This pot experiment was conducted at Department of Soil and Water Science, Yezin Agricultural University during the dry and wet seasons in 2019 and the area is situated at latitude 19˚50'N, longitude 96˚16'E, 102 meters above sea level (Figure 1).

The experiment was set up as split plot design with four replications. The main plots were soil types; Shwebo soil and Yezin soil and sub plots were levels of silicon (Si). Five silicon levels were (0, 100, 150, 200 and 250 Si kg∙ha⁻¹). Yadanartoe rice variety (125 days) was cultivated. The soils were collected from 0 - 15 cm depth in Yezin and Shwebo lowland rice field. Plastic pots were used and filled with 10 kg soil. All treatments were applied with similar recommended dose of fertilizers (RDF) at the rates of 85-12-31 (N-P-K) kg∙ha⁻¹. Urea, Triple superphosphate (TSP) and Muriate of potash (MOP) were used as the sources of these nutrients. Urea was applied at recovery, maximum tillering and heading stages as three equal splits, TSP at basal and MOP at recovery and maximum tillering stages as two equal splits. The calcium silicate fertilizer was used as a silicon source and applied before transplanting. Twenty days old rice seedlings were transplanted with one rice seedling into each pot.

Physicochemical properties of soil samples were analyzed before sowing and

after harvest of the second season of rice. The soil sample was air-dried and ground to pass through a 2 mm sieve. The soil samples were analyzed to determine soil texture, soil pH, available nitrogen, phosphorus and potassium, acid extractable Si, organic matter %, and electrical conductivity. Plant samples were collected at harvest stage and analyzed for the total uptake of N, P, K and Si. Plant and Soil samples were analyzed at the Soil and Plant Analysis laboratory, Soil Science Research Section, Agricultural Natural Resources Management Research Division, Department of Agricultural Research (DAR), Yezin, Nay Pyi Taw.

Growth parameters such that plant height and numbers of tillers per hill were collected two weeks intervals from 14 days after transplanting until harvest. For total dry matter (TDM), the rice plant from each pot was taken after harvesting and dried in an oven at 70˚C for 48 hours. After that, oven dry weight was used and computed for dry matter yield. The number of panicles per hill, panicle length, number of spikelets per panicle, filled grain%, 1000 grain weight and harvest index were measured after harvest. The grain yield was determined from each pot and adjusted to 14% moisture content.

Experimental data were analyzed by analysis of variance (ANOVA) using statistix software (version 8). Mean value was compared using least significant differences (LSD) test at 5% probability level. Least significant difference is used to compare means of different treatments that have an equal number of replications.

Harvest index (HI)

The harvest index was calculated by dividing the economic yield (grain yield) by biological yield (grain + straw yield) [14].

Harvest Index = Economic yield/Biological yield

Nutrient uptake

Nutrient uptakes were calculated by using the following formula [15].

Nutrientuptake = ( Y G × N G ) / 100 + ( Y S × N S ) / 100

where, Y_G = grain yield (g∙pot⁻¹), Y_S = straw yield (g∙pot⁻¹), N_G = nutrient concentration in grain (%), and Ns = nutrient concentration in straw (%)

Agronomic Silicon Use Efficiency (SiUE)

The agronomic SiUE refers to an increase in rice yield (g∙pot⁻¹) per unit of Si applied [15].

AgronomicSiUE = Yield f − Yield 0 / Si

where, Yield_f = rice yield from Si application (g∙pot⁻¹), Yield₀ = rice yield from without Si application (g∙pot⁻¹), Si = amount of Si applied (g∙pot⁻¹).

Plant height in all treatments continuously increased from 14 DAT to 84 DAT (Figure 2(a), Figure 2(b)). Plant height was significantly different in all growth stages between two soils except at 56 DAT. Shwebo soil provided a higher plant height than Yezin soil except at 84 DAT. At 84 DAT, Yezin soil gave significantly greater plant height (124.7 cm) than Shwebo soil (119.5 cm) in dry season. In wet season, there was significant difference in plant height only at 56 DAT. According to results, the effect of soil types on plant height was observed in dry season. Among the silicon levels, there was no significant difference in plant height at all growth stages in dry season and a significant difference at 14 DAT in wet season (Figure 3(a), Figure 3(b)). The effect of silicon fertilizer application on plant height was not significantly different [16]. The interaction between two soils and different silicon levels was not found on plant height in both seasons.

The highly significant difference of tiller numbers per hill was observed between two soils at all growth stages in dry season and significant difference at 84 DAT in wet season (Figure 4(a), Figure 4(b)). Shwebo soil gave significantly higher

tiller numbers per hill than Yezin soil. This result may be due to the effect of soil texture and amount of organic matter content in two soils whereas Shwebo soil had high organic matter content and clay content than that of Yezin soil. In dry season, there was no significant effect of Si-fertilizer application on the number of tillers per hill but significant differences were observed at 56 and 70 DAT in wet season (Figure 5(a), Figure 5(b)). The highest tiller number was observed at Si 250 kg∙ha⁻¹ and the lowest tiller number was found in control treatment. The number of tillers per hill was significantly increased with the application of silicon fertilizer [17]. The beneficial role of Si fertilizer was found in number of tillers per hill [18]. In both season, number of tillers∙hill⁻¹ was not a significant interaction between two soils and different silicon levels.

Mean values of total dry matter was shown in Table 1. There was a significant difference in total dry matter (TDM) between the two soils during dry season and no significant difference during the wet season. The maximum TDM value was resulted from shwebo soil (147.2 g∙pot⁻¹) and the minimum TDM value was obtained from Yezin soil (119.9 g∙pot⁻¹). The Si-fertilizer application was not significantly different on TDM in the dry season but a significant difference at 5% level in the wet season. The highest TDM value (95.3 g∙pot⁻¹) was obtained from the application of Si 250 kg∙ha⁻¹ and the lowest TDM value (77.6 g∙pot⁻¹)

In each column, means followed by a same letters are not significantly different at LSD test 5% level. *Significant difference at 5% level, **Significant difference at 1% level, ns = non-significant difference, Si = Silicon (kg∙ha⁻¹).

was found in control treatment. The increased levels of silicon fertilizer were significant increase in the dry matter∙hill⁻¹ [19].

In both seasons, the number of panicles per hill was a significant difference between the two soils (Table 1). The higher panicle numbers per hill was obtained from Shwebo soil. There was no significant difference in the number of panicles per hill among different silicon levels. The supplemental Si application was not significantly different on the number of panicles per hill [17]. No significant difference of panicle number was found among different silicon levels [20]. No significant interaction was observed between two soils and different silicon levels.

The panicle length of yadanartoe rice variety was a highly significant difference at 1% level in dry season and at 5% level in wet season between two soils (Table 1). Yezin soil recorded significantly higher panicle length than Shwebo soil. There was no significant difference effect of Si-fertilizer application on panicle length at all growth stages in both seasons. Panicle length was not significant different effect by silicon levels [16]. No interaction was observed between two soils and Si application on the panicle length.

Number of spikelets per panicle was highly significant difference at 1% level in dry season and at 5% level in wet season between two soils (Table 1). The higher number of spikelets per panicle was observed in Yezin soil and the lower number of spikelets per panicle was recorded in Shwebo soil. This might be due to panicle length of yezin soil higher than Shwebo soil. Among the silicon levels, mean value of spikelets per panicle was not significantly different in dry season and was significant difference at 1% level in wet season. The maximum number of spikelets per panicle was obtained from the application of Si 250 kg∙ha⁻¹ followed by the application of Si 200 kg∙ha⁻¹. The lowest number of spikelet per panicle was recorded in no silicon fertilizer application.] The silicon fertilizer application was increased the number of spikelets per panicle [21] [22]. The interaction effect between two soils and different levels of silicon was not significantly different in the number of spikelets per panicle.

There was no significant difference of filled grain percent between two soils and different levels of silicon fertilizer application in both seasons (Table 2). The mean values were statistically similar among different treatments. According to the results, the effect of silicon fertilizer application on filled grain percent was not significantly different in two different soils. The mean effect of Si-fertilizer application on filled grain percent was not significant difference [23]. Interaction effect between two soils and silicon fertilizer levels was not found in filled grain percent.

Thousand grain weights showed no significant differences in two soil types, silicon levels and their interaction in both seasons (Table 2). Among the yield components, 1000 grain weight was less influenced by the treatment combinations because it is more or less genetically controlled characteristics. It is usually a stable varietal character and the management practice has less effect on its variation [24]. Interaction between the soil types and silicon levels was not statistically different.

The results showed that the effects of soil types were significantly different on grain yield in dry season and no significant differences in wet season (Table 2). Shwebo soil showed a significantly higher grain yield than that of Yezin soil. This different grain yield may be supported from the Shwebo soil which had

In each column, means followed by a same letters are not significantly different at LSD test 5% level. *Significant difference at 5% level, **Significant difference at 1% level, ns = non-significant difference, Si = Silicon (kg∙ha⁻¹)

higher number of panicles per hill than Yezin soil. There was no significant difference in the grain yield among the silicon levels in dry season but a significant difference at 5% level in wet season. The maximum grain yield was observed from the application of Si 250 kg∙ha⁻¹ and it was not significantly different with the application of Si 200 kg∙ha⁻¹. The lowest grain yield was recorded in Si 0. It can be assumed that the effect of silicon fertilizer application on grain yield was significantly influenced. Significantly highest grain yield of rice was recorded from the application of recommended dose of fertilizers along with silicon at 200 kg∙ha⁻¹ [18]. The interaction effect between the soil types and silicon levels on grain yield was not significant difference in both seasons. The application of silicon fertilizer was increased in grain yield of 0.82% - 6.73% over the control in dry season and 8.5% - 26.4% over the control in wet season.

There was significant difference of straw yield between two soils in dry season and no significant differences in wet season (Table 2). Shwebo soil produced higher straw yield than Yezin soil. In both seasons, the straw yield was not significantly influenced by different levels of silicon fertilizer application. The maximum straw yield (49.3 g∙pot⁻¹) was recorded from the application of Si 250 kg∙ha⁻¹ which did not statistically differ with other Si treatments (100, 150, and 200 kg Si∙ha⁻¹). The lowest straw yield (41.3 g∙pot⁻¹) was observed in contro treatment. Interaction effect between two factors was not found to be significant on straw yield.

Harvest index as affected by two soils was significantly different at 5% level in dry season although no significant difference was observed in wet season (Table 2). The maximum HI was produced by Shwebo soil in dry season. No significant response was observed among the silicon levels in the dry season. In wet season, there was significant difference at 5% levels of harvest index among the silicon levels. The application of Si 200 kg∙ha⁻¹ gave highest HI (0.50) whereas, application of Si 150 kgha⁻¹ and Si 0 kg∙ha⁻¹ showed lowest harvest index (0.47). The response of rice to Si application was increased in harvest index over the control [25].

Nutrient uptakes of yadanartoe rice variety after harvest were presented in Table 3.

In each column, means followed by a same letters are not significantly different at LSD test 5% level. *Significant difference at 5% level, **Significant difference at 1% level, ns = non-significant difference, Si = Silicon (kg∙ha⁻¹).

In dry season, N and Si uptakes were significantly influenced between two soils. Shwebo soil produced significantly highest N and Si uptake. There was no significant difference of Si-fertilizer application on total N, P K and Si uptake in dry season and a highly significant difference in P, K, and Si uptake in the wet season. The application of silicon at 250 kg∙ha⁻¹ was significantly highest total P uptake (0.30 g∙pot⁻¹) followed by the application of silicon at 200 kg∙ha⁻¹ (0.26 g∙pot⁻¹). The lowest P uptake was achieved in the control treatment. According to results, the silicon fertilizer application increased total P uptake by rice plants. This finding might be due to decreased phosphorus retaining capacity of soil and increased solubility of phosphorus by silicon fertilizer application [26]. The significant highest total K uptake (0.97 g∙pot⁻¹) was obtained from the application of silicon at 250 kg∙ha⁻¹ but which was statistically similar with Si 200 kg∙ha⁻¹ and Si 150 kg∙ha⁻¹ (0.86 and 0.87 g∙pot⁻¹ respectively). The lowest value of total K uptake (0.75 g∙pot⁻¹) was observed in control treatment. The application of silicon significantly increased total K uptake by rice plants [26]. Regarding of the Si uptake, the application of silicon at 250 kg∙ha⁻¹ was significantly different on total Si uptake (3.98 g∙pot⁻¹) which was statistically similar to other silicon levels. The lowest total Si uptake (2.66 g∙pot⁻¹) was observed from control treatment. The effect of silicon fertilizer application was influenced on silicon uptake by rice in this study. The application of silicon fertilizer significantly increased total uptake of N, P, K and Si [27].

In both seasons, there were no significantly different of silicon fertilizer application on two soil types, different silicon levels (Table 3). The interaction between the soil types and silicon levels on SiUE was not significant difference in both seasons. The effect of silicon fertilizer application was not significant response on SiUE [28].

The status of soil nutrients was analyzed at initial, and after harvest of the second season (Table 4). The physicochemical properties of Shwebo soil was sandy clay loam with slightly acid condition having soil pH at 6.40. According to the electrical conductivity test, the soil was non-saline condition. The amount of organic matter and available K were medium levels while the levels of available N and P were low. Acid extractable Si was found as high level. The physicochemical properties of Yezin soil was sandy loam with moderately acid and non-saline conditions having soil pH at 5.60 and EC at 0.03 dS∙m⁻¹, respectively. Soil organic matter and available N were found as low levels in this soil. The levels of available P and K were found to be medium. The amount of acid extractable Si was high level.

According to the soil analysis results after harvest of the second season, the

application of silicon fertilizer increased soil pH in both soils whereas the treatment without Si application remained as the initial status. The pH of Shwebo soil increased from slightly acid to neutral in all Si-treated soils. In Yezin soil, the pH increased from moderately acid to slightly acid in Si application rates at 100, 150 and 200 kg∙ha⁻¹ and to neutral in 250 kg Si∙ha⁻¹ application. The application of Si at 100, 150, 200 and 250 kg∙ha⁻¹ using calcium silicate fertilizer containing 30% CaO was equal to the application of lime at 382, 572, 763 and 954 lb∙ac⁻¹. The pH value of soil between 6 and 7 are usually adequate in available Si, while pH value below 6 may be deficient in Si depending on texture, and if pH value is 9.8, maximum adsorption of Si can occur [29]. Soil pH is one of the most important Si solubility in soil solution [30]. Available N value in Shwebo was decreased from low to very low condition and was not changed in Yezin soil. Available P values were maintained and available K status was induced compare to initial value in two soil types. Soil organic matter content in Shwebo soil was decreased and was not changed in Yezin soil. The status of acid extractable Si in two soils can be maintained as the initial value.