Canola and mustard response to short periods of temperature and water stress at different developmental stages

Canola and mustard response to short periods of temperature and water stress at different developmental stages Y. Gan 1, S. V. Angadi 1, H. Cutforth 1, D. Potts 2, V. V. Angadi 3, and C. L. McDonald 1 1 Semiarid Prairie Agricultural Research Centre, Agriculture and Agri-Food Canada, Swift Current, Saskatchewan, Canada S9H 3X2 (e-mail: gan@agr.gc.ca); 2 Saskatchewan Wheat Pool, Saskatoon, Saskatchewan, Canada S7N 4L8; and 3 Department of Agronomy, University of Agricultural Sciences, Dharwad, Karnataka, India-580005. Received 17 June 2003, accepted 16 January 2004. Gan, Y., Angadi, S. V., Cutforth, H., Potts, D., Angadi, V. V. and McDonald, C. L. 2004. Canola and mustard response to short periods of temperature and water stress at different developmental stages. Can. J. Plant Sci. 84: 697 704. Seed yield of Brassica crops in semiarid environments can be increased by minimizing the crops exposure to high temperature and water stress that often occurs during the growing season. A growth chamber study was conducted to determine the effect of short periods of high temperature and water stress at different developmental stages on seed yield and yield components of Brassica crops. Two canola-quality Brassica juncea PC98-44 and PC98-45, a Brassica napus canola Quantum, and a B. juncea oriental mustard Cutlass were grown under 20/18 C day/night temperatures with photoperiod of 16/8 h light/dark. High (35/18 C) and moderate (28/18 C) temperature stress was imposed for 10 d during bud formation, flowering, and pod development. Low (90% available water) and high (50% available water) water stress was imposed in combination with the temperature treatments. On average, the 35/18 C stress reduced main stem pods by 75%, seeds pod 1 25%, and seed weight 22% from the control. Seed yield per plant was reduced by 15% when plants were severely (35/18 C) stressed during bud formation, 58% when stressed during flowering, and 77% when stressed during pod development. Plants stressed at earlier growth stages exhibited recovery, whereas stress during pod development severely reduced most of the yield components. Effect of water stress on seed yield was minimal regardless of crop developmental stage. The four Brassica cultivars responded similarly to water stress. In response to temperature stress, B. juncea produced greater number of pods per plant but had a great rate of pod infertility than B. napus. Seed yield of B. juncea in semiarid environments can be increased by improving pod fertility, whereas the seed yield of B. napus can be increased by improving pod production and retention. Key words: Oilseed, yield components, Brassica species, moisture. Gan, Y., Angadi, S. V., Cutforth, H., Potts, D., Angadi, V. V. et McDonald, C. L. 2004. Réaction du canola et de la moutarde à de brefs stress thermiques et hydriques selon le stade de croissance. Can. J. Plant Sci. 84: 697 704. On peut accroître le rendement grainier des espèces du genre Brassica en milieu semi-aride en réduisant le plus possible l exposition des plantes aux températures élevées et au stress hydrique fréquents durant la période végétative. Les auteurs ont effectué une expérience en phytotron afin de préciser les conséquences de brefs stress thermique et hydrique à divers stades de croissance sur le rendement grainier et sur divers paramètres du rendement des espèces du genre Brassica. Pour cela, ils ont cultivé deux variétés de Brassica juncea de qualité canola (PC98-44 et PC98-45), le canola (B. napus) Quantum et la moutarde joncée (B. juncea) Cutlass à la température diurne/nocturne de 20/18 C durant une photopériode de 16/8 h de clarté/obscurité. Ils ont ensuite soumis les plantes à un stress thermique intense (35/18 C) ou modéré (28/18 C) pendant 10 jours lors de la formation des bourgeons, de la floraison et de la formation des gousses. Au stress thermique s ajoutait un faible (90 % d eau disponible) ou fort (50 % d eau disponible ) stress hydrique. En moyenne, un stress thermique de 35/18 C réduit la proportion de gousses sur la tige principale de 75 %, le nombre de graines par gousse de 25 % et le poids des graines de 22 % par rapport au témoin. Le rendement grainier des plantes soumises à un fort stress (35/18 C) diminue de 15 % quand le stress survient à la formation des bourgeons, de 58 % quand il survient à la floraison et de 77 % quand il survient au développement des gousses. Lorsqu elles subissent ce stress à un plus jeune âge, les plantes récupèrent bien, mais lorsqu il se produit lors du développement des gousses, la plupart des paramètres du rendement connaissent une baisse appréciable. Le stress hydrique n a qu une incidence minime sur le rendement grainier, peu importe le stade de croissance. Les quatre cultivars ont réagi de la même façon au stress hydrique. Pour ce qui est du stress thermique, B. juncea produit plus de gousses par plante que B. napus, mais le taux de stérilité par gousse est plus élevé. On peut accroître le rendement grainier de B. juncea en milieu semi-aride en améliorant la fécondité des gousses tandis qu on peut en faire autant avec B. napus en améliorant la production et la rétention des gousses. Mots clés: Oléagineux, facteurs de rendement, espèces du genre Brassica, humidité Abbreviations: Pod ms, the number of fertile pods produced on the main shoot; Pod br, the number of fertile pods produced on branches; Pod pl, total number of fertile pods per plant; SPP, seeds per pod; Yield ms, seed yield produced on the main shoot; Yield br, seed yield produced on branches; Yield pl, seed yield per plant 697

698 CANADIAN JOURNAL OF PLANT SCIENCE Alternative crops to cereals contribute to the diversification and intensification of the predominantly wheat-based monoculture cropping systems of the semiarid Canadian prairies (Zentner et al. 2002). Early-maturing canola such as B. napus and B. rapa are better suited to the short growing season of the region than other oilseed crops like sunflower and safflower (Miller et al. 2001). Canola is a cool season crop that is sensitive to high temperature stress (Morrison 1993; Brandt and McGregor 1997; Angadi et al. 2000; Morrison and Stewart 2002), while oriental mustard (B. juncea) is better adapted than canola to semiarid prairie conditions (Angadi et al. 2000). Recent success in developing new lines of B. juncea with oil and meal quality similar to canola has renewed interest in this species in the semiarid prairie (Rakow 1995). Temperature and water stress are the two most important abiotic factors limiting productivity of crops around the world (Boyer 1982; Hall 1992). A field-grown crop is under temperature stress for much of the growing season (Mahan et al. 1995). Crop yield reductions due to high temperature stress are evident from the large differences in yield between areas with cooler compared with warmer temperatures (Paulsen 1994). High temperatures accelerate the rate of plant development, reduce the length of the growing period, and reduce the yield potential (Entz and Fowler 1991). Although the effect of high temperature stress on crop yield is substantial, relatively less effort is devoted towards understanding the effect of high temperature stress when water stress also occurs (Hall 1992). High temperature stress and water stress frequently occur simultaneously in many crop production regions of the world. Stresses to crop growth can be synergetic or antagonistic, with one stress increasing the severity of the overall stress in some cases (Machado and Paulsen 2001), while one stress factor can also reduce the severity of overall stress in other cases (Wardlaw 2002). Little information is available on the interactive effect of temperature and water stress on the yield of B. juncea mustard and canola-quality B. juncea. Temperatures above 32 C occur, on average, about 7 d in a given growing season on the Canadian semiarid prairie (McCaig 1997). Temperatures this high can cause substantial yield losses in Brassica species (Angadi et al. 2000; Morrison and Stewart 2002). Water stress is another important abiotic factor in the region (Campbell et al. 1992). Seasonal temperature and rainfall patterns indicate that a combined temperature and water stress usually occurs in spring seeded annual crops during their reproductive stages. Because of the reduction of summerfallow (Zentner et al. 2002) and the expected temperature increases due to global warming (Cutforth et al. 1999; Cutforth 2000), the frequency of water and high temperature stress in the semiarid prairie may increase in the future. Understanding the multiple stress response of canola and mustard, with special emphasis on newly developed canola quality B. juncea, is essential for the sustainability of future cropping systems. High temperature and water stress can reduce crop yield by affecting both source and sink for assimilates (Hall 1992; Paulsen 1994; Mendham and Salisbury 1995). Plant response to abiotic stress depends on the developmental stage. Seed yield potential in Brassica crops depends on the events occurring prior to and during the flowering stage (Mendham and Salisbury 1995), while the reproductive period is most susceptible to stress (Hall 1992; Paulsen 1994). Comparing high temperature stress effect at various crop developmental stages, Angadi et al. (2000) found that the earlier the stress occurs, the greater the opportunity to recover. For example, a plant stressed during bud formation may recover more fully than a plant stressed during the pod development stage. Canola and mustard have indeterminate growth habits; both exhibiting substantial recovery from the stress (Mendham and Salisbury 1995). Yield compensation comes from both increased branching and increased efficiency of pod retention (Angadi et al. 2003). Source limitation for seed yield can arise from reduced photosynthesis and from rapid development of storage organs such as seeds. Abiotic stress at a later stage of reproduction can result in source limitation for seed yield by inducing the shedding of leaves and/or hastening maturity. Therefore, understanding the critical stage for high temperature and water stress and the ability to recover from the stress will enable producers and policy analysts to make more informed management decisions. The objectives of this study were to (i) determine the effect of high temperature and water stress, alone and in combination, on yield components and seed yield for B. napus canola, B. juncea oriental mustard, and canolaquality B. juncea, and (ii) assess the stress susceptibility and the ability to recover from high temperature and water stress imposed at different developmental stages. MATERIALS AND METHODS Plants and Growth Conditions The growth chamber experiment was conducted at the Agriculture and Agri-Food Canada Semiarid Prairie Agricultural Research Centre (SPARC), Swift Current, SK, Canada in 2001. Two canola-quality B. juncea breeding lines PC98-44 and PC98-45 developed at the Saskatchewan Wheat Pool, along with an oriental mustard Cutlass (B. juncea) and an Argentine canola Quantum (B. napus) were used for the study. For the sake of convenience, breeding lines are also referred to as cultivars in the manuscript. Plants were grown in growth chambers (Model GR 96, CONVIRON, Control Environment Ltd., Winnipeg, MB), and every 10 15 d the plants were rearranged within the replication to minimize potential variations within the growth chamber. Swinton silt loam soil collected from the SPARC research farm was placed in 2-L milk cartons. Commercial-grade peat moss was mixed into the surface 25-mm soil layer and into the bottom of the pot to prevent surface crusting and soil drainage. Seeds from all cultivars were treated with Vitavax RS (carbathiin + thiram + lindane) and were planted 25 mm deep. Because phenological development rates for the Brassica cultivars were different, staggered dates of planting were used to coordinate the reproductive phases for all cultivars. The plants were thinned to one per cartoon at the two to three leaf stage. Plants were grown in the growth chambers at 20/18 C day/night temperature until the high temperature

GAN ET AL. CANOLA AND MUSTARD RESPONSE TO TEMPERATURE AND WATER STRESS 699 treatments were imposed. Day/night cycles for temperature and photoperiod were 16 h day and 8 h at night. Beginning 3 wk after seeding, the plants were watered every 2 wk with 100 ml of nutrient solution (5 g of 20-20-20 dissolved in 1 L water). The systemic insecticide Pirliss (pirimicarb) was applied at the recommended rate to control aphids. The photosynthetic flux density at the leaf level (about 0.75 m from light source) was 300 mol m 2 s 1. Experimental Design and Stress Treatments The experiment was conducted in a split-split-split plot design with four replicates. Each of the four Brassica cultivars (main plot) received stress treatments at three different growth stages (sub-plot); bud (stage 5.0), flower (stage 6.0) and pod (stage 7.0) (BBCH Scale, Lancashire et al. 1991). At the desired stage, the plants were imposed to temperature stress (sub-sub-plot) at 35/18 C day/night (high stress), 28/18 C (moderate stress), and 20/18 C (no stress). The temperature stress was accomplished in growth cabinets (Model PGW36, CONVIRON, Control Environment Ltd., Winnipeg, MB). Two water stresses (sub-sub-sub-plot) were applied simultaneously with temperature stress by watering half of the plants to 90% of available soil water (low stress), and the remaining plants to 50% of available water (high stress). The pots were watered twice a day, at 0900 and 1500, with a given amount of water (pre-calculated based on water concentration of the dried soil and its water holding capacity) to bring the water concentration of the cartons to the desired level by weighing the pots individually. After 10 d in the stress treatment, plants were returned to the control chamber. When the pods on the main stem reached maturity (brown to tan color), the trials were terminated. The experiment was run once only with the assumption that the well-controlled experiment is repeatable, and the fact that this type of controlled experiments are very costly. Data Collection and Analysis The observations recorded at harvest included: the number of fertile pods on the main stem (Pod ms ), the number of fertile pods on the branches (Pod br ), seed yield produced on the main stem (Yield ms ), seed yield produced on the branches (Yield br ), mean number of seeds per pod and weight of 1000 seeds. To assess the effect of stress on pod fertility, the last flower opened prior to stress and the last flower opened at the end of the stress period were marked. This marking system allowed accurate counts on sterile (aborted flowers and buds) and fertile pods on the main stem before imposing the stress, during the stress and later in the recovery period. For plants stressed during bud formation, pod fertility was assessed only during and after the stress periods. Data were analyzed using a standard split-split-split GLM procedure (SAS Institute Inc. 1996). Fisher protected LSD was used for all mean separations (Gomez and Gomez 1984). Standard errors of means were calculated for main shoot fertile and sterile pods. Combined analyses were performed for some variables to determine the general trends of the effects whenever interactions were not significant. RESULTS AND DISCUSSION High temperature stress significantly reduced seed yield of Brassica cultivars by affecting most of the yield components, while water stress had little or no effect on yield or yield components (Table 1). Crop developmental stage and the magnitude of the stress influenced the role of yield-related variables in responding to the stress and post-stress recovery. Pod Production High temperature stress significantly reduced pod production both on the main stem (Pod ms ) and branches (Pod br ) (Tables 1 and 2), although the degree of the effect varied with crop developmental stage. On average, the plants stressed at 35/18 C producing only 25% of the Pod ms produced by plants at 20/18 C (Table 2). The plants receiving a moderate temperature (28/18 C) stress produced more than double Pod ms than the plants receiving the high temperature stress. Similarly, Pod br on the plants stressed at high temperature decreased by 10% compared to the control. Observations of fertile and sterile pods formed during stress periods indicated that the high temperature stress increased the proportion of infertile pods significantly (Table 3). Nearly all pods formed during high and moderate temperature stresses were sterile, although the number of fertile pods produced was greater under moderate compared to high temperature stress. For the control, the majority of pods formed were fertile. Similar trends in the pod fertility were observed for conventional canola (e.g., Quantum) and the canola-quality B. juncea (PC98-44), regardless of water status during the temperature stress. Pod ms was lowest for plants stressed during flowering compared to those stressed during either bud or pod formation (Table 4). Flowering is the most sensitive stage for temperature stress damage, probably due to susceptibility of pollen development, anthesis and fertilization (asynchrony of stamen and gynoecium development) to heat damage (Hall 1992). Our observations agree with previous findings (Morrison 1993; Angadi et al. 2000; Craufurd et al. 2003). Stressing Brassica plants during bud formation had the least effect on pod production (Table 4). Plants stressed during bud formation produced more Pod ms during the post-stress recovery period (17 pods plant 1 ) compared to those stressed during flower (8 pods plant 1 ) or pod (3 pods plant 1 ) formation stages (data not shown). Plants stressed during pod development produced the least Pod br or Pod Pl but the greatest Pods ms (Table 4). Pod br was highest on plants stressed at 28/18 C, resulting in the highest Pod Pl. These results indicate that Brassica crops have a strong ability to recover from stress imposed at an earlier growth stage. The recovery is through increased production of pods on branches, reflecting the strong plasticity of Brassica plants. Compared to Argentine canola, oriental mustard and canola-quality B. juncea produced significantly higher numbers of pods on branches (Table 2). As a result, the canola quality B. juncea produced 60 and 44% higher Pod Pl than Quantum canola when stressed at 28/18 and 35/18 C, respectively. The stronger ability of B. juncea to use branches to produce more pods during the post-stress period was responsible for the significant increase in the total number of

700 CANADIAN JOURNAL OF PLANT SCIENCE Table 1. ANOVA mean squares for yield components and seed yield of Brassica crops stressed at different growth stages in a controlled environment Seed Main shoot Main shoot Branches Total Seed weight Source df Sterile Fertile Branch Total (g) (g) (g) pod 1 (g 1000 1 ) Crop (C) 3 1244** 1291** 118234** 106701** 7.36** 8.76** 22.14** 623.7** 0.93** Rep (Crop) [R(C)] (Error a) 12 80 53 1765 1892 0.19 1.21 1.25 2.5 0.12 Stage (S) 2 256 749** 4789* 1775 0.28 39.20** 39.68** 11.9 8.25** S C 6 294 177** 1384 1270 0.36 3.25* 4.45* 12.7 0.56** S R(C) (Error b) 24 143 21 1046 946 0.12 1.13 1.42 6.8 0.10 Temperature (T) 2 25531** 17441** 87753** 132216** 37.00** 168.04** 350.95** 289.9** 18.31** T C 6 915** 455** 8602** 6765** 1.37** 6.54** 5.34** 6.5 0.36* T S 4 3171** 977** 16184** 16758** 2.12** 42.10** 53.54** 62.5** 2.69** T S C 12 324 91 3227** 2983** 0.15 2.33** 2.40* 6.3 0.30* T S R(C) (Error c) 72 181 73 875 890 0.19 0.81 1.02 5.6 0.16 Water (W) 1 508 0.1 902 640 0.01 8.10** 8.03** 1.1 0.42 W C 3 148 81 10 143 0.14 0.45 0.64 4.9 0.16 W S 2 76 13 28 79 0.02 0.13 0.14 11.3* 0.74* W T 2 129 28 881 1359 0.02 4.84** 4.90** 4.2 0.31 W C S 6 241 56 1952 1844 0.02 0.47 0.31 5.3 0.12 W C T 6 42 47 630 727 0.11 1.18 1.29 6.8 0.38* W S T 4 80 86 168 321 0.11 0.52 0.74 3.7 0.22 W C S T 12 125 39 1781 1672 0.03 0.61 0.60 7.3* 0.31* Residual error 108 165 44 1392 1401 0.10 0.71 0.78 3.3 0.17,*, ** represent significant at P < 0.10, P < 0.05 and P < 0.01, respectively. Table 2. Mean number of pods plant 1, seed yield, seeds pod 1 and seed weight for Brassica species with heat and water stress imposed at different developmental stages Fertile pods plant 1 Seed yield (g plant 1 ) Seeds Seed weight Crops Main shoot Branches Total Main shoot Branches Total pod 1 (g 1000 1 ) Temperature 20/18 C 36 183 219 1.42 5.91 7.33 11.19 3.21 28/18 C 24 229 253 0.65 5.41 6.04 7.82 3.31 35/18 C 9 166 175 0.18 3.41 3.57 8.36 2.50 LSD (0.05) 0.7 3.7 4 0.03 0.08 0.09 0.18 0.04 Water Low stress (90 %) 23 193 216 0.76 5.06 5.81 9.24 3.05 High stress (50 %) 23 191 214 0.75 4.75 5.47 9.04 2.97 LSD (0.05) NS NS NS NS 0.07 0.07 NS 0.03 Cultivars Cutlass 17 202 219 0.46 5.20 5.59 8.7 2.85 PC98-44 23 227 250 0.61 4.42 5.03 6.6 3.00 PC98-45 25 206 230 0.74 4.87 5.60 7.8 3.05 Quantum 27 130 157 1.24 5.15 6.37 13.6 3.10 LSD (0.05) 0.7 4.2 4 0.04 0.10 0.10 0.2 0.05 Developmental stage Bud 22 198 220 0.73 5.62 6.38 9.5 3.31 Flower 20 194 215 0.69 4.71 5.40 9.1 2.98 Pod 26 183 210 0.80 4.41 5.16 8.9 2.72 LSD (0.05) 0.7 3.6 3.7 0.03 0.08 0.09 0.2 0.04 pods (data not shown). Brassica crops produced the greatest Pod Pl at 28/18 C, and least Pod Pl at 35/18 C (Table 5). On average, the moderate temperature treatment increased the total number of pods per plant by 16% from the control, probably due to the moderate temperature stimulating the source/sink relationship, and increasing the potential sink size. Mendham and Salisbury (1995) noticed that moderate temperatures during flowering prevent canola from forming pods on the upper nodes of the main stem, which stimulates flower and pod formation on the branches. In contrast, the stress at 35/18 C severely reduced reproductive development, causing pod failure. Severe stress reduces photosynthetic resources (Morrison 1993) and decreases the duration of reproductive growth (Hall 1992). There was a significant temperature cultivar interaction in affecting pod production with stress treatment (Table 1), but the general trend of the responses was similar among the four Brassica cultivars studied (Table 5). Unlike high tem-

GAN ET AL. CANOLA AND MUSTARD RESPONSE TO TEMPERATURE AND WATER STRESS 701 Table 3. Sterile/fertile pod produced on the main stem of Brassica napus Quantum and canola quality Brassica juncea PC98-44 stressed at three temperature levels and at two water levels, at bolting, flower, and pod developmental stages Dry Wet Crop developmental stage 20/18 C 28/18 C 35/18 C 20/18 C 28/18 C 35/18 C Quantum Bolting 7/15 29/0 19/5 7/15 35/1 26/7 Flower 8/34 24/0 7/0 11/22 42/0 18/1 Pod 4/6 11/3 7/0 3/12 11/3 14/1 PC98-44 Bolting 6/6 23/7 19/0 10/6 23/9 22/0 Flower 6/29 33/6 18/0 8/26 35/8 17/0 Pod 0/17 10/14 13/0 1/18 7/13 13/0 Table 4. Mean number of pods plant 1, seeds pod 1, and 1000-seed weight for mustard, canola quality Brassica juncea and canola cultivars with different temperature treatments in a controlled environment Cutlass PC98-44 PC98-45 Quantum LSD (B. juncea) (B. juncea) (B. juncea) (B. napus) (0.05) Pods plant 1 20/18 C 235 245 236 158 9.6 28/18 C 266 293 267 175 5.1 35/18 C 157 213 189 140 8.4 LSD (0.05) 6.8 5.6 6.4 4.8 Seeds pod 1 20/18 C 10.33 9.20 10.06 15.2 0.42 28/18 C 7.83 5.49 6.27 12.2 0.31 35/18 C 7.97 5.15 7.06 13.3 0.63 LSD (0.05) 0.34 0.18 0.25 0.86 Seed weight, g 1000 1 20/18 C 2.93 3.18 3.35 3.39 0.07 28/18 C 3.21 3.28 3.46 3.29 0.08 35/18 C 2.42 2.55 2.38 2.66 0.09 LSD (0.05) 0.07 0.07 0.06 0.11 perature stress, water stress had no effect on pod production either on the main stem or the branches (Tables 1 and 2). The large effect of high temperature stress may have limited the response of the Brassica cultivars to the additional stress by water. Seed Set and Weight High temperature stress reduced seeds per pod (SPP) by 25% from the control (Table 2). Effort of the crops to increase pod production, when stressed, contributed to the decreased SPP. This study did not distinguish between the direct effect of temperature stress through failure of fertilization or seed abortion and the indirect effect through reduced resources (feedback response). Mendham and Salisbury (1995) reported that competition for assimilates among new branches and extra pod formation at the time when seed numbers were being determined was also responsible for decreases in SPP. In our study, the mean SPP was reduced by 13 to 44% at the 35/18 C temperature stress in spite of 11 to 33% reduction in Pod Pl (Table 5), indicating that the direct effect of high temperature stress on seed set may be partially due to pod-to-pod competition for photosynthetic assimilates. In a previous study, Angadi et al. Table 5. Effect of temperature stress imposed at different developmental stages on the pod and yield formation of Brassica cultivars grown in a controlled environment LSD Bud Flower Pod (0.05) Main shoot pods plant 1 20/18 C 31 40 35 1.3 28/18 C 24 18 31 1.5 35/18 C 11 3 12 0.9 LSD (0.05) 1.4 1.3 1.7 Branch pods plant 1 20/18 C 170 183 196 5.4 28/18 C 233 245 208 4.1 35/18 C 194 157 146 4.7 LSD (0.05) 5.3 5.2 4.8 Total pods plant 1 20/18 C 201 223 232 5.1 28/18 C 256 263 239 4.0 35/18 C 206 160 159 4.4 LSD (0.05) 5.2 5.5 4.8 Seeds pod 1 20/18 C 10.1 11.5 12.0 0.4 28/18 C 8.4 7.1 8.0 0.3 35/18 C 9.9 8.5 6.7 0.5 LSD (0.05) 0.4 0.4 0.4 Seed weight, g 1000 1 20/18 C 3.45 3.20 2.98 0.06 28/18 C 3.46 3.14 3.32 0.06 35/18 C 3.04 2.61 1.85 0.07 LSD (0.05) 0.06 0.07 0.07 0 (2000) found that a 35/15 C stress for 7 d decreased dry matter production of B. napus up to 30%, suggesting that limited assimilate contributes to the reduction of SPP. Among cultivars, Quantum canola produced highest SPP; 56% higher than Oriental mustard and 88% higher than B. juncea canola (Table 2), which compensated for its lower pod numbers. Temperature and water stress reduced the seed weight of the Brassica species (Tables 2 and 4). Temperature stress at 35/18 C reduced seed weight by 22% on average, whereas maintaining available water below 50% reduced seed weight by merely 3%, compared to the control. With mod-

702 CANADIAN JOURNAL OF PLANT SCIENCE erate temperature stress, the seed weight tended to be higher for B. juncea but not for B. napus (Table 5). The increased seed weight for B. juncea that was moderately stressed was attributable to the stress effects during pod development rather than during bud formation or flowering (Table 4). During pod development, the moderate (28/18 C) temperature decreased SPP (Table 2), which created a sinklimiting condition that favored increased seed weight (Mendham and Salisbury 1995). Angadi et al. (2000) found that a moderate temperature (28/15 C) treatment for a week during pod development did not reduce the biomass of B. napus, indicating that resources are not the limiting factor. These observations suggest that seed weight is the final opportunity for Brassica plants to recover from previous stress. Stresses imposed at a later stage of development reduce sink size (Mendham and Salisbury 1995), shorten the duration of seed filling (Hall 1992), and decrease the opportunity to recover (Morrison 1993). Seed Yield Temperature stress reduced Yield ms by 54% at 28/18 C and by 87% at 35/18 C, from the control, while water stress had no effect on Yield ms (Table 2). The detrimental effect of the short-period of high temperature stress on seed yield was due to its direct effect on flower development, pollination and fertilization, along with indirect effects on photosynthetic source (Hall 1992). In contrast, the effect of water stress on seed yield is mainly due to altering hydraulic and non-hydraulic properties. The short periods of water stress had a rather smaller effect on yield in the present study. Temperature stress imposed during flowering had the greatest impact on Yield ms (Table 6), with the reduction from the control ranging from 78% at 28/18 C to 96% at 35/18 C. When the stress was applied during pod formation, Yield ms was reduced by 50% at 28/18 C and by 72% at 35/18 C. Morrison (1993) found that late bud to early seed development was the most sensitive growth period to heat stress in canola. Angadi et al. (2000) also found that flowering period was more susceptible to heat damage than pod developmental stage in Brassica species. Temperature stress reduced seed yield per plant (Yield Pl ) by an average of 26% for plants stressed at 28/18 C and by 59% for plants stressed at 35/18 C (Table 2). Seed yield reduction due to high temperature stress was more severe in the present study than those found in a previous study (Angadi et al. 2000) where the duration of temperature stress was 7 d (10 d in the current study). Compared to the control, the average Yield Pl were reduced by 15% when severe stress (35/18 C) was applied during bud formation, 58% during flowering, and by 77% during pod development (Table 6). Plants stressed at an earlier stage exhibited greater recovery. Under field conditions, interplant competition for growth resources may limit the extent of branching in Brassica species (Angadi et al. 2003), while any obstructions of yield formation increase pod formation (McGregor 1981). The ability of Brassica plants to increase pod production in response to an early stage moderate stress decreases a few days after flowering (Mendham and Salisbury 1995). In the present study, the plants receiving Table 6. Effect of temperature stress imposed at different developmental stages on the mean seed yield of Brassica cultivars grown in a controlled environment LSD Bud Flower Pod (0.05) Seed yield on main shoot Seed yield (g plant 1 ) 20/18 C 1.28 1.65 1.34 0.08 28/18 C 0.64 0.37 0.93 0.08 35/18 C 0.36 0.06 0.12 0.02 LSD (0.05) 0.06 0.07 0.08 Seed yield on branches 20/18 C 5.35 6.00 6.37 0.15 28/18 C 6.30 5.01 4.98 0.16 35/18 C 5.28 3.12 1.79 0.19 LSD (0.05) 0.15 0.17 0.15 Total seed yield plant 1 20/18 C 6.62 7.65 7.71 0.15 28/18 C 6.92 5.37 5.91 0.21 35/18 C 5.64 3.18 1.88 0.19 LSD (0.05) 0.14 0.18 0.19 high temperature stress at pod development produced the lowest seed yield (Table 2), partly due to the seeds formed at the later developmental stage not having enough time to develop fully as shown by reduced seed weight with the late stage stress (Table 4). Water stress reduced both Yield br and Yield Pl but it did not influence Yield ms (Table 2), indicating that the shortterm water stress does not affect crop yield directly, but helps in the recovery process (Saini and Westgate 2000). At moderate temperature (28/18 C) stress, water stress reduced both Yield br and Yield Pl, whereas there was no effect of water stress on seed yield at high temperature (35/18 C) stress (Fig. 1). Wardlaw (2002) found in spring wheat that in the presence of high temperature stress, the effect of drought stress on yield was minimal. Savin and Nicolas (1996) also found that seed yield was affected by water stress when temperature stress was not severe and the water stress was long enough to reduce the assimilate supply during reproductive growth. In our study, imposing water stress for a period of 10 d and adjusting the water levels twice a day did not produce a severe water stress. There was a significant cultivar by temperature stress interaction in seed yield (Table 1). When stressed at 28/18 C, B. juncea mustard produced a similar Yield Pl as B. napus canola; both being greater than Yield Pl for canola quality B. juncea (Table 7). However, at 35/18 C, Yield Pl was reduced by 53% in B. juncea mustard, 59% in canola quality B. juncea, and only 38% in B. napus, from the control. The difference among crops in response to the high temperature stress was probably due to different adaptation strategies. When exposed to stress, B. juncea focused on increasing sink size by producing more pods at the expense of SPP and seed weight (Table 5), whereas B. napus main-

GAN ET AL. CANOLA AND MUSTARD RESPONSE TO TEMPERATURE AND WATER STRESS 703 Table 7. Mean seed yield on main shoot, branches and total for mustard, canola quality Brassica juncea and canola cultivars with different temperature treatments in a controlled environment Cutlass PC98-44 PC98-45 Quantum LSD (B. juncea) (B. juncea) (B. juncea) (B. napus) (0.05) (g plant 1 ) Seed yield on main shoot 20/18 C 0.77 1.26 1.51 2.14 0.12 28/18 C 0.48 0.48 0.54 1.17 0.08 35/18 C 0.11 0.07 0.16 0.39 0.04 LSD (0.05) 0.05 0.04 0.07 0.15 Fig. 1. General trend of the effect of temperature and water stress on seed yield produced on main shoot and branches, averaged for four Brassica cultivars. (The vertical bars are standard errors of means). tained pod production and seed development. Wright et al. (1995) reported that B. juncea cultivars are more efficient at increasing number of pods under stressful environments than B. napus, which may contribute to better yield. However, in the present study, B napus Quantum canola produced the highest seed yield and suffered the smallest yield reduction under the most severe stress conditions (Table 7). B. juncea produced more pods under stress conditions than B. napus (Table 5), but the increased pod production did not transfer into a realized seed yield (Table 7) due to the decreased SPP and seed weight (Table 5). One of the reasons for not realizing yield benefits for B. juncea in this controlled environment study was probably due to the lack of interplant competition. With no interplant competition, B. juncea focused on branching and continued forming extra pods, which limited SPP due to within-plant competition for photosynthetic resources. Low seed weight and fewer seeds pod 1 in B. juncea plants at 35/18 C stress indicated that the large number of pods formed did not have enough photosynthates to allocate among them. In contrast, the B. napus Quantum canola produced a limited number of pods per plant (67% of the pods produced by B. juncea), which decreased within plant competition for photosynthetic resources. As a result, Quantum canola significantly Seed yield on branches 20/18 C 6.13 5.75 6.19 5.56 0.25 28/18 C 6.14 4.72 5.21 5.56 0.17 35/18 C 3.25 2.80 3.21 4.38 0.15 LSD (0.05) 0.20 0.13 0.18 0.20 Total seed yield 20/18 C 6.89 7.01 7.70 7.71 0.27 28/18 C 6.62 5.20 5.75 6.68 0.20 35/18 C 3.26 2.88 3.37 4.77 0.16 LSD (0.05) 0.22 0.12 0.23 0.22 increased seed yield by increasing seeds pod 1 and seed weight (Table 5). In an earlier study, Angadi et al. (2003) also observed that B. juncea produced more pods than B. napus under stress, but the magnitude of the pod increase was smaller than those we observed. In the present study, pots were well-spaced in the room, creating a no-competition environment for plant growth. Further studies imitating field growth habits are needed to improve our understanding of the response to temperature and water stress among different Brassica species. In addition, only a limited number of cultivars were used in the present study due to growth room limitation. The different cultivars (such as hybrid and conventional type) within a species may perform differently under stressful field conditions. Detailed research to include more cultivars may further elucidate the response to stresses at different developmental stages. ACKNOWLEDGMENTS We gratefully acknowledge the financial support of Saskatchewan Canola Development Commission, Saskatchewan Wheat Pool, and the Matching Investment Initiative of Agriculture and Agri-Food Canada. We gratefully thank Dr. Peter McVetty and Dr. Van Ripley for their critical review of the manuscript. Angadi S. V., Cutforth H. W., McConkey B. G. and Gan, Y. 2003. Yield adjustment by canola under different plant populations in the semiarid prairie. Crop Sci. 43: 1358 1366. Angadi S. V., Cutforth H. W., Miller P. R., McConkey B., Entz, M. H., Volkmar, K. and Brandt, S. 2000. Response of three Brassica species to high temperature injury during reproductive growth. Can. J. Plant Sci. 80: 693 701. Boyer, J. S. 1982. Plant productivity and environment. Science 218: 443 448.

704 CANADIAN JOURNAL OF PLANT SCIENCE Brandt, S. A. and McGregor, D. I. 1997. Canola response to growing season climatic conditions. Pages 322 328 in Proc. Soils and Crops Workshop, 20 21 Feb. 1997. Extension Div., University of Saskatchewan, Saskatoon, SK. Campbell, C. A., Zentner, R. P., McConkey, B. and Selles, F. 1992. Effect of nitrogen and snow management on efficiency of water use by spring wheat grown annually on zero-tillage. Can. J. Plant Sci. 72: 271 279. Craufurd, P. Q., Prasad, P. V. V., Kakani, V. G., Wheeler, T. R. and Nigam, S. N. 2003. Heat tolerance in groundnut. Field Crops Res. 80: 63 77. Cutforth, H. W. 2000. Climate change in the semiarid prairie of southwestern Saskatchewan: temperature, precipitation, wind and incoming solar energy. Can. J. Soil Sci. 80: 375 385. Cutforth, H. W., McConkey, B. G., Woodvine, R. J., Smith, D. G., Jefferson, P. G. and Akinremi, O. O. 1999. Climate change in the semiarid prairie of southwestern Saskatchewan: Late winterearly spring. Can. J. Plant. Sci. 79: 343 350. Entz, M. H. and Fowler, D. B. 1991. Agronomic performance of winter versus spring wheat. Agron. J. 83: 527 532. Gomez. K. A. and Gomez, A. A. 1984. Statistical procedures for agricultural research. 1st ed. John Wiley and Sons, New York, NY. 680 pp. Hall, A. E. 1992. Breeding for heat tolerance. Plant Breed. Rev. 10: 129 168. Lancashire, P. D., Bleiholder, H., van den Boom, T., Langeluddeke, P., Stauss, R., Weber, E. and Witzenberger, A. 1991. A uniform decimal code for growth stages of crops and weeds. Ann. Appl. Biol. 119: 561 601. Machado, S. and Paulsen, G. M. 2001. Combined effects of drought and high temperature on water relations of wheat and sorghum. Plant Soil. 233: 179 187. Mahan, J. R., McMicheal, B. L. and Wanjura, D. F. 1995. Methods for reducing the adverse effects of temperature stress on plants: A review. Environ. Exp. Bot. 35: 251 258. McCaig, T. N. 1997. Temperature and precipitation effects on durum wheat grown in southern Saskatchewan for fifty years. Can. J. Plant. Sci. 77: 215 223. McGregor, D. I. 1981. Pattern of flower and pod development in rapeseed. Can. J. Plant Sci. 61: 275 282. Mendham, N. J. and Salisbury, P. A. 1995. Physiology: Crop development, growth and yield. Pages 11 64 in D. Kimber and D. I. McGregor, ed. Brassica oilseeds: Production and utilization. CAB International, Slough, UK. Miller, P. R., McDonald, C. L. Derksen, D. A. and Waddington, J. 2001. The adaptation of seven broadleaf crops to the dry semiarid prairie. Can. J. Plant Sci. 81: 29 43. Morrison, M. J. 1993. Heat stress during reproduction in summer rape. Can. J. Bot. 71: 303 308. Morrison, M. J. and Stewart, D. W. 2002. Heat stress during flowering in summer rape. Crop Sci. 42: 797 803. Paulsen, G. M. 1994. High temperature responses of crop plants. Pages 365 389 in K. J. Boote, J. M. Bennett, T. R. Sinclair, and G. M. Paulsen, ed. Physiology and determination of crop yield. ASA, CSSA, SSSA, Madison, WI. Rakow, G. 1995. Developments in the breeding of edible oil in other Brassica species. Pages 401 406 in Proc. the 9th International Rapeseed Congress, Cambridge, UK. Saini, H. S. and Westgate, M. E. 2000. Reproductive development in grain crops during drought. Adv. Agron. 68: 59 96. SAS Institute, Inc. 1996. SAS user s guide: Statistics. 6th ed. SAS Institute Inc., Cary, NC. Savin, R. and Nicolas, M. E. 1996. Effects of short periods of drought and high temperature on grain growth and starch accumulation of two malting barley cultivars. Aust. J. Plan. Physiol. 23: 201 210. Wardlaw, I. F. 2002. Interaction between drought and chronic high temperature during kernel filling in wheat in a controlled environment. Ann. Bot. 90: 469 476. Wright, P. R., Morgan, J. M., Jessop, R. S. and Cass, A. 1995. Comparative adaptation of canola (Brassica napus) and Indian mustard (B. juncea) to soil water deficits: Yield and yield components. Field Crops Res. 42: 1 13. Zentner, R. P., Wall, D. D., Nagy, C. N., Smith, E. G., Young, D. L., Miller, P. R. Campbell, C. A., McConkey, B. G., Brandt, S. A., Lafond, G. P., Johnston, A. M. and Derksen, D. A. 2002. Economic crop diversification and soil tillage opportunities in the Canadian prairies. Agron. J. 94: 216 230.