Competitive Exclusion Principle

The competitive exclusion principle—also known as Gause's Law or the Volterra–Gause Principle—is a foundational proposition in community and population ecology stating that two biological species competing for the exact same limiting resource in a homogeneous, constant ecosystem cannot stably coexist at constant population values. When two species occupy identical ecological niches, even a slight reproductive, physiological, or foraging efficiency advantage enables one competitor to depress shared resource availability below the subsistence threshold of the other, driving the inferior competitor to local extirpation or evolutionary niche divergence.

Although Charles Darwin noted in On the Origin of Species (1859) that competition is severest between closely allied forms occupying similar habits, the principle was first derived mathematically by the Italian mathematician Vito Volterra (1926) and the American biophysicist Alfred J. Lotka (1932). In 1934, the Soviet ecologist Georgii Frantsevich Gause tested the Lotka–Volterra competition equations empirically in his classic monograph The Struggle for Existence using laboratory microcosms of ciliated protozoans (Paramecium) and yeasts (Saccharomyces and Schizosaccharomyces). The phrase "competitive exclusion principle" was subsequently coined in 1960 by the American ecologist Garrett Hardin in Science, who summarized its core corollary: "Complete competitors cannot coexist."

In natural communities, competitive exclusion serves as the primary selective force driving resource partitioning, ecological character displacement, and the divergence between a species' physiological fundamental niche and its ecologically constrained realized niche.

---

To test whether mathematical competition models accurately predict biological outcomes under controlled conditions, G. F. Gause (1934) cultured three species of freshwater ciliates—Paramecium aurelia, Paramecium caudatum, and Paramecium bursaria—in buffered Osterhout balanced salt medium or oatmeal broth replenished daily with a constant suspension of bacterial prey (Bacillus pyocyaneus, now Pseudomonas aeruginosa) or yeast (Saccharomyces exiguus).

1. Monoculture Controls: When grown separately in identical test tubes with constant daily bacterial supply, both P. aurelia and P. caudatum exhibited classic Verhulst sigmoidal logistic growth, stabilizing asymptotically at their respective carrying capacities (K₁ and K₂). Because P. aurelia is smaller in cell volume, it exhibited a higher intrinsic rate of natural increase (r₁ ≈ 1.12 d⁻¹ vs. r₂ ≈ 0.94 d⁻¹ for P. caudatum) and converted bacterial biomass into ciliate density more efficiently.
2. Co-culture of P. aurelia and P. caudatum (Competitive Exclusion): When inoculated together into a single unstratified tube containing only suspended Bacillus pyocyaneus, both populations expanded during the first 4–6 days. Once the shared bacterial supply became limiting, P. aurelia continued to increase toward its carrying capacity while P. caudatum declined steadily, reaching complete local extinction between Day 14 and Day 16 in every replicate.
3. Co-culture of P. caudatum and P. bursaria (Niche Partitioning and Coexistence): When Gause paired P. caudatum with the green algal-endosymbiotic ciliate Paramecium bursaria in medium containing both suspended bacteria and settled bottom yeast, the two species achieved stable long-term coexistence (albeit at equilibrium densities lower than their monoculture K values). Whereas P. caudatum fed exclusively on planktonic bacteria in the upper oxygenated water column, P. bursaria foraged on yeast cells settled in the poorly oxygenated benthic layer, aided by O₂ produced photosynthetically by its intracellular Chlorella endosymbionts. Spatial and resource partitioning reduced interspecific competition below intraspecific competition.

---

Lotka–Volterra Competition Equations and ZNGI Phase-Plane Analysis

The Lotka–Volterra interspecific competition model extends the single-species Verhulst logistic equation to two competing species with population densities N₁ and N₂, intrinsic growth rates r₁ and r₂, and monoculture carrying capacities K₁ and K₂:

dN₁ / dt = r₁ N₁ [ ( K₁ − N₁ − α₁₂ N₂ ) / K₁ ]

dN₂ / dt = r₂ N₂ [ ( K₂ − N₂ − α₂₁ N₁ ) / K₂ ]

Here, the dimensionless competition coefficient α₁₂ measures the per-capita inhibitory effect of an individual of Species 2 on the growth of Species 1 (expressed in units of Species 1 equivalents), and α₂₁ measures the per-capita inhibitory effect of Species 1 on Species 2. Setting dN₁/dt = 0 and dN₂/dt = 0 yields the two linear Zero-Net-Growth Isoclines (ZNGIs) in the (N₁, N₂) phase plane:

• Species 1 ZNGI: N₁ + α₁₂N₂ = K₁ (intercepts at N₁ = K₁ on the horizontal axis and N₂ = K₁/α₁₂ on the vertical axis).
• Species 2 ZNGI: N₂ + α₂₁N₁ = K₂ (intercepts at N₁ = K₂/α₂₁ on the horizontal axis and N₂ = K₂ on the vertical axis).

Depending on the relative magnitudes of the four axis intercepts, exactly four dynamical cases arise:

1. Case 1 — Species 1 Excludes Species 2 (K₁ > K₂/α₂₁ and K₁/α₁₂ > K₂): The ZNGI of Species 1 lies entirely outside (above and to the right of) the ZNGI of Species 2. Regardless of initial densities (N₁(0), N₂(0) > 0), trajectories converge to the boundary equilibrium (K₁, 0).
2. Case 2 — Species 2 Excludes Species 1 (K₂/α₂₁ > K₁ and K₂ > K₁/α₁₂): The ZNGI of Species 2 lies entirely outside that of Species 1, driving all trajectories to (0, K₂).
3. Case 3 — Unstable Priority Effect / Contingent Exclusion (K₁ > K₂/α₂₁ and K₂ > K₁/α₁₂): Each species inhibits the other more strongly than it inhibits itself (α₁₂ > K₁/K₂ and α₂₁ > K₂/K₁). The isoclines cross at an unstable interior saddle point. A separatrix divides the phase plane into two basins of attraction: whichever species starts with an initial numerical advantage excludes the other (priority effect).
4. Case 4 — Stable Interior Coexistence (K₂/α₂₁ > K₁ and K₁/α₁₂ > K₂): Each species inhibits its own conspecifics more strongly than it inhibits the heterospecific competitor (α₁₂ < K₁/K₂ and α₂₁ < K₂/K₁, which requires the necessary condition α₁₂α₂₁ < 1). Both species can invade when rare, and trajectories converge to the globally stable interior equilibrium:

N₁\* = ( K₁ − α₁₂ K₂ ) / ( 1 − α₁₂ α₂₁ ) , N₂\* = ( K₂ − α₂₁ K₁ ) / ( 1 − α₁₂ α₂₁ )

Tilman's Mechanistic *R*\* Resource-Ratio Theory

Whereas the Lotka–Volterra equations treat competition phenomenologically, David Tilman (1982) formulated a mechanistic consumer–resource model in which n species compete explicitly for dissolved or soil resources R. For a single limiting resource R supplied at rate S(R), the population dynamics of consumer i with Monod growth kinetics μ_i(R) = μ_max,*i*R/(K_s,*i* + R), per-capita loss/mortality rate m_i, and resource consumption coefficient c_i are:

dN_i / dt = N_i [ μ_i(R) − m_i ] , dR / dt = S(R) − Σ c_i μ_i(R) N_i

Setting dN_i/dt = 0 defines the break-even resource concentration R_i\* required for Species i to exactly balance growth and mortality (μ_i(R_i\*) = m_i):

R_i\* = m_i K_s,*i* / ( μ_max,*i* − m_i )

Tilman's *R*\* Rule: When multiple species compete for a single homogeneous limiting resource at steady state, the species with the lowest *R*\* draws the ambient resource concentration down to its own R₁\* < R₂\*, starving all competitors (for which dN_j/dt < 0 at R = R₁\*) and competitively excluding them. Tilman verified this quantitatively using Lake Michigan freshwater diatoms: Asterionella formosa has a lower *R*\* for dissolved phosphate (PO₄³⁻) and excludes Cyclotella meneghiniana at high Si:P ratios (Si:P > 90), whereas Cyclotella has a lower *R*\* for dissolved silicate (Si(OH)₄) and excludes Asterionella at low Si:P ratios (Si:P < 6). At intermediate ratios (6 < Si:P < 90), each species is limited by a different resource, permitting stable two-species coexistence.

---

Fundamental vs. Realized Niche (Hutchinson 1957 & Connell 1961)

In his 1957 Cold Spring Harbor Concluding Remarks, G. Evelyn Hutchinson defined the ecological niche as an n-dimensional hypervolume in environmental resource and condition space within which a species can maintain a non-negative intrinsic growth rate (r ≥ 0):

• The fundamental niche is the full hypervolume a species can occupy in the absence of interspecific competitors and predators.
• The realized niche is the reduced hypervolume to which the species is restricted in the presence of superior competitors.

Joseph H. Connell (1961) demonstrated this distinction on the rocky intertidal coast of Millport, Scotland, using two acorn barnacles: Chthamalus stellatus (occupying the upper desiccation-prone zone) and Semibalanus balanoides (occupying the wetter mid-intertidal zone). Although planktonic Chthamalus cyprid larvae settle throughout both zones, fast-growing Semibalanus individuals undercut, pry off, or smother Chthamalus in the lower zone. When Connell experimentally removed Semibalanus, Chthamalus survived and grew thrivingly throughout the mid-intertidal zone—proving that its upper boundary is set physiologically by desiccation (fundamental niche), whereas its lower boundary is set biologically by competitive exclusion (realized niche).

Niche Partitioning and Character Displacement

Coexisting guild members avoid competitive exclusion through behavioral, spatial, temporal, or morphological specialization:

• MacArthur's Warblers (1958): Robert H. MacArthur studied five congeneric wood-warbler species (Setophaga tigrina, S. fusca, S. coronata, S. virens, and S. castanea) that appeared to violate Gause's Law by foraging for caterpillars in the same boreal white spruce (Picea glauca) trees in Maine. By timing foraging observations to the second across height and branch zones, MacArthur showed that each warbler species specialized on a distinct microhabitat section of the spruce crown (apical needles, mid-crown interior, lower lichen-covered limbs, or aerial hawking).
• Character Displacement (Brown & Wilson 1956): Where two competing species occur in sympatry, natural selection favors individuals with divergent trophic morphology that minimize niche overlap, whereas in allopatry the two species retain convergent intermediate traits. Classic examples include beak depth divergence in Galápagos ground finches (Geospiza fortis and Geospiza fuliginosa, documented by Peter and Rosemary Grant) and body-size divergence in Hydrobia mud snails.

Hutchinson's Paradox of the Plankton (1961)

Because open epipelagic ocean and lake waters are well-mixed and contain only a handful of limiting resources (light, NO₃⁻, NH₄⁺, PO₄³⁻, Si(OH)₄, Fe²⁺, and vitamin B₁₂), equilibrium theory predicts that at most 5–7 phytoplankton species should survive competitive exclusion. Yet a single liter of lake or ocean water routinely contains 50–100 coexisting phytoplankton species. Hutchinson (1961) resolved this Paradox of the Plankton by noting that natural aquatic environments are prevented from reaching competitive equilibrium because seasonal and weather-driven abiotic fluctuations occur on the same timescale (~days to weeks) as phytoplankton generation times, continually reversing competitive rankings before exclusion can run to completion. Subsequent work by Jef Huisman and Franz J. Weissing (1999) proved that competition for ≥3 resources can generate endogenous chaotic oscillations in species abundances that allow dozens of species to coexist on just three resources, supplemented by keystone predation and frequency-dependent viral lysis ("kill the winner") that selectively suppress dominant blooms.

---

What is the necessary condition for stable coexistence in the Lotka–Volterra competition model?

For two competing species to coexist stably at an interior equilibrium (N₁\*, N₂\* > 0), intraspecific competition must exceed interspecific competition for both species: α₁₂ < K₁/K₂ and α₂₁ < K₂/K₁ (which implies α₁₂α₂₁ < 1). Ecologically, this means each species limits its own population via a partially distinct resource or microhabitat, allowing either species to maintain a positive per-capita growth rate when rare.

How does Tilman's *R*\* rule predict the winner of resource competition?

In Tilman's mechanistic resource-ratio theory, R_i\* is the minimum equilibrium concentration of a limiting resource (such as nitrate or silicate) at which Species *i*'s reproductive growth rate exactly balances its mortality rate (dN_i/dt = 0). When multiple species compete for a single resource, the species with the lowest *R*\* depletes the ambient environment down to its own *R*\*, a concentration at which all other competitors suffer negative net growth and go extinct.

How do keystone predators prevent competitive exclusion?

A keystone predator (such as the intertidal sea star Pisaster ochraceus or the sea otter Enhydra lutris) or a host-specific pathogen disproportionately consumes the competitively dominant prey species. By holding the dominant competitor's density below its carrying capacity (K₁), trophic cascades prevent resource monopolization and allow subordinate competitors with higher *R*\* values to persist in the community.

---

1. Gause, G. F. (1934). The Struggle for Existence. Baltimore: Williams & Wilkins. https://doi.org/10.5962/bhl.title.4489
2. Hardin, G. (1960). "The competitive exclusion principle." Science, 131(3409), 1292–1297. https://doi.org/10.1126/science.131.3409.1292
3. Hutchinson, G. E. (1961). "The paradox of the plankton." The American Naturalist, 95(882), 137–145. https://doi.org/10.1086/282171
4. Tilman, D. (1982). Resource Competition and Community Structure. Princeton, NJ: Princeton University Press. https://doi.org/10.1515/9780691209654