Harvest stage effects on forage yield and quality for rape and turnip genotypes

The effects of two different harvest stages (full flowering and full podding) on forage yield and quality of ten forage rape (Brassica napus L. var. oleifera) and three turnip (Brassica rapa L. var. rapa) cultivars were evaluated under rainfed conditions in a Mediterranean type climate at Bursa, Turkey, during the 2002/2003 and 2003/2004 growing seasons. Plant height, branches per plant, leaf length and width, plant part components, and dry matter (DM) and crude protein (CP) yield were measured. Significant differences were observed for the main effects cultivar and harvest stage on DM and protein yields. The CP content of the whole plant and vegetative parts of the Brassica genotypes decreased as plant maturity progressed. In general, turnip cultivars showed better performance in DM yield. Fall sown turnip produced 9.10 and 12.1 Mg ha DM yield, with 15.1 and 9.10% CP concentrations, at full flowering and at podding stage, respectively. Signif icant differences were noted among cultivars for these two parameters. Protein content decreased dramatically in stem parts when maturity advanced from full flower to full pod stages. High leaf percentage and high protein content at full flowering suggest that this should be the preferred forage harvest stage for rape and turnip cultivars. Additional key words: Brassica napus L. var. oleifera, Brassica rapa L. var. rapa, crude protein, dry matter yield, plant proportions.

Palabras clave adicionales: Brassica napus L. var.oleifera, Brassica rapa L. var.rapa, proteína cruda, rendimiento de materia seca, proporciones de plantas.Koch and Mitchell, 1985;Guillard and Allinson, 1988).Therefore, forage Brassicas are often used in cool regions for livestock production (Claridge, 1972;Macfarlane Smith et al., 1984).Jung et al. (1986) reported yields of more than 7.01 Mg ha -1 roots and leaves, 90 d after planting in the majority of 19 cultivars of forage Brassica spp.Guillard and Allison (1988) reported total plant yields of 5.10 to 8.30 Mg ha -1 for several Brassica species planted in mid-June and harvested 100 d after planting in Connecticut, USA conditions.The authors concluded that decreasing temperature and light intensity reduced herbage production of rape (Brassica napus L. var.oleifera), but not that of turnip (Brassica rapa L. var.rapa) or turnip hybrids.Wiedenhoeft (1993) reported that the amount of forage produced by Brassica plants averaged 3.10 Mg ha -1 during summer and 3.60 Mg h -1 during fall and early winter.To maximize total dry matter (DM) yield and maintain good protein and energy levels, these crops should be cut any time from early podding stage (just after the flowers have dropped) up to the stage where the lower leaves are starting to drop.When Brassica species were cut at the early to mid-podded stage of development, protein content averaged 14-15% and total digestible nutrients (TDN) were 58-60% in a DM basis (Anonymous, 2004).
In general, fall sowing result in higher yields than spring sowing in the Mediterranean basin.Fall sown pea (Pisum sativum L.) and common vetch (Vicia sativa L.) produced a greater forage and seed yield than when these species were sown in the spring (Cousin, 1976;Aydogdu and Acikgoz, 1995;Uzun and Acikgoz, 1998).Similarly, forage turnip seed was successfully produced in fall seeded plots under rainfed conditions in a Mediterranean-type environment (Bilgili et al., 2003) Trifluralin was sprayed at 1.0 kg ha -1 after sowing for weed control.Hand-hoeing was carried out when necessary.Nitrogen fertilization of 100 kg ha -1 after sowing and 50 kg ha -1 in spring was uniformly applied to all plots as ammonium nitrate.There were no significant problems with pests, diseases or weeds during the course of the study.
Ten plants were removed at random from each plot at full flowering stage (early May) and full podding stage (early June) for morphological measurements.Plant height, branches per plant, leaf length and width (8 th leaf on plant) were measured, as well as morphologic proportional determinations and crude protein (CP) analysis on separate part constituents (leaf, stem and flower/pod).All plant components were dried and weighed.The forage yield was determined from 2.8 m 2 of each plot and 200 g sub-samples were dried in ovens at 70ºC for 48 h for DM yield determination (Martin et al., 1990).Dried samples were ground and nitrogen content determined using the Kjeldahl method.The nitrogen content from each sample was then multiplied by factor 6.25 to calculate CP content.
Analysis of variance was performed on morphological measurements and dry matter yield using MINITAB (Release 14 for Windows, University of Texas at Austin, TX) and MSTAT-C (Version 2.1, Michigan State University, East Lansing, MI) programs.The significance of treatment, main effects, and interactions were determined at the 0.05 and 0.01 probability levels, using the F-test.The F-protected least signif icant difference (LSD) was calculated at the 0.05 probability level.Mean comparisons of B. napus var.oleifera and B. rapa var.rapa were carried out by orthogonal contrasts using JMP 5.0 program (SAS Inst., 2002).
There were significant effects from most sources of variation on the measured parameters at both full flowering and full podding harvest stages of the Brassica cultivars in individual and combined years.Cultivars × year interactions were not significant for DM and protein yield, as well as for most other measured components.Therefore results for the two growing seasons were combined (Table 2).
Plant height, leaf width and leaf length at full flowering and podding stages were significantly influenced by cultivar.In general, plant height differences between rape and turnip cultivars were negligible at both harvest stages.At full podding stage, Emerald and English Giant were the tallest (157 cm), while Bristol was the shortest (109 cm) among the rape cultivars.Polybra was the shortest (122 cm) turnip cultivar.Turnip cultivars, in average, had more branches per plant than rape cultivars at both full flowering (4.87 vs 6.59, respectively), and full podding stages (6.03vs 7.15, respectively).Turnip cultivars also had longer leaves than rape cultivars at both full flowering (16.0 vs 15.0 cm, respectively), and full podding stages (18.0 vs 15.9 cm, respectively).However, turnip cultivars had narrower leaves than rape cultivars at both full flowering (6.12 vs 6.98 cm, respectively), and full podding stages (6.99 vs 7.87 cm, respectively).Plant height and leaf dimensions of both species increased at the later growing stage, as plants developed (Table 2).
Overall, the average DM yield of turnip cultivars (7.78 and 10.3 Mg ha -1 ) was higher than that of rape cultivars (5.78 and 6.32 Mg ha -1 ) at both full flowering and full podding stages, respectively (Table 2).Although significant difference between turnip and rape cultivars was not found for DM yield at both harvest stages in orthogonal contrast tests, Malvira turnip cultivar was consistently superior at both harvesting stages, with average DM yield of 9.12 and 12.1 Mg ha -1 , respectively.On the other hand, Emerald was the most productive rape cultivar at both harvest stages, with DM yield of 7.35 and 8.91 Mg ha -1 , respectively.Crude protein content of Brassica cultivars decreased as plant maturity advanced, with average values for rape and turnip cultivars of 12.6 and 14.2%, respectively, at full flowering, compared to 7.77 and 9.17% when harvest was performed at full podding stage (Table 2).These values indicate that turnip cultivars had significantly higher protein content than rape cultivars.The orthogonal contrast test also confirmed that the difference in CP content between rape and turnip cultivars was significant at both harvest stage.
Total CP yield also decreased as plant maturity advanced, with average values for rape and turnip cultivars of 716 and 1,099 kg ha -1 , respectively, at full flowering, compared to 492 and 914 kg ha -1 when harvesting at full podding stage (Table 2).Orthogonal contrast tests showed that total CP yield difference between rape and turnip cultivars was not significant.However, among the rape cultivars, Emerald consistently had the highest CP yield and Jumbo the lowest at both harvest stages.Among the turnip cultivars, Malvira had the highest CP yield and also consistently out-yielded all other rape cultivars (Table 2).Dry matter yields of Brassica crops in the present study were similar to those of earlier reports.Harper and Compton (1980) reported that harvest date significantly affected DM yield and leaf-stem ratio of Brassica crops.Jung et al. (1986) found that DM yields for 13 of 19 Brassica cultivars exceeded 7.01 Mg ha -1 .Reid et al. (1994) examined the yield potential of f ive Brassica cultivars in USA.They found that DM yield ranged from 5.60 to 10.5 Mg ha -1 , which were generally similar to values of 7.01 and 9.01 Mg ha -1 reported by Jung et al. (1984) for forage turnip and rape, respectively, grown with minimum tillage, and high N and P fertilization levels in Pennsylvania, USA.Rao and Horn (1986) obtained a DM yield of rape near 7.01 Mg ha -1 at El Reno, OK, in the southern US Plains.Altinok and Karakaya (2003) found that the rape cultivar Emerald was the most productive in Turkey, with a DM yield of 5.50 Mg ha -1 .
Since turnip plants developed more quickly in early spring and flowered 12-15 d earlier than rape (mid April vs. beginning of May), turnip cultivars generally produced higher DM yield than rape cultivars in Mediterranean-type environments.In this study, climatic conditions of early spring (average temperature of 6.90°C and rainfall 47.6 mm in March and 11.5°C and 81.2 mm in April) were very suitable for vegetative development of turnip plants.However, rapid onset of high temperatures and water def icit in late spring depressed the vegetative development of the rape cultivars.The mean temperature in May (flowering period of rape) increased to 18.2°C and rainfall decreased to 34.2 mm.Similar to what has been indicated by several other researchers (Guillard and Allison, 1988;Undersander et al., 1991;Rao and Horn, 1995), in the present study fall sown turnip plants developed quickly and produced high DM yield with high protein content under the cool and moist conditions of early spring.
The DM proportions, and CP content and yield for the different plant components, at both harvest stages, are shown in Table 3. Harvest stage had a significant effect on the DM partitioning of plant parts.Even though the leaf component of Brassica cultivars decreased from full flowering to full podding, the flower + pod portion increased.However, the stem component was not affected by the stage at which plants were harvested.The average leaf component of the total plant mass decreased from flowering to podding stage, from 26.5% to 6.81% in the rape cultivars and from 29.1% to 7.84% in the turnip cultivars, respectively.The flower + pod fraction increased from flowering to podding stage, as expected.The general average proportion of flower + pod was 15.6% at flowering stage, and rapidly increased to 38.9% at podding stage.Crude protein content varied signif icantly between plant parts.The flower + pod component had the highest CP content while the stem component had the lowest at both harvest stages.Particularly at podding stage, stems had very low CP content (3.11%), which was a 31.0%decrease from the flowering stage.Leaf and flower + pod contributed to CP yield almost equally at flowering stage (36.4 and 35.4%, respectively).Contrarily, flower + pod parts produced 68.2% of the CP yield at podding stage, while the contribution of leaves and stems to CP yield dropped to 8.80 and 23.0%, respectively (Table 3).
The CP results obtained in the present study were supported by reports from several researchers.For example, Rao and Horn (1995) found CP content values of 16.0% in 1983 and 23.0% in 1984, 95 d after seeding rape.They stated that CP content of rape ranged from 20.0-25.0% in leaves and averaged 10.0% in the stems.They also indicated that the CP content of all plants parts declined with maturity and that this decline was more rapid for the stem component.These authors attributed this rapid decline of CP to increasing DM accumulation rate and N dilution during this growth period.Other researchers concluded that the CP concentrations in the leaves of kale (B.oleracea L.) and turnips also declined with time (Pelletier et al., 1976;Reid et al., 1994).
In conclusion, fall sown Brassica genotypes, particularly turnip cultivars, have the potential to provide additional or supplemental forage with high DM and CP yield in the early summer of a region with quite mild winters.According to our results, DM yield of most Brassica genotypes was slightly increased by later harvest.However, full flowering stage with high leaf percentage and high CP content must be the preferred harvest stage of Brassica forage production.

Table 1 .
Monthly rainfall, mean temperature and relative humidity in 2002Monthly rainfall, mean temperature and relative humidity in  /2003Monthly rainfall, mean temperature and relative humidity in  , and 2003Monthly rainfall, mean temperature and relative humidity in  /2004Monthly rainfall, mean temperature and relative humidity in   growing seasons, and long  term (1928Monthly rainfall, mean temperature and relative humidity in  -2001) )averages Georg Lembke KG, Germany); Emerald, English Giant, Hobson, and Interval (Advanta Seeds Ltd., UK); B. rapa var.rapa cultivars Lenox and Malvira (Hans-Georg Lembke KG, Germany), and Polybra (Mommersteeg International, The Netherlands) were used.Each plot consisted of 8 rows, 10 m long, with 17.5 cm row spacing.Seeding rate was 5 kg ha -1 for all cultivars.The experiments were sown with an oyjort experimental drill on 5 November 2002 and 24 October 2003.

Table 2 .
Effect of harvesting at two growth stages on the forage yield and quality of several cultivars of two Brassica species (average of two growingseasons, 2002/03 and 2003/04) a Orthogonal contrast B. napus var.oleifera vs B. rapa var.rapa.*, ** Significant at the 0.05 and 0.01 level.ns: not significant.

Table 3 .
Effect of harvesting at two growth stages on the crude protein content and yield in different plant components of several cultivars of two Brassica species (averages of 2 yr) a Orthogonal contrast B. napus var.oleifera vs B. rapa var.rapa.*,** significant at the 0.05 and 0.01 level.ns: not significant.