Monarch butterfly | Credit: iStock

The Trouble With Roundup

If there were a contest to name the most-hated chemical, Roundup, Monsanto’s brand name for the chemical glyphosate, would have a very good chance of winning. It has been identified not only as a plant poison, but also a destroyer of monarch butterflies, an impoverisher of developing-world farmers, and a probable carcinogen. These are serious issues. But they are also complex, and I will argue, they are not settled (except for being a plant poison). On the other hand, there is a different, large, and growing problem in Roundup-dependent agriculture.

The Butterfly Effect

One of the environmentally damaging effects ascribed to Roundup is the decline of monarch butterflies, as described on a huge number of internet sites. There is indeed a connection between Roundup and the beautiful monarch butterfly, but it’s not as described on most of those sites.

North American monarch butterflies are an incredible species. The annual life cycle of those that migrate (many do not) goes like this. In the winter, they are in an altered metabolic state, called diapause, which is somewhat similar to hibernation. They are gathered in massive clusters of tens of millions in high mountainous regions of Mexico. The total number of butterflies in Mexico in the winter (a smaller number spend the winter in southern California) is in the hundreds of millions. These clusters cover every available surface.

butterfly-tree
Wintering monarch butterflies on a pine tree in Mexico.

Early in Spring, they awaken in response to environmental signals and begin a journey of several months’ duration, thousands of kilometers northward, reaching ultimately into Canada. But the overwintering butterflies will not reach the end of that migration. In March and April they mate and lay eggs on milkweed plants in the southern and central United States. Then they die. The newly hatched caterpillars eat the milkweed, and undergo transformation within a chrysalis, to emerge as the beautiful monarch we see in the garden. They are the first generation of the new year. This generation continues the northward migration, but it needs to act quickly; it will live for only a few weeks, and during this time it must produce the next generation. The eggs laid will hatch in May and June to produce the year’s second generation. These insects repeat the process to produce the third, which begets the fourth. The first, second, and third generations live for only a few weeks or months. But the fourth, miraculously, lives long enough to migrate to Mexico, overwinter, and begin the cycle again the next Spring.

Perhaps the most remarkable thing about the life cycle of the migrating monarch butterfly is that it is the great-great-grandchildren who are returning to the overwintering sites. How do they know where to go? Some scientists think there is a sensory organ that can read the earth’s magnetic field, and allow navigation by that. But our understanding is limited. The monarchs are born with a genetically-determined ability to migrate.

There’s a worrying problem with these migrating monarch butterflies: they seem to be disappearing. It’s difficult to count butterflies directly, but estimates have been made based on the area their winter roosts occupy. Between 1996 and 2013, the estimated number of monarchs overwintering decreased by as much as 90%. The winter of 2015-16 did see a three-fold surge in that number (1), but then the next year, there were 30% fewer, perhaps due to a storm in the overwintering area.

What makes these fluctuations even more puzzling is that they are not reflected in the numbers of butterflies seen during the summer further north. But between mid-summer and the time of arrival in Mexico, the number plummets. Some people question the existence of a problem in monarch butterfly numbers. But the list of experts who consider the monarch a threatened species includes the eminent scientist Lincoln Brower, who has spent a lifetime studying them (there’s an interview with him on the topic here).

A scientific analysis of the summer populations sounds a warning note (2). Researchers have found that in the American midwest, there are far fewer milkweeds in farmers’ fields than before, and correspondingly fewer butterfly eggs. The combination of fewer milkweed plants and fewer eggs per plant leads them to an estimation of an 81% decline in monarch production in the Midwest from 1999 to 2010, which happens to correspond to the reduction in the number of overwintering butterflies in Mexico during that time.

The most commonly discussed threats to monarch butterflies are degradation of the overwintering grounds in Mexico (illegal deforestation), climate change (premature warming throws them off their migration cycle), and the use of Roundup on American farms. The last of these is linked to the development of Roundup-resistant crops such as corn and soybeans. Such genetically-engineered (GMO) crops are sprayed with Roundup to kill weeds, including milkweed, to reduce the cost of production, and perhaps yield.

The linkage from GMO plants (many developed and sold by Monsanto), the use of Roundup (also initially a Monsanto product, but off-patent since 2000, and now provided by other companies as well) and the reduction of migrating monarch butterfly numbers is less compelling than its advocates suggest. In the first place, the reduction in milkweed is also a result of a reduction in “waste” agricultural land — ditches and margins that have been turned into cropland, largely eliminating milkweed production. In the second place, observers have not noticed a significant reduction of butterflies over the summer months (3). “Citizen scientists” who go into the fields and try to estimate butterfly populations agree that there isn’t a huge decline in the breeding pairs. The puzzle is, why are so few reaching Mexico in the Fall? The story is not finished, but Roundup is clearly not yet implicated as a major destructive force on the monarch butterfly population.

Roundup, Monsanto and cancer

The three words of this heading are the perfect trifecta for many critics of modern agriculture. The linkage between them (i.e. Monsanto makes Roundup, Roundup causes cancer) was immeasurably strengthened by a report in 2015 from the International Agency for Research on Cancer (IARC), a semi-autonomous arm of the World Health Organization. That study placed glyphosate, the generic name of Roundup, into the category of “probable human carcinogen”, as reported in Lancet Oncology and just about every major newspaper in the western world. It suggested that farmers, applicators, casual users (like me) and handlers were now at significant increased risk of getting cancer.

But there is a huge shadow over the IARC conclusion, and on the agency itself. Mother Jones, no friend of industrial agriculture and environmental pollution, distributed a report in June of this year that described contradictory evidence on the glyphosate-cancer linkage, and the rather weird circumstances under which that conclusion arose. The Mother Jones article was based on an investigative report by the Reuters news agency, which thoroughly examined the glyphosate-cancer link claimed by IARC (the Reuters report is online here). The details of this investigation are interesting, and revealing.

The problem with the IARC report

The mandate of IARC is to examine the literature about the possible linkage of agents and activities to cancer. It tries to determine “hazard”, any possible connection, whether supported by data on humans or laboratory animals. It does not propose “risk”, that is, the likelihood that people will get cancer (which depends on hazard and exposure time, and, most importantly, dose). In reviewing the world literature on glyphosate, it found some data that suggested a possible cancer risk in laboratory animals, but very little human data, and none that pointed to a human risk.

iarc-headq
Imposing headquarters of IARC in Lyon, France.

One of the few human studies was a 2005 research paper that looked for a glyphosate-cancer link in herbicide applicators, men (mostly) who mix and apply glyphosate in agriculture (4). If anyone was at risk, it should be them. The study included over 57,000 subjects and was carried out prospectively (people were followed from the beginning, rather than being asked to remember what they had done afterwards; this is considered to be the better way to collect epidemiological data). It found no association with overall cancer rates. This does not completely settle the issue. For example, the incidence of cancers follows exposure to risk in time, so longer follow-up may be necessary (for example, smoking and lung cancer incidence are separated by 30 years on average). The time dependence question is partially answered by studying a large number of subjects; if there’s any risk, it should be reflected in at least some cancers by, say, 10 years. Another possibility could be that some rare cancers are in fact increased, but because they are rare this would not affect the overall cancer rate in a statistically-significant way. More study needed.

That’s where it gets bizarre. In its 2015 deliberations, IARC did not take into consideration a follow-up to the 2005 results which concluded even more strongly that there was no link between glyphosate and human cancer (there were over 89,000 subjects in this study). The reason given was that IARC only considers published, peer-reviewed data, and the new data hadn’t been published yet. Fair enough. But the head of the panel, one Aaron Blair, was a lead author on that research report, and knew that it hadn’t been published only because it was part of such a large data set that not everything could be included in one report. The world’s experts on any potential glyphosate-cancer link were on the IARC panel, and could easily have reviewed the quality of the data. But they didn’t.

The existence of the new data came to light when Reuters looked at court documents that were part of a suit, filed by hundreds of Americans who claimed that glyphosate had given them, or a loved one, a rare cancer called non-Hodgkin’s lymphoma (NHL). They based their claim on the 2015 IARC report. In the disclosed documents were the data from the study that Blair had participated in. When questioned by lawyers, Blair agreed that the new data showed no link between glyphosate and NHL. He also agreed that if the new data had been available, IARC’s decision would probably have been different. Actually, the data were available; they resulted from research Blair himself had been instrumental in carrying out.

Meanwhile, other scientists and agencies concerned with biosafety were metaphorically scratching their heads, puzzled as to why the new data had not been submitted for publication before the 2015 IARC analysis. They now have been submitted, and experts who have seen them think they are solid; but IARC says it has no plans to revise its conclusion of “probable human carcinogen” for glyphosate. (It’s worth noting that IARC almost never concludes that a chemical it’s reviewing isn’t a human carcinogen. However, they have reversed an earlier decision about coffee, moving it into the “probably not carcinogenic for humans” column.) The U.S. Environmental Protection Agency, the European Food Safety Authority, Health Canada’s Pest Management Regulatory Agency, New Zealand’s Environmental Protection Authority and Japan’s Food Safety Commission have all reviewed the safety of glyphosate regularly, and they all say that it is unlikely to cause cancer in humans. In 2016, a meeting of the Food and Agriculture Organization (of the UN) concluded that glyphosate is unlikely to pose a carcinogenic risk to humans. A raft of international experts has said essentially the same thing. (Incidentally, none of the sources quoted here have any connection to Monsanto.)

Roundup’s existential problem

Roundup may yet have a lot to answer for in the decline of the migrating monarch butterfly population, although farming practices that insist on maximizing the portion of the land under cultivation must bear at least some of the blame. And glyphosate may yet turn out to be carcinogenic for humans, although the available evidence strongly suggests it isn’t. But there is a fundamental problem for the continued use, and usefulness, of glyphosate/Roundup: before long, there will be so much resistance to it in the world of weeds that it no longer has any beneficial effect.

A genetic mutation is responsible for the glyphosate resistance of GMO plants genetically engineered for it. That gene came, not from a plant, but from bacteria selected for resistance. Bacteria are like plants in that they need to synthesize their so-called aromatic amino acids (tyrosine, phenylalanine and tryptophan). Humans get these amino acids from the protein they eat (ultimately, from plants). Glyphosate inhibits just one enzyme on the pathway for aromatic amino acid synthesis, EPSP synthase. But that inhibition stops the growth of any plant, and it will die, which is why Roundup is such an effective weed killer. Bacteria with the right mutation in EPSP synthase can grow in the presence of Roundup, and that mutated gene was engineered into a variety of food plants back in the 1990s. This made these cultivars resistant to Roundup, and Monsanto started making a lot of money from them. The fact that the target of glyphosate is an enzyme that is not present in animals makes it a very specific growth inhibitor.

But evolution never rests. Genetic variations arise all the time. Glyphosate-resistant “somaclonal variations” in maize (corn) were discovered in 1995, and similar mutations exist in cotton. No genetic engineering was required; these variations in the EPSP synthase enzyme occurred spontaneously. It was inevitable that some weeds, somewhere, would develop the same mutation, and this has, indeed, happened. Such weeds are not affected by glyphosate and have been selected for by glyphosate in their environment.

Glyphosate-resistant “superweeds” are becoming a bigger problem all the time. They have led to the use of increasing levels of glyphosate (250 million pounds of Roundup were sprayed on crops in the USA in 2012). Glyphosate-resistant weeds were essentially unknown before the introduction of Roundup-resistant crops in 1996, but now infect somewhere around 50 million hectares (nearly 125 million acres) in the United States alone. If anti-GMO ideology guides some people’s resistance to Roundup, it is this practical outcome which provides an objective limit to its use.

field-waterhemp
Waterhemp, one of the most common glyphosate-resistant weeds, growing in a field of soy.

Like the overuse of antibiotics, which has led to drug resistant pathogenic bacteria becoming a major health hazard, overuse of glyphosate has led us to the present condition. Roundup may have only a limited lifetime of usefulness, before careful, organic methods of farming will be the only way to continue.

Work cited

  1. Grant, B. “Migrating Monarch Numbers Rebound” The Scientist, March 1, 2016.
  2. Pleasants, J. M. and K. S. Oberhauser “Milkweed Loss in Agricultural Fields Because of Herbicide Use: Effect on the Monarch Butterfly Population” Insect Conservation and Diversity (2012) 6(2): 135-144
  3. Ries, L., D. J. Taron, and E. Rendón-Salinas “The Disconnect Between Summer and Winter Monarch Trends for the Eastern Migratory Population: Possible Links to Differing Drivers” Ann. Entomol. Soc. America (2015) 108(5): 691-699.
  4. DeRoos, A. J. et al., “Cancer Incidence Among Glyphosate-Exposed Pesticide Applicators in the Agricultural Health Study” Environ. Health Perspect. (2005) 113(1): 49-54.

Back to the front page

Leave a Reply

Your email address will not be published. Required fields are marked *