Lh. Rieseberg et Nc. Ellstrand, WHAT CAN MOLECULAR AND MORPHOLOGICAL MARKERS TELL US ABOUT PLANT HYBRIDIZATION, Critical reviews in plant sciences, 12(3), 1993, pp. 213-241
The study of hybrids and their evolutionary significance is often base
d on a number of tacit assumptions regarding character expression in h
ybrids. This article examines morphological, chemical, and molecular c
haracter expression in hybrids to determine whether traditionally reco
gnized properties of hybrid plants, such as hybrid intermediacy and ch
aracter coherence, are actually supported by empirical evidence, and a
lso examines the impact of hybrids on phylogenetic analyses. We show t
hat hybrids are a mosaic of both parental and intermediate morphologic
al characters rather than just intermediate ones, and that a large pro
portion of first (64%) and later generation hybrids (89%) exhibit extr
eme or novel characters. Chemical character expression in hybrids is m
ore predictable, with predominantly additive or complementary expressi
on for both first generation hybrids (68%) and hybrid taxa (54%). Like
wise, the genetic basis, and thus the expression of molecular characte
rs, is well-worked out and predictable, although non-Mendelian inherit
ance has been reported in a few instances for molecular markers as wel
l. There is even less empirical support for the concept of character c
oherence than hybrid intermediacy. Although morphological character co
herence has been reported in both artificial and natural hybrid popula
tions, it appears to be the exception rather than the rule. This idios
yncratic relationship among morphological characters is shown to be th
e norm for the relationship of morphological to molecular characters,
as well as for the relationship among molecular characters. Hybrids ar
e shown to have little impact on the topology of nonhybrid taxa in phy
logenetic trees. However, the expression of primitive vs. derived char
acter states, the placement of hybrids in cladograms, the number of eq
ually parsimonious trees produced, and the effects of hybrids on phylo
genetic topology do not appear to be predictable. Thus, cladistic iden
tification of hybrid taxa is difficult and may not be possible based o
n phenotypic data, regardless of the analytical tools used.