Seed Dormancy and Germination Responses to Different Temperatures of Leptochloa chinensis (L.) Nees: A Case Study with 242 Populations Collected from Rice Fields in East China

Abstract

Leptochloa chinensis (L.) Nees is a troublesome rice weed. We collected 242 L. chinensis populations from rice fields in eastern China and studied the duration of seed dormancy and the seed germination ability at different temperatures. All L. chinensis populations studied exhibited seed dormancy. The periods required to reach 50% germination under optimal conditions were 31–235 days, with an average of 96 days. None of the populations germinated at 15 °C. Under constant temperatures of 20, 25, 30, and 35 °C, the average germination percentages of 242 populations were 0%, 71%, 79%, and 60% at 2 days after treatment (DAT), and were 56%, 84%, 88%, and 88% at 14 DAT, respectively. The duration of seed dormancy, as well as the germination ability of seeds, were significantly (p < 0.05) influenced by the agricultural region and the longitude and latitude of the collection locations. Under constant temperatures of 20 to 35 °C, the average germination percentages of seeds collected from transplanted rice fields were significantly higher than those collected from direct-seeded fields. This is the first study on seed germination biology of L. chinensis with multiple populations systematically collected from rice fields on a regional scale.

1. Introduction

Chinese sprangletop [Leptochloa chinensis (L.) Nees] is a troublesome weed in paddy fields worldwide that seriously affects rice quality and yield [1]. Leptochloa chinensis produces up to about 45,000 seeds per individual plant [2,3], and 25 plants m⁻² caused a 69% yield loss [4]. Chemical control is one of the main methods of L. chinensis management in rice fields. Leptochloa chinensis populations in different rice-growing areas have evolved resistance to the main post-emergence rice herbicides, such as cyhalofop-butyl and metamifop [5,6]. It is insensitive to other herbicides used for grass control in rice, such as florpyrauxifen-benzyl, bispyribac-sodium, pyribenzoxim, and penoxsulam [5]. Various pre-emergence herbicides showed good efficacy against L. chinensis [7], which mainly target the growing stage from germination to the two-leaf stage [6]. Knowledge of seed dormancy and germination is the basis for predicting the occurrence of L. chinensis on-field.

To date, different conclusions on seed dormancy in similar studies with different populations of L. chinensis have been reported. Benvenuti et al. reported that L. chinensis seeds did not show seed dormancy [8]; however, Dong et al. reported that L. chinensis seeds showed dormancy [9]. The heterogeneity of seed dormancy duration makes it difficult to determine the optimal application period of herbicides, which greatly decreases the efficacy of applied herbicides. For example, pre-emergence herbicides are typically applied 3–5 days after rice seeding, 3–5 days before rice transplanting, or 7–10 days after rice transplanting. The seed dormancy of L. chinensis is reported to be physiological dormancy [9,10], which might be overcome with different periods for different seeds. Thus, variations in seed dormancy frequently result in the heterogeneity of seedling emergence. In eastern China, L. chinensis seedlings frequently emerge continuously for one month in rice fields, which causes great difficulties for pre-emergence chemical control practices [11]. Revealing the variations in seed dormancy durations of L. chinensis with multiple populations could be very important for its integrative management. Moreover, temperature is an important climate factor that plays a fundamental role in the determination of the emergence of plants. As a rice weed, seed germination of L. chinensis occurs in higher temperatures. Lower temperatures may constrain or delay its seed germination, which should be managed with different strategies accordingly. To date, germination patterns of L. chinensis responding to different temperatures are still unclear.

In 2022, we collected 242 L. chinensis populations (Figure 1) from rice fields in eastern China and compared their seed dormancy, germination percentage, and rate at different temperatures. We aimed to (1) reveal the range and general characteristics of the seed dormancy periods of L. chinensis, (2) reveal the range and general characteristics of L. chinensis seed germination at different temperatures, and (3) explore correlations among seed dormancy, germination percentage at different temperature, and longitude, latitude, agricultural region, and rice planting methods of the seed collecting rice fields.

2. Materials and Methods

2.1. Plant Material

In October 2022, we collected 242 L. chinensis populations from rice fields in eastern China, including 222 populations collected from Jiangsu Province and 20 populations from Jiaxing City, Zhejiang Province. Panicles with mature seeds were randomly collected from more than 100 individuals of each population with a pollen bag (100 mesh, 30 cm by 45 cm), and mature seeds were collected by hand. An interval of >5 km was set for adjacent populations (Figure 1, Table S1). All counties cultivating rice in Jiangsu Province were visited. Jiangsu Province was divided into six agricultural regions: Xuhuai (XH), Yanhai (YH), Lixiahe (LXH), Ningzhenyang (NZY), Yanjiang (YJ), and Taihu (TH) [12]; and Jiaxing City, which belongs to Zhejiang Province, was set as another agricultural region (Figure 1; Table S1). The longitude, latitude, agricultural regions, and rice planting methods of each L. chinensis population were recorded. The collected seeds were stored in kraft paper envelopes at room temperature, fluctuating from 20 to 25 °C in a lab with air conditioners to control the temperature. The lemmas of mature L. chinensis seeds were usually not closed and were easily separated. Before each experiment, candidate seeds of each population were put in a pollen bag and gently rubbed by hand to gain seeds without lemmas. Then, plumb and intact seeds were used for the experiments.

The seed-collecting areas referred to 14 cities in eastern China. We checked the temperatures during the ordinary rice growing season (1st June to 15th October) and annual precipitation in 2022 of these 14 cities. Temperature data were cited from www.tianqihoubao.com, and annual precipitation data were cited from http://tjj.zj.gov.cn/ (Jiaxing city, accessed on 10 September 2024) and http://www.jiangsu.gov.cn/col/col84736 (13 cities of Jiangsu Province, accessed on 10 September 2024). The average maximum temperatures of the 14 cities ranged from 28.2 to 30.2 °C during the rice growing season in 2022, which were significantly and negatively correlated with latitude (Figure 2A) but not significantly correlated with longitude (Figure 2B). Annual precipitations ranged from 633.5 to 1189.2 mm in 2022 (Figure 2), which were significantly and negatively correlated with latitude (Figure 2C) and significantly and positively correlated with longitude (Figure 2D).

2.2. Seed Dormancy

The experiments were conducted in the laboratory of Yangzhou University (E119.423; N32.388) from October 2022 to August 2023. Six treatments were set up in the experiment, which was tested at 3, 15, 30, 120, 210, and 300 days of seed storage. Storage conditions were the same as mentioned above, and storage duration (seed age) was calculated according to the period from seed collecting to the day of incubation in Petri dishes. Seeds of 43 populations were not transported to the lab within 3 days and were not used in this experiment. Thus, 199 populations were used in the dormancy experiments. For each treatment, thirty-five intact and undamaged seeds were selected and placed in a 9 cm diameter Petri dish with two layers of filter paper moistened with 5 mL distilled water. Each treatment was replicated with three Petri dishes. Each Petri dish was sealed with cling film, and water was added as needed. The seed germination percentage for each treatment was then determined in illumination boxes (CFG-400CH, Changzhou Haibo Instrument Equipment Co., Ltd., Changzhou, China) with a constant temperature of 30 °C (light 12 h/dark 12 h, 12,000 lx). The number of seeds that germinated in each Petri dish was recorded every 2 days until 14 days after treatment.

2.3. Seed Germination under Different Temperatures

The experiments were conducted in the laboratory of Yangzhou University (E119.423359; N32.388486) in August 2023. Seeds (300 days of seed aging) of 242 L. chinensis populations were used for testing seed germination at different temperatures, using the same method as mentioned above. Dishes were placed in illumination boxes (two of CFG-400CH, Changzhou Haibo Instrument Equipment Co., Ltd.; three of HP1000GS-B, Wuhan Ruihua Instruments & Equipment Co., Ltd., Wuhan, China) with temperatures of 15, 20, 25, 30, and 35 °C (light 12 h/dark 12 h, 12,000 lx). Each treatment was replicated with three Petri dishes.

2.4. Statistical Analysis

Data are presented as mean ± standard error. To analyze variations among populations in different indices, the coefficient of variation (CV) was calculated [13]. To analyze the influences of the seed storage period, treated temperature, agricultural region, and rice planting method of seed collecting sites on germination percentage, a generalized linear model (GLM) univariate analysis in SPSS 16.0 was applied. The relationship between longitude, latitude, and germination percentage was analyzed by linear regression of SPSS 16.0. To analyze the significant influences of seed storage, temperatures, and agricultural regions on average germination percentages, data were subjected to a one-way analysis of variance using SPSS 16.0, which was checked for normality and constant variance before analysis. The germination percentage data were log-transformed before analysis. Non-transformed means for germination percentages are reported, with statistical interpretation based on the transformed data. Treatment means were separated using Fisher’s protected LSD test at p = 0.05. An independent sample t-test was used to compare the seed germination percentages of L. chinensis populations collected from direct-seeded rice fields and those collected from transplanted rice fields.

A three-parameter logistic function was fitted to test the response of germination percentage to the seed storage period. A four-parameter logistic function was fitted to test the response of germination percentage to temperature treatment, which both using the ‘drc’ add-on package in R 3.1.3 [14]:

logistic, 3 parameter: Y 1 = (a − d)/[1 + (x/e₁)^b]

(1)

logistic, 4 parameter: Y 2 = d + (a − d)/[1 + (x/e₂)^b]

(2)

Y 1 and Y 2 represent seed germination percentage; x is the seed storage period or treatment temperature for germination tests; a is the upper limit; d is the lower limit; e₁ is the number of days of seed storage when the germination percentage reaches 50% (DR₅₀); e₂ is the germination temperature at which the germination percentage reaches 50% (TR₅₀); and b is the slope.

3. Results

3.1. Seed Dormancy Duration

All 199 L. chinensis populations collected from paddy fields in eastern China showed seed dormancy 30 days after collection. At seed storage for 3–30 days, the mean germination percentages of 199 populations were 0.3–1.3% (Figure 3A–C). Among the 199 populations, 38.7–66.3% of the populations showed germinated seeds, with germination percentages all <25%. One population showed 24.6% seed germination at 30 days after collection.

The mean germination percentage of 199 L. chinensis populations increased significantly (p < 0.05) with extending seed storage (Figure 3), and meanwhile, the CVs among populations decreased. After storing for 120 days, seed germination percentages of the 199 populations ranged from 6.7 to 100%, averaging 64.6% (Figure 3D), and 74% of populations showed seed germination percentages >50%. After storing for 210 d, 95% of the populations showed seed germination percentages >50% (Figure 3E). After storing for 300 days, the mean germination percentage of 199 populations was 88.0%, and 98% of populations showed seed germination percentages >50% (Figure 3F). The DR₅₀ of the 191 L. chinensis populations was 31–235 days, with an average of 96 days (Figure 3G), and the germination percentages of the other eight populations did not conform to the logistic model used. DR₅₀ periods for 16.8% of populations showed DR₅₀ periods <50 days, 62.8% of L. chinensis populations ranged from 70 to 120 days, and 8.9% showed DR₅₀ periods >150 days.

In May 2024, another experiment suggested that the average germination percentage of 89 populations (randomly selected from the seven agricultural regions, stored for 570 d, and treated with a constant temperature of 30 °C, light 12 h/dark 12 h, and 12,000 lx for 14 days) ranged from 56.4 to 98.8%, averaging 88.0% (Figure S3A; unpublished data).

3.2. Influences of Location and Planting Methods on Seed Dormancy Durations

The seed storage, agricultural region, and seed storage × agricultural region showed significant (p < 0.05) influences on germination percentages of L. chinensis seeds (

Table 1. Results (F-values) of GLMs among germination percentages of Leptochloa chinensis (L.) Nees seeds at different seed storage, agricultural regions, and rice planting methods of seed collecting fields.

Influencing Factors	F
Seed storage	128.5 *
Agricultural region	21.1 *
Rice planting method	0.1
Seed storage × agricultural region	7.4 *
Seed storage × rice planting method	2.1
Agricultural region × rice planting method	0.8
Seed storage × agricultural region × rice planting method	1.6

Note: The incubating temperature was constant at 30 °C (12/12 h light/dark). Seed storage treatments included 120, 210, and 300 days after collecting; agricultural regions were the same as in Figure 1; rice planting methods include direct-seeded and transplanting. *: p < 0.05. The same as below.

). The average germination percentage of populations collected from the YH agricultural region was the highest among all seed dormancy experimental treatments (

Table 2. Germination percentages of L. chinensis seeds stored for different periods and collected from different agricultural regions.

Agricultural Region	Days after Storage
	120	210	300
XH	71.0 ± 2.9 ab	80.3 ± 1.3 b	92.0 ± 0.6 a
YH	80.5 ± 5.7 a	89.2 ± 1.5 a	92.2 ± 0.8 a
LXH	59.7 ± 3.9 b	78.9 ± 2.0 b	90.2 ± 1.0 ab
NZY	66.5 ± 3.1 b	68.5 ± 2.4 d	87.6 ± 1.1 bc
YJ	65.4 ± 3.6 b	72.6 ± 1.3 cd	87.4 ± 1.0 bc
TH	62.8 ± 3.1 b	77.6 ± 1.4 bc	83.8 ± 1.1 c
JX	41.9 ± 6.4 c	67.2 ± 2.5 d	85.7 ± 1.4 c

Note: Different letters within the same column indicate significant differences at p < 0.05. The incubating temperature was constant at 30 °C (12/12 h light/dark).

and Table S2). At seed storage for 120 days, the average germination percentage of populations collected from the YH agricultural region was significantly the highest among the seven regions, followed by the XH agricultural region, and the average germination percentage of populations collected from the JX agricultural region was significantly the lowest, 38.6% lower than that of YH. At seed storage for 210 days, the NZY and JX agricultural regions were significantly the lowest, and the TH and JX agricultural regions were significantly the lowest at 300 days of seed storage. The variation in germination percentages among the different agricultural regions decreased significantly. Thus, the variations in germination percentages among different agricultural regions were mainly due to the duration of seed dormancy.

The collection sites of the 199 L. chinensis populations ranged from 4.1° longitude to 4.1° latitude. The latitude of the collection site significantly (p < 0.05) influenced seed germination percentages after being stored for different periods (

Table 3. Regressions between longitude, latitude, and germination percentages of L. chinensis seeds stored for different periods.

Seed Storage Period (Day)	Longitude		Latitude
	Formula	R2	Formula	R2
120	y = −2.92x + 414.59	0.014	y = 5.35x − 109.21	0.077 *
210	y = −0.54x + 140.22	0.001	y = 3.30x − 31.37	0.069 *
300	y = −0.94x + 200.55	0.008	y = 2.39x + 10.33	0.081 *

Note: The longitude and latitude ranges of the 199 L. chinensis populations studied were 117.5–121.6° and 30.4–34.6°, respectively. *: p < 0.05. The incubating temperature was constant at 30 °C (12/12 h light/dark).

; Figure S1). The germination percentages at 120, 210, and 300 days of seed storage were positively correlated with the latitude. Seed dormancy of the L. chinensis population in the low longitudinal (western) and high latitudinal (northern) areas was released faster, and the germination percentage was higher after overcoming dormancy.

Among seed storage with 120, 210, and 300 days, the average germination percentage of L. chinensis seeds collected from transplanted rice fields (76.1%) was slightly and significantly higher than that of L. chinensis seeds collected from direct-seeded rice fields (74.8%).

3.3. Seed Germination Percentage under Different Temperatures

The optimal germination temperature was 30 °C, and seeds germinated well under 25 °C and 35 °C, according to germination percentages on 2, 4, and 14 days after treatment. All 242 populations did not germinate at 15 °C (Figure 4A). The seed germination percentage of L. chinensis increased significantly (p < 0.05) with the temperature increasing from 20 to 30 °C, and the CV of germination percentages of 242 populations decreased significantly with the increasing temperature (Figure 4). At 20 °C, germination percentages of 242 populations ranged from 3.4 to 96.9%, averaging 56.1% (Figure 4B), and 63% of populations showed seed germination percentages >50%. No seed from any of the populations germinated two days after treatment, and four days after treatment, the average germination percentage of 242 populations was 22.4%. At 25 °C, the germination percentage of 96% of populations was >50%, and the average germination percentage of all populations was 84.3% (Figure 4C). Under 30 °C treatment, the mean germination percentage of all populations was 88%, ranging from 41.5% to 100% (Figure 4D). At 25 and 30 °C, the average germination percentage of all populations at two days was >70%. At 30–35 °C, the germination percentage did not increase significantly (Figure 4D,E). Seed germination of L. chinensis was inhibited in the first two days at 35 °C, compared with those at 30 °C. The TR₅₀ of 240 L. chinensis populations ranged from 15.1 to 33.8 °C, with an average of 19.7 °C (Figure 4F). The germination data for the two populations did not fit the logistics model. Among the 240 L. chinensis populations, 82% showed a TR₅₀ between 18 and 21 °C.

3.4. Influences Factors on Seed Germination under Different Temperatures

The temperature, agricultural region, rice planting method, and temperature × agricultural regions of the L. chinensis seed collection sites showed significant influences (p < 0.05) on the germination percentage 14 days after treatments (

Table 4. Results (F-values) of GLMs among germination percentages of L. chinensis seeds at different temperatures, agricultural regions, and rice planting methods of seed collecting fields.

Influencing Factors	F
Temperature	205.6 *
Agricultural region	21.0 *
Rice planting method	4.3 *
Temperature × agricultural region	10.3 *
Temperature × rice planting method	0.8
Agricultural region × rice planting method	1.6
Temperature × agricultural region × rice planting method	0.8

Note: Seeds were stored for 300 days before the experiment. The temperature treatments include 20, 25, 30, and 35 °C. Seeds of L. chinensis did not germinate at 15 °C. The agricultural regions are the same as in Figure 1. The planting methods include direct-seeded and transplanting. *: p < 0.05.

). At 20 °C, the germination percentages of seeds collected from the YH agricultural region were significantly the highest among the seven regions, and the TH agricultural region was the lowest (

Table 5. Germination percentages of L. chinensis seeds under different temperatures and collected from different agricultural regions.

Agricultural Region	Treated Temperature (°C)
	20	25	30	35
XH	55.5 ± 3.2 b	87.7 ± 1.9 ab	90.1 ± 1.0 ab	91.8 ± 0.8 a
0.8YH	84.6 ± 2.2 a	90.8 ± 1.5 a	92.8 ± 0.8 a	91.9 ± 1.0 a
LXH	65.7 ± 6.6 b	89.6 ± 1.4 a	89.3 ± 1.4 ab	90.6 ± 1.3 a
NZY	63.0 ± 3.0 b	85.7 ± 2.3 ab	86.5 ± 1.4 b	86.9 ± 1.4 ab
YJ	57.9 ± 3.6 b	81.4 ± 1.8 bc	85.5 ± 1.5 b	88.4 ± 1.5 a
TH	37.8 ± 2.7 c	79.8 ± 2.3 bc	88.7 ± 1.1 ab	82.4 ± 1.8 b
JX	63.4 ± 3.2 b	76.2 ± 3.7 c	87.2 ± 2.6 b	86.3 ± 2.1 ab

Note: Different letters within the same column indicate significant differences at p < 0.05. Note: Seeds were stored for 300 days before the experiment.

and Table S3). The differences in seed germination of L. chinensis populations among different agricultural regions were substantially lower at 25–35 °C.

At 20–25 °C, the germination percentage at 4 and 14 days after treatment were significantly and negatively correlated with the longitude of the seed collection site and significantly positively correlated with the latitude (

Table 6. Regressions between latitude or longitude and germination percentages of L. chinensis seeds at 2, 4, and 14 days after treatment with different temperatures.

Temperature °C	Days	Longitude		Latitude
		Formula	R2	Formula	R2
20 °C	4	y = −3.06x + 388.89	0.016 *	y = 2.78x − 67.99	0.020 *
	14	y = −4.96x + 649.83	0.035 *	y = 2.87x − 36.96	0.018 *
25 °C	2	y = −6.18x + 811.15	0.093 *	y = 6.57x − 142.16	0.160 *
	4	y = −3.11x + 453.99	0.039 *	y = 3.90x − 45.51	0.092 *
	14	y = −3.04x + 448.15	0.044 *	y = 3.78x − 38.50	0.104 *
30 °C	2	y = −0.12x + 92.98	0.000	y = 3.52x − 35.19	0.058 *
	4	y = −0.10x + 97.49	0.000	y = 2.57x + 2.60	0.055 *
	14	y = −0.48x + 145.46	0.002	y = 2.64x + 1.84	0.089 *
35 °C	2	y = −2.85x + 401.40	0.017 *	y = 4.47x − 85.23	0.064 *
	4	y = −0.40x + 132.80	0.001	y = 0.86x + 57.13	0.009
	14	y = −0.24x + 116.90	0.001	y = 0.98x + 56.51	0.020

Note: The longitude and latitude ranges of the 242 populations studied were 117.5–121.6° and 30.4–34.6°, respectively; seeds were stored for 300 days before the experiment. Seeds did not germinate after 2 days under 20 °C. *: p < 0.05.

; Figure S2). At 30 °C, the germination percentage determined at 2 and 14 days after treatment was significantly and positively correlated with the latitude of the seed-collection sites. There was a significant and negative correlation between germination percentage and longitude at 35 °C for 2 days. Moreover, the germination percentages of seeds collected from transplanted rice fields (81.5%) were significantly higher than those collected from direct-seeded fields (78.7%).

4. Discussion

4.1. Different Periods Were Required to Release Seed Dormancy in Different Populations

Together our study suggested that all L. chinensis populations collected from eastern China showed physiological dormancy, which required different after-ripening periods. Moreover, L. chinensis populations continuously and gradually released seed dormancy. Thus, L. chinensis generally continuously emerges seedlings in rice fields, which effectively facilitates the escape of this weed species from chemical control strategies. It is important for rice growers to know the seed dormancy duration of L. chinensis in their own fields. For example, in fields with a great part of L. chinensis seeds holding dormancy durations >210 days, the seeds may possibly germinate continuously for more than two months; and pre-emergence herbicides targeting this weed species should be applied several times, such as before rice planting, 3–5 days after planting, and 15–20 days after planting. Whereas, in fields with L. chinensis holding shorter durations of seed dormancy, most seeds may germinate in 2 days after irrigation or rainy weather; thus, one-time pre-emergence chemical control targeting this weed species may be enough. Leptochloa chinensis biotypes show more significant variation in germination timing, which could be an adaptation to variable climates [15]. Many weed species, such as Halenia elliptica D. Don [16], Convolvulus arvensis L. [17], and Capsella bursa-pastoris (L.) Medik. [18], can potentially germinate under a wide range of dormancy durations.

The lemmas of mature L. chinensis seeds were usually not closed. Germination percentages increased with the release of dormancy. Thus, we supposed that ungerminated seeds were mostly under dormancy. Regarding seed germination experiments with temperature treatments, the germination of seeds was constrained by unsuitable conditions. In another experiment, the average germination percentage of L. chinensis seeds (seed storage for 570 days) of 31 populations treated with a water potential of −0.3 MPa for 14 days was 31.3%, and the average germination percentage of ungerminated seeds of these populations increased to 80.1% (Figure S3B) after incubation under optimal conditions for another 14 days (unpublished data).

4.2. Different Populations Showed Different Seed Germination Patterns under Different Temperatures

Under optimal conditions, a majority of L. chinensis populations tested showed germination percentages >50% in 2 days, which could be sufficient for infesting rice fields. Specifically, farmers frequently finish soil preparation of all paddy fields and then finish transplanting or direct-seeding of rice in several days, which is frequently named “field waiting for planting period”. Many L. chinensis seedlings may emerge and grow quickly during the “field waiting for planting period”, which might cause failures of pre-emergence chemical control. For example, pretilachlor, pendimethalin, clomazone, pyraclonil, and mefenacet exhibit good control effects on various L. chinensis populations [7]. However, these five pre-emergence herbicides are ineffective in managing L. chinensis seedlings after the two-leaf stage [5]. Moreover, applying post-emergence herbicides at early stages after rice seeding or transplanting frequently causes serious rice injury. Therefore, in rice-planting areas infested by L. chinensis, pre-emergence herbicides should be applied in periods between soil preparation and rice planting. For example, in transplanted rice fields, applying pre-emergence herbicides during the “field waiting for planting period” and transplanting rice seedlings 3–5 days after pre-emergence chemical control frequently results in good control efficacy against L. chinensis. In direct-seeded rice fields, modern seeding methods with high efficiency to shorten the “field waiting for planting period” could be important for L. chinensis management, such as mechanized seeding with drills or unmanned aerial vehicles [19].

Benvenuti et al. found that the optimum temperature for L. chinensis seed germination was 25–35 °C [8]. Our study suggested that the optimal temperature was 30 °C [20], while different populations showed different seed germination patterns responding to temperatures. Intraspecific differences in seed germination in response to different temperatures have also been observed in Arabidopsis thaliana [21]. Liu et al. [22] found that four Dalbergia odorifera populations exhibited different seed germination percentages and speeds at different temperatures. Zhang et al. [23] found that seed germination of temperature requirements of Bidens pilosa differed significantly among populations.

4.3. Intraspecific Variations in Seed Dormancy and Germination

Overall, the experiments found intraspecific variations in seed dormancy and germination biology under different temperatures, among which the agricultural region, longitude, latitude, and rice planting methods of seed-collecting fields all showed significant (p < 0.05) influences. Seed germination can be affected by environmental conditions experienced by mother plants, and differences in germination of seeds from different maternal environments might be due to the epigenetic memory inherited from mother plants [24,25]. Brainard et al. found that in the same area, Amaranthus powellii seeds from vegetable fields typically have more extended dormancy than those from dairy farms [26]. Gafni found that the germination percentages of nine Amaranthus albus populations from different regions of Israel were significantly different at different temperatures [27]. There are differences in rice planting habits, terrain distribution, soil type, cultivated land quality, and irrigation conditions between different agricultural regions [28,29], which might show different influences on L. chinensis seed dormancy and germination under different temperatures. The YH and XH agricultural regions are located in the northern part of the seed-collection area, and the TH and JX agricultural regions are located in the southern part. Thus, it appears that L. chinensis populations collected from southern locations had a longer duration of seed dormancy. Intraspecific variations in the dormancy characteristics of weed species at different latitudes and longitudes have been reported [15]. Similar to the findings in this study, Cheng et al. found that Spartina alterniflora populations collected from high-latitude locations needed cold stratification to release dormancy, while low-latitude populations did not [30]. Shihan et al. found that the duration of summer dormancy in Dactylis glomerata tended to decrease from South to North [31]. Variations in the duration of L. chinensis seed dormancy might be due to the adaptive phenotypic plasticity of maternal plants or the gene dominance and recessiveness of seeds [32,33]. In addition, geographically related variations in seed germination percentage and speed have been observed in different plant species [34]. Dwiyanti et al. found that seeds of Miscanthus sinensis, a grassy weed species, collected from locations at higher latitudes germinated faster under 30 °C/20 °C (day/night) and 15 °C/10 °C (day/night) [35]. In our study, L. chinensis seeds collected from locations at higher latitudes and lower longitudes germinated rapidly, with higher germination percentages at a lower temperature (20 °C). Interestingly, in seed collecting areas of this study, areas with higher latitudes and lower longitudes tended to hold lower precipitations with lower temperatures. Weng and Hsu found that the seed germination percentage of Lilium formosanum collected at lower latitudes was higher than those collected at higher latitudes [36]. Zhou et al. found that the seed germination periods of Ambrosia artemisiifolia were significantly delayed with increasing latitude and longitude of collection sites, whereas the germination percentages increased with increasing latitude and longitude of collection sites [15].

Our results suggested that germination percentages of L. chinensis seeds collected from transplanted rice fields were significantly and slightly higher than those collected from direct-seeded fields, which may be related to the maternal effect [27]. There are many differences in applying periods and dosages of herbicides between the two kinds of rice fields [5]. In transplanted rice fields, rice seedlings at the four- to five-leaf stage are transplanted, which have a higher tolerance to many herbicides; thus, chemical herbicides with higher toxicity and efficacy against weeds could be used with higher doses and acceptable weed control efficacies could be frequently achieved with an application of pre-emergence herbicides and an application of post-emergence herbicides. Whereas, in direct-seeded rice fields, rice seedlings emerge from seeds together with weed seedlings. Seedlings of direct-seeded rice suffer longer periods and higher pressures of weed competition at early growing stages, which are frequently injured by various herbicides. Farmers usually apply chemical herbicides more times and with higher total dosages to effectively control weeds on the basis of avoiding rice seedling injury, frequently applying various herbicides three to five times. Hence, we suppose that differences in germination percentages among L. chinensis populations collected from the two different kinds of rice fields might be related to the differences in chemical control strategies. The mechanisms under the above differences between L. chinensis seeds collected from direct-seeded and transplanting rice fields need more studies.

5. Conclusions

Together, our results suggested that L. chinensis seeds of all populations studied exhibited physiological dormancy, while the dormancy duration of different populations showed high intraspecific variation. The periods required to reach 50% germination under optimal conditions were 31–235 days, with an average of 96 days. The temperature required for 50% seed germination ranged from 15.1 to 33.8 °C among different populations, with an average of 19.8 °C. None of the populations germinated at 15 °C. The optimal temperatures for L. chinensis seed germination were 25 to 30 °C. Under optimal conditions, a majority of L. chinensis populations tested showed germination percentages >50% in 2 days and >80% in 4 days, which could be sufficient for infesting rice fields. Moreover, the agricultural region, longitude, latitude, and rice planting methods (direct-seeded or transplanting) of seed-collecting fields all showed significant (p < 0.05) influences on seed dormancy and germination percentages under different temperatures. The commonness and uniqueness of L. chinensis populations are valuable for improving its integrative management strategies and are worth further investigation.