The tetraploid Conostylis stylidioides
(n= 16) has been proposed to be a stabilised
hybrid between the diploid (n = 8) species
C. prolifera and C. candicans
because of morphological and geographical intermediacy, as well as a polyploid
chromosome number. To test this hybrid-origin hypothesis, we used the
DNA-fingerprinting technique amplified fragment length polymorphism (AFLP) and
measured genetic variation within these taxa, as well as the putative
outgroups C. robusta and
C. aculeata. One AFLP primer pair generated 192 dominant
markers for 36 samples from these species, of which 189 were polymorphic.
Polymorphism within populations was uniformly high for all species, with
66–86% of all fragments polymorphic and estimates of
heterozygosity ranging from 0.36 to 0.41. Ordination, UPGMA and maximum
parsimony analyses of these genetic data consistently clustered species,
supporting the current species’ level taxonomy. The intermediate
placement of C. stylidioides between
C. proliferaand C. candicans on
the maximum parsimony tree supports the hybrid-origin hypothesis, although
other interpretations are possible. The phenetic results for AFLP data, in
which C. stylidioides is not strictly intermediate
between C. prolifera and
C. candicans, are either concordant with recent research
suggesting that rapid intra- and inter-genomic rearrangements occur with the
origin of polyploid taxa, or indicate an ancient hybridisation event. While
our results do not reject the hybrid origin hypothesis, the extremely high
levels of genetic variation detected with AFLP within these populations, in
combination with extensive genomic reorganisation with the origin of
C. stylidioides and the possibility of independent
origins for different populations, make it difficult to confidently exclude
other scenarios.