In order to study the effect of sulphur nutrition levels and time of application on seed yield and oil quality of canola, an experiment was undertaken at the Agronomy Research Farm, The University of Agriculture Peshawar (2013–2014). The experiment consisted of five sulphur levels (S1= 15, S2 = 30, S3 = 45, S4 = 60, S5 = 75; kg·ha^-1) and three different application times (AT1 = application at seedling growth stage, AT2 = application at bolting growth stage and AT3 = application at flowering growth stage). In addition, a control was maintained with each of the replicated treatments (no sulphur application was made). A randomized complete block design was used and replicated four times. Ammonium sulphate (NH₄SO₄) from the Fuji fertilizer private Ltd, Lahore, Pakistan was used as the source of sulphur. Canola seeds (Abasin-95) were provided by the Nuclear Institute for Food and Agriculture, Pakistan and sown in winter. 5 m × 3.2 m plots were made and replicated, consisting of 8 rows with 0.4 m row-to-row distance. Phosphorus (50 kg·ha^-1) was applied during sowing in the form of single super phosphate, and nitrogen (75 kg·ha^-1) was applied in the form of urea (half dose during sowing time and half dose during flowering stage) after subtracting the amount of nitrogen supplied through ammonium sulphate. After complete emergence, hand thinning was done at the four leaf stage, maintaining 50 plants per row. All other cultural practices, including irrigation, weeding and hoeing etc. were carried out uniformly in all the plots. An average soil analyses of the experimental site showed a sandy loam texture, pH (7.4), EC (0.204 dS·m^-1), bulk density (1.52 g·cm^-3), moisture (7.8%), organic matter (0.50 mg·kg^-1), phosphorous (0.45 mg·kg^-1) and low sulphur (0.035 mg·kg^-1) at 0–15 cm depth. Mean maximum, minimum temperature (^oC), humidity (%) and rainfall (mm) for the growing period of the canola crop are presented in Figure 1. The experiment was carried out from October 2013 to May 2014. Seed yield (kg·ha^-1) was determined in each of the replicated plots where the four central rows were harvested at maturity and dried under sun light, manually threshed, weighed and coverted to kg·ha^-1.

Biological yield (kg·ha^-1) was calculated after harvesting the four central rows in each of the replicated plots at harvest maturity of the canola crop, dried under sun light, weighed and converted in kg·ha^-1.

To determine the glucosinolate (μ mol g^-1), erucic acid (%), oil content (%), protein content (%), oliec acid (%) and linoliec acid (%) in the canola seed, a five (5) gram sample obtained from each treatment plot was analyzed by a Near Infra-red Reflectance Spectroscopy System, (TR-3657-C Model 6500) as described by Ahmad et al., (2015) at Oilseed Quality Laboratory, Crop Breeding Division, Nuclear Institute for Food and Agriculture Peshawar, Pakistan.

The recorded data was statistically analyzed according to the analysis of variance techniques used for randomized complete block design and least significant difference (LSD) was applied at a 5% level of significance (P ≤ 0.05) upon significant F-test through the procedure described by Jan et al., (2009).

During the growing season the temperature ranged from 11.8 ºC minimum to 25.8 ºC maximum (Figure 1). Relative humidity ranged from 52.5 to 72.4%, while total rainfall was recorded as 14.1 mm. It is clear that minimum and maximum temperature gradually decreased from October to January and then increased. Maximum rainfall was recorded in March (3.9mm) and April (3.8mm).

Seed is the ultimate output of a crop which determines the profitability of the crop production enterprise. Sulphur levels and time of sulphur application had significant effects on the seed yield of canola and their interaction was found to be non significant (Table 1). Sulphur application increased seed yield compared to the control. Higher seed yield (2452 kg·ha^-1) was recorded in plots where 60 kg·ha^-1 sulphur were applied as compared to the 15 kg·ha^-1 sulphur plots (1376 kg·ha^-1). These results are in agreement with Chauhan et al., (1996), who reported that sulphur increased the seed yield of rapeseed. The findings of Begum et al., (2012) revealed that sulphur up to 60 kg·ha^-1 application produced a higher seed yield of canola. Sulphur application timings had significantly affected the seed yield. More seed yield (1998 kg·ha^-1) was obtained at the bolting stage, while lower seed yield was obtained when sulphur was applied at seedling and bolting stages (1907 and 1922 kg·ha^-1, respectively). The crops grown in sulphur fertilized plots obtained higher seed yield (1942 kg·ha^-1), while lower seed yield (1134 kg·ha^-1) was recorded from the control plots. Canola seed yield was increased when sulphur was applied at the rate of 40 kg·ha^-1 (Varényiová et al., 2017). The findings of Malhi and Gill (2002); Malhi and Leach (2000) indicated that the seed yield of canola was lower when sulphur was applied at the flowering stage compared to that obtained when sulphur was applied at the bolting stage. Hocking et al., (1996) also found that sulphur applied at bolting resulted in the same yield although there was significant reduction in seed yield when sulphur application was delayed until the flowering stage. The interaction between S x AT revealed non significant effects on seed yield (Table 6).

The data revealed that sulphur levels have a significant effect on the biological yield of canola (Table 1). Increasing sulphur rates up to 60 kg·ha^-1 significantly increased the biological yield. Maximum biological yield (11461 kg·ha^-1) was obtained from 60 kg sulphur·ha^-1 followed by 75 kg sulphur·ha^-1 (Table. 3). A higher cofficient of variation (12.03) was seen in the biological yield (Table 7). Collectively, the highest biological yield (10574 kg·ha^-1) was recorded in the sulphur-treated plots compared to the control plots (8469 kg·ha^-1). The application timings and interaction of S x AT had no significant effect on biological yield. These results are comparable to those of Khandkar et al., (1991) in that biological yield was enhanced significantly with the application of sulphur levels. Malik et al., (2004) also reported that biological yield was significantly affected by sulphur levels.

The glucosinolate content of canola as influenced by various sulphur levels and application times is presented in Table 2. The statistical analysis of data showed a significant effect of sulphur levels on the glucosinolate of canola. Average values for the data showed that significantly higher glucosinolate (31.2 μmol·g^-1) was obtained from 75 kg sulphur·ha^-1 treated plots, whereas minimum glucosinolate (19.7 μmol·g^-1) was obtained from 15 kg sulphur·ha^-1 plots. Application times and interaction of sulphur x application times was found non significant (Table 5). The sulphur levels showed a positive influence on the glucosinolate content of canola. Mailer (1989) reported that sulphur was essential up to some level for normal plant growth and affected the glucosinolate content of canola. It was also elaborated that a sulphur deficiency lowered glucosinolate content in control plots. Wang et al., (1997) also determined that an increase in sulphur application enhanced the glucosinolate content of canola. The 60 kg·ha^-1 sulphur application enhanced the glucosinolate content of canola, the safe and acceptable limit, which is 30 μmol·g^-1. Sulphur is involved in the synthesis of chlorophyll and also required in Cruciferae for the synthesis of volatile oil which accumulates as glucosinolate (Marschner 1986 and Ahmad et al., 2007).

The mean values for the data showed that sulphur application enhanced erucic acid (1.6%) compared to the control (1.2%). Significantly higher erucic acid (2.1%) was recorded for 75 kg sulphur·ha^-1 (Table. 2). The application times and the interaction of sulphur x application times was found to be non significant for the erucic acid (%) of canola seeds. However, erucic acid gradually increased with increasing doses of sulphur (Table 2). Manaf and Fayyaz (2006) reported that eurcic acid was at its maximum when the highest dose of sulphur (25 kg·ha^-1) was applied. Non-essential fatty acids like palmitic, stearic and erucic acid were increased in the canola seed with decreased sulphur levels (Begum et al., 2015).

The sulphur levels significantly affected the oil content in canola seeds (Table 3). The comparison of sulphur fertilized plots with the control plots showed that the sulphur fertilized plots obtained significantly higher oil contents (44.1%) than the control plots (41.1%) (Table 3). The oil content was enhanced from 42.8 to 45.2% as the sulphur rate was boosted up to 60 kg·ha^-1. Kumar and Trivedi, (2012) recorded increasing oil percentages with increasing sulphur applications. However, a further sulphur application of up to 75 kg·ha^-1 did not enhance the oil content. Application timings and the interaction of S x AT had no significant affect on the oil content. Similar results were obtained in earlier studies by Mailer et al., (1989), who concluded that sulphur application enhanced the oil content of canola. The oil contents of canola were enhanced by sulphur application up to 30–50 kg·ha^-1 (Subhani et al., (2003); Malhi and Leach, 2000). The control plots showed reduced oil contents compared to the 60 kg·ha^-1 sulphur-applied plots. Higher oil content was in sulphur-applied plots, as oil contains many fatty acids and these fatty acid structures contain sulphur compounds (Malik et al., 2004).

The seed protein content (%) of canola is the basic parameter with respect to the quality. The statistical analysis of the data indicated that sulphur levels showed a positive influence on protein content (%) (Table 3). The 60 kg·ha^-1 sulphur application resulted in higher seed protein content (24.8%) in canola. Protein content was increased from 22.4% to 24.8% when sulphur levels increased from 15 to 60 kg·ha^-1, while protein content decreased (23.6%) when the sulphur level was further increased up to 75 kg·ha^-1. Canola seed protein and oil content were increased with increasing doses of sulphur (Kandil and Nadia, 2012). The crops grown without sulphur had lower protein contents (20.6%) than the sulphur-treated plots (23.4%). The time of application and interaction of S x AT had no significant effect on the protein content in the seeds (Table. 5). The sulphur application enhanced the seed protein contents of canola due to the protein content of rapeseed containing large amounts of amino acids such as cysteine and methionine which are constituents of sulphur (Holmes, 1980). These results confirm the findings of Rashid (1996) in that amino acids (methionine, cystine and cysteine) contain sulphur as an important structural part and therefore essential protein synthesis occurs due to sulphur. The results obtained regarding the response of sulphur application to the protein content of canola seed are also in line with the earlier findings of Wang et al., (1997) who stated that sulphur applied at up to 40 kg·ha^-1 increased the protein content of canola seeds.

The data on the oleic acid (%) of canola as affected by sulphur levels, application timings and their interaction are given in Table 4. The control plots resulted in less oleic acid (52.65%) as compared to the sulphur fertilized plots (55.82%). The statistical analysis of the data showed a non significant effect of sulphur levels and application timing on the oleic acid (%) of canola. Ahmed and Abdin (2000), concluded that sulphur levels had no significant effect on oleic acid. However, oliec acid (55.23% to 56.18%) was increased with increasing sulphur doses from 15–75 kg·ha^-1. Maximum oleic acid (229.6 mg·g^-1) was obtained from sulphur-treated plots as compared to the control (Shoja et al., 2018). Maximum oleica acid contents were observed at 60 kg·ha^-1 when applied at the bolting stage. Oleica acid was increased when sulphur doses were increased from 30–60 kg·ha^-1 (Ray et al., 2015).

Sulphur levels, time of application and their interaction have no significant effects on the linoliec acid of canola (Table 4). A lower linoleic acid content (9.5%) was found in the control plots than the plots fertilized with sulphur (10.5%). The results are in line with those of Malhi and Gill (2002), and Malhi and Leach (2000). Increasing doses of sulphur from 15–60 kg·ha^-1 increased linoliec acid (10.3–10.8). However, further a increase beyond 60 kg sulphur ha^-1 decreased the linoliec acid content. Ray et al., (2015) reported that increasing sulphur doses from 30–60 kg·ha^-1 increased linoleic acid percentage. Manaf and Hassan 2006 concluded that linolienic acid was increased when sulphur fertilization increased from 15–25 kg·ha^-1. Sulphur applied at seedling and bolting stage induced more linoliec acid (10.5) compared to flowering stage.