24
series and highest level of polyploidy, from 2 x to 12 x (Table 2.1 ). The neopolyploids
are of two kinds: auto- and allopolyploids. The autopolyploids can be divided further
into two types: typical autopolyploids, characterized by multivalent pairing and mul-
tisomic inheritance, and bivalent-forming autopolyploids, characterized by exclusive
bivalent pairing and presumably disomic inheritance. Several species of the latter
group contain a lower DNA content than expected from the additive sum of the dip-
loid parent (Eilam et al. 2009 ).
The taxonomy of the tribe is complicated by several special factors such as allo-
polyploidy and ancient and recent inter-specifi c and inter-generic hybridization s
that are largely responsible for the blurred boundaries between species and even
between genera. Being a relatively young tribe, it shows an exceptional capacity for
inter-generic hybridization involving most of its members, which creates problems
both in the theoretical concept of genetic rank, and in the practical construction of
keys (Clayton and Renvoize 1986 ). It also implies a more close-knit reticulate pat-
tern of relationships between the various genera. Most of the species, if not all, can
be crossed with wheat, barley, and rye , and thus, their gene pool can serve as an
important source of useful traits for the improvement of the domesticated cereals.
Because of their economic importance, species of this tribe have been subjected
to intensive cytogenetic, genetic, molecular, and phylogenetic studies. Cytogenetic
research has provided an extensive knowledge of the tribe’s genomic constitution
(Love 1982 , 1984 ; Dewey 1982 , 1984 ). Consequently, the generic makeup of the
Triticeae has varied widely. Since there were essentially no genetic barriers between
the Triticeae genera, Stebbins ( 1956 ) and Bowden ( 1959 ) recommended merging all
these genera into one genus or, at least, unite Aegilops and Triticum in one genus,
Triticum. Morris and Sears ( 1967 ) and Kimber and Feldman ( 1987 ) included the
two genera in one genus, Triticum , since tetraploid wheat contains one genome that
derived from an Aegilops species and hexaploid wheat contains two such genomes.
On the other hand, based on the taxonomic philosophy that a system of classifi cation
should refl ect phylogeny and biological relationship (Love 1982 , 1984 ), the classi-
fi cation of most genera was drastically reorganized, mainly based on the genome
constitution of each taxon (Love 1982 , 1984 ; Dewey 1984 ; Wang et al. 1994 ). Hence
the tribe was divided, on the genome basis, into 38 different mono- generic genera.
A problem with this idea stems from the diffi culty to reach agreement on what is
a similar genomic constitution. Baum et al. ( 1987 ) criticized this classifi cation by
presenting a number of arguments against genomic genera. Some of these include
the observation that the genomic system is in effect a single- character classifi cation
and that genomic genera are not recognizable morphological units. Baum et al.
( 1987 ) claim that the genomic system makes far too many monotypic genera;
a genomic system is unstable, necessitating changes with every new genome com-
bination recognized; and that genomes are not good taxonomical characters anyway.
The genome classifi cation, though attractive in theory, is sometimes diffi cult to
translate into practical morphological diagnoses, i.e., there is incongruence between
the genomic and morphological data. The net result of this is that the generic
classifi cation of the Triticeae is currently in a state of fl ux, subject to major disagree-
ments whose outcome is still uncertain. In this article, the genera were classifi ed
according to Clayton and Renvoize ( 1986 ), who based their classifi cation on
M. Feldman and A.A. Levy