A Thermal Time Basis for Comparing the Germination Requirements of Alfalfa Cultivars with Different Fall Dormancy Ratings

Abstract

Fall dormancy plays important roles in the evaluation of alfalfa’s winter hardiness and in the selection of alfalfa breeding. A rapid and effective method to estimate the fall dormancy rating of alfalfa will shorten the breeding cycle. The purpose of this study is to test the correlations between the germination thermal time model parameters and the fall dormancy ratings and to evaluate the potential of the thermal-based fall dormancy methodology. Alfalfa cultivars with a series of fall dormancy ratings were used to study the responses of seed germination at six constant temperatures (5, 10, 15, 20, 25, 30 °C). The results showed that all cultivars had a relatively high germination percentage at all temperatures and the optimal temperature is 25 or 30 °C. Germination rate and base temperature significantly increased with the fall dormancy rating of alfalfa cultivars while thermal time (θ_T) decreased with the fall dormancy rating. The extremely significant linear regression relationships between the germination rate, base temperature (T_b), θ_T, and fall dormancy rating indicated that it is convenient and straightforward to predict the fall dormancy rating of unknown cultivars or lines using thermal time model parameters. This method can significantly shorten the selection and breeding cycles in alfalfa cultivation.

1. Introduction

Alfalfa (Medicago sativa L.) is one of the most extensively cultivated perennial forage legumes worldwide [1,2]. Its primary utilization involves serving as hay, pasture, silage, or fresh feed for a diverse range of livestock and poultry. It is widely acknowledged for its remarkable capacity to provide abundant forage, resistance to frequent cutting, good persistence, and high levels of protein and nutritional value compared to grasses; thus, it is called “the king of fodder crops” [3,4]. Alfalfa also has good biological nitrogen fixation capacity, which can be used to improve soil fertility [5]. Therefore, interseeding alfalfa in natural and artificial grasslands is a highly adopted grassland management practice to enhance grassland productivity and forage quality [5].

The vast genetic and phenotypic variations in alfalfa permit its cultivation in large geographic regions with diverse climate and environmental conditions [6,7,8,9]. Alfalfa cultivars that developed from different geographic regions have distinct characteristics in fall dormancy [9]. Fall dormancy is a growth characteristic resulting from the physiological response to shorter photoperiods and declining temperature in northern latitudes during autumn [1,9]. Fall dormancy induces changes in morphological type and production capacity, and the transition in plant growth from upright to creeping ultimately leads to a decrease in yield [1,9]. Typically, alfalfa genotypes are grouped into nine fall dormancy ratings (FD), which are usually classified into three types including dormant (FD 1–3), semi or intermediate dormant (FD 4–6), and non-dormant (FD 7–9) [10]. Fall dormant alfalfa (FD 1–3) exhibits reduced stem growth during the autumn season, slower regeneration after mowing, and a creeping growth habit. Conversely, non-dormant types (FD 7–9) display accelerated stem growth in autumn, faster post-mowing regeneration, and an upright growth habit. Semi-autumnal varieties (FD 4–6) demonstrate intermediate characteristics between the dormant and non-dormant types [10].

Fall dormancy plays an important role in the adaptation to particular environments of alfalfa, owing to its association with winter survival [1]. The winter survival of fall dormant alfalfa is higher than for non-dormant alfalfa, and a higher fall dormancy rating of alfalfa indicates a lower ability for winter survival [11,12,13]. The strong correlation between fall dormancy and winter survival has been proved in various studies [1,11,12,13]. For example, three cycles of divergent phenotypic selection have been conducted on the basis of plant height in autumn, and the results showed that selection for greater fall dormancy decreased the height of alfalfa and markedly improved its winter hardiness [11]. Therefore, fall dormancy has been considered as a selection criterion to enhance the winter resilience of alfalfa and plays an irreplaceable role in the evaluation and screening of the overwintering ability of alfalfa varieties or germplasm [1,9].

Fall dormancy in alfalfa is also related to other important traits including initiation of spring growth, growth habit, root growth traits, stand persistence, and herbage yield and quality [4,7,14,15,16]. In major alfalfa-producing countries such as the United States and Canada, fall dormancy is treated as the primary indicator for assessing cultivar characteristics [17]. However, using the standard test to estimate the fall dormancy rating of alfalfa requires measuring regrowth height after the last autumn cutting and comparing it with standard cultivars, which is both time- and space-consuming [17]. Researchers have attempted new methods to predict fall dormancy ratings. For example, the near-infrared reflectance spectrum has been used to assess the fall dormancy series of alfalfa, but the method faced challenges in accurate distinguishment between the fall dormancy of cultivars with similar FD ratings [18]. Liu et al. [19] have found a strong correlation between internode length and number of internodes with fall dormancy. However, these studies have only focused on growth traits and just proved the correlation between those growth traits and fall dormancy, which cannot be used as a method for identifying the fall dormancy of different alfalfa cultivars or lines. Molecular methods have also been attempted in order to isolate and identify the fall dormancy rating. For example, SRAPs (related sequence amplification polymorphisms) have be used to directly classify the relative fall dormancy rating of populations with unknown characteristics and to isolate alfalfa populations and varieties based on the amplification of DNA [20,21]. However, these methods are costly and not widely used.

Temperature plays a major role in determining the periodicity of seed germination, plant growth, and the distributions of species [22,23,24]. For example, each plant species has a base temperature (or minimum temperature, T_b), below which it will not germinate, and an optimum temperature (or range of temperatures, T_o) at which it germinates fastest, and a maximum temperature (or ceiling temperature, T_c) above which seeds cannot germinate [25,26]. At sub-optimal and supra-optimal temperatures, the germination rate, that is, the reciprocal of the time to germination, often increases linearly with germination temperature [27,28]. Therefore, the germination response to sub-optimal temperature has been described by a thermal time model, which assumes that a seed requires the accumulation of a certain thermal time or heat sum (θ_T), above a base temperature (T_b), to germinate, under which the germination rate is theoretically zero [29]. Within a seed population or a seed lot, T_b is often constant and θ_T generally varies among seeds within a population [30,31].

A wide range of independent and comparative studies have indicated that the thermal time characteristics of plant species are influenced by their geographic origins because of local adaptation [31]. Tropical species usually require warmer growing conditions than temperate species and have higher T_b [31]. It has also been found that intro- and inter-specific variations exist in base temperatures for growth in warm-season grass, and that the variations probably are closely related to different regions of origin or at least adaptation to specific environmental pressure [32]. However, little information is available concerning the relationship of alfalfa fall dormancy and the germination responses to temperature, especially on a thermal time basis. Only Jungers et al. [28] measured T_b and the thermal time constant for germination across a range of temperatures and alfalfa cultivars with FD 2–6 and found that fall dormancy was related to T_b, not to the thermal time constant. One possible reason for the result was the relatively large interval of germination observation in days, leading to inaccurate model fitting and parameter estimation. Alfalfa usually germinates very fast in the first few days and finishes germination within 7 d above 10 °C [28]. Thus, it should be observed more frequently at least during the first few days of the germination duration of alfalfa using the thermal time model analysis. The correlation between the fall dormancy of alfalfa and the germination thermal traits has the potential application significance in selection and screening procedures for alfalfa breeding programs.

The aim of this study was to test how the germination of alfalfa seeds with different fall dormancy ratings responded to temperatures, to use the thermal time model to estimate T_b and θ_T for six cultivars with different fall dormancy ratings, to analyze the correlation of T_b and θ_T values with the fall dormancy ratings of alfalfa cultivars, and to determine how accurate and effective this method of predicting the full dormancy ratings is. We predict that T_b is positively related to the fall dormancy of alfalfa cultivars and θ_T is negatively related to the fall dormancy of alfalfa.

2. Materials and Methods

Six cultivars of alfalfa from three dormant groups were used, representing FD ratings from 1 to 9 (

Table 1. Cultivars, fall dormancy rating, and fall dormancy type of alfalfa.

Cultivar	Breeding Country	Fall Dormancy Rating	Fall Dormancy Type
AC Caribou	Canada	1	Dormant
53Q60	United States	3
Saranac	United States	4	Semi-dormant
Tango	United States	6
Dona Ana	United States	7	Non-dormant
CUF101	United States	9

). Seeds of all cultivars were stored at 4 °C until used. Intact seeds of good quality were selected for the germination experiment.

2.1. Experimental Design

Seeds were germinated at constant temperatures of 5, 10, 15, 20, 25, 30 °C. Germination tests were carried out in incubators (HPG-400, Haerbin, China) in 16 h light and 8 h dark photoperiod conditions. Four replicates of 50 seeds were used for each temperature treatment of each cultivar. Seeds were germinated in 9 cm diameter Petri dishes containing two layers of filter paper moistened with 7 mL distilled water. The Petri dishes were randomly arranged in the incubators. Distilled water was added to keep the filter paper moist during the experiment. Germination was recorded one, two, three, or four times per day depending on germination rates. For example, we checked germination per six hours at the first two days for seeds germinating at 25 °C conditions. We observed germination per day when 50% of seeds had germinated for each treatment. Seeds were considered to have germinated when the radicle protruded from the seed coat. Germinated seeds were removed at each counting. The duration of the germination was 21 days when no seeds germinated for three days.

2.2. Thermal Evaluation

The time required to achieve 50% germination was calculated by interpolation from the cumulative germination curve. The germination rate was calculated as the inverse of the number of days required to reach 50% germination. We transformed germination time from hour to day in model calculation because thermal time in day is more realistic and practical in alfalfa production and breeding. A linear regression equation was derived to relate germination rate with temperature in the sub-optimal temperature range. The thermal time constant (θ, °C·d) was estimated as the inverse slope of the regression line, and the base temperature (T_b) was calculated by extrapolation to the point where the germination rate was zero [33].

2.3. Statistical Analysis

The germination percentage was arcsine square root transformed before analysis to ensure homogeneity of variance. Two-way ANOVA was used to test the effects of temperature and cultivar on the final germination percentage and the rate of 50% germination. Linear regression analysis was used to test the relationships between the germination characteristics and fall dormancy ratings of the alfalfa cultivars. The formula of the relationship between germination rate and fall dormancy under different temperatures was obtained by regression analysis. All data were analyzed using SPSS (Version 19.0 for Windows).

3. Results

3.1. Germination Responses of Alfalfa Cultivars to Temperature

Germination percentage and germinate rate were significantly affected by temperature, cultivar, and their interaction (p < 0.05;

Table 2. Analysis of variance showing effects of temperature, cultivar, and their interaction on the germination of alfalfa.

Source		Final Germination Percentage		Germination Rate (1/T50)
	d.f.	F Value	p	F Value	p
Temperature (T)	5	2.77	0.02	1469.99	<0.001
Cultivar (C)	5	8.54	<0.001	69.32	<0.001
T × C	25	1.69	0.03	5.69	<0.001

).

The final germination percentage of the six cultivars showed similar patterns in response to temperature. More than 80% of seeds germinated from 5 to 30 °C for all cultivars (Figure 1a). The final germination percentage of cultivar Dona Ana at 5 °C was significantly lower than those at other temperatures (p < 0.05), but there was no significant difference in final germination percentage among temperatures for the other five cultivars (Figure 1a).

The germination rate of all cultivars was the lowest at 5 °C. The germination rate gradually increased with the increase in temperature and reached the maximum at 25 °C for cultivars AC Caribou, 53Q60, Saranac, and Dona Ana or at 30 °C for cultivars Tango and CUF101 (Figure 1b). The germination rate was significantly different among cultivars. Among the cultivars, the germination rate of cultivar AC Caribou was the lowest and the germination rate of cultivar CUF101 was the highest. The average germination rate was ranked as follows: AC Caribou < 53Q60 < Saranac < Tango < Dona Ana < CUF101 following the order of the fall dormancy rating.

Generally, the alfalfa seeds of all the cultivars needed more time to germinate at low temperatures (5, 10, and 15 °C), whereas seeds at high temperatures (25 and 30 °C) germinated rapidly. At 5 °C, alfalfa seeds needed 93–104 h to reach 50% germination, whereas at 25 °C, the time needed to reach the 50% germination varied from 15 to 19 h for different cultivars (

Table A1. Germination time for 50% germination (in hours) of alfalfa cultivars at different temperatures.

Temperature (°C)	Cultivar
	AC Caribou	53Q60	Saranac	Tango	Dona Ana	CUF101
5	104 ± 1.9 a *	100 ± 3.9 a	103 ± 1.4 a	96 ± 0.8 a	98 ± 1.7 a	93 ± 0.9 a
10	49 ± 5.2 b	49 ± 4.3 b	51 ± 2.1 b	47 ± 2.9 b	45 ± 2.2 b	42 ± 1.3 b
15	30 ± 1.3 c	29 ± 1.1 c	27 ± 1.5 c	25 ± 0.3 c	25 ± 0.7 c	22 ± 0.8 c
20	23 ± 0.8 d	21 ± 0.8 d	21 ± 0.9 d	19 ± 0.6 d	19 ± 0.7 d	17 ± 0.1 d
25	19 ± 0.5 e	19 ± 0.1 e	19 ± 0.5 d	18 ± 0.3 de	16 ± 0.1 e	15 ± 0.4 e
30	21 ± 0.4 de	21 ± 0.6 de	19 ± 0.5 d	17 ± 0.5 e	16 ± 0.2 e	15 ± 0.1 e

* Indicates that the same letter in a column was not significantly different at p < 0.05 level; and the different letters indicate significant differences in a column (different temperatures).

). The time required for germination had significant differences between different cultivars. Among the six cultivars, cultivar CUF101 took the shortest time to reach 50% germination at all temperatures, while cultivar AC Caribou took the longest time to reach 50% germination (Table A1).

3.2. The Relationship between Fall Dormancy of Alfalfa Cultivars and Germination and Thermal Parameters

Base temperature (T_b) and thermal time (θ_T) for 50% germination of six alfalfa cultivars were estimated using the thermal time model, and the model fitted very well for all cultivars (p < 0.001,

Table 3. Estimated base temperature (T_b) and thermal time (θ_T) for 50% germination of six alfalfa cultivars with different fall dormancy ratings.

Cultivar	Fall Dormancy Rating	Tb (°C)	θ (°C·d)	R2 *	p
AC Caribou	1	0.89	17.95	0.96	<0.001
53Q60	3	1.39	16.5	0.97	<0.001
Saranac	4	1.46	16.26	0.96	<0.001
Tango	6	1.82	14.35	0.98	<0.001
Dona Ana	7	1.72	14.29	0.97	<0.001
CUF101	9	2.23	12.15	0.98	<0.001

* Indicates the determination coefficient of the linear regression relationship between the germination rate 1/T₅₀ and the sub-optimal temperature range, which means the coefficient of determination for the equations we used to calculate T_b and θ_T.

). The dormant cultivars AC Caribou and 53Q60 had low T_b, which were 0.89 °C and 1.39 °C, respectively. Meanwhile, the non-dormant cultivars required relatively high T_b for germination, which were 1.72 °C and 2.23 °C, respectively. The base temperatures of the semi-dormant cultivars Saranac and Tango were between those two groups, which were 1.46 °C and 1.82 °C, respectively (Table 3). Contrary to the trend of the base temperature, the thermal time constant for the germination of all cultivars decreased with the increase in the fall dormancy rating. Cultivar CUF101 (FD = 9) required the lowest thermal time (12.15 °C·d), while cultivar AC Caribou (FD = 1) required the highest thermal time for germination (17.95 °C·d, Table 3).

Although the correlation analysis showed that there was no significant correlation between the final germination percentage and fall dormancy at each temperature (p > 0.05,

Table 4. Correlation analysis of alfalfa fall dormancy rating and germination percentage.

Temperature (°C)	r	p
5	−0.446	0.38
10	−0.783	0.07
15	−0.346	0.50
20	−0.584	0.22
25	−0.691	0.13
30	−0.526	0.28

), the regression analysis demonstrated a strong positive correlation between germination rate and fall dormancy rating across all temperature conditions (Figure 2). The minimum R² value of the regression equation is 0.85, with the highest correlation observed at 30 °C (R² = 0.97, p < 0.01, Figure 2f). Similarly, the base temperature of germination increased significantly with the fall dormancy rating (R² = 0.95, p < 0.01), but the thermal time required for germination significantly decreased with the increasing fall dormancy (R² = 0.98, p < 0.01, Figure 2g,h).

4. Discussion

All alfalfa cultivars exhibited a relatively high germination percentage at all the tested temperatures including 5 °C in our study, although the effect of temperature, cultivars, and their interaction on the final germination percentage was significant. Previous studies have also shown that alfalfa germinates at a wide range of temperatures. For example, alfalfa can germinate within the temperature range of 0–1 °C and can be treated as one of the most cold-resistant leguminous forages [1]. Alfalfa can also germinate at high temperature. Alfalfa showed an excellent germination at 36 °C in a previous study [34]. Brar et al. [35] have investigated the germination of 20 leguminous forages including alfalfa and indicated that the optimal temperature for germination was between 15–25 °C. The best temperature for germination in distilled water of the alfalfa cultivar CUF101 was also shown to be 25 °C [36]. In the present study, the optimal temperature for germination of all the cultivars was 25 or 30 °C considering both germination percentage and germination rate. This may be an adapting strategy of seeds to the environment, and germination is prone to occur rapidly when seedlings can survive [37].

In our study, the effect of temperature on the germination rate of alfalfa cultivars was substantially greater than that on the final germination percentage, which is consistent with previous studies [38,39]. The alfalfa cultivars took around four days to reach 50% germination at 5 °C, while they reached 50% germination within one day at 20–30 °C. However, there were no significant differences in the final germination percentages between 5 °C and 20–30 °C. These results indicated that low temperature retarded the radicle extension of alfalfa and increased the lag time for germination, but did not induce seed damage or reduce the final germination percentage. The germination rates of all alfalfa cultivars increased with the increasing temperature until the optimum temperature at 25° for the cultivars AC Caribou, 53Q60, Saranac, and Dona Ana, and 30 °C for the cultivars Tango and CUF101.

The germination rates of different alfalfa cultivars responded differently to the temperature in our study. This might be correlated with the genetic variation or the adaptability of each cultivar to their habitat or origin [40]. Below the optimal temperature, differences have been shown in the rate of germination and root growth among alfalfa cultivars, and some cultivars exhibit better establishment potential than others [41,42]. One study has revealed that rapid germination occurs more frequently in non-wintering varieties compared to wintering varieties, although there are both fast- and slow-germinating varieties of alfalfa in all winter hardiness categories [43]. It has also been found that the germination percentage and rate of rice under low temperature treatment reflected differences in cold resistance among cultivars [44]. In our study, a significant difference in the germination rate was observed among the cultivars, and the germination rate was significantly positively correlated with the fall dormancy rating under various temperatures. Cultivars with high fall dormancy exhibited a higher germination rate, whereas cultivars with low fall dormancy showed a lower germination rate. This is consistent with the previous study on rice [44]. However, different from previous studies, the greatest variation in germination rate between cultivars was observed at a relatively high temperature (30 °C), instead of at a low temperature. On one hand, cultivars with a relatively lower germination rate decreased from 25 °C to 30 °C, while cultivars with a relatively higher germination rate continued to increase. Therefore, the variation among cultivars was enlarged, which might be the reason for the greatest correlation between germination rate and fall dormancy at 30 °C. On the other hand, the germination rate in this study was calculated as the reciprocal of the time taken to reach 50% germination, and all tested cultivars showed 50% germination within 21 h under 25 or 30 °C. We counted germination every six hours for the fast germination period. However, most of the previous studies observed germination each day and may have missed the significant differences among cultivars under high temperatures. For example, 22 h and 15 h are both within one day; germination rates calculated based on them have a significant difference, e.g., the germination rate at 15 °C and 25 °C for cultivar CUF101 (Figure 1b and Table A1). Klos [45] showed that alfalfa seeds with faster germination selected under low temperatures can result in increased plant height in the field after two growth cycles. In combination with the findings of our study, alfalfa cultivars with higher fall dormancy levels exhibited relatively fast germination under all temperatures. This may be a consequence of breeding for alfalfa with a high fall dormancy rating (non-dormant type), and thus, could be reflected in increased plant height in field conditions.

The significant relationship between germination rate and fall dormancy makes it possible to apply the thermal time model and the model parameters for predicting the fall dormancy rating. It has been reported that the base temperature is the characteristic for a seed lot or even for a species and the variation arises from disparities in their respective points of origin [46]. For example, the base temperature of winter annual species is found to be lower than that of summer annuals [47] and the base temperature of tropical species is higher than that of temperate species [31]. Consistent with the previous studies on wild species, our study showed that the base temperature for fall dormant cultivars adapted to colder regions was higher than that for non-dormant cultivars adapted to warmer regions. Specifically, for the cultivar AC Caribou with a fall dormancy level of 1, the minimum germination temperature was even below 1 °C. For another important parameter, thermal time constant, some studies suggest that the thermal time constant may vary within seed lot and exhibit normal distribution. Other studies found that tropical species with high T_b had lower thermal time required, whereas temperate species with low T_b had higher thermal time required [48,49]. Opposite to the base temperature–fall dormancy relationship, there was a significantly negative correlation between the thermal time requirement and the fall dormancy rating in our study. The cultivars with higher fall dormancy ratings had lower thermal time requirements compared to cultivars with lower fall dormancy ratings. Jungers et al. [28] also found that the thermal time was negatively correlated with T_b in alfalfa, which supports the hypothesis that seeds with lower T_b require more thermal time to germinate.

There are strong correlations between germination rate, base temperature and cumulative temperature requirement with alfalfa fall dormancy rating.The coefficient of determination (R²) for the linear regression equations predicting fall dormancy based on the germination rate at 30 °C, base temperature, and thermal time are 0.97, 0.95, and 0.98, respectively, which are close to the 0.995 achieved by traditional harvesting methods [50] and surpass the 0.94 obtained from near-infrared reflectance spectroscopy [18]. This suggests that determining fall dormancy rating based on germination rate, thermal time model parameters are a highly accurate and effective method. For instance, if a specific alfalfa variety exhibits a germination rate of 1.15 at 30 °C, the corresponding fall dormancy rating for this variety is approximately 2 (calculated based on the equation depicted in Figure 2f). Similarly, if the germplasm of a particular alfalfa genotype has a T_b value of 1.6 °C, its fall dormancy rating would be determined as 5 (according to the equation presented in Figure 2g). Using these germination characteristics to predict the fall dormancy rating of unknown cultivars or lines is straightforward and can significantly shorten the selection and breeding cycles. In the future, further experiments with a wider range of varieties with varying fall dormancy ratings may be needed to validate these relationships, allowing for broader applications of the thermal time model predicting alfalfa fall dormancy.

5. Conclusions

In conclusion, germination rates and base temperature are positively related to fall dormancy ratings and thermal time constant is negatively related to the fall dormancy ratings of the six alfalfa cultivars we used. For accurate model parameters evaluation, frequent observation of alfalfa seed germination is needed, especially for the first few days of experiment and for relative high temperature treatments. The thermal time model based method is effective to predict fall dormancy of unknown alfalfa cultivars or germplasm, which is paramount and useful for alfalfa selection and breeding. This method deserves further studies on more alfalfa varieties with wider fall dormancy rating range and on field validation.