Journal of Applied Botany and Food Quality 88, 202 - 208 (2015), DOI:10.5073/JABFQ.2015.088.029 1 Grassland Institute of Animal Science and Technology College, China Agricultural University, Beijing, China 2 DLF Beijing Office, No. 8 Beichendong St., Beijing, China 3 USDA-ARS Forage and Range Research Lab, Utah State University, Logan, USA Germination response of Apocynum venetum seeds to temperature and water potential Yuping Rong1*, Hongxiang Li2, Douglas A. Johnson3 (Received November 6, 2014) * Corresponding author Summary Apocynum venetum (commonly known as luobuma or rafuma) is a shrub that is native to Eurasia. It is economically important for sand fixation, forage production, honey production, and for the production of medicine, fiber and fuel. Rapid and uniform seed germination is critical for successful crop establishment and vegetation restoration. The purpose of this study was to determine the germination responses of A. venetum seeds to temperature and water availability using hydrotime, thermal time and hydrothermal model analysis. Seed germination was relatively high for A. venetum from 25 °C to 30 °C. The base (Tb), optimum (To) and ceiling temperatures (Tc(50)) of A. venetum seed germination were 16.6, 27.0 and 45.9 °C, respectively. Values of base water potential (Ψb(g)) shifted to zero with increasing temperature, which was reflected in the greater effect of low Ψ on germination for temperatures above 30 °C. Hydrotime analysis suggested that Tb may not be independent of Ψ, and Ψb(g) may change as a function of temperature at temperatures below 30 °C. The interaction effects of Ψ and temperature reduced the ability of the hydrothermal time model to predict germination performance across temperature and Ψ conditions. Introduction Apocynum venetum is a perennial semi-shrub species that is widely distributed in the temperate regions of Asia and Europe, and is found in Iran, Afghanistan, India, Russia and China. It commonly grows in barren saline soil, desert edges, riverbanks, alluvial plains and areas surrounding reservoirs and is well adapted to desert climates (Thevs et al., 2012). A. ventum is of economic importance and is used in sand fixation and a forage, tea, medicine, and fiber. However, wild populations of this species are currently in danger of being over- exploited (Thevs et al., 2007; WesTermann et al., 2008). Previous research on A. venetum has been conducted covering fiber extraction from its stems (GonG and Fu, 2001) and processing tea and medicine from its leaves (ma et al., 2003). Seed germination is critical for its use in sand fixation and crop production. A. venetum can be propagated from seeds or cuttings; however, propagation by cuttings limits its use compared to propagation by seeds. Studies have shown that sowing seeds in a nursery and transplanting the saplings after the second year is the best way to propagate this species (TanG, 2008). Temperature and soil water content are critical factors that influence seed germination in both greenhouse and field environments. In arid environments, the water required for germination is only available for short periods. Consequently, successful crop establishment de- pends not only on rapid and uniform seed germination, but also the seed’s ability to germinate under low water availability (Fischer and Turner, 1978; Windauer et al., 2012). Knowledge of the influence of water and temperature on seed germination in A. venetum is fundamental for its use in crop production and sand stabilization. Previous research was shown that seed germination of A. venetum is difficult under conditions of high soil water content, whereas high temperatures can increase its germination (Liu, 2010). ZhanG et al. (2007) found that low concentrations of NaCl (≤50 mmol/L) promote seed germination in A. venetum, whereas high NaCl concentrations (≥200 mmol/L) reduce germination. However, limited research has been conducted on the interaction affects of water and temperature on the germination and seedling establishment of this economically species. Population-based threshold models have been used to model the ecophysiological responses of seed germination to environmen- tal factors (Finch-savaGe and Leubner-meTZGer, 2006). Seed germination attributes can be quantified using the thermal time model (θT) (Garcia-huidobro et al., 1982), hydrotime model (θH) (Gummerson, 1986) and hydrothermal time model (θHT) (Gum- merson, 1986). Seed germination responses to temperature can be characterized using three cardinal temperatures (beWLey and bLack, 1994): a base temperature (Tb) below which germination of the seed lot does not proceed, an optimal temperature (To) at which the process occurs with the highest speed and a maximum (ceilings) temperature (Tc) over which the germination process does not pro- ceed. When temperature remains constant, but water is suboptimal, progress towards the completion of germination can be quantified in hydrotime (Finch-savaGe and Leubner-meTZGer, 2006). In addition, the thermal and hydrotime models can be combined to produce a hydrothermal time (HTT) model (Gummerson, 1986; aLvarado and bradFord, 2002). The objective of this study was to characterize the germination responses of A. venetum seeds to temperature and water availability using thermal time, hydrotime and hydrothermal time analysis to test the validity of these models to A. venetum seeds. Materials and methods Seed Seeds of A. venetum were collected during October 2009 in a desert environment (41.08°N, 85.97°E, 878 m asl) of the middle Tarim River in southern Xinjiang, China. Seeds were cleaned and stored at 4 °C in a sealed glass bottle until needed for experimantation (March 2010). The mass of 1000 seeds of A. venetum used in our study was 0.36 g. Germination assays Germination assays were conducted in the Forage and Turf Grass Seed Laboratory at the Grassland Science Department, China Agricultural University, Beijing, China. Seeds were sterilized for 5 min with 10 % NaClO and then washed with distilled water. Seeds were placed on two layers of filter paper saturated with distilled water or polyethylene-glycol (PEG) solution in glass Petri dishes (90-mm inner diameter). Four replicates per treatment with 50 seeds each were placed in a growth chamber under 8 h light and 16 h dark at constant temperatures of 20, 25, 30 and 35 °C (±1 °C). Apocynum venetum seed germination 203 The water potential (Ψ) of the germination medium was controlled by different solutions of polyethylene-glycol 6000 (PEG 6000) and prepared according to micheL and kauFmann (1973) so that Ψ was -0.3, -0.6, -0.9, -1.2 and -1.5 MPa at the respective temperature. The actual Ψ at all temperatures was measured using a vapour pressure osmometer (Wescor, Inc., Logan, Model 5100C). Seeds incubated on solutions containing PEG were transferred to fresh solutions every 2 d to maintain constant water potential in the germination medium. The germination (radicle protrusion) was scored daily, and the germinated seeds were removed. Germination experiments were terminated when no new germination for three consecutive days was recorded in the four replicates of a treatment. Germination analysis Germination rates were calculated as the inverse of the time to radicle emergence. Germination times for specific percentiles of the seed population (GR50) were calculated by interpolation using curves fit to the time course data. To determine the optimal germination temperature, germination rates at 30 and 35 °C were compared. If germination at 30 °C was significantly greater than germination at 35 °C, the optimal temperature was assumed to be 30 °C. The germination time course data were analyzed using repeated probit regression analysis, as described by bradFord (1990) and dahaL and bradFord (1990, 1994). Probit analyses were conducted using the PROC PROBIT routine of the SAS statistical package, which employs a maximum-likelihood weighted regression method (SAS 9.2). Thermal time In the sub-optimal temperature range (i.e., between Tb and To), the thermal time to germination at fraction g (θT1(g)) can be characterized using the following thermal time (θT1, °C d) equation: θT1(g)=(T-Tb)tg (1) where T is the germination temperature, Tb is the base temperature and tg is the germination time of fraction g. This equation indicates that for a given seed fraction g, θT1(g) is constant at all sub-optimal temperatures when expressed on a thermal time basis as the degrees in excess of Tb multiplied by the actual time to germination. Values of Tb and θT were determined through repeated probit regression analysis (Garcia-huidobro et al., 1982) by regressing all observed germination percentages on a probit scale versus log thermal times logθT1(g) to germination, varying the value of Tb until the best fit was obtained, according to the equation: probit (g)={log[(T-Tb)tg]-log[θT1(50)]} / σθT1 (2) Where probit (g) is the probit transformation of cumulative ger- mination percentage g, θT1 (50) is the median thermal time to germination, and σθT1 is the standard deviation in logθT1 among individual seeds in the population. Above To, the germination rate decreases almost linearly until Tc is reached, which is also known as thermoinhibition (hiLLs and van sTaden, 2003). Thus, in the supra-optimal temperature range, the equation used is as follows (eLLis and buTcher, 1988): θT2=[Tc(g)-T]tg (3) Germination time courses (θT2, Tc) in the supra-optimal temperature range can be predicted in a similar way to those in the sub-optimal temperature range (Equation 4); however, the germination rate decreases with temperature, and Tc (g) varies among fractions, while the thermal time to radical protrusion is constant for all seeds: Probit (g)=[log(T+θT2)/tg-logTc(50)]/σθT2 (4) Where probit (g) is the probit transformation of cumulative ger- mination percentage g, Tc (50) is the median ceiling temperature to germination, and σθT2 is the standard deviation in log Tc (50) among individual seeds in the population. Hydortime The germination response to reduced Ψ was analyzed by the hydrotime model (bradFord, 1990; Finch-savaGe and Leubner- meTZGer, 2006) according to the equation: θH=[Ψ −Ψb(g)]tg (5) where θH is the hydrotime (MPa d) of the seeds required for germination, Ψ is the actual water potential of the germination medium (MPa), Ψb(g) is the theoretical threshold or base water potential that will just prevent the germination of fraction g and tg is the germination time (d) of fraction g. The model assumes that Ψb varies among fractions of a seed population following a normal distribution with mean Ψb(50) and standard deviation σΨb. θH is considered constant for a seed population (bradFord, 1990). The parameters in the hydrotime model were estimated according to the equation: Probit (g) = [(Ψ-θH/tg)-Ψb(50)]/σΨb (6) where Ψb(50) is the median Ψb, and σΨb is the standard deviation in Ψb among seeds within population. The parameters from the hydrotime model can be used to normalize germination time courses for the effects of reduced Ψ. Germination time course at any Ψ can be normalized to the time course that would occur in water (0 Mpa) for the seed population by multiplying the actual time to germination tg (Ψ) by the factor [1-(Ψ/Ψ(g))] (bradFord, 1990). This normalization can evaluate the ability of the model to describe the germination behavior (bradFord, 1995). All normalized data from all temperatures were normalized on a common thermal time scale, using the estimated Tb at 0 Mpa. Hydorthermal time The thermal and hydrotime models were combined to produce a hydrothermal time (HTT) model to describe germination rates when temperature and Ψ both varied. The HTT model at the sub- optimal temperature where θHT is the HTT constant (MPa °C d) was calculated according to the equation: θHT =(T-Tb) [Ψ-Ψb(g)]tg (7) This equation describes germination time courses at any combina- tion of sub-optimal temperature and Ψ (aLvarado and bradFord, 2002). The following modified hydrothermal time model was pro- posed by aLvarado and bradFord (2002) to describe the germi- nation timing and percentages across all T from Tb to Tc: θHT={Ψ-Ψb(g)-[KT(T-To)]}(T-Tb)tg (8) where [KT(T-To)] applies only when T>To and in this supra-optimal range of T; the value of Ψb(g) is set equal to Ψb(g)To and T-Tb is set equal to To-Tb. The values of KT, Tb, To and θHT in this model can be obtained by repeated probit regressions using germination time course data according to the equation: Probit (g)=[(Ψ-θHT/(To-Tb)tg)-KT(T-To))-Ψb(50)]/σΨb (9) The values of KT and To were varied for germination time courses at T>To until a fit was obtained that resulted in θHT, Ψb(50) and σΨb values close to those obtained at or below To (aLvarado and bradFord, 2002). 204 Y. Rong, H. Li, D.A. Johnson Results Germination responses to temperature and water potential The cardinal temperature of A. venetum seeds was determined by germination in distilled water at different constant temperatures. Germination of A. venetum seeds in water progressed more rapidly as T increased within the sub-optimal range (20-30 °C). In contrast, the final germination percentage in water decreased between 30 and 35 °C. The final germination percentage in the temperature range of 25 to 30 °C was relatively high with the values of Ψ used in our study, which ranged from 70 to 89 %, respectively (Fig. 1a). The germination rate increased with temperature in the range of 20 to 30 °C, and decreased above 30 °C. The germination rate (1/t50) was highest for seeds incubated at 30 °C (Fig. 1b). As a result, the effect of the PEG solutions was tested at temperatures of 20 and 25 °C, considered as sub-optimal temperatures, and 30 and 35 °C considered as supra-optimal temperatures. Thermal time analysis Accumulated daily germination percentages at 20 and 25 °C for distilled water were transformed to probit and regressed on logθT1(g)=log[(T-Tb)tg]. The number of degree days necessary for 50 % of the seeds to germinate in the sub-optimal temperature range was 29.4 °C d (θT1) (R2=0.65). The value of Tb that produced the best fit was 16.6 °C, which was taken as the minimum temperature for germination of A. venetum seeds. In the supra-optimal temperature range, the value of probit (g) was related to logTc(g)=log[(T+θT2/ tg)], and the best fit was obtained with Tc(50)=45.9 °C (R2=0.87) (Tab. 1). In the supra-optimal interval, the value of θT2=43.2 °C d was assumed to be constant, and the limiting factor for germination would be the distribution of Tc (g) within the population. The value of Tc varies among fractions of a seed population, following a normal distribution with mean Tc (50) and standard deviation (σTc). Similar analyses of germination data under of -0.3, -0.6, -0.9, -1.2 and -1.5 MPa were also performed (Tab. 1). The R2 values indicated that the thermal model for sub-optimal temperatures occurred at lower values of Ψ (R2=0.69-0.90), but Tb was underestimated and θT1(50) was overestimated. Tb decreased with decreasing Ψ, especially below -0.6 MPa. Hydrotime analysis The hydrotime model accounted for most of the variation in germination time at reduced Ψ values at various temperatures (R2=0.93-0.95) (Tab. 2). The predicted germination time courses at the various Ψ values (Fig. 2) generally fit the observed data well, except for some cases in which the predicted response deviated Fig. 1: Effect of temperature on the final germination (a) and germination rate (b) for A. venetum seeds incubated at various water potentials (Ψ) (0– -1.5MPa). Tab. 1: Parameter estimates of the thermal time model describing seed ger- mination across a range of water potentials. Water potential Tb θT1(50) σθT1 R2 (MPa) (°C) (°d) (°d) Temperature at 20, 25 Probit(g)={log[(T-Tb)tg]-log[θT1(50)]}/σθT1, θT1(g)=(T-Tb)tg 0 16.6 29.4 0.51 0.65 -0.3 15.9 42.6 0.57 0.69 -0.6 16.1 50.2 0.64 0.76 -0.9 12.9 89.1 0.45 0.90 -1.2 2.4 466.0 0.86 0.77 -1.5 0.3 773.1 0.90 0.86 Temperature at 30, 35 Probit(g)=[log(T+θT2)/tg-logTc(50)]/σθT2, θT2=[T-Tc(g) ] tg Water potential θT2 Tc(50) σTc R2 (MPa) (°d) (°C) (°d) 0 43.2 45.9 0.12 0.87 -0.3 65.7 45.7 0.19 0.95 -0.6 103.5 48.4 0.19 0.98 -0.9 78.6 40.2 0.19 0.97 -1.2 70.0 31.5 0.20 0.95 -1.5 64.6 29.4 0.16 0.99 Tb, base temperature; θT1(50), thermal time to germination of 50 % at sub- optimal T; θT2, thermal time constant at supra-optimal T; Tc(50), ceiling tem- perature to germination of 50 %; σθT1, standard deviation for θT1(50); σTc, standard deviation for Tc (50). R2, coefficient of determination. considerably from actual germination. Estimated values of θH, Ψb(50) and σΨb for various germination temperatures are shown in Tab. 2. θH decreased from 14.2 MPa d at 20 °C to 4.0, 3.1 and 3.1 MPa d at 25, 30 and 35 °C, respectively. In addition, the values of Ψb(50) increased as temperature increased from 20 °C to 35 °C, becoming less negative; at 35 °C, Ψb(50) was -0.8 MPa. The pre- dicted responses in Fig. 3 are based upon the ψb(g) threshold distributions from the hydrotime model. In general, the predicted responses described the distributions of the observed cumulative germinations percentage relatively well for the treatments at 20, 25 and 35 °C and -0.3, -0.6 and -0.9 MPa, whereas the predicted values showed poor agreement with the observations at water potentials of Apocynum venetum seed germination 205 -1.2 and -1.5 MPa. Predicted σΨb varied considerably from 1.01 to 1.95 MPa. The normalized germination curves for the hydrotime model (Fig. 4) showed that the observations from various values of Ψ at the various experimental temperatures merged into a common curve. When the interaction between temperature and Ψb was taken into account using individual estimates of Ψb(50) and σΨb at sub-optimal temperatures, the observations normalized more consistently into a common curve, whereas distinct groupings of observations remained at sub-optimal temperatures (Fig. 4a). However, the normalized curves revealed that observations fell into a distinct group at 35 °C, at which temperature the germination time was long and final germination values were low (Fig. 4b). This indicated that the hydrotime estimates interacted with temperature, and consequently, the grouping of observations was the most profound in the seed population with the largest shift in Ψb with temperature (Tab. 2). Hydrothermal time analysis The hydrothermal model of each sub-optimal temperature at diffe- rent Ψ was regressed on θHT=(Ψ-Ψb(g))(T-Tb)tg, which described the germination responses at constant sub-optimal temperatures in the Ψ range of 0 to -0.6 MPa well. The values producing the best fit are presented in Tab. 3. According to the hydrothermal model, Tb and Ψb(50) were 16.6 °C and -1.30 MPa, respectively, at sub- optimal temperatures. In the supra-optimal temperature interval, θHT=32.2 MPa d, predicted by the hydrothermal time model (Tab. 3), was used to fit the germination data across various Ψ at each T (30, 35 °C). Because the distributions of Ψb for various temperatures were pooled into one common distribution in the hydrothermal time model, estimates of Ψb(50) and σΨb in the hydrothermal time model (Tab. 3) generally agreed with estimates of the hydrotime model at sub-optimal temperatures. However, the estimated values were not consistent with those of the hydrotime model at supra-optimal temperatures. Discussion In the present study, we used A. venetum seeds stored for approxi- mately 10 months, which generally exhibit a lower germinability than newly collected seeds (95 %) (hu et al., 2002b). The germination of our A. venetum seeds was 88 %, which was similar to those of hu et al. (2002a) who found that germination decreased to 81-85 % after 6 months of storage and then remained stable for more than 10 years at room temperature under sealed conditions. Many studies have shown that responses of seed germination to temperature are related to the geographical and ecological distri- bution of the particular species studied (Grime et al., 1981; schüTZ Tab. 2: Parameter estimates of the hydrotime model (Probit(g)=[(Ψ-θH/tg)- Ψb(50)]/σΨb, θH=(Ψ-Ψb(g))tg) describing seed germination. Temperature θH Ψb(50) σΨb R2 (°C) (MPa d) (MPa) (MPa) 20 14.2 -1.76 1.95 0.93 25 4.0 -1.32 1.15 0.95 30 3.1 -1.19 1.01 0.93 35 3.1 -0.80 1.23 0.94 θH, hydrotime; Ψb(50), base water potential for 50 % seed germination; σΨb, standard deviation for Ψb (g); R2, coefficient of determination. Fig. 2: Germination time course of A. venetum seeds for a range of water potentials at various constant temperatures. Symbols indicate actual data, and lines indicate values predicted by probit analysis. 206 Y. Rong, H. Li, D.A. Johnson Fig. 3: Distribution of the base water potential of A. venetum seeds at vari- ous temperatures. Tab. 3: Parameter estimates of the hydrothermal time model describing seed germination. Sub-optimal temperature (20, 25 °C): Probit(g)=[Ψ-(θHT)/(T-Tb)tg)-Ψb(50)]/σΨb, θHT=(Ψ-Ψb(g))(T-Tb)tg Tb θHT Ψb(50) σΨb R2 (°C) (MPa d) (MPa) (MPa) 16.6 32.2 -1.31 1.19 0.93 Supra-optimal temperature (30, 35 °C): Probit(g)=[Ψ-(θHT)/(To-Tb)tg)-KT(T-To))-Ψb(50)]/σΨb, θHT={Ψ-Ψb(g)-[KT(T-To)]}(To-Tb)tg To θHT Ψb(50) σΨb KT R2 (°C) (MPa d) (MPa) (MPa) (MPa °C-1) 27.0 32.2 -1.54 1.07 0.10 0.89 To, optimal temperature; θHT, hydrothermal time constant; Ψb(50), base water potential for 50% seed germination; σΨb, standard deviation for Ψb (g); R2, coefficient of determination. Fig. 4: Normalized time courses of the hydrothermal time model at sub-optimal temperatures (a), supra-optimal temperatures (b) and all temperatures com- bined (c). and rave, 1999; Liu et al., 2011). The thermal requirement of seed germination also was related to the life history strategy of the species (ProberT, 2000). A. venetum grows in Central Asia and is adapted to desert climates, surviving even under extremely arid conditions (<50 mm mean annual precipitation) by exploiting groundwater to meet its water demands (chen et al., 2007; Thevs et al., 2012). In our study, temperature had a marked effect on the germination of A. venetum at Ψ studied, and the effect was described by the thermal time model using probit analysis (Tab. 1). A. venetum seeds germi- nated in a relatively narrow temperature range with the final germi- nation reaching 70-88 % in the interval of 25-30 °C. The estimated Tb and θT1 were 16.6 °C and 29.4 °C d in distilled water, respec- tively, corresponding to a relatively high Tb and low accumulated temperature during seed germination. This is consistent with some studies reporting a negative relationship between Tb and θT1 (anGus et al., 1981; TrudGiLL et al., 2000). Species originating from tropi- cal regions have higher Tb and lower θT1 (TrudGiLL et al., 2005). Tc was high in our study, which indicated that seeds of A. venetum are able to tolerate quite high temperatures. This germination charac- teristic of A. venterum is important for desert environments, which have highly variable temperatures, precipitation and soil water avail- ability. The thermal time model did not fit the data at lower Ψ (e.g., -1.2 MPa and -1.5 MPa). The low estimate of Tb and considerably high θT1 (50) at low values of Ψ are probably not biologically signi- ficant. The thermal time analysis in our study failed to describe ger- mination response at the later phases of germination. One possible reason could be that maximal germination was not possible during Apocynum venetum seed germination 207 the 15 days used in our study. According to hu et al. (2002a), the storage of A. venetum seeds led to lengthening the germination time period at lower temperatures, which in their study was 21 days. The parameters from the thermal time analysis can be used for production conditions for A. venetum. First, Tb was estimated to be 16.6 °C; this value is quite high and precludes the sowing of these seeds in soils where the temperature does not exceed 16.6 °C. Secondly, germination rate (GR50) was found to be the highest at 30 °C, demonstrating that maximum germination rate would be at- tained at soil temperatures approaching this value. However, our analysis also revealed that final germination percentage decreased sharply at temperatures higher than 30 °C. Results from the hydrotime analysis indicated that the θH constant was reduced from 14.2 to 3.1 MPa d when the incubation tempera- ture increased to 30 °C and then remained stable at 35 °C. Seed germination was relatively low even at 20 °C, the germination rate was faster at higher temperatures (25, 30 and 35 °C), with seed ger- mination beginning on the third day. The hydrotime analysis also indicated that Ψb(50) were less negative at 30 and 35 °C than at 20 and 25 °C. According to hydrotime theory, the Ψb(50) value of a seed population gives an indication of its ability to avoid stress. Less negative values of Ψb(50) at supra-optimal temperatures result in declines of both seed germination rate and percentage germination, eventually reaching zero at a ceiling temperature. Several studies documented that changes in the dormancy state are related to changes in Ψb (meyer et al., 2000; aLvarado and bradFord, 2002; roWse and Finch-savaGe, 2003), with Ψb(50) increasing with rising temperatures. Many native species in central Asia apparently have no pronounced dormancy or lack deep dor- mancy with species following an opportunistic strategy that allows them to germinate whenever physical conditions become suitable (Wesche et al., 2006). Soil water availability is critical for seed germination with higher temperatures inducing thermoinhibition (hiLLs and sTaden, 2003). When the distribution of Ψb overlaps with 0 MPa, the proportion of seeds with Ψb larger than 0 MPa is inhibited at the given temperature (Larsen et al., 2004; WaTT et al., 2011). In our study, Ψb values above 0 MPa were observed at 35 °C (Fig. 3), which is consistent with a reduction in maximum final germination percentages with temperature. At value of less than -0.6 MPa at 25 and 30 °C, values of the θH constant and σΨb increased. A similar response was reported by dahaL and bradFord (1994) for tomato seeds, who ascribed this effect to physiological changes produced by prolonged exposure to the osmoticum. Because 30 °C was found to be optimal for seed germination (i.e., GR50 was highest at this temperature), some seeds within the population probably germinated so quickly that they escaped the water stress effect of this temperature. The hydrotime model has been effective at explaining the cardinal temperatures for seed germination in potato seeds (aLvarado and bradFord, 2002) to investigate the effect of fluctuating temperatures on the termina- tion of dormancy (benech-arnoLd et al., 2000). From a crop pro- duction standpoint, the Ψb(50) values (-1.86 to -0.80 MPa) deter- mined in our analysis suggest that seeds of A. venetum have a strong tolerance to water stress. According to aLvarado and bradFord (2002), the decrease of germination rate and percentage in the supra-optimal temperature range is due to an increase (less negative values) in the Ψb(g) thres- holds for germination. When T exceeded To, the Ψb(50) of the seed population shifted to -0.8 MPa d, which was the highest value ob- served in our study. Values of σΨb ranged from 1.01 to 1.95 MPa d, which was higher than values reported in other studies and indicates that seed germination time varied substantially. This may be a sur- vival adaptation to harsh desert environments. The extended germi- nation time for A. venetum may avoid mass germination following suitable environment conditions by allowing a few seeds to rapidly germinate under initial favorable temperature and soil water avail- ability, and then wait a bit longer for confirmation of safe germina- tion conditions before the remaining seeds germinate (baTLLa et al., 2009; WaTT et al., 2011). Conclusion This study characterized A. venetum seed responses to tempera- ture and water availability through the application of thermal time, hydrotime and hydrothermal models. Interactions between Ψ and temperature affected the ability of the hydrothermal time model to predict germination responses across temperature and Ψ conditions. Results also revealed some possible limitations in seed germination that need to be considered in working with A. venetum for crop pro- duction. The narrow thermal range for seed germination and non- uniform germination in response to Ψ and temperature variation are the most important characteristics during A. venetum seed germina- tion. These factors might sufficiently delay or even prevent germina- tion in arid and semi-arid environments. Plant breeding and selection in A. venerum may be useful in modifying these responses. Acknowledgments We gratefully acknowledge the Key Laboratory of Grassland Science for excellent technical assistance and Gebao Company in Xinjiang, China for assistance with field seed collection work. Financial support This work was funded by National Forage Production System Proj- ect (CARS-35) in China. Conflict of interest None. References aLvarado, v., bradFord, K.J., 2002: A hydrothermal time model explains the cardinal temperatures for seed germination. Plant Cell Environ. 25, 1061-1069. anGus, J.F., cunninGham, r.b., moncur, m.W., mackenZie, D.H., 1981: Phasic development in field crops. I. Thermal response in the seedling phase. Field Crops Res. 3, 365-378. baTLLa, d., Grundy, a., denT, k.c., cLay, h.a., Finch-savaGe, W.E., 2009: A quantitative analysis of temperature-dependent dormancy changes in Polygonum aviculare seeds. Weed Res. 49, 428-438. benech-arnoLd, r.L., sa´ncheZ, r.a., ForceLLa, F., kruk, b.c., Ghersa, C.M., 2000: Environmental control of dormancy in weed seed banks in soil. Field Crops Res. 67, 105-122. beWLey, J.d., bLack, M., 1994: Seeds: physiology of development and germination. New York, Plenum Press. bradFord, K.J., 1990: A water relations analysis of the seed germination rates. Plant Physiol. 94, 840-849. bradFord, K.J., 1995: Water relations in seed germination. In: Kigel, J., Galili, G. (Eds.), Seed development and germination, 351-394. Marcel Dekker, Inc., New York. chen, y., Li, G., menG, J., cao, M., 2007: Effect of sodium chloride stress on seed germination and seedling growth of Apocynum venetum L. Chi- nese Wild Plant Resources 26, 49-51 (In Chinese with English abstract). dahaL, P., bradFord, K.J., 1990: Effects of priming and endosperm inte- grity on germination rates of tomato genotypes. II. Germination at re- duced water potential. J. Exp. Bot. 41, 1441-1453. dahaL, P., bradFord, K.J., 1994: Hydrothermal time analysis of tomato seed germination at suboptimal temperature and reduced water potential. 208 Y. Rong, H. Li, D.A. Johnson Seed Sci. Res. 4, 71-80. eLLis, r.h., buTcher, P.D., 1988: The effects of priming and ‘natural’ dif- ferences in quality amongst onion seedlots on the response of the rate of germination to temperature and the identification of the characteristics under genotypic control. J. Exp. Bot. 39, 935-950. Finch-savaGe, W.e., Leubner-meTZGer, G., 2006: Seed dormancy and the control of germination. New Phytol. 171, 501-523. Fischer, r.a., Turner, N.C., 1978: Plant productivity in the arid and semi- arid zones. Ann. Rev. Plant Physiol. 29, 277-317. Garcia-huidobro, J., monTeiTh, J.L., squire, G.R., 1982: Time, tempera- ture and germination of pearl millet (Pennisetum typhoides S.& H.). I. Constant temperature. J. Exp. Bot. 33, 288-296. GonG, T.y., Fu, H.X., 2001: Development of care yarn of Apocynum spe- cies. Silk Technology Application (1), 32-33 (In Chinese with English abstract). Grime, J.P., mason, G., GurTis, a.v., rodman, J., band, s.r., moWForTh, m.a.G., neaL, a.m., shaW, S., 1981: A comparative study of germina- tion characteristics in a local flora. J. Ecol. 69, 1017-1059. Gummerson, R.J., 1986: The effect of constant temperature and osmotic po- tentials on the germination of sugar beet. J. Exp. Bot. 37, 729-741. hiLLs, P.n., van sTaden, J., 2003: Thermoinhibition of seed germination. S. Afr. J. Bot. 69, 455-461. hu, r.L., Lin, L.m., donG, Zh.J., 2002a: The effect of temperature on seeds germinating ability of kyndyt. Acta Botany Boreal-Occident Sinica 22, 6-10 (In Chinese with English abstract). hu, r.L., ZhanG, Sh.W., qian, X.Sh., Zhu, h.P., JianG, L.X., hao, Y.R., 2002b: Reserve condition of kyndyt seed. Acta Botany Boreal-Occident Sinica 22, 70-78 (In Chinese with English abstract). Larsen, s.u., baiLLy, c., côme, d., corbineau, F., 2004: Use of the hydro- thermal time model to analyse interacting effects of water and tempera- ture on germination of three grass species. Seed Sci. Res. 14, 35-50. Liu, P., 2010: The study on influential factors about the germination seeds of Apocynum. Northern Horticulture 3, 92-94 (In Chinese with English abstract). Liu, k., baskin, J.m., baskin, c.c., bu, h.y., Liu, m.X., Liu, W., du, G.Zh., 2011: Effect of storage conditions on germination of seeds of 489 species from high elevation grasslands of the eastern Tibet Plateau and some implications for climate change. Am. J. Bot. 98, 12-19. ma, m., honG, c.L., an, s.q., Li, B., 2003: Seasonal, spatial and inter- specific variation in quercetin in Apocynum venetum and Poacynum hendersonii, Chinese traditional herbal teas. J. Agr. Food Chem. 51, 2390-2393. meyer, s.e., debaene-GiLL, s.b., aLLen, P.S., 2000: Using hydrothermal time concepts to model germination response to temperature, dormancy loss, and priming effects in Elymus elymoides. Seed Sci. Res. 10, 213- 223. micheL, b.e., kauFmann, M.R., 1973: The osmotic potential of polyethy- lene glycol 6000. Plant Physiol. 51, 914-916. ProberT, R.J., 2000: The role of temperature in the regulation of seed dor- mancy and germination. In: Fenner, M. (ed.), Seeds: The Ecology of Regeneration in Plant Communities, 261-293. 2nd edition. CAB Interna- tional, Wallingford, U.K. roWse, h.r., Finch-savaGe, W.E., 2003: Hydrothermal threshold models can describe the germination response of carrot (Daucus carota) and onion (Allium cepa) seed populations across both sub- and supra-optimal temperatures. New Phytol. 158, 101-108. schüTZ, W., rave, G., 1999: The effect of cold stratification and light on the seed germination of temperate sedges (Carex) from various habitats and implications for regenerative strategies. Plant Ecol. 144, 215-230. TanG, X.Q., 2008: Study on the resource of Wild P Pictum Ball. and planting technology in Chaidam Basin. Qinghai Agriculture 17, 48-50 (In Chi- nese with English abstract). Thevs, n., Zerbe, s., GahLerT, F., muiT, m., succoW, M., 2007: Producti- vity of reed (Phragmites australis Trin. ex. Steud.) in continental-arid NW China in relation to soil, groundwater, and land use. J. Appl. Bot. Food Qual. 81, 62-68. Thevs, n., Zerbe, s., kyosev, y., roZi, a., TanG, b., abdusaLih, n., noviTskiy, Z., 2012: Apocynum venetum L. and Apocynum picutm Schrenk (Apocynaceae) as multi-functional and multi-service plant spe- cies in Central Asia: a review on biology, ecology, and utilization. J. Appl. Bot. Food Qual. 85, 159-167. TrudGiLL, d.L., honek, a., Li, d., van sTraaLen, N.M., 2005: Thermal time-concepts and utility. Ann. Appl. Biol. 146, 1-14. TrudGiLL, d.L., squire, G.r., ThomPson, K., 2000: A thermal time basis for comparing the germination requirements of some British herbaceous plants. New Phytol. 145, 107-114. WaTT, m.s., bLoomberG, m., Finch-savaGe, W.E., 2011: Development of a hydrothermal time model that accurately characterizes how thermoinhi- bition regulates seed germination. Plant Cell Environ. 34, 870-876. Wesche, k., PieTsch, m., ronnenberG, k., undrakh, r., hensen, I., 2006: Germination of fresh and frost-treated seeds from dry Central Asian steppes. Seed Sci. Res. 16, 123-136. WesTermann, J., Zerbe, s., ecksTein, D., 2008: Age structure and growth of degraded Populus euphratica floodplain forests in NW China and per- spectives for their recovery. J. Integrat. Plant Biol. 50, 536-546. Windauer, L.b., marTineZ, J., raPoPorT, d., Wassner, d., benech- arnoLd, R., 2012: Germination responses to temperature and water potential in Jatropha curcas seeds: a hydrotime model explains the dif- ference between dormancy expression and dormancy induction at dif- ferent incubation temperatures. Ann. Bot. 109, 265-273. ZhanG, X., Li, r., shi, F., 2007: Effects of salt stress on the seed germination of Apocynum venetum. Acta Scientiarum Universitatis Nankaiensis 40, 13-18 (In Chinese with English abstract). Address of the corresponding author: Department of Grassland Science, China Agricultural University, 2 Yuan- mingyuan West Rd., 100193 Beijing, China. E-mail: rongyuping@cau.edu.cn © The Author(s) 2015. This is an Open Access article distributed under the terms of the Creative Commons Attribution Share-Alike License (http://creative- commons.org/licenses/by-sa/4.0/).