Abstract:
A biosystematic investigation of Cryptogramma section
Cryptogramma in western North America has been directed towards
resolution of two separate taxonomic questions.
Results of this research support the hypothesis that a
previously undescribed diploid (2n = 30 II) taxon, C. cascadensis, is
present along with the common and widespread C. acrostichoides. The
new species is distinguished by a number of subtle but constant
morphological features, including sterile fronds that are soft and
deciduous (compared to the coriaceous, evergreen sterile leaves of C.
acrostichoides); surficial rather than sunken hydathodes; absence of
laminar trichomes, and significantly smaller spores. Data from enzyme
electrophoresis confirm the separation of C cascadensis as a distinct
species, as 6 of 13 loci scored were "marker loci", with no alleles
held in common between the two species. While different populations
of the same species were genetically similar, with genetic identities
> 0.85, the mean genetic identity for interspecific population
comparisons was 0.36, demonstrating a substantial amount of genetic
divergence between the two species. Both species were found to be
primarily outcrossing, and the existence of marker loci and a high
degree of allozymic differentiation, even in mixed populations,
suggests that the two species are reproductively isolated.
Also investigated was the taxonomy and evolutionary origin of
Cryptogamma sitchensis, a taxon previously treated as a variety of C.
acrostichoides. Cryptogamma sitchensis was found to be a tetraploid
species, with 2n = 60 II, the first report of polyploidy in the genus
from North America. Data from enzyme electrophoresis showed fixed
heterozygosity, a characteristic of allopolyploid species. Allozyme
banding patterns showed that C. sitchensis combines genomes of C.
acrostichoides and another distinct species not present in North
America. Morphological characters, particularly the dissection of
sterile leaves, suggested that this second parent is the eastern Asian
C. raddeana. Frequent triploid hybrids, which can be identified by
their abortive spores, occur where the ranges of C. sitchensis and C
acrostichoides overlap. Hybrids blur the morphological distinctions
between the two species, and are probably responsible for leading
earlier workers to conclude that the taxa are only varietally
distinct.
These results have implications for systematic treatment of
other taxa in section Cryptogramma, suggesting that the best taxonomic
approach may be to treat geographic segregates as distinct species,
rather than as varieties or subspecies of a single circumboreal
species, C. crispa, as has been proposed by some workers.
