Crop Improvement: Trends in
Crop Improvement: Trends in
Crop Improvement: Trends in
In
Crop Improvement
R. K. BEHL
N. MAHERCHANDANI
Plant breeding efforts have substantially increased yields mainly of crops grown
under (near) optimum conditions, especially during the last half century. This,
with or without deliberate efforts, has resulted in an increase in adaptability of culli-
vars and stability of grain yields, even in abiotic stress environments. In order to
ensure nutritional security for the growing population, especially in the tropics with
its dwindling and degraded land resources, we constantly need newer crop production
tzchnologies. Even in highly developed temperate agricultural systems, 60 to 80%
of the seasonal variation in crop productivity is attributed to weather fluctuations
(Thompson 1975; Boye 1082). In the tropics where the intensification of crop produ-
ction and extension of crops to more margrnal areas are the order o f the day, iticrcased
research on crop productivity, especially in stress environments is callcd for.
In the short run, crop management may be as efficient in achieving stress arnelio-
ration as is genetic improvement. Because certain plant tram or char~cterrst~cs arc
now recognised as important for abiotic stres; resistance of cropsgeneticii~lpr~veillc~~t
as a m:rns of stress a m : l ~ o r ~ tis~ becoinrn:
~n more fedsible. In the long run Ihrr~lrrs
must be given optcons co~iibin~rig b ~ t hgznet~cand management components of crop
production In dry areas, These options are brietly discussed here.
The extent of adaptation, is limited to the ranges of environmental factors ,e. g.,
minerals or water availability) in which plants have evolved during t h e ~ rexistence.
Wh;n subj;cted to excess (e. g., water logging) or deficiency of these factors (e, g ,
nitrogen deficiency), the plant is regarded to be under stress. For example, use of
high population densities and nitrogenous fertilizer will result in higher demand for
water and other nutrients, as aell as on radiation to sustain maximurn growth rates.
Iicnce definition of 'stress' in agricultural context, is any environmental condition
$hat results in limiting a crop (and plant) for realizing its potential for growth, deve-
lopment, and reproduction. Optimum conditions for field crops can not be fully
extrapolated from results obtained with the single plants in controlled environments,
and vary significantly with developmenfal stages, and the changes in other significant
elements of the environment,
Stress pllysiologists generally choose the most limiting factor for their study, but
application of such studies may not have sufficient relevance to practical crop produ-
ction unless the nature and quantum of interaction with other factors is also elucida-
ted. Most physiologists study the individual processes in isolation, and poorly
quantify fhe relative significance of each process in stress tolerance. Processes such
I:lT~ILNG CKOI' CENO'TYPE3 TO ABIOTIC STRESS 139
a s cell enlargement and cambial activity respond rapidly to water strcss than metabo-
lic processes or photosynthesis. For example, 111 sorghurn, brief stress periods during
differentiation of pistil and stamen primordia can irreversibly reduce seed number
arid yield, but photosynthcs~s may seturn to eorinal following relief from stress
(Eastin el 01. 1983). Althougl~cost- bent fit-rations of adaptatios t o stress by specilic
processes have been investigated by several (Gutscllick 1987) there are only a few
instances where they 11,rve b ~ c napplied in practic;~l crop improvement program.
Relative merits of adaptations also change wit11 agc; for example, ~ncreasedroot-shoot
ratios as a means of drought avoidance are agronomically more sound at seedling
stage than thiit at panicle development or grain tilling stages.
The above stepwise approach is illustrated and discussed in the foliowing sections
using the problem of crop establisbment.
emergence are two important stages susceptible to these factors and may limit crop
production. This phase of crop establishment has to be examined as a good seedling
stand forms the first step towards the realisation of a good crop and high yield under
local conditions. We provide few examples of such a treatment of the problem,
below.
Soil crusting
Soil crusting could limit seedling emergence and establishment. A technique bas
developed in an Alfisol (with 54% coarse sand in the upper 3.1 m) at ICRISAT
Center which crusts naturally when rainfall is followcd by bright sunshine (Soman
et of., 1984).
The soil is disced and rotovated. Ten 1.5 m wide broad bed are prepared
between the sprinkler lines and smothed with a bedshaper. Seeds are sown at a
specific depth (50 mm) with four-cone commercial planter in 2 m long plots. After
sowing, the beds are again smoothed to make the soil surface even. Thirty five mm
of water is then applied using two parallel lines of the sprinklers. The plots are left
to dry for three days, after which the surface becomes firm. The crust in the control
treatment is broken without damaging the plumules of germinating seeds using a
roller with nails mounted on it.
We measure crust strength, bulk density, moisture, and temperature of the soil
and record the number ofseedlings emerged in both the crust and control treatments.
'ihe genotypic variation is assessed on the ability to emerge through crust over
different years.
Soil temperature
Soil surface temperature similarly inhibits seedling emergence. This problem may
occur more widely than soil crusting in the tropics. In India, and West Africa soil
surface temperatures commonly exceed 45°C which inhibit the emergence of seedlings
resulting in poor stand.
The technique for above stuty is described in detail by Soman and Peacock
(1985). Briefly, porous clay pots, 0.3 m in length and 0.1 m in diameter, are filled
with sieved top soil (10-20 cm depth) from a n Alfisol field. The pots are placed in a
tall water tank such that only the top 0.07 m of !he pot is above the water level, The
FITTING CRCP GENOTYPES TO ABIOTIC ~STRGSS 141
soil surface is 20 mm below the top of the pot. The soil in the pots were heated by
lamps fitted to a frame above the tank. The temperature of the soil surface can be
altered by varying the height of the frame above the soil surface, to obtain tempera-
tures of 35,40, 45 and 50°C measured at 20 mrrl depth below the soil surface. The
wet soil column in the pots provided a steady water $upply for the seedlings while
allowing theni to be aflected by the teriiperature of the soil. The design is simple, and
the total cojt of the unit is approximately one tenth of :I cornmcrcial growth charnbei
Seeds are sown at 50 mm depth in the soil in the pots. Soil temperature is
measured every two hours in each pot using copper-conatantan th:rrnoco~ples. Water
is added to th+ tank daily to rnaintain a constant level. Emerging seedlings are
counted 5 and 6 days a f k r sowing (DAS:, and the percentage of emergence calcula-
fed based on the nurilber of herds sown. rhrs ligure is used to differentiate (or
group) genotypes.
Soi! moisture
Lack of sufficient moisture in the soil and inhibits both germination and enler-
gerice thus affecting crop establishment. A technique was developed in the field
where a line-source gradient irrigation (-30 mm, maximum near the sprinkler line;
0.0 m m a t 12 m away from it on either side) was applied to seeds sown in dry soil
(ICRISAT 1987). This set up, under a high evaporative demand condition in the
field simulates a series of (2 to 4) combinat~onsof soil moisture levels and soil tem-
peratures. This technique provides an example of an interactive environment where
more than one factor is involved; the levcl of interaction can be controlled by the
experimenter by choosing the appropriate level of irrigation, type of soil, and other
factors.
The problem solving approach elucidated above in three case studies have
several distinct features. Not only the identification of desired traits that are present
in widely varying geaetic background, btit also the availability of a particular trait
i a combination with others are eaph,lsized. This latter information could be very
useful in fitting genotypes for problematic environments.
It is now appreciated that timing, intensity and probability of drought and its
elfecb o n crops - -both in quantitative and quant~tative terms - can vary widely
depending upon locations and cultural conditions. Hence crop genotypes and cro-
pping s y s t e d suited t o the local conditions have been duly emphasized. Better
analysis o f climatic and edaphic data, and synthesis of knowledge of various crop
production aspects through crop simulation modeling (Huda et al. 1986) have been
increasingly put to practical use.
141 SEETITARMA AND SOMAN
Progress in breeding for drought rcsistance in crop plants is still not satisfdctory,
especially if we exclude drought escape by early maturity in many recently released
cultivars. Whilc some may still consider that breeding for drought resistance is a
waste of resources (Arnon 1980), a more realistic approach is possible if one is
willing to accept the reasons for slow progress in this field. A more pragmatic
approach of combining the traditional fragmented approaches of empirical screening
(numbers game of breeders), and isolated studies on a few selected components of
drought rcsistance believed to contribute to growth under stress (mechanistic app-
roach of of physiologists) is called for (Seetharama et al. 1982). Multilocation and
multiple testing of a large numb-r of selections is the backbone of most breeding
programmes; however, timing and severity of stress over years are seldom sufhciently
uniform to provide reliable, and repeatable test environments with respect to mois-
ture, as well as other princip:lI fclctors essential for determining the degre: of
success. Use of rainout shelters and dry locatio~rswhere one can manipulate stress
profiles are helpful in simulating most probable conditions of target environment.
Drought intensity for selection should be decided based on expression of crltlcal
response studied, and the probabil~tyof that response being useful across dilrerent
growing seasons.
The trade-off between yield potential and drought resistance is well discussed
(Seetharama at al. 1983). As no immunity t o physical stresses is possible, the ques-
tion one would ask is how much less susceptible chosen genotype can be. Only a
few characters, that too t o a apecifled degree, c a n be selectively 'switched on' (in-
duced) under stress, .thus avoiding drainage of plant's energy for this adaptation,
or opportunity t o increase productivity under good condition of growth. Leaf
rolling serves as a n example of such a trait, but'one should note that its significance
is only within certain ranges of stress, w d useful only if conditions are e x ~ e c t e d to
improve later.
FITI'ING CROP GENOTYPES TO ABIOTIC STRESS 143
I. 'The range or set of conditions under which the character is useful (or countcr-
productive) should be known.
2. The character must havc a demonstrated role in drought resistance.
3. Measurement of character must be simple, rapid, cost-effective, and prcfcrnbly
capable of being used during early growth stages.
4. There must be sufficient genotype variation for the character.
5. Assessment of character should preferably involve single (set of) measurements
rather than multiple measurements.
6. Reasonablejsr~fficientknowledge about the inheritance of the character and its
interaction with other characters, and adverse side-effects (or pleiotropic effects)
must exist.
7. The inclusion of the character should fit into jhc over-all plant improvement
strategy for the target environment.
144 SEETHARAMA AND SOMAN
Equal yields could be achieved by different mechanisms, nnd hence the need for
phy:iolkigicalanalysis of yield under stress continues to exist. With a 1 ighly selected.
battery of' tests uniquely suited to the needs of a target environment, the researcher
can ensure that the parents chosen for crossing have sever1 desired characteritics.
Carc must be taken to ensure t h a t the progenies retain useful combinations of traits
in them by manipulating reasonable repeatability of stress pattern and severity of t l ~ e
target locatiou, each year during varietal development,
Though breeding for stress tolerant genotypes with agronomic eliteness is the
ideal solution, most researchers would attemt the crop management appraach in the
short run, especially under highly variable and harsh environments. Tbe biggest
challange is to bring together diverse adaptive features showing complex interactions
with an array of crop growth environments faced with stresses of different kinds
I aryiog in space (plant organs) and time (growth stages). Physiologists must also
note the differences between tests for presence of a trait or response associated with
drought, and field tests that finally prove the stress resistance of a cultivar. Attempts
to integrate many facets of plant activity under stress should be activcly considered
while working on plant parts or processes under stress.
FITTING CROP GENOTYPES TO ABIOTIC STRESS 145
REFERENCES