Effect of canola seed size on crop development and yield

Take home messages

• There was no yield difference between small and large sized canola seeds in any of the
  varieties tested.
• Small-seeded ATR Bonito had significantly lower establishment than the large-seeded ATR Bonito and the small and large seeded hybrid varieties.

Background

Seed size in canola can be highly variable, even in certified seed which has been cleaned, graded and then treated. Retained seed, particularly after a dry season, can be significantly smaller than seed purchased directly from the breeder.

The seed is the energy source available to the plant during germination and early root formation, up until the plant produces leaves and is able to photosynthesise. A smaller seed will have less stored energy and therefore reduced ability to establish and possibly lower vigour during the early stages of growth. Any reduction in early establishment and growth has the ability to translate into a yield penalty later in the season, either directly through yield, or indirectly through increased susceptibility to weed, insect or disease pressure.

Hybrid canola has a series of beneficial traits so it generally yields higher and has better establishment and early vigour than open-pollinated (OP) varieties. These advantages in early growth may mean hybrids are less likely to be affected by any negative effects of seed size. Previous studies in this area have found less dry matter accumulation, and reduced establishment as a result of small seed size in open pollinated varieties compared to hybrids (Brill, Jenkins and Gardner 2014). However, these trials are predominantly lab or glasshouse based and generally focus on early development rather than yield.

Aim

To determine how seed size from purchased seed effects crop development and yield, and whether this differs between hybrid and open pollinating varieties.

Paddock details

Trial details

Trial inputs

Method

A replicated field trial was sown using a complete randomised block trial design. Assessments included establishment counts, Normalised Difference Vegetation Index (NDVI) assessments at three different timings, crop biomass at flowering, grain yield, quality measurements and 1000 grain weights. Purchased canola seed was graded into sizes using a two millimetre screen and seed was selected from above and below the screen.

Results and interpretation

There was no yield difference attributed to seed size in any of the varieties tested despite early differences in establishment and biomass (Table 1). The site received high spring rainfall and it is possible that in a lower rainfall year the effects of seed size may persist and translate into yield. The good break and warm winter meant growth was rapid which likely negated any small differences at sowing. There was no statistical difference between the yields of the three varieties although there was a varietal difference in biomass and the two later NDVI timings.

At establishment ATR Bonito (OP) with a seed size of <2mm, had significantly fewer plants per meter than the hybrid varieties and the large seed size OP, which all had similar establishment. This shows that seed size does have the ability to reduce establishment and the hybrid varieties were more resistant to the effect of seed size at establishment.

All NDVI measurements showed a reduction in biomass and greenness with small seed size up to early flowering. There was an indication of interaction between variety and seed size, however this response wasn’t quite significant at the three-leaf stage and only lasted up to the five-leaf stage. In general, ATR Bonito was the most affected by seed size. In Hyola559TT the effect was present but reduced and Hyola474CL showed little to no difference between seed size treatments.

The difference in NDVI in ATR Bonito may have been due to lower establishment in the small seed treatments, but Hyola559TT had no difference at establishment, suggesting that reduced vigour was the cause.

ATR Bonito and Hyola559TT showed reduced NDVI with smaller seed size at flowering. This effect however, was not evident in the biomass samples taken at the same time. Therefore, it could possibly be attributed more to differences in crop colour, flowering or senescence rather than biomass.

Seed size had no effect on grain quality. Hyola559TT had a mean oil content of 40.6 per cent, Hyola474CL was 39.6 per cent and ATR Bonito was 39.4 per cent.

Although there was no yield response, there was a significant indication that small seeds result in small seeds, even after good growing conditions. When small seed was used 1000 grain weight was lower in ATR Bonito and Hyola559TT (Table 1). This effect was not the same in the case of Hyola474CL.

Figure 1. Mean establishment (plants/m2). Error bars indicate significant difference (LSD). Stats: P=0.022, LSD=11.0 plants/m2, CV=18.6%.

Table 1. Yield, establishment NDVI and biomass results for seed size and variety.

Commercial practice

Canola can be an extremely forgiving crop in terms of recovery from reduced establishment and vigour, provided there is adequate nutrients and moisture.

In this trial, grain yield was not influenced by seed size in any of the varieties despite a reduction in establishment and NDVI with small-seeded canola. This could reduce yields in a drier season as has been seen in other research where the effect of seed size is more pronounced under drought conditions (Mian and Nafziger 1994).

Using hybrid seed helped to reduce the negative effects at establishment but there was still some loss of early vigour in Hyola559TT. This reduction in establishment or early vigour did not affect yield this year. Small-seeded ATR Bonito had lowest establishment but still performed as well as the other varieties and large seed sizes.

On-farm profitability

This trial used certified purchased seed graded into seed size. If establishment of small seeded canola is a concern it may be worth selecting for larger seed size, either by purchasing hybrid varieties ($22/ha) which tend to have larger seed size, or screening open-pollinated seed on farm ($32/t).

GRDC recommends cleaning with a two millimetre screen if retaining open pollinated seed (Brill et al. 2014). Open pollinated seed is less costly ($11/ha) but may have reduced performance compared to hybrid seed. It may be possible to select larger seeded varieties, however this may be a balance between increasing establishment with larger seeds and reducing it with less plants per hectare.

If using larger seed, it may be worth increasing the weight per hectare but this will also incur extra costs.

References

Brill R., Jenkins L., and Gardner M., 2014, GRDC Research Papers, ‘Canola establishment; does size matter?’.

Mian M., and Nafziger E., 1994, Crop Science 34(1), ‘Seed size and water potential effects on germination and seedling growth of winter wheat’.

Acknowledgements

This trial was funded by Agritech Rural and BCG members through their membership.