Hirata and Ulanowicz1986). This whole group of characteristics has been brought together under theumbrella concept of biological temperature (Johnson 1981, 1990; Vanriel andJohnson, in prep.).
A hierarchical structure lends great support to the postulate of a resonantorigin of organisms. In such a structure, the various species function as a systemof resonators. The dominant species, being most heavily heavily damped by theaccumulation of energy-time, resonates to a wide range of environmentalsignals, whereas species lower in the hierarchy dissipate their energy rapidly,responding only to a relatively narrow band of frequencies.
Of great ecological significance is the fact that the climax represents aminimax state. The sides of the curves leading to "least" and "most" action aresteep, whereas at the "minimax" the curve is flat. This implies a condition of"stationary time" in which small deviations from the minimax path do notchange the action. This accounts for the climax state being quite variable andrelatively easily displaced. Furthermore, Curtis and McIntosh (1951) state that:
...although the tolerance range of each species may overlap that of many otherspecies, the range for optimum development of each species is different from that of
all other species. There are therefore, no groups of tree species which regularly occurtogether, and only together, except for accidental duplications of narrow environmen-tal ranges. Rather, tree species occur in a continuously shifting series of combinationswith a definite sequence or pattern, the resultant of a limited floristic complementacting on, and acted upon, by a limited range of physical environmental potentialities.
To summarize this rather bewildering complex of processes, many of which areoccurring simultaneously, a general schema is presented:
I. Characteristic of the Ecodeme1. Ecodemes or species populations, tend to assume a state of most action, or
least dissipation, relative to their energy input.2. Individual ecodemes function as cohesive entities not merely collections of
individuals (Johnson 1983).3. Within an ecodeme, cohesion and least dissipation are maintained by
equipartitioning the available energy.4. The equivalence of energy and time allows heterochronic change (de Beer
1958; Gould 1977; McNamara 1988). Heterochrony at the ecological level,within the ecodeme (Parker and Johnson 1991), allows changes of theenergy-time component (action) to meet new conditions, thus impartingflexibility to the structure.
II. Characteristics of the Ecosystem1. In an undisturbed, autonomous ecosystem a stationary stable state is
reached; this state is recognized as the "climax".2. Ecosystems have properties that are adaptations at the system level (Dunbar
1972; Ott 1981).3. Individual ecodemes form a hierarchy according to their ability of acquire
and conserve energy.4. At the head of the hierarchy is the the dominant species (defined as the
species with the "best" integration of energy acquisition and conservation).This is particularly evident in the dominant species in undisturbed,autonomous ecosystems (Johnson 1976, 1983, 1990, in press).
5. The dominant species may be either a plant, when it is also the mainprimary producer, or an animal, when it is terminal in the food chain. In asimple linear food chain with an animal dominant, specific energy fluxdeclines at each stage in the hierarchy from primary producer to terminalpredator (Vanriel 1989; Vanriel and Johnson, in press).
6. The dominant species has the broadest niche width (McNaughton and Wolf1970).
7. The salient characteristics of an ecosystem are largely those of one or a few
dominant species (Frank 1968), thus, the ecosystem as a whole tends toward'most action' ('least dissipation' or minimum attainable value of the PI/Bratio (Margalef 1968; Odum 1969).
8. In many ecosystems, during the course of succession, species diversity isinitially low, it then increases with time reaching a point of greatestdiversity in the penultimate state, before generally declining as the climaxstate is approached (Clements 1905; Whittaker 1969). In certain animaldominated communities, maximum diversity is attained at the climax.
III. Characteristics of the Biosystem1. The earth as a whole forms a closed thermodynamic system exchanging
energy and entropy with the external universe.2. Within the Earth's closed system various subsystems or ecosystems formed
within partial boundaries. Ecosystems exhibit a greater or less degree ofautonomy depending on the nature of their boundaries.
3. The salient changes over evolutionary time have been an increase indiversity and an increase in average complexity (Raup and Sepkoski 1982;Simpson 1949, 1953, 1969). Most authorities agree that, initially, there wasonly a single species of organism (Darwin 1859; Dobzhansky 1968).
4. There is an increase in diversity from the poles to the tropics (Wallace1878; Connell and Orias 1964; Fischer 1960; Pianka 1966).
5. Specific energy flux increases with decreasing latitude (measured by theProduction/Biomass ratio (Connell and Orias 1964; Golley 1965, 1972;Mann and Brylinsky 1975; Reiners 1972; Burger 1981).
6. Because specific energy flux increases with decreasing latitude anddiversity increases with decreasing latitude, it is concluded that specificenergy flux increases with increasing diversity. However, this is concurrentwith the gradients of increasing energy availability and increasing tempera-ture, which also run from the poles to the tropics. Given the multiplicity ofgradients, this may be provisionally accepted as "strong inference" (Platt1964).
7. Total energy flux has increased over evolutionary time owing an increase ofsome 25% in luminosity over the lifetime of the sun (Newman and Rood1977), and improved assimilation mechanisms.
8. If specific energy flux increases with increasing diversity, geographically,then it may be assumed that specific energy flux has increased withincreasing diversity over evolutionary time.
9. Specific energy flux increases from the poles to the tropics; therefore, therewill be a tendency for energy to be captured by ecosystems with higher fluxfrom those with, relatively, lower flux, i.e., there will be an overall energydrift toward the tropics, (reflected in migration patterns). This energy drifttoward the equator is contrary to the energy flow within an ecosystem(within an ecosystem, energy flows from high PI/B to low PI/B).
10. Many taxa, over the course of their evolutionary history show an increase in
size (Cope's Rule) (Stanley 1973). An increase in size implies a decline inspecific metabolic rate (Kleiber 1961; Lindstedt and Calder 1981) thereforegreater biomass per unit energy assimilated; the effect of this is in opposi-tion to proposition 8.
11. Over the course of evolution, there have been long periods of stasispunctuated with periods of relatively rapid change (Eldredge and Gould1972; Gould and Eldredge 1977; Stanley 1981; Boucot 1983).
The above sets of observations are consonant with the proposition that ecosys-tem structure results from the antagonism between trends to least action and tomost action. The trend to most action dominates system behaviour in the short-term; the trend to least action dominates the long-term behaviour of the system.This is comparable with the Solar System where the cyclic motion of planetsaround the sun is maintained with little change, although, at the same time, theentropy of the whole system is gradually increasing. The major difference is thatorganisms, unlike the planets, receive a continuous input of new energy.
The above properties of the biosystem can be subsumed under three prin-ciples. These principles are in addition to the laws of thermodynamics and donot replace them.
1. The principle of association. Free energy (energy capable of performingwork), in the appropriate circumstances, enters into a temporary association withmaterials to the greatest extent possible.
2. The principle of conservation. When free energy enters into association withmaterials it is conserved in cyclic (reversible) processes. That is, energy isdelayed in its transformation from the free form to low-grade heat.
3. The principle of least action. The tendency for the specific energy flux toincrease to the greatest extent possible.
To these principles must be added the rider that, for any given configuration,the forces generated by the principle of conservation must exceed or equal theforces generated by the principle of least action.
The following extensions or corollaries of the above principles can bedeveloped:1. Organisms originated through the resonant amplification of reversible
fluctuations occurring in the vicinity of thermodynamic equilibrium. Thedelay in energy transformation resulting from the forces of conservation,allowed work to be done with the formation of ergodynamic structures.("the devil makes hands for idle work".) These resonant structures imposedorder and damping on the generative fluctuations.
2. The two energy transport processes established (cyclic and rectilinear)interfered with each other. In the vicinity of thermodynamic equilibrium,
this interference ensured that the system assumed a state of least dissipation.3. The initial atomisms interacted symmetrically, equipartitioning the avail-
able energy. In this stable state they formed a tabula rasa capable ofidentifying and encoding environmental signals. This new state is adeparture from thermodynamic equilibrium that may be termedergodynamic equilibrium. This implies that the initial symmetry of thetabula rasa was distorted but not broken. Initially, at ergodynamic equi-librium the opposing forces were weak and symmetric: energy inputequalled output and least work was done.
4. The principle of least action induces increased specific energy flow. Thiswas effected by an increase in the number of signals identified and anincrease in complexity. Thus, the system tended to explore the total signalspace of the environment. Although moving away from thermodynamicequilibrium the system remained at ergodynamic equilibrium through theperformance of work.
5. Identification of multiple signals was possible because, through errors inreplication, ergodynamic structures are capable of change. Change may beeither in the direction of increased conservation or increased dissipation.Any viable change will allow the system to reach a short-term stationarystate in the vicinity of ergodynamic equilibrium.
6. The probability that either acceleration or deceleration of the energy flowwill occur as a result of change is slightly asymmetrical; that is, there is asomewhat greater probability that any change will accelerate energy flowrather than decelerate it. This difference in probabilities imposes direc-tionality on the system.
7. The two antagonistic forces being nearly equal, the system tends to settledown in a position mid-way between zero and near-instantaneous dissipa-tion. This will be a position of maximum power (work done per unit time)and 50% efficiency (output/input) (Odum and Pinkerton 1955).
8. The three principles together function as a ratchet which gradually winches-up the system to greater energy input and more elaborate acquisition andstorage mechanisms through the development of increasing power, butwithin the constraint of remaining in the vicinity of ergodynamic equi-librium. This ratchet is natural selection.
9. Ecosystems thus move toward increasing entropy production over evolution-ary time through the development of greater complexity and more workdone, and not, as under the second law of thermodynamics, toward increas-ing entropy. At the same time there is a movement toward increased energyaccumulation and greater stability in the face of "normal" environmentalfluctuations experienced. Irreversible evolutionary processes (Dollo's Law)(Brooks, Cumming, and LeBlond 1988) are thus imposed on what are,essentially, reversible or cyclic processes. In the short-term, the oppositionof the two principles leads to succession, over the long term to abscession;interaction between the two processes results in evolution.
10. A biological hierarchy results from the differing individual capacities of
each species to acquire and conserve energy. The hierarchy is formed as aresult of the conflict between the two fundamental characteristics of"order": sequence and uniformity (or rank and row). Sequence acceleratesenergy flow, uniformity tends to retard it. A hierarchy involves the follow-ing propositions:
i. Processes operating at higher levels proceed more slowly than thoseat levels relatively lower (Bastin 1968; Simon 1973; Allen and Starr1982; O'Neill et al. 1986).
ii. The more the action (energy-time) accumulates in the dominantspecies the greater the control of resources achieved and the greaterthe damping moment of the species on the system.
iii. The more heavily damped a resonator, the broader is the band widthof signals to which it can respond. Thus, a more heavily dampedspecies tends to encroach on the signal series to which a less heavilydamped species responds (interspecific competition).
iv. Maintenance of a biological hierarchy requires that a state of leastdissipation (attained through cohesion and uniformity) be attainedwithin the "holons" (Allen and Starr 1982; Koestler 1967), and at thelevel of the hierarchy.
v. The more complex the hierarchy the greater the internal workrequired for its support (i.e., the greater the power required).
vi. Two species responding to different generative signal series mayinteract at a near-symmetric level. This interaction may subsequentlydevelop to their mutual benefit, thereby leading to co-evolution (e.g.,bird and insect pollinators of angiosperm plants).
11. The structure of an ecosystem at the climax reflects the recent individualhistories of the species in an ecosystem as they interact within the primarysignal series. An ecosystem may be regarded as an communicationsnetwork (MacArthur 1975) yielding information on the complex of signalsidentified. Once the primary tabula rasa had been distorted with theidentification of a second signal, all species tend toward a state of leastdissipation, thus dampening the signal identified and thus tending torespond to a wider signal band width. Each species will tend to adapt to thecentral peak of the signal or signal series identified. Species that improvetheir acquisitive and conservative propensities will encroach more and moreon the "tails" of signals identified by other species. Thus certain species willexpand their niche width and assume dominance ("K-selection"), others willbe "forced" to specialize in the vicinity of the peak of the signal or signalseries identified ("r-selection").
The arresting feature of this outline is the way in which the various physicalprinciples interact and overlap to produce a harmonious and internally consistentwhole. In fact, the physics of biology, as postulated above, represents theconsilience of a wide range of inductions, one of the most compelling reasonsfor the acceptance of any theoretical construct.
These principles provide a new outlook on some of the central themes ofbiology. Perhaps, most significant is the role of time and the inherent asymmetryof the underlying forces. Energy-time is thus a vital component of living tissuecomparable with space-time in cosmology. At the highest level of generality,symmetrical interaction leads to a decline in entropy production; asymmetricalinteraction leads to increased entropy production and acceleration of the energyflow. As Pasteur suggested life is dependent on a fundamental asymmetry(Dubos 1960; Kondepudi 1988).
Light is cast also on "teleology," the belief that progress is directed toward thegoal of some perfect state. This question has confused and confoundedbiologists since the time of Aristotle, particularly in recent times with therejection by most scientists of any supernatural guidance to the evolutionaryprocess; yet despite all arguments, there has been an overall trend that can onlybe termed progressive. Undoubtedly there has been progress, but, as Simpson(quoted by Dobzhansky (1968)) says,
...evolution is not invariably accompanied by progress, nor does it seem to becharacterized by progress as an essential feature. Progress has occurred within it but isnot of its essence... Within the framework of the evolutionary history of life therehave been not one but many different sorts of progress.
The "goal" turns out to be increased work concomitant with increased energyacquisition, overall directionality being ensured by the somewhat greaterprobability that any specific change will result in more rapid energy dissipationrather than in conservation. "Progress" may be regarded as resulting either froma trend to dominance through more generalized behaviour patterns and greateraction, or, alternatively, to specialization through closer adaptation to a specificset of signals and less action. Specialization will tend to increase fluctuation asspecies track specific fluctuations in energy input. However specialization inone era may lead to later generalization, as in the lungfishes, which, specializingin drought adaption, ultimately gave rise to air-breathing land vertebrates.
Dissection of the total signal series will proceed as far as possible in a givenset of environmental conditions, within the constraint that conservation mustdominate short-term behaviour. The less the environmental variability at themacro level (both within and between seasons), and the greater the variability atthe micro-level, the greater will be the diversity, because of the existence of agreater number of identifiable signals. Diversity at the micro-level will be self-augmenting. Hence, diversity increases toward the tropics. Dominance also, will
be self-augmenting. Dominance will be favoured by a variable environment (atthe macro level), or a "lean" energy environment.
A significant change in the global environment may cause a drastic decline indiversity, and allow opportunism free-reign when the system settles down oncemore. The new species complement will be gradually tempered by the long-termattenuation of individual energy resources as diversity increases and the forcesof natural selection becomes more stringent.
The immediate need is for better understanding of ecosystem mechanics inthe hope that management of the world's biosystem may be improved. Whilethis schema may contribute to understanding ecosystem structure and behaviour,it will, unfortunately have little predictive capacity with respect to detail.Reliance for improvement will have to remain largely a matter of judgment.
If the above propositions are correct it will be evident that they have a far-reaching effect on the philosophy of biology and man's relationship with theremainder of the living world, as well as having implications for economics(Vogel 1988, 1989). The living world in a near-stable state is a finely balancedstructure, maintained by a great deal of internal work; it is on the continuance ofthis work that stability depends. The balance is evident in the vicinity of theclimax, where the value of the action is insensitive to small changes in theenergy path. Virtually any displacement of the system beyond the immediatevicinity of the climax will tend to accelerate energy dissipation. Any disturbanceof a relatively small nature, will, if continued, drive the system to a new stablestate. If the disturbance is sufficiently large, it may lead to great structuralchange; but provided the system remains within certain limits, it will thenestablish a completely new, but less evolutionarily mature, ergodynamicequilibrium.
This potential for change is clearly demonstrated by man's current activitiesas the dominant species. He is accelerating energy flow at all subordinate levelsin the hierarchy, either by shortening and straightening food chains in naturalecosystems, or by replacing natural systems with agricultural systems. At thesame time he is accumulating energy-time within his own population through anincrease in abundance, an increase in individual biomass and an increase inlongevity. This leads inexorably to the elimination of a large number of plantand animal species. His dependence on an extremely wide signal band width andthe mechanical equipment available to him to facilitate signal capture, ensuresthat he is a formidable competitor. This highly asymmetrical process, largelysupported by a massive input of energy in the form of "fossil fuel," at presentshows little sign of approaching a new ergodynamic equilibrium.
It is frequently stated that ethics has no biological base (Provine 1982), but Ibelieve that the above conclusions tend to confound this view. It has beenstressed that individuals in species populations in an ecosystem must maintain adegree of uniformity and cohesion, through near-symmetric interaction andcommunication. In natural systems this is controlled without conscious effort.Ethics is therefore the conscious implementation of such precepts often to one'sown immediate disadvantage. This view supports the Aristotelian position of
"the way of the mean" or that of the major religions, as exemplified in theGolden Rule (Matthew, vii, 12):
Therefore all things whatsoever ye would that men should do to you, do ye even so tothem: for this is the law and the prophets
or the Taoist concepts embodied in the interaction of "yin" and "yang":Nothing is only yin or yang. All natural phenomena are manifestations of a continuousoscillation between the two poles (...). The natural order is one of dynamic balancebetween yin and yang (Capra 1983, p. 35, quoted by Balon 1988).
Perhaps the greatest ethical dilemma with which we are faced, is the fact thatshort-term solutions to immediate problems frequently conflict with long-termstability.
Allen, T. F. H. and T. B. Starr: 1982, Hierarchy, University of Chicago Press, Chicago.Balon, E. K.: 1988, 'Tao of Life: Universality of Dichotomy in Biology', Rivista di
Biologia 81 (2), 185-230.Bastin, T.: 1968, 'A General Property in Hierarchies', in C. H. Waddington (ed.), Toward
a Theoretical Biology, Aldine, Chicago, pp. 252-264.Boucot, A. J.: 1983, 'Does Evolution Take Place in an Ecological Vacuum?' Journal of
Paleontology 57, 1-30.Brooks, D. R., J. Collier, B. A. Maurer, J. D. H. Smith, and E. O. Wiley: 1989, 'Entropy
and Information in Evolving Biological Systems', Biology and Philosophy 4,407-432.
Brooks, D. R., D. D. Cumming, and P. H. LeBlond: 1988, 'Dollo's Law and the SecondLaw of Thermodynamics: Analogy or Extension?', in B.H. Weber, D.J. Depew, andJ.D. Smith (eds.), Entropy, Information, and Evolution, M.I.T. Press, Cambridge, MA,pp. 189-224.
Brylinsky, M.: 1980, 'Estimating the Productivity of Lakes and Reservoirs', in E. D.LeCren and R. H. Lowe-McConnell (eds.), The Functioning of Freshwater Ecosys-tems, Cambridge University Press, Cambridge, pp. 411-453.
Burger, C.: 1981, 'Why Are There so Many Kinds of Flowering Plants?', Bioscience 31,572-591.
Capra, F.: 1983, The Turning Point: Science, Society and the Rising Culture, BantamBooks, Toronto.
Clements, F. E.: 1905, Research Methods in Ecology, Nebraska University Press,Lincoln.
Connell, J. H.: 1978, 'Diversity in Tropical Rain Forests and Coral Reefs', The AmericanNaturalist 98, 399-414.
Connell, J. H. and E. Orias: 1964, 'The Ecological Regulation of Species Diversity', TheAmerican Naturalist 98, 399-414.
Coulter, G. W.: 1970, 'Population Changes Within a Group of Fish Species in LakeTanganyika Following Their Exploitation', The Journal of Fish Biology 2, 235-259.
Coulter, G. W.: 1976, 'The Biology of Lates Species (Nile perch) in Lake Tanganyika,and the Status of the Pelagic Fishery for Lates Species and Luciolates stappersii(Blgr.)', The Journal of Fish Biology 9, 235-259.
Curtis, J. T. and R. P. McIntosh: 1951, 'An Upland Forest Continuum in the Prarie-Border Region of Wisconsin', Ecology 32, 476-496.
Darwin, C.: 1859, The Origin of Species (lst Ed.), John Murray, London (Penguin).de Beer, G. R.: 1958, Embryos and Ancestors, Clarendon, Oxford.Dickie, L. M., S. R. Kerr, and P.R. Boudreau: 1987, 'Size-Dependent Processes
Underlying Regularities in Ecosystem Structure', Ecological Monographs 57,233-250.
Dobzhansky, T.: 1968, 'On Some Fundamental Concepts of Darwinian Biology', in T.Dobzhansky, M. K. Hecht, and W. C. Steere (eds.), Evolutionary Biology, Appleton-Century-Crofts, New York, pp. 1-34.
Dubos, R.: 1960, Pasteur and Modern Science, Anchor Books, Doubleday, New York.Dunbar, M. J.: 1972, 'The Ecosystem as a Unit of Natural Selection', Transactions of the
Connecticut Academy of Arts and Sciences 44, 111-130.Eldredge, N. and S. J. Gould: 1972, 'Punctuated Equilibria: An Alternative to Phyletic
Gradualism', in T. J. M. Schopf (ed.), Models in Paleobiology, Freeman, SanFrancisco, pp. 82-115.
Feynman, R .P.: 1963, 'The Principle of Least Action', in Feynman, R. P., R. B.Leighton, and M. Sands (eds.), The Feynman Lectures on Physics, Vol. 2, Addison-Wesley, Reading, Mass., pp. 19-1 to 19-4.
Fischer, A. G.: 1960, 'Latitudinal Variation in Organic Diversity', Evolution 14, 64-81.Frank, P. W.: 1968, 'Life Histories and Community Stability', Ecology 49, 355-357.Gauldie, R. W., I. F. West, and N. M. Davies: 1989, 'K-Selection Characteristics of
Orange Roughy (Hoplosthethus atlanticus) Stocks in New Zealand Waters', JournalApplied Ichthyology 5, 127-140.
Gaymer, R.: 1968, 'The Indian Ocean Giant Tortoise, Testudo gigantea, on Aldabra',Journal of Zoology (London) 154, 341-363.
Gnaiger, E.: 1986, 'Optimum Efficiencies of Energy Transformation in AnoxicMetabolism: The Strategies of Power and Economy', in P. Callow (ed.), EvolutionaryPhysiological Ecology, Cambridge University Press, Cambridge.
Golley, F. B.: 1965, 'Structure and Function of an Old-Field Broomsedge Community',Ecological Monographs 35, 113-117.
Golley, F. B.: 1972, Energy Flux in Ecosystems, Oregon State University Press,Corvallis, Oregon.
Gould, S. J.: 1977, Ontogeny and Phylogeny, Bellknap Press, Harvard.Gould, S. J. and N. Eldredge: 1977, 'Punctuated Equilibria', Paleobiology 3, 115-151.Grubb, P. J.: 1971, 'The Growth Ecology and Population Structure of Giant Tortoises on
Aldabra', Philosophical Transactions of the Royal Society, London, Series (B), 260,327-372.
Harper, J. L.: 1967, 'A Darwinian Approach to Plant Ecology', Journal of Ecology 55,247-270.
Harper, J. L.: 1969, 'The Role of Production in Vegetational Diversity', in G. M.Woodwell, and H. H. Smith (eds.), Diversity and Stability in Ecological Systems,Brookhaven Symposium on Biology 22.
Hirata, H. and T. Fukao: 1977, 'A Model of Mass and Energy Flow in Ecosystems',Mathematical Bioscience 33, 321-334.
Hirata, H. and R. E. Ulanowicz: 1986, 'Large Scale System Perspectives on EcologicalModeling and Analysis', Ecological Modeling 31, 79-104.
Hutchings, J. A. and R. L. Haedrich: 1984, 'Growth and Population Structure in TwoSpecies of Bivalves (Nuculanidae) from the Deep Sea', Marine Ecology ProgressSeries 17, 135-142.
Johnson, L.: 1964, 'Marine-Glacial Relicts of the Canadian Arctic Islands', SystematicZoology 13, 76-91.
Johnson, L.: 1972, 'Keller Lake: Characteristics of a Culturally Unstressed SalmonidCommunity', Journal of the Fisheries Research Board Canada 29, 731-740.
Johnson, L.: 1975, 'The Dynamics of Arctic Fish Populations', in Proceedings of theCircumpolar Conference on Northern Ecology, National Research Council, Ottawa,
pp. III-81 to 90.Johnson, L.: 1976, 'The Ecology of Arctic Populations of Lake Trout, Salvelinus
namaycush Lake Whitefish, Coregonus clupeaformis, and Arctic char, S. Alpinus, andAssociated Species in Unexploited Lakes of the Canadian Northwest Territories',Journal of the Fisheries Research Board Canada 33, 2459-2488.
Johnson, L.: 1981, 'The Thermodynamic Origin of Ecosystems', Canadian Journal ofFisheries and Aquatic Science 38, 571-590.
Johnson, L.: 1983, 'Homeostatic Characteristics in Single Species Fish Stocks in ArcticLakes', Canadian Journal of Fisheries and Aquatic Science 40, 987-1024.
Johnson, L.: 1985, 'Hypothesis Testing: Arctic Charr, Giant Land Tortoises, Marine andFreshwater Molluscs and Tawny Owls', Proceedings of the Third Workshop on ArcticCharr, Tromso, Institute of Freshwater Research, Drottningholm.
Johnson, L.: 1990, 'The Thermodynamics of Ecosystems', in O. Hutzinger (ed.),Handbook of Environmental Chemistry, Vol. 1, Part E, The Natural Environment andBiogeochemica Cycles, Springer-Verlag, Heidelburg, pp. 1-47.
Johnson, L.: in press, 'The Far-From-Equilibrium Ecological Hinterland', in B. C. Pattenand S. E. Jorgensen (eds.), Complex Ecology, Prentice Hall, Englewood Cliffs.
Jones, E. W.: 1945, 'The Structure and Reproduction of Virgin Forest of the NorthTemperate Zone', New Phytologist 44, 130-148.
Kleiber, M.: 1961, The Fire of Life, Wiley, New York.Koestler, A.: 1967, The Ghost in the Machine, Macmillan, London.Kondepudi, D.: 1988, 'Parity Violation and the Origin of the Biomolecular Chirality', in
B. H. Weber, D. J. Depew, and J. D. Smith (eds.), Entropy, Information, andEvolution, M.I.T. Press, Cambridge, MA.
Landsberg, P. T.: 1987, 'Entropy and the Unity of Knowledge', Journal of Non-Equilibrium Thermodynamics 12, 45-60.
Lindstedt, S. L. and W. A. Calder: 1981, 'Body Size, Physiological Time, and Longevityof Homeothermic Animals', Quarterly Review of Biology 56, 1-16.
Lotka, A. J.: 1922, 'Contribution to the Energetics of Evolution', Proceedings of theNational Academy of Science 8, 147-154.
MacArthur, J. W.: 1975, 'Environmental Fluctuations and Species Diversity', in M. L.Cody and J. M. Diamond (eds.), The Ecology and Evolution of Communities,Bellknap, Harvard, pp. 74-80.
Mann, K. H. and M. Brylinsky: 1975, 'Estimating Productivity of Lakes and Reservoirs',in T.W. M. Cameron and L.W. Billingsley (eds.), Energy Flow; Its BiologicalDimension, Royal Society of Canada, Ottawa.
Margalef, R.: 1968, Perspectives in Ecological Theory, University of Chicago Press,Chicago.
McIntosh, R. P.: 1980, 'The Relationship Between Succession and the Recovery Processin Ecosystems, in J. Cairns (ed.), The Recovery Process in Ecosystems, Ann ArborScience, Ann Arbor, pp. 11-62.
McIntosh, R. P.: 1985, The Background of Ecology: Concept and Theory, CambridgeUniversity Press, London.
McNamara, K.: 1988, 'The Great Evolutionary Handicap', New Scientist (16 September),47-51.
McNaughton, S. J. and L. L. Wolf: 1970, 'Dominance and the Niche in EcologicalSystems, Science 176, 131-139.
Newman, M. J. and R. T. Rood: 1977, 'Implications of Solar Evolution for the Earth'sEarly Atmosphere', Science 198, 1035-1037.
Odum, E. P.: 1969, 'The Strategy of Ecosystem Development', Science 164, 262-270.Odum, H. T. and R. C. Pinkerton: 1955, 'Time's Speed Regulator: The Optimum
Efficiency for Maximum Power Output in Physical and Biological Systems' AmericanScientist 43, 424-443.
O'Neill, R. V., D. L. De Angelis, J. B. Waide, and T. F. H. Allen: 1986, A Hierarchical
Concept of the Ecosystem, Princeton University Press, Princeton.Onsager, L.: 1931, 'Reciprocal Relations in Irreversible Processes', Physics Reviews 37,
405-562.Ott, J.A.: 1981, 'Adaptive Strategies at the Ecosystem Level: Examples from Two
Benthic Marine Systems', Marine Ecology 2, 113-158.Paine, R. T.: 1966, 'Food Web Complexity and Species Diversity', American Naturalist
100, 65-75.Parker, H. H. and L. Johnson: 1991, 'Population Structure, Ecological Segregation and
Reproduction in Non-Anadromous Arctic Charr, Salvelinus alpinus, in a Series ofUnexploited Lakes in the Canadian Arctic', Journal of Fish Biology 38, 123-147.
Pianka, E. R.: 1966, 'Latitudinal Gradients in Species Diversity', American Naturalist100, 33-46.
Pierce, J. R.: 1961, Symbols, Signals, and Noise: The Nature and Process of Communica-tion, Harper and Row, New York.
Platt, J. R.: 1964, 'Strong Inference', Science 146, 347-353.Polanyi, M.: 1968, 'Life's Irreducible Structure', Science 160, 1308-1312.Prigogine, I.: 1978, 'Time, Structure, and Fluctuations', Science 201, 777-785.Prigogine, I. and I. Stengers: 1984, Order Out of Chaos, Bantam, New York.Prigogine, I. and J. M. Wiame: 1946, 'Biologie et Thermodynamiques des Ph6nomenes
Irreversible', Experientia 2, 451-453.Provine, W. B.: 1982, 'Influence of Darwin's Ideas on the Study of Evolution',
Bioscience 32, 501-506.Raup, D. M. and J. J. Sepkoski: 1982, 'Mass Extinctions in the Marine Fossil Record',
Science 215, 1501-1503.Reiners, W. A.: 1972, 'Structure and Energetics of Forests', Ecological Monographs 42,
71-94.Saunders, G. W., K. W. Cummins, D. Z. Gak, E. Pieczynska, V. Straskrabova, and R. G.
Wetzel: 1980, 'Organic Matter and Decomposers', in E. D. LeCren and R. H. Lowe-McConnell (eds.), The Functioning of Freshwater Ecosystems, Cambridge UniversityPress, Cambridge.
Scheer, B. T.: 1970, 'A Universal Definition of Work', Bioscience 26, 505-506.Schneider, E. D.: 1988, 'Thermodynamics, Ecological Succession and Natural Selection:
A Common Thread', in B. H. Weber, D. J. Depew, and J. D. Smith (eds.), Entropy,Evolution and Information: New Perspectives on Physical and Biological Evolution,M.I.T. Press, Cambridge, MA., pp. 107-138.
Simon, H. A.: 1973, 'The Organization of Complex Systems', in H.H. Pattee (ed.),Hierarchy Theory, Braziller, New York, pp. 3-27.
Simpson, G. G.: 1949, The Meaning of Evolution: A Study of the History of Life and of ItsSignificance to Man, Yale University Press, New Haven.
Simpson, G. G.: 1953, The Major Features of Evolution, Columbia University Press,New York.
Simpson, G. G.: 1969, 'The First Three Billion Years of Community Evolution', in G. M.Woodwell and H. H. Smith (eds.), Diversity and Stability in Ecological Systems,Brookhaven Symposium on Biology, 22.
Slobodkin, L.B., F.E. Smith, and N.G. Hairston: 1967, 'Regulation in TerrestrialEcosystems and the Implied Balance of Nature', American Naturalist 101, 109-124.
Soodak, H. and A. Iberall: 1978, 'Homeokinetics: A Physical Science for ComplexSystems', Science 201, 579-582.
Stanley, S. M.: 1973, 'An Ecological Theory for the Sudden Origin of Multicellular Lifein the Late Precambrian', Proceedings of the National Academy of Sciences 70,1486-1489.
Stanley, S. M.: 1981, A New Evolutionary Timetable, Basic Books, New York.Swingland, I. R. and M. Coe: 1978, 'The Natural Regulation of Giant Tortoise Popula-
tions on Aldabra Atoll: Reproduction', Journal of Zoology (London) 186, 285-309.
Swingland, I. R. and M. Coe: 1979, 'The Natural Regulation of Giant Tortoise Popula-tions on Aldabra Atoll: Recruitment', Philosophical Transactions of the Royal SocietyLondon, Series (B) 286, 177-188.
Tamm, C. O.: 1948, 'Observations on Reproduction and Survival of Some PerennialHerbs', Botanical Notes 3, 305-321.
Tamm, C.O.: 1956, 'Further Observations on Reproduction and Survival of SomePerennial Herbs', Oikos 7, 273-292.
Vanriel, P.: 1989, 'Thermodynamic Relationships Between Trophic Levels in a SmallLacustrine Ecosystem in the Canadian Arctic', M.Sc. Thesis, Department of Zoology,University of Manitoba.
Vanriel, P. and L. Johnson: in prep., 'Thermodynamics and the Structure of Ecosystems:The Small Arctic Lake as a Reference System'.
Vogel, J. H.: 1988, 'Evolution as an Entropy Driven Process: An Economic Model',Systems Research 5, 200-312.
Vogel, J. H.: 1989, 'Entrepreneurship, Evolution, and the Entropy Law', The Journal ofBehavioral Economics 18, 185-204.
Waddington, C.H.: 1968, 'Towards a Theoretical Biology', Nature (London) 218,525-527.
Wallace, A. R.: 1878, Tropical Nature and Other Essays, MacMillan, London.Watson, A.: 1986, 'Physics - Where the Action Is', New Scientist (January), 42-44.Watt, K. E. F.: 1971, 'Dynamics of Populations: A Synthesis', in P. J. Den Boer and
G. R. Gradwell (eds.), Dynamics of Populations, Centre for Agricultural Publishingand Documentation, Wageningen, pp. 568-580.
Weber, B. H., D. J. Depew, C. Dyke, S. N. Salthe, E. D. Schneider, R. E. Ulanowicz, andJ. S. Wicken: 1989, 'Evolution in Thermodynamic Perspective: An EcologicalApproach', Biology and Philosophy 4, 374-405.
Weber, B. H., D. J. Depew, and J. D. Smith (eds.): 1988, Entropy, Information andEvolution: New Perspectives on Physical and Biological Evolution, M.I.T. Press,Cambridge, MA.
Werner, P. A., and H. Caswell: 1977, 'Population Growth Rates and Age Versus Stage-Distribution Models for Teazel (Dipsacus sylvestris Huds.), Ecology 58, 1103-1111.
Whittaker, R. H.: 1965, 'Dominance and Diversity in Land Plant Associations', Science147, 250-260.
Whittaker, R. H.: 1969, 'Evolution of Diversity in Land Plant Communities', in G. M.Woodwell and H. H. Smith (eds.), Diversity and Stability in Ecological Systems,Brookhaven Symposium on Biology 22, pp. 178-196.
Winberg, G. G.: 1972, 'Etudes sur le bilan biologique 6nergetique et la productivity deslacs en Union Sovittique', Verh. Internat. Verein. Limnol. 18, 39-64.
Winberg, G. G., V. A. Babitsky et al.: 1972, 'Biological Productivity of Different LakeTypes', in Z. Kajak and A. Hillbricht-Ilkowska (eds.), Productivity Problems in FreshWaters, Polish Scientific Publishers, Warszawa, pp. 383-404.
Zee, A.: 1986, Fearful Symmetry: The Search for Beauty in Modern Physics, Macmillan,New York.