Agriculture Reference
In-Depth Information
results of the fragmentation of a continuous distribution. At other time scales, the impacts of other
events of earth history on earthworm distributions can be examined. Glaciation is one example of
such an event.
The second subject I use as an illustration of biogeographical research looks at the distributions
of Nearctic earthworms in relation to the maximal extent of the Wisconsin
glaciation and
how the process of postglacial recovery affects the composition of earthworm fauna at different
points from south to north. This may also provide some basis for figuring out the larger modern
ecological consequences of the processes affecting earthworm distributions.
Work to date on the subject of glaciation and earthworm distributions has resulted in limited
distributional data (Gates 1977; Reynolds 1994, 1995) and a plausible explanation for the distri-
bution patterns (Gates 1977; Reynolds 1994), still leaving a need for an analytical approach.
Analytical biogeography requires that hypotheses be formulated and their predictions tested. For
example, it could be stated as a hypothesis that earthworms cannot survive beneath glaciers. This
is testable, although it would be difficult to release test earthworms at the underside of a glacier
and even harder to go back and look for them. It may be preferable to examine the predictions of
the hypothesis: There should be no native earthworms in areas recently uncovered by receding
glaciers, regardless of preglacial history of earthworm presence on that site.
However, even if the observed distributions are in accord with the above historical model (the
glaciers made it impossible for earthworms to be there), this does not prove a causal link. In the
present case, there are conflicting data and alternative explanations to consider. For instance,
Schwert (1979) found a fossil earthworm cocoon in postglacial sediments of southern Ontario,
Canada. This is well north of the extent of maximum glaciation, and the discovery is old enough
to indicate that colonization must have been more rapid than conventional estimates allow (about
6 to 10 m per year from points of introduction). It also worth noting that periglacial conditions
(permafrost, for example) also affected earthworm distributions, such that simply marking the
locations of terminal moraines is not sufficient to tell the whole picture (Schwert 1977, 1990).
Finally, semiaquatic species have such as
(Wrm)
natural distributions in glaciated
areas in the eastern United States (Schwert and James, unpublished data). It inhabits wetlands and
so may have different propagule dispersal mechanisms than terrestrial worms.
Returning to the alternative hypotheses, it is conceivable that modern-day earthworm species
Eisenoides lnnbergi
in
North America were not capable of surviving the climate any farther north than they presently
occur. Second, it is conceivable that North American earthworms were present farther north before
the colonization of North America by settlers from Europe, but that European earthworms (or some
combination of European earthworm invasion and habitat destruction) eliminated the native species
by competition. An additional hypothesis needs to be made that climate affects the outcome of the
process of competition, rendering it possible for native species to persist in the southerly areas.
To meet these complaints against what seems a simple and obvious narrative explanation, I
conducted experiments and made several altitudinal transects to test the hypothesis of climate limi-
tation. Transects are comparative, not experimental, and so do not constitute a strong class of evidence.
To address the climate question, I transplanted
species
from Pennsylvania and southern Iowa, respectively, to forest and grassland sites in northern Minnesota
in 1990. All the species survived the two winters of the experimental period and were still present
on the site in 1994. Juveniles were found in both Minnesota sites, indicating that reproduction had
taken place. Similarly, on altitudinal transects in the Appalachian Mountains of Virginia, West Virginia,
and North Carolina, native earthworms were found in the highest elevations attainable, in vegetation
zones characteristic of low elevation areas much farther north and presently not occupied by native
species in those northern locations. Thus, climate limitation (and vegetation type) has been eliminated
for the species in question and is unlikely to be a significant determinant of the northern boundary
of native North American earthworm distributions.
The evidence favors abiotic historical factors to explain modern North American earthworm
distributions: Glaciation removed earthworms from areas once covered by ice or underlain by
Eisenoides carolinensis
and several
Diplocardia
