Contribution of Alfalfa Hard Seed to Stand and Yield in the Field

Ken Albrecht and Dan Undersander, Department of Agronomy, University of Wisconsin, Madison, WI 53706; Nick Dagenhart, formerly Cal/West Seeds, West Salem, WI 54669; Jim Moutray, formerly ABI Alfalfa, Napier, IA 50014; Mark McCaslin, Forage Genetics International, Savage, MN 55378

Abstract

The contribution of hard seed to alfalfa (Medicago sativa L.) stand establishment has not been clearly defined. A study was established at three locations (Arlington and West Salem, WI, and Napier, IA) to examine the effect of the proportion of hard seed in alfalfa seed lots at time of planting on time of seed emergence and yield. In each of two years scarified and un-scarified seed lots with either low, medium, or high proportions of hard seed from four commercially available alfalfa cultivars were planted under field conditions. Alfalfa emergence was monitored during the following months for a year. Seventy percent of the unscarified seed emerged within 2 weeks, 6% in 2 to 4 weeks, 3% at 8 weeks, and none after that. No seedlings emerged the spring after sowing. Scarification increased (P ≤ 0.05) slightly the number of seedlings that emerged within 2 weeks but did not affect overall establishment. In a second study, treatments were sown at 12 lb pure live seed/acre in small plots and yield measured at each of the three locations. Level of hard seed and seed scarification had no effect on yield in the seeding year or the year after. Hard seed contributes to alfalfa stand establishment similarly to other alfalfa seed.

Introduction

Hard seed is that seed which does not absorb water (swell) or germinate under standard seed testing conditions. However, hard seed has long been known to be live seed (6,9,11) and, in fact, is included in the final germination value reported from standard seed tests (2). The importance of the hard seed when establishing an alfalfa (Medicago sativa L.) stand has long been debated. Bass et al. (3) argued that hard seed does not contribute to a stand. Others (8,10,12) concluded that hard seed was nearly as important as permeable seed in contribution to alfalfa stands. Questions about the value of hard seed in alfalfa are becoming more important with increased seed production in the northwestern United States and Canada where seed produced tends to have higher levels of hard seed than alfalfa seed produced farther south. (4).

The objectives of this research were to: (i) determine when alfalfa hard seed emerges from soil; (ii) determine if hard seed emerges differently from genetic backgrounds; and (iii) determine effect of hard seed on forage yield.

Examining Effects of Hard Seed in Alfalfa Stand Establishment

Two studies were conducted to determine the effect of hard seed on alfalfa: (i) A field study in which germination and emergence were monitored with time; and (ii) a field experiment in which we measured the forage yield of cultivar/hard seed combinations in small plots. For both studies seed of four varieties with three levels (low, medium, high) of hard seed from different breeding programs were used (see Table 1). Varieties were selected by the participating companies for seed lots with a range in hard seed. The selected cultivars represented three major breeding programs and thus had a range of genetic variation that would be expected among the commercially available cultivars of the semi-dormant types of alfalfa marketed. All varieties had a 3 or 4 fall dormancy rating and were adapted to region of study.

Table 1. Hard seed lots used for studies in 1992 and 1993.

Year	Cultivar	Treatment	% hard seed
Year	Cultivar	Treatment	Fall	Spring
1992	Starmaster Starmaster Starmaster Venture Venture Venture DK-122 DK-122 DK-122 G-2833 G-2833 G-2833	Low Medium High Low Medium High Low Medium High Low Medium High	11 26 35 16 30 45 15 33 55 12 22 38	12 20 38 12 14 20 8 27 50 6 12 23
1993	Alfagraze Alfagraze Alfagraze Multiplier Multiplier Multiplier DK-122 DK-122 DK-122 Verta+ Verta+ Verta+	Low Medium High Low Medium High Low Medium High Low Medium High	9 30 50 6 22 60 10 23 36 5 17 32	12 28 45 5 10 40 9 9 16 3 6 13
Mean		Low	10.5a^*	8.4a
		Medium	25.4a	15.8b
		High	43.9a	30.6b

* Means within a row followed by the same letter do not differ
(P ≤ 0.05).

Germination tests were run on all seed lots by Cal/West Seeds using standard AOSA (2) procedures. Lots were tested in the fall to allow selection of three levels of hard seed and a germination test was run again prior to planting to determine percentage hard seed at planting. An additional treatment was scarification vs. not for each level of hard seed within each variety. Both studies were established at three locations in 1992 and 1993: (i) the University of Wisconsin Arlington Agricultural Research Station (43.30°N, 89.35°W) on a Plano silt loam (fine-silty, mixed, superactive, mesic Typic Argiudoll); (ii) the ABI Research Station, Napier, IA (41.98°N, 93.70°W) on Webster silty clay loam (fine-loamy, mixed, superactive, mesic Typic Endoaquoll); and (iii) at Cal/West Seeds, West Salem, WI (43.89°N, 91.10°W) on a Toddville silt loam (fine-silty, mixed, superactive, mesic Typic Argiudoll).

In the first study, three replicates of 100 seeds of each lot were hand seeded in the field and covered with ½ inch of soil during late April at each site each year. Each replicate was surrounded by a plastic ring placed about 1 inch into the soil for identification of treatment. Counts of emerged seedlings were made approximately 2, 4, 6, and 8 weeks after seeding and the following spring. Emerged plants were removed at each counting period to reduce competition to later emerging plants.

In the second study, 3-ft by 16 to 20-ft plots were seeded with the same seedlots at 12 lb pure live seed/acre in late April and managed for maximum yield at three locations. Pure live seed according to the standard AOSA (2) definition is purity times germination (which includes hard seed). Yields were measured in the year of establishment and the following year. At each site 2 to 3 cuttings were taken in the seeding year and 3 cuttings the following year. Studies were conducted under high soil fertility and plots sprayed for insects, primarily potato leaf hopper (Empoasca fabae ), as needed.

Each study was analyzed as a split plot in time using PROC GLM (SAS Institute Inc., Cary, NC) with block, year, and scarification assumed to be random effects and location, varieties and hard seed level assumed to be fixed effects.

The Contribution of Hard Seed

The range of hard seed in seed lots at the time of selection for the low, medium, and high levels varied somewhat but averaged 10.5, 25.4, and 43.9%, respectively (Table 1). Higher levels of hard seed are attributed by the seed industry to more northern environments. The seed industry recognizes that alfalfa seed produced in California is generally lower in hard seed than that produced in northern states or Canada. Within a location the industry recognizes that hard seed level varies from year to year due to environmental conditions.

All seed lots declined in hard seed from after cleaning in the fall to before spring seeding, averaging 8.4, 15.8, and 30.6% hard seed for the low, medium, and high levels, respectively, in the spring. The decline in hard seed with time has previously been reported (1). The amount of hard seed decline from fall to spring increased as the level of hard seed in the fall increased. There was no decline at the low level of hard seed (2.1%), whereas at the high level of hard seed, the average decline was 13.3% (Table 1). All cultivars showed the same trends in decline of hard seed levels however, the actual amount of decline was variable as shown in Table 1 and previously reported (5). It appears that hard seed varies appreciably among seed lots in terms of its persistence in storage.

Seedling emergence data in the first study were averaged across the three sites because fews differences occurred among the sites (Table 2). Most seedlings emerged within the first 2 weeks after planting, including a portion of the hard seed because the total emergence was greater than the germination percentage minus the hard seed. Another 3 to 7% of the seedlings emerged between 2 and 4 weeks after planting with minimal seed emergence after that. At 4 weeks after planting, the difference in total seedling emergence was little affected by level or hard seed or whether or not the seed was scarified.

Table 2. Effect of hard seed on alfalfa emergence at 2, 4, and 8 weeks after planting. Data are means of three sites (Arlington, WI, West Salem, WI, and Napier, IA) and over two years.

Hard seed	Scarified	2 weeks	4 weeks	8 weeks	total
Hard seed	Scarified	Percent emergence
Low	No	66c	7a	3a	77b
Medium	No	69c	6a	2a	77b
High	No	74a	6a	0a	81a
Mean	No	70	6	2	78
Low	Yes	73b	4b	1a	79a
Medium	Yes	72b	4b	1a	77b
High	Yes	75a	3b	1a	80a
Mean	Yes	73	4	1	78

* Means within a column followed by the same letter do not differ (P ≤ 0.05).

Scarification increased the emergence of alfalfa seedlings during the first 2 weeks after planting. Scarified seed, however, had lower emergence during the 2 to 4 week period so that no effect of scarification was noted at the end of 4 weeks. No other effect of scarification was observed.

Emergence of both scarified and unscarified seed at 4 weeks after planting varied from 0 to 3% and was not related to the natural level of hard seed. Further, only about 1% of scarified seed emerged after 4 weeks. It should be noted that this late emergence occurred in the absence of any other growth from earlier seed emergence. In the field, where growing seedlings would be competing with the late emerging seed, it is unlikely that any of the late emerging plants would survive to contribute to the stand. Thus, to answer farmer’s questions about the usefulness of hard seed in filling a thin stand, our data indicate that hard seed is of limited value, because very few seedlings emerge after the first month and those that do would suffer significant competition from early emerging seedlings.

There were no differences in emergence among cultivars with varying levels of hard seed (Table 3). The cultivars represented a range genetic variation that would be expected among the commercially available cultivars of the semi-dormant types of alfalfa marketed; however, the cultivars do not represent the total genetic variation that might exist within the Medicago species and may not represent what occurs in, for example, M. falcata types.

Table 3. Effect of alfalfa cultivars on total seedling emergence
with varying levels of hard seed. Data are means across three
levels of hard seed and three locations.

Year	Cultivar	Total percent emergence
1992	DK-122	76
	G-2833	73
	Starmaster	73
	Venture	75
1993	DK-122	84
	Alfagraze	84
	Multiplier	82
	Verta+	81

Means within a year are not significantly different (P ≤ 0.05).

The most important question is: did hard seed levels affect alfalfa yield? In this study all plots were seeded at 12 lbs pure live seed/acre, which included varying levels of hard seed. We purposely chose a rate that most universities recommend and lower than what many farmers seed to increase likelihood of detecting differences if hard seed had an impact on yield. As shown in Table 4, neither level of natural hard seed nor scarification affected yield in either the seeding year or the year after (P ≤ 0.1). These data suggest that growth and forage production are similar regardless of hard seed level. If hard seed has an effect on yield, the greatest yield differences are likely to occur as the stand is establishing during the seeding year. We did not observe yield differences in the seeding year or in the year after seeding due to level of hard seed or whether or not the seed was scarified.

Table 4. Effect of hard seed and seed scarification on alfalfa yields.

Sowing		1992				1993
Harvest		1992		1993*		1993		1994
Scarified?		no	yes	no	yes	no	yes	no	yes
Hard seed level (tons/acre)	Low	3.24	3.30	4.81	4.77	2.57	2.62	6.52	6.57
	Medium	3.18	3.21	4.69	4.79	2.62	2.57	6.56	6.66
	High	3.17	3.24	4.68	4.75	2.62	2.62	6.59	6.51

* 1993 harvest means of 1992 sowing are for two sites only.

Means within a column do not differ (P ≤ 0.1).

In summary, while hard seed is considered a dichotomous category for purposes of seed testing, it is really a continuum. Hard seed in some species (e.g., some Trifolium spp.) is very resistant to germination (7). The authors have personally observed red clover (Trifolium pratense L.) seed remain dormant in soil for more than 7 years and then emerge from the soil. Further, there is likely some variation of hard seed in alfalfa as evidenced by the differential rate of hard seed decline over winter (some of the variation may be environmental and some genetic). Greater seed hardness may occur in alfalfa seed produced in northern regions; however, our studies indicate that it has little impact on seedling establishment in commercial semi-dormant cultivars. Our research also demonstrates that seed scarification is not necessary to improve seedling emergence for alfalfa. This is an important finding because viability of scarified seed declines faster in storage than unscarified seed.

Literature Cited

1. Acharya, S. N., Stout, D. G., Brooke, B., and Thompson, D. 1999. Cultivar and storage effects on germination and hard seed content of alfalfa. Can J. Plant. Sci. 79:201-208.

2. AOSA. 2006. Rules for testing seeds. Sec 4.2d and 4.9k(6). Association of Official Seed Analysts (AOSA), Ithaca, NY.

3. Bass, L. N., Gunn, C. R., Hesterman, O. B., and Roos, E. E. 1988. Seed physiology, seedling performance and seed sprouting. Pages 961-983 in: Alfalfa and alfalfa improvement. A. A. Hanson, K. D. Barnes, and R. R. Hill, Jr., eds. ASA, CSSA, and SSSA, Madison, WI.

4. Fairey, D. T., and Lefkovitch, L. P. 1991. Hardseed content of alfalfa grown in Canada. Can J. Plant Sci. 71:437-444.

5. Hall, J. W., Stout, D. G., and Brook, B. M. 1998. Alfalfa seed germination tests and stand establishment: The role of hard (water impermeable) seed. Can J. Plant Sci. 78:295-300.

6. Hopkins, E. F. 1923. The behavior of hard seed of certain legumes when subjected to conditions favorable to germination. Proc. Assoc. Seed Anal. N. Am. 14:46-48.

7. Kendall, W. A., and Stringer, W. C. 1985. Physiological aspects of clover. Pages 112-115: Clover Science and Technology. N. L. Taylor, ed. Am. Soc. Agron., Madison, WI.

8. Leggatt, C. W. 1927. The agricultural value of hard seeds of alfalfa and sweet clover under Alberta conditions. Sci. Agric. 8:243-266.

9. Leggatt, C. W. 1928. Investigation into the agricultural value of hard seeds of alfalfa under Alberta conditions. Proc. Assoc. Seed Anal. N. Am. 19:37-39.

10. Legatt. C. W. 1930. Further studies on the hard seed problem: Alfalfa and sweet clover. Sci. Agric. 11:418-427.

11. Lute, A. M. 1942. Clipping alfalfa seed and result on germination. Agron. J. 34:90-99

12. Weihing, R. B. 1940. Field germination of alfalfa seed submitted for registration in Colorado and varying in hard seed content. Agron. J. 32:944-949