Crop Science Illumina
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 September 2007
Published in Crop Sci 47:2089-2097 (2007)
© 2007 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by DeVries, M.
Right arrow Articles by Moore, K. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by DeVries, M.
Right arrow Articles by Moore, K. J.
Agricola
Right arrow Articles by DeVries, M.
Right arrow Articles by Moore, K. J.
Related Collections
Right arrow Seed Quality
Right arrow Plant and Environment Interactions
Right arrow Seed Physiology

SEED PHYSIOLOGY, PRODUCTION & TECHNOLOGY

Determining Seed Performance of Frost-Damaged Maize Seed Lots

Mindy DeVriesa, A. Susana Goggia,* and Kenneth J. Mooreb

a Iowa State Univ., 166 Seed Science Center, Ames, IA 50011
b Iowa State Univ., Agronomy Hall, Ames, IA 50011. This journal paper of the Iowa Agricultural and Home Economics Experiment Station, Ames, Project no. 3638, was supported by the Hatch Act, State of Iowa Funds

* Corresponding author (susana{at}iastate.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Seed quality of maize (Zea mays L.) can be negatively impacted by a fall frost event. It is important for the seed industry to detect frost damage early and to make marketing decisions before the seed lots are conditioned for sale. This study compared several seed quality tests (standard germination [WG], accelerated aging [AA], saturated cold [SC], and soak tests) for their ability to quantify frost damage. Additionally, these tests were used to predict field emergence under poor and average to good field conditions. Two genotypes (B73 x IRF311 and Mo17 x IRF311) were harvested at three moisture contents (300–350, 400–450, and 500–550 g H2O kg–1 fresh weight). An artificial frost treatment was applied to the seed and damage was determined by testing seed after approximately 0, 1.5, 3, 4.5, and 6 mo of storage. The artificial frost treatment significantly decreased viability and vigor of all the seed lots except Mo17 x IRF311 harvested at 300 to 350 g H2O kg–1 fresh weight. As seeds matured, the damage associated with frost treatment decreased. Laboratory tests did not accurately predict field emergence of frost-damaged seed under poor field conditions; however, AA at 0 mo, SC at approximately 0, 1.5, and 3 mo, and soak at approximately 0, 1.5, and 3 mo had strong relationships to field emergence under average to good field conditions. These results indicate that frost damage in seed lots is quantifiable using the SC and soak tests during approximately the first 3 mo of storage and both tests accurately predict field emergence.

Abbreviations: AA, accelerated aging • SC, saturated cold • WG, standard germination

Determining Seed Performance of Frost-Damaged Maize Seed Lots

Mindy DeVriesa, A. Susana Goggia,* and Kenneth J. Mooreb

a Iowa State Univ., 166 Seed Science Center, Ames, IA 50011
b Iowa State Univ., Agronomy Hall, Ames, IA 50011. This journal paper of the Iowa Agricultural and Home Economics Experiment Station, Ames, Project no. 3638, was supported by the Hatch Act, State of Iowa Funds

* Corresponding author (susana{at}iastate.edu).

Seed quality of maize (Zea mays L.) can be negatively impacted by a fall frost event. It is important for the seed industry to detect frost damage early and to make marketing decisions before the seed lots are conditioned for sale. This study compared several seed quality tests (standard germination [WG], accelerated aging [AA], saturated cold [SC], and soak tests) for their ability to quantify frost damage. Additionally, these tests were used to predict field emergence under poor and average to good field conditions. Two genotypes (B73 x IRF311 and Mo17 x IRF311) were harvested at three moisture contents (300–350, 400–450, and 500–550 g H2O kg–1 fresh weight). An artificial frost treatment was applied to the seed and damage was determined by testing seed after approximately 0, 1.5, 3, 4.5, and 6 mo of storage. The artificial frost treatment significantly decreased viability and vigor of all the seed lots except Mo17 x IRF311 harvested at 300 to 350 g H2O kg–1 fresh weight. As seeds matured, the damage associated with frost treatment decreased. Laboratory tests did not accurately predict field emergence of frost-damaged seed under poor field conditions; however, AA at 0 mo, SC at approximately 0, 1.5, and 3 mo, and soak at approximately 0, 1.5, and 3 mo had strong relationships to field emergence under average to good field conditions. These results indicate that frost damage in seed lots is quantifiable using the SC and soak tests during approximately the first 3 mo of storage and both tests accurately predict field emergence.

Abbreviations: AA, accelerated aging • SC, saturated cold • WG, standard germination


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
FREEZING INJURY causes irreversible damage in maize (Zea mays L.) seed. Every four or five years, an early fall frost in the Upper Midwest causes substantial monetary losses for the seed industry due to lower seed germination and vigor (Burris and Knittle, 1985). The loss of seed germination and vigor is the product of various physical, mechanical, biochemical, and physiological changes. The severity of these changes depends, among other things, on the developmental stage of the seed at the time of the frost, the genetic background of the maternal parent, and the extent and location of ice formation.

The plant's response to frost closely depends on the developmental stage of the plant before freezing (Kiesselbach and Ratcliff, 1920). As maize seeds mature and dehydrate, they undergo a series of physiological changes called acquisition of desiccation tolerance (Bewley and Black, 1994). The lack of acquisition of desiccation tolerance and, thus, seed immaturity, are critical factors in determining the extent of freezing injury in corn (Kiesselbach and Ratcliff, 1920; Hartwigsen, 1999).

The genetic background of the seed also plays an important role in freezing tolerance. The maternal parent has greater influence on freezing tolerance than the paternal parent, and endosperm characteristics have greater effect than the pericarp (Rossman, 1949). Recent research, however, assigns less importance to the characteristics of the endosperm in preventing or reducing the severity of frost damage in corn (Woltz et al., 2006).

Another maternal trait is husk protection, but its importance in preventing freezing injury is not fully understood. The husks act as an insulator and "buffer" against temperature changes (Rossman, 1949; Fick, 1989). The incidence of freezing injury in seed is less for ears frozen with husks intact than for ears with husks removed (Fick, 1989). Although genetic background influences the effect of frost, the cause of this difference is not fully explained by endosperm composition or husk phenotype.

The extent and location of freezing damage in the seed is also important. Freezing within plant tissues can occur in both the intracellular (Kiesselbach and Ratcliff, 1920; Burke et al., 1976) and extracellular (Mazur, 1969) regions. Ice nucleation begins in the extracellular region and, as the ice crystals expand, additional water from within the cells migrates into the extracellular region and freezes (Mazur, 2004; Steponkus, 1984), essentially causing dehydration stress. Extracellular ice formation and cell water migration impose a severe dehydrative force and, consequently, freezing and desiccation stresses share some mechanisms of adaptation. In contrast to slow propagation of ice, when temperature is very low or drops quickly, ice crystals form in the intracellular region, which causes direct damage to the cells, reducing seed viability and vigor (Rossman, 1949; Burke et al., 1976).

To determine the extent of freezing damage, several seed quality tests are used to determine viability and vigor in seed corn. The cold test is the most commonly used vigor test for seed corn in the Midwest. Cold tests are used to predict early field emergence; however, the different protocols available and degree of correlation with field emergence vary greatly (Burris and Navratil, 1979). The saturated cold test (SC) is considered the most stressful and it has been widely adopted by seed companies (Gutormson, 1996). In the SC, the embryo is placed into a soil that is saturated with water, leading to respiratory stress due to limited gas exchange (Hoegemeyer and Gutormsen, 2000). Soaking seed corn in water before germination, the soak test, is another stress test that inhibits seed germination by the accumulation of CO2 (Martin et al., 1991). The SC test has higher correlation to field emergence, however, than the soak test (Martin et al., 1988). The accelerated aging (AA) test, originally developed to predict seed storability (Delouche and Baskin, 1973), is not widely used in the seed industry. The results from this test, nevertheless, correlate well with field emergence (Egli and TeKrony, 1996).

Freeze-damaged symptoms in seed are difficult to evaluate immediately after the frost event. Early detection of frost-damaged seed lots would allow seed companies to make timely marketing decisions. There is little information in the literature, however, about the relationship between seed quality tests and early evaluation of frost damage in seed corn during the first 6 mo of storage, and their relationship to field emergence.

The objectives of this study were to determine which seed quality tests will identify frost-damaged seed lots during the first 6 mo of storage, and which tests can best predict field emergence of frost-damaged seed.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Seed Production and Frost Treatment
The field experiments were planted in 2003 and 2004 near Ames, IA. In 2003 an isolation plot was planted at the agronomy research farm west of Ames using a randomized complete block design (RCBD) with five blocks. In 2004 two locations were planted at the agronomy research farm. The design at each location was a RCBD, one with three blocks and the other with four blocks. The planting dates, growing conditions, and harvest periods for each seed production environment are shown in Table 1 . Each experimental unit was a seed lot harvested from a field plot that had been assigned a block, female parent, and moisture content at harvest.


View this table:
[in this window]
[in a new window]

  		Table 1. Agroclimatic conditions for the seed production environments near Ames, IA, in 2003 and 2004.

 
Field plots were planted using a three-point mounted Almaco twin-row planter (Almaco, Nevada, IA), and seeded at 61,750 seeds ha–1, with a final stand between 49,000 and 54,000 plants ha–1. Hybrid seed was produced by open pollination in isolation specified for certified seed production. Plots were planted with six female rows to four male rows. Males were planted at two different dates to ensure adequate pollen supply and "nick". The first two rows of the male were planted on the same date as the female; the second two rows were planted 10 to 15 d after the female rows. The female inbreds used in the cross were B73 and Mo17. A common male parent, IRF311, was selected because of its synchrony of flowering with both female parents. Inbred B73 is considered relatively cold tolerant at the seedling stage (Mock and McNeill, 1979). Inbred Mo17 is considered relatively cold sensitive at the seedling stage. The pollinator IRF311 was obtained from the Iowa Research Foundation.

Field plots were harvested at 500 to 550, 400 to 450, or 300 to 350 g H2O kg–1 fresh weight (Table 2 ). The final harvest was done after the seed reached physiological maturity, as determined by black layer formation according to Hunter et al. (1991). Seventy-five to 100 ears were harvested from the four center rows of the six-row plots in each field plot with the husk and shank intact. Once harvested, the ears from each plot were divided in two groups and randomly either assigned to the frost treatment or designated a control sample.


View this table:
[in this window]
[in a new window]

  		Table 2. Mean moisture content (MC) and standard deviation of the mean for samples representing each harvest MC category in 2003 and 2004.

 
Moisture content of the seed was determined using the oven-drying method. Twenty-five seeds were shelled from the center rows of the ears. Three ears were used per plot. Seeds were weighed and dried at 100°C for 48 h in an Isotemp oven (Fisher Scientific, Pittsburgh, PA). Seeds were weighed and moisture content was calculated in grams H2O per kilogram on a fresh-weight basis. The three replications were averaged to estimate seed moisture content at harvest.

Frost Treatments
All ears were placed in 10°C for 2 to 4 h to prechill the ears and ensure that the freezing rates for all harvests were similar. In 2003 control ears remained in the 10°C chamber while the frost-treated ears were placed in a Conviron growth chamber (Controlled Environment Ltd., Winnipeg, MB, Canada) with a 24-h programmed freeze cycle. In 2004 control ears were husked and dried following the 2- to 4-h prechill to avoid the low level of damage observed in all control samples in 2003 (DeVries and Goggi, 2006). An artificial frost cycle (Fig. 1 ) was developed to mimic typical early fall frost temperature patterns in Iowa (Dr. Elwynn Taylor, personal communication, 2003). The entire cycle lasted 24 h, beginning and ending at 10°C. It included a total of 9 h below 0°C, with 2 h of this time at –5°C. This temperature cycle was severe enough to be considered a hard killing frost. The programmed cooling rates were –1.4°C h–1 from 9 to 0°C, and –0.83°C h–1 from 0 to –5°C (Fig. 1), which is considered slow, causing extracellular freezing of water and desiccation damage (Mazur, 1969; Steponkus, 1984). Thermocouples were placed inside seeds in the middle of the ear to record ice formation events in the seed tissue.


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		Figure 1. Artificial frost cycle developed to mimic an autumn severe killing frost in the Midwest. Segment 1 cooling rate is –1°C h–1 from 10 to 9°C, Segment 2 cooling rate is –1.4°C h–1 from 9 to 0°C, Segment 3 cooling rate is –0.83°C h–1 from 0 to –5°C, Segment 4 is a constant –5°C for 2 h, Segment 5 thawing rate is 4.7°C h–1 from –5 to 9°C, and Segment 6 thawing rate is 1°C h–1 from 9 to 10°C.

 
In 2003 all ears were placed in the 10°C chamber for an additional 24 h to thaw before being husked. In 2004 frost-treated ears were husked immediately following the frost cycle to avoid the low levels of seed damage observed in the control samples in 2003 (DeVries and Goggi, 2006). After ears were husked, they were placed in small-scale laboratory dryers, as described by Navratil and Burris (1982), and dried with forced air at temperatures below 35°C until the seed was at 120 g H2O kg–1 fresh weight. Drying temperature was monitored on a daily basis. Dried ears were shelled using a Custom Seed sheller (Custom Seed Engineering & Equipment, Altoona, IA), and seed was stored in 10°C and 50% relative humidity to simulate normal storage conditions used by the seed industry until the seed quality tests were conducted. Seed testing was conducted at approximately 0, 1.5, 3, 4.5, and 6 mo after the beginning of storage.

The 12 field blocks were used as replications (or blocks) in the lab. Each field block had 12 treatment combinations of female x moisture x frost treatment for a total of 144 seed lots. One hundred seeds from each seed lot were used for each seed quality test.

Seed Quality Tests
Standard germination tests were done according to the Association of Official Seed Analysts (AOSA) rules for testing seeds (AOSA, 2004). Each replication consisted of 100 seeds per seed lot and treatment. In 2003 rolled towel media (Anchor, Hudson, WI) were used for the standard germination test. In 2004 crepe cellulose paper media (Kimberly Clark, Neenah, WI) were used for the standard germination test instead of rolled towels. Both methods are consistent with the AOSA (2004) approved methodology. The final evaluation was made at 7 d according to the AOSA (2004). Standard germination tests were conducted at constant temperature (25°C) and alternating light (4 h on, 4 h off) for a total of 12 h of light per day. Light photon flux density (400–700 nm) inside the chamber was 25 µmol s–1 m2. Light measurements were made with a LI-1400 data logger with a LI-190SA spectral response point quantum sensor (LI-COR, Lincoln, NE).

Saturated cold tests (SC) were conducted as described by Martin et al. (1988) with the following changes. After 7 d at 10°C, the trays were moved to 25°C and alternating light for 4 d before being evaluated. Only 100 seeds were used for each seed lot. Seedlings were evaluated according to the AOSA rules for testing seeds (AOSA, 2004).

Accelerated aging tests were done according to the seed vigor testing handbook of the AOSA (2002) with the following exceptions. Samples of 100 seeds were placed on a mesh screen above 40 mL of water inside a plastic box (approximately 100% relative humidity) and aged at 43°C for 72 h. After the aging process, seeds were placed in a crepe cellulose paper (Kimberly Clark, Neenah, WI) germination test for 7 d. Seedlings were evaluated according to the AOSA (2004).

The soak tests were conducted as cited in Martin et al. (1991) with the following changes. Two replications of 50 seeds were placed in 0.27-L (9-oz.) plastic cups with 150 mL of deionized, distilled water for 48 h at room temperature (20–25°C). Water was drained and the two replications were combined, for a total of 100 seeds per seed lot, and germinated on crepe cellulose paper (Kimberly Clark, Neenah, WI) for 7 d. Seedlings were evaluated according to the AOSA rules for testing seeds (2004).

Field Emergence
One hundred seeds from each seed lot were planted in a completely random design in the spring. There were two planting dates for each year. The earlier planting was under poor field conditions and the later planting was under average to good field conditions. A summary of the field emergence conditions is shown in Table 3 . The number of seedlings was counted at approximately the V2 stage.


View this table:
[in this window]
[in a new window]

  		Table 3. Agroclimatic conditions for the field emergence experiments in 2004 and 2005.

 
Statistical Analysis
Plots for seed production were planted in a randomized complete block design with three environments, one in 2003 and two in 2004. The combination of years and locations was treated as a single factor (environment), and was considered a random effect. The ANOVA and mean comparisons were done using PROC MIXED (SAS Institute, Cary, NC). The ANOVA was performed by time and the model used block nested within environment as the random effect. Fixed effects were female parent, moisture, and frost treatment. Testing time did not have a significant interaction with female parent or moisture, thus data were pooled across the five testing periods. The effect of frost treatment for each combination of location, female parent, and moisture content was calculated as the average difference between the frost treatment and control from the same field plot. Comparisons of frost effects between female lines or between moisture levels were made using ANOVA and pairwise comparisons on the differences (control minus frost treatment) from each field plot as the response. Because the control treatments were not the same in 2003 and 2004, the analysis of frost effects was done separately for each environment.

To determine their prediction capabilities, the results of the seed quality tests were compared with field emergence data. Regressions were done using PROC REG in SAS. Slopes of the regression lines were compared using t-tests (Zar, 1996).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Table 1 presents a summary of the agroclimatic conditions in the seed production environments. In 2003 weather conditions were very dry and warm during the period of grain fill, which negatively impacted seed quality across the Midwest (Goggi et al., 2006). In 2004 however, there was adequate precipitation and very few stress days. Although there was a 2-wk difference in planting dates between the two fields in 2004, the growing conditions were very similar (Table 1). The total precipitation and number of stress degree days between planting and harvesting were similar.

The seed quality of the seed lots did not change significantly during the 6 mo of storage of these experiments. All seed samples for these experiments were stored in a cold room (10°C and 50% relative humidity). The use of cold storage is considered normal practice in the seed corn industry. Under this cold storage environment, however, the decline in seed quality was too slow to measure by the tests used in these experiments. Because there was no significant decline in seed quality with time, the seed quality test data were pooled across the five testing periods for the analysis.

The percentage of normal seedlings, abnormal seedlings, and dead seeds were recorded for all tests; however, only the percentage of normal seedlings and dead seeds changed significantly (data not shown). Thus, data analyses for this study were conducted on the percentage of normal seedlings. Similar results were obtained in a previous study (DeVries and Goggi, 2006), where the total number of viable seeds in the tetrazolium test decreased in frost-damaged seed samples, but not the number of low-vigor seeds (or possible abnormal seedlings in the germination test).

The main factors in this study were two hybrids with different female parents (B73 and Mo17) and moisture content at harvest. Because the female parent x moisture, female x frost treatment, and moisture x frost treatment interactions were significant for several of the tests and testing times, data were analyzed by female parent and moisture content. All data analyses are presented by female parent and moisture content for all tests, even though the female parent x moisture interaction was not significant for the saturated cold test at any of the testing times (Table 4 ).


View this table:
[in this window]
[in a new window]

  		Table 4. Analysis of variance (P values) for standard germination and saturated cold test after approximately 0, 1.5, 3, 4.5, and 6 mo in storage.

 
Seed Quality Test Results
Table 5 presents the mean percentage of normal seedlings for control and frost-treated seed lots analyzed by female parent and moisture content. The standard germination test (WG) evaluates seed germination under ideal growing conditions, while vigor tests (SC, AA, and the soak test) measure germination under stressful growing environments. The WG percentage of most commercial maize seed lots is >95% and their range of seed vigor is between 75 and 85% (Ferguson Spears, 1995). The germination percentage of B73 x IRF311 control samples was >95% for most samples tested. A few control seed samples were <95% germination, but these samples were harvested before physiological maturity (Table 5). The WG percentage of control samples of Mo17 x IRF311 harvested at all moisture contents, however, was <95% The vigor tests results for most control samples of both hybrids were above the acceptable range (75–85%) (Table 5 and 6). Only a few control samples harvested at 500 to 550 g H2O kg–1 fresh weight were below this seed vigor range. This lower seed vigor of control samples harvested at 500 to 550 g H2O kg–1 fresh weight could be associated with premature harvest and drying. Cordova-Tellez and Burris (2002) found that some genotypes harvested at high moisture content (500–550 g H2O kg–1 fresh weight) were sensitive to premature drying, even when proper seed drying techniques were used. Results from the present study support this conclusion, and suggest that both B73 x IRF311 and Mo17 x IRF311 hybrids may be sensitive to drying when harvested at moisture contents >500 g H2O kg–1 fresh weight. Therefore, it is likely that sensitivity to early harvest and drying are reasons for the lower seed quality observed in some control samples. These observations were true for all environments (data not shown).


View this table:
[in this window]
[in a new window]

  		Table 5. Percentage of normal seedlings in the standard germination and saturated cold tests for two hybrids harvested at three moisture contents. Female parents of the hybrids are B73 and Mo17, with IRF311 as the common male parent.

 
The artificial frost treatment significantly decreased WG percentages of both hybrids harvested at all moisture contents except for the hybrid seed lot obtained with Mo17 as a female parent at physiological maturity (300–350 g H2O kg–1 fresh weight; Table 5). Saturated cold test results of Mo17 x IRF311 frost-treated seed samples were not significantly lower in seeds harvested at 300 to 350 g H2O kg–1 fresh weight, but were significantly lower in seeds harvested at 400 to 450 and 500 to 550 g H2O kg–1 fresh weight. The SC test results for frost-treated seed samples of B73 x IRF311 were significantly lower than the results for the control samples for all moisture contents.

Results from the AA and soak tests are presented in Table 6 . The artificial frost treatment significantly decreased the AA germination percentage of Mo17 x IRF311 harvested at 400 to 450 and 500 to 550 g H2O kg–1 fresh weight; seed samples harvested at 300 to 350 g H2O kg–1 fresh weight were not significantly affected. The AA test percentage of B73 x IRF311 seed samples harvested at 300 to 350, 400 to 450, and 500 to 550 g H2O kg–1 fresh weight were significantly lower when exposed to the frost treatment. The soak test percentages of all B73 x IRF311 frost-treated seed samples and Mo17 x IRF311 frost-treated seed samples harvested at 400 to 450 and 500 to 550 g H2O kg–1 fresh weight decreased significantly at all harvest moisture contents. The soak test results of Mo17xIRF311 seed samples harvested at 300 to 350 g H2O kg–1 fresh weight, however, were not significantly impacted by the frost treatment.


View this table:
[in this window]
[in a new window]

  		Table 6. Percentage of normal seedlings in the accelerated aging and soak tests for two hybrids harvested at three moisture contents. Female parents of the hybrids are B73 and Mo17, with IRF311 as the common male parent.

 
In general, B73 x IRF311 seed quality test results of frost-treated seed samples were lower for all moisture contents. The seed quality of Mo17 x IRF311 of frost-treated seed lots was significantly lower in samples harvested before physiological maturity (400–450 and 500–550 g H2O kg–1 fresh weight). Frost-treated samples of Mo17 x IRF311 harvested at 300 to 350 g H2O kg–1 fresh weight (on or about physiological maturity), however, were not affected by the frost treatment. This tolerance to the frost treatment was observed in all environments (data not shown).

Female Parent and Moisture Content
Table 7 presents the moisture content and genotype interaction mean difference between the control and frost treatment. When seeds were at physiological maturity, the hybrid with Mo17 as the female parent showed almost complete tolerance to frost regardless of the germination or vigor assay that was used. The mean difference between control and frost treatments was lowest for the hybrid with Mo17 as a female parent at physiological maturity. The frost-induced decrease in the SC and soak test percentages at physiological maturity was in the range of 10 to 11% for the B73 hybrid and 1 to 5% for the Mo17 hybrid. These results indicated that the Mo17 hybrid is more tolerant to frost at physiological maturity than the B73 hybrid. Previous studies found that, at the vegetative stage, B73 is cold tolerant (Stewart et al., 1990). Recent studies however, showed that these two inbreds behave differently at the reproductive and vegetative stages. Jorgensen et al. (1992) found that Mo17 was more heat tolerant than B73 at the vegetative stage. At the reproductive stage, however, high temperature significantly affected seed set, development, and the number of endosperm cells in Mo17, while B73 was unaffected (Commuri and Jones, 2001). These researchers concluded that B73 was more tolerant to high temperatures than Mo17 during the reproductive and seed-development stages. The results from the present study showed that the frost treatment negatively affected the seed quality of the two hybrids before physiological maturity. Other researchers have shown a strong genetic component associated with seed quality as estimated by the AA (Woltz and TeKrony, 2001; Munamava et al., 2004), SC (Munamava et al., 2004), and soak tests (Martin et al., 1991; Cerwick et al., 1995; Munamava et al., 2004).


View this table:
[in this window]
[in a new window]

  		Table 7. Moisture content and genotype interaction means of the difference between control and frost treatment for standard germination, saturated cold, accelerated aging, and soak tests for two hybrids harvested at three moisture contents. Female parents of the hybrids are B73 and Mo17, with IRF311 as the common male parent.

 
Seed moisture content at harvest was important for determining seed quality and the effect of frost. Both hybrids showed the greatest damage when seeds were at 500 to 550 g H2O kg–1 fresh weight for all tests. As seeds matured, the detrimental effect of frost decreased in both hybrids (Table 7). The test results from control samples for the hybrid with B73 as the female parent harvested at 400 to 450 g H2O kg–1 fresh weight in Environment 3 were lower than in all other environments (data not shown). This resulted in smaller percentage mean differences between the control and frost treatments (Table 7). These lower test results for control samples of the B73 hybrid in Environment 3 did not change the effect of frost, they simply made the difference between control and frost-treated samples smaller.

Regressions between Seed Quality Tests and Field Emergence
To determine which seed quality test best predicts field emergence of frost-damaged seed, the lab test results were independently regressed to both early and late field emergence (Table 8 ). Only data from frost-damaged seed lots were used to calculate the regressions; however, the non-frost-treated seed lots followed the trend line of the linear regression (data not shown). The simple regressions between laboratory tests and early field emergence were weak, and the adjusted r2 values ranged from 0.08 to 0.32, indicating that the regression equations accounted for very little of the variation observed in the data (Table 8). The ideal slope for a regression between a lab test and field emergence would be 1.0, implying that the germination percentage in the laboratory test was equal to field emergence. The slopes of these regressions (0.38–0.62), although significantly different from zero (P < 0.05), indicate that the lab tests were overestimating early field emergence, or that field conditions early in the spring were too variable for a strong relationship between field emergence and laboratory test results to be obtained. TeKrony and Egli (1977) also found that regressions between laboratory tests and early field emergence were very poor. In their study, the cold test and accelerated aging did not predict emergence under poor field conditions in soybean [Glycine max (L.) Merr.]. In a later study, Woltz and TeKrony (2001) found similar results in corn. Because of variation in field conditions, the lab test results often overestimate field emergence.


View this table:
[in this window]
[in a new window]

  		Table 8. Intercept, slope, adjusted r2, and error of the estimate for simple linear regressions between the seed quality tests{dagger} of frost-damaged seed samples at 0 and 1.5 mo of storage and early field emergence (FE1), and at approximately 0, 1.5, and 6 mo of storage and late field emergence (FE2).

 
The regressions between the lab tests and the field emergence of the second planting date were much stronger. All regressions and their slopes were significant (P < 0.05). Standard germination results after approximately 0, 1.5, and 3 mo of storage were regressed to late field emergence (Table 8). The adjusted r2 values were between 0.56 and 0.75. The slopes of the regression were 0.64, 0.75, and 0.72 for approximately 0, 1.5, and 3 mo of storage, respectively. The intercepts of these regressions were significant (P < 0.05). The slopes and significant y intercepts indicated that standard germination overestimates the field emergence of frost-damaged seed. The standard germination test is designed to give maximum germination under ideal conditions (AOSA, 2002). Field conditions are rarely ideal; therefore, we would expect the standard germination test to overestimate field emergence.

The AA test at 0 mo was better than the standard germination test for predicting field emergence of frost-damaged seed. The adjusted r2 value was 0.81 and the slope of the regression was 0.90 (Table 8). The y intercept was not significant. After storing frost-damaged seed for approximately 1.5 mo, however, the adjusted r2 decreased to 0.62 and the y intercept was significantly different from zero. After approximately 3 mo of storage, the adjusted r2 was 0.78 and the y intercept was significant (Table 8). Therefore, the AA test was more sensitive than the standard germination test. It is possible that during the early months of storage, some seed recovery and physiological and metabolic repair occurs. This repair could lead to higher seed quality at approximately 1.5 and 3 mo and, thus, seed quality test results that overestimate field emergence. Our results, however, indicate that the AA test is a good indicator of seed vigor and a good predictor of field emergence immediately after harvest and drying.

The best seed quality tests for predicting field emergence under average to good field conditions were the SC and soak tests (Table 8). The adjusted r2 values of the regressions between late field emergence and the SC and soak tests ranged from 0.77 to 0.86. There were no significant y intercepts for either of the tests during approximately the first 3 mo of storage. The slopes for both tests indicated a very strong relationship between the SC and soak tests and field emergence. The slopes of the regressions for SC were 0.89, 0.90, and 0.96 at approximately 0, 1.5, and 3 mo of storage, respectively. The slopes for the regressions for the soak test were 0.95, 0.90, and 0.93 at approximately 0, 1.5, and 3 mo of storage, respectively. By performing t-tests on the slopes of these regression lines (Zar, 1996), it was determined that the slope of the regressions for SC at approximately 0, 1.5, and 3 mo were not significantly different from each other. The slopes of the soak test were also not different, indicating that these two tests were not changing during approximately the first 3 mo of storage. When comparing the slopes of the regressions between these two tests and late field emergence, there were no significant differences between tests. These results confirmed that the two tests are equal in their ability to predict field emergence of frost-damaged seed within approximately the first 3 mo of storage.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The extent of frost damage is strongly influenced by both maternal parent and moisture content at harvest. The hybrids produced from B73 and Mo17 as maternal parents responded differently to an artificial frost treatment during seed maturation. The ability of Mo17 to tolerate a frost event at physiological maturity with no significant seed quality reduction has very important seed production implications. These novel results suggest that when a natural frost event occurs, some female parents could remain in the field and possibly have no significant damage to seed quality. Other female parents at the same moisture content, however, may suffer damage that can lower seed quality. Therefore, hybrids with a sensitive female parent should be harvested first.

Frost damage is less severe as seeds mature and dry down. These results highlight the importance of harvesting the immature seed lots before a frost event. Mature seed lots (300–350 g H2O kg–1 fresh weight) would be less impacted by the frost and thus could remain in the field. This strategy, however, is only successful to a point. At high moisture content (≥500 g H2O kg–1 fresh weight), premature harvest can lower the seed vigor of some genotypes even if proper drying, conditioning, and storage conditions are maintained. The seed viability and vigor of control seed samples for the Mo17 hybrid decreased as harvested moisture content increased. The control samples for the B73 hybrid harvested at high moisture content (500–550 g H2O kg–1 fresh weight) were viable in the WG test, but had low vigor as determined by the SC and AA tests. The lower vigor of these seed samples indicates that, at high moisture content, there is a trade-off between damage due to premature harvest and damage due to frost. A mild frost may not cause enough damage to warrant early harvest. A severe killing frost similar to our artificial frost event, however, can cause enough damage to justify harvesting early.

Standard germination and AA tests have strong relationships to field emergence under average to good field conditions; however, both tests overestimate field emergence. The SC and soak tests are the best predictors of field emergence of frost-damaged seed lots. Both tests are consistent, and can accurately predict field emergence of frost-damaged seed lots. Seed samples can be tested immediately after the seed lots are dried and shelled, or within approximately the first 3 mo of storage. Management decisions can be based on the strong relationship between the SC and soak tests and field emergence under average to good field conditions.

Results from these experiments advance our understanding of the relationship between seed moisture content and maternal parent, and their importance in determining the extent of frost damage in immature maize seeds. Seed companies can make harvesting decisions based on the probability and severity of a fall frost forecast. If the seed lots are frost damaged, common seed quality tests can be used to predict field emergence. These seed quality tests are important tools for making management decisions about hybrid seed production in the event of an early fall frost.


   ACKNOWLEDGMENTS
 
We express our gratitude to the Seed Science Center and Iowa State Seed Laboratory for providing technical assistance and supplies for seed testing. We also thank Nate LeVan, Heather Hall, and Jay DeVries, who provided help with the field and laboratory work.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication January 3, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
A. S. Goggi, P. Caragea, L. Pollak, G. McAndrews, M. DeVries, and K. Montgomery
Seed Quality Assurance in Maize Breeding Programs: Tests to Explain Variations in Maize Inbreds and Populations
Agron. J., February 26, 2008; 100(2): 337 - 343.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by DeVries, M.
Right arrow Articles by Moore, K. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by DeVries, M.
Right arrow Articles by Moore, K. J.
Agricola
Right arrow Articles by DeVries, M.
Right arrow Articles by Moore, K. J.
Related Collections
Right arrow Seed Quality
Right arrow Plant and Environment Interactions
Right arrow Seed Physiology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome