Genetic diversity of 33 Elymus caninus accessions was investigated using is
ozyme, RAPD, and microsatellite markers. The three assays differed in the a
mount of polymorphism detected. Microsatellites detected the highest polymo
rphism. Six microsatellite primer pairs generated a total of 74 polymorphic
bands (alleles), with an average of 15.7 bands per primer pair. Three gene
tic similarity matrices were estimated based on band presence or absence. G
enetic diversity trees (dendrograms) were derived from each marker techniqu
e, and compared using Mantel's test. The correlation coefficients were 0.20
4, 0.267, and 0.164 between isozyme and RAPD distance matrices, RAPD and mi
crosatellite distance matrices, and between isozyme and microsatellite dist
ance matrices, respectively. The three methodologies gave differing views o
f the amount of variation present but all showed a high level of genetic va
riation in E. caninus. The following points may be drawn from this study wh
ether based on RAPD, microsatellite, or isozyme data: (i) The Icelandic pop
ulations are consistently revealed by the three dendrograms. The congruence
of the discrimination of this accession group by RAPD, microsatellite, and
isozyme markers suggests that geographic isolation strongly influenced the
evolution of the populations; (ii) The degree of genetic variation within
accessions was notably great; and (iii) The DNA-based markers will be the m
ore useful ones in detecting genetic diversity in closely related accession
s. In addition, a dendrogram, which took into account all fragments produce
d by isozymes, RAPDs, and microsatellites, reflected better the relationshi
ps than did dendrograms based on only one type of marker.