Biology Reference
In-Depth Information
The demonstration that interspecific competition occasionally occurs in
type II communities is not sufficient evidence that it limits local species
richness. Another problem with distinguishing the two types of commu-
nities is that local and regional richness are statistically not independent
(for references see Lawton 2000 ). Also, Rosenzweig and Ziv ( 1999 )have
shown that a linear relationship between local and regional richness cannot
always be distinguished from a power curve. According to them, local
versus regional richness patterns are ''echoes'' of species-area curves. These
echoes are nearly linear, in fact so close to linear for certain parameters that
they cannot be distinguished from linear although they are, in fact, power
curves. Furthermore, Caswell and Cohen ( 1993 ) have shown that type I
patterns can also arise in communities with strong competition, if patches of
species are knocked out by environmental disturbances, and Godfray and
Lawton ( 2001 ) presented a model in which type I patterns may exist even if
competition limits species numbers. Interactive communities may have no
limits to species numbers in certain models (e.g., Tilman 1999 ); Lawton
interprets this as meaning that it is not interspecific competition per se that
matters, but that competition must be sufficiently strong to limit species
numbers. Finally, Shurin ( 2000 )andShurinet al.( 2000 )haverecently
demonstrated that communities of zooplankton exhibit linear relationships
between local and regional diversity, but that such relationships are not
incompatible with strong local interactions. Furthermore, such relation-
ships are highly scale-dependent (Shurin and Srivastava, in press). All this
shows that, as already pointed out by Lawton (see above), linear or curvi-
linear relationships between local and regional richness do not say anything
about the mechanisms responsible for the patterns.
Many authors have used the distinction between the two types of
community to test for saturation (e.g., Cornell 1985a , b ;Aho 1990 ; Tonn
et al. 1990 ; Hawkins and Compton 1992 ; Kennedy and Gu´gan 1994 , 1996 ;
Oberdorff et al. 1998 ), but a null model as discussed above has not been used
in any of these studies. Also and importantly, as pointed out above, several
studies have shown that a linear relationship between local and regional
speciesrichnessdoesnotexcludethepossibility of saturation per se, and
Shurin and Srivastava (in press), evaluating empirical evidence from the
literature, believe that - in spite of the predominance of linear relationships -
theoverallpictureseemstobemoreconsistent with saturation. However, it
is important to check these assertions using other evidence (not related to
local versus regional diversity patterns; see pp. 39-48).
Pool exhaustion may explain the asymptotic relationship between local
and regional diversity, even in the absence of competition, as pointed out
Search WWH ::




Custom Search