Naturalized fodder sorghums in Queensland, and their role in shattering in grain sorghum


Simon, B.K.

Queensland Journal of Agricultural and Animal Sciences 36(1): 71-86

1979


In an attempt to understand the nature of shattering off-types of Sorghum bicolor in grain sorghum crops in Queensland, the taxonomy of the related naturalized fodder sorghums is outlined, and the possible role of these species in the evolution of shattering off-types is discussed. The species concerned are S. halepense, S. almurn, S. verticilliflorum, S. sudanense, S. miliaceum and S. brevicarinatum.

SHATTERING
IN
GRAIN
SORGHUM
71
QUEENSLAND
DEPARTMENT
OF
PRIMARY
INDUSTRIES
DIVISION
OF
PLANT
INDUSTRY
BULLETIN
NO.
797
NATURALIZED
FODDER
SORGHUMS
IN
QUEENSLAND,
AND
THEIR
ROLE
IN
SHATTERING
IN
GRAIN
SORGHUM
by
B.
K.
SIMON,
M.Sc.
SUMMARY
In
an
attempt
to
understand
the
nature
of
shattering
off-types
of
Sorghum
bicolor
in
grain
sorghum
crops
in
Queensland,
the
taxonomy
of
the
related
naturalized
fodder
sorghums
is
outlined,
and
the
possible
role
of
these
species
in
the
evolution
of
shattering
off-types
is
discussed.
The
species
concerned
are
S.
halepense,
S.
almurn,
S.
verticilliflorum,
S.
sudanense,
S.
miliaceum
and
S.
brevicarinatum.
I.
INTRODUCTION
One
of
the
most
common
requests
received
at
the
Botany
Branch,
Brisbane
is
for
the
identification
of
specimens
of
the
three
sorghum
species
Sorghum
verticilliflorum
(wild
sorghum),
S.
halepense
(Johnson
grass),
and
S.
almum
(Columbus
grass).
Recently
however
(since
the
1976-77
season)
the
need
to
understand
the
relationship
between
these
and
closely
related
sorghum
species
has
grown
because
of
their
involvement
in
the
appearance
of
shattering
off-types
in
grain
sorghum
crops
in
Queensland
(see
figure
4),
and
more
recently
(1977-78
season)
in
New
South
Wales
(J.
Strang
and
T.
Dale
personal
communication).
The
genetically
controlled
characteristic
of
seed
shedding
can
be
found
in
any
Sorghum
species
(Swarbrick
1977),
including
those
of
section
Sorghum
(formerly
incorrectly
called
section
Eu-Sorghum),
which
includes
all
species
discussed
in
this
paper.
Three
other
species
less
commonly
collected
but
likely
to
be
con-
fused
with
the
three
common
species,
are
Sorghum
sudanense
(Sudan
grass),
S.
brevicarinatum
and
S.
miliaceum.
Section
Sorghum
is
divided
taxonomically
into
subsection
Halepensia,
characterized
by
the
possession
of
elongated
rhizomes
and
a
somatic
chromosome
number
of
40
(rarely
20),
and
subsection
Arundinacea,
in
which
the
plants
are
tufted
and
have
a
somatic
chromosome
number
of
20
(Snowden
1955;
Doggett
1970).
Sorghum
halepense,
S,
almum
and
S.
miliaceum
belong
to
subsection
Halepensia,
and
S.
verticilliflorum,
S.
sudanense,
S.
brevicarinatum
and
cultivated
grain
sorghum
(S.
bicolor)
belong
to
subsection
Arundinacea.
The
spikelets
and
inflorescence
of
both
sub-sections
are
extremely
similar
in
their
morphology
and
the
only
reliable
morphological
character
separating
these
two
groups
is
the
presence
or
absence
of
an
elongated
rhizome;
for
this
reason
it
is
essential
the
whole
plant
including
all
and
not
only
some
of
the
underground
portions,
be
used
for
identification.
Queensland
Journal
of
Agricultural
and
Animal
Sciences
Vol
36
(1)
1979
72
B.
K.
SIMON
II.
KEY
The
following
key
applies
to
grain
sorghum
and
wild
fodder
sorghums
naturalised
in
Queensland:
1.
Panicle
axis
tough
and
much
thicker
than
the
branches;
grains
large,
usually
exposed
at
maturity,
with
the
glumes
gaping
..
t.
S.
bicolor
(grain
sorghum)
1.
Panicle
axis
fragile,
slightly
thicker
than
the
branches;
grains
small,
enclosed
by
the
glumes
at
maturity.
2.
Plant
with
an
elongated
rhizome,
sometimes
with
prop
roots.
3.
Sessile
spikelet
4-5-5.0
x
1-8-2.0
(-2.3)
mm,
becoming
yellowish
though
reddish
to
purple
and
occasionally
black
and
shiny
in
exposed
parts
at
maturity;
lower-most
panicle
branches
2-4-nate;
some
rhizome
internodes
with
a
tendency
to
be
long
and
thin.
4.
Leaf
blades
0.5-2-0
cm
wide;
culms
0.5-1.8
m
tall,
up
to
5
mm
thick.
2.
S.
halepense
(Johnson
grass)
4.
Leaf
blades
2-4
cm
wide;
culms
2-3
m
tall,
up
to
1
cm
thick.
3.
S.
miliaceum
3.
Sessile
spikelet
5.0-6.5
x
2.0-5.0
mm,
becoming
black
and
shiny
at
maturity;
lowermost
panicle
branches
(3-)
4-9-nate;
culms
up
to
3.5
m
tall;
rhizome
internodes
with
a
tendency
to
be
short
and
thick
..
..
4.
S.
almum
(Columbus
grass)
2.
Plant
tufted,
often
with
prop
roots,
without
rhizomes.
5.
Sessile
spikelet
4.5-5.5
mm
long
.
.. ..
5.
S.
brevicarinatum
5.
Sessile
spikelets
(5.5-)
6.0-7.5
mm
long.
6.
Perennial,
loosely
tufted,
sometimes
annual;
raceme
disarticulating
readily
at
maturity
with
the
sessile
spikelets
breaking
clean
from
the
apex
of
the
rhachis
internode;
pedicelled
spikelet
falling
free
from
the
pedicel
6.
S.
verticilliflorum
(wild
sorghum)
6.
Annual,
raceme
not
disarticulating
readily
at
maturity
but
finally
breaking
up
with
the
upper
portion
of
the
rhachis
internode
attached
to
the
sessile
spikelet;
pedicelled
spikelet
persisting
on
the
pedicel
..
7.
S.
sudanense
The
morphological
differences
between
all
these
species,
as
seen
from
the
key
and
figures,
are
extremely
small
and
often
it
is
difficult
to
accurately
identify
plants.
The
ease
with
which
the
species
cross,
results
in
hybrids
and
hybrid
swarms
between
species.
It
could
be
that
the
intermediate
plants
are
those
most
often
sent
in
for
identification.
III.
ENUMERATION
OF
SPECIES
Sorghum
bicolor
(L.)
Moench.
Grain
sorghum,
cultivated
sorghum,
sorghum
(figures
1,
2A,
3H,
4,
13,
14,
and
15).
Much
morphological
variation
is
covered
by
this
binomial.
Plants
vary
from
forms
to
3
m
tall
with
very
open
heads
to
very
short
commercial
cultivars
only
0.5
m
tall
with
tightly
clustered
heads.
Snowden
(1936)
recognized
31
species
within
this
taxon
but
it
has
since
been
shown
that
variability
is
continuous
between
them
with
as
many
intermediate
forms
as
there
are
species.
For
this
reason,
and
for
the
fact
that
all
Snowden's
species
readily
cross-breed,
it
is
con-
sidered
more
meaningful
to
treat
all
grain
sorghums
under
the
one
binomial.
Although
introgression
from
the
fodder
sorghums
does
occur,
grain
sorghums
have
been
found
always
to
differ
morphologically
from
fodder
sorghums
by
the
posses-
sion
of
a
much
thicker
panicle
axis
and
by
the
development
of
large
grains
which
become
exposed
at
maturity.
SHATTERING
IN
GRAIN
SORGHUM
73
3
2
E
0
S.
almum
S.
halepense
S.
miliaceum
3
2
1
0
S.
bicolor
S.
verticilliflorum,
S.
sudanense,
S.
brevicarinatum
Figure
1.
Habit
ideograms
of
fodder
and
grain
sorghums.
74
B.
K.
SIMON
A
II
oC
i
\
Bt
Cl
1
11
,
tl
r
rl
I(,
I
I
/1
'
'
f
'
.
,
t,
4
C2
Figure
2.
Spikelet
pairs
of
A.
S.
bicolor,
shattering
off-type
at
mature
stage
(Finlay
and
Barnes
4);
B.
S.
halepense
(Walsh
76-1);
C.
S.
miliaceum
(Finlay
2-1962);
D.
S.
&mum
(Everist
5/1965);
all
x
8.
Spikelet
pairs
are
shown
in
back
view
(1)
and
front
view
(2).
A
is
shown
in
front
view.
SHATTERING
IN
GRAIN
SORGHUM
75
Sorghum
halepense
(L.)
Pers.
Johnson
grass
(figure
1,
2B,
5,
and
6).
This
troublesome
weed
with
its
system
of
aggressive
rhizomes
is
naturally
distributed
from
the
Mediterranean
to
Pakistan
and
India.
It
was
introduced
to
the
United
States
by
1830
although
the
exact date
of
its
first
introduction
is
not
certain
(McWhorter
1971).
A
Colonel
Johnson
is
reported
to
have
taken
it
into
cultivation
early
as
a
fodder
grass.
It
was
widely
grown
in
the
southern
United
States
and
became
a
troublesome
weed.
It
was
probably
introduced
first
to
Australia
in
the
1860's
or
1870's,
as
it
is
mentioned
in
Flora
Australiensis
(Bentham
1877),
but
there
were
further
introductions
after
this.
It
occurs
in
Queensland
in
coastal
and
sub-coastal
localities
from
Wallangarra
to
Mossman.
There
is
only
one
record
in
the
Brisbane
herbarium
from
inland
Queensland
and
that
is
from
Cloncurry
where
the
species
was
recorded
as
a
garden
weed.
Emphasis
in
the
past
has
been
placed
on
its
long
creeping
rhizome
as
a
character
to
distinguish
it
from
S.
almum
which has
been
described
as
having
shorter
rhizomes
which
soon
ascend.
However,
this
characteristic
has
not
been
found
to
be
consistently
developed
in
herbarium
material,
where
there
are
specimens
of
S.
almum
with
extensive
elongated
rhizomes.
There
is
however,
a
tendency
for
some
rhizomes
of
S.
halepense
to
have
longer
and
thinner
internodes
than
those
of
S.
almum.
More
reliable
differences
are
the
smaller
spikelets
and
shorter
stature
of
S.
halepense.
It
differs
from
S.
verticilliflorum
by
its
smaller
spikelets
and
height
and
also
by
its
possession
of
rhizomes.
Sorghum
miliaceum
(Roxb.)
Snowden
(Figures
1,
2C,
7
and
8.)
This
is
a
very
robust
species
allied
to
S.
halepense
and
in
its
native
habitat
is
a
grass
of
rain-forest
clearings
in
northern
and
eastern
India.
Some
authors
have
treated
it
as
a
variety
of
S.
halepense
(e.g.,
Magoon
et
al.
1967).
It
has
been
collected
in
Australia
only
in
areas
around
the
Burdekin
delta
and
on
the
Darling
Downs
and
its
method
and
date
of
introduction
are
unknown.
It
has
the
narrow
spikelets,
elongated
rhizomes
and
long
thin
rhizome
internodes
of
Johnson
grass
but
it
is
much
taller
with
prop
roots
usually
developed
and
has
much
wider
leaf
blades
than
that
species.
Sorghum
almum
Parodi
Columbus
grass
(Figures
1,
2D,
9
and
10).
This
species
was
described
in
Argentina
in
1943
from
cultivated
material
and
is
postulated
to
have
arisen
from
a
cross
between
S.
halepense
and
a
member
of
the
section
Arundinacea,
the
latter
being
either
a
grass
fodder
sorghum
(Parodi
1943)
or
a
cultivated
grain
sorghum
(Endrizzi
1957).
It
has
broader
spikelets
than
those
of
S.
halepense
and
is
taller.
It
has
been
reported
to
differ
from
S.
halepense
by
having
shorter
rhizomes
which
are
not
horizontally
developed
(Everist
1974).
However,
as
mentioned
under
S.
halepense,
this
difference
is
not
consistently
developed
and
there
are
many
intermediate
stages
of
rhizome
development
between
the
two
species
with
some
rhizome
internodes
tending
to
be
shorter
and
thicker
than
in
S.
halepense.
It
has
an
inflorescence
very
similar
to
that
of
S.
verticilliflorum,
although
the
sessile
spikelets
are
slightly
smaller.
It
has
been
introduced
to
Australia
many
times
by
C,S.I.R.O.
since
1946
(Griffith
Davies
and
Edye
1959)
and
is
widely
grown
for
fodder
in
Queensland
in
areas
of
500-900
mm
rainfall.
Naturalized
occurrences
are
sub-
coastal
in
south-eastern
Queensland
but
are
more
coastal
in
the
north
of
the
State.
76
B.
K.
SIMON
\
'4
1
E2
V
iii
I
I
2
/P
.
Figure
3.
Spikelet
pairs
of
E.
S.
brevicarinatum
(Schmid
50);
F.
S.
verticilliflorum
(Hubbard
2378;
G.
S.
sudanense
(White
9736);
H.
S.
bicolor,
younger
stage
than
A.
(Henzell
16).
All
x
8.
Spikelet
pairs
are
shown
in
back
view
(1)
and
front
view
(2).
SHATTERING
IN
GRAIN
SORGHUM
77
Figure
4.
Shattering
off-type
of
S.
bicolor
growing
in
a
field
of
grain
sorghum
(Photo
by
Mr
T.
Dale,
New
South
Wales
Department
of
Agriculture).
Sorghum
brevicarinatum
Snowden
(Figures
1
and
3E).
Like
S.
verticilliflorum
and
S.
sudanense
this
is
a
tufted
grass
without
rhizomes
but
differs
from
these
species
by
its
shorter
spikelets.
The
few
Queensland
records
of
this
species
come
from
Mt.
Isa
and
the
Darling
Downs.
The
history
of
its
introduction
to
Australia
is
unknown.
Sorghum
verticilliflorum
(Steudel)
Stapf
Wild
sorghum
(Figures
1,
3F,
11
and
12).
Although
called
wild
sorghum
in
Australia
it
is
not
native
here
and
comes
originally
from
the
savannah
areas
of
tropical
Africa
and
sub-tropical
southern
Africa.
Its
exact
date
of
introduction
to
Queensland
is
not
known
but
in
1937
it
was
recorded
to
have
been
here
for
many
years
(White
1937).
It
is
a
vigorous
loosely
tufted
perennial,
sometimes
annual,
and
usually
has
distinctive
prop
roots;
it
differs
from
S.
almum
and
S.
halepense
by
its
larger
spikelets
and
by
the
lack
of
a
rhizome.
It
is
less
favoured
for
forage
purposes
than
the
closely
related
Sudan
grass
due
to
its
fragile
racemes
which
shed
their
spikelets
readily
at
maturity.
It
occurs
in
Queensland
in
coastal
and
sub-coastal
localities
from
the
New
South
Wales
border
to
north
of
Cairns
and
in
scattered
inland
localities.
A
N
al
r
yr
'"'
Figure
5.
Plants
of
S.
halepense
with
flowering
cu/ms
ranging
in
height
form
0.5
to
1.8
m.
Figure
6.
Rhizomes
of
S.
halepense.
SHATTERING
7N
"GRAIN
SORGHUM
9
,
Figure
7.
Plants
of
S.
miliaceum,
with
flowering
cu/ms
to
3
m.
Sorghum
sudanense
(Piper)
Stapf
Sudan
grass
(Figures
1
and
3).
This
annual
forage
grass
was
introduced
to
Australia
in
1913
(Breakwell
1915)
from
the
United
States
where
it
was
introduced
from
the
Sudan
in
1909.
It
is
morphologically
very
similar
to
wild
sorghum,
differing
from
it
in
its
uniformly
annual
habit,
its
ability
to
produce
more
tillers
and
its
persistent
racemes.
Further,
when
the
sessile
spikelets
fall
at
maturity,
they
do
so
with
a
rhachis
fragment
remaining
attached
to
their
base
as
opposed
to
the
spikelets
breaking
off
cleanly
from
the
rhachis
apex
(see
figures
3F
and
3G).
Sudan
grass
has
become
naturalized
in
both
coastal
and
inland
areas
of
Queensland,
probably
as
escapes
from
cultivation,
but
how
long
the
populations
persist
in
the
naturalized
condition
is
not
known.
80
B.
K.
SI
ON
Figure
8.
Rhizomes
of
S.
miliaceum.
Figure
9.
Flowering
heads
of
S.
almum.
fp'
s
.
Figure
10.
Base
of
S.
almum
showing
rhizomes.
Figure
11.
Plants
of
S.
verticilliflorum
with
flowering
culms
to
2.3
m.
82
B.
K.
SIMON
IV.
ORIGIN
OF
SHATTERING
TYPES
Consideration
of
the
possible
origin
of
shattering
off-types
in
grain
sorghums
in
Australia
must
include
cyto-taxonomic
aspects
of
the
section
Sorghum.
Native
Australian
sorghums
belong
to
other
sections
(Garber
1950,
where
sections
are
treated
as
subgenera)
and
are
not
likely
to
cross
with
cultivated
sorghums.
Sub-
section
Arundinacea,
which
contains
both
non-rhizomatous
wild
fodder
sorghums
and
cultivated
grain
sorghums,
has
only
diploid
species
and
all
members
cross
readily
with
each
other.
For
this
reason
some
authors
(Harlan
&
De
Wet
1972)
have
used
one
binomial,
S.
bicolor,
to
cover
all
forms
of
this
complex,
but
it
has
been
pointed
out
(Purseglove
1972)
that
hybrids
between
cultivated
and
wild
sorghums
rarely
persist
in
nature
due
to
their
reduced
fertility.
An
alternative
suggestion
(Purseglove
1972)
is
a
grouping
of
all
wild
members
of
subsection
Arundinacea
under
the
binomial
S.
arundinaceum
with
taxa
that
are
usually
treated
as
species
being
given
varietal
rank.
Subsection
Halepensia,
containing
the
rhizomatous
wild
fodder
sorghums
has
tetraploid
species
in
most
cases
and
members
likewise
cross
readily
with
fellow
members.
An
extreme
case
for
lumping
has
been
taken
in
the
classification
of
De
Wet
and
Huckabay
(1967)
where
both
subsections
are
put
together
under
S.
bicolor.
The
rhizomatous
forms
are
called
S.
bicolor
subsp.
halepense
and
non-rhizomatous
fodder
forms
S.
bicolor
subsp.
bicolor,
which
are
further
divided
into
four
varieties
and
one
of
these
(var.
bicolor)
is
divided
into
four
races.
However,
the
classification
of
De
Wet
(1978)
re-establishes
S.
halepense
as
a
distinct
species
and
divides
S.
bicolor
into
three
subspecies
one
of
which
represents
the
cultivated
grain
sorghums,
with
separate
races,
and
the
other
two
represent
fodder
sorghums.
Although
these
classifications
may
better
reflect
the
relationship
of
the
entities,
separate
binomials
are
kept
for
the
different
taxa
in
this
paper
purely
for
practical
reasons
of
com-
munication
in
that
these
taxa
are
known
in
Queensland
by
these
names.
Members
of
the
subsection
Halepensia
show
little
tendency
to
hybridize
with
the
cultivated
grain
sorghums
(Snowden
1955)
primarily
due
to
the
difference
in
chromosome
numbers.
For
this
reason
shattering
strains
of
S.
almum,
S.
halepense
and
S.
miliaceum
are
an
unlikely
source
of
"shattering
genes"
in
recently
discovered
shattering
off-types
of
grain
sorghum.
This
therefore
leaves
diploid
S.
verticilli-
forum
as
a
likely
donor.
However,
crosses
between
members
of
the
two
sub-
sections
have
been
made
(Celarier
1958;
Endrizzi
1957;
Hadley
1958;
Pritchard
1965;
R.
G.
Henzell
personal
communication)
so
the
shattering
Halepensia
members
cannot
be
ruled
out
as
a
source
of
genes
causing
shattering.
Introgression
between
the
two
subsections
probably
occurs,
and
it
has
been
mentioned
(Doggett
1970)
that
"there
is
steady
intercrossing
between
the
diploid
Arundinacea
and
S.
halepense,
though
at
a
low
level."
S.
sudanense
can
probably
be
dismissed
from
consideration
as
a
donor
of
genes
causing
shattering
in
that
it
is
not
a
shedder
or
the
type
that
breaks
cleanly
at
the
base
of
the
sessile
spikelet.
Shattering
off-types
collected
from
crops
to
date
have
large
grains
loosely
enclosed
by
the
glumes
(see
figure
2A)
and
possess
inflorescence
types
ranging
from
fairly
open
to
the
semi-clustered
types
approaching
that
of
forms
of
cultivated
S.
bicolor
(see
figure
13).
If
S.
verticilliflorum
is
responsible
for
the
shattering
character
the
shattering
forms
could
be
represented
nomenclaturally
by
the
hybrid
formula
S.
bicolor
x
S.
verticilliflorum.
However,
the
leaf-blades
of
the
off-types
are
as
wide
as
those
of
cultivated
sorghum,
whereas
S.
verticilliflorum
has
narrower
leaf
blades.
My
recent
survey
of
the
Burdekin
delta
region
revealed
several
colonies
of
tall
open-headed
wide-leaved
S.
bicolor
with
prop
roots
up
to
the
third
node
growing
on
the
headlands
of
sugar
cane
plots
(see
figures
14
and
15).
Most
SHATTERING
IN
GRAIN
SORGHUM
83
of
these
"wild-type"
strains
of
S.
bicolor,
are
strong
shatterers,
and
are
more
likely
to
be
a
putative
parent
of
the
shattering
off-types
of
cultivated
sorghum
found
in
the
Burdekin
delta
region
than
is
S.
verticillifiorum.
However
until
the
source
of
"shattering
genes"
is
definitely
established
it
is
probably
preferable
to
refer
to
the
shattering
forms
as
the
shattering
off-type
of
S.
bicolor.
A
state-wide
survey
of
all
grassy
sorghums
is
presently
being
undertaken
by
the
Agriculture
Branch
of
the
Queensland
Department
of
Primary
Industries
(W.
H.
Hazard,
personal
communication).
Included
in
the
data
being
collected
is
whether
the
plant
has
shattering
or
non-shattering
spikelets
and
a
rhizomatous
or
non-rhizomatous
habit.
The
aim
is
to
establish
which
species
could
be
involved
in
the
evolution
of
the
shattering
off-types
and
where
they
occur.
Dormancy
tests
of
all
collections
are
also
being
undertaken
by
the
Standards
Branch
of
the
Department
of
Primary
Industries
(R.
L.
Harty,
personal
communication).
Figure
12.
Base
of
S.
vertientifiorum
showing
prop
roots.
84
B.
K.
SI
ON
Figure
13.
Range
of
inflorescence
types
of
shattering
off-type
S.
bicolor
found
in
grain
sorghum
(Photo
by
Mr
T.
Dale,
New
South
Wales
Department
of
Agriculture).
Figure
14.
Plants
of
robust
S.
bicolor
from
the
Burdekin
delta
region.
SHATTERING
IN
GRAIN
SORGHUM
5
Figure
15,
Base
of
robust
S.
bicolor
showing
well-developed
prop
roots.
V.
ACKNOWLEDGEMENTS
I
would
like
to
acknowledge
the
help
of
officers
of
the
Agriculture
Branch,
Queensland
Department
of
Primary
Industries
for
making
available
much
sorghum
material
during
the
off-type
sorghum
survey
of
1977-78.
I
also
thank
Mr.
Moray
Finlay,
Department
of
Primary
Industries,
Ayr
for
showing
me
the
areas
of
shattering
sorghums
in
the
Burdekin
delta
region.
86
B.
K.
SIMON
REFERENCES
BENTHAM,
G.
(1877).-Flora
Australiensis
7:540.
L.
Reeve,
London.
BREAKWELL,
E.
(1915).-Sudan
Grass
and
its
advantages
over
Johnson
Grass.
Agricultural
Gazette
of
New
South
Wales.
Miscellaneous
Publication
1815.
CELARIER,
R.
P.
(1958).-Cytotaxonomic
notes
on
the
subsection
Halepensia
of
the
genus
Sorghum.
Bulletin
of
the
Torrey
Botanical
Club
86:49-62.
°toss,
D.
0.
(1932).-Aids
in
establishing
the
identity
of
Johnson
Grass
and
Sudan
Grass.
Agricultural
Gazette
of
New
South
Wales.
Miscellaneous
Publication
2897.
DE
WET,
J.
M.
J.,
and
HARLAN,
J.
R.
(1971).-The
origin
and
domestication
of
Sorghum
bicolor.
Economic
Botany
25:128-135.
DE
WET,
J.
M.
J.,
and
HOCKADAY,
J.
P.
(1967).-The
origin
of
Sorghum
bicolor.
II.
Distribu-
tion
and
domestication.
Evolution
21:787-802.
DE
WET,
J.
M.
J.
(1978).-Systematics
and
evolution
of
Sorghum
sect.
Sorghum
(Gramineae).
American
Journal
of
Botany
65:477-484.
DocoETT,
H.
(1970).-`Sorghum'
Longman,
London.
ENDRIZZI,
J.
E.
(1957).-Cytological
studies
of
some
species
and
hybrids
in
the
Eu-Sorghums.
Botanical
Gazette
119:1-10.
EvERisr,
S.
L.
(1974).-`Poisonous
Plants
of
Australia'
Angus
&
Robertson,
Sydney.
p.
245-248.
GARBER,
E.
D.
(1950).-Cytotaxonomic
studies
in
the
genus
Sorghum.
University
of
California
Publications
in
Botany
23:282-348.
GRIFFITH
DAMES,
J.
and
EDYE,
L.
A.
(1959).-Sorghum
allmum
Parodi-a
valuable
summer-
growing
perennial
grass.
Journal
of
the
Australian
Institute
of
Agricultural
Science
25:117-127.
HADLEY,
H. H.
(1958).-Chromosome
numbers,
fertility
and
rhizome
expression
of
hybrids
between
grain
sorghum
and
Johnson
grass.
Agronomy
Journal
50:278-282.
HARLAN,
J.
R.,
and
DE
WET,
J.
M.
J.
(1972).-A
simplified
classification
of
cultivated
Sorghum.
Crop
Science
12:172-176.
HAZARD,
W.
H.,
and
HENZELL,
R.
G.
(1978).-Off-type
sorghums
as
weeds
of
cultivation.
Proceedings
of
the
First
Conference
of
the
Council
of
Australian
Weed
Science
Societies
203-206.
MAGOON,
M.
L.,
SUDASIVAIAH,
R.
S.,
and
RAYYAB,
M.
A.
(1967).-Cytogenetic
studies
in
the
sub-section
Halepensia
of
the
genus
Sorghum.
Der
Zuchter
Genetics
and
Breeding
Research
37:221-225.
MCWHORTER,
C.
G.
(1971).-Introduction
and
spread
of
Johnson
grass
in
the
United
States.
Weed
Science
19:496-500.
PARODI,
L.
R.
(1943).-Una
neuva
especie
de
"Sorghum"
cultivada
en
la
Argentina.
Revista
Argentina
de
Agronomia
10:361-372.
PRITCHARD,
A.
J.
(1965).-Cytological
and
genetical
studies
on
hybrids
between
Sorghum
almum
Parodi
(2n
=
40)
and
some
diploid
(2n
=20)
species
of
Sorghum.
Euphytica
14:307-314.
PURSEGLOVE,
J.
W.
(1972).-Tropical
Crops,
Monocotyledons'
I.
J.
Wiley
&
Sons,
New
York.
SNOWDEN,
J.
D.
(1936).-`The
Cultivated
Races
of
Sorghum'
Adlard
and
Son,
London.
SNOWDEN,
J.
D.
(1955).-The
wild
fodder
sorghums
of
the
section
Eu-Sorghum,
Journal
of
the
Linnean
Society
of
London-Botany
55:191-260.
SWARBRICK,
J.
T.
(1977).-Shattercane.
Weed
Society
of
Queensland
Newsletter
10:8-12.
WHITE,
C.
T.
(1937).-Johnson
Grass
and
Wild
Sorghum.
Queensland
Agricultural
Journal
47:365-368.
(Received
for
publication
27
June
1978)
The
author
is
a
Senior
Botanist
in
the
Queensland
Department
of
Primary
Industries
and
is
stationed
at
Indooroopilly,
Q.,
4068.