Germination requirements: time

Not all seeds in a germination test will germinate at the same time, and it is important to give seeds that do not germinate immediately the chance to germinate over a longer period.

In module 2, you explored seed survival curves for the viability of seeds in storage, and learned how these patterns can be modeled to make predictions. Just as patterns of seed viability change over time in a predictable way, it is also possible to describe patterns of seed germination using mathematical formulae. Figure 7 (below) draws on the work of Joosen et al., who modeled the germination of thale cress, Aridopsis thaliana, over time:

  • Gmax is the final number of seeds that germinated at the end of germination testing.
  • t50 is the time it takes for half of the seeds to germinate.
  • U7525 is uniformity - the time to go from 25 to 75% germination.
The image shows a typical S-shaped curve for germinating seeds and the information that can be read from it. The x axis is the percentage of seeds germinating. The y axis is the time spent in the germination test. A horizontal line ruled from 50% germination meets the curve at 60 hours – this is t50, the time it takes for half the seeds to germinate. A horizontal arrow at the top of the curve indicates Gmax, the final number of germinated seeds. A horizontal line spanning the steepest part of the S-shaped curve indicates U7525, time it takes to go from 25% to 75% germination, also known as the uniformity. The area under the curve is shaded and labelled as “AUC”.
Figure 7: germination over time

Figure 7 mirrors what happens in the real world, for instance, t50 and U7525 can help you plan when to score germination. The area under the curve, AUC, provides useful information about both the speed and the final germination rate.

  

Pause for reflection

The beauty of mathematical modeling of germination is that it allows the possibility of predicting the number of seeds that can be expected to germinate at any particular point in time. In future, procedures involving visual imaging and machine learning could take the drudgery out of germination testing, by automating scoring. Although these processes are still experimental, in future they could allow high-throughput scoring of germination tests. Take five minutes to think about how this technology could affect genebanks.

Germination requirements: light

Scoring germination