Tải bản đầy đủ - 0trang
V. Sugar Beet Improvement Entering New Era
G . H.
OWEN, A N D DEWEY STEWART
companies are continuing their cooperation and financial support of the
federal research. I n addition to the federal breeding investigations, the
major beet sugar companies of western United States have extensive
research programs aimed at breeding varieties adapted to their respective areas, and these varieties have had important effect on the general
improvement in sugar beet production of the respective districts concerned. Interchange of ideas among sugar beet breeders is fostered by
a Breeders’ Forum sponsored by the Beet Sugar Development Foundation. This forum has centered its attention upon an appraisal study of
inbreds, participants contributing such breeding material as available
for replicated trials whereby a catalogue of important characteristics of
available inbred stocks could be compiled.
The research on sugar beet agronomy conducted in Europe, as s u marized by Roemer (1927), and that on genetics and breeding, as reported by Munerati (1929) , Vilmorin (1923), and Schneider (1939)
constitute the starting points for many of the advances made in sugar
beet investigations in the United States. The investigations on the
morphology of the vegetative and reproductive organs of the sugar beet
by Artschwager (1926, 1927) are built upon the earlier European
studies, especially those of De Vries (1879) and others. The European
research on wind and insect transport of pollen has been summarized by
Archimowitsch (1949). Investigations by Stewart (1946) in the United
States have centered on wind pollination as of greater significance. The
presence of viable pollen in the upper air currents has been demonstrated by Meier and Artschwager (1938) and this is a factor in obtaining strict isolations. In the sugar beet seed field, however, the dispersal
of wind-borne pollen, as shown by Stewart and Campbell (1952), is
highly localized and shows a curvilinear relationship, the amount of
pollination from a given pollen source being inversely proportional to
the distance in feet. The study of methods of obtaining pollen-free isolations as initiated by Nilsson (1922, 1923) a t Svalof, Sweden, was extended by Brewbaker (1934) ; as a result of these studies, investigations
in the United States have chiefly employed bagging techniques with
sugar beets coming into flower, in which individual branches are enclosed in parchment or Kraft paper bags. Only limited use has been
made of the cage-type of isolations suggested by Munerati (1920) or the
cloth bag types of Vilmorin (1923).
2 . Polycross Method Applicable in Breeding Sugar Beets
Breeding and genetics research with sugar beets has drawn heavily
on the advances made with other crop plants, both for theory and
methodology. Thus, as previously noted, the polycross method as de-
SUGAR BEET IMPROVEMENT
vised for alfalfa improvement by Tysdal and Kiesselbach (1944) is
being applied in the breeding of sugar beets for black root resistance.
Brewbaker and Wood (1948) reported results with sugar beets obtained
essentially by this method.
The first step in the polycross method conforms to the principle originally set forth by Vilmorin, that the best criterion of the breeding
value of a selected plant is the performance of its progeny. Open-pollinated seed, produced from selected plants which are brought to flower
in an isolated plot where random pollination is expected to occur, are
used in field tests to evaluate the breeding value of the mothers. The
genotype of the mothers is propagated through selfed seed or by clones
which can be established without difficulty from stalk cuttings, as indicated by Owen (1941). The outstanding mothers, as indicated by the
field performances of the open-pollinated progeny, are intercrossed to
produce a synthetic variety, or possibly a strain for a second cycle of
selection. Repeating the breeding cycle would be essentially recurrent
selection, which has been found, for other crops, an effective method
3 . Application of Other Breeding Methods
Tester strains to evaluate inbreds have been investigated by Deming
(1942), who suggested utilization of the red garden beet as the topcross parent, because of the ease of recognition of the hybrid. This
method has been given further study by Oldemeyer (1954). Marker
characters, especially the simple Mendelian character R which governs
red hypocotyl and red bud color in sugar beets, have been utilized for
identification of the hybrid between RR, Rr, and rr genotypes. The
inheritance of hypocotyl color was first worked out by Linhard and
Iverson (1919) and confirmed by Keller (1936). Nuckols (1931)
showed that hypocotyl color did not influence yield of roots.
I n Europe and the United States, the correlation of morphological
characters with yield and quality has been studied. Pritchard (1916a)
and Pack ( 1930) have reviewed European literature and reported their
own findings. The correlation of leaf, stem, and root characters, or of
readily noted growth habit with sucrose percentage, has not been fruitful in revealing indices for high sucrose. Artschwager (1930, 1952) has
suggested a positive correlation of high ring density and core type of the
root with high sucrose.
As discussed earlier, European breeding methods for sugar beets,
particularly those of the seed-breeding establishments, were less influenced by Johannsen’s pure-line concept than those of the United
States; hence, in any comparison of breeding materials available for
G . H. COONS, F.
OWEN, A N D DEWEY STEWART
analyses of the role of genetic factors, this difference in approach
should be borne in mind. Apparent contradictions between European
and American results frequently can be explained by the consideration
that, with the former, conclusions often were drawn from a population
rather than from a single biotype. The European research on sugar beet
improvement and the related studies on cytogenetics constitute a body
of knowledge that should not be neglected. This review, although centering on American contributions, has made some attempt to tie in these
with the pertinent European contributions.
4 . Hetel-osis
It was generally postulated as, for example, by Vilmorin (1923),
following the discovery of heterosis in Zea mays, that sugar beet hybrids
would show a similar type of response. Only observational evidence was
adduced in this regard prior to the experiments at Fort Collins in 1932
by Stewart et al. (1940).
I n the Fort Collins study, 41 F, hybrids, obtained by mating inbred
sugar beets in isolated seed plots, were used. To produce the hybrid, the
two strains to be crossed were brought to seed production in a location
at considerable distance from any other sugar beets. In the progenies
grown from the various hybrid seed lots (usually reciprocals were
taken), there occurred not only F, plants but selfs of the inbred strain
that was the seed bearer. Where possible, the data on performance of a
progeny were based upon the F, class of plants, but in many cases the
entire population, regardless of extent of hybridization, was taken. The
comparisons were made in 8X or 6X replicated tests, the quantity of
seed dictating the type of test that could be conducted. In 31 of the 41
cases tested, the root weight of the hybrid was significantly greater than
the root weight of the parent strains appropriate for the comparison.
The average gain in root weight of the hybrid over the parental mean
was 42.5 per cent, but it was recognized that this percentage is greatly
influenced by the relative yielding abilities of the inbreds entering a
given cross. The average sucrose percentage of the hybrids did not
differ significantly from the average of the inbred parents. I n the tests,
effects attributable to the resistance to leaf spot of certain inbreds and
hybrids could not be separated from effects associated with hybridity.
The conclusion was drawn that direct appraisal of the effects of hybrid
vigor in increasing root weight could not be made, but the gains in root
weight of certain hybrids over the commercial brand used as a check
were enough beyond those reasonably attributable to superior disease
resistance to indicate definite increase in productivity from heterosis.
A further study of heterosis in sugar beet single crosses was con-
SUGAR BEET IMPROVEMENT
ducted a t Ault, Colorado, in 1942 and 1943 by Stewart et al. (1946),
using 35 sugar beet hybrids obtained by mating (1 ) 11 inbred strains
and (2) an open-pollinated variety (US 22) with each of 3 pollen parents, inbred US 215, inbred US 216, and EUROPEAN CHECK.In two tests
(one 7x replication, the other 8 x ) ,root yields, sucrose percentages, and
sugar production were studied in hybrids and parents. Leaf spot was
not a factor in these tests, and the data were drawn exclusively from
identified F, plants. All the inbred strains were relatively high yielding.
Futhermore, the variety known as EUROPEAN CHECK is very productive
under conditions where leaf spot is not a factor. With such initial
breeding material, relatively few hybrids would be expected significantly to exceed, in the attributes measured, the means of parents or
FIG. 8. Hybrid vigor in sugar beets. Representative roots of the hybrid US
215 X 216 are shown at the center. Piled at the right and left are similar numbers
of roots of US 215 and US 216, respectively, the parents. (Arlington Farm, Virginia,
October 25, 1939.)
EUROPEAN CHECK. It was found that, as a class, the hybrids were significantly superior to the parents. Definite heterosis effects were shown
to occur in sugar beets, but in comparison with the higher yielding inbreds and with a high-yielding commercial variety, relatively few
hybrids gave significantly high performances. Six of the hybrids were
significantly above EUROPEAN CHECK in production of sugar. There was
evidence from the experiments that, with many inbreds to evaluate,
EUROPEAN CHECK might prove effective as a tester. Doxtator and
Skuderna (1946) and Kohls (1950) have shown in their reports a
similar tendency for heterosis to occur in sugar beet hybridizations.
An illustration of the type of response obtainable from appropriate
matings is given in Fig. 8; the hybrid, US 215 X 216, is shown at the
center with the parental inbred varieties US 215 and US 216 at right
and left respectively.
G . H . COONS, F.
OWEN, A N D DEWEY STEWART
With the curly-top-resistant varieties, there is clear-cut evidence
that hybrid varieties, properly chosen for adaptation to the particular
districts concerned, hold the key to increased productivity. I n particular, the data from Utah, Idaho, and Washington experiments on the
performance of hybrids are very impressive. For example, at Twin
Falls, Idaho, in 1952, US 22/3 produced 31 tons of sugar beet per acre,
whereas the hybrid produced 35 tons with an improved sugar content.
I n a series of tests conducted in Utah and Idaho ovei- a four-year period,
the yields of the better hybrids have shown consistent gains over
US 22/3, ranging from 10 to 20 per cent in acre yields of sugar.
It must be recognized, however, that the hybrid variety as made between two inbred strains does not have the range of adaptability shown
by an open-pollinated variety, and more years of tests are required to
fit the various hybrids into the districts for which they are best suited.
A hybrid that was highly productive in a number of tests suffered
badly from blister beetle attack when planted in Washington. This illustrates what may happen when the genetic base is narrowed in the
hybrids from what is presented in the open-pollinated variety. It seems
safe to forecast that research on hybrids shortly will have progressed
far enough so that dependable, high-yielding hybrids can be released
€or the curly top area.
5. Male-Sterility in Sugar Beets
Male-sterility produced by combining cytoplasmic and genic inheritance was found by Owen (1945) in cross-pollinated varieties of
sugar beets bred for resistance to curly top, notably in US 1, US 33,
and other selections from this basic breeding stock. Complete malesterility was characterized by the occurrence in the flowers of white,
empty anthers (Fig. 9). Semi-male-sterility also was found. Assuming
two types of cytoplasm, S for male-sterility and N for normal, and two
Mendelian factors, X and 2, the majority of the evidence, including
striking differences from reciprocal crosses, indicates a genic constitution of male-sterile and semi-male-sterile beets, as follows: Sxxzz types
are completely male-sterile; S X x z z or SxxZz are semi-male-sterile, usually without viable pollen; and SXxZz are semi-male-sterile, usually
with some viable pollen, and sometimes indistinguishable from the normal hermaphrodite.
The male-sterile factor in beets, therefore, conforms with the pattern
described by Gairdner (1929) for flax, and that for onions by Jones
and Clarke ( 1943), but in both of these plants only one Mendelian factor is involved. Cytoplasmically inherited male-sterility off ers many
advantages to the sugar beet breeder. However, to be most useful, it
SUGAR BEET IMPROVEMENT
must be easily obtained in progenies in which 100 per cent of the plants
are completely male-sterile. Progenies approaching this condition are
not only possible but are common among controlled crosses to the malesterile beets. Some crosses to male-sterile beets do not produce uniformly
FIG.9. Sugar beet flowers. A , Male-sterile; ~ 1 2 B. , Normal, immediately after
dehiscing of anthers; x12. C, Cross section of male-sterile flowers in bud stage;
x21. D , Cross section of normal flower in bud stage; x21.
male-sterile off spring, and genic effects have been recognized which appear to modify the expression of cytoplasmic inheritance.
Owen (1948) has designated three general types of pollen parents
as 0, I, and 11. These pollen parents carry N type of cytoplasm and can
be used either as females or pollen parents. They cannot be dis-
G. H. COONS, F. V. OWEN, AND DEWEY STEWART
t,inguished by their appearance, but they represent different genotypes
and therefore differ in breeding behavior. If crossed to a cytoplasmically
male-sterile plant ( S x z z z ) , type 0 ( N x x z z ) gives an offspring that is
all male-sterile; type I gives mostly male-sterile offspring, but 5 to 30
per cent may be semi-male-sterile; and type I1 gives a progeny that
is 50 per cent male-sterile, 25 per cent semi-male-sterile, and 25 per
cent more or less normal with respect to pollen production. Obviously,
for effective use of the male-sterile factor, types I and 11, which are
heterozygous genotypes, are undesirable. Owen (1950) has pointed out
the need of proved stocks of Sxxzz genotype as male-sterile parents in
matings to establish the three types. Some plants which may be considered as SxxzZ in constitution may appear completely male-sterile
owing to environmental influences.
By a series of backcrosses, the male-sterile factor can be incorporated
in a variety, once 0 types are found. After two or three backcrossings,
using the 0 type as the recurrent parent, of course, the male-sterile
plants resemble the pollinator, and for practical purposes the third
or fourth backcross generations may be considered its pollen-sterile
equivalent. The practical utilization of the male-sterile equivalent requires that an 0 type pollinator be isolated and preserved. With it a
constant and 100 per cent male-sterile progeny can be produced essentially representative of its perfect flowered prototype.
The male-sterility factor puts in the .hands of the plant breeder a
very powerful tool. As mentioned previously, the production of hybrids
between inbred lines is relatively inefficient if intercrossing must be
left to chance. By virtue of their very nature and the means employed
in their purification, inbreds, after many years of inbreeding, tend to
become self-fertile and as hermaphrodites are ineffective in hybrid seed
production. Now that cytoplasmic male-sterility is available, selffertility of an inbred strain is an asset in preserving the 0 type after
it has been located and is no handicap when used as a pollinator.
In addition to cytoplasmic male-sterility, Mendelian sterility has
been found in sugar beets. Owen (1945) had noted this type and
in more recent work (1952) has isolated plants carrying either genes
a, or a,, the symbols chosen to designate abortion of pollen. He
has proposed the following utilization of both types of male-sterility
in the production of improved varieties to give a four-way hybrid
( A X B) X ( C X D) carrying desirable characters. This may be cited
as illustrative of the use of male-sterility in sugar beet improvement.
As proposed for production of a curly-top-resistant variety, strain A
can be a male-sterile phase (Sxxzz) of a variety carrying curly top
resistance, and it is mated with an appropriate 0 strain ( N z z z z ) B as
SUGAR BEET IMPROVEMENT
pollinator to give a completely male-sterile hybrid progeny. The C
grandparent may be a Mendelian male-sterile ( a a ) from curly top
strains having high sucrose percentage, such as US 35, and rogued to
give only pollen-sterile plants. The D grandparent would be the pollinator ( A A ) to give the hybrid C x D. Sample crosses of the respective
types are now under production, the first-named single cross ( A X B)
being male-sterile and the latter Aa, a heavy pollen producer that is extremely vigorous and known to transmit desirable attributes to the offspring.
Thus, the production of double hybrid sugar beets is feasible and
will not add greatly to the cost of growing seed over present methods.
The pollinator indicated as C x D may be discarded before harvesting
the Commercial seed, since it is required in small proportions, but satisfactory yields can be obtained from the seed bearer A x B. Malesterility conditioned by Mendelian genes in grandparent C can be produced to the extent of 50 per cent of the population by appropriate
crosses. Therefore, the job of roguing to produce the pollinator C x D
is not a serious problem.
Other combinations are also under test. If monogerm research is
successful, the female parent A x B in the double cross may be taken
from monogerm types. The successful production of male-sterile monogerm types has been reported by Owen et al. (1945). The pollinator
C x D need not be monogerm, for commercial seed is taken from the
monogerm male-sterile parent, A x B.
In addition to using a gene for pollen abortion, Owen suggests utilizing gene B for annual habit (Owen, 1952; Owen et al., 1954). With the
gene B present, thermal induction is no longer necessary and three or
four generations can be grown in a single year under warm temperatures and continuous illumination. After the desired degree of homozygosity has been reached in successive backcrosses and the desired gene
or genes have been transferred from the nonrecurrent to the recurrent
parent, genes B for bolting and a for abortion of pollen may be eliminated by selfing. No injury should result to the final inbred by carrying gene B for bolting up to the last backcross operation. So long as the
annual habit is produced by a single gene and segregation is clear-cut,
this should not influence the bolting tendency of the final inbred.
6 . Monogerm Sugar Beets
The seed ball of the beet is a glomerule containing one, two, three,
to many true seeds. A sugar beet plant that would bear single-germed
fruit has long been sought and, fully 50 years ago, experimental work
to this end was started. It had no success, but the advantages of mono-
G. H. COONS, F. V. OWEN, A N D DEWEY STEWART
germ seed as stated then as a means of saving hand labor in our beet
fields are just as cogent today as when first proposed by Palmer (1918).
During World War 11, as a labor-saving device, Bainer (1942) devised
a method of shearing multigerm seed balls to produce planting stock
that was prevailingly single-germ. This was widely adopted, and
Bainer’s improved method (Bainer and Leach, 1946) of decorticating
seed, plus the breaking up of large seed balls, is now used to provide
planting stock for a high percentage of United States acreage. Processed
seed as now used is obtained by rubbing off the corky ridges of the seed
balls, by breaking the large seed balls, and by screening, so that the
inch in diameter, broken
planting stock ranges between 6/s4 and
pieces and small seeds being sifted. out and discarded and larger ones
screened off for further size reduction. A seed piece usually contains
from one to two germs to, occasionally, three germs. As a result, carefully sized seed can be used in precision drills set to drop a single seed
piece at about l-inch intervals in the drill row.
Processed seed and the mechanized operations that are made possible by it have made definite contributions as valuable labor-saving
methods. The importance of these to industry has been summarized by
Smith (1950). The sparse stand that is obtained with processed seed
can be thinned by down-the-row stand-reducing machines that take out
at random, by means of rotating cutting knives, a predetermined portion of the row, the grower being able to go over the row a second or
even a third time to accomplish a greater degree of stand reduction, if
such is desired.
It is generally conceded, however, that the processing of seed is
wasteful and expensive. Many farmers enthusiastic for processed seed
still have laborers perform by hand the customary operations to give the
desired stand of single plants uniformly spaced and thoroughly weeded.
Others furnish their laborers with long-handled hoes to go over the
fields for trimming out surplus plants and to give additional weed control.
For these reasons, the industry had great interest in the discovery of
monogerm seed by V. F. Savitsky and his colleague, Bordonos (1941),
in the U.S.S.R., and took steps, when the former was driven out of that
country, to make it possible for him to continue his genetical research
in the United States. Knowing what to look for, Savitsky found in
Oregon in a variety designated as MICHIGAN HYBRID 18, a few plants
the seed balls of which were truly monogerm. From these, Savitsky
(1950) obtained his monogerm race SLC 101 and a companion variety
SLC 107, both considered true monogerm plants. Savitsky (1952a, b)
has shown that, in SLC 101, monogermness is controlled by a single re-
SUGAR BEET IMPROVEMENT
cessive Mendelian factor. Hybrids have been made by Owen, et al.
( 1954) between the monogerm types (mm) and multigerm varieties
( M M ) having high curly top resistance. Stewart (1952) has made hybrids with multigerm leaf-spot-resistant varieties. The monogerm varieties found by Savitsky tended to be of a slow-bolting type-a desirable
character in itself, but one which has complicated the production of
hybrids with monogerm stocks.
Incorporation of the monogerm character has been accomplished
either by selection of the segregates in the F, and F, generations following the first cross or by backcrosses followed by reselection for monogerm types (Savitsky, 195213).
It is too early to predict when monogerm lines will be available for
commercial use. It is obvious that, with a commercial variety at a given
stage of development, several years are required to make that variety
monogerm. In the meantime, further improvement of the multigerm
variety may make it desirable to start afresh. Once varieties are stabilized for major characters, the job of making them monogerm can be
Owen has suggested a method that has been used in seed production,
by which male-sterile monogerm types of reasonable productivity and
with some curly top or leaf spot resistance, as the case may be, can be
immediately utilized. The male-sterile monogerm type is used as the
seed bearer and the pollinator can be any multigerm type chosen because of qualities it may contribute to the hybrid. The F, seed produced
on the male-sterile monogerm seed bearer will, of course, be monogerm.
New stocks of seed must be made each year by the same procedures,
since any direct increase of the hybrid would give a multigerm product.
In the more northern latitudes of Europe and America, bolting in
early-sown sugar beets has been of great concern for many years.
Munerati (1942) gave a review of the literature of more than 250
references to bolting, as well as the results of his own researches. He
established strains resembling the sugar beet that passed from the vegetative to the reproductive phase of growth in only a few weeks after
germination, and with continuous light exposure, as many as five generations could be obtained in a single year. Annualism, as established
in this strain, was transmitted as a dominant character in crosses with
sugar beets of the normal biennial cycle of growth. Abegg (1936), using
an annual strain supplied by Munerati, has shown that the annual
tendency is conditioned by a single dominant gene, B.
If given a favorable environment €or growth, the vegetative and
G . H. COONS, F. V. OWEN, A N D DEWEY STEWART
reproductive phases of the annual beet can be controlled readily by
length of the photoperiod (Munerati, 1929). The annual beet and the
sugar beet are long-day plants, but for the sugar beet a period of ’herma1 induction at temperatures below 20° C. is a basic requirement Tor
flowering, as has been demonstrated by Chroboczek ( 1934). According
to Stout (1946), temperatures of storage for beet roots above loo C.
tend to reverse the induction process. Thus, roots of a nonbolting type
of sugar beet stored for 43 days at 9.Io C. bolted 98 per cent, whereas
storage at 10.6O C. resulted in only 79.5 per cent bolting after 43 days,
and storage at 13.2O C. showed, after 43 days, only 23 per cent bolting. The thermal induction period required for abundant seed production may be reduced by long photoperiods, as shown by Bell and Bauer
(1942, 1943) and Bell (1946). They also proposed that a combination of
photoperiod and thermal induction may be used as an effective means
of selection for nonbolting types of sugar beets. Stout and Owen (1942)
tested vernalization techniques with sugar beet seeds and found it a
practical measure for hastening reproduction under controlled greenhouse conditions, but probably impractical in the field because warm
conditions in the seed field might reverse the process. Gaskill (1952a)
demonstrated that a combination of continuous light and low temperature exposure to seedlings can be used to bring about two generations
a year of biennial sugar beets. The principle that cool temperature and
photoperiod are complementary influences in hastening sexual maturity in the sugar beet has received wide applications by sugar beet breeders in the United States as a means of reducing the time required in
reaching an objective in breeding programs.
The mode of inheritance of nonbolting in the sugar beet has had several interpretations. No doubt some of the difficulty in reaching a satisfactory explanation has been the failure to recognize the complementary effect of photoperiod and low temperature, which has been designated by Owen et al. (1940) as photothermal induction. These investigations have shown that there is genetic variability with regards to
temperature and photoperiod response in both annual and biennial
beets. They recognized a genetic factor for bolting, designated B’, which
is allelic to the factor B for the annual tendency described by Abegg
(1936). Genetic factors B and B’ are dominant to b associated with the
nonbolting tendency. Linkage, as shown by Abegg ( 1936) and by Owen
and Ryser (1942), between B’ and the color factor R made the genetic
Tetraploid cells in roots of the sugar beets were reported from
Europe as occurring naturalIy, but current production of completely