1 Morphology, forage and seed yield of soybean cultivars of different maturity grown as a forage crop in Turkey Suzan Altinok 1, İlker Erdoğdu 1, and Istvan Rajcan 2 1 Department of Field Crops, Faculty of Agriculture, University of Ankara, Diskapi-Ankara, Turkey; 2 Department of Plant Agriculture, Crop Science Bldg., University of Guelph, Guelph, Ontario, Canada N1G 2W1 ( Received 27 January 2003, accepted 18 September Altinok, S., Erdoğdu, ÿ. and Rajcan, I Morphology, forage and seed yield of soybean cultivars of different maturity grown as a forage crop in Turkey. Can. J. Plant Sci. 84: Soybean [Glycine max (L.) Merr.] has long been used as a forage crop but its morphological development and forage potential have not been studied extensively. A study was conducted using oilseed soybean cultivars of different maturity to determine their morphological development, forage yield, seed components and seed yield at the University of Ankara, Turkey. Five soybean cultivars OAC Salem (earliest), OAC Bayfield (midearly), OAC Eclipse (mid-early but later than OAC Bayfield), OAC Glencoe (full-season soybean) and SA.88 (late), were used in the study conducted in 2000 and The full-season cultivar OAC Glencoe was the tallest cultivar, and had the highest levels of fresh and dry matter yield and crude protein yield among soybean cultivars in both years. Therefore, OAC Glencoe is considered to be the most suitable soybean cultivar for forage production when intercropped with corn in Ankara region of Turkey. However, attention should be given to the earliest maturing cultivar OAC Salem that had the highest seed yield and better relative seed composition of all the cultivars. Key words: Forage, soybean, cultivar, morphology, development Altinok, S., Erdoğdu, ÿ. et Rajcan, I Morphologie et rendement fourrager et grainier des cultivars de soja à précocité variable récoltés comme provende en Turquie. Can. J. Plant Sci. 84: On utilise depuis longtemps le soja [Glycine max (L.) Merr.] comme provende sans avoir examiné en profondeur sa morphologie ni son potentiel fourrager. En 2000 et 2001, les auteurs ont étudié des cultivars oléagineux de soja à précocité variable en vue d en préciser la morphologie, le rendement fourrager, la composition de la graine et le rendement grainier à l Université d Ankara, en Turquie. À cette fin, ils ont retenu les cinq cultivars suivants : OAC Salem (le plus hâtif), OAC Bayfield (mi-hâtif), OAC Eclipse (mi-hâtif mais moins que le précédent), OAC Glencoe (pleine saison) et SA.88 (tardif). La variété de pleine saison OAC Glencoe avait les plants les plus hauts et a donné le plus important rendement de matière humide et sèche et le meilleur rendement en protéines brutes des cinq cultivars, les deux années de l étude. On estime donc qu il convient le mieux à la production de provendes lorsqu il est cultivé avec le maïs dans la région d Ankara. On devrait néanmoins porter une certaine attention au cultivar le plus précoce, OAC Salem, dont le rendement grainier était le plus élevé et dont les graines présentaient la composition la plus intéressante comparativement aux autres variétés. Mots clés: Fourrage, soja, cultivar, morphologie, développement 181 Normally, soybean [Glycine max (L.) Merr.] is grown as an oilseed crop, but its vegetative and seed components combined make good forage (Munoz et al. 1983). Although currently grown almost entirely as an oilseed crop, soybean was a popular summer annual forage legume in the past (United States Department of Agriculture, 1940). Soybean is desirable as a forage crop because it contains higher levels of protein than many other forages and has a dinitrogen fixation capability (Redfearn et al. 1999). Little is known about the influence of management practices on nutrient partitioning and composition of soybean plant components and, therefore, on whole-plant forage quality (Hintz and Albrecht 1994). Soybeans can produce forage similar in quality to alfalfa when using management practices typically used for forage production (Hintz et al. 1992). Use of adapted, highyielding soybean cultivars with good agronomic characteristics can help increase forage soybean production. There is limited information on morphological characters, yield and quality of forage soybean cultivars. Available data often come from oilseed (grain) soybean cultivars that were evaluated for their forage potential. Among oilseed type soybean cultivars, dry matter yield and forage quality were closely related to maturity (Redfearn et al. 1999). It was reported that the forage yield of soybean increased from 2.4 Mg ha 1 to 7.4 Mg ha 1 when harvested at R7 (Fehr et al. 1971) instead of the R1 stage of development (Hintz et al. 1992). However, forage quality declined between stages R1 and R5, followed by an increase from R5 to R7 as pods developed and seeds filled (Hintz et al. 1992). In a study conducted in Iowa, Darmosakoro et al. (2001) compared 13 forage and 5 grain soybean cultivars for forage and agronomic performance. They reported that the forage cul- Abbreviations: BD, number of days to flowering; BN, number of branches plant; CHU, crop heat units; CP, percent crude protein; CPY, crude protein yield; DMY, dry matter yield; FY, fresh yield; HD, the number of days to harvest; LR, percent leaves per plant; PH, plant height; PR, percent seed pods per plant; SR, percent stem weight per plant; SY, seed yield
2 182 CANADIAN JOURNAL OF PLANT SCIENCE Table 1. Monthly precipitation, mean temperature and relative humidity in Ankara, Turkey Precipitation (mm) Temperature ( C) Relative humidity (%) Month Average z Average Average January February March April May June July August September October November December Total Mean z Average of 65 yr. tivars produced more dry matter than grain cultivars but had a lower leaf stem 1 ratio and leaf + pod stem 1 ratio in August and September, which could reduce forage quality (Darmosakoro et al. 2001). The results also indicated that the significant difference observed between forage and grain cultivars would warrant development of breeding programs aimed at improving forage quality of soybeans (Darmosakoro et al. 2001). Tall forage soybean cultivars in maturity groups V, VI and VII have recently been developed to supply forage (Devine and Hatley 1998; Devine et al. 1998a, 1998b). When these cultivars were tested for yield and forage quality in the upper midwestern United States, it was found that decreasing the row width from 76 cm to 25 cm increased forage yield by 0.8 Mg ha 1 but had no effect on total herbage quality (Sheaffer et al. 2001). In Europe, where soybean is considered as an alternative forage crop, there is very little research on optimizing forage soybean production. The objective of this research was to test several oilseed soybean cultivars with different maturity to determine the potential of Canadian oilseed cultivars of different maturity to be grown successfully as forage crops. MATERIAL AND METHODS This research was carried out at the experimental field of the University of Ankara, Faculty of Agriculture, Department of Field Crops, during 2000 and The soil was clay loam, alkaline in nature and contained approximately 1% organic matter and 5% CaCO 3 The area receives about 343 mm precipitation and experiences an average monthly temperature of 12 o C and 60% relative humidity (Table1). In this study, five soybean cultivars with different maturity durations, OAC Salem (earliest, 2500 crop heat units [CHU; Ontario Ministry of Agriculture and Food (OMAF), 1993]; Tanner et al. 1998a), OAC Bayfield (mid-early, 2650 CHU; Tanner et al. 1998b), OAC Eclipse (mid-early 2700 CHU) and OAC Glencoe (full-season soybean, 3000 CHU) from the University of Guelph in Guelph, Ontario and, cultivar SA.88 (mid-early, oilseed cultivar from maturity group III) supplied by Sapeksa seed company were used. The seeding dates were 2000 May 01 and 2001 Apr. 30. The experimental design was a randomized complete block with three replications. Each plot consisted of 12 rows, 5 m long. Between row spacing was 35 cm and the seeding rate was seeds ha 1. Soybean seeds were inoculated with Nitragin brand (Brookfield, WI) granular inoculants of Bradyrhizobium japonicum. Weeds were removed by hoeing as needed in both years. Plots were flood irrigated three times during the growing season: at seedling (V1), beginning bloom (R1) and full bloom (R2) stages of the plant development in both years. Before the forage harvest, observations and measurements were taken on a subsample consisting of 10 plants plot 1, which were chosen randomly and labelled within the harvest area of each plot. Data were collected on number of days to flowering, plant height, number of branches per plant and number of days to harvest. After forage harvest, the subsample of 10 plants was separated into leaf, stem and pod components and weighed individually to determine the relative contribution of each component to the total plant biomass. Plots were harvested for forage when most seeds reached full size and pods were green (full seed, R6) from a 4-m 2 area. The fresh weight of all plants from this area was measured. In each plot, another subsample of 12 harvested soybean plants was weighed (fresh) and then dried at 80 C for a minimum of 72 h to determine dry matter content (Martin et al. 1990). Dried samples were ground and the amount of N was found using the Kjeldahl method with the Vapodest 2000 apparatus (Bremner 1982). The amount of N from each sample was multiplied by a factor of 6.25 (D. J. Hume, Univerisity of Guelph, personal communication) to calculate percent crude protein. Fresh weight obtained from each plot, dry matter yield and crude protein content were used for calculating fresh, dry matter and crude protein yields per hectare. Seed yield was determined when most of the pods in a plot turned tan (full maturity, R8). Seed was harvested from a 4-m 2 area in the remaining part of each plot. After seed harvest, the following observations were made on another 10 randomly selected soybean plants from each plot. They are: number of pods per plant, weight of pods per plant, weight of seeds per plant, number of seeds per pod, height of first seed pod, harvest index and 1000-seed weight. After harvest, seeds were threshed and seed yield was taken from
3 ALTINOK ET AL. SOYBEAN CULITIVARS IN TURKEY 183 Table 2. Yearly mean plant characteristics of five soybean cultivars grown in Ankara, Turkey, in 2000 and ND z PH BN LR SR PR HD FY DMY CP CPY Cultivar (d) (cm) (no.) (%) (%) (%) (days) (kg ha 1 ) (kg ha 1 ) (%) (kg ha 1 ) 2000 OAC Salem 53e 60c 3.7a 31a 22b 47a 100b 22543bc 9100a 15b 1377a OAC Bayfield 56d 76b 3.0b 28a 27b 45ab 104b 19127c 6043b 17a 1027a OAC Eclipse 60c 75b 2.5bc 35a 25b 39b 105b 26193a 8223ab 15b 1253a OAC Glencoe 64b 97a 2.2c 34a 26b 40b 108b 28190a 9113a 15b 1370a SA.88 75a 94a 1.3d 34a 34a 32c 139a 24860ab 7890ab 17a 1353a Mean 62A 80A 2.5A 33A 27A 41A 111A 24183A 8074A 16A 1276A CV OAC Salem 42c 64c 3.2a 32a 23b 45ab 89e 15923c 5177b 15a 740b OAC Bayfield 46bc 59d 3.4a 28b 23b 48a 92d c 5590b 15a 837ab OAC Eclipse 50b 63c 2.4b 34a 22b 44ab 97c 20360ab 7810a 15a 1143a OAC Glencoe 51b 97a 2.3b 33a 26b 41b 102b 22523a 7987a 13a 1043ab SA.88 63a 91b 1.4c 33a 39a 28c 131a 18493bc 6597ab 11a 720b Mean 51B 75B 2.6A 32A 27A 41A 102B 18664B 6612B 14B 897B CV Mean CV z ND = number of days to flowering; PH = plant height; BN = number of branches per plant; LR = % leaf weight per plant; SR = % stem weight per plant; PR = % pod weight per plant; HD = number of days to harvest; FY = fresh biomass yield; DMY = dry matter yield; CP = crude protein content; CPY = crude protein yield. a e, A,B Means of cultivars in a year within a column followed by the same lower case letters and means of 2 yr, followed by the same upper case letter, are not significantly different (P = 0.05, DMRT). Table 3. Two year mean plant characteristics of five soybean cultivars grown in Ankara, Turkey, grown in 2000 and 2001 ND z PH LR SR PR HD FY DMY CP CPY Cultivar (d) (cm) BN (%) (%) (%) (d) (kg ha 1 ) (kg ha 1 ) (%) (kg ha 1 ) OAC Salem 48e 62d 3.5a 32a 23c 45a 95d 19233cd 7138b 15ab 1058ab OAC Bayfield 51d 67c 3.2a 28b 25bc 47a 98cd 17575d 5817c 16a 932b OAC Eclipse 55c 69c 2.5b 34a 23c 42b 101bc 23277ab 8017ab 15ab 1198a OAC Glencoe 58b 97a 2.3b 34a 26b 41b 105b 25357a 8550a 14b 1207a SA.88 69a 92b 1.3c 34a 37a 30c 135a 21677bc 7193b 14b 1037 ab Mean CV Effect w Years ** ** NS NS NS ** ** ** ** ** ** Cultivars ** ** ** ** ** ** ** ** NS NS NS Year cultivar NS ** NS * NS NS NS * * * NS z ND = number of days to flowering; PH = plant height; BN = number of branches per plant; LR = % leaf weight per plant; SR = % stem weight per plant; PR = % pod weight per plant; HD = number of days to harvest; FY = fresh biomass yield; DMY = dry matter yield; CP = crude protein content; CPY = crude protein yield y Effect denotes components of variance from the combined ANOVA over 2 yr; * is F significant at P 0.05, ** is F significant at P 0.01, and NS is F not significant (P > 0.05). a e Means followed by the same letter are not significantly different at P = 0.05, according to DMRT. each plot. The seeds were ground using a coffee grinder and the amount of N was measured as described above for forage soybean. On the ground seed sample crude oil content was determined using the standard Soxhelet method with the Soxtherm 2000 aparatus (American Oil Chemistry Society 1961). Data were analyzed using ANOVA procedure of the SAS program (SAS Institute, Inc. 1989) at the P 0.05 and 0.01 levels of significance and means were compared using Duncan s Multiple Range Test (DMRT) at P = RESULTS AND DISCUSSION Plant Morphology and Forage Yields In both years, highly significant differences (P 0.01) were found for the cultivars (Table 2) for the number of days to flowering (ND), plant height (PH), number of branches per plant (BN), percent stem weight per plant (SR), percent seed pods per plant (PR), the number of days to harvest (HD), and fresh yield (FY). Cultivar differences were also significant (P 0.05) for dry matter yield (DMY) and percent crude
4 184 CANADIAN JOURNAL OF PLANT SCIENCE Table 4. Yearly mean seed characteristics of soybean cultivars grown in Ankara, Turkey, in 2000 and 2001 PN z PW SW SN FSP HI TSW SY CP CO Cultivar (no.) (g) (g) (no.) (cm) (%) (g) (kg ha 1 ) (%) (%) 2000 OAC Salem 31a 14a 8.7a 3.3a 11cd 44a 133a 1680a 29a 27a OAC Bayfield 26a 11abc 6.5b 3.1ab 12c 38ab 115b 1040b 27a 24ab OAC Eclipse 26a 10bc 6.2b 3.0bc 9d 37ab 117b 947bc 27a 24ab OAC Glencoe 31a 12ab 6.7ab 3.0bc 22b 33b 117b 977bc 28a 23b SA.88 19b 8c 5.3b 2.8c 30a 30b 113b 643c 28a 21b Mean 27A 11A 6.7B 3.0B 17A 36A 118A 105A 28B 24A CV OAC Salem 27ab 16a 10.7a 3.3a 8d 47a 131a 1674a 35a 20a OAC Bayfield 25bc 11b 6.0b 3.3a 10cd 35b 115bc 1269a 27c 23a OAC Eclipse 26ab 11b 7.6b 3.4a 11c 44a 116b 1139ab 33ab 22a OAC Glencoe 31a 12b 7.3b 3.2a 20b 36b 109c 1082ab 2 c 23a SA.88 21c 10b 6.4b 2.8b 29a 34b 119b 745b 31b 21a Mean 26A 12A 7.6A 3.2A 15B 39A 118A 1100A 31A 22B CV Mean CV z PN = number of pods per plant; PW= weight of pods per plant; SW= weight of seeds per plant; SN = number of seeds per pod; FSP = height of first seed pod; HI = harvest index; TSW = 1000-seed weight; SY= seed yield; CP = crude protein concentration in seeds; CO = crude oil concentration in seeds. a d, A,B Means of cultivars in a year within a column followed by the same lower case letters and means of 2 yr, followed by the same upper case letter are not significantly different (P < 0.05, DMRT). protein (CP) in 2000, and percent leaves per plant (LR) and crude protein yield (CPY) in 2001 (Table 2). The crude protein contents are similar to those of forage soybean varieties OR 13-12, OR and PA but lower than those of grain cultivars Sturdy and WISC.BK grown in Minnesota (Sheaffer et al. 2001). Soybean cultivars with later flowering and maturity duration were taller. A combined analyses over 2 yr (2000 and 2001) revealed, highly significant (P 0.01) year effect for ND, PH, HD, FY, DMY, CP and CPY (Table 3.). Cultivar effect was significant for all morphological and forage yield parameters, but not for forage quality traits such as CP or CPY (P > 0.05). The year cultivar interaction effect was significant only for PH (P 0.01), SR, DMY and CP (P 0.05) over the 2 yr (Table 3). The OAC Glencoe was the most productive cultivar over the two years. In general, later-maturing cultivars had higher fresh and dry matter yield than the earlier-maturing cultivars in this test. Although OAC Glencoe was an earlier-maturing cultivar than SA.88, its fresh and dry matter yields over 2 yr were higher than SA.88. This discrepancy may be due to differential reaction of the genetic background of these cultivars to environmental stimuli in the Ankara region. Differences among varieties for plant characteristics and forage yields appear to be related to the maturity group of cultivars in 2000 and In both years, the full-season cultivar OAC Glencoe ranked as the tallest and its fresh and dry matter yields were higher than that of any other cultivar (Table 2). Ocumpaugh et al. (1981) and Martin et al. (1998) observed superiority of late-maturing cultivars over earlymaturing types when soybean cultivars were intercropped with a grass and corn, respectively. Hintz et al. (1992) also found that late-maturing cultivars (Maturity Group III vs. Maturity Group II) produced greater forage yields but lower quality when harvested on the same stage of development. Earlier-maturing varieties are expected to set seed and their leaves senesce by the time corn is at its optimum stage for ensiling when soybean is intercropped with corn. The selection of an appropriate cultivar of soybean is, therefore, important for the success of the intercrop (Martin et al. 1998). The results of the present 2-yr study indicates that OAC Glencoe is the most suitable soybean cultivar for its forage yield and optimum growth stage that would coincide with the milk stage of silage corn. This hypothesis is being tested in a follow-up study. Seed Yield and Yield Components The soybean cultivars showed significant differences for seed yield and other seed components (Table 4). Earlier-maturing cultivars generally produced higher seed yield than late-maturing types. The earliest-maturing cultivar, OAC Salem, had significantly larger seed (TSW) and seed yield (SY) over the 2- yr test period (Table 4). The differences in 1000 seed weight are consistent with previous reports indicating relatively larger seed of the cultivar OAC Salem (Tanner et al. 1998a) compared with cultivar OAC Bayfield (Tanner et al. 1998b). The crude protein levels observed in this study were somewhat lower than the level reported earlier for OAC Salem and OAC Bayfield (Tanner et al. 1998a, b); however, it is within the range commonly found in Ontario oilseed cultivars [Ontario Oil and Protein and Seed Crop Commitee (OOPSCC) 2003]. The possible reason for the relatively high crude protein concentration for Ontario grown soybean may be due to the methodology used for determination of protein concentration in Ontario, where protein is estimated using the near infra-red reflectance machine (GrainSpec) and standard callibrations (Association of Official Analytical Chemists 1990).
5 ALTINOK ET AL. SOYBEAN CULITIVARS IN TURKEY 185 Table 5. Seed characteristics of soybean cultivars grown in Ankara, Turkey over combined over 2 yr ( ) PN z PW SW SN FSP HI TSW SY CP CO Cultivar (no.) (g) (g) (no.) (cm) (%) (g) (kg ha 1 ) (%) (%) OAC Salem 29ab 15a 9.7a 3.3a 9c 46a 132a 1477a 33a 24a OAC Bayfield 26c 11b 6.3b 3.2ab 11c 36bc 115b 1154b 27d 23a OAC Eclipse 26c 11bc 6.9b 3.2ab 10c 40b 117b 1043b 30b 23ab OAC Glencoe 31a 12b 7.0b 3.1b 21b 34c 109b 1029b 28cd 23ab SA.88 20d 9c 5.9b 2.8c 29a 32c 116b 694c 30bc 21b Mean CV Effect y Years NS NS * ** ** ** ** ** NS ** Cultivars ** ** ** NS * NS NS ** ** NS Year cultivar NS NS NS ** NS NS NS ** NS * z PN = number of pods per plant; PW= weight of pods per plant; SW = weight of seeds per plant; SN = number of seeds per pod; FSP = height of first seed pod; HI = harvest index; TSW = 1000-seed weight; SY = seed yield; CP = crude protein concentration in seeds; CO = crude oil concentration in seeds. y Effect denotes components of variance from the combined ANOVA over 2 yr, whereas, * is F significant at P = 0.05; ** is F significant at P = 0.01; and, NS is F not significant (P > 0.05). a c Means fallowed by the same letter are not significantly different at P = 0.05, according to DMRT. The combined analyses showed significant cultivar effect for all traits except crude oil content (Table 5). SA.88, a later-maturing cultivar than the Canadian cultivars, produced significantly lower seed yield than the other cultivars. This may have been caused by relatively high temperatures during the later part of the growing season profoundly affecting plant growth and development for SA.88 more than that of other cultivars under Ankara conditions. This may also have adversely affected forage yield of SA.88 causing it to produce less biomass than the early-maturing Canadian cultivars, including the latest, OAC Glencoe. The earliest cultivar OAC Salem had the highest seed yield and a superior protein concentration compared to other soybean cultivars (Table 5). However, the height of first seed pod for this cultivar was lower than that of the other cultivars. Because of this, its harvest could be problematic, specially when harvesters with floating cutter bars are not available. High seed yields for early-maturing soybean cultivars is not surprising because short season oilseed soybean cultivars tend to be best adapted and able to reach the full seed production potential in areas with a short growing season, such as the Ankara region of Turkey. Under such conditions, the late maturing cultivars tend not to flower at the optimum time, do not produce as many flowers per plant as a short-season cultivar, often resulting in a smaller number of pods and seeds being set. The relative adaptation of soybean cultivars is the main reason for the existence of five CHU (OMAF 1993) zones (OOPSCC 2003) in the Ontario soybean producing area, where most of the tested cultivars were developed. In conclusion, five oilseed soybean cultivars of different maturity durations showed significant differences in morphological development, forage and seed yield components. The full-season cultivar OAC Glencoe was ranked as the latest and tallest among the tested cultivars and produced the highest fresh and dry matter yield. Therefore, when selecting a soybean cultivar for intercropping with corn, OAC Glencoe seems to be most suitable because of its ability to produce high biomass yield approximately at a time when silage corn is at the milk stage. However, consideration should also be given to the earliest cultivar OAC Salem, as it produced the highest seed yield and best relative seed composition among all the cultivars tested at Ankara, Turkey. American Oil Chemist Society Official method Aa Official and tentative methods of the American Oil Chemists Society. AOC, Campaign, IL. Association of Official Analytical Chemists Official methods of analysis. AOACS, Washington, DC. Bremner, J. M., Total nitrogen. Pages in A. L. Page, R. H. Miller, and D. R. Keeney, eds. Methods of soil analysis, part 2. ASA Madison, WI. Darmasarkoro, W.H., Harbur, M. M., Buxton, D. R., Moore, K. J., Devine, T. E. and Anderson, I. C Growth, development and yield of soybean lines developed for forage. Crop Sci. 93: Devine, T. E. and Hatley, E. O Registration of Donegal forage soybean. Crop Sci. 38: Devine, T. E., Hatley, E. O. and Starner, D. E. 1998a. Registration of Derry forage soybean. Crop Sci. 38: Devine, T. E., Hatley, E. O. and Starner, D. E. 1998b. Registration of Tyrone forage soybean. Crop Sci. 38: Fehr, W. R., Caviness, C. E., Burmond, D. T. and Pennington, J. S Stage of development description for soybeans. Crop Sci. 11: Hintz, R. W. and Albrecht, K. A Dry matter partitioning and forage nutritive value of soybean plant components. Agron J. 86: Hintz, R. W., Albrecht, K. A. and Oplinger, E. S Yield and chemical composition of soybean forage as affected by cultivar and management practices. Agron J. 84: Martin, R. C., Astatkie, T. and Cooper, J. M The effect of soybean variety on corn soybean intercrop biomass and protein yields. Can. J. Plant Sci. 78: Munoz, A. E., Holt, E. C. and Weaver, R. W Yield and quality of soybean hay as influenced by stage of growth and plant density. Agron J. 75: Ocumpaugh, W. R., Matches, A. G. and Luedders, V. D Sod-seeded soybean for forage. Agron. J. 73: Ontario. Ministry of Agriculture Crop heat units for corn and other warm-season crops in Ontario. [Online], Available: gov.on.ca/omafra/english/crops/facts/ htm [25 July 2003].
6 186 CANADIAN JOURNAL OF PLANT SCIENCE Ontario Oil and Protein Seed Crop Committee Ontario soybean variety trials. [Online], Available: org/vartrial.html [25 July 2003]. Redfearn, D. D., Buxton, D. R. and Devine, T. E Sorghum intercropping effects on yield, morphology, and quality of forage soybean. Crop. Sci. 39: SAS Institute, Inc SAS/STAT user s guide. Version 6. 4th ed., Vol. 2. SAS Institute, Inc., Cary, NC. Sheaffer, C. C., Orf, J. H., Devine, E. and Jewett, J. G Yield and quality of forage soybean. Agron. J. 93: Tanner, J. W., Luzzi, B. M., Montminy, W., Gostovic, P. and Hume, D. J. 1998a. OAC Salem soybean. Can. J. Plant Sci. 78: 629. Tanner, J. W., Luzzi, B. M., Montminy, W., Gostovic, P. and Hume, D. J. 1998b. OAC Bayfield soybean. Can. J. Plant Sci. 78: 625. United States Department of Agriculture Agricultural statistics. Government Printing Office, Washington, DC.