Tải bản đầy đủ - 0trang
VII. Breeding for Lodging Resistance
LODGING IN WHEAT, BARLEY, AND OATS
less grain than the check varieties (and was also more susceptible to leaf
rust and mildew). It may be concluded that better lodging resistance was
achieved by deliberate breeding as well as through the selection for high
yield under fertile conditions.
The breeding of short-strawed varieties has contributed considerably to
lodging resistance but has not eliminated the problem of lodging. Under
abundant moisture and nutrient supply even the semidwarf varieties are
prone to lodge and thus prevented from yielding commensurate with their
potential. Data presented by Ephrat et al. (1965) indicate that the superiority in grain yield of a semidwarf wheat variety over a tall one decline
from 1040 kg/ha ( 2 1 % ) in a year with little or no lodging, to 250 kg/ha
when more lodging occurred. Growth regulants, such as CCC, have also
not yet presented an entirely satisfactory solution to the problem and in
any case their application involves extra expense. Therefore, lodging resistance still presents an important challenge in any cereal breeding
Resistance to lodging, unlike resistance to disease or pests, is similarly
to grain yield, a character of the plant population rather than of the single
plant. Moreover, it is strongly affected by environmental conditions. Therefore, its evaluation is rather complicated and plant breeders have made
use of various lodging indices as selection criteria. Selection according to
plant character associated with lodging resistance has also been applied,
especially in the evaluation of initial breeding material from rather comprehensive collections (Basistov, 1970; Kohli et al., 1967; Strutsovskaya,
1 . Direct Assessment in the Field
Lodging may sometimes not occur at all, or only very slightly, so that
none of the tested lines or varieties are affected. On the other hand, under
conditions extremely favorable to lodging, all the lines may lodge-which
again will not enable any distinction to be made. Moreover, lodging is
greatly influenced by variety X environment interaction effects (Fig. 5 ) .
Consequently, no single “lodging nursery” may provide valid information
for selection, which should rather be performed on the basis of evaluation
over a wide range of environmental conditions. This has already been
pointed out by various authors (e.g., Atkins, 1937; Baier, 1965; Hamilton,
1951; Murphy et al., 1958). In Denmark it was suggested to test cereal
varieties at five N levels in order to obtain an estimate of lodging resistance
(Pedersen, 1964). Considering the effect on lodging of the growth stage
MOSHE J. PINTHUS
at which the lodging-induced conditions occur, selection should be performed within groups of lines of similar maturity.
The grading for lodging resistance requires a reliable rating of the severity of lodging which should consider the prevalence as well as the degree
of lodging. Unfortunately, this has not always been strictly observed and
no standard quantitative method has been adopted. Furthermore, the grading should distinguish between stem lodging and root lodging (Section 11,
A), and should be repeated several times throughout the period from head-
ZIG. 5. Lodging (rates from 0 = no lodging, to 4 = complete lodging, averaged
ove three replications) of ten spring wheat varieties on four different fields at
St. Paul, Minnesota, 1958.
ing to harvest. A single rating at harvest time may overestimate late lodging
resulting from culm breakage of dry straw, and underestimate lodging prior
to heading because of the recovery from it. Murphy et al. (1958) found
a high correlation ( r = 0.89) for oats between the percentage of culm
curvature, i.e., culms which had recovered from early lodging, and the
prevalence of lodging at full ripeness. However, recovery from early lodging may sometimes increase the resistance to later lodging (Section 11, C) .
Therefore, it is suggested that the rating of lodging should be based on
the position of the lower culm internodes rather than on the inclination
of the upper part of the culm and of the head. This is based on the assumption that if the external factors conducive to lodging had operated
LODGING I N WHEAT, BARLEY, AND OATS
at a later growth stage, “unrecoverable” lodging of similar severity would
All these difficulties in making a valid direct assessment of lodging resistance in the field, warrant the attempts of breeders to find alternate
methods. This is of vital importance for the evaluation of early breeding
material consisting of a great number of lines, each with only a small quantity of seed.
2 . Auxiliary Methods
Lodging resistance and associated characters of cereal plants grown in
pots have been evaluated in wind tunnels (Bauer, 1964), and mobile wind
tunnels have been used for this purpose in field plots (Section 111, A, 4).
However, these methods are complicated and costly and their reliability
with regard to root lodging is doubtful.
Turkova and Suan (1966) suggested that lodging resistance be estimated by the degree of bending of the second culm internode following
plant treatment with auxins and with reductases. Varietal differences in
lodging had indeed been established following spraying of wheat with
heteroauxin (Petinov and Urmantsev, 1964).
The attempts to identify lodging resistance at early growth stages are
of particular interest. Percival (1921 ) claims that wheat varieties with
young spreading shoots are firmly anchored to the soil and thus resistant
to root lodging, whereas erect young shoots indicate a poor spread of the
root system and consequent lodging susceptibility. The relations between
lodging resistance in the adult plant and mechanical strength in the seedling
stage has been investigated by Baier (1965) and by Holienka and HruSka
( 1962). The latter investigators could distinguish the most susceptible varieties, whereas Baier concluded that this relation was not close and consistent enough to provide a reliable criterion of evaluation.
The snap test (Section IV, C, 1) has proved useful for the evaluation
of lodging resistance of differing lines of oats (Hess and Shands, 1966).
3. Lodging Indices
The recognition that lodging resistance cannot be attributed to any single
plant character, the correlations which have been established between
varietal differences in lodging and various characters and those found
among these characters, have led to the combining of several plant characters into lodging indices. Some of these indices were closely correlated with
lodging in the field and could therefore be utilized as selection criteria of
The ratio of culm length to diameter of basal internodes was found to
be rather closely correlated with lodging resistance ( r = -0.67 to -0.85)
MOSHE J. PINTHUS
in various studies of the three cereals (Basistov, 1970; Hamilton, 1941;
Hansel, 1957; Wav, 1960). A little more sophisticated modification of this
index is the ratio of area of culm wall cross section to culm length (Wag,
1970). Indices which take the root development into consideration are
of particular interest. The ratio of culm length: number of coronal roots
per culm was applied by Hansel (1957) but he did not find it superior
to the first-mentioned index. The top weight :root weight ratio has been
found to give an indication 01 lodging in wheat (Malkani and Vaidya,
1956) and oats (Sechler, 1961 ) . This index could be improved by adding
to it an estimate of breaking strength (Malkani et al., 1959), and still
better results were obtained when culm length was also included (Vaidya
and Singh, 1963).
Applying discriminate function analysis, Hamilton ( 195 1) developed an
index in which 10 x culm diamter ( m m ) 5 x root rating (on a 1-10
1 x plant height (inches) were combined. [The detailed analytical procedure in this case is presented as Example 17-2 by Goulden
( 1952) .] With this index he could readily differentiate lodging-resistant
oat varieties from susceptible ones. Similarly, ZeniSEeva and Lekes (1966)
combined culm length, length of second internode, and breaking strength
of culms and roots into an index ( Y ) which was found, with all three
cereals, to be very closely correlated ( r = 0.90) with lodging resistance.
Successful ranking of lodging resistance of winter wheat varieties was
achieved by Miller and Anderson (1963) with the quotients LBC/Ht or
LBC/(log Ht/3)?, where LBC (load-bearing capacity) is the product of
the weight (g) supported by a culm and the horizontal displacement (cm)
of the spike from the vertical position, and Ht = culm length (cm). LBC
is determined by hooking a chain, of known weight per link, to the head
of the plant.
High values of r (-0.8 to -0.9) were found by Oda et al. (1966)
for the correlation between lodging resistance of wheat and barley and
the lodging index L = Z"W/wM, where 1 = culm length (cm), W = total
fresh weight per tiller (g), w = culm dry weight per tiller (g), and
M = bending moment at breaking ( 1 0 0 g-cm), which was determined
by a special apparatus.
The most widely used measure for estimating lodging resistance seems
to be the cLr factor devised by Grafius and Brown (1954) and found
by them to be effective for the selection of lodging-resistant lines of oats.
cLr = F / b , where b = culm length (cm), and F = bending force (g),
which is determined at the point where the culm stops its downward bending, as the weight of the links of a chain hanging down from the head
of the plant to the ground. Close correlations between this factor and lodging resistance were also found for barley (Grafius, 1958) and for wheat
(Roman and Muszyhska, 1962; Waa, 1960), whereas no such correlation
was found by Mukherjee et al. (1967). In spite of the close correlation
of the cLr with lodging resistance of oats, it was found inferior in this
respect to the snap test (Section IV, C , 1) by Murphy et al. (1958) and
more expensive than the snap test by Frey et al. (1960). Jellum (1962)
found that the cLr values of oats changed in relation to maturity, declining
during a 15-day period following anthesis and then leveling off. Bauer
(1964) indicated that in wheat the cLr factor overestimates the role of
culm rigidity in lodging resistance, and underestimates the role of culm
length; he suggested to multiply the factor by 100/culm length.
The correlation between the cLr and lodging resistance of oats was
found to be affected by differences in the area of the heads (Grafius,
1958). This index should therefore apply primarily for the evaluation of
lodging resistance within groups of similar panicles.
An increase in the value of lodging rates assessed in a single field trial
was attempted by combining them with the respective snap scores, cLr
values, and estimates of culm curvatures, into a lodging-resistance index
(Murphy et al., 1958).
The practical value of any of the above-mentioned indices is dependent,
in addition to its correlation with lodging in the field, on the number of
lines which can easily and inexpensively be evaluated by it. None of them
can replace, in the final stages of a breeding program, the assessment of
lodging resistance in the field under a wide range of environmental
The earliest relevant work seems to be that of Howard and Howard
(1912) in India. They investigated F, populations of crosses of common
wheat varieties with strong roots and weak straw X varieties with weak
roots and strong straw and concluded that strength of roots and strength
of straw were independently inherited and that lodging resistance was due
to the combination of both characters.
From the few studies which are available on the inheritance of lodging
resistance proper, the following points are apparent. Different degrees of
dominance have been observed (Aastveit, 1962; Nasr et al., 1972;
ZeniSEeva, 1968). Rather high heritability estimates ( > 5 0 % ) were found
in some cases (Hess and Shands, 1966), intermediate (25-50%) ones in
other cases (Nasr et al., 1972), and low values in still others (ZeniSEeva,
1968). Maternal effects on lodging resistance were observed in reciprocal
crosses of barley (Zhivotkov, 1968). Maternal effects have also been ob-
MOSHE J. PINTHUS
served on the inheritance of the length and diameter of the basal culm
internodes in progenies of crosses between lodging-resistant and susceptible
wheat varieties (Roman, 1962a).
In other crosses between lodging-resistant and susceptible barley varieties (ZeniSEeva, 1968), incomplete dominance was found of resistance,
estimated through the lodging index Y. Various degrees of dominance as
well as recessiveness of lodging resistance, estimated by the cLr method,
were obtained by D. A. Wheeler in different oat crosses (Jensen, 1961).
Rather low heritability estimates (15% ) were obtained in this study for
the F,-F, comparison; this was also the case in another study of the cLr
values in segregating oat crosses (-Frey and Norden, 1959). In a cross
of a lodging-resistant x susceptible wheat variety, Atkins (1938) found
intermediate weight per unit length of culm with some tendency to the
higher parent in F,. Significant correlations (r = 0.6) were found between
the expression of this character in F, plants and F, progeny lines, and
between F, plants and F, progenies.
In a certain wheat cross, stem breakage was found to be controlled by
a single gene, the strong straw being slightly dominant (Boyce, 1948).
In another study (Kohli et al., 1970), breaking strength as well as culm
diameter revealed a quantitative mode of inheritance, and transgressive
segregation for both characters was observed in F,.
Inheritance of stem solidness in wheat has been studied in relation to
sawfly resistance. This and the inheritance of some other culm characters
of wheat have been reviewed by Ausemus et al. (1967). Inheritance of
the number of vascular bundles and of the size of the sclerenchyma ring
was studied in crosses between lodging-resistant and susceptible wheat
varieties by Jankovib ( 1966b), who also reviewed some of the earlier literature on this subject. Both characters show polygenic inheritance and certain degrees of dominance of the greater values. Similar results for vascular
bundles and culm-wall thickness were obtained by Roman ( 1962a).
The number and thickness of coronal roots in F, populations of crosses
between lodging-resistant and susceptible wheat varieties were found to be
intermediate by Roman (1962b). In barley, dominance of root tensile
strength was observed by ZeniSEeva (1968). Inheritance of the length of
the root crown of oats was found to be controlled by at least three factor
pairs and also to be affected by many modifying genes; dominant factors
may be involved in the determination of both long- and short-root crowns
(Sechler, 1961 ) .
It may be concluded that the inheritance of most root and culm characters associated with lodging resistance is rather complex and has not yet
been sufficiently investigated. A conspicuous exception, however, is the
culm length character of wheat, which has been studied extensively (Briggle and Vogel, 1968; Johnson el al., 1966; Woo and Konzak, 1969). Fac-
BARLEY, AND OATS
tors influencing stem length were found on many chromosomes; several
major, as well as many minor, genes have been shown to control this character in different crosses; the dwarfing factors are generally recessive, but
dominant ones have also been observed. From the point of view of lodging
resistance it is of major importance that there are indications of the existence of independent factors controlling the length of the different internodes (Paquet, 1968).
The combination of lodging resistance with other favorable agronomic
traits through the intentional cross breeding of parental varieties excelling
in these characters has been obtained in wheat (Pal, 1944), barley (Ambastha, 1962), and oats (Ponomarenko, 1971 ). More often, however, the
improvement of lodging resistance should be ascribed to selection for this
trait in general breeding material. In this respect the snap test (Section
IV, C, 1) has been particularly effective as a selection criterion for oats
(Murphy et al., 1958).
Considerable success has been achieved in the selection of lodging-resistant mutant lines in all three cereals. This subject has been reviewed
in detail by Scarascia Mugnozza (1965). The Swedish barley variety Pallas
is one of the high-yielding and lodging-resistant varieties obtained through
mutation breeding (Gustafsson and Ekman, 1967). Several other lodgingresistant barley and oat varieties, selected directly after mutagenic treatment or out of the progenies of crosses between the mutants and other
varieties, have already been released (Sigurbjornsson and Micke, 1969).
Rather impressive success in mutation breeding of lodging resistance has
been achieved with durum wheat in Italy (Scarascia Mugnozza and Bozzini, 1968). Several of the resulting mutant lines have excelled in lodging
resistance and grain yield in extensive trials at many locations throughout
the Mediterranean region and the Middle East (Bogyo et al., 1969).
The increased lodging resistance of the mutants has been associated in
many cases with reduced culm length, but this has not always been the
case. Moreover, several lodging-resistant durum mutant lines were significantly higher than their mother varieties (Scarascia Mugnozza, 1965).
Mutants of the Austrian durum wheat ADUR showed less reduction in grain
yield when their lodging resistance was not accompanied by reduction in
height (Nagl, 1971 ) , The improved lodging resistance of various mutants
could be attributed to several characters other than culm shortness: number of elongating internodes, relative length of the different internodes,
culm anatomy and solidness, and structure of the basal internodes.
Breeding for various anatomical culm characters related to lodging resistance and particularly for those of the basal internodes, has been sug-
MOSHE J. PINTHUS
gested (Nitr, 19160; ZeniSEeva and LekeS, 1971) . However, the greatest
achievements of lodging resistance through the breeding of any single character have been accomplished by reducing culm length. In addition to the
famous semidwarf wheat varieties (Borojevic, 1968; Briggle and Vogel,
1968), new barley and oat varieties which owe their lodging resistance
to reduced culm length have been bred. BARTELbarley in Arizona (Day
et al., 1972) and RANDOM oats in Canada (Kaufmann, 1971) are only
two examples of recently released varieties.
Semidwarf varieties may suffer from several defects, e.g., reduced seedling emergence and increased susceptibility to various diseases, which
present urgent challenges to current breeding programs (Berbigier, 1968;
Briggle and Vogel, 1968). Under certain agricultural conditions their reduced straw yield has also been disadvantageous (Pal, 1966). Nevertheless, the availability of suitable genetic material, the ease of selection, and
the possibility of starting selection in early generations, in addition to the
previous achievements, justify and encourage the improvement of lodging
resistance through the breeding of reduced culm length.
Lines with improved lodging resistance were obtained following two or
three cycles of selection, out of established winter wheat varieties, of plants
with well-developed root systems (Il'inskaya-CentiloviE and TeterjatEenko,
1961); when selecting, the spread, number, and thickness of the coronal
roots were considered. These authors claim that the improved root system
of the selected types was accompanied by an increase in the thickness of
cell walls and in the number of vascular bundles in the sclerenchyma of
However, generally root characters have been neglected in the work
of breeding for lodging resistance. This has obviously been due to the practical difficulties involved with the evaluation of the root system. However,
the causal relationship between this character and root lodging, and the
weak variety X environment interaction effects on the spread of the roots
in the upper soil layer (Pinthus, 1967a), warrant some breeding efforts
in this direction. This may be particularly rewarding with the improvement
of high-yielding, short-strawed varieties which nevertheless are prone to
lodge under conditions of high fertility.
Increased Exploitation of Yield-Promoting Factors Due to the
Prevention of lodging
The exploitation of yield-promoting factors, such as N fertilization or
irrigation, is dependent on the prevention of lodging. This may be illustrated by the following examples.
In an experiment in Italy (Scarascia Mugnozza et al., 1965), the durum
LODGING IN WHEAT, BARLEY, AND OATS
wheat CAPPELLI and its lodging-resistant mutant B 132, without N-fertilization, yielded 2628 and 2643 kg/ha, respectively. After the application
of 120 kg/ha N, the grain yield of the mutant, which did not lodge, was
increased by 82 % , whereas that of Cappelli, which lodged considerably,
was increased by only 28%. Similarly, certain trials in Britain have shown
that the differential response of barley varieties to high rates of nitrogen
could be attributed almost entirely to varietal differences in lodging
(Holmes, 1969). In India, grain yield of a semidwarf lodging-resistant
wheat variety could be increased considerably through adequate irrigation,
whereas the response of tall varieties to irrigation was considered to be
limited because of lodging (Misra et al., 1969).
In many cases, the prevention of lodging by the application of CCC
enabled wheat to benefit from high N fertilizer rates (Humphries, 1968a;
Wunsche, 1970). Similarly, under certain field conditions in Israel, CCC
has increased the beneficial effects of irrigation on wheat yield.
Up to the level of grain yield which has hitherto been achieved under
the most favorable conditions, lodging indeed constitutes the main factor
limiting the response of cereals to yield-promoting agents. Consequently,
the prevention of lodging through the application of CCC as well as
through the cropping of semidwarf varieties, may enable the attainment
of this grain yield level. However, beyond this level, any further increase
in grain yield will depend on the more effective use of ambient light. This
may perhaps be achieved through the development of hybrid varieties.
However, considering the dominant nature of the height-controlling genes
(Section VII, B ) and some other information on hybrid wheat (e.g., Johnson and Schmidt, 1968), the hybrids are expected to be tall and therefore
susceptible to lodging. On the other hand, the cereal “ideotype,” proposed
by Donald (1968), wiIl be a weak competitor, to be planted densely, having a single culm with a big awned head. It may be added that long upper
internodes, which should complete their elongation by the onset of grain
development, would also be advantageous for the assimilation of carbohydrates and their supply to the grain. The yielding capacity of such an ideotype will obviously be limited by lodging.
Consequently, further efforts to prevent lodging, whether by chemical
means or by breeding lodging resistance, will be needed in future in order
to benefit from the anticipated increase in grain yield potential. It may
be concluded that there is a permanent requirement for increased lodging
resistance which is linked to the continuous process of grain yield
Mr. D. Trifon, C.E. contributed considerably to Section 11, B. Mrs. V. Priel
read and corrected the manuscript. Their assistance is greatly appreciated.