How we classify collections

Every Zomia field report classifies its seed collections by the tests on this page. Each report names the version of the method it used. It also gives its own evidence for each test in a table.

1. What we mean by landrace

We follow the definition of a landrace by Casañas and his colleagues. A landrace is a cultivated variety that has evolved, and may go on evolving, within a defined area and under the influence of the local human culture. It may be grown in old ways or new ones. It may be selected by farmers or by breeders, with traditional or modern techniques.¹ The word does not necessarily mean the population is pure or unchanged. Landraces are always evolving under natural selection and under the selection of the people who grow them. In some cases they may also mix with other landraces and varieties.¹

2. Three counts

We classify each collection on three counts. The first is its class. The second is whether it is autochthonous, meaning grown in the area from local seed for at least fifty years (section 4). The third is whether other varieties have been crossed into it. Each finding rests on something we saw, something we were told or something we read. Each report’s classification table says which.

3. Class

A collection falls into one of three classes. A landrace meets the definition in section 1. To count as one it must pass three tests. People sow and tend it. It has evolved in a defined area. The people who grow it shape it through how they choose, keep and use the seed. A landrace that has had other varieties crossed into it is still a landrace; the crossing is recorded as introgression. A hybrid is cultivated seed of a modern bred variety, or of a cross with one, that has not yet been shaped in this way in the area. A feral collection comes from plants that grow and set seed without anyone sowing or tending them. For a feral collection we also say how the plants compare with the cultivated plants nearby.

4. Autochthonous

Zeven calls a landrace autochthonous when it has been grown for a long time in the farming system where it is found. He calls it allochthonous when it is a landrace from another region that was brought in recently.^2: 137 Whether an allochthonous landrace becomes autochthonous depends, Zeven says, on the number of generations it has been grown since and on how often its seed is changed.^2: 137 Mayr required more than a century in the same region.^2: 132 We use fifty. Cannabis is an annual, so one generation is one year. In France, wheat populations drawn from one mixed source population were grown from 1984 at sites from Toulouse to Vervins, with no deliberate selection. After ten generations the southern populations headed earlier and needed less winter cold than the northern ones. The authors conclude that local climate greatly influenced this.³ Fifty generations is five times as long. The extra time allows for traits controlled by many genes, such as tolerance of drought or cold, which are expected to respond to selection more slowly.^4: 967 No study we know of has measured this in cannabis, so the figure is our judgement.

We count a landrace as autochthonous when it has been grown from local seed in the area for at least fifty years before the collection, with no known point at which its seed was replaced with seed from outside the area. Seed passed between farmers or villages in the same area does not count as replacement. We accept two kinds of evidence. One is a written record of cannabis grown in the area at least fifty years before the collection. The other is growers who learned to grow it from parents or grandparents who grew it there. A landrace brought in within the last fifty years is allochthonous. If the evidence falls short either way, we record it as unknown.

5. Introgression

Introgression is genes from another variety getting into a population through crossing. We grade it in four steps:

• None seen or reported.
• Reported nearby: seed from outside was said to have been sown in the area, but not into this population.
• Reported in this population.
• Seen in the plants: some plants in the field show the traits of another variety.

Without a genetic test we can say only that we found no sign of introgression. We cannot say that there is none.

6. References

Footnotes

1. Casañas, Francesc; Simó, Joan; Casals, Joan; Prohens, Jaime (2017). “Toward an Evolved Concept of Landrace”. Frontiers in Plant Science. 8 145. doi:10.3389/fpls.2017.00145. PMC 5296298. ↩ ↩²

2. Zeven, A. C. (1998). “Landraces: a review of definitions and classifications”. Euphytica. 104: 127–139. doi:10.1023/A:1018683119237. ↩ ↩² ↩³

3. Goldringer, Isabelle; Prouin, Claire; Rousset, Michel; Galic, Nathalie; Bonnin, Isabelle (2006). “Rapid Differentiation of Experimental Populations of Wheat for Heading Time in Response to Local Climatic Conditions”. Annals of Botany. 98 (4): 805–817. doi:10.1093/aob/mcl160. ↩

4. Bellon, Mauricio R.; Dulloo, Ehsan; Sardos, Julie; Thormann, Imke; Burdon, Jeremy J. (2017). “In situ conservation: harnessing natural and human-derived evolutionary forces to ensure future crop adaptation”. Evolutionary Applications. 10 (10): 965–977. doi:10.1111/eva.12521. PMC 5680627. ↩