I was vaguely aware of the Composite Cross II (CCII) long-term experiment with barley, not least because of a Brainfood entry a couple of years ago. But I didn’t know a whole lot about it, so when a link to a recent thesis by Jill Marzolino at UC Riverside referencing it popped up in my feeds, I decided to look into it a bit more.
Harry Harlan and Mary Martini set up the experiment way back in 1929, in an effort to come up with barley varieties better adapted to the Californian environment. They started out with 28 diverse barley varieties from all over the world, made all possible crosses among them (though in only one direction), bulked together all the seeds they got, and planted a random sample of the mixture at Davis, California. The next year, they harvested the resulting crop, and sowed a sample of the seeds they obtained again in the same place.
And so on, for decades. Researchers following in the footsteps of Harlan and Martini planted 5,000-20,000 seeds year after year and left them to it, for 58 generations, saving a sample of the harvest along the way. This is called a composite cross population, or sometimes a composite hybrid mixture. When used in crop improvement, the process is sometimes referred to as evolutionary plant breeding, and has been proposed as a useful strategy for adapting crops to climate change.
Anyway, in the days of DNA sequencing, you can also see how CCII is an incredible resource for just trying to figure out how — and how fast — evolution works.
The paper I referenced in Brainfood in 2024 did just that. In Natural selection drives emergent genetic homogeneity in a century-scale experiment with barley, the authors showed that natural selection pretty quickly and drastically narrowed the genetic diversity of the original diverse population, affecting especially genes regulating the timing of flowering and reproduction. That basically allowed the plants to avoid drought. Yield also doubled, though that was less than plant breeders were able to achieve over the same period.
The thesis that caught my eye today is entitled Uncovering the Genetic Basis of Local Adaptation With Long-Term Evolution Experiments in Barley, and comes from the same lab. Dr Marzolino also looked at flowering time, but the chapter that really struck me was the one on “genetic rescue.” She found that when she planted the CCII at a different site, in Bozeman, some genetic variants that had been incredibly rare at Davis suddenly exploded in frequency. In her words:
This result indicates that very rare maladapted types persist in the CCII. It is not clear how they are maintained, but perhaps fluctuating environmental conditions from year to year alter the fitness of these types enough for them to persist… It remains unclear how long these rare types can be maintained in a population. Maintenance of rare types could be critical for the longer term survival of many plant species in a changing world.
Something for genebank managers to ponder. The normal practice is to regenerate genebank accessions under environmental conditions which are as close to ideal for the particular germplasm in question as possible, in order to ensure a good harvest of high quality seeds. But this result suggests it might actually be worth considering occasionally subjecting sub-samples to a few regeneration cycles in a contrasting environment. Might that not rescue some interesting — not to mention useful — genes?