Introduction

Physiological aspects of photosynthesis and respiration to demonstrate how much remains to be accomplished in sorting out the regulation and function of various components. If there are inefficiencies in these systems and if causes can be identified, they would represent legitimate targets for genetic manipulation. Properties of rubisco, alternative oxidase, and photorespiration process already loom as opportunities for genetic manipulation. Those factors evolved and have survived during millions of years, however, indicating utilities of which we are as yet uncertain. Serious reinvestigation of foliage canopies offers promise for important gains in photosynthetic productivity of crops. First cycle of such research, begun over 50 yr ago, demonstrated importance of  strong advantages of erect leaves in dense canopies and minimum interception by emergent reproductive structures. Those properties are now credited with contributing to yield progress in maize (Fischer and Evans, 1999)

Plant production is driven by photosynthesis. Key elements in the system are (i) the interception of photosynthetically active radiation (PAR, 400-700 nm spectral band), (ii) use of that energy in the reduction of CO₂ and other substrates (photosynthesis), (iii) incorporation of assimilates into new plant structures (biosynthesis and growth), and (iv) maintenance of plant as living unit. Achieving high yield is conceptually simple-maximize the extent and duration of radiation interception; use captured energy in efficient photosynthesis; partition new assimilates in ways that provide optimal proportions of leaf, stem, root, and reproductive structures; and maintain those at minimum cost.

Crop yield comprises only a portion of biomass that accumulates over a crop cycle. Effective root and canopy systems (including stem structure for foliage display), for example, generally must be established before onset of reproductive effort. In addition, cost of maintenance increases as vegetative biomass accumulates during  season. Because crops are at mercy of spatial and temporal variations in weather, plant spacing, and supplies of water and nutrients, and in occurrence of pests and disease, flexibility in morphogenesis and acclimation of physiological systems is a key requirement for achieving high and stable performance. Whether biological efficiency of these processes has, or might be, improved through breeding are important questions.

Biomass and Other Morphological Traits

Just as impact of Green Revolution can be attributed mostly to improved partitioning of  products of photosynthesis to grain yield, progress in yield is strongly associated with improved HI. Morphological traits associated with increased yield potential include grain number and HI Even if HI could be raised to 60% from its current maximum value (50%), it implies that yields could only be increased by a further 20% using HI as a selection criterion, unless total crop biomass is also raised. Furthermore, improved partitioning by greater reduction in plant height is unlikely since research suggests that optimal plant heights have already been achieved. Some studies have shown increased biomass to be associated with yield increases.

Photosynthesis and Related Traits

By definition, improved yield cannot be attributed to better overall radiation use efficiency (RUE) in cases where total biomass has not been improved. (RUE in a crop context represents ratio of total energy present in crop's biomass to that of solar energy incident on crop across its growth cycle; mass is not, is perhaps explained by lower preanthesis vegetative growth rates observed for more modern cultivars in this study. Expression of higher photosynthetic rate in absence of significant changes in biomass could be a pleiotropic effect of improved partitioning to yield driven by high demand for assimilates during grain filling. Canopy temperature depression is a direct function of evapotranspiration rate, which itself is determined largely by stomatal conductance. These traits could also be pleiotropic effects of genetic variability among lines for a number of physiological and metabolic processes including sink strength, photosynthetic rate, vascular capacity, and hormonal signals.

Improvement in Nutrient Use Efficiency

Genetic gains in N use efficiency (NUE), defined as grain yield per unit of N available to plant (Ortiz-Monasterio et al., 1997). While NUE almost doubled with introduction of height reduction (Rht) genes in early 1960s, progress since Green Revolution has continued at a lower rate in parallel with more modest improvements in partitioning to yield. Improvement in NUE has been associated with improvements in both total N uptake, as well as efficiency of utilization in terms of grain yield. This study also revealed interesting and controversial fact that Green Revolution varieties demonstrated genetic gains in yield even under severely N-limited conditions, that is 2 to 2.5 t ha^-1 yield levels. This trend has continued since 1966 with varieties of mid 1980s yielding more than 3 t ha^-1 under same conditions. Work in Argentina has demonstrated improvement in both N and P use efficiency in modern varieties of wheat.

Adaptation to Density

Idea that higher yield potential could be achieved by designing a plant type that is well adapted to commercial practice of sowing high density monocultures was introduced 30 yr ago by Donald (1968). Improvement in yield potential would appear to be more a function of improved adaptation to canopy microenvironment, rather than macroenvironmental factors such as climate. Several studies have shown that selection for yield potential in early generations can be enhanced by reducing interplant competition between genotypes in bread wheat (Lungu et al., 1987), and durum wheat (T. turgidum L.) (Mitchell et al., 1982). They do suggest that conventional breeding approaches may lose yield potential by selecting against it.

 Improving the Ideotype

Relatively few systematic studies, such as those described above, have been conducted to examine physiological basis of yield improvement in post–Green Revolution era. Perhaps for this reason, further investment is made in understanding which plant traits could be optimized to further improve yield. Many traits have been suggested in  literature as having potential to raise yield, but very few have been examined in a systematic way for their potential to increase genetic gains when used as selection criteria. They have not generally been introgressed into high-yielding backgrounds, and little if any work has been conducted to assess potential complementarity between many morphological and physiological traits which have potential to improve the crop ideotype.

Source and Sink

It is widely believed that yield gains are most likely to be achieved by simultaneously increasing both source (photosynthetic rate) and sink (partitioning to grain) strengths. While most experiments indicate that yield is primarily limited by growth factors prior to anthesis, source capacity may have become more limited in modern cultivars. For example, experiments on a historic series of spring wheats from Russia indicated that, while sink capacity has been improved in post–Green Revolution period, improvement has also resulted in modern lines that are now more source limited than those in previous eras. Using field-grown plots, spikelets from one side of the spike were completely removed at flowering. Potential yield (i.e., in the absence of source limitation during grain filling) was calculated from doubled grain weight of semidegrained spikes. Data showed that, while potential yield per spike had been improved (i.e., the doubled grain weight of semidegrained spikes was higher in more modern lines), extent to which  potential was realized (i.e., potential vs. actual spike yield) had declined in modern lines.

Reproductive stages of development, from initiation of floral development to anthesis, are pivotal in determining yield potential, and especially rapid spike-growth phase which has a duration of 25 d in irrigated spring wheat in northwest Mexico and Argentina. During this period, final grain number is determined; a major factor determining subsequent partitioning of assimilates to yield, as well as heavily influencing assimilation rate of photosynthetic apparatus during grain filling. Duration of spike growth relative to other phenological stages shows genetic variation. This is associated with sensitivities to photoperiod, vernalization, and developmental rate independent of these stimuli (i.e., earliness per se).Fischer (1975, 1985) established critical nature of the rapid spike-growth phase in determining yield. Based on this, it has been suggested that possibility exists of improving final grain number and yield potential by manipulating genes associated with sensitivity to photoperiod (Ppd) and vernalization (Vrn), as well as earliness per se (Slafer et al., 1996). Hypothesis is based on the idea that by increasing the partitioning of assimilates to spike growth, and therefore spike biomass, potential floret survival will be increased and hence yield potential raised (Bingham, 1969). Experiments in which different radiation regimes were compared during this critical phase are consistent with hypothesis (Fischer, 1985; Abbate et al., 1997). Recently, duration of rapid spike growth has been successfully manipulated using photoperiod, revealing a strong relationship between its duration and number of fertile florets/spike (Miralles and Richards, 1999). By maintaining plants at a relatively short photoperiod during this growth phase, number of days from terminal spikelet to heading was increased from 50 to 70 d, with 13- and 9-h photoperiods, respectively, while number of fertile florets per spike increased from 77 to 108.

From a practical point of view, breeders have tried to modify sink capacity of wheat by modifying spike morphology. A good example of this approach was reported by Dencic (1994) who crossed genotypes using single, back, and top crossing, and desirable lines selected using a pedigree approach. with branched tetrastichon (two spikelets per node of  rachis) with high-yielding lines that contained other desirable traits such as high yield, disease resistance, and quality. After 10 yr of breeding and selection, 229 lines with desirable characteristics were yield tested, of which yield superior to the standard checks of the four lines was 13% (i.e., 1t ha^-1) higher than standards (Jugoslavija and Skopljanka), and following morphological traits were improved over standards: spike length (16%), spikelets per spike (10%), grains per spikelet (9%), grains per square meter (18%). This progress in yield was achieved in spite of fact that tetrastichon donor lines had problems of empty florets or shriveled grain with very low kernel weight.  

Source and Sink: seed Size

Genetic progress in yield potential is strongly associated with increases in grain number while weight per grain has generally declined. Nonetheless, some studies have shown increased kernel weight has contributed to improved yield potential in irrigated wheat. Understanding the physiological and genetic basis of potential kernel size remains an obvious challenge for yield improvement. An important question is whether grain weight potential can be increased independently of increases in grain number. Simplistically, it can be argued that this inverse relationship is a necessary  tradeoff when more grains are competing for limited assimilates during grain filling. However, studies that have examined the relationship between kernel sizes and number at different spike positions using lines from different eras conclude that size of kernels at low potential weight spikelet positions are independent of kernel number or year of release. It is suggested that grain weight is colimited by both source and sinks, such that grain weight potential would be most likely determined during spike growth, resulting in different potential sizes at different spike positions. Realization of potential would be determined by assimilate availability during grain filling.

Very recent research with synthetic hexaploid wheat, which tends to have larger kernel weights than conventional cultivars, has demonstrated significant increases in grain weight when assimilate supply was increased by partial degraining treatments during rapid spike growth. Effect was greatest at grain positions generally showing lower grain weight. No effect on grain size was apparent when degraining occurred a week after anthesis. Data confirmed physiological potential for increasing kernel weight at distal spikelet positions by as much as 16%, strongly endorsing objective of breeders to raise yields through increasing grain weight potential. On basis of these data, it is proposed that extending duration of rapid spike-growth phase may increase yield potential not only by increasing potential grain number, but also by increasing grain weight potential through extending window of opportunity for individual kernel development (Calderini and Reynolds, 2000).

Photosynthetic Production 

Whether a canopy (amount of leaf area, LAI, and its manner of display) is optimal for photosynthesis in a particular environment is reciprocally linked with development and properties of individual leaves, including their longevity. According to a leaf's position in canopy, variations occur in the components of its photosynthetic system, its acclimation to changing conditions, and its protection from excess photon flux density (PFD).

Leaf Components

Solar-energy-capturing apparatus of higher plants is located in thylakoid membranes of chloroplasts. As summarized in Fig.1 , it consists of light-harvesting antennae complexes composed of carotenoids and chlorophylls a and b connected to Photosystem (PS) I and II reaction centers, a cytochrome b₆f complex, and ATP synthase. The b₆f complex transfers electrons from PSII, the water-oxidizing center, to PSI leading to NADP⁺ reduction. The proton gradient that develops across thylakoid between an interior lumen and the exterior stroma is employed by ATP synthase (coupling factor complex, CF_o-CF₁) to produce ATP from ADP. Photosynthetic reductive pentose phosphate cycle ("dark reactions" involving CO₂ assimilation) is found in stromal solution. Key enzyme, rubisco, catalyzes both oxygenation and carboxylation of ribulose-1,5-bisphosphate (RuP₂). Rubisco's activity as an oxygenase, initial step in process of photorespiration, increases as  ratio [O₂]/[CO₂] at enzyme and/or temperature increase. Also in stroma are enzyme systems that manufacture and repair chloroplast constituents, reduce nitrite and sulfite, and synthesize starch. Q₁₀ of reductive cycle is 2 and low temperature limits  CO₂ reduction unless the capacity is increased through increases in enzyme concentrations.

All chloroplasts of C₃ plants contain full set of enzymes for CO₂ assimilation. In C₄ crop plants, rubisco and most of C reduction cycle occur only in chloroplasts of bundle sheath cells. C₄ mesophyll cells, lack rubisco but rely on an important cytosolic enzyme, PEP carboxylase, to assimilate CO₂. In C₄ plants of NADP-malic enzyme type (maize, and sugarcane), PEP carboxylase fixes CO₂ into oxaloacetate, which is then reduced to malate. Malate is transferred to bundle sheath cells where it is decarboxylated to pyruvate, thus concentrating dilute supply of CO₂ around rubisco and greatly reducing photorespiration.  pyruvate returns to mesophyll chloroplasts where it is converted to PEP through conversion of ATP to AMP. The cycle is completed when the PEP returns to mesophyll cytosol. An additional complexity in C₄ plants is that 3-phosphoglycerate (PGA) is also exported from bundle sheath cells and is reduced in mesophyll chloroplasts, thus utilizing reducing power that is available there.

Leaf Angle

The erectophile leaf canopy has been proposed as a trait that could increase crop yield potential by improving light use efficiency in high radiation environments. A number of studies support the hypothesis. It has been associated with a 4% yield advantage in wheat isolines in U. K. More erect leaf posture was associated with higher grain number and higher stomatal conductance. In barley two varieties contrasting in leaf angle were compared for photosynthetic rate at different depths of canopy. The erect leaf variety showed a more even distribution of photosynthetic rate throughout the canopy, as well as higher rates of stem photosynthesis (Angus et al., 1972).  

Stem Reserves and Green Leaf Area Duration

There are a number of additional physiological traits that have implications for yield potential and are related to increasing assimilate availability (i.e., source). One is ability to reach full ground cover as early as possible after emergence to maximize interception of radiation (Richards, 1996). Another is remobilization of soluble carbohydrates (stem reserves) during grain filling. A third is ability to maintain green leaf area duration ("stay-green") throughout grain filling. Direct evidence for contribution of these traits to high yield potential is lacking. Stem reserves apparently make a greater contribution to performance in relative low-yielding lines where contrasting lines have been examined (Austin et al., 1980b). It is been suggested that use of stem reserves and stay-green may be mutually exclusive, since loss of chlorophyll and stem reserve mobilization seem to be consequences of plant senescence. A greater understanding of genetics of these traits is called for to establish potential for breaking such linkage. As yield potential is raised by improving reproductive sinks, extra assimilates gained by increasing early ground cover could contribute to increased stem reserves and be tapped at later reproductive stages to enhance potential kernel number and size.

 Acclimation
Crop plants are exposed to widely fluctuating conditions of light and temperature, and supplies of water and nutrients, and have evolved with a leaf-level photosynthetic apparatus that is highly flexible in structure and activity. Depending upon environment, leaves develop with different numbers and sizes of cells, different numbers of chloroplasts per cell, and with variations in amounts and proportions of thylakoid and carbon-reduction-cycle components. Changes in these factors seem more related to photosynthetic activity per unit C and N invested in leaf structure than per unit leaf area. Acclimative changes depend on light environment and position in the canopy and continue on a time scale of days to weeks throughout the life of a leaf.

Acclimation to light is proportional to mean daily irradiance of leaf rather than to peak irradiance (Chabot et al., 1979). This ability is important because new leaves generally emerge at top of a crop in full sun and later are submerged into shade of canopy as other leaves develop above them. C₃ leaves in full sun typically have more of their leaf N involved in electron transport and carbon reduction and less in light harvesting (and fewer grana stacks) than is case for shade leaves. These properties also vary with depth within  leaf from its sunlit surface.  In  Evans' (1993) study of alfalfa canopies, soluble proteins (e.g., rubisco) declined more with depth in the canopy than did thylakoid proteins. In addition, chlorophyll a/b decreased with increasing depth in canopy, reflecting a decline in reaction centers relative to light-harvesting antennae. These changes resulted in a decline in chloroplast N/chlorophyll in a way that maintained photosynthetic capacity per unit N, a key trait for optimal distribution of N within the canopy.Leaf adjustments to limiting supplies of N are especially important, involving changes in number and size of new leaves as well as in proportions of thylakoid and carbon-reduction-cycle components with depth in canopy. Optimal distribution of N among leaves within a canopy is important and has received attention in recent years. Problem also involves canopy architecture, solar track, sky condition, and time remaining in  season . For young crops with small leaf area, increasing leaf area for greater radiation interception provides more benefit than increasing photosynthetic capacity of existing leaves (through greater N content per unit leaf area).

Protection

Leaves exposed to full sun encounter challenges in balancing electron transport with their capacity for carbon reduction and/or the supply of CO₂. For crop plants well supplied with water and nutrients, present low atmospheric CO₂ concentration (near 360 µmol mol^-1 air) is a greater problem than reduction capacity. Daily photosynthesis of such crops can be equivalent to all of the CO₂ in 100 m of air. Although atmospheric turbulence extends mixing to much greater heights and ensures that concentrations within canopy generally remain above 250 µmol CO₂ mol^-1 air, CO₂ concentration within leaves (C_i) limits maximum photosynthesis rates in C₃ crops. Rubisco has low affinity for CO₂.As light flux increases and C becomes limiting to carbon reduction cycle, light-response curve of CO₂ uptake departs from a linear increase (minimum number of quanta required per CO₂ reduced) and plateaus at a maximum, light-saturated rate (A_max). Absorption of light energy continues but excitation energy cannot be dissipated in usual way because ADP and NADP⁺ substrates are not available to accept electrons; i.e.,  reaction centers are "closed". Limitations to CO₂ supply because of stomatal closure (e.g., drought stress) or low capacity for CO₂ reduction (e.g., N deficiency or low temperature) increase the likelihood that PFD will be in excess of rate of use.

Closure places reaction centers at risk of short- or long-term photodamage. D1 protein in core of PSII is particularly vulnerable given large redox potential (about 1.17 eV) developed there and presence of singlet oxygen (¹O₂). As a result, this protein turns over very rapidly (once or twice per hour) in illuminated leaves. Repair is expensive:  protein must be ejected from PSII center, rebuilt, and reinserted. Such photodamage increases the minimum quantum requirement and reduces A_max. With small leaf area, canopy photosynthesis is maximum with spreading leaves despite high levels of light saturation & potential for photodamage. With more leaf area, canopy architectures that result in less irradiance per unit leaf area (e.g., erect leaves) are effective in limiting these problems. Some crops, notably legumes, alter leaf display during day in ways that reduce light absorption in high light and, in all crops, small portions of the excess energy may be expended in nitrite or sulfite reduction or in leaf maintenance.

In C₃ leaves, reaction centers may be kept open and excess energy dissipated by utilization of NADPH and ATP in photorespiration. In this, rubisco catalyzes condensation of O₂ with RuP₂, rather than CO₂, and glycolate (a potentially toxic compound) and PGA are generated. In terrestrial plants, glycolate is metabolized in peroxisomes and mitochondria (where CO₂ is released), producing glycerate. Glycerate is taken up by chloroplasts where it and PGA participate in regeneration of the RuP₂ substrate used by rubisco. In this way, three of every four C atoms are recycled allowing another cycle of O₂ or CO₂ reduction unless serine (which is involved in a mitochondrion-to-peroxisome transfer) is "drained" off to protein synthesis. Unfortunately, photorespiration increases with temperature because of decreased solubility of CO₂ in the stroma and decreased affinity of rubisco for CO₂ relative to O₂ (i.e., decreased specificity for CO₂). Even in dim light, when there is no need for protection, photorespiration increases the quantum requirement for CO₂ reduction (qr) of C₃ plants from a minimum value near 11 mol photons mol^-1 CO₂ at low temperature (<15°C) to about 25 at temperatures near 35°C. The slow rate of photorespiration at low temperature means that it cannot offer much protection against over-excitation under those conditions. The extent rubisco acts as carboxylase or oxygenase depends on relative concentrations of CO₂ and O₂ presented to the enzyme. Under elevated atmospheric CO₂,  CO₂  concentration within C₃ leaves increases and oxygenase activity is suppressed. C₄ plants suppress photorespiration in ambient air by the same principle—the CO₂-concentrating action of their mesophyll cells keeps rubisco (in bundle sheath cells) well supplied with CO₂. Oxygenase activity is then only a small percentage (2–6%) of the net CO₂ flux.

Another form of protection based in Mehler and ascorbate-peroxidase reactions, which operate well in water-stressed plants, as perhaps more important than photorespiration for protection of C₃ leaves. This involves formation at PSII of peroxide (H₂O₂), which is then reduced by ascorbate peroxidase. The oxidized ascorbate is regenerated by a reductase with expenditure of NADPH. These reactions keep both reaction centers open without a net exchange of O₂. Because study of this pathway depends on examining oxygen isotope discrimination, which is difficult, little is yet known about its protective role.Another system of protection exists in interconversion of carotenoid pigments found in association with chlorophyll in light-harvesting complexes. With high PFD, pH changes in thylakoid lumen induce conversion of violaxanthin to zeaxanthin, which can accept resonance energy from chlorophyll. When reaction centers close, excitation energy passes from chlorophyll to zeaxanthin and converted to thermal energy (increasing leaf temperature) before it reaches PSII reaction center. Leaf thermal energy is dissipated to  environment through convection, transpiration, and long-wave(IR) radiation emission. Such "nonphotochemical quenching" of excited chlorophyll is assayed easily through measurements of variable chlorophyll fluorescence. In dim light, zeaxanthin recycles to violaxanthin, which cannot intercept energy from chlorophyll. nonphotochemical quenching as a major mechanism for protection of PSII from excess PFD.

Little is yet known about the need for photoprotection by crops. Young crops with spreading leaves are at the most risk. Within canopies, leaves are displayed at angles to the sun's rays that greatly limit the amounts of excess PFD they absorb. In addition, most crop plants are "sun" plants with a large capacity for photosynthesis and less need for photoprotection than plants having less capacity. Sun plants generally have high stomatalconductance that helps maintain C_i and allows sunlit leaves to dissipate heat by transpiring freely. Yield advance in CIMMYT wheat lines, has been associated with increased stomatal conductance.  Osmond and Grace(1995) viewed photorespiration and ascorbate-peroxidase reactions as main protection. Andrews and Baker(1997) suggested photorespiration buffers against imbalances in PFD absorption by PSI and PSII. Long (1998), saw little merit in photorespiration and view leaf acclimation, canopy architecture and the xanthophyll cycle as the principal means of protection. Suggested that photorespiration persists only because evolution has reached a "barrier" for improvements in rubisco's affinity for CO₂.  RUBP in some Rhodophyta has greater affinity for CO₂ & might be a source of genetic material.

Radiation-Use Efficiency

Crop growth rate and yield are functions of canopy photosynthesis and they generally correlate poorly if at all with maximum photosynthesis rates of individual leaves. Given the oblique display and mutual shading of leaves within canopies, few leaves exposed to PFD sufficient to achieve A_max. Other reasons for discrepancy can be found in acclimation to radiation level, temperature and stress, and amount of standing crop (and thus amount of maintenance respiration). As a result, crop physiologists have sought other measures that would relate yield and canopy photosynthesis. Light-conversion efficiency ( radiation-use efficiency, RUE) has received most attention. RUE is measured and reported in various units, e.g., g new biomass produced MJ^-1 radiation intercepted or absorbed by leaves. A useful feature of RUE is that experimental values can be compared with estimates of potential rates of dry matter production that might be possible by a canopy of well-acclimated leaves. Potential RUE would be attained with all leaves exposed to only moderate PFD (little or no light saturation and, thus, minimum qr). If that crop at same time intercepted most of incoming radiation, its rate of biomass production per unit land area would also be maximized. We recently reexamined potential RUE for C₃ plants in light of modern understanding of quantum requirements and C losses in respiration. Calculations were done for a C₃ crop with 1000 g m^-2 standing biomass (midseason amount for a crop producing 10 Mg grain ha^-1), moderate maintenance respiration, and a growth yield in biosynthesis (Y_G) of 0.72 g new biomass g^-1 assimilate consumed. Calculated RUE varied from 4.1 g MJ^-1 solar radiation absorbed for qr = 10 to only 1.1 at qr = 30. This wide range of qr embraced the large increase in photorespiration that occurs in C₃ leaves with increasing temperature. 

Small amount of photorespiration in C₄ leaves, only a small range of qr values around 16 is considered. minimum possible qr for C₄ plants is> for C₃ species because ATP is expended in malate production, a cost that is increased by leakage of concentrated CO₂ from bundle sheath leading to "over-cycling" of malate minimum values for several C₄ species in a narrow range around qr = 15.4 mol photons mol^-1 CO₂. That high efficiency seems to require participation of  thylakoid Q-cycle (enhanced ATP production) to offset cost of malate over-cycling. With qr = 16, 1400 g biomass m^-2, and Y_G = 0.74, Calculated RUE values are sensitive not only to variation in qr (which varies with radiation level) but also to variation in maintenance respiration, which depends on temperature, maintenance coefficient, and size of standing crop. For system with qr = 16, variations in maintenance lead to following: Maintenance (mmol CH₂O):0, 20, 40, 60 RUE(g MJ^-1 solar): 2.7, 2.4, 2.0, 1.5 respectively. where CH₂O represents carbohydrate with MW = 30 and maintenance is per megajoule solar radiation intercepted. RUE is also sensitive to variation in Y_G, which declines as concentrations of protein and/or lipid increase; i.e.,amount of assimilate used per gram new biomass increases. Thus, largest RUE occurs for crops with a large content of carbohydrates (cellulosic material, starch, sugars). Where comparisons are to be made between crops differing in composition, RUE should be expressed either in glucose equivalents required in synthesis or in energy content of  biomass. At qr = 16, RUE of 2.3 g biomass MJ^-1 solar radiation intercepted can be expressed as equivalent to 104/6 = 17.3 mmol glucose MJ^-1. Biomass of this composition would have a heat of combustion near 17.6 kJ g^-1 corresponding to over 4% of absorbed solar radiation (and near 9% of absorbed PAR).

RUE values for C₄ crops generally exceed those of C₃ crops, and, for maize, smaller values were found during cool weather (Andrade et al., 1993) and under stress, such as high evaporative demand (Kiniry, 1999). In one of the few successful comparisons of cultivars, Tollenaar and Aguilera (1992) found that postanthesis RUE of a maize hybrid released in 1988 exceeded that of a 1959 hybrid. The difference was related to the "stay-green" trait possessed by the 1988 hybrid. Contrary to the hopes of crop physiologists, the variable nature of RUE and difficulties in its measurement prevent it from serving, as a sensitive measure for exploring fine structure of photosynthetic systems. Uncertainty about what constitutes a "reliable" value of RUE also raises serious questions about its wide use as a driving variable for crop simulation models. Perhaps the RUE era should be closed. It persists, however, because the alternatives for study of canopy photosynthesis are difficult or expensive and because crop modelers resist replacing it by submodels that simulate canopy photosynthesis.

Roots

 Work with Rht isolines of wheat in Argentina indicated that shorter lines had a higher investment in root length and dry weight at anthesis than the tall ones, in the top 30 cm of soil (Miralles et al., 1997). The proportion of total respiration resulting from root activity has apparently declined in more modern Russian spring wheat (Koshkin and Tararina, 1989). While progress in measuring and understanding root anatomy and its relationship to yield is likely to be slow, the discovery of root signals should raise a note of caution for breeders wishing to increase yield in irrigated wheat. Biochemical signals, elicited by reduced soil water potential, have been shown to cause reduced stomatal conductance well in advance of leaf water deficit (Davies and Zhang, 1991). The trait probably evolved to increase the likelihood of completing a genotype's life cycle under unpredictably dry conditions. However, researchs need to test hypothesis that root signalling may limit productivity in irrigated crops by reducing stomatal conductance (and therefore the potential for CO₂ assimilation) as roots detect progressively drying soils prior to scheduled irrigations.

Respiration and Biosynthesis

The Respiratory System

Respiration in higher plants is commonly viewed as a sequence of enzymic steps. With hexose as a generic substrate, flow of carbon can be traced through glycolytic pathway (found incytosol and plastids) to tricarboxylic acid (TCA) cycle in matrix solution of mitochondria. Mitochondria, like chloroplasts, are enclosed by an outer membrane that encompasses a convoluted inner membrane, inside which is matrix. An interesting point is that pyrophosphate (PP_i), rather than ATP, can be used in phosphorylating hexose at the start of glycolytic scheme and this may be an essential process in some plants  in mitochondrial electron transport chain. Protons may enter matrix through membrane leaks, but F_o-F₁ ATP synthase (similar to the CF_o-CF₁ATP synthase in is main route of proton entry. Apparently, one ADP is  F₀-F₁ ATP synthase. ADP required for ATP formation enters mitochondrial matrix only as ATP exits matrix through an antiporter. A symporter couples transport of P_i and H⁺ into matrix phosphorylated when three protons passthrough.

Products of glycolysis (pyruvate and malate) can be completely oxidized in TCA cycle with production of ATP and reduced nucleotides. The great bulk of ATP production then occurs through oxidation of nucleotides by protein complexes located in mitochondrial inner membranes. If all reducing agents produced by glycolysis and TCA cycle are employed in ATP production, a total of about 30 mol ATP mol^-1 glucose can be produced (This amount is<36 mol ATP commonly quoted in older biochemistry texts.) About half of free energy of hexose is captured in ATP when a hexose molecule is completely oxidized in respiration; rest is lost as heat. Most of "retained" energy is also lost as heat when ATP is subsequently used (hydrolyzed).

Alternative Oxidase

The alternative oxidase found in mitochondria deserves special comment. Electrons that pass to this oxidase bypass two sites (complexes III and IV) of proton translocation with the result that less ATP is formed. There is no evidence that it has a significant impact on crop performance.

Promising New Approaches to Accelerate Yield Gains

To date, breeding programs worldwide have achieved significant genetic gains in yield potential. Nonetheless, there is consensus among breeders as well as physiologists that while the contribution from physiology has been modest, its contribution to breeding is expected to be larger in the next 20 yr (Jackson et al., 1996). There are perhaps two main reasons for this. One is based on need.This means that current trends in the improvement of genetic yield potential are too low to keep pace with future demand. The second is probably based on justifiable optimism. Several studies suggest that some new selection technologies have real potential to complement conventional breeding programs in the areas of biotechnology (Tanskley and Nelson, 1996) and physiology (Richards et al., 1996; Fischer et al., 1998; Reynolds et al., 1998). At a recent consultation on raising yield potential, successful breeders suggested a number of strategies for increasing genetic gains in yield potential. While the pivotal role of recombining elite germplasm was recognized, it was also agreed that significant jumps in yield potential will almost certainly require introgression of new genes from diverse sources. This will permit evaluation of new yield-determining genes in different backgrounds. Development of improved early-generation selection criteria was also among the recommendations highlighted by the group.

 Using Physiological Tools to Complement Empirical Selection

While morphological traits associated with yield, such as grain number and HI, can be used in visual selection of breeding lines, neither trait is reliably expressed in small plots, or at low density in early generations. However, there is now good evidence that certain physiological traits have potential for improving selection efficiency. For example, under warm, irrigated conditions, CTD measured on yield trials in Mexico was significantly associated with yield variation in situ, as well as with the same lines grown at a number of international testing sites (Reynolds et al., 1994b).An integrated CTD value can be measured almost instantaneously using an infrared (IR) thermometer on scores of plants in a small breeding plot, thus reducing error normally associated with traits measured on individual plants. Leaf temperatures are depressed below air temperature when water evaporates from their surface. The trait is affected directly by stomatal conductance, and therefore indirectly by many physiological processes, including vascular transport of water as well as C fixation and other metabolic activity. As such, CTD is a good indicator of a genotype's fitness in a given environment. Canopy temperature depression measured during grain filling also seems to be influenced by the ability of a genotype to partition assimilates to yield. This is indicated by the fact that CTD frequently shows a better association with yield and grain number than it does with total aboveground biomass. Investigations into methodology (Amani et al., 1996) have shown that CTD was best associated with performance when measured at higher vapor pressure deficits (i.e., on warm, sunny afternoons). Irrigation status was not a confounding factor within the normal frequencies of water application.

The possibility of combining selection for both CTD (on bulks) and stomatal conductance (on individual plants) is another interesting possibility. In fact, work which evaluated a number of indirect early generation (F₂) selection criteria for yield demonstrated the value of stomatal (i.e., leaf) conductance as a predictor of yield more than 20 yr ago (Wall, 1977). In more recent work at CIMMYT (Gutierrez et al., 2000), stomatal conductance was measured on individual plants in F_2:5 bulks and showed significant phenotypic and genetic correlation with yield of F_5:7 lines.

Measuring Canopy Temperature Depression with Aerial Infrared Imagery

In terms of selection technologies on horizon, one may be use of aerial IR imagery to significantly increase efficiency of conventional selection for yield. Work conducted recently in northwestern Mexico showed that aerial IR images had sufficient resolution to detect CTD differences on relatively small yield plots (1.6 m wide). Data were collected using an IR radiation sensor mounted on a light aircraft that was flown at a height of 800 m above plots. Information from image was subsequently digitized to provide individual plot canopy temperatures to with accuracy of 0.1°C. Data of plot temperatures showed significant correlation with final grain yield for random derived recombinant inbred lines as well as advanced breeding lines, and a set of elite varieties. Data from IR imagery were compared with a spot reading taken a few days earlier with a handheld IR thermometer under clear and sunny conditions. Considering that conditions were suboptimal at time of IR imagery measurement (intermittent cloud cover introduced significant error into measurements), correlation with yield compared quite favorably with that of data from handheld IR thermometers^. For both methodologies, correlation with yield was higher with random derived lines than with advanced or elite lines that had already been screened for performance. (This is to be expected since nurseries would be skewed in favor of  physiologically superior lines after selecting for yield.) results validated potential of aerial IR imagery as means of screening thousands, of breeding plots in a few hours for CTD, and hence for their genetic yield potential.

Spectral Reflectance

Another promising technology is spectral reflectance, which can be used to estimate a range of physiological characteristics including plant water status, leaf area index, chlorophyll content, and absorbed PAR (Araus, 1996). The technique is based on the principle that certain crop characteristics are associated with the absorption of very specific wavelengths of electromagnetic radiation (e.g., water absorbs energy at 970 nm). Solar radiation reflected by the crop is measured, and calibrated against light reflected from a white surface. Different coefficients can be calculated from specific bands of the crop's absorption spectrum, giving a semiquantitative estimate (or index) of a number of crop characteristics. In preliminary experiments, the indexes NDVI (normalized difference vegetation index), WI (water index), SR (simple ratio), and SIPI (structural independent pigment index) all showed significant correlation with yield, biomass, and leaf area index. The measurements were made during grain filling on 25 advanced lines selected for diverse morphology, with yields ranging from 5 to 9 t ha^-1 in an irrigated spring wheat environment in northwestern Mexico (Reynolds et al., 1999). Performance was best correlated with NDVI and was a little higher  for biomass than for yield . For further explanation of these and other spectral reflectance indexes, the reader is referred to Araus et al. (1996, 1999). Spectral reflectance devices record the intensity of reflected radiation from 400 to 1200 nm, producing a unique SR signature for a genotype. Given the wealth of information present in a genotype's SR signature, the possibility exists of using parallel processing to search out characteristic reflectancepatterns associated with performance, to complement some of the existing indexes mentioned above.

Conclusions

With respect to understanding the physiological basis of yield improvement perhaps the most interesting developments are: (i) the direct manipulation of spike growth duration which resulted in substantially increased spike fertility and (ii) the observation that yield is not necessarily associated with improved partitioning of assimilates to spike growth challenging the hypothesis that a simple increase in biomass of the juvenile spike would necessarily result in yield gains. If we are to understand the physiology of the wheat crop, perhaps we also need to consider the evolution of the wheat species. Wheat evolved and was subsequently selected under relatively low-yielding conditions. Physiological traits conferring survival were strongly favored for most of the crop'sevolution. Blum (1996) suggests that subtle expression of "conservative" traits may still hold back yield potential in modern wheat. As a result, any degree of competition for assimilates from alternate sinks, for example root and tiller growth, osmotic adjustment, or carbohydrate reserves in stems, may reduce partitioning of assimilates to grain yield. Obviously, these types of traits are not advantageous when a single genotype is grown at highdensity with ample water and nutrients. Perhaps another example of a conservative trait is root signalling, which can cause reduced stomatal conductance in response to soil water deficits that are not actually limiting potential evapotranspiration (Davies and Zhang, 1991). If in response to an environmental cue, the water relations of a plant can be regulated, it is conceivable that subtle stresses at critical growth stages may also lead to conservative responses in reproductive growth. Nitrogen availability is another critical factor determining growth potential, and spike fertility apparently can respond directly to N supply to the spike (Abbate et al., 1995). Whether or not significant genetic diversity exists for sensitivity to subtle stresses needs investigation, as does the potential role of these conservative traits in determining yield potential in modern wheat varieties.

REFERENCES

• Abbate P.E., Andrade F.H., Culot J.P. The effects of radiation and nitrogen on number of grains in wheat. J. Agric. Sci. (Cambridge) 1995;14:351-360.
• Abbate P.E., Andrade F.H., Culot J.P., Bindraban P.S. Grain yield in wheat: Effects of radiation during spike growth period. Field Crops Res. 1997;54:245-257.
• Abbate P.E., Andrade F.H., Lázaro L., Briffi J.H., Berardocco H.G., Inza V.H., Marturano F. Grain yield increase in recent Argentine wheat cultivars. Crop Sci. 1998;38:1203-1209.[Abstract/Free Full Text]
• Amani I., Fischer R.A., Reynolds M.P. Canopy temperature depression association with yield of irrigated spring wheat cultivars in hot climate. J. Agron. Crop Sci. 1996;176:119-129.
• Angus J.F., Jones R., Wilson J.H. A comparison of barley cultivars with different leaf inclinations. Aust. J. Agric. Res. 1972;23:945-957.