Master 2, specialty “Seed Science and Plant Propagation”. Duration: about 2 hours. Work in pairs.

At the end of this session you should be able to:

• explain how temperature and water potential control the rate and the final percentage of germination;
• explain why seeds of the same lot do not all germinate at the same time, using the concept of a population threshold (base water potential);
• read the hydrothermal time equation and calculate the germination time of a single seed;
• compare two seed lots and explain why a standard germination test may hide differences in vigour;
• predict and interpret germination curves under stress, in a virtual germination experiment.

You do not need to be a programmer. Every change you are asked to make is a change of one value at the top of the file.

1. Start GroIMP and open the germination project (File → Open).
2. Open the code in the text editor (Panels → Explorers → Files, then double-click the file).
3. The 3D view shows a germination tray of 100 seeds: the left half is seed lot A, the right half is seed lot B.
4. Each click on the step button advances the simulation by one hour. Use the run/loop button to run until the end (200 h). A run takes only a few seconds.
5. To start again, press the reset button, or save the code (Ctrl+S): the model is recompiled and a new tray is sown.

What you see:

• Seeds change colour from brown to light green as they progress towards germination. A germinated seed turns green and grows a white radicle.
• The chart Cumulative germination shows the percentage of germinated seeds of each lot over time.
• The XL Console shows the conditions of the experiment, the expected final germination of each lot, the germination percentages every 10 hours, and at the end the final germination and the t50 (time until 50 % of the sown seeds have germinated).

Germination starts with the uptake of water by the dry seed (imbibition) and ends when the radicle protrudes through the seed coat. Water uptake follows three phases: a fast physical uptake (phase I), a plateau during which metabolism is reactivated (phase II), and a new increase in water content when the radicle starts to grow (phase III). Germination in the strict sense ends at the beginning of phase III.

Two environmental factors control the rate of germination:

• Temperature: below a base temperature (T_b) seeds do not germinate; the rate increases up to an optimal temperature (T_opt) and decreases above it, until a maximum temperature (T_max).
• Water potential (ψ, in MPa) of the substrate: pure water has ψ = 0 MPa; a drying soil or a salt or PEG solution has a negative ψ. Each seed has a base water potential ψ_b below which it cannot complete germination.

In a seed lot, the seeds are not identical: some are fast, some slow, some never germinate. The hydrothermal time model (Gummerson 1986; Bradford 1990) explains this with a single idea: all seeds need the same amount of “hydrothermal time”, but each seed has its own base water potential.

Questions (answer in 2–3 lines each):

1. Q1. What is the difference between imbibition and germination? Can a dead seed imbibe?
2. Q2. Give two ways to create a substrate with a water potential of −0.5 MPa in the laboratory.
3. Q3. A standard germination test (ISTA) is carried out on moist paper at an optimal temperature. What is a vigour test, and why do seed companies need both?

Every seed accumulates hydrothermal time θ at each hour t:

dθ/dt = (ψ − ψ_b(g)) · (T − T_b)          (only when both terms are positive)

and germinates when θ reaches the constant THETA_HT (540 MPa·°C·h), which is the same for all seeds. ψ_b(g) is the base water potential of the seed. In a seed lot, ψ_b(g) follows a normal distribution with median ψ_b(50) and standard deviation σ.

In the code:

sd:Seed ::> {
    float T = temperature(time_hours);
    float psi = waterPotential(time_hours);
    float psiBase = sd[psib] + psiBaseShift(T);
    sd[theta] += DT * Math.max(0, psi - psiBase) * thermalTerm(T);
}

and a seed germinates when theta >= THETA_HT.

The two seed lots differ only in the distribution of ψ_b(g):

• Q4. Calculate the germination time of a seed with ψ_b = −1.0 MPa at 20 °C in pure water (ψ = 0, T_b = 2 °C). Then for a seed with ψ_b = −0.5 MPa.
• Q5. Same two seeds, but on a substrate at ψ = −0.5 MPa. What happens to the second seed?
• Q6. Why does the model need only one value of θ for the whole population, but a distribution of ψ_b? What does this distribution produce in the germination curve?
• Q7. Seed ageing shifts ψ_b(50) upwards (towards 0) and widens the distribution. Using Q4–Q6, predict how an aged lot behaves (a) in pure water and (b) under water stress.

float thermalTerm(float T) {
    if (T <= T_BASE || T >= T_MAX) return 0;
    return Math.min(T, T_OPT) - T_BASE;
}
float psiBaseShift(float T) {
    return (T > T_OPT) ? K_T * (T - T_OPT) : 0;
}

Below T_opt (20 °C), warmer means faster. Above T_opt, the thermal term stays constant, but the base water potential of every seed rises by K_T = 0.05 MPa per °C (Alvarado & Bradford 2002).

• Q8. At 30 °C, what is the base water potential of a seed that has ψ_b = −1.0 MPa at 20 °C? What does this mean for germination at high temperature (thermoinhibition)?

Under constant conditions, a seed can germinate only if ψ_b(g) < ψ. The console therefore prints the expected final germination of each lot, which is the share of the normal distribution of ψ_b(g) lying below ψ.

• Q9. Using the parameters of lot A, what proportion of seeds has ψ_b < −0.5 MPa? (Hint: how many standard deviations is −0.5 above −1.0?)

Run the model with the default values (20 °C, pure water) until the end (200 h).

• Q10. Both lots pass a standard germination test (≥ 85 %). Do they have the same quality? Which differences do you see?
• Q11. Why is the germination curve S-shaped? Link it to the normal distribution of ψ_b(g).

Method, for each experiment: write your prediction first, change one value at the top of the file, save, run to 200 h, record the results, then restore the original value.

Change PSI_0 (water potential of the substrate):

• Q12. How do the rate (t50) and the final percentage respond to water stress? Which lot is more affected? Why?

Change T_MEAN (PSI_0 back to 0):

• Q13. Plot (on paper) the germination rate 1/t50 of lot A against temperature. Where is the optimum? Why does germination decrease above 20 °C in this model, although the thermal term does not decrease?
• Q14. At 5 °C, the final germination is low after 200 h. Is this a real loss of viability? How would you check?

In the field, conditions change over time. Try:

• Day/night temperatures: T_AMP = 8 (temperature varies between 12 and 28 °C around a mean of 20 °C).
• A drying seedbed: DRY_RATE = 0.005, then 0.01 MPa per hour (the substrate loses water after sowing).

• Q15. Why does a fluctuating temperature around 20 °C slow down germination compared with a constant 20 °C?
• Q16. In a drying seedbed, a lot can be “caught” by the drought before it has germinated. Which lot suffers most? What could a farmer do?

• Q17. You must choose a seed lot for an early sowing in a cold, dry spring. Which lot do you choose, and which experiment(s) of this session support your choice?
• Q18. Why do seed laboratories use vigour tests (for example germination under osmotic stress or cold test) in addition to the standard germination test? Use your results.
• Q19. List three processes that the model does not include, although they affect germination in real seed lots (think of dormancy, the imbibition phases, seed-borne pathogens, and the difference between germination and seedling emergence).

For fast groups. In priming, seeds are imbibed at a water potential just too low for germination, then dried back. They keep part of the progress made towards germination.

In the model, this means that primed seeds start with part of THETA_HT already accumulated. In init(), the seeds are created with Seed(drawPsiB(lot), 0, …): the second value is the initial θ.

1. Add a constant PRIME_FRACTION = 0.5 at the top of the file.
2. Replace the initial 0 by PRIME_FRACTION * THETA_HT.
3. Compare t50 and the spread of germination times with the reference run, in pure water and at ψ = −0.5 MPa. Does priming raise the final germination percentage? Why (not)?

This exercise is not handed in, but the questions above cover the kind of reasoning expected in the exam: explaining how temperature and water control germination, calculating with the hydrothermal time model, and predicting and interpreting germination curves. Keep your answers and tables as revision notes.