Water-soluble Carbohydrate Reserves Research Articles

Pasture resilience reflects differences in root and shoot responses to defoliation, and water and nitrogen deficits

The yield of a pasture is directly proportional to the amount of light plants intercept and allocate to different organs. When plants are carbon (C) limited, due to defoliation, they allocate more C preferentially to shoots to restore leaf area. In contrast, water and nitrogen (N) limitations lead to a greater allocation of C to roots. Changes in the root:shoot ratio therefore reflect changes in C and N partitioning and indicate their relative priority. A major factor that influences plant responses to stress is their ability to store and remobilise reserves to restore leaf area. Species with tap roots, like lucerne, have a large potential C and N storage capacity that is utilised seasonally for storage and remobilisation. This has been used to develop seasonally based grazing management rules. Similarly, recommendations to graze perennial ryegrass at the 2- or 3-leaf stage are based on the balance between maximizing growth rates and the need to replenish water-soluble carbohydrate reserves. However, perennial ryegrass has lower levels of perennial reserves than other grass species. This reduces its resilience to concurrent water deficits or N deficiency. Under these conditions maintaining the recommended 3-leaf grazing intervals and/or leaving higher post-grazing pasture masses are recommended to assist canopy recovery. Other grass species, such as cocksfoot and tall fescue, provide more resilience, particularly in response to water deficits.

Decreasing Defoliation Frequency Enhances Bromus valdivianus Phil. Growth under Low Soil Water Levels and Interspecific Competition

Bromus valdivianus Phil. (Bv) is a water stress-tolerant species, but its competitiveness in a diverse pasture may depend on defoliation management and soil moisture levels. This glasshouse study examined the effect of three defoliation frequencies, based on accumulated growing degree days (AGDD) (250, 500, and 1000 AGDD), and two soil water levels (80–85% of field capacity (FC) and 20–25% FC) on Bv growth as monoculture and as a mixture with Lolium perenne L. (Lp). The treatments were applied in a completely randomised block design with four blocks. The above-ground biomass of Bv was lower in the mixture than in the monoculture (p ≤ 0.001). The Bv plants in the mixture defoliated more infrequently (1000 AGDD) showed an increase in root biomass under 20–25% FC compared to 80–85% FC, with no differences measured between soil water levels in the monoculture. Total root length was highest in the mixture with the combination of infrequent defoliation and 20–25% FC. Conversely, frequent defoliation treatments resulted in reduced water-soluble carbohydrate reserves in the tiller bases of plants (p ≤ 0.001), as they allocated assimilates mainly to foliage growth. These results provide evidence that B. valdivianus can increase its competitiveness relative to Lp through the enhancement of the root growth and the energy reserve in the tiller base under drought conditions and infrequent defoliation in a mixture.

How extreme summer weather may limit control of Festuca paniculata by mowing in subalpine grasslands

Background: The tussock grass Festuca paniculata can become strongly dominant in subalpine grasslands after cessation of mowing. The depletion of water-soluble carbohydrate (WSC) reserves has been suggested as a mechanism by which mowing can contain this species. By affecting plant physiology and especially by favouring WSC accumulation, extreme summer weather (i.e. exceptionally hot and dry) could however counterbalance the effects of mowing on WSC reserves in F. paniculata. The relevance of this hypothesis needs to be tested in the current context of climate and land-use changes. Aims: We investigated (1) the physiological mechanisms that control the growth of F. paniculata, (2) how they are affected by mowing and (3) whether extreme summer heat and drought could influence physiological mechanisms and thereby the ecological response of F. paniculata to mowing. Methods: In a field experiment we manipulated weather and mowing during two summers. For current summer weather (W0), ambient temperature was unchanged and precipitation was adjusted on the past 30-year average. Extreme summer weather (W+) corresponded to a seasonal change (+1 °C, –80% in precipitation compared to W0) and a three-week heatwave (+4.3 °C) in the first year. In addition, vegetation was either mown at 5 cm in late summer (M) or left unmown (U). Concentrations and absolute contents of WSC contained in tiller bases, leaf nitrogen concentration (LNC), vegetative multiplication, plant growth and leaf senescence were measured from one to four times, depending on the variable considered, throughout the summer of the second year of the experiment. Results: As compared to the unmown treatment, late-summer mowing decreased tillering, tussock size and LNC, regardless of the summer weather treatment. However, it depleted WSC pools, including fructans, only under current summer weather (W0). Conclusions: These results suggest that extreme summer heat and drought could alleviate the sensitivity of F. paniculata to mowing. They raise the question of the consequences of recurrent summer extremes for conservation management in subalpine grasslands.

Effect of short-term heat stress prior to flowering and at early grain set on the utilization of water-soluble carbohydrate by wheat genotypes

In Mediterranean environments, cereal crops are often exposed to short periods of elevated temperatures in spring when crops are approaching flowering and grain filling. Stem water-soluble carbohydrate (WSC) could be an important carbon bank for supporting grain filling in wheat especially when carbon assimilation is hampered by heat stress. However, there is relatively little information available in the literature on the effects of short periods of heat stress on grain growth and the importance of WSC in mitigating the effects of heat stress. Therefore, field and controlled environment studies were undertaken to determine the effects of short-term heat stress on the peduncle WSC content and its contribution to grain size of wheat in the field. Wheat genotypes were exposed to heat stress on a single-day at two different stages: (H1) near flowering or green anther stage and (H2) early grain set or 7–10 days after anthesis (DAA). On each occasion crops were enclosed in a portable heat chamber and the temperature was gradually increased to a maximum of 35°C. This single day heat stress event caused a significant reduction in individual grain mass (IGM) and grain number in both years. There was no discernible change in IGM until 14 days postanthesis, after which time grain growth in the heat-stressed plants was reduced. On average, as compared to the unheated control, the reductions in IGM in wheat genotypes ranged from 10 to 25%. The maximum WSC content was reduced by heat stress in all wheat genotypes. Heat stress (average of H1 and H2) reduced peduncle WSC content by 26% and mobilized WSC content by 15% across all studies. Mobilization of peduncle WSC content was also significantly reduced by the heat stress treatments and mobilization appeared to start at 21 days postanthesis when the grain was about one third of the final mass. IGM of wheat genotypes with higher maximum peduncle WSC content showed lower sensitivity to heat stress in both years of this study (H1: r=−0.82; p<0.001 and H2: r=−0.84; p<0.001). Over both seasons, heat stress treatments (H1, H2) and all genotypes, greater mobilized WSC was associated with increased grain size (r=0.70; p<0.01). It could be argued that selection of wheat genotypes with higher stem WSC reserves and greater mobilization of WSC could be used to buffer grain growth and development under heat stress conditions, which are a common occurrence in the Australian wheat belt.

The morphological and physiological responses of perennial ryegrass (Lolium perenne L.), cocksfoot (Dactylis glomerata L.) and tall fescue (Festuca arundinacea Schreb.; syn. Schedonorus phoenix Scop.) to variable water availability

AbstractThere is a growing interest in the use of deficit irrigation and perennial pasture species other than perennial ryegrass (Lolium perenne L.) in temperate agriculture, in response to the decreasing availability of irrigation water. Deficit irrigation requires an understanding of plant responses to drought stress to ensure maximum dry‐matter return on water applied. A glasshouse study was undertaken to investigate some of the morphological and physiological responses of perennial ryegrass, cocksfoot (Dactylis glomerata L.) and tall fescue (Festuca arundinacea Schreb.; syn. Schedonorus phoenix Scop.) to varied moisture availability. One water treatment involved frequent applications of water to maintain a soil water potential of approximately −10 kPa (100% treatment), and three other treatments involved applications at the same frequency, but using 33, 66 or 133% of the water applied in the 100% treatment. The water treatments continued over two plant regrowth cycles, followed by a ‘recovery’ phase of a single regrowth cycle during which all plants received the same water allocation as the 100% treatment. Depletion and replenishment of stubble water‐soluble carbohydrate (WSC) differed between the three species in response to soil moisture availability. By the second regrowth cycle, stubble WSC concentration and content in moisture‐stressed cocksfoot plants had increased, followed by a decrease during the subsequent recovery phase when the stored WSC reserves were utilized to support regrowth. The changes in stubble WSC reserves corresponded to the maintenance of relatively stable (i.e. the smallest reduction in leaf DM in response to moisture stress), but consistently lower DM production for cocksfoot compared with the other species. In contrast, moisture stress had no effect on the stubble WSC reserves of perennial ryegrass and tall fescue, with the exception of a significant decrease in WSC concentration under the 33% water treatment for perennial ryegrass. Perennial ryegrass achieved an intermediate DM yield and maintained positive growth rates throughout the study, even when watered at 33% of the requirement for optimal soil moisture levels. However, a more pronounced reduction in leaf DM in plants under moisture stress compared with the other species, combined with declining WSC reserves and the death of daughter tillers, highlighted the vulnerability of perennial ryegrass to poor persistence under prolonged drought conditions. Tall fescue appeared to have the greatest scope under moisture stress in terms of maintaining productivity and displaying attributes that contribute to persistence. Its leaf DM was consistently greater than that of the other species, displaying a smaller decline in growth under water stress compared to perennial ryegrass and an ability to recover faster upon re‐watering. This study has expanded the information available that compares and defines the potential of each species under moisture stress and emphasizes the importance of balancing short‐term DM production with long‐term persistence in choice of pasture species.

Impact of defoliation severity on photosynthesis, carbon metabolism and transport gene expression in perennial ryegrass.

Defoliation severity affects grass regrowth. The changes to biological processes affecting regrowth induced by severe defoliation are not fully understood, nor have they been investigated at a molecular level in field-grown plants. Field-grown perennial ryegrass (Lolium perenne L.) plants were defoliated to 20, 40 or 60mm during winter. Throughout regrowth, transcript profiles of 17 genes involved in photosynthesis and carbon metabolism or transport were characterised in stubble and lamina tissue. Although defoliation to 20mm reduced residual lamina area and stubble water-soluble carbohydrate reserves compared with plants defoliated to 40 or 60mm, net herbage regrowth was not reduced. Transcript profiles indicated a potential compensatory mechanism that may have facilitated regrowth. At the one-leaf regrowth stage, plants defoliated to 20mm had greater abundance of photosynthesis-related gene transcripts (rca, rbcS1, rbcS2, fba, fbp and fnr) and 20% greater stubble total nitrogen than plants defoliated to 60mm. A greater capacity for photosynthesis in outer leaf sheaths may be one potential mechanism used by severely defoliated plants to compensate for the reduced residual lamina area; however, this premise requires further investigation.

Effect of Defoliation Management on Water-Soluble Carbohydrate Energy Reserves, Dry Matter Yields, and Herbage Quality of Tall Fescue

There is limited information on the effect of leaf stage based defoliation management on the regrowth of tall fescue (Festuca arundinacea Schreb.). The aim of this study was to investigate the physiological changes in tall fescue during its regrowth cycle up to the five-leaf stage, and the effect of repeated defoliation at the one-leaf, two-leaf, and four-leaf stages on herbage quality, water-soluble carbohydrate (WSC) energy reserves and the rate of subsequent plant regrowth. Crude protein (CP) and metabolizable energy (ME) concentrations decreased with increased leaf regrowth stage, from 27% and 11.3 MJ kg−1 dry matter (DM) at the one-leaf stage, to 16.1% and 9.2 MJ kg−1 DM at the five-leaf stage, respectively. Acid detergent fiber (ADF) and neutral detergent fiber (NDF) concentrations increased during the regrowth cycle by 16 and 9%, respectively. Frequent defoliation therefore maximized the CP and ME concentrations and minimized the ADF and NDF concentrations of tall fescue, but limited DM yields during regrowth. Defoliation at the four-leaf stage resulted in 30% higher stubble WSC concentration and 20% higher leaf DM yield than defoliation at the two-leaf stage of regrowth, but compromised herbage quality. The stubble was confirmed as the primary storage organ for WSC reserves, while leaf and root growth were found to have an equally high priority for available energy following defoliation.

Effect of Defoliation Management on Water‐Soluble Carbohydrate Energy Reserves, Dry Matter Yields, and Herbage Quality of Tall Fescue

There is limited information on the effect of leaf stage based defoliation management on the regrowth of tall fescue (Festuca arundinacea Schreb.). The aim of this study was to investigate the physiological changes in tall fescue during its regrowth cycle up to the five‐leaf stage, and the effect of repeated defoliation at the one‐leaf, two‐leaf, and four‐leaf stages on herbage quality, water‐soluble carbohydrate (WSC) energy reserves and the rate of subsequent plant regrowth. Crude protein (CP) and metabolizable energy (ME) concentrations decreased with increased leaf regrowth stage, from 27% and 11.3 MJ kg−1 dry matter (DM) at the one‐leaf stage, to 16.1% and 9.2 MJ kg−1 DM at the five‐leaf stage, respectively. Acid detergent fiber (ADF) and neutral detergent fiber (NDF) concentrations increased during the regrowth cycle by 16 and 9%, respectively. Frequent defoliation therefore maximized the CP and ME concentrations and minimized the ADF and NDF concentrations of tall fescue, but limited DM yields during regrowth. Defoliation at the four‐leaf stage resulted in 30% higher stubble WSC concentration and 20% higher leaf DM yield than defoliation at the two‐leaf stage of regrowth, but compromised herbage quality. The stubble was confirmed as the primary storage organ for WSC reserves, while leaf and root growth were found to have an equally high priority for available energy following defoliation.

Patterns of leaf and root regrowth, and allocation of water‐soluble carbohydrate reserves following defoliation of plants of prairie grass (Bromus willdenowii Kunth.)

AbstractThis study utilized leaf stage‐based defoliation intervals to describe the concentrations and contents of water‐soluble carbohydrate (WSC) and nitrogen (N) in stubble and root reserves and their effect on the regrowth of prairie grass (Bromus willdenowii Kunth.) plants. The priority sequence for allocation of WSC reserves during the regrowth period was also investigated. There were substantially higher concentrations of WSC and N in the stubble compared with the roots following defoliation, confirming the stubble as the primary site for energy storage, with roots playing a lesser role. However, high R2 values for the relationships between WSC concentration in roots and regrowth variables suggested that plants of prairie grass were reliant on WSC reserves from the roots in addition to the stubble to meet the energy requirements of plants until adequate photosynthetic tissue had been produced. The sequence of priority for allocation of WSC reserves followed the order of leaf growth, root growth and tillering during the regrowth period. Although WSC reserves were identified as the primary contributor to plant regrowth following defoliation, there was also a strong relationship between stubble N concentration and regrowth variables.

Effect of defoliation management, based on leaf stage, on perennial ryegrass (Lolium perenne L.), prairie grass (Bromus willdenowii Kunth.) and cocksfoot (Dactylis glomerata L.) under dryland conditions. 1. Regrowth, tillering and water‐soluble carbohydrate concentration

AbstractA field study was undertaken between April 2003 and May 2004 in southern Tasmania, Australia to quantify and compare changes in herbage productivity and water‐soluble carbohydrate (WSC) concentration of perennial ryegrass (Lolium perenne L.), prairie grass (Bromus willdenowii Kunth.) and cocksfoot (Dactylis glomerata L.) under a defoliation regime based on leaf regrowth stage. Defoliation interval was based on the time taken for two, three or four leaves per tiller to fully expand. Dry‐matter (DM) production and botanical composition were measured at every defoliation event; plant density, DM production per tiller, tiller numbers per plant and WSC concentration were measured bimonthly; and tiller initiation and death rates were monitored every 3 weeks. Species and defoliation interval had a significant effect (P &lt; 0·05) on seasonal DM production. Prairie grass produced significantly more (P &lt; 0·001) DM than cocksfoot and ryegrass (5·7 vs. 4·1 and 4·3 t DM ha−1 respectively). Plants defoliated at the two‐leaf stage of regrowth produced significantly less DM than plants defoliated at the three‐ and four‐leaf stages, irrespective of species. Defoliation interval had no effect on plant persistence of any species during the first year of establishment, as measured by plant density and tiller number. However, more frequent defoliation was detrimental to the productivity of all species, most likely because of decreased WSC reserves. Results from this study confirmed that to maximize rates of regrowth, the recommended defoliation interval for prairie grass and cocksfoot is the four‐leaf stage, and for perennial ryegrass between the two and three‐leaf stages.

Effect of defoliation interval on water-soluble carbohydrate and nitrogen energy reserves, regrowth of leaves and roots, and tiller number of cocksfoot (Dactylis glomerata L.) plants

This study investigated the influence of leaf stage-based defoliation interval on water-soluble carbohydrate and nitrogen energy reserve status, regrowth of leaves and roots, and tiller number of cocksfoot (Dactylis glomerata L.) cv. Kara plants up to 24 days (3.5-leaf stage) following defoliation. Treatments were based on defoliation intervals of 1-, 2-, and 4-leaf stages of regrowth, with treatments terminated when the 1-leaf defoliation interval had been completed 4 times, the 2-leaf interval 2 times, and the 4-leaf interval once. Selected plants were destructively harvested prior to commencement of treatments (H0), immediately following cessation of treatments (H1), and at 5 days (H2), 10 days (H3), and 24 days (H4) following H1. Leaf, root, and tiller dry matter yield were determined at each harvest event, as well as tiller number/plant. Levels of water-soluble carbohydrate and nitrogen reserves in plant stubble and roots were determined at each destructive harvest. Initiation and death of daughter tillers were monitored from H0 to the completion of the study. More frequent defoliation of cocksfoot plants resulted in reduced water-soluble carbohydrate assimilation and therefore leaf, root, and tiller dry matter accumulation during the subsequent recovery period. Defoliation at the 1-leaf stage severely limited the regrowth potential of cocksfoot plants, whereas defoliation at the 2-leaf stage was adequate for plant recovery, but did not maximise regrowth. The results of this study showed that a defoliation interval based on the 4-leaf stage maximises water-soluble carbohydrate reserves, tillering, and leaf and root dry matter yields. The priority sequence for allocation of water-soluble carbohydrate reserves followed the order of leaf growth, root growth, and tillering during the regrowth period. Nitrogen energy reserves were found to play a minor role in the regrowth of cocksfoot plants following defoliation.

Changes in the physiology and feed quality of cocksfoot (Dactylis glomerata L.) during regrowth

Abstract A glasshouse study was undertaken to determine the physiological and morphological changes in cocksfoot (Dactylis glomerata L.) during regrowth after defoliation. Individual plants were arranged in a mini‐sward in a randomized complete block design. Treatments involved harvesting each time one new leaf had expanded (one‐leaf stage), up to the six‐leaf stage, with the plants separated into leaf, stubble (tiller bases) and roots. Stubble and root water‐soluble carbohydrate (WSC), stubble and leaf dry matter (DM), tiller number per plant and leaf quality (crude protein (CP), estimated metabolizable energy (ME) and mineral content) were measured to develop optimal defoliation management of cocksfoot‐based pastures. WSC concentration in stubble and roots was highest at the five‐ and six‐leaf stages. Mean WSC concentration (g kg−1 DM) was greater in stubble than roots (32·7 ± 5·9 vs. 9·4 ± 1·5 respectively). There was a strong positive linear relationship between plant WSC concentration and leaf DM, root DM and tillers per plant after defoliation (Adj R2 = 0·72, 0·88 and 0·95 respectively). Root DM plant−1 and tiller DM tiller−1 decreased immediately following defoliation and remained low until the three‐leaf stage, then increased from the four‐leaf stage. Tillers per plant remained stable until the four‐leaf stage, after which they increased (from 9·9 ± 0·5 to 15·7 ± 1·0 tillers plant−1). Estimated metabolizable energy concentration (MJ kg−1 DM) was significantly lower at the six‐leaf stage (11·01 ± 0·06) than at any previous leaf regrowth stage, whereas CP concentration (g kg−1 DM) decreased with regrowth to the six‐leaf stage. Both the levels of ME and CP concentrations were indicative of a high quality forage throughout regrowth (11·37 ± 0·04 and 279 ± 8·0 for ME and CP respectively). Results from this study give a basis for determining appropriate criteria for grazing cocksfoot‐based pastures. The optimal defoliation interval for cocksfoot appears to be between the four‐ and five‐leaf stages of regrowth. Delaying defoliation to the four‐leaf stage allows time for replenishment of WSC reserves, resumption of root growth and an increase in tillering, and is before herbage is lost and quality falls due to onset of leaf senescence.

Priority for allocation of water‐soluble carbohydrate reserves during regrowth of Lolium perenne

A glasshouse study was undertaken to determine the priority within the perennial ryegrass (Lolium perenne L.) plant for leaf and root growth and daughter tiller initiation after defoliation, in relation to various levels of water‐soluble carbohydrate (WSC) reserves at defoliation. Individual plants were arranged in mini‐swards, and underwent varying defoliation frequencies and ambient temperatures before defoliation, and harvest heights at defoliation, to obtain a gradient of WSC content at H1, the date when all plants were defoliated. Defoliation interval consisted of defoliating either three times at the one new leaf tiller–1 stage (1‐leaf stage) of regrowth, or once only at the 3‐leaf stage, up to H1, while night temperature in the week prior to H1 was altered from 15°C to either 8 or 20°C. At H1, plants were defoliated to a stubble height of either 20 or 50 mm. Plants were subsequently destructively harvested at days 4, 6, 8, 12, 18 and 27. Leaf and root extension and tiller dynamics were also measured.On a regrowth timescale, tiller initiation was most sensitive, root regrowth moderately sensitive, and leaf regrowth relatively insensitive to a decrease in WSC. The time of daughter tiller initiation also coincided with replenishment of stubble WSC levels.In contrast to this sequence of regrowth events following defoliation, the quantitative effects on growth were different, with elongation and survival of roots most affected by reduced WSC levels. A 30‐fold difference in stubble WSC at H1 between high and low WSC plants (1·52 vs. 0·05 mg tiller–1) produced only a 4‐fold increase in leaf dry matter (DM) after 27 d (2·2 vs. 0·6 g plant−1), while tiller number plant−1 increased 6‐fold (138 vs. 23% increase in tiller number from H1). Root elongation rate was 59 times higher in the high than in the low WSC plants (1·18 vs. 0·02 mm d−1).From a pasture management perspective, the study confirms that defoliation, coinciding with the 3‐leaf stage of regrowth and around a stubble height of 50 mm, optimizes persistence and productivity of perennial ryegrass. By allowing more rapid replenishment of WSC reserves, this optimal defoliation strategy enables a greater proportion of WSC to be allocated to maintain a more active root system, and promotes tillering, compared with more frequent and close defoliation.

Leaf number as a criterion for determining defoliation time for Lolium perenne: 2. Effect of defoliation frequency and height

AbstractThe object of this study was to determine the importance of frequency and height of defoliation on regrowth potential of Lolium perenne. Defoliation interval was based on stage of the regrowth cycle, as indicated by leaves per tiller.Simulated swards of Lolium perenne cv Yatsyn were grown as individual plants in a glasshouse kept at a day/night temperature of 25°C/15°C.Treatments imposed were defoliation at 2, 5 or 12 cm residual height, and low and high water soluble carbohydrate (WSC) level obtained by varying defoliation interval, i.e. defoliating at the 1‐leaf or 3‐leaf stage of the regrowth cycle. Regrowth after frequent short defoliations was only 65% of the less frequently defoliated plants taken over the full regrowth cycle. This was associated with a lower stubble WSC content (2·15 vs 17·5% in stubble) and a twenty‐seven‐fold difference in the amount of WSC in the stubble per plant. This difference in total WSC was a combined effect of more and heavier tillers and higher WSC content in stubble of plants defoliated less frequently at the end of the regrowth cycle. The regrowth of plants with WSC levels depleted by frequent defoliation when defoliated at 2 cm was significantly below that of those defoliated at 5 and 12 cm.The results indicate the desirability of defoliating plants at the 3‐leaf stage of the regrowth cycle. This not only allows the full regrowth potential to be expressed in that growth cycle, but also in the next cycle, by allowing the replenishment of WSC reserves and optimizing tiller status. The potential to regrow appears then to be based more on the total amount of WSC than the proportion of WSC in stubble.

Effects of summer drought and spring defoliation on carbohydrate reserves, persistence and recovery of two populations of cocksfoot (Dactylis glomerata) in a Mediterranean environment

SUMMARYThe ability to survive severe summer water stress determines the persistence, and therefore autumn yield, of temperate grasses in the Mediterranean environment. In the island of Corsica, sward survival and changes in carbohydrate content were investigated over 2 years, in two contrasting populations of cocksfoot (Dactylis glomerataL.): KM2, an early-flowering population of Mediterranean origin and Lutetia, a late-flowering cultivar bred in North West France.In Expt 1 (1987/88) three spring defoliation intensities were imposed. In Expt 2 (1989/90) three levels of water deficit were imposed under a rain shelter.In spring, dry matter (DM) yields and water-soluble carbohydrate (WSC) concentrations in leaf bases were similar (20–35% of DM) for both populations. From June to September, WSC reserves tended to decline in Lutetia, whereas accumulation of WSC was maintained until August (45% of DM) in KM2. Fructans formed the major component, and therefore followed the same seasonal pattern, as total WSC. Contents of trisaccharide and sucrose increased in late summer as drought progressed. Contents of glucose and fructose were lower but increased in August or in September according to treatment.In both field experiments during summer, KM2 had 50% more live tillers/m2than Lutetia, possibly because it remained in positive carbon balance for a longer period. WSC levels in summer and regrowth in autumn were reduced both by intense defoliation in spring and by severe water stress in summer. On average, autumn yields of KM2 were five times greater than those of Lutetia and the recovery yields were closely correlated with fructan contents of leaf bases in summer (r =0·80;P&lt; 0·001).The role of fructans and low DP-sugars as solutes in osmotic adjustment and as substrates for regrowth are discussed, and the differences between populations are analysed.