I have had a few enquiries this autumn about the fate of black-grass seed after shedding. This type of question is fostered by rather imprecise language describing the number of black-grass seeds formed on a head of the weed. On average one head of autumn germinating black-grass sets 100 
seeds; a convenient number. But are all these seeds viable as some claim?
The simple answer is no. Although, of necessity there is a lot of variability involved in determining the fate of black-grass seed, the average figures are intriguing. The best source of such data is page 86 of HGCA Project Report 466 – Integrated Management of Herbicide Resistance.
The data suggest that on average only 55% of the seeds are viable. This still seems a horrifying situation; on every black-grass head there are 55 viable seeds. However, be (relatively) comforted by the huge subsequent losses.
Only 45% of the seeds survive the period between shedding and sowing a following crop in late September. So 45% of the 55 viable seeds per head make it to a stage where they can potentially become a weed in the following crop. However, many are buried by cultivations to a depth from which they will not emerge, assumed to be more than 5 cm. For instance, non-inversion tillage carried out to a depth of 20 cm buries 40% of freshly shed seeds deeper than 5 cm. As you can see, the losses keep mounting and there is more to come.
Only a proportion (15%) of these freshly shed seeds in the top 5 cm will actually germinate and establish plants in the following crop. All this means that for every head of autumn germinating black-grass that survives to produce seed in June, approximately 2.2 plants will emerge in a following late September sown crop established after non-plough tillage to a depth of 20 cm.
Of course life (certainly the life of black-grass) is not that simple. Some of the seeds shed will survive up to three or more years in the soil. There is a loss of about 70% of viable seeds a year but this is partly compensated for by the fact that a higher proportion (30%) of over-yeared viable seeds will germinate, because of the loss of dormancy, if they are in the top 5 cm of soil.
This complicates the maths but when it is all worked through, the seeds shed from one head will result in around three plants being formed over the following three years in continuous late September sown crops established after non-inversion tillage to 20 cm depth. So there is some comfort to be had from this information on seed losses but please note that these are average figures.
The factor that I have not mentioned until now is the number of fertile tillers (i.e. heads) per plant. Generally, the earlier the autumn sowing the more heads per plant are formed. In addition, extreme summer weather events can make a huge difference. In 2012, the continuous wet conditions resulted in low black-grass tiller losses and so very high heads/plant. In contrast, the very dry spring of 2011 resulted in tiller losses right up to head emergence, leading to very low heads/plant.
In conclusion, and in the context of total seeds set, natural losses in continuous late-September sown cropping are higher than those that can now be achieved with cereal herbicides. However, these losses are a long way from being high enough to avoid the use of chemicals.
There are a lot of research and field observations being carried out to see how the number of natural losses can be increased, for example by later drilling. Head numbers per plant can be reduced by higher crop seed rates and the weed’s lifecycle can be disrupted by spring sowing. It is heartening that we can do things that enhance natural losses but they all cost money and, for many, will result in systems that are less reliable than continuous early-autumn sown crops.