HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Klos, K. L.E. Articles by Brummer, E.C. Search for Related Content PubMed Articles by Klos, K. L.E. Articles by Brummer, E.C. Agricola Articles by Klos, K. L.E. Articles by Brummer, E.C.

CROP BREEDING, GENETICS & CYTOLOGY

Field Response to Selection in Alfalfa for Germination Rate and Seedling Vigor at Low Temperatures

Department of Agronomy, Iowa State University, Ames, IA 50011 USA

brummer{at}iastate.edu

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Successful establishment of alfalfa (Medicago sativa L.) stands in early spring may require emergence and seedling growth at soil temperatures below 10°C. The objectives of this experiment were to evaluate changes in emergence and seedling height after laboratory selection in six cultivars at suboptimal temperature for (i) early germination (EG), (ii) high seedling vigor (HSV), (iii) EG + HSV, or (iv) late germination and low seedling vigor (LG + LSV). Cycles (C) 0 to 2 were evaluated for emergence 8 d after planting, seedling height (SH) 27 d after planting, forage dry matter yield, and other agronomic traits in field trials at Ames and Nashua, IA, in early spring 1998. Emergence was improved in all selected populations at Ames but not at Nashua; HSV and EG + HSV selection were most effective at improving emergence. After two cycles, seedling height was increased 21% by HSV selection and 9% by EG selection; however, the response among cultivars was highly variable. Combined EG + HSV selection was less effective than HSV alone at increasing height. Selection for LG + LSV did not reduce seedling height in the field even though large decreases were previously observed in the laboratory. Most gain from selection was realized with C1, possibly because the variable seed production environment in the greenhouse may have limited more consistent responses in C2. Seedling height selection at suboptimal temperatures in the laboratory successfully improved seedling height in the field in some alfalfa populations without changing other agronomic characteristics.

Abbreviations: C, cycle • EG, early germination • GT, germination time • HSV, high seedling vigor • LG, late germination • LSV, low seedling vigor • SH, seedling height

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
MOST ALFALFA is planted in the north-central USA during the spring to take advantage of available moisture and to allow harvest in the seeding year (Barnes and Sheaffer, 1995; Vough et al., 1995). Alfalfa seed planted under optimum conditions emerges rapidly, usually within 17 d after planting, but spring soil temperatures in this region are often suboptimal for alfalfa germination and growth (Brar et al., 1991; McElgunn, 1973; Vough et al., 1995; Weihing, 1941). At suboptimal temperatures, alfalfa cultivars differ in germination and root growth rates, and some cultivars may have better potential for good stand establishment than others (Brar et al., 1990, 1991; Klos, 1999).

Field emergence of alfalfa 8 d after planting is positively correlated with radicle growth rate and negatively correlated with germination time (GT) measured in the laboratory (Klos, 1999). Germination time was defined at the average days from planting to germination for a population of seeds (Klos, 1999). Germination rate assessed at low temperatures in the laboratory has been reported to be a good test for field tolerance to temperatures suboptimal for germination and growth in other legumes, including soybean, Glycine max (L.) Merr., and common bean, Phaseolus vulgaris L. (Dickson and Boettger, 1984; Littlejohns and Tanner, 1976; Szyrmer and Szczepanska, 1981).

Selection under laboratory conditions for GT and seedling vigor under low temperatures was conducted in six alfalfa cultivars (Klos and Brummer, 2000). Two cycles of selection for early germination at 5°C decreased the average GT in the laboratory by 29%. Improvement was also observed for seedling vigor as estimated by SH measured at a fixed period after germination. Seedling height increased by 15% after two cycles of selection at 10°C. McConnell and Gardner (1979) used phenotypic recurrent selection on two maize populations to increase germination percentage under laboratory conditions at 7.2°C by 8.8 and 9.9% per cycle, but field emergence and seedling vigor (on a scale of 1–9) of these populations were unchanged, possibly because of warm weather during testing.

The objectives of this study were to assess changes in emergence and seedling height in the field after two cycles of phenotypic recurrent selection for GT and seedling vigor at suboptimal temperatures under laboratory conditions and to compare these results with those observed in laboratory evaluations (Klos and Brummer, 2000).

	Materials and methods

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
The six populations chosen for recurrent selection represent commonly available cultivars in the midwestern USA in Fall Dormancy Groups 2 through 4 (Table 1) . Untreated seed was obtained either from commercial sources or from Dr. T.A. Campbell, USDA-ARS.

Selection for Early Germination at 5°C
Three hundred scarified seeds of each population were germinated on blotting paper in an environmentally controlled chamber at 5°C and 100% relative humidity. The first 15 (5%) seeds to germinate were selected.

Selection for High Seedling Vigor at 10°C
Three hundred scarified seeds of each population were planted in the greenhouse at 25°C for 3 d to allow even germination and then transferred to a growth chamber at 10°C with 20 h of light. After 45 d in the chamber, seedling vigor was estimated by seedling height, and the tallest 15 (5%) plants were selected. Stem diameter was used as a secondary criterion to select between plants of equal height.

Selection for Combined Early Germination and High Seedling Vigor
Three hundred scarified seed of each population were germinated at 5°C as described above for EG selection. The first 30 (10%) seeds to germinate were transferred to a growth chamber as described for HSV selection. After 45 d, the tallest 15 (50%) were selected.

Selection for Combined Late Germination and Low Seedling Vigor
This method was conducted as EG + HSV selection except that the last 30 (10%) seeds to germinate at 5°C were moved to the 10°C growth chamber for 45 d, when the shortest 15 (50%) of seedlings were selected.

The 15 selected plants for each population and method were transplanted into 3.5-L pots in the greenhouse. Plants within each selected population were intercrossed by hand tripping florets without emasculation. Equal amounts of seed were bulked from each plant in the selected population for use in the next cycle of selection. Fifty-four populations were created after two cycles of selection.

In addition to the four selection methods, control populations of `5454' and `Innovator +Z' were developed by intercrossing 15 randomly chosen plants per cycle to create the Control–C1 and Control–C2 populations.

Evaluation
The experiment was planted at the Iowa State Univ. Agronomy and Agric. Eng. Research Farm west of Ames, IA, on 4 Apr. 1998 on a Nicollet loam soil (fine-loamy mixed, superactive, mesic Aquic Hapludoll) and at the Northwest Research Farm near Nashua, IA on 14 Apr. 1998 on a Readlyn loam soil (fine-loamy mixed mesic Aquic Hapludoll). The 70 entries included the original six cultivar populations from which selection was initiated (i.e., Cycle 0), the 48 C1 and C2 populations from all six cultivars for the four different selection methods, seven C3 populations, two Control–C1 and two Control–C2 populations, and five check cultivars (which are not discussed further). The entries were planted in a rectangular triple lattice design. An experimental unit consisted of 50 scarified seeds of each population planted 1 cm deep in rows 1 m long and 75 cm apart. Seedling emergence (number of plants) was recorded 8 d after planting. At 27 d after planting, SH was recorded, and plots were thinned to 10 plants, uniformly spaced within the row. Seedlings were counted 27 d after planting at Ames; no count was recorded at Nashua, where insect damage occurred after the 8-d count, biasing the resulting data. Maturity at first harvest was estimated as the mean developmental stage of the plot (Kalu and Fick, 1981). The average height of the tallest upright stem of each plant, dry matter yield, and average regrowth height 7 d after harvest were recorded on all plots for three harvests at each location.

For the analysis of variance, locations and entries were considered fixed effects. The response to selection was calculated as the difference between the C1 or C2 and the C0, or, for 5454 and Innovator +Z, between the C1 or C2 and the Control–C1 or Control C–2 populations. Statistical analyses were conducted using the MIXED, CORR, and GLM procedures of the SAS statistical software package (Littell et al., 1996; SAS Institute, 1996). Mean separations were determined using Fisher's protected LSD (P < 0.05) (Steel and Torrie, 1980). Correlations among agronomic traits were computed using population means at each location.

	Results and discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Averaged across all C0 populations, seedling height 27 d after planting, first-harvest plant height, first-harvest dry matter yield, and total dry matter yield were greater at Nashua than at Ames (Table 2) . No differences in 8-d emergence or fall regrowth were present between locations. A location x population interaction was observed for all traits, except maturity, plant height at first harvest, and total dry matter yield, primarily as a result of changes in magnitude rather than rank (data not shown).

Control (Unselected) Populations
No differences in SH were observed between C0 populations and unselected control populations of 5454 and Innovator +Z (Table 3) . The lack of change suggests that the changes in SH that we observed in some selected populations were probably not caused by genetic drift or inbreeding depression. However, because only a single control population was developed and evaluated in only two of the cultivars, we cannot conclusively eliminate either of these causes in interpreting our results. Control populations and C0 populations did not differ in number of plants emerged 8 d after planting, except the Control–C2 populations developed from 5454 and Innovator +Z at Ames, which had significantly greater emergence than their C0s (data not shown). Genetic drift could have been a cause of increased emergence in this population, but this is unlikely in view of similar increases observed in the majority of populations produced in the greenhouse as compared with the C0 populations, which had been produced under various field conditions (Table 2). Because the C0 and Control populations did not differ substantively, we are confident that comparisons between C0 and the various selected populations are valid.

View larger version (21K): [in this window] [in a new window]   Fig. 1 Average daily soil temperatures at a depth of 10 cm for Ames, IA and Nashua, IA, from 1 Apr. 1998 through 15 May 1998

High Seedling Vigor Selection
After the first cycle of HSV selection, the average 27-d SH increased 25% at Ames and 19% at Nashua from C0 (Table 2). The second cycle of selection was ineffective at further increasing SH at either location. All cultivars, except Magnum IV, had a SH increase in at least one location, with Alfagraze and Amerigraze 401+Z having the most consistent responses (Table 3). Only Amerigraze 401+Z had a positive increase in SH from C1 to C2, which was observed only at Nashua. The C1 population of Innovator +Z was improved compared with both C0 (P < 0.05) and the Control–C1 (P < 0.05 at Ames; P < 0.10 at Nashua), but the C2 population was inferior to the C1 (P < 0.10 at Ames; P < 0.05 at Nashua). Similar to the other selection methods, emergence was improved with HSV selection (Table 2). Selection for HSV did not affect any of the other traits studied (Table 2).

Bidirectional Combined Selection
On average, EG + HSV selection increased 27-d SH compared with C0 at both locations (Table 2). The location x population interaction for SH was significant. At Ames, the EG + HSV–C1 populations, but not the C2, were taller than C0. At Nashua, EG + HSV–C2 populations, but not C1, were taller than C0 (Table 2). Response to EG + HSV selection among the six cultivars was inconsistent, with significant increases in 27-d SH observed at only one location for Alfagraze, Amerigraze 401+Z, and WL252HQ (Table 3), none of which had corresponding increases in SH when evaluated in the laboratory (Klos and Brummer, 2000). The selection intensity used in the EG + HSV method may not have been great enough to produce a consistent change in field performance for either of these traits. Alternatively, improving the two traits simultaneously may be difficult due to linkage or pleiotropy. Despite large reductions in height and delayed germination when evaluated in the laboratory (Klos and Brummer, 2000), selection for LG + LSV did not result in substantial changes in field performance. Compared with C0, plant height at first harvest decreased for LG + LSV–C1 at Ames, but C2 was not different from C0 (Table 2). At Nashua, LG + LSV–C2 populations, but not C1, were shorter than C0 (Table 2).

Correlations among Traits
The number of plants emerged 8 d after planting was significantly (P < 0.05) but weakly correlated with first-harvest yield at both locations, first-harvest height at Nashua, and total yield and fall regrowth at Ames (Table 4) . Seedling height 27 d after planting was positively correlated with first-harvest height and yield, total yield, and fall regrowth. None of the correlations was strong, suggesting that increased emergence and seedling vigor can improve plant growth at later stages of maturity to only a limited extent. Among the other traits evaluated, only first-harvest yield and total yield were highly correlated.

Across all four selection methods, no correlation between laboratory germination time and 8-d field emergence were observed based on the mean percentage of change across all populations for each cycle (r = -0.020; P = 0.96). However, correlations of SH response in the laboratory and field were high (r = 0.91; P < 0.001). These relationships suggest that SH would be a better trait on which to base predictions for field performance than GT if the traits were to be measured in the laboratory.

Prospects for Use in Breeding Programs
Because the soil is often cold when alfalfa is seeded in the spring, selection for improved germination and growth at suboptimal temperatures could enhance stand establishment. We examined the possibility of using laboratory selection for improved germination and/or seedling height to improve emergence and seedling height in the field. Although the first cycle of selection of the various methods was generally successful in altering the desired traits, the second cycle often contributed no further gain and in some cases reversed the gain observed in C1. The seed used to initiate the selection program was grown in commercial seed production fields, but the C2 parents were selected from greenhouse-produced seed, which was larger in mass than C0 and C2 seed. The different seed production environments may have resulted in different seed quality, which hindered our ability to make selections accurately for C2. However, unselected control populations, seed of which was produced in the greenhouse, performed similarly to the C0 base populations. Thus, we suggest that much of the selection gain we observed from C0 to C1 is real.

Gain from selection was often inconsistent across populations. Considering all populations, however, most methods of selection were effective at improving GT and/or SH (Table 5). If we consider the six cultivars as replicated selections, then these methodologies have obvious promise. The variation among individual cultivars could have resulted from experimental problems, such as controlling the greenhouse environment when producing seed. Other causes for the variability in results could have been genetic drift and inbreeding depression. Genetic drift can be significant in small populations (Falconer and Mackay, 1996), though it is probably less severe in tetraploid than in diploid populations of the same size since the former have twice as many chromosomes (and alleles). Inbreeding depression, though potentially severe in alfalfa, was probably not great after only two generations (Busbice, 1969). No evidence for either drift or inbreeding depression was observed in the control populations.

Selection by any of these methods had no affect on the first-harvest yield, total yield, maturity, or fall dormancy of the cultivars. Improvement in field SH with the various methods of selection was highly variable among cultivars, but was clearly observed in some populations. Though the percentage improvement was quite good in a number of cases, the absolute increase in seedling height or germination time was relatively small. Whether this improvement could have a significant effect on stand establishment under more adverse conditions than we tested needs to be evaluated. A location x population interaction was observed for SH, due to both rank and magnitude differences between locations, suggesting that future evaluations of response to selection for this trait should be performed at multiple locations. In cases where variability due to environment is high and heritability is low, progeny test methods of selection may be more effective than individual plant selection (Hill et al., 1971; Hill and Haag, 1974). Exploring alternate selection schemes to maximize gain from these procedures is warranted.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Journal Paper no. J-18410 of the Iowa Agric. and Home Econ. Exp. Stn., Ames, IA 50011, Project no. 2569, supported by Hatch Act and State of Iowa Funds.

Received for publication July 12, 1999.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 

• Barnes, D.K., and C.C. Sheaffer. 1995. Alfalfa. In R.F Barnes, et al. (ed.) Forages. Vol. 1: An introduction to grassland agriculture. Iowa State Univ. Press, Ames.
• Brar G.S., Gomez J.F., McMichael B.L., Matches A.G., Tailor H.M. Root development of 12 forage legumes as affected by temperature. Agron. J. 1990;82:1024-1026.[Abstract/Free Full Text]
• Brar G.S., Gomez J.F., McMichael B.L., Matches A.G., Tailor H.M. Germination of twenty forage legumes as influenced by temperature. Agron. J. 1991;83:173-175.[Abstract/Free Full Text]
• Busbice T.H. Inbreeding in synthetic varieties. Crop Sci. 1969;9:601-604.[Abstract/Free Full Text]
• Dickson M.H., Boettger M.A. Emergence, growth, and blossoming of bean (Phaseolus vulgaris) at suboptimal temperatures. J. Am. Soc. Hort. Sci. 1984;109:257-260.
• Falconer D.S., Mackay T.F.C. Introduction to quantitative genetics, 4th ed Harlow, UK: Longman, 1996.
• Hill R.R., Jr., Haag W.L. Comparison of selection methods for autotetraploids. I. Theoretical. Crop Sci. 1974;14:587-590.
• Hill R.R., Jr., Pedersen M.W., Elling L.J., Cleveland R.W., Graham J.H., Frosheiser F.I., Starling J.L. Comparison of expected genetic advance with selection on the basis of clone and polycross progeny-test performance in alfalfa. Crop Sci. 1971;11:88-91.
• Kalu B.A., Fick G.W. Quantifying morphological development of alfalfa for studies of herbage quality. Crop Sci. 1981;21:267-271.
• Klos, K.L.E. 1999. Variation in alfalfa (Medicago sativa L.) for germination and seedling vigor at suboptimal temperatures; and laboratory and field responses to selection within six alfalfa populations. Ph.D. diss. Iowa State Univ., Ames. (Diss. Abstr. ISU 1999 K58)
• Klos K.L.E., Brummer E.C. Response of six alfalfa populations to selection under laboratory conditions for germination and seedling vigor at low temperatures. Crop Sci. 2000;40:959-964 [July-Aug].[Abstract/Free Full Text]
• Littell R.C., Milliken G.A., Stroup W.W., Wolfinger R.D. SAS system for mixed models. Cary, NC: SAS Inst, 1996.
• Littlejohns D.A., Tanner J.W. Preliminary studies on the cold tolerance of soybean seedlings. Can. J. Plant Sci. 1976;58:373-375.
• McConnell R.L., Gardner C.O. Selection for cold germination in two corn populations. Crop Sci. 1979;19:765-768.[Abstract/Free Full Text]
• McElgunn J.D. Germination response of forage legumes to constant and alternating temperatures. Can. J. Plant Sci. 1973;53:797-800.
• SAS Institute. 1996. SAS/STAT guide for personal computers. Version 6.12 ed. SAS Inst., Cary, NC.
• Steel R.G.D., Torrie J.H. Principles and procedures of statistics: A biometrical approach, 2nd ed New York: McGraw-Hill, 1980.
• Szyrmer J., Szczepanska K. Screening of soybean genotypes for cold-tolerance during germination. Z. Pflanzenzucht. 1981;88:255-260.
• Weihing R.M. Field germination of alfalfa seed submitted for registration in Colorado and varying in hard seed content. J. Am. Soc. Agron. 1941;32:944-949.
• Vough, L.R., A.M. Decker, and T.H. Tailor. 1995. Forage establishment and renovation. In R.F Barnes, et al. (ed.) Forages. Vol. 2: The science of grassland agriculture. Iowa State Univ. Press, Ames.

This article has been cited by other articles:

				M. D. Casler and D. J. Undersander Selection for Establishment Capacity in Reed Canarygrass Crop Sci., April 25, 2006; 46(3): 1277 - 1285. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Klos, K. L.E. Articles by Brummer, E.C. Search for Related Content PubMed Articles by Klos, K. L.E. Articles by Brummer, E.C. Agricola Articles by Klos, K. L.E. Articles by Brummer, E.C.