Assessment of Optimal Seeding Rate for Fine and Coarse Rice Varieties Using the Direct Seeded Rice (DSR) Method

Abstract

Rice (Oryza sativa L.) is one of the most crucial cereal crops worldwide, serving as a staple food for a significant portion of the global population. Rice is the second most important staple food crop in Pakistan after wheat, and it is also a major export commodity. Concerning this, the current study aimed to evaluate the effects of different seed rates on the yield and yield-contributing parameters of rice varieties. The experiment was conducted over two consecutive kharif summer seasons, from 2020–21 and 2021–22, at the Pakistan Agricultural Research Council (PARC) Rice Program experimental area in Kala Shah Kaku, Lahore, Pakistan, by following a factorial randomized complete block design with three replications using coarse rice (KSK-133) and fine rice (Super Basmati) varieties. Different seed rates, including 27 kg/ha, 22 kg/ha, 17 kg/ha, and 12 kg/ha, were tested during the experiment. Different growth and yield-related attributes, such as plant height (cm), the number of productive tillers per plant, panicle length (cm), the number of grains per panicle, and grain yield (m⁻²), were recorded. The results showed that for KSK-133 and Super Basmati, the maximum grain yield was achieved at a sowing rate of 27 kg/ha in direct seed rice (DSR). The lowest yield was observed at a seeding rate of 12 kg/ha for KSK-133 and Super Basmati in DSR. Both basmati (Super Basmati) and coarse-grain (KSK-133) varieties exhibited similar responses to seed rate treatments, with the optimal performance observed at the highest seed rate of 27 kg/ha for both seasons. Grains per panicle and thousand grain weight emerged as critical determinants of yield, highlighting the need to balance vegetative growth with reproductive development. Breeding programs should focus on developing varieties that balance vegetative traits like tiller production and panicle length with reproductive traits to enhance overall yield. Based on these findings, it is concluded that using an optimal seeding rate of 27 kg/ha for direct-seeded fine and coarse rice varieties is beneficial in terms of tillers and higher yield.

1. Introduction

Rice (Oryza sativa L.) is one of the most crucial cereal crops worldwide, serving as a staple food for a significant portion of the global population [1]. In Pakistan, rice is not only a staple food crop after wheat; a huge quantity is exported as well. Globally, more than half of the world’s population ingests rice as a food, particularly in Asia, where rice consumption is highest [2]. According to the U.S. Department of Agriculture (USDA), about 45% of Pakistan’s rice production is consumed locally, while the rest is exported. This makes rice an essential component of the country’s agricultural export economy [2,3]. Achieving optimal rice production is essential for ensuring food security and meeting the nutritional needs of growing populations [4]. Rice cultivation in Pakistan remains a promising sector, especially in the 2023–24 season, where the crop area expanded significantly by 22.2%, reaching 3.6 million hectares from 3.0 million hectares. This expansion, coupled with favorable monsoon rains, higher rice prices, and better export prospects, led to a substantial increase in rice production by 34.8%, resulting in 9.9 million tons compared to 7.3 million tons in the previous year. Rice now contributes 0.6% to the GDP and 2.5% to agriculture value addition, underscoring its economic importance [5].

Direct Seeded Rice (DSR) has emerged as a crucial method in modern rice farming, addressing challenges like water scarcity and labor shortages. Research indicates that varying seed rates in DSR can significantly impact weed control, plant growth, and yield outcomes. For instance, increasing seed rates can suppress weed growth, reducing yield losses due to competition [6]. Studies have shown that both hybrid and inbred rice varieties achieved the highest grain yields at lower seeding rates of 80 seeds per square meter, with higher rates leading to reduced nitrogen efficiency and lower yields [7].

Optimal seeding strategies have been found to improve canopy growth uniformity, reduce disease incidence, and minimize lodging, resulting in higher yields [8]. These findings highlight the importance of integrating optimal fertilization and seeding rates in DSR systems to maintain stable yields and enhance crop resilience under changing agricultural conditions [9,10]. Specifically, a seed rate of 27 kg/ha has been identified as beneficial for maximizing productive tillers and yield in both fine and coarse rice varieties, like IR-6 and Super Basmati [11].

The KSK-133 and Super Basmati are two prominent rice varieties cultivated in Pakistan. KSK-1333 is a high-yielding rice variety developed by the Rice Research Institute (RRI), Kala Shah Kaku, Pakistan. It is renowned for its medium grain size, excellent cooking quality, and resilience to environmental stress and disease. KSK-133 is popular for domestic consumption and is also exported to several countries, contributing to Pakistan’s rice export sector [12]. Super Basmati, on the other hand, is celebrated for its unique aroma, soft texture, and long grains [13]. Primarily grown in the Punjab region, it is considered among the finest basmati rice varieties globally. Its quality makes it a premium product in international markets, particularly in regions like the Middle East, Europe, and North America [14]. Both KSK-133 and Super Basmati play a critical role in Pakistan’s rice production, significantly contributing to the country’s economy. In 2023, Pakistan exported more than 4.5 million metric tons of rice, with an export value surpassing 2.5 billion USD [15]. These rice varieties remain favorites among local farmers for their yield potential and market value.

The objective of this study is to evaluate the impact of different seed rates on the growth, yield, and yield-contributing parameters in coarse and fine rice varieties under Direct Seeded Rice (DSR) cultivation methods. The study aims to identify optimal seeding rates that maximize grain yield. Additionally, the research seeks to provide insights that can inform integrated crop and weed management strategies, contributing to the sustainability and productivity of DSR in the context of evolving agricultural challenges such as water scarcity, labor shortages, and climate change.

2. Materials and Methods

2.1. Experimental Site

The trials were conducted over two consecutive kharif seasons (May–September), 2020–21 and 2021–22, at the Pakistan Agricultural Research Council (PARC) Rice Program experimental area in Kala Shah Kaku, Lahore, Pakistan.

2.2. Experimental Design and Treatments

The experiments employed a factorial randomized complete block design with three replications, focusing on coarse-grain rice (KSK-133) and fine-grain rice (Super Basmati) varieties. Seed rates of 11, 9, 7, and 5 kg/acre were tested on plots measuring 27 × 9 m.

2.3. Crop Husbandry

For weed control in the Direct Seeded Rice (DSR) method, pendimethalin 30% EC was applied at a rate of 2.47 L/ha to saturated soil as a pre-emergence herbicide, which was followed by the application of Pyranex Gold 30WDG at 296 g/ha, 25 days after seeding, under similar soil conditions. A standard fertilizer dose of NPK 121–89–59 kg/ha (in the form of Urea, Diammonium phosphate (DAP), and Sulphate of potash (SOP), respectively) was administered with the full dose of phosphorus and potash, along with one-third of the nitrogen, applied at sowing. The remaining nitrogen was split into two applications at 35 and 55 days after seeding. Irrigation was provided every 4–6 days, from seeding until maturity, with water withheld 25 days before harvesting. Standard pest and disease management practices were followed, and the crop was mechanically harvested at full maturity.

2.4. Data Recording

Five rice plants were randomly selected from each plot to count tillers per plant at the physiological maturity stage. Plant height in centimeters (cm) was measured from base to tip with a measuring meter rod, and panicle length (cm) was measured from the base of the panicle to the tip. Thousand grain weight, in grams, was weighed using a digital weight scale. Grains per panicle were counted to evaluate the grain productivity of individual panicles. The yield in grams per square meter (g/m²) was calculated by harvesting and weighing grains to evaluate overall productivity.

2.5. Statistical Analysis

Analysis of variance (ANOVA) was performed in Statistix 10.1, and correlation was performed in the available open-access software R Studio v4.2.0 (Agricolae package of R version). Pearson’s Correlation coefficient matrix and graph were performed using the ‘Metan’ (Multi-Environment Trial Analysis) package in R Studio. The comparison of treatment means was conducted using the least significant difference (LSD) test at a probability level of 0.05 [16].

3. Results

3.1. Analysis of Variance (ANOVA)

The results of the ANOVA (Analysis of Variance) for the seasons 2020–21 and 2021–22 growing seasons provide a clear insight into the variation observed in the studied traits across different rice varieties and the effectiveness of treatments. Statistically evaluated data are presented in Table 1, Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8. The results highlight that varietal differences played a major role in determining the performance of key traits such as plant height, tillers per plant, panicle length, grains per panicle, thousand grain weight, and grain yield. The treatments applied also had a significant impact, especially on traits like tiller production, grain number, and yield. The interaction effects between treatments and varieties were significant for some traits in the 2021–22 season, indicating that the response to treatments may vary depending on the rice variety. The low coefficient of variation (CV) values across traits suggest a high level of precision in the experiment, ensuring the reliability of the results.

3.2. Season 2020–21

Plants Traits and Their Correlation

Plant height showed a non-significant variation across all seed rate treatments, but between varieties, it showed a highly significant variation (p ≤ 0.001), and its interactions with treatment and varieties were also highly significant (p ≤ 0.001) (Table 1). The basmati variety showed the highest mean value (132.5 cm) of plant height under the highest seed rate treatment, which was 27 kg/ha, followed by 9 kg/acre (131.57 cm), and the minimum mean value (129.87 cm) of plant height was recorded in a 5 kg/acre seed rate treatment. But the coarse variety showed the highest mean value (117 cm) of plant height under the lowest seed rate treatment, which was 5 kg/acre, followed by 7 kg/acre (131.57 cm), and the minimum mean value (114 cm) of plant height was recorded in a 27 kg/ha seed rate treatment (Table 2). Plant height had shown a highly significant negative correlation with grain yield, thousand grain weight, and grains per panicle (−0.91, −0.95, and −0.76), respectively, but it also showed a highly significant positive correlation with tillers per plant (0.72) and panicle length (0.77 cm) (Table 3 and Figure 1a). The direct effect of plant height on grain yield was negative, with a coefficient value of (−1.78) and an indirect effect through tillers per plant, and panicle length was also negative, with coefficient values of (−1.34 and −1.53), but it had a positive indirect effect through grains per panicle and thousand grain weight (1.46 and 1.74), respectively (Table 4).

Tillers per plant showed a highly significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and its varieties was significant (p ≤ 0.05) (Table 1). The basmati variety showed the highest mean value (28) of tillers per plant under the highest seed rate treatment, which was 27 kg/ha, followed by 9 kg/acre (27), and the minimum mean value (23) of plant height was recorded in a 5 kg/acre seed rate treatment. The coarse variety showed the highest mean value (24) of tillers per plant under the highest seed rate treatment, which was 27 kg/ha, followed by 9 kg/acre (22), and a minimum mean value (19) of tillers per plant was recorded in a 5 kg/acre seed rate treatment (Table 2).

Panicle length showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and its varieties was significant (p ≤ 0.05) (Table 1). The basmati variety showed the highest mean value (29.75 cm) of panicle length under the lowest seed rate treatment, which was 12 kg/ha, followed by 7 kg/acre (28.06 cm), and a minimum mean value (26.33 cm) of panicle length was recorded in 9kg/acre followed by 27 kg/ha, having a mean value of (26.5) for seed rate treatments. The coarse variety showed the highest mean value (24.8 cm) of panicle length under seed rate treatment of 7 kg/acre, followed by 12 kg/ha (24.23 cm), and the minimum mean value (23.6 cm) of panicle length was recorded in 09kg/acre, followed by 27 kg/ha, having a mean (23.8 cm) seed rate treatment (Table 2). Panicle length showed a highly significant negative correlation with grain yield, thousand grain weight, and grains per panicle (−0.86, −0.76, and −0.45), respectively, but showed a highly significant positive correlation with plant height (0.77) and tillers per plant (0.38) (Table 3 and Figure 1a). The direct effect of panicle length on grain yield was negative, with a coefficient value of (−1.45), and the indirect effect, through plant height and tillers per plant, was also negative, with coefficient values of (−1.25 and −0.63), but it had a positive indirect effect through grains per panicle and thousand grain weight (0.81 and 1.26), respectively (Table 4).

Grains per panicle showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and its varieties was significant (p ≤ 0.05) (Table 1). The basmati variety showed the highest mean value (85) of grains per panicle under the lowest seed rate treatments, which were 12 kg/ha and 7 kg/acre, followed by 27 kg/ha (76) and a minimum mean value (73) of grains per panicle was recorded in 9 kg/acre (Table 2). The coarse variety showed the highest mean value (115) of grains per panicle under a seed rate treatment of 7 kg/acre, followed by 12 kg/ha (101), and a minimum mean value (91) of grains per panicle was recorded in 9 kg/acre, followed by 27 kg/ha, having a mean (99) seed rate treatment (Table 2). Grains per panicle showed a highly significant positive correlation with grain yield and thousand grain weight (0.65 and 0.84), respectively, but showed a highly significant negative correlation with plant height (−0.76), tillers per plant (−0.77), and panicle length (−0.45) (Table 3 and Figure 1a). The direct effect of grains per panicle on grain yield was positive, with a coefficient value of (0.58), and the indirect effect through plant height, tillers per plant, and panicle length, was negative, with coefficient values of (−0.48, −0.49, and −0.32), respectively, but they had a positive indirect effect through thousand grain weight (0.49) (Table 2 and Table 4).

Thousand grain weight showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and varieties was significant (p ≤ 0.05) (Table 1). The basmati variety showed the highest mean value (22.7 g) of thousand grain weight under a 17 kg/ha seed rate treatment, followed by a 12 kg/ha (22.2 g), and a minimum mean value (21.8) of thousand grain weight was recorded in 9 kg/acre, followed by 27 kg/ha (22 g) (Table 2). The coarse variety showed the highest mean value (29 g) of thousand grain weight under a seed rate treatment of 9 kg/acre, followed by 17 kg/ha (28.67 g), and a minimum mean value (27.6 g) of thousand grain weight was recorded in 27 kg/ha, followed by 12 kg/ha, having a mean of (28.4) under a seed rate treatment (Table 2). Thousand grain weight showed a highly significant positive correlation with grain yield and grains per panicle (0.87 and 0.84), respectively, but showed a highly significant negative correlation with plant height (−0.95) and tillers per plant (−0.77) and panicle length (−0.76) (Table 3 and Figure 1a). The direct effect of thousand grains weight on grain yield was negative, with a coefficient value of (−2.88), and the indirect effect, through plant height, tillers per plant, and panicle length, was positive, with coefficient values of (2.82, 2.32, and 2.5), respectively, but they had a negative indirect effect through thousand grain weight (−2.88) (Table 4).

Grain yield showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and varieties was significant (p ≤ 0.05) (Table 1). The basmati variety showed the highest mean value (403.94 g) of grain yield/m² under a 27 kg/ha seed rate treatment, followed by 22 kg/ha (388.51 g), and the minimum mean value (335.94 g) of grain yield/m² was recorded in 12 kg/ha, followed by 17 kg/ha (371.71 g). The coarse variety showed the highest mean value (583.06 g) of grain yield/m² under a 27 kg/ha seed rate treatment, followed by 09 kg/acre (525.89 g), and a minimum mean value (475.43 g) of grain yield/m² was recorded in 12 kg/ha, followed by 17 kg/ha (505.81 g) (Table 2). Grain yield/m² showed a highly significant positive correlation with thousand grain weight and grains per panicle, (0.87 and 0.65), respectively, but showed a highly significant negative correlation with plant height (−0.91), tillers per plant (−0.45), and panicle length (−0.86) (Table 3 and Figure 1a). Plant height (−1.78), tillers per plant (−0.32), panicle length (−1.45), and thousand grain weight (−2.88) had a negative direct effect on grain yield/m², and grains per panicle had a positive direct effect on grain yield/m² (0.58). Plant height, tillers per plant, and panicle length had a positive indirect effect through thousand grain weight with coefficient values of (2.82, 2.32, and 2.5), respectively. Grains per panicle and thousand grain weight had a positive indirect effect on grain yield/m² through the plant heights of (1.46 and 1.74), tillers per plant of (0.27 and 0.26), the panicle lengths of (0.81 and 1.26), and the grains per panicle of (0.58), respectively (Table 4).

Table 1. ANOVA of all traits (Season 2020–21).

Source	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
Replication	2.00	27.12	0.9472	51.26	0.225	30.00
Treatment	0.51 ns	2480.2 **	5.8059 **	348.99 **	0.814 *	8057 **
Varieties	1366.55 **	11,201.8 **	75.0834 **	2845.99 **	233.75 **	130,577 **
Treatments × Varieties	12.9 **	98.9 *	2.7073 *	56.88 *	0.704 *	673 **
Error	1.49	30.2	0.9489	12.41	0.167	42
Grand mean	123.46	237.6	25.894	90.527	25.296	448.79
CV	0.99	2.31	3.76	3.89	1.61	1.44

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01; ns—non-significant.

Table 1. ANOVA of all traits (Season 2020–21).

Source	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
Replication	2.00	27.12	0.9472	51.26	0.225	30.00
Treatment	0.51 ns	2480.2 **	5.8059 **	348.99 **	0.814 *	8057 **
Varieties	1366.55 **	11,201.8 **	75.0834 **	2845.99 **	233.75 **	130,577 **
Treatments × Varieties	12.9 **	98.9 *	2.7073 *	56.88 *	0.704 *	673 **
Error	1.49	30.2	0.9489	12.41	0.167	42
Grand mean	123.46	237.6	25.894	90.527	25.296	448.79
CV	0.99	2.31	3.76	3.89	1.61	1.44

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01; ns—non-significant.

Table 2. Mean values and homogeneous groups of all traits under all treatments (Season 2020–21).

Treatments	Varieties	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
27 kg/ha	Basmati	132.50 A	28 A	26.50 BC	76 E	22.00 CD	403.94 E
22 kg/ha	Basmati	131.57 AB	27 A	26.33 C	73 E	21.80 D	388.51 F
17 kg/ha	Basmati	130.10 B	26 B	28.07AB	85 D	22.70 C	371.71 G
12 kg/ha	Basmati	129.87 B	23 D	29.75 A	85 D	22.20 CD	335.94 H
27 kg/ha	Coarse	114.00 E	24 C	23.83 D	99 B	27.60 B	583.06 A
22 kg/ha	Coarse	115.00 DE	22 D	23.60 D	91 C	29.00 A	525.89 B
17 kg/ha	Coarse	117.67 C	21E	24.833 CD	115 A	28.66 A	505.81 C
12 kg/ha	Coarse	117.00 CD	19 F	24.23 D	101 B	28.40 A	475.43 D
S.E.		0.99	2.24	0.79	2.87	0.34	5.27
CV		2.14	4.81	1.70	6.16	0.72	11.30
CTV		2.15	2.14	2.14	2.14	2.15	2.145

The different letters in the same column indicate differences among treatments. Basmati—Super Basmati; Coarse—KSK-133; S.E—Standard Error for Comparison; CV—Critical Value for Comparison; CTV—Critical T Value.

Table 2. Mean values and homogeneous groups of all traits under all treatments (Season 2020–21).

Treatments	Varieties	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
27 kg/ha	Basmati	132.50 A	28 A	26.50 BC	76 E	22.00 CD	403.94 E
22 kg/ha	Basmati	131.57 AB	27 A	26.33 C	73 E	21.80 D	388.51 F
17 kg/ha	Basmati	130.10 B	26 B	28.07AB	85 D	22.70 C	371.71 G
12 kg/ha	Basmati	129.87 B	23 D	29.75 A	85 D	22.20 CD	335.94 H
27 kg/ha	Coarse	114.00 E	24 C	23.83 D	99 B	27.60 B	583.06 A
22 kg/ha	Coarse	115.00 DE	22 D	23.60 D	91 C	29.00 A	525.89 B
17 kg/ha	Coarse	117.67 C	21E	24.833 CD	115 A	28.66 A	505.81 C
12 kg/ha	Coarse	117.00 CD	19 F	24.23 D	101 B	28.40 A	475.43 D
S.E.		0.99	2.24	0.79	2.87	0.34	5.27
CV		2.14	4.81	1.70	6.16	0.72	11.30
CTV		2.15	2.14	2.14	2.14	2.15	2.145

The different letters in the same column indicate differences among treatments. Basmati—Super Basmati; Coarse—KSK-133; S.E—Standard Error for Comparison; CV—Critical Value for Comparison; CTV—Critical T Value.

Table 3. Pearson’s Correlation (Season 2020–21).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Tillers/Plant	0.72 **	-
Panicle length	0.77 **	0.38 **	-
Grains/panicle	−0.76 **	−0.77 **	−0.45 *	-
Thousand grain weight	−0.95 **	−0.77 **	−0.76 **	0.84 **	-
Grain yield/m2	−0.91 **	−0.45 *	−0.86 **	0.65 **	0.87 **

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01.

Table 3. Pearson’s Correlation (Season 2020–21).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Tillers/Plant	0.72 **	-
Panicle length	0.77 **	0.38 **	-
Grains/panicle	−0.76 **	−0.77 **	−0.45 *	-
Thousand grain weight	−0.95 **	−0.77 **	−0.76 **	0.84 **	-
Grain yield/m2	−0.91 **	−0.45 *	−0.86 **	0.65 **	0.87 **

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01.

Table 4. Direct (diagonal) and indirect effects of parameters on grain yield per m² (Season 2020–21).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Plant Height	−1.78	−0.24	−1.25	−0.48	2.82
Tillers/Plant	−1.34	−0.32	−0.63	−0.49	2.32
Panicle length	−1.53	−0.14	−1.45	−0.32	2.50
Grains/panicle	1.46	0.27	0.81	0.58	−2.44
Thousand grain weight	1.74	0.26	1.26	0.49	−2.88

Residual effect^2 = 0.05224989.

Table 4. Direct (diagonal) and indirect effects of parameters on grain yield per m² (Season 2020–21).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Plant Height	−1.78	−0.24	−1.25	−0.48	2.82
Tillers/Plant	−1.34	−0.32	−0.63	−0.49	2.32
Panicle length	−1.53	−0.14	−1.45	−0.32	2.50
Grains/panicle	1.46	0.27	0.81	0.58	−2.44
Thousand grain weight	1.74	0.26	1.26	0.49	−2.88

Residual effect^2 = 0.05224989.

3.3. Season 2021–22

Plants Traits and Their Correlation

Plant height showed a highly significant variation across all seed rate treatments and also between varieties (p ≤ 0.001), and its interaction with the treatments and varieties was also highly significant (p ≤ 0.001) (Table 5). The basmati variety showed the highest mean value (127.37 cm) of plant height under the second highest seed rate treatment, which was 22 kg/ha, followed by 17 kg/ha (127.07 cm), and a minimum mean value (126.27 cm) of plant height was recorded in both remaining 27 kg/ha and 12 kg/ha seed rate treatments (Table 6). But the coarse variety showed the highest mean value (125.33 cm) of plant height under the lowest seed rate treatment, which was 5 kg/acre, followed by 7 kg/acre (125.27 cm), and a minimum mean value (119.47 cm) of plant height was recorded in both remaining 27 kg/ha and 22 kg/ha seed rate treatments (Table 6). Plant height showed a highly significant negative correlation with grain yield, thousand grain weight, and grains per panicle (−0.939, −0.698, and −0.733), respectively, but showed a highly significant positive correlation with tillers per plant (0.509) and panicle length (0.826 cm) (Table 7 and Figure 1b). The direct effect of plant height on grain yield was negative, with a coefficient value of (−0.094), and the indirect effect, through tillers per plant and panicle length, was also negative, with coefficient values (−0.053 and −0.084), but it had a positive indirect effect through grains per panicle and thousand grain weight (0.073 and 0.068), respectively (Table 8).

Tillers per plant showed a highly significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatments and varieties was significant (p ≤ 0.05) (Table 5). The basmati variety showed the highest mean value (29) of tillers per plant under the highest seed rate treatment, which was 27 kg/ha, followed by 9 kg/acre (28), and a minimum mean value (24) of plant height was recorded in a 5 kg/acre seed rate treatment. The coarse variety showed the highest mean value (23) of tillers per plant under the highest seed rate treatment, which was 27 kg/ha, and a minimum mean value (19) of tillers per plant was recorded in all remaining seed rate treatments (Table 6). Tillers per plant showed a highly significant negative correlation with grain yield, thousand grain weight, and grains per panicle (−0.59, −0.872, and −0.871), respectively, but showed a highly significant positive correlation with plant height (0.509) and panicle length (0.724) (Table 7 and Figure 1b). The direct effect of tillers per plant on grain yield was negative, with a coefficient value of (−0.37), and the indirect effect, through plant height and panicle length, was also negative, with coefficient values (−0.21 and −0.27), but it had a positive indirect effect through grains per panicle and thousand grain weight (0.33), respectively, Table 8).

Panicle length showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and its varieties was also highly significant (p ≤ 0.001) (Table 1). The basmati variety showed the highest mean value (27.8 cm) of panicle length under the lowest seed rate treatment, which was 12 kg/ha, followed by 27 kg/ha (27.4 cm), and a minimum mean value (27 cm) of panicle length was recorded in both remaining 9 kg/acre and 7 kg/acre seed rate treatments (Table 6). The coarse variety showed the highest mean value (25.8 cm) of panicle length under the seed rate treatment of 7 kg/acre, followed by 12 kg/ha (25.4 cm), and a minimum mean value (24.4 cm) of panicle length was recorded in 9 kg/acre, followed by 27 kg/ha, having a mean (24.6 cm) seed rate treatment (Table 6). Panicle length showed a highly significant negative correlation with grain yield, thousand grain weight, and grains per panicle (−0.93, −0.921, and −0.919), respectively, but showed a highly significant positive correlation with plant height (0.509) and tillers per plant (0.724) (Table 7 and Figure 1b). The direct effect of panicle length on grain yield was negative, with a coefficient value of (−0.845), and the indirect effect, through plant height and tillers per plant, was also negative, with coefficient values of (−0.747 and −0.634), but it had a positive indirect effect through grains per panicle and thousand grain weight (0.787 and 0.789), respectively, Table 8).

Grains per panicle showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatment and its varieties was also significant (p ≤ 0.001), but its interaction with treatment and varieties was not significant (p ≤ 0.05) (Table 5). The basmati variety showed the highest mean value (76) of grains per panicle under 12 kg/ha and 27 kg/ha, followed by 9 kg/acre (74), and a minimum mean value (75) of grains per panicle was recorded in 7 kg/acre. The coarse variety showed the highest mean value (121) of grains per panicle under a seed rate treatment of 9 kg/acre, followed by 17 kg/ha (118), and a minimum mean value (109) of grains per panicle was recorded in 27 kg/ha, followed by 12 kg/ha, having a mean (121) seed rate treatment (Table 6).

Grains per panicle had a significant positive correlation with grain yield and thousand grain weight (0.849 and 0.992), respectively, but showed a highly significant negative correlation with plant height (−0.733), tillers per plant (−0.871), and panicle length (−0.919) (Table 7 and Figure 1b). The direct effect of grains per panicle on grain yield was positive, with a coefficient value of (2.67), and the indirect effect, through plant height, tillers per plant, and panicle length was negative, with coefficient values (−2.1, −2.44, and −2.21), respectively, but grains per panicle had a positive indirect effect through thousand grain weight (2.69) (Table 8).

Thousand grain weight showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), but its interaction with the treatments and varieties was not significant (p ≤ 0.05) (Table 5). The basmati variety showed the highest mean value (23.7 g) of thousand grain weight under a 22 kg/ha seed rate treatment, followed by 17 kg/ha (23.5 g), and a minimum mean value (23.1) of thousand grain weight was recorded in 11kg/acre, followed by 12 kg/ha (23.2 g). The coarse variety showed the highest mean value (28.2 g) of thousand grain weight under the seed rate treatment of 9 kg/acre, followed by 17 kg/ha (28.1 g), and a minimum mean value (27.4 g) of thousand grain weight was recorded in 12 kg/ha, followed by 27 kg/ha, having a mean value of (27.6) for seed rate treatment (Table 6). Thousand grain weight showed a highly significant positive correlation with grain yield and grains per panicle (0.83 and 0.992), respectively, but showed a highly significant negative correlation with plant height (−0.698), tillers per plant (−0.872), and panicle length (−0.921) (Table 7 and Figure 1b). The direct effect of thousand grain weight on grain yield was negative, with a coefficient value of (−2.4), and the indirect effect, through plant height, tillers per plant, and panicle length, was positive, with coefficient values of (1.73, 2.14, and 2.22), respectively, but it had a negative indirect effect through thousand grain weight (−2.38) (Table 8).

Grain yield showed a significant variation across all seed rate treatments and varieties (p ≤ 0.001), and its interaction with the treatments and varieties was also significant (p ≤ 0.001) (Table 5). The basmati variety showed the highest mean value (264.04 g) of grain yield/m² under a 27 kg/ha seed rate treatment, followed by 22 kg/ha (255.98 g), and a minimum mean value (244.11 g) of grain yield/m² was recorded in 12 kg/ha followed by 17 kg/ha (251.56 g) (Table 6). The coarse variety showed the highest mean value (436.16 g) of grain yield/m² under 27 kg/ha seed rate treatment, followed by 22 kg/ha (401.87 g), and a minimum mean value (316.84 g) of grain yield/m² was recorded in 12 kg/ha, followed by 17 kg/ha (320.40 g) (Table 6). Grain yield/m² showed a highly significant positive correlation with thousand grain weight and grains per panicle (0.83 and 0.849), respectively, but it also showed a highly significant negative correlation with plant height (−0.939), tillers per plant (−0.59), and panicle length (−0.93). Plant height (−0.094), tillers per plant (−0.37), panicle length (−0.845), and thousand grain weight (−2.377) had a negative direct effect on grain yield/m2, and grains per panicle had a positive direct effect on grain yield/m² (2.67) (Table 7 and Figure 1b). Plant height, tillers per plant, and panicle length had a positive indirect effect through thousand grain weight, with coefficient values of (1.73, 2.14, and 2.22), respectively. Grains per panicle and thousand grain weight had a positive indirect effect on grain yield/m² through plant height (0.072 and 0.068), respectively, tillers per plant (0.335 and 0.333), respectively, panicle length (0.787 and 0.789), respectively, and grains per panicle (2.699) (Table 8).

Table 5. ANOVA of all traits (Season 2021–22).

Source	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
Replication	0.686	220.9	0.0079	3.26	0.063	6.7
Treatments	113.97 **	29,952.8 **	30.375 **	9707.45 **	118.148 **	79,194.2 **
Varieties	16.587 **	2182.7 **	0.9894 **	43.80 **	0.665 **	6813.5 **
Treatments × Varieties	18.39 **	509.6 *	0.7694 **	42.64 **	0.046 ns	4065.7 **
Error	0.97	134	0.0422	3.03	0.023	57.1
Grand mean	124.56	236.39	26.158	95.31	25.602	311.37
CV	0.79	4.9	0.79	1.83	0.6	2.43

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01; ns—non-significant.

Table 5. ANOVA of all traits (Season 2021–22).

Source	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
Replication	0.686	220.9	0.0079	3.26	0.063	6.7
Treatments	113.97 **	29,952.8 **	30.375 **	9707.45 **	118.148 **	79,194.2 **
Varieties	16.587 **	2182.7 **	0.9894 **	43.80 **	0.665 **	6813.5 **
Treatments × Varieties	18.39 **	509.6 *	0.7694 **	42.64 **	0.046 ns	4065.7 **
Error	0.97	134	0.0422	3.03	0.023	57.1
Grand mean	124.56	236.39	26.158	95.31	25.602	311.37
CV	0.79	4.9	0.79	1.83	0.6	2.43

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01; ns—non-significant.

Table 6. Mean values and homogeneous groups of all traits under all treatments (Season 2021–22).

Treatments	Varieties	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
27 kg/ha	Basmati	126.27 AB	29 A	27.4 B	76 D	23.1 D	264.04 D
22 kg/ha	Basmati	127.37 A	28 AB	27 C	74 D	23.7 C	255.98 DE
17 kg/ha	Basmati	127.07 A	27 B	27 C	75 D	23.5 C	251.56 DE
12 kg/ha	Basmati	126.27AB	24 C	27.8 A	76 D	23.2 D	244.11 E
27 kg/ha	Coarse	119.47 C	23 C	24.6 F	114 B	27.6 B	436.14 A
22 kg/ha	Coarse	119.47 C	19 D	24.4 F	121 A	28.2 A	401.87 B
17 kg/ha	Coarse	125.27 B	19 D	25.8 D	118 A	28.1 A	320.40 C
12 kg/ha	Coarse	125.33 B	19 D	25.4 E	109 C	27.4 B	316.84 C
S.E.		0.80	9.45	0.16	1.42	0.12	6.17
CV		1.72	20.27	0.35	3.04	0.26	13.2
CTV		2.14	2.14	2.14	2.14	2.14	2.14

The different letters in the same column indicate differences among treatments. Basmati—Super Basmati; Coarse—KSK-133; S.E—Standard Error for Comparison; CV—Critical Value for Comparison; CTV—Critical T Value.

Table 6. Mean values and homogeneous groups of all traits under all treatments (Season 2021–22).

Treatments	Varieties	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight	Grain Yield/m2
27 kg/ha	Basmati	126.27 AB	29 A	27.4 B	76 D	23.1 D	264.04 D
22 kg/ha	Basmati	127.37 A	28 AB	27 C	74 D	23.7 C	255.98 DE
17 kg/ha	Basmati	127.07 A	27 B	27 C	75 D	23.5 C	251.56 DE
12 kg/ha	Basmati	126.27AB	24 C	27.8 A	76 D	23.2 D	244.11 E
27 kg/ha	Coarse	119.47 C	23 C	24.6 F	114 B	27.6 B	436.14 A
22 kg/ha	Coarse	119.47 C	19 D	24.4 F	121 A	28.2 A	401.87 B
17 kg/ha	Coarse	125.27 B	19 D	25.8 D	118 A	28.1 A	320.40 C
12 kg/ha	Coarse	125.33 B	19 D	25.4 E	109 C	27.4 B	316.84 C
S.E.		0.80	9.45	0.16	1.42	0.12	6.17
CV		1.72	20.27	0.35	3.04	0.26	13.2
CTV		2.14	2.14	2.14	2.14	2.14	2.14

The different letters in the same column indicate differences among treatments. Basmati—Super Basmati; Coarse—KSK-133; S.E—Standard Error for Comparison; CV—Critical Value for Comparison; CTV—Critical T Value.

Table 7. Pearson’s Correlation (Season 2021–22).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Tillers/Plant	0.509 *
Panicle length	0.826 **	0.724 **
Grains/panicle	−0.733 **	−0.871 **	−0.919 **
Thousand grain weight	−0.698 **	−0.872 **	−0.921 **	0.992 **
Grain yield/m2	−0.939 **	−0.590 *	−0.930 **	0.849 **	0.830 **

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01.

Table 7. Pearson’s Correlation (Season 2021–22).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Tillers/Plant	0.509 *
Panicle length	0.826 **	0.724 **
Grains/panicle	−0.733 **	−0.871 **	−0.919 **
Thousand grain weight	−0.698 **	−0.872 **	−0.921 **	0.992 **
Grain yield/m2	−0.939 **	−0.590 *	−0.930 **	0.849 **	0.830 **

*—F test significant at p ≤ 0.05; **—F test significant at p ≤ 0.01.

Table 8. Direct (diagonal) and indirect effects of parameters on grain yield per m² (Season 2021–22).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Plant Height	−0.09465	−0.20908	−0.7473	−2.07339	1.733041
Tillers/Plant	−0.05355	−0.36956	−0.63385	−2.44616	2.144322
Panicle length	−0.08368	−0.27713	−0.84525	−2.5154	2.219695
Grains/panicle	0.072687	0.334839	0.787516	2.69979	−2.36967
Thousand grain weight	0.068988	0.333293	0.789101	2.690758	−2.37763

Residual effect^2 = 0.018675.

Table 8. Direct (diagonal) and indirect effects of parameters on grain yield per m² (Season 2021–22).

Traits	Plant Height	Tillers/ Plant	Panicle Length	Grains/ Panicle	Thousand Grain Weight
Plant Height	−0.09465	−0.20908	−0.7473	−2.07339	1.733041
Tillers/Plant	−0.05355	−0.36956	−0.63385	−2.44616	2.144322
Panicle length	−0.08368	−0.27713	−0.84525	−2.5154	2.219695
Grains/panicle	0.072687	0.334839	0.787516	2.69979	−2.36967
Thousand grain weight	0.068988	0.333293	0.789101	2.690758	−2.37763

Residual effect^2 = 0.018675.

4. Discussion

In both the 2020 and 2021 seasons, plant height showed very significant differences between different seed rates and varieties and in their interactions. Research has shown that the height of rice plants can vary significantly due to genetic factors and environmental conditions, including planting density and water or nutrient management [17,18]. A relevant study examines the interaction between plant height, seeding rates, and varietal characteristics in rice (Oryza sativa L.), highlighting that plant height is influenced by both genetic factors and planting density, which in turn affects agronomic traits such as production and yield the tillering components [19]. Research supports the claim that taller plants of certain varieties may result in different trade-offs in resource allocation and competition, affecting overall productivity and growth dynamics [20,21].

Plant height had a highly significant negative correlation with grain yield, thousand grain weight, and grains per panicle in both years, indicating that taller plants tended to produce lower yields, lighter grains, and fewer grains per panicle. This negative relationship could be attributed to the fact that taller plants might invest more energy in vertical growth at the expense of reproductive output [22]. This negative correlation between plant height and yield parameters has also been linked to hormonal regulation in rice, particularly involving brassinosteroids (BRs). Genetic manipulation of BR pathways has shown that changes in plant height often come with corresponding impacts on grain size and yield traits [23]. This suggests a delicate balance between vegetative and reproductive growth, with plant height serving as an important determinant of overall yield performance [22,23]. However, plant height positively correlated with tillers per plant and panicle length, suggesting that taller plants might have more tillers and longer panicles, which could be beneficial for certain agronomic traits. Research has confirmed that plant height is positively associated with traits like panicle length and tiller number, especially in rice varieties [22,24]. This correlation suggests that improving one of these vegetative traits might simultaneously enhance others, which could be targeted for selection in breeding programs aimed at optimizing yield potential under different environmental conditions [25].

Tillers per plant showed a significant variation across all seed rate treatments and varieties in both years, with the highest values recorded at the highest seed rate treatments. This trend was more pronounced in the basmati variety compared to the coarse variety, suggesting that higher seed rates promote tillering in basmati rice. A relevant study highlights that tillering in rice is influenced not only by seed rates but also by nitrogen availability [26]. The significant negative correlation of tillers per plant with grain yield, thousand grain weight, and grains per panicle suggests that excessive tillering might lead to competition among tillers, reducing the overall yield and grain quality [27].

Panicle length also showed a significant variation with seed rates and varieties, with the basmati variety having longer panicles under lower seed rates. This trend suggests that lower seed rates may reduce competition among plants, allowing them to allocate more resources to panicle development [28]. In contrast, some varieties show a different response depending on genetic factors that influence panicle architecture and grain-filling characteristics under different planting densities [26,27,28]. The highly significant negative correlation between panicle length and grain yield, thousand grain weight, and grains per panicle indicates that longer panicles do not necessarily translate into higher yield or better grain quality. The positive correlation of panicle length with plant height and tillers per plant suggests that panicle length is associated with overall vegetative growth [29,30]. Studies on rice (Oryza sativa L.) confirm this relationship, where an increase in panicle length often correlates with a reduction in grain-filling capacity, impacting the total grain yield negatively.

Grains per panicle varied significantly across seed rates and varieties, with higher values generally observed at lower seed rates, particularly in the coarse variety. This suggests that lower seed rates may reduce intra-plant competition, leading to better grain filling and more grains per panicle [17,31]. The significant positive correlation of grains per panicle with grain yield and thousand grain weight indicates that grains per panicle is a critical determinant of yield and grain quality. However, its negative correlation with plant height, tillers per plant, and panicle length suggests a trade-off between vegetative growth and reproductive output [31]. These studies underline the complexity of optimizing seed rates and the need for careful management of genotype-environment interactions to achieve high yields across different varieties. Thousand grain weight (TGW) showed significant variation across seed rates and varieties, with higher values typically observed under moderate seed rates. This trend was more evident in the coarse variety, indicating that optimal seed rates are crucial for maximizing grain weight. Previous research has identified genetic factors like OsMADS56 as vital for regulating TGW in rice, impacting grain size and yield [32]. The positive correlation of thousand grain weight with grain yield and grains per panicle underscores its importance in determining yield potential. The negative correlation with plant height, tillers per plant, and panicle length suggests that heavier grains are associated with less vegetative growth [32].

Grain yield per square meter exhibited significant variation with seed rates and varieties, with the highest yields recorded at the highest seed rates in both years. The coarse variety consistently produced higher yields than the basmati variety, indicating its superior yield potential under varying seed rates. Previous studies examined the effects of different seed rates on grain yield and quality traits in rice [33]. It found that higher seed rates significantly improved yield, with the coarse varieties outperforming basmati in terms of yield potential under various conditions. The results highlighted the critical role of optimal seed rates in maximizing grain yield [34]. The review article discusses the genetic basis of grain size and weight in cereals, emphasizing that grain weight is positively correlated with yield. This source elaborates on how grain filling and developmental traits influence overall yield [34,35]. The positive correlation of grain yield with thousand grain weight and grains per panicle highlights the importance of these traits in achieving higher yields. The negative correlations with plant height and panicle length suggest that excessive vegetative growth may detract from yield potential.

5. Conclusions

Optimizing seed rates is crucial for maximizing rice yield and quality. This study underscores the complex interplay between seed rates, varieties, and various agronomic traits in determining rice yield. The basmati and coarse varieties responded almost the same to seed rate treatments, showing better performance under a higher seed rate, which was 27 kg/ha. The 27 kg/ha seed rate treatment generally promoted better grain filling and higher grain weights. There were significant trade-offs between vegetative growth and reproductive output, with taller plants and more tillers often associated with lower yields and grain weights. Grains per panicle and thousand grain weight emerged as critical determinants of yield, highlighting the need to balance vegetative growth with reproductive development. Thus, breeding programs should focus on developing varieties that balance vegetative traits like tiller production and panicle length with reproductive traits to enhance overall yield. Furthermore, optimizing seed rates can significantly improve yield outcomes, particularly in coarse varieties that are less responsive to high densities. Future research should explore genetic manipulation techniques that can enhance traits associated with grain filling and yield while maintaining optimal plant height. Implementing precision agriculture practices to monitor growth dynamics and resource allocation can further inform best practices for maximizing rice yield. By integrating these strategies, rice production can be significantly improved to meet the demands of growing populations and changing climatic conditions.