- The Lebanese and Syrian genebanks in the news. For good reasons, for now at least.
- Wild American apples should be more in the news. And probably more in genebanks.
- Community seed banks could be good news in fragile states.
- Good news for India’s banana diversity. Yes, it now has a genebank!
- All those genebanks need breeders, like Mina Nešić.
- Genebanks are nice of course, but it’s even better news when the agrobiodiversity gets out and about.
Underselling breeding, and conservation
Crops with massive … importance, clear biological upside, and real demand for better genetics — but a system where breeding remains small, underfunded, and structurally difficult to scale.
What crops, you ask? “Opportunity crops,” perhaps? Fonio, say, or Bambara groundnut, or any number of African leafy vegetables.
Those would have been good guesses, but actually I cheated, so no. The word hidden by the ellipsis is actually “economic,” and the quote comes from a Reddit post on coffee breeding. 1
That of course makes the observation even more amazing. As the Reddit poster goes on to point out:
We’ve built a ~$100B global industry that depends on plant genetics… while seemingly allocating only a negligible fraction of that value to actually improving those genetics.
And, I would add, allocating an even more negligible fraction to conserving those genetics (despite the fact that there’s a pretty solid strategy for how to do that). Which goes for opportunity crops too, come to think of it.
Rare today, relevant tomorrow: new lessons from a really old barley experiment
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?
Beat the heat with seeds
I haven’t yet had a chance to read the full FAO–WMO joint report on Extreme heat and agriculture, but some preliminary skimming reveals that agrobiodiversity does seem to be addressed, at least to some extent:
No mention of genebanks, mind you. I guess you can’t have everything, but you’d have thought the following snippets could easily have been used to make the case very explicitly for ex situ conservation of crop diversity.
For domesticated agricultural species, human influence on the genome through selective breeding for enhanced performance in increasingly homogenous production environments has resulted in a loss of natural genetic variability that have accentuated many species vulnerability to temperature extremes.
…
It is only through innovation and the implementation of adaptative measures (e.g. selective breeding, making changes in the physical environment and altering management practices) that the global community can shelter agricultural activities from the larger forces of planetary human induced climate change.
…
Switching to more resilient species to extreme heat may result in reduced genetic diversity, increasing the vulnerability of crops and livestock to large-scale losses due to a narrower genetic base.
The seeds of tropical fodder grass development
Usually, if plant breeders do anything at all with wild species, they use them to try to improve the domesticated relative in some way. But in Bajra–Napier Hybrids (BNH), it’s actually the crop that is used to improve a wild (or at least wildish) relative.
That’s more than just a fun fact. BNH are actually pretty important forages in tropical and subtropical livestock systems, though you don’t hear too much about them other than from specialists. I certainly hadn’t, until a recent social media blitz from ICRISAT.
They are made by crossing the crop pearl millet (bajra, Pennisetum glaucum) with the related forage Napier grass (Pennisetum purpureum). This has the effect of combining nicely complementary traits into a highly productive fodder plant.
The best thing about BNH is their high yield of biomass. Under ideal conditions, annual green fodder production can exceed 200–300 tonnes per hectare, which comfortably outperforms other forage grass options. This productivity is due to fast growth, profuse tillering, and efficient nutrient uptake. For smallholder dairy systems, where land is usually at a premium, such a yield advantage translates pretty quickly into higher milk output per area. Also, BNH are perennial, which reduces costs over time, as fields can remain productive for several years with proper care.
And the nutritional profile of the fodder is pretty good. Crude protein is typically 8–14%, depending on management and cutting stage, while digestibility remains ok if the plants are harvested relatively early, before they start getting woody, say at 45–60 day intervals.
BNH are resilient, being tolerant to drought and intermittent water stress, a trait inherited largely from pearl millet, though they also respond well to irrigation and fertilization. That makes them widely suitable, everywhere from low-input rainfed systems to intensive peri-urban dairying.
All that said, there are drawbacks. Perhaps the main one is that BNH are typically sterile, not producing seeds, and therefore have to be propagated vegetatively, through stem cuttings or root splits. This means farmers depend on planting material supply chains that are often weak or informal. Diseases can also be transmitted more easily via vegetative material. Plus high biomass production demands big nutrient inputs, particularly nitrogen, with inadequate fertilization quickly eroding both yield and quality. That can be expensive.
In response, an important recent line of research has focused on developing seed-propagated BNH. Seeds simplify dissemination, reduce transport costs, and mitigate the spread of vegetatively transmitted diseases. They also enable more formal seed sector engagement, including developing new varieties.
Making fertile hybrids is technically tricky. Sterility in the classic hybrids is due to genomic incompatibilities between the parental species, basically their different ploidies, or numbers of chromosomes. So breeding strategies have explored chromosome doubling, intermediate ploidy levels, and backcrossing to restore partial fertility while retaining the desirable forage traits.
This has been reasonably successful, but trade-offs remain: some seed-propagated lines show lower biomass yields or less persistence compared to established clonal hybrids, and ensuring consistent performance across environments is still a work in progress. So it’s good to see ICRISAT and its partner still on the case, hard at work.
ICRISAT is advancing forage research with trials on seed-propagated Bajra–Napier Hybrids (BNHs) a shift from traditional stem cuttings to a scalable, seed-based approach.
Bajra–Napier, a key fodder crop for dairy systems, combines the high yield and perennial nature of Napier… pic.twitter.com/kKBQ58IuIn
— ICRISAT (@ICRISAT) April 20, 2026
LATER: Dr Chris Jones, program leader for feed and forage development at ILRI, who should know, tells me that the currently accepted names for the parents of BNH are Cenchrus americanus (pearl millet) and C. purpureus (Napier grass). Something to do with Cenchrus being nested within Pennisetum evolutionarily speaking, so the best bet was to merge the genera, but under the name Cenchrus because that is the older one.