3.7 THE NUMBER OF SPECIES OR THEIR IDENTITY

20.3.7 The number of species or their identity?

P

Active

Keystones again

plant

tissue

Indeed, it is clear that the whole concept of a keystone species

Litterfall and

Consumption

(see Section 20.2.6) is itself a recognition of the fact that the effects

translocation

of a disturbance on structure or function are likely to depend

very much on the precise nature of the disturbance – that is, on

Inactive

HD

Defecation

which species are lost. Reinforcement of this idea is provided by

organic

Heterotrophs

Decomposition

matter

a simulation study carried out by Dunne et al. (2002), in which

they took 16 published food webs and subjected them to the

Transport

Respiration

sequential removal of species according to one of four criteria:

(i) removing the most connected species first; (ii) randomly

removing species; (iii) removing the most connected species

(b)

first excluding basal species (those having predators but no prey);

Pond

and (iv) removing the least connected species first. The stability

of the webs was then judged by the number of secondary extinc-

tions that resulted from the simulated removals, such extinctions

Freshwater

occurring when species were left with no prey (and so basal species

spring

Temperate

were subject to primary but not secondary extinction). In the

Tropical

deciduous

first place, the robustness of community composition in the face

forest

Salt marsh

Rate of recovery after perturbation (arbitrary units)

of species loss increased with connectance of the communities

– further support for an increase in community stability with

Tundra

complexity. Overall, however, it is also clear that secondary extinc-

10

^–2

10

^–1

10

⁰

10

¹

10

²

tions followed most rapidly when the most connected species

Energy input per unit of

were removed, and least rapidly when the least connected species

standing crop (energy units)

were removed, with random removals lying between the two

(Figure 20.13). There were, moreover, some interesting exceptions

Figure 20.12 (a) A simple model of a community. The

when, for example, the removal of a least connected species led

three boxes represent components of the system and arrows

to a rapid cascade of secondary extinctions because it was a basal

represent transfers of energy between the system components.

species with a single predator, which was itself preyed upon by

(b) The rate of recovery (index of resilience) after perturbation

a wide variety of species. This, finally in this section, reminds

(as a function of energy input per unit standing crop) for

us that the idiosyncrasies of individual webs are likely always to

models of six contrasting communities. The pond community

undermine the generality of any ‘rules’ even if such rules can be

was most resilient to perturbation, tundra least so. (After

agreed on.

O’Neill, 1976.)