Importance of species replication in understanding plant invasions into North American grasslands - Invasive Plants: Ecological and Agricultural Aspects - page 50

not finding overall differences), I based differences on the means for all native
or exotic species. I only had enough data to test for above-ground growth rate
variables because of the small number of studies that replicated species. In 11
studies that replicated either native or exotic species or both (Fig. 1a), I found
a significant difference among outcomes (chi-squared exact test,

2 = 6.3, 2
d.f., P < 0.05). Majority of studies (65%) found that exotic species had higher
growth rates than natives (Fig. 1). Only 12% of studies found that native spe-
cies had higher growth rates than exotics. Thus, this analysis supports the
hypothesis that growth rate overall is higher in exotics than it is in natives.
When I analyzed the data set that included studies that did not replicate spe-
cies (Fig. 1b), I found no significant difference among the three outcomes (chi-
squared exact test,

χ

2 = 0.1, 2 d.f., P > 0.10), which does not support the
hypothesis that exotics differ from natives. Taken together, the difference
between these two data sets suggests that conclusions about exotic species may
change depending on whether species are replicated within groups. If multiple
species were used, the (correct?) generalization reached was that exotics had
higher growth rates than natives. If un-replicated species pairs were used, the
(incorrect?) generalization is that there was no overall effect. This result is not
entirely surprising. By using species pairs, one is less likely to find a differ-
ence between natives and exotics because of the very high variability among
species. Replicating species leads to a more precise estimate of mean differ-
ences between native and exotic species within sites, which is an important
variable to managers.

χ

Grazing tolerance in exotic species

Many plant species were introduced into North and South America,
Australasia and elsewhere to improve grazing lands. In many cases, introduc-
tions were made of species that tolerate grazing well. For example, grasses
from East Africa were introduced to many places because they evolved with
large populations of grazing mammals [31, 32]. Tolerance is defined as having
a smaller reduction (or even an increase) in relative growth rate due to com-
pensatory growth after grazing or simulated grazing (i.e., clipping) [33-35].
An intolerant plant would have larger reductions in relative growth rate. A few
influential early studies found that a native species was less tolerant of defoli-
ation than an invading exotic species [8, 32]. Again, both of these studies used
only a single native and exotic species. Based on these studies, the authors
concluded that an exotic Agropyron (now Pseudogneria sp.) species was
spreading in grazed grasslands of the inter-mountain western USA and an
African grass was spreading across South America due their greater tolerance
to grazing [8, 32].
I reviewed studies cited by Daehler [24] that included data on grazing or
clipping tolerance in exotic and native species. By including several studies in
my analysis, I could test the generality of the hypotheses of Caldwell et al. [8]