Elmer Stobbe, Jack Moes, Yantai Gan,

Harry Ngoma, and Luc Bourgeca

Department of Plant Science

University of Manitoba

Introduction

Seed vigour refers to the ability of seed to germinate, emerge, and produce a good crop under a wide range of environmental conditions. Our studies, initiated in 1986 demonstrated that pedigreed wheat seedlots of the same cultivar vary in vigour, that is, they vary in respect to emergence and productive potential. Yield differences between the best and worst seedlots were as much as 12%, and on average, seedlots yielded 5.3% less than the best seedlot (Table 1). Subsequent research was initiated on the hypothesis that seed crop management could be altered to improve the vigour of pedigreed seedlots.

Several characteristics of cereal seed (as well as seed of many other species) seem important determinants of seed vigour. These can be grouped into two broad categories: innate physical characteristics of the seed such as large seed size, high density, and high protein; and various types of seed damage, such as weathering, mechanical (threshing and handling), and drying damage. In our research, we have approached concerns about seed vigour from two points of view: that of the grain producer buying pedigreed seed, in terms of the effect of the vigour determining characteristics on crop performance under Manitoba conditions; and that of the seed producer, in terms of identifying management practices which are most likely to be conducive to producing high vigour seed - ie. large, dense, high protein seed which is free from weathering, mechanical, and drying damage.

Research Results

Large, High Protein Seed

Crops grown from large kernels consistently yield higher than crops grown from small kernels of the same cultivar, for both wheat and barley. For Roblin wheat for example, increased yield is associated with increasing seed size (Table 2). Seed sizing offers an important low- or no-cost way of improving yield potential - ie. if only the largest kernels are used for seeding, yield potential is maximized.

Screening more rigorously for seed size increases the amount of screenings below the minimum desired size. A seed grower obviously wishes to minimize screenings and retain the maximum amount of seed for sale. Consequently, some of our research has focussed on management practices to produce the maximum percentage of seed, or conversely the minimum percentage of cleanout. In particular, we have looked at seeding rate, nitrogen management, and foliar fungicide applications.

Seeding Rate: At low seeding rates, cereal plants tend to produce tillers. As seeding rate increases, the production of tillers declines until the plant density is such that only main stems will produce heads. At low plant densities, kernels on main stems are relatively large, while progressively later appearing tillers produce progressively smaller kernels. At high plant densities, where kernels are produced on the mainstem only, few very large or very small kernels are produced - ie kernels within a seedlot are more uniform. At very high plant densities, average kernel size may decline as competition between plants for light, water, and nutrients becomes severe.

We have conducted studies to investigate the effect of seeding rate on seed size distribution in wheat. In general, percent cleanout (<6/64") tended to decline as seeding rate increased (Fig. lc). Seed yield (>6/64') tended to increase with seeding rate (Fig. lb). With respect to differential cultivar responses, in terms of seed yield, the two cultivars Katepwa and Roblin, responded similarly to increases in seeding rate, although for Katepwa, seed yield tended to peak at 200 seed/m², while for Roblin, seed yield tended to peak at 300 seeds/m² (data not shown).

Seasonal rainfall amount and distribution appear to contribute to overall cleanout levels and seed yield (Table 3). June of 1988 was very dry, resulting in a low yield potential being set, and the dry conditions also led to a high proportion of small seed being produced. In 1989, June rainfall was very good, but July was dry, therefore, although the yield potential was set high in June, dry conditions during July led to production of a relatively high proportion of small seed. In 1990, an excellent yield potential was set in June, and filling conditions in July were good; therefore, seed yield was very high and clean out was very low.

We now recommend a seeding rate of 300 viable seeds/m² to optimize seed yield and minimize percent cleanout for Roblin wheat, whereas for Katepwa, 200-250 viable seeds/m² may be adequate. For wheat such as Katepwa, with a thousand kernel weight of 34 g, and standard germination of 98%, a seeding rate of 250 viable seeds/m² corresponds to 78 lb/acre. In barley, our field trials suggest that we use a similar seeding rate as for wheat, that is 300 viable seeds/m².

Foliar Fungicides: Late season fungicide applications have been shown to increase kernel size in both wheat and barley. Usually this is attributable to protection from foliar diseases, but there have been reports of beneficial effects that were not attributable to disease control.

Fungicide application to barley under severe disease pressure frequently caused large increases in the yield of large seed (>6/64") (data not shown). Yield increases for wheat are smaller, but under severe disease conditions, may also warrant application of fungicide to control foliar diseases.

Weathering and Seed Vigour

Field weathering exposes seeds to cycles of wetting and drying and to high atmospheric humidity. Weathering causes an irreversible decrease in kernel density, reduced kernel weight, lower bulk density, and premature enzyme activation, especially alpha-amylase. Studies were conducted to determine the effect of field weathering on seed vigour of Katepwa wheat and Heartland barley.

Laboratory Vigor Tests: Weathering of wheat or barley seed affected results of both standard and cold germination tests (Table 4). In particular the severely weathered lots of wheat in 1990 and barley in 1989.These results would suggest that performance of these seed lots may be different under field conditions.

Weathered Seed-Field Performance: Our studies show that severe field weathering can adversely affect

seed vigour as determined by laboratory tests. However, our results from evaluations of field performance of weathered seed have been less clear. Field weathering appeared to affect barley more than wheat.

For wheat, no significant differences in emergence could be detected among seed lots under any seedbed condition (Table 5). At the 3 inch seeding depth, the unweathered seed tended to have a higher emergence count than the severely weathered seed (data not shown). However, severely weathered seed appeared to emerge more quickly than unweathered seed for both wheat and barley.

In 1990, total emergence of unweathered barley seed was significantly higher than severely weathered seed only under conventional tillage, particularly when deep seeding was practiced (Table 6). In1991, total emergence was not affected by weathering. The spring of 1991 was warm at time of seeding, thus providing ideal conditions for germination.

In wheat, the relationship of seed weathering to grain yield was variable (Table 7). No meaningful differences among wheat seed lots in grain yield could be detected in either year.

Although not consistent, significant yield differences were observed between barley seed lots (Table 7). In 1990 in a Fortress treated area, the most weathered seed produced a significantly lower yield than unweathered seed. Under zero tillage, severely weathered seed tended to have the lowest yield. This yield pattern was more consistent with deep seeding than with shallow seeding under all conditions. In 1991, a reverse trend was observed, where severely weathered barley seed had significantly higher yield than the unweathered lot under all seed bed conditions. This difference was not related to total seedling emergence (Table 5), but was related to speed of emergence.

Our data indicate that weathering may not be a critical factor under favourable germination environments. However, field weathering may reduce plant stand, and consequently yield, especially in marginal conditions.

Threshing and Seed Vigour

Impact forces such as those experienced during the threshing process can cause external seed damage, but also may cause internal damage that can compromise both viability and vigour. In a rub-bar type threshing mechanism, impact forces on seeds increase with cylinder speed.

Kernel water content (KWC) has a great influence on kernel mechanical properties, therefore, it also influences the degree to which kernels are damaged in the threshing process. Kernels at low KWC are susceptible to breakage, while those at high KWC are susceptible to permanent deformation. The range from 14-16% KWC appears to be optimum for minimal external- threshing damage.

Severe threshing may be manifested not only in visible damage, but also in invisible damage, which in turn may result in decreased viability and vigour of the seed. High cylinder speed is a key factor in causing both visible and invisible damage in wheat, with invisible damage being manifested in reduced germination.

Seed Damage during Threshing: Two major types of combines were evaluated: the conventional type with a transverse cylinder, in which material flow is tangential to the cylinder; and the axial-flow (or rotary), in which material flow is parallel to the axis of the cylinder.

For both combines, samples threshed at low speed had a high percentage of white caps and a low percentage of broken seed and the lowest grain yields (Table 2a,b). The number of kernels recovered from a spike was therefore best at high cylinder speed. However, high cylinder speed increased the impact forces on kernels resulting in high number of broken seeds.

The conventional combine had lowest percentage of total dockage (dockage, broken seeds and white caps) at the optimum cylinder speed (Fig. 2a,b). Thus the yield of visually undamaged seeds was equivalent for both high and optimum speeds. The rotary combine resulted in higher yields of visually sound seeds than for the conventional combine at any speed, because the dockage percent was reduced.

Cylinder Speed and Seed Vigour: For both years and both combines, the most dramatic loss of germination occurred in samples harvested at high cylinder speed (Fig. 3a, b). In cold tests, germination of visually undamaged seeds decreased with increasing cylinder speed. For seeds threshed at high speed with a conventional combine, 15 to 25% of germinated seedlings exhibited abnormalities such as stunted or missing seminal roots, or damaged coleoptiles (Fig. 3c,d). These abnormal seedlings can not be expected to emerge as well as normal seedlings under typical stress conditions in the field. Non-visual damage was also indicated by increased leachate conductivity with increased threshing cylinder speed (Fig. 3e,f).

Field emergence decreased as cylinder speed increased, but to a lesser extent for the rotary combine than for the conventional combine (Fig. 4a). Differences in field emergence were larger than differences in cold germination percentages. The failure of abnormal seedlings to emerge may account for this.

Final yield was related to cylinder speed (Fig. 4c), with yield decreasing as cylinder speed increased for both combines. The low yield of plants produced from seed harvested at high cylinder speed may be due to low vigour of abnormal seedlings during establishment.

Combine Type and seed Quality: Yield differences were observed between the combines and among the cylinder speeds. Cylinder speed had a relatively small effect on yield for the rotary combine. In contrast, minor modifications of cylinder speed for the conventional combine resulted in relatively large yield differences.

For the conventional combine, the highest vigour seed was obtained with low cylinder speed. Unfortunately, low cylinder speed resulted in insufficient bu/acre. Conventional combine users are confronted with a dilemma: obtaining high vigour seed with low cylinder speed leads to inefficient threshing and yield loss. Therefore, the optimum cylinder speed for both years was the best compromise between threshing efficiency and seed quality.

The best seeds were obtained with the rotary combine. Differences between cylinder speeds in seed quality and threshing efficiency were relatively small. However, even with the rotary combine, we recommend setting the cylinder speed as low as possible (less than 800 RPM) because higher cylinder speeds did reduce seed vigour.

Seeding Depth and Crop Performance.

Seeding Depth and Emergence: Emergence exceeded 80% in both wheat and barley at shallow seeding in 1990 (Fig. 5 a,b). Increased seeding depth resulted in significantly lower plant populations for both crops. Emergence time was delayed by three days when planting depth was increased from 1 to 3 inches. Results were similar in 1991 (data not shown). With deep seeding, barley coleoptiles appeared to rupture below the soil surface. The leaves below the soil surface were unable to push up through the soil.

Seeding Depth and Yield: Seeding depth had a significant effect on grain yield for both wheat and barley in1990 (Fig. 6a,b) and 1991 (not shown). Shallow seeding resulted in consistently higher yield than medium or deep seeding. Yield differences between depths were less in 1991 compared to 1990 because of weather differences. In 1991, drought at grain filling may have reduced yield in the shallow seeded plots. Yield differences between seeding depths were more pronounced in barley than in wheat.

Table 1. Grain yield produce from seed coming from different seed growers throughout Manitoba in 1986 and 1987 (grown at Portage la Prairie).

Table 2. Effect of seed size on grain yield of Roblin wheat (Portage la Prairie MB, 1990).

Table 3. Rainfall data for growing seasons at Portage and Carman experimental sites.

Table 4. Standard and cold germination percentages for weathered seedlots of Katepwa wheat and Heartland barley.

H Means followed by different letters within each column and year are significantly different by LSD with P# 0.05.

Table 5. Emergence (plants/0.30 m²) of wheat and barley at Portage la Prairie.

H Means followed by different letters within each row and year are significantly different by LSD with P<0.05.

Table 6. Final emergence (plants/0. 30 m²) of barley seed lots at 7.5 cm.

H Means followed by different letters within each row and year are significantly different at P< 0.05 (LSD) test.

Table 7. Yield of spring wheat and barley at Portage la Prairie.

H Means followed by different letters within each row and year are significantly different at P<0.05 (LSD) test.

I WCT=Wheat conventional tillage; WFR=Wheat conventional tillage with Fortress; WZT=Wheat zero tillage; BCT=Barley conventional tillage; BFR=Barley conventional tillage with Fortress; BZT=Barley zero tillage.

Fig.1

Fig.2

Fig.3

Fig. 4

Fig. 5

Fig.6