Download 1987_mini_3

Not to be cited without prior reference to the authors.
..•
International Council for the
Exploration of the Sea
POPULATION REGULATION AND SPECIATION IN THE OCEANS
by
M. Sinclair
Department of Fisheries and Oceans
Scotia-Fundy Region
Halifax Fisheries Research Laboratory
Biological Sciences Branch
Invertebrates, Plants, and Environmental Ecology Division
Halifax, Nova Scotia B3J 2S7
Canada
."
and LD. Iles
Departmentof Fisheries and Oceans
Scotia-Fundy Region
Biological Station
Biological Sciences Branch
Marine Fish Division
St. Andrews, New Brunswick EOG 2XO
Canada
Abstract
•
The recruitment problem is a component of the broader question of,the
regulation of animal populations. The population regulation question is
rephrased to include geographie pattern, richness, abundance,"and .
variability. For marine fish species with complex life histories, the
specific patterns in spawning location, and the associated degree cf
population richness, are interpreted to be a function of the requirement for
coherence at the early life-history stages in the face of the diffusive and
advective characteristics of the oceans. Such observations generate a
population regulation hypothesis (Member/Vagrant) to' account for the four
defined characteristics. Two categories of constraints associated
respectively with food-chain processes and the physical geography itself are
defined. Losses from populations due to vagrancy are argued to frequently
limit successful life-cycle closure. Intraspecific competition for resources
for species with complex life histories is inferred to be often of minor
significance. The two categories of losses from populations are argued to be
the ecological mechanisms of respectively energetics adaptation and the
origin of reproductive isolation (speciation). Reproductiveisolation at the
population and species levels is inferred to be generated as a matter of,
course during population regulation without geographic'barriers; The
hypothesis accounts for why there are a specific number of. populations for
different marine species, and provides an ecological basis for the origin of
reproductive isolation between species.
.
2
•
•
Resume
Le recrutement fait partie du probleme plus vaste de la regulation des
populations animales. La question de la regulation des populations est
reformule pour tenir campte des facteurs de distribution geographique s
richesse s abondance et variabilite. Pour lesespeces de poissons marins au
cycle de vie complexe s 1'emplacement des frayeres etla richesse des
populations correspondante dependraient des besoins de coherence pendant les
premiers stades de vie en fonction des caracteristiques de diffusion et
d'advection de l'ocean. Ces observations ont permis l'elaboration d'une
hypothese de regulation des populations (membre/vagabond) tenant campte des
quatre facteurs definis. Deux categories de contraintes s associees
respectivement a la chalne alimentaire et,aux caracteristiques physiques du
milieus ont ete mises .en evidence. 11 est frequent que le succes de la
conclusion du cycle de vie soit limite a cause des poissons vagabonds qui
quittent les populations. La competition intraspecifique pour les ressources
chez les especes au cycle de vie complexe sembleetre un facteur de moindre
importance. Les deux categories de pertes encourues par lespopulations
seraient respectivement les mecanismes ecologiques de 1 adaptation
energetique et ceux qui sont a l'origine de l'isolement reproductif
(speciation). En 1'absence de barrieres geographiques s l'isolement
reproductif se produirait taut naturellement.au cours.du processus de
regulation des populations s et ces ta nt aU,niveau de la population que de
l'espece. L'hypothese explique pourquoi il existe un nombre specifique de
populations pour ,differentes especes marines et apporte un fondement
ecologique a l'origine de l'isolement reproductif des especes.
•
I
Introduction
One theme of the Mini-Symposium on Recruitment Processes in Marine
Ecosystems involves conceptual issues. This paper contributes to that
component of the Mini-Symposium~ Recruitment variability and its causes are
part of the broader problem of the regulation of abundance of populations.
In this speculative essay we attempt to provide an integrated conceptual
framework within which both physical-oceanographic and food-chain processes
have a direct role in the regulation of abundance. For a particular
populations it is to be expected that,certain processes are relatively more
important than others in generating losses from populations. There should s
however s be. a continuums from dominance. of physical oceanographic processes
to dominance of food-chain processes s in the generation of recruitment
variability for populations characterized by different life cycles. The
mechanisms of.population regulation proposed for the aceans are discussed in
an evolutionary context.
In ecologys the term "population" has more than one meaning. Most
frequently it represents the animals or plants of a species of interest in a
particular study area. As suchs the population is an abstraction of the
ecologist in relation to the particular question being addressed s rather than
a natural organizational unit. In other studies a population is a relatively
self~sustaining component of a species which s if accurately described s can be
defined without reference to an ecological question. It is this latter
definition of the term "population" that is addressed here. There are four
4It
3
aspects of such populations that may be usefully considered when the question
of regulation of,numbers is addressed - spatialpattern, richness, absolute
abundance, and temporal vai"iabilitY,,(Fig; 1). The first two aspectsare
species-1evel,characteristics. A' species is composed,of re1ative1ydiscrete
self-sustaining populations that,are distributed in particular spatia1
patterns. In the,oceans these spatia1 patterns in the distribution of
populations are frequent1y observedto persist on eco10gica1 time sca1es.
To'the degree that such patterns are notephemeral they shou1d be ,
interpretable; Further, there'are marked differences between species'in the
number of component,populätio!1s,,from arie (for, so-ca1,led parirriictic species)
to many; This species-1eve1 characteristic is defined here as population
richness~ ,
,
,
•
Of the four aspects of the population reguiation question, the ,latter
two are population-level characteristics. Some populations of a particular
species contain many.individua1s; others a much smaller number. , For'
individual marine fish species the range of absolute abundances between
comporient populations has been observed to cover severa1 orders of magnitude
(Blaxter 1985; Iles and Siric1air 1982). 'In the aceans, just as geographic
patterns in populations are observed to persist, populationsurider moderate
fishing pressure have been observed to have characteristic absolute
. abundances that also persist. Final1y, the number of individua1s in a
population varies from year to year, usua11y within an order cf magnitude for
age-structured,populations., It isthis 1atter aspect ofthe pop~lation, '
regulation question,because,of the economic impacts on the fishing industry,
that has received the bulk of the attention in the fisheries literature. It
may have been premature to have addressed this fourth aspect of population
regulationwithout prior understanding of the processes regu1ating pattern,
richness; and absolute abundance~'
,,
Evol uti onary bi 01 ogy and ecology overl ap1 n important, ways on the
population regulation question; Pattern and richness are the very features
of the bio10gica1 species concept that were a critica1 contribution by the
systematists to the evol~tionary synthesis (Chetverikov 1926; Renschl929;
Dobzhansky 1937; Mayr 1942).' The morphological differences between, ,
populations were interpreted as evidence of speciatian in action, and the
geographical dis~ontinuities as barriers to gene flow. However, the
between-species differences,in population richness have not been accounted
fore Their interpretation may thus be of general interest. In addition, the
processes irivolvedin the regulation of abundanceare hypothesized tri be
central toadaptation and speciationthrough natural se1ection; Thus, a
clearer,understanding of the processes regulating abundance may have
implications on evolutioriary questions.
,
,
,
The äbove-defined aspects of thepopulation ,regulation question were
recently addressed for Atl anti c herri ng CI 1es and Si nc1 ai r 1982) ~ Here the
concepts developed for,herring are genera1ized for sexually reproducing,
marine species having comp1ex life histories. In comp1ex life histories,
eggs, larvae, arid juveniles,have,identifiable geographic or spatial
distributions'that are usually different than that characteristic rif the
adult phase. _Population regulation is hypothesized to bei comprised of two
sorts ofconstraints - ,those generated by respectively the physical geography
itself and the food-chain processes. ' It is argued that competition for
4
•
limited resources, even ultimately, is not necessary for population
regulation of sexually reproducing marine specieshaving complex life
histories. Speciation and food-chain-related adaptation are considered to be
generated respectively by these two separate, albeit contemporaneous,
components of population regulation.
First a summary of population patterns,and richness for marine fish
species in the northern Atlantic is made. The marine fisheries literature
provides a unique accumulation of empirical observations on the four
components of populations that are being addressed~- Since the turnof the
century the population has been the preferred unit of study; andsampling
programs have been designed on the appropriate time and space scales for
their analysis. Such information on a range of species does not exist for
any other animal category. These observations suggest a hypothesis on
population regulation.
-Marine Fish Populations
The fisheries literature clearly demonstrates that there are marked
differences between species in population richness (Sinclair and Iles 1986;
Sinclair 1987). There is a continuum in this characteristic of species, from
Atlantic salmon, for,example, which is highly population rich, to Atlantic
cod (moderately population rich), to Atlantic,mackerel (population poor) , to
European eel (a single population throughout the distributional limits of the
species). In a comparative analysisit is clear that in 'general terms the
degree of population richness is defined ,at the early life-history phases in
relation to well-defined physical oceanographic (or geographic) features
(rivers, estuaries, coastal embayments, tidal-induced circulation features,
banks, coastal current systems, major ocean currents, oceanic-scale gyres).
When there are manysuch physical features (to,which the early life history
is behaviorally adapted to permit retention) within the distributional range
of the species, the species is population rich. , Withfew appropriate
physical features ~ the speci es i s popul ati on poor. The argument i s; not
tautological in the sense that the numbers of populations havebeen'defined
in most cases independently of the early life-history distributional
descriptions.
The evidence is not as complete as one would want for any particular
species, but when a comparative approach is taken the evidence is persuasive.
For Atlantic salmon, American shad, and striped bass for which the egg and
larval phases are completed in a particular river system, that river system
itself supports a relatively self-sustaining population. For smeltj whose
eggs and larvae drift from the individual spawning rivers to a common
downstream estuarine or coastal zone larval distributional area, the complex
of rivers flowing into the larval retention area (rather than the individual
rivers) is the geographic unit which supports a self-sustaining population.
Atlantic herring is an open-ocean species characterized by many
populations. The very existence of populations, and the underlying reason
for their precise locations of spawning, has been hypothesized to be a
function of the existence arid location of geographically stable or
predictable larval retention areas (Iles and Sinclair 1982). Atlantic
•
•
5
,
herring fixes its eggs to the bottom of the ocean in relation to a variety of
temporally persistent water column recirculation features. There are many
such physical oceanographic features within the distributional limits cf the
species, 'and it is hypothesized that for this reason there is the potential
for many populations.
,
•
•
"
....
,
, Atlantic cod is an open-ocean species which is moderately population
rich. ,Eggs ~nd larvae,for populations of this species are interpreted to be
retained on offshore banks and within coastal embayments and fjords. Again,
there are many such physical oceanographic systems (although less numerous
than,for the bottom-spawning herring)~ and the potential for many
populations. Atlantic,mackerel, in contrast; is population poor~ with only, .
five or six populations within thedistributional limitsof the species. The
eggs and larv~eof populations ofAtlantic mackerelareretained within
large-scale seas and bights, of,which there are only a fewin the northern
Atlantic~ European and American eel represent the oppositeextreme from
salmon, shad, and herring. Thereis but a,single population for each of the
eel species within their overall .distributional limits. Eggs and larvae of
these eel species are initially retained within the North Atlantic Central
Gyre~ ,There is only orie such'physical oceanographic system within the
distributional limits of the ,two species;
. The extremes in the'continuum forcefully and simply demonstrate that it
is the earlylife-history discreteness itself, in conjunction with the
ability of adult fish to horne, that defines the population riehness.The
life,eyeles of Atlantie salmon andEuropean eelare almost mirror,images, arid
for these two speei es popul atiOn ri ehness i s generated by the early
life-history discreteness rather than at the juvenile phase and the ,
non-reproductive components of the ,adult phase. There is extensive mixing
between populations at the adult phase for salmon, 'and isolation of parts of
a panmietic population at the,adult phase of eel. It is the continuity/
discontinuity at theearly life-history phase that. defines; with homing for
reproduetion, tlie population ,riehness. There,is in additionsufficient
evidence to joiri these extreme examples, and thus describe the,interspeeies
patterns in population richness as a eontinuum (Fig~ 2)~ ,In the past ,the
anadromous/eatadromous life-history distributions have not been eonsidered as
part of a life-history eontinuum"but as somewhat of a paradox (Baker 1978).
From the continuum perspective introduced here the underlying eonstraints on
population riehness (in this case from one to many populations) are simply
interpreted.
Perhaps it is useful to view the population riehness,charaeteristie of
, marine speeies from a plankton perspeetive. The other life forms of a
population (post-larvae,juvenile, and adult phases) are from this view parts
of a eyele that are neeessary for the long-term temporal persistence,of the
larval distribution in partieular geographie spaee. This shift in emphasis
from adult to larvae puts homing; whieh permits population riehness, into a
different perspeetive;' Homing of,adults to specifie spawning sites al10ws
temporal,persistenee of the ,larval distribution,in particular geographie
spaee~ The weak linki;' theeyele of many species may be the planktonie
phase~, For anadromous species; however, the self-contained nature of rivers
reduces the dispersive losses duringthe egg arid ,larval phases; and the '
homing of adults to their natal river may well be the major challenge to
6
life-eyele elosure •.The ability of the egg and larval eomponents of the life
eyele of a population to persist and maintain an aggregated distribution at
shorter time seales (weeks to months) is the key, in this view, to the
population riehness question. The spatial seale at whieh the plankton phase
of the life cyele ean be coherent, andthe corresponding number of physical
oeeanographie (or geographie) systems underlying thateoherenee within the
distributional limits of the speeies, defines the degree of population
richness. Further, it is the particular behavioral charaeteristics of the
speeies (both short term within a partieular stage of the life eycle and the
changes in behaviorat different life~history stages) in relation to the
physical oceanographic features of the environment that define the spatial
scales of coherence at the planktonic.phase of the life eyele. It is clear
from a comparison of the elosely related speeies of Atlantie eod arid haddock
(cod is considerably more population rich) that subtle differences in
behavior must be involved.
It is this area, mechanisms of persistence of plankton distributions in
partieular geographie space (be it an upwelling zone, the fringes of an
oceanic island, or an estuary), that is well developed in the flanking
literature to fisheries biology. Representative studies from respeetively
the estuarine, the oeeanie island marine fauna, and,marine zooplankton
literature amply demonstrate that plankton ean persist in relatively fixed
geographie locations in spite of the average residual eireulation (the
literature is reviewed .in Sinclair 1987)~ It is inferred that the animals
are much more coneerned about the varianee in the eireulation and mixing
characteristies of their partieular environment, and the vertieal structure
of these features, than they are with the depth-averaged residuals.
.
•
.'
.
In sum, the fisheries literature indicates that population richness is
defined at the planktonie phase of thelife eycle. The zooplankton
'literature on population ecology, and the speeial environments literature
(estuaries and oceanie islands, which are amenable to detailed study),
indieates how plankton ean use behavior in relation ,to the partieular physics
of thearea of distribution to persist in a highly diffusive environment.
From consideration of the combined literature; both the questions of
population riehness and pattern may in· general terms be interpreted.
The empirieal observations on the geographie pattern arid richness of
marine populations suggest a hypothesis on population regulation~ Before the
hypothesis is stated the constraints imposed on population dynamies by sexual
reproduction are discussed.
Distributional Order Out of Chaos
Population patterns in geographie space, in partieular for mobile
species, are not obviously defined in relation to geographie barriers.
Individuals from diverse populations routinely share a common distributional
area during non-reproductive phases of both the life eyele and the annual
adult migration eyele. Yet at spawning time the animals have been observed
to aggregate in particular spaWning populations. Animals for. many, if riot
most, sexually reproducing speeies are members of the speeifie populations of
their birth. At maturity they horne to the appropriate geographie area for
•
7
their particular population. Thus, opportunities for life-:-cycle closure; ..
rather than, geographie barriers between spawnlng eomponents, may define, the
spatial patterns in spawning populations. On eeologieal time seales,there is
abundant empirieal evidence in the marine fish literature that relative,
reproduetive isolation betwee~ populations is sustainedwithout geographie ,
barriers. Other proeesses must be involved. The eonstraint of free~erossing
for sexually reprodueing species maY,itself,ultimately generate the diverse
patterns in populations observed in the oceans~
AreeeritlY proposed hypothesis for the,use of sex provides,a useful
perspective for,eonsideratiori of the definition of these spatial ,patterns of
spawni ng popul ati ons in the o.ceans. Sexual reproducti on i s the foeus of the,
biologieal speeies eoneept itself. Speeies are defined in relationto sexual
reproduetion, arid speeies usually eonsist of groups af.populations (whieh are
themselvesdefined in relation to free-crossing)~ Inthis sense an .'
understanding of the "use of sex" is germane to both the species arid the
population question. The repair hypothesis addresses tWo problems in.
replieatirig,genetie information: DNA damage, and mutation (Bernstein et ale
1985). It is argued that the two principal features of sex, reeombination
'arid outerossing, are maintained respectively by the advantage ofrepairing
damage arid masking mutations. The importance of this eonceptualization of,
the origin and predominance of the sexual mode of reproduetion on the species
eoneeptwas also stressed (Bernstein et ale 1984). This line of thought is
followed at the population level. The argument is not speeifieallydeperident
on the repair hypothesis; bu~ rather that sex provides some advantages.
•
..
-,
The ,evolution of sexual reproduct10n in m~rine.orgariisms several b1liion
years,ago, with the associated eonstraint of free~crossing of genetie,
material, introdueed a new mode of seleetion. The primary ehallenge to
~ersistenee of a population of a given marine ~pecies became not just
survival ,of the individual to maturity, as isthe.ease for asexual speeies,
but the ability to find a mate having a similar, genome at this time. ,In
arder to introduce the implications of this eonstraint, and its dramatie
.
impact on distributional pattern, populations of species whieh eomplete their
life eyele as plankton are easier to eonceptualize. Subsequently, the
.
concepts can be enlarged to include life cycles involving either a nekton
phase, (fish and pelagie invertebrate speeies) or a relatively fixed phase at
the sea floor interface (diverse benthie species).
In the highly dispersive oeean environment, which is also dilute,from
the animals' perspective~ it is not diffieult to envision that with.time,
subsequent tO.birth of a small planktonie form the chance of encounter,with
another individual with similar genetie material deereases monotonieälly.
Diffusion,itself fram a pointsöurce for.non-mobile drifting organisms, or
randommovement,for moöile organisms, will minimize the frequeney of sexual
encounter that is necessary to allow persistence of the population. A new
phenomenon in the evolution of life,thus:was,generatedeoineident.with the
emergence of sex - the eonstraint of physieal ,relationship between
,.
individuals of populations to ensure persistence. More fundamentally, the
very existence of populatio~s as. defined in relation to the shared gene pool
emerged with the evolution of the sexual mode of reproduction~ Following the
interpretation of the repair hypothesis for the use of sex (or alternate
hypotheses on the advantages of the sexual mode of reproduction); the
8
advantages of masking of ehemiea11y generated variability inthe gerietie
material (through the sexual modeof reproduetion) was gained at the eost of
freedom of the individual in geographie spaee. Spaee re1ationships with
ethers of· a simi1ar genome sudden1y beeame a eonstraint.
C1ear1y any behavior,by an anima1 that will enhanee retention of the
individua1s of the population in re1ative1y fixed geographie spaee in
proximity to the souree of reproduetion itse1f, and,thus inerease the
probability of sexual eneounter, will be high1y se1eeted fore Surviva1
itse1f is not the on1y issue; rather, finding a mate in a diffuse environment
at di1ute eoneentrations beeomes the additional, perhaps more-eritiea1,
eha11enge. The anima1' that out-eompetes others for resourees may survive
longer; but if a mate iS.not eneountered, obvious1ynothing in the genome ean
be se1eeted fore This simple eoneept is the starting point of the member/
vagrant hypothesis of population regulation.
The bifureation from asexua1 to sexual reproduction generated severa1
emergent properties: 1) a new mode of se1eetion, that invo1ving ensuranee of
encounterwith a'mate at maturity; 2) the generation of appropriate responses
between individua1s of similar genomes (i .e. re1ationa1 behavior of like
individua1s); 3) the generation of persistent population patterns in
geographie space;and thus 4) the generation of the bio10gica1speeies
itse1f, or the Rassenkreis of Rensch. As a resu1t of the additional mode of
se1eetion, 1ife eyc1esof sexua11y reproducing anima1s may be eonsidered to
be primari1y meehanisms toensure persistence of populations in specifie
geogra~hie space. Theseeond phenomenon invo1ves the diverse behaviora1
eharaeteristics of anima1s in relation to what Patersan (1982) has usefu11y
defined as the Speeific Mate Recognition System (SMRS)~ The third and fourth
phenomena, the generation of persistence in population patterns on an
eeo1ogiea1 time sea1e, are start1ing. In a very real sense it ean be said
that fo110wing the evolution cf sex, order (in this ease the generation of
persistent geographie patterns in population structure based on eircu1ation
and mixing features of the oceans) was ereated out of ehaos (the inferred,
eontinua11y ehanging geographiea1 patterns that are generated by asexua1
reproduetion).
The Member/Vagrant Hypothesis on Marine Population Regulation
There are two points in a preamb1e to the population regulation
hypothesis:
1. Due to. the constraint of free-:crossing in sexua11y reprodueing species in
the oceans, theeomp1ex 1ife histories that are frequent1y observed are
eonsidered "1ife-eyc1e solutions" that ensure temporal persistence of
populations in re1ative1y fixed geographie space.
2.
Populations ea" exist only in those geographie loeations within whieh
there can be c10sure of the 1ife,eyc1e [i.e. in a geographiea1 setting
within whieh retention (membership) exceeds lasses (vagrancy) in same
integrated sense for the life eyele as a whole].
•
9
The member/vagrant hypothesis comprises three statements:
1.
Population pattern and r1chness are functions' of the,number 'and'location
of geographlc settings (wlthln the overall distributioria1"area of the
species) within which the particular1ife cycle is capable of closure~
2.
Absolute abundance 'is scaled according to,the,size of the geographic area
to which there is c10sure of the 1ife cycle of the free-crossing
popul ation;
,
•
•
'
3. Temporal variabi1ity in abundance is a function of the intergenerational
lossesof,lndlvlduals(vagrancy and morta1ity) from the appropriate
distributional area that will ensure membership within a given
population. For populations witha planktonic stage-ofthe·life-cyc1e
physical oceanographic processes may dominate in gerierating the,
variabi1ity in losses from the appropriate distributiorial area •.
The hypothesis emphasizes that membership ina popu;ation1n the,oceans
requires being in the appropriate place at,the various parts of,the ,life
eYc1e. It implies that anima1s can be lost fromtheirpopu1ation~ arid'thus
become vagrants. Life cycles are corisidered continuity solutions within
particu1ar geographical settingswhich impose spatial constraints.,. This
concept of complex 1ife histories of populations is represented schematica11y
in Fi gure 3. At each stage of the 1ife cycl e there are opportuni ti es to
become lost,from the population. The area for successful reproduction is
considered to be geographically constrained in relation to the life cycle as
a whole. In this illustration sexual reproductionatother geographic
locations along the central line does not allow the progeny to recruit to the
population.
-
Aconvoluted trend in the evolution of life histories involves the .
internalization,of eggs, larvae; and juveniles within, or closely attached
to, mature adults~ ,This process is in acertain sense a collapse of complex
life cycles that reduces theimplications of,the cost of rarity that is . "
associated with the sexual, mode of reproduction; .It is'tö beexpected that
the constraints of the physical oceanographic environment to life-cycle
closure, and thus the relative magnitude of vagrancy, would be less of a
problem with populations of species characterized by collapsed life cycles
(i.e. amphipods, sharks, and marinemammals).The trend in the collapse of
1ife cycles is more pronounced at interfaces, but poorly developed, in the
pe1agic realm of the oceans. Due to,this condition, vagrancy, might be
expected,to be of considerable importance in the open ocean, but perhaps of
less importance at interfaces.
Two categories of losses from the population can be defined: .losses due
respective1y to "spatia1" and"energetics" processes. We define energetics
processes to include predation; ,disease, and starvation" (and their
','
interactions). Suchlosses from a marine sexually reproducing population
with a complex 1ife history are listed in Tab1e 1. The losses from a .
population due to spatial .processes,are indicated with an asterisk. It is
inferred that· for marine species with complex life' histories..the proportion
of the totallosses from the population dueto spatial processes maybe much
1arger than losses due to energetics processes. Further, it is noted that
10
density dependence in A* + C* + E* + G* + 1* can be sufficient for population
regulation. There is no necessary requirement for limitation of abundance of
the population by food availability or predators [Bernsteln et al. (1985)
have already made this point]. Losses due to spatial processes.can be
sufficient for the regulation of absolute abundance and temporal variability.
For this to be the case all that is required is that there be some
density-dependent losses associated with the spatial processes themselves.
Vagrancy has to be density-dependent on occasion. The existencein principle
of this sort of control is i1lustrated in Figure 3. by the liniited space for
successful sexual reproduction~ Individuals reproducing outside of the
appropriategeographic area for continued membership of the progeny in the
population do not contribute to .th~next generatiori.The point being made
here is an extreme one. It is not being suggested that food limitation,
predation, and disease are not important in the oceans. Rather, it is being
argued that such so-called energeticsprocesses are not required for
population regulation. All of the density-dependence couldbe generated by
spatial processes alone. Further, it is stressed that a large component of
the losses from the population due to energetics processes may be density
independent, just as a proportion of the losses due to spatial processes may
be density dependent. This point is illustrated in Figure 4. Both
categories of losses from populations can be either density independent or
density dependent.
It is perhaps useful in a conceptual sense to identify separately the
two categories of losses from sexually reproducingpopulations (i~e. losses
due respectively to spatial versus energetics processes). Spatial losses do
not involve competition for food resourcesor avoidance ofpredators. Yet
they can be density dependent. In this sense carrying capacity of a
population in the marine environment could be defined below the level of food
limitation and without predator control. Density-dependent mortality
resulting from biological interactions involving food or predation is not in
principle required for the definition cf absolute abundance (i.e. carrying
capacity), or for stability and persistence. This conclusion·is
counterintuitive if population patterns are not considered in relation to
geography. Once life cycles of populations are associated with particular
features of the physical geography imposed by the constraint of
free-crossing, however, the argument is a simple one.
Leaving out individuality (individuals are members of particular
populations whichrequire being at the right place at the right time) and
space may have narrowed the focus on population regulationquestions in an
inhibiting manner. ' The two types of processes (i;e. energetics and spatial)
may be involved in separate aspects of evolution. In this conceptualization
the energetics or food-chain processes lead to intraspecific competition for
resources. This component of the regulation of populations generates what
may be usefully defined as food-chain selection for a certain class of
adaptations assöciated with energetic constraints, whereas the spatial
processes generated by the constraints,of free-crossing in sexually
reproducing animals result in relational events (membership, vagrancy). This
component of the regulation of populations is considered to generate what may
be defined as life-cycle selection. The spatial component of the total
losses from populations has the potential to select directly for reproductive
isolation (i.e. speciation) without the requirement of geographic barriers.
~
~
~
"
11
Rep)icated physical,oceanographic features that,permit pöpulation persistence
provide opportunities for,life-cycle closure rather than generating barriers,
between populations. Life-cycle selection as conceptualized herein relation
töthe spatial losses listed in Table 1 includes sexual and mate selection;
Oarwin (1859) defines sexual selection as being ,a separate process from
natural selection that does not involve the struggle for existence. He
states,
'
IIThis depends, no't onthe struggle for existence~ but on a
struggl e between the mal es for possessi on of the femal es; the
result is not death to the unsuccessful competitor, but
few or no offspring.
1I
•
In the, same sense, life-cycle selection involves a metaphorical struggle to
, maintain membership in the population, but does not necessarily'involve
mortality at thetime of vagrancy. Life-cycle selection does not involve
intraspecific competition for limited food resources or the avoidance of
predators. These contemporaneous population regulation processes, and their
association,with so-called energetics adaptation and speciation, are.
summarized in Figure 5. ,The spatial selection concept, and its putative role
in speciation, is now considered.
Spatial Selection Concept arid Speciation
,
,
There are considered to be four constraints involvedin,speciation of
sexually reproducing animals.
1.
Information for metabolism and developmerit)s chemically mediated,and is
in a discrete rather than,a continuous form. This in,itself leads,ta
particulate inheritance of information across generations.
2.
Biochemical information degrades through time. The replication of
information is not perfect.
3; Sexual reproduction between ind~viduals serves the ~unction of.conserving
information content, or of retarding or masking the errors in its '
repl ication.
4. The geographical ,or spat1al framework underlying the ~elationai,phenomena
between similar individuals of populations (because of the constraint of '
sexual reproduction) changes due to non-equilibrium physical conditions
in the oceans at a range of time scales.
From the starting point of,these foür'constraints~speciation is'perceived to
be the result of trying not to change,in the face of dissipation of
informatiori in the biochemical hereditary material,and a shiftingspatial
matrix within which populations are striving to persist. The spatial matrix
includes the physical geography as well aS,the,structure provided byother
species of plants and animals. Life cycles, as statedabove, and thus
population patterns of sexually reproducing animals; are considered in the
member/vagrant hypothesis tO,be defined in relation,to particular
geographical or spatial constraints thatensure persistence~ It is this
12
spatial infrastructure for the population that is eonsidered to change
ehanges in vertieal structure of the water eolumn or in eirculation
patterns) •
(i.e~
Life-cycle selection is a logical extension of the member/vagrant
hypothesis on population regulation, just as ,the originally defined eoncept
of natural selection is an extension of the Malthusian doctrine and the
hypothesis on population regulation through competition for limited
resources. The member/vagranthypothesis focusses on the losses from
populations during eyclical sexual reproduction in a fixed spatial eontext;
life~cycle selection focusses on the changes infrequency distribution of
traits in populations generated by membership versus vagrancy during
life-cycle closure. In essence what is selected for is the ability to
continue to be a member of a spatially defined population.The emphasis is
on those aspects of. the 1i fe eye1e as a who1e that ensure popul ati on
persistenee in relation to the,spatial dimension. As stated above,
life-cyele selection eneompasses sexual seleetion and mate seleetion whieh
deal with reproductive proeesses at.the adult stage of the life eyele
(ensuring eoupling of mates in partieular geographical spaee). It also
encompasses distributional and behavioral phenomena at other stages of the
life cycle. It acts at the individual level. This spatial selecticin process
is argued to be a mechanism contributing to speeiation in the oeeans with or
without geographie barriers.
The eoupled observations of relative stability of numbers of individuals
of a particular species and high reproduetive output led Darwin and Wallaee
to infer or deduce the existenee of intense eompetition for limited resources
between indivi dual s of the same speeies J and the avoi dance of predati on).
This inferenee is the basis of,natural seleetion. It is argued here that the
inference or deduction was partially flawed in the sense thatin the oceans
there is considerable evidence that the inferred intense eompetition for
limited resourees does not necessarily take place, and that spatial losses
from the population may account for much of the excess feeundity. A portion
of the accumulated life-eycle mortality is due to intraspeeifie eompetition
for resources and predation, but it is argued to. be a small portion for many
speeies with eomplex life histories. This very minor shift in emphasis from
food-chain eonstraints to spaee has subtle but perhaps signifieant
implications. Reproductive continuity andits opposite (reproduetive
isolation) are the very foeus of both the member/vagrant hypothesis of
population regulation and the life-eyele selection eoncept of the'origin of
reproductive isolation (i.e. speeiation). Reproductive isolation reinforced
by life-eycle selection is not an incidental by-produet of energetics
adaptations of other aspects of the life history (as is presently the case
within the evolutionary synthesis), but is the very foeus of the meehanism of
population regulation. Speciation is a result'of life-cyele selection in a
non-steady-state system, with degradation and changes in information
internally at the biochemical or molecular level, and physieal changes
externally at the spatial orgeographicallevel; energeties adaptation is a
result of food-chain selection involving differential mortality associated
with energeties processes.
. life-cycle selection is a conservative process generating persistence of
populations of sexually reproducing animals. Two creative processes are
•
•
'"
13
I'
I
•
..
e
14
It is inferred above that the subtle shift inemphasis on the nature of
population regulation in the oeeans, involving life-eyele eontinuity'in
relation to physieal geography, and the related eoneept of life-eyele
seleetion, may help aeeount for several eharaeteristies of speeies (and thus
speeiation!extinetion proeesses). First, life-eyele seleetion aets direetly
on the reproduetive proeess. Reproduetive isolation and sterility of hybrids
are the eentral foei of speeiation. The meehanism of the origin of
reproduetive isolation and assoeiated hybrid sterility is a logieal extension
of the member/vagrant hypothesis of population regulation (just as
morphologieal adaptations are a result of intraspeeifie eompetition for
limited resourees and predator avoidanee). Genetie drift within populations
reinforeed by life-eyele seleetion (i.e. the selective lossesfrom the
appropriate populations due to differential vagraney}may in itself generate
reproduetive isolation. With population regulation being eonsidered a
spatially limited phenomenon, rather than a eompetitive one, the meehanism of
the origin of reproduetive isolationis the foeusof .the seleetion proeess
for speeiation (and not an ineidental by-produet of other adaptations). The
eeologieal proeesses underlying the evolution of reproduetive isolation
durlng speeiation were not satisfaetorily addressed in the evolutionary
synthesis (Dobzhansky 1937, Chapters 8 and 9; Mayr 1942, p. 270-273).
In addition, population riehness as a key charaeteristie of speeies
(from one, to many populations) is simply interpreted underthe member/vagrant
hypothesis. One does not have to invoke the existence of different genetie
systems for monotypic species as has been hypothesized by Carson (1975) to
. account.for aspects of population richness. Richness is a function of
replieation of the appropriate physieal basis required for life-eyele
elosure •. Also, the absolute abundance eharaeteristie of populations and
species is eonsidered to be a funetion of the spatial scale of the life cycle
that permits eontinuity and persistence.
Coneluding Remarks
The member!vagrant hypothesis addresses eomponents of the population
regulation question that have rarely been eonsidered in the reeent eeologieal
literature (i .e. the differences. between speeies .in the geographie patterns
of the eomponent populations as well asthe marked species-speeifie
differences in population riehness). The debate on population regulation,
whieh peaked inthe 1950's and 1960's, predominantly focussed on the relative
importanee of food-ehain processes versus physical environmental processes on
the control of density. Self-sustaining populations having characteristic
absolute abundances were not usually the foeus of study. The debate was not
resolved. It may well be that the question of the regulation of abundanee
was not suffieiently deeomposed to permit its resolution. From the
perspeetive developed here,.physical processes can be associated with density
dependence in vagrancy, just as food-chain processes can have a densityindependent effect (Fig. 4). The member/vagrant hypothesismay provide a
useful framework to address the density-dependenee!density-independenee issue
in the regulation of abundance.
The trend in the evolution of the collapseof eomplex life cycles,
involving the internalization of early life-history stages, reduces the
•
. '.
'.
",
,~
' . ; ....
~
.... too
0"
-II'
I
.15
..
•
-,4"-,
,-
e
i~~ortance of~patiai processes on popul~tion regulation~ Inth~ pelagic
environment, this trend, however, has not been pronounced, with complex life
histories being predomi.nant. ~"Thus, .spatial processes should be frequently
critical to population regulation~"At'interfacei the internalization is more
frequently observed~ One wouldexpect, as the depenrlence on spatial
processes for life-cycle closure is minimized, that food-chain processes
would beniore im~ortant .in population regulation; In parallel, evolutionary
everitsmay be associated with morphological changes reflectingfood-chain
specialization; A comparison of theresults of evolution in amphipods (which
have a partially collapsed life cycle) to copepods (which do not) is .
supportive~ Relative freedorn.from the constraints of free-crossing by the
collapse of life cycles should be associated with a gre~ter role of
food;chain processes in the regulation of abundance and thus a larger role
for energetics adaptations in evolution; Following this reasoning, the
member~vagrant hypothesis predicts that amphipod and shark speciation
(involving s~ecies with relatively collapsed lifecycles and a greater role
offood-chain selection relative to life-cycle selection) should reflect more
morphological adaptations than, say, copepod and gadoid speciation; It is of
interest that Frost and Fleminger (1968), Fleminger (1967), and Fleminger and
Hulseman (1974) indicate that pelagic copepod speciation frequently has
involv~d.no obvious food~chain-related adaptations. Not to puttoo fine a
point on it atthis stage, the .relative role of food-chain and spatial losses
on population regulation should vary in relation to the degree.of IIcollapsell
in the ·life cycle~ Thisreasoning rnay provide a simple test of.the
member/vagranthypothesis (i;e. an analysis of themorphological differences
betweeri species of genera having respectively complex and IIcollapsed" life
cycles).
Themember/vagrant hypothesis and the associated concept of sp~ti~l
selection may also clarify one of the enigmas of the evolutionarysynthesis.
Evolution involves both adaptation,and speciation. Food-chain selection
(which .is synonymous ,with the original Darwinian definition of natural .
selection), involving intraspecific competitionforlimited resources and the
avoidance of pr~dators, addresses the ecological component of the processes
generating energetics adaptations; It does not address the ecological
component of the origin of reproductive isolation"th~ key component of
speciation itself. Rephrasing the population regulation question to include
both spatial and energetics processes; the former imposed by the constraint
of free-crossing, may clarifythis enigma within the syrithesis. Geographie
barri ers niay" not necessarily be requi red for speci ati on under 1ife-cycl e
selection. Rather, oceanographic and geographie features provide distinct
opportunities for life-cycle closure of populations. The effect iS,a degree
of lsolabon between populations at spawning if, there is more than,a single,
appropriate physical infrastructure available within the distributional range
of the speci es., Further, the empi ri calobservati ons on popul ati on patterns
ingeographic,space may not be evidence of speciation in action as inferred
within the sYnthesis.
.
References
Baker, R.R. 1978. The Evo1utionary.Eco10gy of Animal Migration. Holmes and
Meier Pub1ishers, New Vork:36-43;
.,
16
Bernstein, H., H.C. Byer1y, F.A. Hopf, and R.E. Michod. 1984.
J. Theor. Bio1. 110:323-351.
Origin of sex.
------------------------------------------------------- 1985. Genetic
damage, mutation, and the evolution of sex. Science 229:1277-1281.
B1axter, J.H.S. 1985. The herring:
Aquat. Sci. (Supp1. 1):21-30.
A successfu1 species? Can. J. Fish.
Carson, H.L. 1975. The genetics of speciation at the diploid level. Am.
Nat. 109:83-92.
Chetverikov, S.S. 1926. On certain aspects of the evo1utionary process from
the standpoint of modern genetics. J. Exp1t. Bio1. (Russian) A2:3-54.
[Eng1ish trans. (1961). Proc. Amer. Phi1. Soc. 105:167-195.]
Darwin, C. 1859. The Origin of Species (first edition).
Library Issue 1982:416 p.
Penquin Eng1ish
Dobzhansky, Th. 1937. Geneticsand the Originof Species. Co1umbia
University Press, New Vork:364 p.
. .
F1eminger, A. 1967. Taxonomy, distribution, and po1ymorphism in the
Labidocera jo11ae group with remarks on evolution within the group
(Copepoda:Calano1da). Proc. U.S. Nat. Mus. 120:1-61.
F1eminger, A. and K. Hu1semann. 1974. Systematics and distributions of the
four sib1ing species comprising the genus Ponte11ina Dana (Copepoda,
Ca1anoida). Fish. Bu11. 72:63-120.
Frost, B. and A. F1eminger. 1968. Arevision of the genus C1ausoca1anus
(Copepoda:Ca1anoida) with remarks on distributiona1 patterns 1n
diagnostic characters. Bull. Scripps Instit. Oceanogr., Univ. Calif. San
Diego 12:1-99.
Hardy, A.C. 1954. Escape from specialization, p.122-142. In J. Hux1ey,
A.C. Hardy, and E.B. Ford, eds. Evolution as a Process.--George Allen &
Unwin Ltd., London.
I1es, T.D. and M. Sinc1air. 1982. At1antic herring: Stock discreteness and
abundance. Science (Wash., D.C.) 215:627-633.
Mayr, E. 1942. Systematics and the Origin of Species. Columbia University
.
Press, New York:334 p.
Paterson, H.E.H. 1982. Perspectives on speciation by reinforcement.
S. African J. Science 78:53-57.
Rensch, B. 1929. Das Prinzip Geographischer Rassenkreise und das Problem der
Artbildung. Borntraeger Verl., Berlin.
Sinclair, M. 1987. Marine Populations: Sea Grant/Univ. Washington, Seattle.
Sinclair, M. and T.D. Iles. 1986. Population richness of marine fish.
species. Int. Counc. Explor. Sea C.M.1986/M:22.
•
17
Table 1. Some spatial and energetics loss processes in the life cycle of a
sexually reproducing species (spatial processes are indicated with
an asterisk).
A.* Egg and larvallosses due directly to physical processes of diffusion
and advection (i.e. losses from appropriate distributional areas of the
population).
B. Egg and larval mortality due to predation, disease, and starvation (and
their interaction).
C.* Juvenile losses from the population due directly to spatial constraints
(juvenile vagrants).
D.
Juvenile mortality due to predation, disease, and starvation.
E.* Adult losses from the population due directly to spatial constraints
(adult vagrants).
F. Adult mortality due to predation, disease, and starvation.
G.* Adults who do not successfully sec ure a mate, or who reproduce in the
wrong place relative to life-cycle continuity, even though they are not
vagrants yet due to spatial constraints.
H.
Adult reduced fecundity due to food limitation.
1.* Offspring which abort or are infertile due to biochemical (spatial)
constraints.
18
-,
Population Spatial PATTERN
and RICHNESS (6 populations)
..................
.....
....
.
....-'.
..
"
"
.
:
.
'..-"-.
.
..•••.....•••.....
....
•••.•..........•••••
'.0':
.......
'.-'.
......l
:" B :-A: ..'........... -
........
..
.' .'
C
...
~
~
"',
.
.. •••
~
.
E
'•\
~
::.
.'.: ....
" . .~:
.'"
..' " ::
... F.:
:
... ..-:
~.......
...•...•.........
.
.
....
.•............•
Population ABUNDANCE and VARIABILITY
w
o
~1000
POPULATION
E
POPULATION
B
POPULATION
A
o
Z
:;::)
Ln
<t
W
100
~
:;::)
..J
o
cn
m
<t
10
L-
~
TIME
Fig. 1.
Schematic representation of the four components of the
population regulation question - pattern, richness,
abundance, and variability. The nesting of small populations
within the distributional range of large populations is a
common observation for Atlantic herring and Atlantic coda
•
Atlantie salmon
Ameriean shad
Atlantle herrlng
Atlantie cod
......
1.0
Ralnbow smelt
Haddock
Atlantlc mackerel
•
g
Atlantlc menhaden
Euröpean eel
I.....---r-------
one
Fig. 2.
NUMBER OF LARVAL RETENTION AREAS
-.--~
many
Continuum in population richness of selected anadromous and marine
species in the northern Atlantic.
20
Life Cycle Solution with
Particular Sp.atial Constraints
".
,/
MEMBERS
!
~~\~
1 X
SEXUAL
\
.....-------~R~E~P~R~ODUCTION
U
~
·
"t
~GRANTSJ'
.,--._.
.,
i
/.
."",.
Fig. 3.
Schematic representation of life-cycle closure of a
marine population with a complex life history in
relation tospatial constraints. It is inferred
that there are chances of becoming a vagrant at
different phases of the life cycle.
•
..
21
Total Annual Losses (vag~y and mortality~
From a Pop-ulatlon
SPATIAL PROCESSES
1
I
I
I
I
,,,
,
I
I
,,
:
,
,,
I
Density-dependent
\ ,.processes
'v'
Densit y - independent
processes
Fig. 4.
Schematic representation of proportionallosses due to
spatial and energetic processes. Each of these categories
of losses from a populatoin may involve density dependence
and density independence.
,
22
•
Contemporaneous Population Regulation Processes
Energy
Space
Food limitation,
predation, disease
conttraint of
free-crossing due to
evolution of sex
t
t
Competition
t
Regulation of abundance
t
Food-chain selection
t
Energetics adaptation
Fig. 5.
t
Relational processes,
membership, vagrancy
t
Regulation of abundance
t
t
Speciation
Life-cycle selection
Schematic representation of the inferred evolutionary effects of
energy and space constraints on population regulation of sexually
reproducing species in the oceans. It is proposed that the two
components of population regulation are part of mechanisms leading
to respectively adaptation and speciation.