Making the improbable happen

I was recently reminded of a post I wrote in 2006, and thought to myself that it could have been written yesterday. The trigger for this memory was a long piece in the New York Times that appeared more or less yesterday.

The subject is superweeds; that is, weeds resistant to one or more herbicides. In 2006, I was writing about waterhemp (Amaranthus tuberculatus), which had recently been shown to have developed resistance to an entire class of weedkillers. The thrust of it was that resistance involved mutations in two separate genes, taken together a one in a billion billion (1,000,000,000,000,000,000) chance. But it happened. Agriculture selects improbable events.

The NYT article was about another amaranth, Palmer amaranth (A. palmeri). The thrust of it was that Palmer amaranth is now resistant to at least six different classes of herbicide.¹ In 2006 it was resistant only to glyphosate. That leaves farmers with almost no options to control Palmer amaranth, and control it they must.

In 2008, researchers scattered 20,000 seeds of glyphosate-resistant Palmer amaranth into a 1 metre diameter circle in the middle of four different cotton fields that had no history of Palmer amaranth. The experimental sowing was intended to represent survival and maturity in the field of a single resistant female Palmer amaranth plant.² After that, they managed the glyphosate-resistant cotton as per recommendations, spraying to control the weeds.

One year later, in one of the fields, Palmer amaranth had moved 114 m from the original site in one of the fields. Two years later, the weed had spread to the boundaries of all the fields and covered 20% of the field area. “Three years after the introduction (2010), Palmer amaranth infested 95 to 100% of the area in all fields, resulting in complete crop loss since it was impossible to harvest the crop.”

That paper goes on to discuss some possible management options, suggesting a “zero-tolerance threshold” to eradicate every weed. I doubt anyone even tried. For its part Monsanto, which had developed glyphosate-resistant seeds, worked to stack resistance to another weedkiller — dicamba — into its genetically engineered seeds, a pointless exercise. As the NYT reports, “The agribusiness giant took a decade to develop that product line. The weeds caught up in five years.”

A squandered resource

The evolution of resistance to some life-threatening challenge is axiomatic in biology and it doesn’t matter whether the threat is an antibiotic, a herbicide, fungicide or insecticide, or even a predator. Anything that gives an organism even the slightest competitive edge in its ability to reproduce will in the end be selected. The problem is certainly not unique to genetically modified organisms. In the mid 1990s, wild oats resistant to three and four classes of weedkiller appeared on the Canadian prairies. Agriculture Canada blamed farmers who ignored advice to rotate crops and herbicides.³ But genetic engineering has exacerbated the problem many times over by giving natural selection so many more opportunities to do its inexorable thing.

Antimicrobial resistance is, belatedly, gaining a little recognition. Herbicide resistance might just be heading in the same direction, if the New York Times is taking an interest. These problems are, to some extent, a manifestation of a mismanaged commons; to begin with, using the stuff confers a benefit on the individual, but as everyone does so, everyone begins to suffer. In agriculture, they’re also a reflection on efficiency at any cost. Good weed control means good preparation of the soilbed, physical weeding three or four times during the life of the crop and other practices that take time and, therefor, money. How much easier to pay over the odds for seeds, buy weedkillers from the people who lent you the seeds, spray and pray. The same goes for antibiotics as growth promotors. Shave a fraction of a percent off feed costs and multiply that across millions of animals and you create a powerful incentive to abuse antibiotics. And just as it isn’t genetic engineering per se that creates problems of resistance, it isn’t agriculture per se that makes the improbable certain. It is the way agriculture is conducted.

What to do

Scientists have offered lots of advice on how to minimise the problems of resistance, some of which have even made it into policy. Just recently, the Food and Drug Administration tightened up a bit on its Guidance for Industry on the use of antibiotics on farms. From June 2023, a veterinary prescription will be needed for all antibiotics. I’m sure no veterinarians will be tempted to issue prescriptions that might not be absolutely necessary.

As it happens, organic farmers don’t use chemical herbicides on their crops, and superweeds are much less prevalent on organic farms. An excellent long article in Civil Eats uses one organic farmer’s fight against glyphosate-resistant giant ragweed as a jumping off point for a thorough discussion of the difficulties of controlling weeds in the oversimplified farm systems of the US corn belt. The key is hybrid rye, which overwinters and helps to smother ragweed as it emerges in the spring. Control can be achieved, but as the article makes clear, there are lots of obstacles, not least the difficulty of finding uses for hybrid rye. Longer rotations, increased diversity, more hands-on management, local cooperation: all are necessary, and all go against the grain for the majority of today’s corn belt farmers. Nevertheless, something is going to have to happen, and another technical fix is unlikely to be any kind of solution.

If you can’t beat them …

There is an additional approach, at least for Palmer amaranth. Its leaves, stems and, especially, seeds are edible, delicious, and highly nutritious (as they are for many amaranths). In theory, one could, perhaps, turn the noxious weed into a nutritious addition to the diet. And even though it is glyphosate resistant (not to mention the other five weedkiller classes), it come by that resistance natural, rather than through meddlesome genetic engineering, so there couldn’t possibly be any objection to eating it.