Published in Crop Sci. 44:535-541 (2004).
© 2004 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
SEED PHYSIOLOGY, PRODUCTION & TECHNOLOGY
Vigor Tests Used to Rank Seed Lot Quality and Predict Field Emergence in Four Forage Species
Y. R. Wang*,a,b,
L. Yub,
Z. B. Nanb and
Y. L. Liua
a The College of Life Science, Nanjing Agricultural University, Nanjing 210095, China
b Gansu Grassland Ecological Research Institute, Lanzhou University, P.O. Box 61, Lanzhou 730020, China
* Corresponding author (yrwang66{at}public.lz.gs.cn).
 |
ABSTRACT
|
|---|
Laboratory and field tests were conducted in 1999 and 2000 in Gansu Province, China, to investigate the suitability of various laboratory vigor tests, to rank quality of commercial seed lots, and to predict seedling field emergence (FE) of four forage species. Species used included two legumes, purple vetch (Vicia benghalensis L.) and alfalfa (Medicago sativa L.), and two grasses, sudangrass [Sorghum bicolor (L.) Moench subsp. drummondii (Steud.) de Wet ex Davidse] and Siberian wild ryegrass (Elymus sibiricus L.). Results showed that among all tests, the electrical conductivity (EC) test provided the best estimate of seed vigor for the two legume species, both for ranking seed lots quality and predicting FE, but gave the worst indication of seed lot vigor for the two grass species. The EC result was not only poorly related to FE, but also poorly related to the standard germination (SG) of a wide range of seed lots of the two grass species that varied in viability. The controlled deterioration (CD) test generally better indicated seed lot vigor for test species than did SG, except that it lacked sensitivity in ranking seed lot quality of alfalfa and was not significantly related to Siberian wild ryegrass FE. The germination index (GI) better indicated seed lot quality and predicated FE than SG of sudangrass over the 2-yr results, but not consistently better than SG for other species. Initial count of standard germination (SGi) generally performed more poorly than the other vigor tests. From this study and previous work on forage species, we conclude that EC test for forage legumes and CD tests for both forage legumes and grasses have the potential to be developed as improved vigor tests for ranking seed lot quality and predicting seeding performance.
Abbreviations: CD, controlled deterioration • EC, electrical conductivity • FE, field emergence • GI, germination index • RL, radical length • SG, standard germination • SGi, initial count of standard germination
|
INTRODUCTION
|
|---|
THE OBJECTIVE of the SG test is to estimate the germination potential of a seed lot, which can then in turn be used to compare the quality of different lots and also estimate the value of field planting (ISTA, 1999). The methodology of SG has been refined to a high level of reproducibility and reliability; however, SG does not always indicate seed lot potential performance, especially if field conditions are less than optimal (Hampton and TeKrony, 1995). Seed lots that do not differ in germination may differ in deterioration level (Delouche and Caldwell, 1960) and they may differ substantially in field or storage performance (Perry, 1980; Naylor, 1981; Powell and Matthews, 1984; Kolasinska et al., 2000). Seed vigor tests therefore have been proposed to detect differences in potential seed lot performance. The critical requirements of a vigor test include (i) it must provide a more sensitive index of seed quality than does SG (McDonald, 1980; Perry, 1984) and (ii) it must better predict planting value of high germinating seed lots than does SG (Hampton and TeKrony, 1995). Although some vigor test methods have been developed and successfully used for several major agronomic crops (AOSA, 1983; Hampton and TeKrony, 1995), these methods have not been studied extensively in forage crops (Wang et al., 2001).
Wang and Hampton (1989)(1991) reported that CD and EC tests were more sensitive and accurate for predicating red clover (Trifolium pratense L.) storability and field emergence than the SG and germination index tests. Similar results were obtained for alfalfa (Wang et al., 1992, 1996). Happ et al. (1993) demonstrated that seedling growth, EC, and accelerated aging tests could differentiate seed quality among commercially acceptable perennial ryegrass (Lolium perenne L.) seed lots. Marshall and Naylor (1985) found that CD, respiration, and moisture stress tests could indicate field performance of Lolium multiflorum Lam.
The objectives of this study were to examine various seed vigor tests for their ability to rank seed lot quality and to predict seedling field emergence in each of four forage species—purple vetch, alfalfa, sudangrass, and Siberian wild ryegrass—under semiarid sowing conditions. These species are important forage crops in semiarid regions of the world and they play a significant role in livestock production and environmental protection. Furthermore, little information is currently available on seed vigor tests for these species.
|
MATERIALS AND METHODS
|
|---|
Laboratory and field tests were conducted in 1999 and 2000 in Gansu province, China. All laboratory tests were conducted at the Herbage Seed Testing Center (HSTC) of Gansu Grassland Ecological Research Institute (GGERI). Field trials of 1999 were done at Lanzhou, those of 2000 were done at Yuzhong, both in Gansu province.
Seeds
Seed samples were provided by HSTC and were submitted originally as commercial seed lot samples by various seed companies during the previous 3 yr. All samples were kept at 4°C before use. Number of lots for each of test in each year is indicated in the Table 1. The laboratory germination and CD tests were performed in both years.
Laboratory Germination
The laboratory germination test was done on four replicates of 50 seeds per seed lot. Each day, normal seedlings at least 1.5 cm long were counted and removed until no further seeds germinated. Germination index (GI) was calculated by the formula of Maguire (1962)
where Gt = germination percentage at tth day, Tt = day of germination test.
In addition, number of normal seedlings at the initial count (Table 2) was used to assess seed quality (SGi) further. Finally, percentages of normal seedlings (SG) and hard seeds in legumes were calculated at the end of the test following ISTA rules (1999); viability of legume seeds was expressed as a total sum of SG and hard seed. Radical length was measured for 10 randomly selected seedlings per replicate for each lot.
Controlled Deterioration
The CD test used 210 seeds from each lot. Initial moisture content was measured gravimetrically according to ISTA rules (ISTA, 1999). Seeds of each replicate were placed in an aluminum bag and moisture content was adjusted to 100 g kg–1 by adding water. The amount of water to be added was calculated by the following formula:
where V = water to be added, mL; MCo = initial moisture content, %; MCr = required moisture content, %; W = seed weight, g.
After adding water, bags were turned about 10 times to allow seeds to mix well with the water. Bags were sealed and kept at 10°C for 24 h to allow a slow and even imbibition by seeds. The sealed packets were then transferred to 40°C (±0.5°C) for another 24 h. The standard germination test was subsequently conducted on the aged seeds in four replicates of 50 seeds each (Hampton and TeKrony, 1995).
Electrical Conductivity
The EC test was performed with three replicates of 50 seeds per lot for purple vetch and sudangrass, and 0.5 g of seeds for the small seeded species of alfalfa and Siberian wild ryegrass. Each replicate was weighed and soaked in 100 mL distilled water in a 150-mL flask. Flasks were then stirred, covered, and held at 20°C for 24 h. Conductivity of the water was measured with a DST-A conductivity meter (No. 2 Analytical Instrument Plant, Tianjing, China) and results were expressed in µS cm–1 g–1. Hard seeds in each sample at the end of test were examined following ISTA Rules (1999), removed, surface dried, and weighed. This weight was subtracted from the initial weight of each replicate of seeds for the two forage legumes for calculation of leachate EC (Hampton and TeKrony, 1995).
Several low quality seed lots were additionally subjected to SG and EC tests to compare different responses to EC test between legumes and grasses. The numbers of lots added were 14 for purple vetch, 10 for alfalfa, 9 for sudangrass, and 7 for Siberian wild ryegrass.
Field Emergence
In 1999, FE was tested at an experimental site of GGERI at Lanzhou City from 14 May to 25 June. Soil was sandy loam belonging to sierozem of aridisols (Li et al., 1994) and was irrigated before sowing. After planting, plots were irrigated only once at 7 d after sowing and no rainfall occurred during the test.
In 2000, FE was tested in Yuzhong County about 60 km southwest of Lanzhou city from 9 August to 8 September. The soil was loess belong to sieronzem of aridisols (Li et al., 1994). There was one 20-mm rain before sowing and one 10-mm rain at 10 d after sowing.
In both years, field plots were laid out in a randomized complete block design for each species. Four replicates of 100 seeds per lot were sown by hand at a depth of 1 cm, with spacing of about 1 cm between seeds and 40 cm between rows. Emergence counts started when the first seedling was visible and continued at 2-d intervals until no further seedlings emerged. Soil moisture content (SMC) and climatic conditions are listed in Table 3. The SMC in the 0- to 10-cm layer was measured weekly from two replicates of 10-g soil samples collected each week. Samples were taken randomly with a 50-cm-diam soil auger. The SMC was determined gravimetrically after drying at 105 to 110°C for 6 h and expressed as percentage of fresh sample weight.
Statistical Analysis
Statistical analysis was performed by the Statistica program (1992, StatSoft, Inc., Tulsa, OK). Analysis of variance with LSD was performed to rank the quality of seed lots, and arcsine transformation was applied to percentages before analysis. Simple correlation coefficients were calculated to evaluate the associations among laboratory tests and seed field emergence for each species.
|
RESULTS
|
|---|
Quality of Seed Lots Evaluated by Different Tests
Qualities of seed lots as evaluated by the different tests are given in Table 1. The SG results showed that all seed had germination or viability 
70%, which is commercially acceptable for seed in China. Laboratory tests used were able to rank the seed lots into various quality groups on the basis of LSD means separation; however, the numbers of groups ranked for each species differed among tests. We assume that a powerful test identifies more seed lot groups than a weak test.
For the two legume species, EC provided the most sensitive index in ranking seed lot quality over both years. For instance, in 1999, EC divided seed lots of purple vetch into four groups and alfalfa into three groups, whereas SG divided lots of the two species into only two groups. Similarly, in 2000 EC divided seed lots of the two legumes into three or four groups, whereas SG divided each of the two species into two groups. The CD test ranked seed lots into more groups than SG in both years for purple vetch, but not for alfalfa. The GI differentiated seed lots of each legume species into more groups than SG only in 1999. The other laboratory tests did not show better or consistently better lot discrimination than SG (Table 1).
The most sensitive parameter in ranking seed lot quality of sudangrass was GI, dividing the seed lots into three groups in both years compared with one group for SG in 1999 and two groups for SG in 2000. For Siberian wild ryegrass, GI, RL and CD tests were more sensitive than SG in discriminating quality among seed lots. The other tests did not constantly show an advantage over SG for the two grasses (Table 1).
Mean FE was generally lower than that of SG in both years and the differences varied between the species and sowings. Values of FE expressed as a percentage of SG (relative field emergence) generally followed the pattern: sudangrass > purple vetch > alfalfa or Siberian wild ryegrass (Table 1).
Correlations between Standard Germination, Vigor, and Field Emergence
The relationships among SG, vigor, and FE are listed in Tables 4 and 5. For purple vetch, all laboratory test results significantly correlated with FE except SG, SGi, and GI in 2000 and RL in both years. However, the CD and EC tests predicted FE better than did SG and the other tests. Two-year correlation coefficients of FE with other tests were smaller than those with SG (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 4. Correlation coefficients (r) between standard germination or vigor and field emergence of seed lots for two legume species tested in 1999 and 2000.
|
|
View this table:
[in this window]
[in a new window]
|
Table 5. Correlation coefficients (r) between standard germination or vigor and field emergence of seed lots for two grass species tested in 1999 and 2000.
|
|
For alfalfa, SG, CD, and EC significantly correlated with FE in both years, but correlation coefficients of CD and EC with FE over 2 yr was higher than those with SG. Some of the other vigor tests correlated significantly with FE and predicated FE better than did SG in one year, but this relation did not hold in the other year (Table 4).
In sudangrass, SGi and GI significantly correlated with FE in both years and r values of these tests were higher than those of SG. The CD correlated significantly with FE in 1999 only. The other laboratory tests were not correlated to the FE in either year (Table 5).
In Siberian wild ryegrass, no vigor tests predicted FE better than did SG (Table 5).
|
DISCUSSION
|
|---|
Most forage seeds are small and some species have dormancy. They are generally difficult to germinate under adverse field conditions (Wang et al., 2001). Field experiments both in 1999 and 2000 were conducted under semiarid conditions. We expected that FE would be variable and substantially lower than SG. This was confirmed in both years particularly for the small seeded species alfalfa and Siberian wild ryegrass. This result supports previous findings on other forage species, such as Lolium multiflorum (Marshall and Naylor, 1985), Bromus biebersteinii Roemer & Schultes (Hall and Wiesner, 1990), Lotus corniculatus L. (McKersie et al., 1981), and alfalfa (Wang et al., 1996). Poor establishment of the small seeded forage species is common worldwide; more research will be needed to address this problem.
Electrical conductivity provided a more sensitive parameter in ranking seed lot quality and showed a higher correlation with FE than did SG in two legume species over 2 yr (Table 1 and 4). Data from previous research on forage legumes similarly indicated that, compared with SG, EC was more sensitive in ranking seed lot quality in alfalfa (Wang et al., 1992, 1996) and Trifolium pratense (Wang and Hampton, 1989, 1991), and EC could provide a more reliable index for predicating FE of alfalfa in drier soils (Wang et al., 1996). Our results confirmed that herbage seed lots differ in vigor (Hampton, 1991) and that cell membrane integrity, as estimated by electrolyte leakage by the EC test, is a fundamental cause of vigor degradation (Powell, 1988). In addition, the EC test has the tremendous advantage of simplicity and rapidity, and meets the requirements for a good vigor test (Hampton and Coolbear, 1990). This test is one of two tests that have been approved by the 26th ISTA Congress held in France in 2001 to be included in the International Seed Testing Rules as a standard method for testing seeds of Pisum sativum L. (TeKrony, 2001).
Compared with SG, the EC test was less sensitive in ranking seed lots for quality (Table 1) and poorly correlated with FE for the two grass species tested (Table 5). These findings were supported by the earlier research on Lolium perenne (Bennett et al., 1998; Ching and Schoolcraft, 1968; Cookson et al., 2001), Lolium multiflorum (Marshall and Naylor, 1985), and Bromus biebersteinii (Hall and Wiesner, 1990). However, current results differed from the work on Festuca arundinacea Schreber reported by Han et al. (1995), who found that EC correlated significantly with FE both in glasshouse (r = –0.950) and field (r = –0.934) studies. Our study, by adding some results from low viability seed lots to the test data set, further demonstrated that the EC was also not correlated with SG results of the two grasses (Fig. 1)
. Seeds with poor germination or even dead seed had a similar EC as seeds with high germination. This pattern differed from the results of legumes (Fig. 2)
in which EC was significantly correlated with FE for both species. These results may help to explain why EC was significantly related to FE for the legumes, but not for the grasses in this research. However, the explanation for the poor correlation of EC with FE and SG in most grass seeds is unknown and warrants further investigation.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Relationship between standard germination and conductivity in two forage grass species. Seed lots included the total of those for vigor study in 1999 and 2000, plus several additional low viability lots.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Relationship between standard germination and conductivity in two forage legume species. Seed lots included the total of those used for vigor study in 1999 and 2000, plus several additional low viability lots. ***Significant at the 0.001 probability level.
|
|
The controlled deterioration test was initially developed to evaluate potential seed lot performance for small seeded vegetable species (Matthews, 1980; Powell and Matthews, 1981; Powell and Matthews, 1984). Subsequent research demonstrated the ability of this test to rank seed lot vigor in several forage species, for example, Trifolium pratense (Wang et al., 1994), alfalfa (Wang et al., 1996), and L. multiflorum (Marshall and Naylor, 1985). In our experiments, CD generally indicated seed lot vigor better than did SG for the test species (Tables 1, 4, and 5), with exception of being less sensitive ranking seed lot quality for alfalfa and not significantly correlated to FE for Siberian wild ryegrass. This result may have occurred because of interactive effects of hard seed content in alfalfa, which was 16% in 1999 and 10% in 2000, and low seed moisture content (100 g kg–1) used in this study. As a result, CD did not produce a wide range of germination after the aging treatment for alfalfa (Table 1). Whether a better result with CD could be achieved by changing seed moisture content during the aging treatment needs further study.
Among the other combinations between the species and laboratory tests used, GI ranked sudangrass seed lots into more groups and better predicated FE than did SG for both years (Table 1 and 5). Further work with more seed lots and sowing conditions is needed to confirm this.
From this study and previous work on forage species, we conclude that EC test for forage legumes and CD tests for both forage legumes and grasses have the potential to be developed as improved vigor tests for ranking seed lot quality and predicting seed performance.
|
ACKNOWLEDGMENTS
|
|---|
We are grateful for financial support provided by the Ministry of Science and Technology, and the Natural Science Foundation of China through the research projects G20000487044 and 30170672. The authors also thank Prof. Alison A. Powell from University of Aberdeen, UK, and Dr. Chang Gui Wan, from Texas Tech University, USA, for their critical evaluations of the paper.
Received for publication April 30, 2002.
|
REFERENCES
|
|---|
- Association of Official Seed Analysts (AOSA). 1983. Seed Vigor Testing Handbook. Contribution No. 32 to the Handbook on Seed Testing. AOSA, Las Cruces, NM.
- Bennett, J.S., J.S. Rowarth, and Q.F. Jin. 1998. Seed nitrogen and potassium nitrate influence browntop (Agrostis capillaries L.) and perennial ryegrass (Lolium perenne L.) vigour. J. Appl. Seed Prod. 16:77–81.
- Cookson, W.R., J.S. Rowarth, and J.R. Sedcole. 2001. Seed vigour in perennial ryegrass (Lolium perenne L.). Seed Sci. Technol. 29:255–270.
- Ching, T.M., and I. Schoolcraft. 1968. Physiological and chemical differences in aged seeds. Crop Sci. 8:407–409.
- Delouche, J.C., and W.P. Caldwell. 1960. Seed vigour and vigour tests. Proc. Assoc. Off. Seed Anal. 50:124–129.
- Hall, R.D., and L.E. Wiesner. 1990. Relationship between seed vigor tests and field performance of Regar meadow bromegrass. Crop Sci. 30:967–970.[Abstract/Free Full Text]
- Hampton, J.G. 1991. Herbage seed lot vigour—Do problems start with seed production. J. Appl. Seed Prod. 9:87–93.
- Hampton, J.G., and P. Coolbear. 1990. Potential versus actual seed performance—Can vigour testing provide an answer? Seed Sci. Technol. 18:215–228.
- Hampton, J.G., and D.M. TeKrony. 1995. Handbook of vigor test methods. 3rd ed. International Seed Testing Association, Zurich, Switzerland.
- Han, J.G., P.S. Mao, X.C. Pu, and M. Li. 1995. A study on the seed vigour of tall fescue. Acta Agrestia Sinica 3(4):269–275.
- Happ, K., M.B. McDonald, and T.K. Danneberger. 1993. Vigour testing in perennial ryegrass (Lolium perenne L.) seeds. Seed Sci. Technol. 21:375–381.
- International Seed Testing Association (ISTA). 1999. International rules for seed testing. Seed Sci. Technol. 27(Suppl.):27–32.
- Kolasinska, K., J. Szyrmer, and S. Dul. 2000. Relationship between laboratory seed quality tests and field emergence of common bean seed. Crop Sci. 40:470–475.[Abstract/Free Full Text]
- Li, F.X., L.H Chen, X.M. Di, B.R.Wu and A.M. Zhu. 1994. Diagnostic features of representative sierozem profiles in Lanzhou and Yili region and their position in taxonomic classification. p. 49–57. In Resent Treatises on Chinese Soil Taxonomic Classification. Editorial Committee for Research in Chinese Soil Taxonomic Classification (ed.) China Science Press, Beijing.
- Maguire, J.D. 1962. Speed of germination aid in selection and evaluation for seedling emergence and vigour. Crop Sci. 2:176–177.[Free Full Text]
- Marshall, A.H., and R.E.L. Naylor. 1985. Seed vigor and field establishment in Italian ryegrass. Seed Sci. Technol. 13:781–794.
- Matthews, S. 1980. Controlled deterioration: A new vigour test for crop seeds. p. 647–660. In Seed Production, P.D. Hebblethwaite (ed.) Butterworths, London.
- McDonald, M.B. 1980. Assessment of seed quality. HortScience 15:784–788.
- McKersie, B.D., D.T. Tomes, and S. Yamamoto. 1981. Effect of seed size on germination, seedling vigor, electrolyte leakage, and establishment of bird's-foot trefoil (Lotus corniculatus L.). Can. J. Plant Sci. 61:337–343.
- Naylor, E.E.L. 1981. An evaluation of various germination indices for predicting differences in seed vigor in Italian ryegrass. Seed Sci. Technol. 9:593–600.
- Perry, D.A. 1980. Seed vigour and field establishment. Adv. Res. Technol. Seeds 5:9–42.
- Perry, D.A. 1984. Commentary on ISTA Vigour Test Committee collaborative trial. Seed Sci. Technol. 12:301–308.
- Powell, A. 1988. Seed vigour and field establishment. Adv. Res. Technol. Seeds 11:29–61.
- Powell, A.A., and S. Matthews. 1981. Evaluation of controlled deterioration, a new vigour test for small seeded vegetables. Seed Sci. Technol. 9:633–640.
- Powell, A.A., and S. Matthews. 1984. Prediction of the storage potential of onion seed under commercial storage conditions. Seed Sci. Technol. 12:641–647.
- State Bureau of Quality Control of China (SBQCC). 2001. Rules for forage seed testing. Standards Press of China, Beijing.
- TeKrony, D.M. 2001. Report of the vigour test committee 1998–2001. Seed Sci. Technol. 29(Suppl. 1):259–263
- Wang, Y.R., and J.G. Hampton. 1989. Red clover (Trifolium pratense L.) seed quality. Proc. Agron. Soc. NZ 19:63–68.
- Wang, Y.R., and J.G. Hampton. 1991. Seed vigour and storage in ‘Grassland Pawera’ red clover. Plant Var. Seeds 4:61–66.
- Wang, Y.R., J.G. Hampton, and M.J. Hill. 1994. Red clover vigor testing—Effects of three test variables. Seed Sci. Technol. 22:99–105.
- Wang, Y.R., L. Yu, and J.H. Chen. 1992. The relationship between standard germination, vigor, moisture content and storability in alfalfa seed lots. Pratacultural Science 9(6):34–38.
- Wang, Y.R., L. Yu, and Z.B. Nan. 1996. Use of seed vigor tests to predict field emergence of Lucerne (Medicago sativa). N.Z. J. Agric. Res. 39:255–262.
- Wang, Y.R., Y.L. Liu, and Y.X. Sheng. 2001. An evaluation of seed vigour tests for forage species. Acta Prataculturae Sinica 10(1):48–57.
TOP
ABSTRACT
INTRODUCTION
