by Jordan Wilkerson
figures by Brian Chow

Summary: In the history of agriculture, no technology has been adopted so quickly and completely as genetically engineered crops. Particularly useful crops are ones that have an engineered resistance to herbicides. These crops have alluring benefits: reduced crop damage when herbicides are sprayed, easier weed management, and even the potential for environmental benefits. So what’s the problem? Herbicide-resistant weeds. The benefits gleaned from these crops begin to disappear as these superweeds gain prominence on farmlands across world. However, to fully appreciate the current predicament, it is necessary to understand what led to the difficult problem of superweeds. And it starts with the most common herbicide used in agriculture: Roundup.

Roundup and the crops that resist it

Chances are you’ve heard of Roundup, a common weed killer used by farmers and homeowners alike. Roundup is the brand name for the herbicide, glyphosate. Glyphosate works by preventing plants from being able to make the proteins they need to survive. Since virtually all plants make these essential proteins the same way, glyphosate affects nearly all plants. For this reason, glyphosate is deemed a “broad-spectrum” herbicide [1].

Of course, this is a double-edged sword. While Roundup is a great weed killer, its broad-spectrum effects make it a decent crop killer, too. That’s no good. If our crops aren’t resistant to Roundup, can we figure out a way to make them resistant? In fact, the agriculture company Monsanto did just that.

In 1996, Monsanto introduced the Roundup Ready soybean, a genetically engineered crop resistant to glyphosate. In the few years after, Roundup Ready cotton, maize, and various other crops also made their debut. While almost all plants are susceptible to glyphosate’s grip, the beautiful thing about genetic engineering is that the genes need not be from similar organisms [2]. In this case, the gene allowing resistance to glyphosate was taken from a type of bacteria called Agrobacteria [3]. With the introduction of this organism’s relevant gene into the desired plant genetics, a Roundup Ready crop is born (Figure 1).

Figure 1. Roundup contains glyphosate, which is toxic to standard plants that contain its target protein. The plants containing this protein are destroyed upon exposure to glyphosate, indicated by the red X. Roundup Ready crops have been engineered to contain a gene from Agrobacteria, making them immune to the herbicide.

While companies are developing these brand new plants, one question that naturally arises is this: do Roundup Ready crops directly affect the environment? Based on the large amount of scientific evidence so far, the short answer is no. There are two ways in which scientists and the public speculated these crops could have an effect on their surrounding environment [4].

One is the possibility that Roundup Ready crops may crossbreed with weeds to produce glyphosate-resistant weeds. This is slightly possible for some crops such as oilseed rape and sugar beet, but in most cases, the weeds present are not sexually compatible with the crops present, so crossbreeding cannot occur [2].

The other possibility is that dead, Roundup Ready plant matter could be different from normal plant matter. This could be a problem if this difference results in plant tissue that is toxic to organisms in the surrounding ecosystem. However, evidence strongly suggests that the two plant types have very similar composition, and their fast breakdown renders the final products that enter the environment basically identical [2].

A win for farmers and the environment

Assuming that farmers only use Roundup at the recommended rates in their weed management techniques, results show that farming practices associated with Roundup Ready crops actually have a lower environmental impact.

One reason for this is that they are able to reduce their pesticide use and replace more harmful pesticides with glyphosate [1]. The harmfulness of a pesticide is assessed by exposing organisms to higher and higher concentrations of the pesticide until it appears to inhibit their ability to survive. Glyphosate is considered less harmful because much higher concentrations are required to hurt aquatic animals and plants than with other herbicides [5]. When herbicides are considered relatively non-toxic to ecosystems, this means that the concentrations that could enter the environment from the recommended application of the herbicide is usually lower than the minimum concentration for that chemical to be toxic.

Another reason Roundup Ready crops theoretically result in a new environmental benefit is that farmers no longer have to till their cropland. While tilling has other benefits, its primary purpose is to mechanically destroy weeds present on the land before the crops are planted. However, this comes with a price. Tilling the land loosens the soil, which causes more of it to run off into nearby water bodies when there’s rain or even a strong wind. Soil on a farm plot is different from natural soil in that it contains much higher concentrations of fertilizer and residual pesticides [6].

With Roundup Ready crops, farmers can spray glyphosate on their land instead of plowing their entire area to get rid of the weeds. Furthermore, since their crops are resistant to glyphosate, farmers can replace outdated, more harmful herbicides with glyphosate, which breaks down quickly and is not very toxic to aquatic organisms [6]. Thus, runoff is reduced overall, and the small amount that may occur has less harmful effects.

Possible concerns about Roundup Ready crops

These benefits only occur if spraying Roundup is the only weed management technique required. As early as the 1960s, scientists have known that the tactic of only spraying one pesticide has strong potential to push weeds to develop resistance to it. Resistance to an herbicide occurs when one weed has a mutation that allows it to survive a Roundup spraying. If this resistant weed reproduces, it can pass its mutation to its offspring. Soon, a cluster of superweeds can quickly spread across acres of cropland, rendering Roundup largely ineffective [7].

One strategy that would reduce the probability of weeds developing resistance is to periodically use another herbicide that kills weeds through a different route. Therefore, resistant weeds would have to establish two mutations to survive the herbicidal spray [6,7]. However, this strategy would periodically eliminate the Roundup Ready crop, too, making it not much better than using non-resistant crops.

Another approach that is even more environmentally friendly is to change what crop is grown on the land each year. Different crops are planted at different times, deplete different nutrients, and root into the soil differently. This makes the environment much less stable and therefore less habitable for weeds to prosper [8]. However, this technique requires much more work of farmers.

Soon after the introduction of Roundup Ready crops, several research articles were published concluding that there is no benefit to using a portfolio of weed management strategies such as crop rotations and herbicide rotations mentioned above, one of which was cited by Monsanto in an advertorial they published [9]. Therefore, most farmers believed that employing much less convenient weed management just seemed unnecessary.

Rise of superweeds, return to old farming practices

Unfortunately, these studies had one major problem: the croplands they created for their experiments were rather small [1]. Industrial croplands, on the other hand, can be on the order of a million acres. Therefore, even though developing resistance to Roundup is not probable for any individual plant, there are a massive number of weeds growing that have the potential to resist its poison: this significantly increases the probability of at least some Roundup resistant weeds developing [7,8].

We do not have to go into detail about probabilities to assess whether superweeds will form – we already have confirmation that they have. Twenty-four cases of glyphosate-resistant weeds have been reported around the world, 14 of which are in the United States [7]. Farmers are now back to tilling their farmlands and spraying more toxic herbicides in addition to Roundup in an attempt to control the superweeds spreading across their farmlands [8].

Additionally, because many of the superweeds can still be killed by glyphosate if it is sprayed in higher doses, many farmers are spraying more glyphosate and other herbicides to combat the weeds. The attraction is that this is much less labor intensive than plowing and handpicking weeds out of the soil [9]. Consequently, a report drawing from US Department of Agriculture data on pesticide use estimated that an additional 383 million pounds of herbicides have been used than if Roundup Ready crops were never introduced [6]. This increased use of glyphosate heightens the likelihood of higher concentrations of the chemical running off into nearby ecosystems. At these elevated concentrations, glyphosate may be capable of causing environmental damage.

Furthermore, the practices of tilling and increased herbicide use are similar to what they were 20 years ago (with even more reported herbicide use). This is both an environmental problem and a financial problem for farmers who must now revert to spending more on herbicides and labor costs to till the land [10].

While Roundup Ready crops themselves have not caused environmental damage, they are certainly responsible for the Roundup-intensive weed management practices that have accompanied them. The environmental benefits – reduced tilling and reduced use of more toxic herbicides – are fading because the weeds Roundup was supposed to control have sprung up in revolt.

More complex weed management strategies than the ones mentioned here can be employed to help reverse this, and they’re already being used by some farmers [6]. These and other techniques mentioned previously may be worthwhile for realizing the benefits of adopting Roundup Ready technology on a larger scale. However, it is up to farmers to decide whether to invest in complex weed management practices. They are less convenient, but can help reduce risk of resistance in the future. This was a hard decision when Roundup was so effective. However, the advent of glyphosate-resistant weeds has shown that just spraying Roundup is not sufficient – not for farmers and not for our environment.

Jordan Wilkerson is a third year graduate student in the Department of Chemistry at Harvard University.

This article is part of the August 2015 Special Edition, Genetically Modified Organisms and Our Food.


  1. Devos, Y., et al., 2008. Environmental impact of herbicide regimes used with genetically modified herbicide-resistant maize. Transgenic Res. 17:1059-1077.
  2. Dale, P., et al., 2002. Potential for the environmental impact of transgenic crops. Nature Biotechnology 20: 567-574.
  3. Padgette S. R. , Kolacz K. H. , Delannay X. , Re D. B. , LaVallee B. J. , Tinius C. N. , Rhodes W. K. , Otero Y. I. , Barry G. F. , Eichholz D. A. , et al. (1995) Crop Sci. 35:1451–1461.
  4. Ackerman, J. Food: How Altered? National Geographic.
  5. Herbicides. United States Environmental Protection Agency.
  6. Eight Ways Monsanto Fails and Sustainable Agriculture. Union of Concerned Scientists.
  7. Selection Pressure, Shifting Populations, and Herbicide Resistance and Tolerance. University of California, Agriculture and Natural Resources.
  8. Boerboom, C and Owen, M. Facts about Glyphosphate Resistant Weeds. The Glyphosphate, Weeds, and Crops Series.
  9. Hartzler, B. Weed Science. (December 2004).
  10. Neuman, W. and Pollack, A. Farmers Cope with Herbicide Resistant Weeds. New York Times. (May 2010)

Note that a previous version of this article stated that the estimate of 383 million pounds of increased pesticide use came directly from the USDA. This has been corrected.

98 thoughts on “Why Roundup Ready Crops Have Lost their Allure

  1. You’ve done a good job of presenting this issue, but for one item. Glyphosate has reduced the environmental impact of crop production and GM crops are now the most sustainable form of crop production in existence. An article published in 2011 based on a large GM canola farmer survey in Canada revealed that the environmental impact of GM canola production was 53% lower in 2006 compared with 1995 when no GM canola was grown. The reference is:
    Smyth, S. J., M. Gusta, K. Belcher, P. W. B. Phillips and D. Castle. 2011. Changes in Herbicide Use Following the Adoption of HR Canola in Western Canada. Weed Technology 25: 3: 492-500.

    Organic crops require tillage to control weeds, which is incredibly environmentally damaging. In 2006, 67% of canola was produced using zero tillage, up from 11% in 1995. The high toxicity levels of chemicals that the organic industry are allowed to use in their crop production are multiple times higher than glyphosate which has an environmental impact quotient of 15.3.

    When combining the changes in tillage and chemical EIQ’s, GM crop production is the most sustainable means of producing crops.

    1. Nice article Jordan. I agree that more sophisticated levels of weed control management are required. What are these strategies? Different chemicals and crop rotation? The problems run so much deeper than switching chemicals and crops in any given year. How do you think that would work exactly (besides being “more work for farmers?” You are basically proposing a chemical merry go round. Would it not just be easier to create and grow more stacked GE crop varieties that you can alternate sprays on? I think you need to talk to more farmers in different areas of the country to get a better idea (if that’s what you are interested in) of what farming entails. While you present a fine case with some fine citations (although some of them are not the best).
      Also, the vast majority of farmers still disc or chisel (or both) their land every year. Farming no or low till is more common but requires specialized equipment. We can talk all we want about the environment, but round up for most farmers is convenient and easy to use and that is why they use it. They do not use it because it is better for the environment or to till their land less.

      1. Actually, many farmers do use glyphosate to reduce tillage. They use glyphosate to control weeds, not just for the enjoyment. You mention a “chemical merry go round”. Are you concerned about the tillage merry go round of organic farming. There are multiple trips per year destroying the soil structure.

          1. JC , no you prefer organic corn with worse pesticides then glyphosate, neither will effect you though since its only trace amounts.

          2. Roundup is applied at <1lb/acre. An acre is approx the size of a football field including endzones. If that is considered "bathed", I'd hate to consider your hygene habits.

          1. I actually farm 36,000 acres in southwest Kansas 15,000 acres in eastern Colorado and 12,000 acres in Washington state I actually quit using round up ready grains to the fact that I have noticed I have a hard time getting a good kill with round up like I used to. Round up is an absolutely safe fantastic product for no till Applications or for borders and or corners of circles pivot points field roads etc. The worst product that we have to use to control overgrowth which most people in large cities don’t understand you can’t just go out and pull every weed. That product is called Gramoxone/paraquat/agent orange since Vietnam they have added die and smell but we use this on a consistent basis because it burns down quick. The chemical that’s in Gramoxone has been used for thousands of years to poison people and or a suicide Potion! We had a field covered in grasshoppers one time Ariel Applied Gramoxone I went out there a week later and there was dead grasshoppers everywhere. When we handle this product on the farm we use Tyvek suits facemasks and respirators as well as rubber gloves look up the ingredients do your research! But due to the fact that the federal government and the liberal activists have pushed most Farms across the country to go no till versus old conventional farming Because of the dust pollution now you guys have to deal with the chemical pollution we put out in the air dust settles and hits the ground. because of the dust pollution now you guys have to deal with the chemical pollution we put out in the air dust settles and hits the ground. Just left wing liberal agenda tree huggers thinking they know everything have actually done more harm to this planet and good by not thinning out the forest Causing forest fires almost every year but the good thing is it’s created lots of jobs with DNR. Nobody gets it and nobody ever will tell there’s nothing left! The government pretty much pays us to no till it’s covered over 70% of our equipment costs allowing us to ramp up production by picking up more ground we have produced more jobs that way but again no till for the most part creates what Farmers call chemical foul and or Chem follow Rather than just summer foul which we kept weed free with a Rod Weeder Or a summer plow/sweep. Anyways hope this helps!

    2. The trouble with this whole issue is the use of the word “sustainable”. We can take the same data, and arrive at opposite conclusions, not because of the data itself, but what that data is indicating. For example, if less tillage was the only indicator of sustainability, then GM is the way to go. The same is true for less herbicide use, and so on. But to reach the conclusion that GM crops are more “sustainable” requires ignoring a whole host of other indicators; increased cancer rates, reduction of mineral content in food, etc.

      1. GM crops do not have increased rates of cancer and there is no reduction in nutrient content. Some particular GM crops specifically have higher nutrient content.

        1. Where are the proven facts upon which you are making a bold statement produce the facts & data that will prove your statements on cancer and nutrient content and we’ll see what the results are .

        2. roundup is cancerous. It ends up in our food. When 95%+ of our corn, wheat, and rice production is treated with roundup, there is no way to dodge the bullet. If you eat food, you now are forced to consume some amount of this cancerous roundup. Even applied in the best practice and even when food is washed thoroughly before processing, inevitably the herbicide will end up being consumed to a certain degree. I find this to be a sad description of ‘sustainability’.

          1. David,
            Its not proven even to even be a probable carcinogen and your exposure is very small from food. You seem extremely worried about this so i ask you, do you live in a city and use sealed gas mask when you go outside ? because particles from traffic is actually proven to increase risk of cancer. Do you fry food in a pan or eat chips ? The science for this being carcinogenic is much stronger than glyphosphate although actual risk is not established.

          2. People have lived thousands of years sustainably in America without herbicides and here we are after only a few hundred years of living amongst these invasive lazy slave drivin foreigners they want us to believe that we absolutely need herbicides…I say its not invasive plants we should worry about…Its invasive people by far who are our biggest threat…

    3. You neglect the environmental damage to soil critters, bacteria and fungal mycorrhiza by glyphosate. Interruption of the mutual relationships within soil communities is a recipe for disaster:

    4. I thought the point of organic crops and promotion of them is that the farmers are not allowed to use chemicals ? Organic crops all a con ?

    5. Woolworths South Africa has moved to near organic production for its fruit & veg for all 540 of its stores. Woolworths went on to win best supermarket chain in the world in 2014 for quality of produce, and their produce throwout rate reduced by 58%, saving them tens of millions due to the higher nutritional value in the produce, making it last longer on the shelf.

        1. Reading this thread 2 years later. I hope you have learned to chill out since then. I am not comforted by hearing that POPs on my food are there in “trace amounts”

  2. The 53% decrease in environmental impact describes the herbicides used, not the intrinsic environmental impact of the HR crops. As I state in my article, the HR crops themselves do not affect the environment any differently than conventional crops. Therefore, the assessment in the paper you cite is based on weed management practices. It is true that herbicides such as glyphosate are more benign than older herbicides, which I do state in the article. Therefore, this weed management practice is better for the environment as you stated in your comment (and as I stated in the article, actually). However, spraying these new herbicides alone strongly pressures the development of herbicide-resistant weeds. When these weeds dominate the cropland, the weed management practice described in the paper (just spraying the new herbicides) is no longer tenable, and the reduced environmental impact is largely lost because farmers must revert to tilling and intensifying their herbicide use, which includes the application of outdated herbicides. To be clear, my ultimate point is not that we should eschew HR crops. It is that we should consider adopting more complex weed management practices to reduce the frequency that HR weeds develop on croplands, so we can get closer to realizing the full potential of HR crops’ environmental superiority.

    1. Your view of weed control is over simplistic. Organic production relies heavily on tillage which increases erosion potential, requires greater amounts of fuel, causes compactions, . . .

  3. @Jordan Wilkerson

    This appears to be the best, most informative, and least biased article I’ve ever read on this subject. Thank you.

    @Stuart Smythe

    > … GM crops are now the most sustainable form of crop production in existence.

    Maybe it’s sustainability of topsoil with corporate farming you meant. That would make more sense. I don’t mean to put down what you said. It’s just when people read ‘sustainable’, certain things immediately come to mind.

    GM crops are typically patented. Patented crops are not sustainable without permission (if something requires permission and/or money to buy license to use it, that’s a lot of risk and overhead). Also, it sounds like many of these crops are sterile, which further degrades their sustainability (since they can’t be reproduced without their creator corporation, who for all we know, may not be around a few years from now, especially if there’s an apocalypse, a severe war, or something). Sustainability implies independence. GM crops, as they currently stand, imply reliance on whoever owns the intellectual property, and the (probably) expensive/proprietary means to make them. This isn’t an inherent trait of GM crops, though, but it is an inherent trait of how they’re being made and used today.

    Additionally, glyphosphate seems to suffer similar problems with sustainability.

    1. And nobody has mentioned anything about the dangers of GMO products consumed by other organisms including humans. Several studies have proven that if Monsanto, Pioneer, and other GMO grain seed providers had continued testing their grain feed for just a few more months on laboratory rats; they would’ve found high numbers of their test subjects had developed cancerous tumors.

  4. In the caption for Figure 1, I believe you have a mistake. Roundup is an herbicide, not an insecticide.

    1. Thanks! You’re correct–while the figure shows Roundup acting as an herbicide, it says insecticide–oops!

  5. By now the health impact of using glyphosate pre-harvest on nearly all the food staples of the USA is clearly visible. Please Youtube the lectures of Dr Stephanie Seneff.

    1. Stephanie Seneff has a PhD in computer science, not medicine, pathology, physiology, or any other biological science. She is unqualified fear-monger.

      1. She switched her focus and educated herself in their, just like we all could if we decided to . William Albrecht started off as a medical student but switched his major to AG when he decided he could do more to help more people in the world by promoting healthier farming practices and producing more nutritious food. Look him up. He has the answer to the NPK farming frailties and super insecticides. Soil minerals are pretty cheap but provide the basics that help plants resist insects. If I didn’t see it with my own eyes, I might not believe it, but we all have to test these theories for our own satisfaction.

        1. I am very familiar with the work of William Albrecht, but you obviously don’t. He promoted the use of cation exchange capacity and base saturation for making fertilizer recommendation. You should look him up and actually try to understand what he wrote. You should consider a very basic soil science class if you are going to chat about this topic.

      2. Larry, what is your educational background focused in? If you don’t have a PhD in computer science, I don’t think you are qualified to post comments on an internet forum such as this. Without a background in computer science, anything you post really has no validity.

        Makes sense, right???

        1. You make no sense. Seneffs MO ia to dredge through data and make correlations and then claim that “correlation proves causation”. This is an error that anyone with even an extremely basics stats class would be aware of.

          Does that make sense??? If not let me know and I will try to simplify the explanation even more.

        2. I have a PhD in agronomy which is the topic of this article. So, what I post is more scientifically valid than what a computer programmer writes about pesticides.

          Makes sense, Right???

  6. Can you direct me to peer reviewed studies comparing the effects on humans of synthetic pesticides like glyphosate and 2,4D to the natural pesticides used in organic farming? Can you direct me to studies showing the long term environmental impact of the above mentioned pesticides both natural and synthetic?

  7. This is a great explanation! As a health professional, I find myself reminded of the similar problem of antibiotic resistance in bacteria. It makes me wonder if the farming community will do better at combating this looming issues than the medical community has with antibiotic overuse.

    Or, as Ian Malcolm would say, Life finds a way… That sneaky b******.

  8. Lovely article. I only wish it had addressed the potential human health impacts of having so much glyphosate in our food supply. With so much corn, soy, wheat, canola oil, and sugar being eaten by Americans each day, exposure levels and their consequences deserve a closer look.

  9. Hi Jordan, in parts of this article you use the words ‘herbicide’ and ‘pesticide’ interchangeably. Was this intentional? I’m struggling to get my head around this rather vital concept, especially in the context of pharmacokinetics/dynamics and safety testing. The main question being is the mechanism for glyphosate’s destructive effect on plants the same mechanism that can cause problems in mammalian cells?

  10. Ok. This was a well-written calmly argued and well substantiated article. However, would like to make a couple general comments. One, the development and spread of a lot of herbicide resistant weeds, particularly pigweed (amaranth and waterhemp), in the US is a fact and has posed challenges for a number of farmers to the point that they are having difficulty managing them with chemicals. Two, in the long run having vast monocultures is not sustainable as they are inherently prone to being disrupted by various pathogens, insects, and undesired plants.

    1. Hey, KMS!

      That’s a good point. I cited Union of Concerned Scientists. However, the number, 383 million pounds, is directly from the USDA. In fact, it’s from the source you provide in your comment. This is the quote from the Preface of that GE13Years Report: “The most striking finding is that GE crops have been responsible for an increase of 383 million pounds of herbicide use in the U.S. over the first 13 years of commercial use of GE crops (1996-2008).”

      1. yes! I asked the author to answer that one–please see his response and reply if you have further questions.

        1. Yes, I received the author’s reply. Thank you so much. I am doing an extensive report on GM Foods for nutrition courses and doing as much research as I can.

          I’d like to get your thoughts about Ronald Bailey’s statement in his article, The Top 5 Lies About Biotech Crops, where he writes,
          “Benbrook [author of The First 13 Years] largely got his 2012 results by making some strategic extrapolations of herbicide use trends to make up for missing data from the U.S. Department of Agriculture. In fact, the USDA does not provide herbicide use data for corn in 2004, 2006, 2007, 2008, 2009, or 2011, for soybeans in any year after 2006, and for cotton in 2002, 2004, 2006, 2009, and 2011. (The USDA’s National Agricultural Statistics Service is expected to issue a report updating national herbicide and insecticide usage later this year.) Retrieved from:

          Thank you for all that you do!

          1. I don’t think it’s currently true that there is no herbicide use data for those years (not sure when that post was written)–there’s a report that goes up to 2008 here ( that claims: “By 2008, over 90 percent of soybean acres were planted with HT seeds (fig. 14). HT soybean production sharply boosted glyphosate use on soybeans from 0.17 pound per planted acre (a total of 11 million pounds applied) in 1996 to 1.26 pounds per planted acre (95 million pounds) in 2008. Pounds of all other herbicides applied to soybeans declined considerably from 1.02 pounds per planted acre in 1996 to 0.14 pound in 2008.”

            So glyphosate-resistant crops lowered the use of some (arguably more toxic herbicides), but increased the use of glyphosate. And then, as the article explains, the use of a single pesticide mixed with other less-than-ideal farming practices resulted in weeds that developed glyphosate resistance. Insecticide use also decreased with the introduction of Bt GE crops (as, I think, did total herbicide use). So the issue here is not total pesticide use, but rather increased use of a single pesticide–which on its own might be okay, as it is less toxic than many other herbicides–but it has unfortunately led to the development of resistant weeds.

  11. Also, did the resistant weeds (superweeds) get their resistance from incorporating the resistant gene from the the RoundUp ready crop into their own DNA? In other words, is this a case of gene transfer from RoundUp Ready plant to the average weed? If not, then superweeds could form on fields using conventional crops as long as pesticides are being used, correct? So then, it is my understanding that this is not a problem of genetically-engineered plants, it’s a problem of pesticide use. These distinctions need to be made very clear. Please correct my understanding if need be. Thank you.

    1. That’s a great question! Resistant weeds developed resistance themselves–they did not incorporate the RoundUp resistant gene. So yes, you’re correct that this is primarily a problem of pesticide use. However, RoundUp resistant crops make it easier to use lots of pesticide (specifically RoundUp). As the author suggests, with more thoughtful weed management RoundUp resistant crops could have some environmental benefit by reducing the need for tilling. Also, RoundUp is less toxic than other pesticides, which provides its own environmental and health benefits (

  12. Like KMS, I’m questioning the author’s assertion that USDA reported that RR technology resulted in “an additional 383 million pounds of herbicides have been used than if Roundup Ready crops were never introduced.” There are several things wrong with this:

    1) A poor source was chosen for this citation. Union of Concerned Scientists can’t be considered a high-quality, unbiased source.

    2) If this information was from a USDA publication, you need to cite the USDA publication. Your citation contains links for its information, but none of them lead back to a USDA report. I’ve followed your citation through 5 links and still have not come to the USDA report this claims to be from. In subsequent papers you write, you should use this as a good clue that you’ve found a biased or false source.

    3) Your paper seems very reliant on popular press articles, which we all know have a very low standard for review and veracity. You also rely heavily on lobbying or special interest groups for information. If you are to take science communication seriously, you need to use peer-reviewed sources for your information, not unreviewed special interest websites.

    The bio here states that you are a 3rd-yr graduate student at Harvard University.
    I would hope that a scholar who has reached your level at such an institution as Harvard would know a little more about science communication, or at the very least understand citation of sources better. As a member of the scientific community, I suggest that you correct the errors in your essay, or remove it so that it doesn’t become part of the “fake news” narrative.

    1. Did you see the author’s response to KMS? The 383 million number is from a report drawing from USDA data:

      Of course this is an estimate (we can’t strictly speaking know how much glyphosphate use would increase if GE crops weren’t introduced), but based on the report, it seems that this number is based on the increase in the use of glyphosphate. Ideally, this estimate should also somehow control for any increase in production, but that becomes kind of circular (since glyphosate resistant crops increase yield etc).

      1. Author here!

        I’d actually like to describe my citations a bit in response to your comment that the article is ‘very reliant on popular press articles.’ I should be clear that about a third of my sources are from peer-reviewed journals, and they are heavily cited in the article. But why not exclusively use journal articles as my sources? It’s a good point, and I’d like to address it. Journal articles are normally behind pay walls; readers therefore cannot learn more from these sources because they cannot access them. Although I read more journal articles than what I cited, I chose to include more open-access sources instead to make it easier to learn more about the topic. The majority of these remaining sources are universities and government organizations. The only 2 sources I would consider ‘popular press’ are National Geographic and New York Times. They’re each only cited once.

        1. Thanks for replying to my concerns, Jordan. The problem with quoting popular press articles and special interest sources (like Union of Concerned Scientists) is that they have either a very low or no standard of review, so they can publish whatever claims they wish and have no mechanism to ensure that those claims are valid. As you can see, there’s a lot of trouble verifying the UCS claim of increased herbicide use. Although they made that claim with very little evidence, people expect that authors from Harvard would have a higher standard and would find the truth before publishing such figures.

          I can appreciate your concern that the common man may not have access to journal articles. However, the risk you take when you choose to cite secondary or tertiary sources, like special interest and popular press, instead of reputable primary sources is that if your secondary or tertiary sources contain false information, your essay will also contain false information. This is what has happened in your case.

          The more foundational problem, as far as your essay writing goes, is that you’re attributing a figure to a source that your didn’t cite. You read something in UCS’s opinion piece and you attributed the figure to USDA. We learned in high school composition class that when you claim that information comes from a particular source (USDA, for example), you need to cite a USDA publication. If you can’t find a USDA publication with that information, you may say that UCS claims that figure (and cite UCS), but you must also try to corroborate the information. If you can’t find corroboration, you then must either state in the essay that it can’t be corroborated, or you must leave it out. If UCS says that NASA tells them there’s monkeys on Mars, but you can’t find anything from NASA saying there’s monkeys on Mars, you can’t reprint what UCS wrote, claim it came from NASA, then cite UCS.

          1. Hey, John
            I appreciate that. That specific sentence has been modified to more precisely describe that number’s origin, which is based on an analysis performed on USDA data. To be clear, though, UCS is a nonprofit science advocacy organization. The current chair of its board of directors is a distinguished environmental science professor at Dartmouth. Consequently, their reports are not considered merely opinion pieces. Maybe we should step back a bit, though. The important question is presumably this: is it true that herbicide use has increased due to glyphosate tolerant crops on the order of hundreds of million pounds? I’m providing an additional link that corroborates this information (this time, a peer-reviewed article):

            The results are provided in increased herbicide use per hectare. You can couple these findings with the USDA data on soybean and maize acreage to get an estimate on total change in herbicide use:

            With this article and USDA acreage data, you get an estimate comparable to the UCS report – on the order of hundreds of million pounds of herbicides since these GT crops were introduced. As a reminder, glyphosate is the herbicide that primarily composes this increase, and glyphosate is considered less toxic than most other herbicides (as I state in the article). This dampens the deleterious effects of increased herbicide use and is why some reports (but not all) find the environmental impact of this increased herbicide use is still lower than if conventional crops had been used instead.

          2. Again, we come back to two main themes:

            1) You cited source that didn’t support the words you wrote.
            If you’re going to communicate science and cite sources, you need to cite sources that actually support the ideas you’re writing. It really looks like you wrote your own ideas independent of the scientific body of knowledge, then cited the first or simplest thing that came up in a Google search. The best way to ensure that you get the topic right is to read the existing literature first, then compose your essay as a synthesis of the information you read. It’s obvious that wasn’t done here. If it were, you would have cited in your essay the 3 papers you mention here, instead of a biased webpage that doesn’t even contain the information that you were writing about.

            2) You’re saying different things in the comment section than the additional sources you cite are saying. None of the papers you cited here define a “natural soil,” but you claim to know what a “natural soil” is, claim that they all are the same, and claim to know the nutrient status of all of the them. These papers *do* say that land used to produce specific crops can have various characteristics, but they don’t make the judgment that you made.

            You also make a couple of subjective statements just in this comment that aren’t founded;

            1) Nitrous oxide is not solely an anthropogenic phenomenon that only occurs in agricultural production. Nitrous oxide production is a NATURAL gas produced during the NATURAL nitrification process. Rain forests produce tremendous amounts of nitrous oxide during leaf litter decomposition. Your comment here makes the lay person think that nitrous oxide is some synthetic byproduct of agricultural fertilizers, but that’s not true at all. It’s a natural process.

            2) What criteria are you using to determine that fertilizer application is “excess”? Remember that words have meaning. Scientific communication is different from opinion pieces because we need to follow specific definitions so that our meaning isn’t lost in varied interpretation. Using the words that you did, you’ve made the accusation that more fertilizer than necessary is used to produce all crops. If you’re going to make that accusation, you need to have some data to support that. You don’t have that here. You also don’t have data to support your assertion that agricultural soils have more residual pesticide in them than other soils. I might submit that there’s more residual pesticide in the soil around a single walnut tree (in the form of allelopathic secretions) than in most farm fields in the US.

            The bottom line here is that scientific communication needs to be based on facts, rather than your opinion.

          3. I think this conversation is getting kind of out of the scope of utility. Of the major conclusions of the article, which do you think are not based on fact, but rather on the author’s opinion?

            It feels a little bit like you’re picking small points to nit-pick here and essentially accusing the author of lying (as he said that he read more peer-reviewed literature when writing this article than what he cited here and you said he “obviously” didn’t do that). It’s kind of hard to have a civilized conversation about science with these kinds of accusations, as they just make everyone feel defensive and are thus not particularly helpful.

            Jordan is providing you with peer reviewed sources when you ask for them–if you disagree that the sources support his claims, that is a valid thing to bring up, but please stop with the accusations and personal attacks.

      2. I saw the author’s response. The problems are:

        1) The author didn’t cite that report in his essay. If that’s where the information comes from, it needs to be cited.

        2) The Organic Center essay is a biased report and doesn’t provide any data or references for its “383 million pound” claim. Wilkerson’s essay claims this to be a USDA-reported value, but neither his primary nor his secondary citation refer to the USDA value. If this value is from the USDA, an actual USDA report needs to be cited. If one can not be found, USDA can not be referenced. Wilkerson could claim that the value came from the Organic Center and reference his source, but he runs the risk of citing untrue information that came from a biased source.

        I hope the point that Wilkerson is taking from this is that information and sources need to be adequately vetted in scientific communication.

        1. We agree that it was misleading to attribute that figure directly to the USDA, and we have corrected that in the text. Thanks for pointing that out! Also, the original Organic Center report actually has a whole supplemental table about where its data came from.

          1. I understand from your comment above that you feel like I’m being “picky,” but it’s important for those in scientific communication to understand that their words have meaning. One perfect example is that labeling of fertilizer use as excessive. To say that in scientific communication, the author must have some data showing that more fertilizer is being used than is necessary to produce a particular crop yield. If that data doesn’t exist, this is no longer a scientific piece — instead, it’s an opinion piece.

            The problem with playing fast and loose with terms like that is that they can be easily misinterpreted. Just like Jordan misinterpreted the “383 million pound” claim from an opinion source that was passing it off as fact (it really has no basis in fact at all), others might be inclined to take the term “excess fertilizer” and create the opinion that US farmers use too much fertilizer, when that isn’t based in fact at all. When people see that in a Harvard paper, written by a Harvard grad student, and posted on a Harvard website, they assume that there’s data to back it up. Why would Harvard publish something without data to back it up? They’re at the highest standard of science and knowledge — they wouldn’t possibly write something without the data to back it up, right?

            So, it’s all really quite simple. In scientific communication, every word matters, we need to be very specific about what we say, and we need to have proper data for every assertion we make. Just like Jordan did, the public is going to read an article like this and take every word to be the solemn truth. As a science communicator, it’s his responsibility to make sure that every word he writes is impartial and backed by evidence. Words matter.

          2. I have no problem with you being “picky”–I have definetely sent corrections in or left comments about similarly small issues. I agree with you that “words have meaning,” and I mean, really, who doesn’t?

            The problem with your comments is that instead of directly responding to the data and peer reviewed sources that Jordan sends you, you seem to just want to lecture him about how to communicate and attack him personally instead of first actually understanding whether or not he is in fact communicating false information. Of course I can’t know what your motivations are, but the tone and content of your comments makes you come off as more invested in insulting Jordan than in actually coming to an agreement with him on what the science says. When he provides sources, for example, it would be helpful to identify SPECIFICALLY why you disagree that they support what he’s saying rather than immediately dismissing them and reminding Jordan again of the obvious truth that “words have meaning.”

            If your goal is for people to get the right idea about this, wouldn’t it be more helpful to convince Jordan that he’s actually saying something that is incorrect?

  13. Thought you might be interested in my three questions to the USDA and their reply:
    1) Was there ever a response from the USDA to the Organic Center’s, The First 13 Years report. If so, I would love to get the link to that.

    2) Has the USDA ever published a similar report, which looked at the pesticide use of conventional crops versus GM crops? Again, if you can provide links that would be great.

    3) In his article, “The Top 5 Lies About Biotech Crops” Ronald Bailey writes,
    “Benbrook [author of The First 13 Years] largely got his 2012 results by making some strategic extrapolations of herbicide use trends to make up for missing data from the U.S. Department of Agriculture. In fact, the USDA does not provide herbicide use data for corn in 2004, 2006, 2007, 2008, 2009, or 2011, for soybeans in any year after 2006, and for cotton in 2002, 2004, 2006, 2009, and 2011. (The USDA’s National Agricultural Statistics Service is expected to issue a report updating national herbicide and insecticide usage later this year.)” Retrieved from:

    Can the USDA confirm Bailey’s statements?

  14. USDA Reply:
    For the first question, see the USDA Fact Sheet at

    For the second question, the ERS report at addresses some of the issues raised by the AC21 report. Tables 4-5 of the report addresses that with respect to corn and soybeans. If you have further questions, please contact the authors of the report. You may also find use several other recent ERS reports on biotechnology at

    USDA surveys farm chemical use for different crops each year but highlights of the latest data are available at

  15. I don’t think the final sentence of the 9th paragraph is accurate (“Soil on a farm plot is different from natural soil in that it contains much higher concentrations of fertilizer and residual pesticides [6].”) That information isn’t mentioned anywhere in the cited source. Is there an unbiased, reputable source that supports your assertion that farm soil has much higher concentrations of fertilizer and residual pesticides than “natural soil.” What is a “natural soil”, anyhow? Forest soils are often have much higher levels of phosphorus and potassium (among other nutrients) than agricultural soils.

    I think this is where the author is inserting his own opinion and bias, instead of writing a fact-based essay. The scientific literature supports the statement made before this one (that less tillage results in less off-site transport of nutrients via soil runoff). But, the scientific literature doesn’t define “natural soil” and doesn’t make claims about nutrient status of different soils. Soils under agricultural production can’t be painted with the same brush. Even soils just a few feet apart on the same farm can have wildly different nutrient profiles. Painting soils not under agricultural production with the same broad brush is also misleading and incorrect — unless you have soil test data to prove that.

    The author should either change that passage to be correct and use proper sources, or he should strike it from this essay.

    1. Does Anyone else besides me feel that ‘John’ is expressing myopic troll-like behavior? Not seeing the forest for the trees.. To me his hyperfocus and jumping to yet another personal remonstrance -when the author and site manager have done yet more due diligence on yet more (and smaller) minutiae brought up by ‘John’- is disingenuous, built to discredit for it’s own sake -classic troll behavior, and not actually interested in any earnest desire to improve the article’s quality. Please sit down ‘John’, and, you know…. 👏😎

    2. John, please study (if you are actually interested in learning & not simply trying to disrupt) KMS’s contributions… His criticism remains constructive, and he takes the time to find answers to his own questions! Cheers.

  16. Hey, John

    Again, I cited an open access source, but the basic premise that agricultural practices alter soil’s natural composition is well-established. Here are 3 example peer-reviewed sources asserting that soil nitrogen content (which is from the fertilizers applied) on agricultural lands are higher than typical natural soils:
    Subbarao, Guntur & Ito, Osamu & Sahrawat, Kanwar & L Berry, W & Nakahara, Kazuhiko & Ishikawa, T & Watanabe, Takeshi & Suenaga, K & Rondon, Marco & Rao, Idupulapati. (2006). Scope and Strategies for Regulation of Nitrification in Agricultural Systems—Challenges and Opportunities. Critical Reviews in Plant Sciences.
    DOI: 10.1080/07352680600794232



    Also, most of nitrous oxide (the 3rd most influential anthropogenic greenhouse gas) emitted by humans is from agricultural soils. The EPA specifically states this is because of the loading of fertilizers onto these soils:

    So it’s not just that we happened to grow crops on soils with naturally higher nitrogen content. The higher nitrogen content is a consequence of humans adding excess fertilizer to the soil.

  17. I am going to be very Naive here. But surely messing around with crops and mutating them by genetically modifying them (even though there are superior benefits). Surely nature over time will adapt with that and create it’s on weed strain to combat. Or maybe the soil will change and you won’t be able to grow them.
    Again I am Naive on this subject but also feel mother nature always wins over time.

    1. Many crop varieties came from natural mutations which were desireable and propagated – seedless bananas.

      Some varieties (including organically grown ones) were developed using radiomutagenesis (a slegehammer approach). GMO are development is much more controlled.

  18. Hey, Jack!

    It sounds like your ultimate concern is humans creating organisms that wouldn’t otherwise exist in nature. Even though genetic engineering is new, humans have been creating unnatural organisms since Roman times. This is through the technique called selective breeding (also known as artificial selection). Domesticated cats and dogs are both examples of this.

    However, we’re concerned about crops, so let’s look at an example of that: corn. Corn cobs used to be ~10 times smaller than they are today. Their growth didn’t happen naturally. For a long time now (starting around 10,000 years ago), farmers found larger corn cobs more desirable, so they would collect their seeds and plant those for the next year. Over time, this effectively altered the genetic make-up of corn. Importantly, modern corn – even corn labeled organic – is not really a natural crop. The crop only exists because humans tinkered with corn breeding to their benefit over centuries.

    Genetic engineering is a much different technique than this. However, if your only concern is the consequence of creating organisms that don’t exist naturally, humans have a long history of doing that with our food.

    If you’re interested, here’s a more detailed discussion of the history of artificial selection of maize by University of Utah’s Genetic Science Learning Center:

    1. Love these questions. I have a couple…

      “For a long time now (starting around 10,000 years ago), farmers found larger corn cobs more desirable, so they would collect their seeds and plant those for the next year.” If the larger seeds were already produced naturally, how is simply planting them ‘not natural’ as your explanation details above?

      “However, we’re concerned about crops, so let’s look at an example of that: corn. Corn cobs used to be ~10 times smaller than they are today. Their growth didn’t happen naturally.” The above seems pretty natural. Unless I’m missing something.

      “Genetic engineering is a much different technique than this.” YES, in 20th and 21st c. capitalist America, and in an already delicate and compromised bio environment, it is indeed a VERY different thing . I’m extremely concerned about the environmental and soil microbiome consequences that have yet to show themselves.

      A finely written article. But I do agree with the man who is pointing out the somewhat indiscriminate word disqualifiers you used. I noticed them too, and I’m not even a scientist, but my objectivity meter is on high with controversial topics like these. Nonetheless, I’ve bookmarked this one!


  20. From the Washington Post in July of 2019: “Although the judge substantially cut the damages, she supported the jury’s finding that Monsanto was to blame for Alva and Alberta Pilliod’s cancer and that the company repeatedly tried to bury evidence about Roundup’s health risks at the expense of consumer safety.” ( )

    Wilkerson writes above, “While Roundup is a great weed killer, its broad-spectrum effects make it a decent crop killer, too.”

    And people-killer. We’re toying with plant genetics in order to make it easier for us to spread toxins far and wide. I mean, what could possibly go wrong?

    And I won’t even get into the issue of Monsanto’s making the plants sterile so that farmers have to buy seeds from them every year.

  21. I will add that this statement here, that “Roundup Ready crops themselves have not caused environmental damage” is not valid science or logic in that it is not provable. You cannot so confidently state or prove this kind of a negative assertion without access both to far more data and to much better data assimilation and analysis. How does the author know for sure that this is true? He doesn’t. So a more accurate statement would be, “I believe that Roundup Ready crops have not caused environmental damage.”

    Over the centuries, we have been confident time and again that our inventions would be safe, and time and again that has proven not to be true. From lead paint to DDT to margarine, the list is nearly endless. Just because the research is not (yet?) definitive certainly does not mean that there is no danger, which is why Europe employs the precautionary principle ( ).

    It is VERY hard statistically to analyze such a complex system — nature itself, or the WHOLE “environment” — because the scientific method requires an testable control group. That the effects are hard to measure does not mean that common sense suggests that nature prefers genetically-modified crops to crops that evolved to be in harmony with it, or at least that were bred via more typical means of reproduction. It’s one thing to encourage two breeds of dog to produce a preferred type of offspring, and another altogether to tweak genes to make their puppy glow in the dark.

    And again, this is common sense: genes that spring forth naturally are much more likely to be robust and tested in nature in the past, whereas brute-force-tweaks are much more likely to have unforeseen consequences. Even if it’s a random mutation, in nature it has a chance to start small and further genetically adjust before it’s scaled way up. Here, we just make a brute-force tweak and then mass-produce it and spread it around the world. That’s not in tune with how genetics is supposed to work, and eventually our hubris in this area will come back to haunt us.

    Our desire to scale up our inventions, often because of profit, has also cost us, and that’s very evident in the world of antibiotics, where similar resistance is neutralizing the efficacy of these once-much-more-valuable tools.

    The question isn’t, do I think that this one specific GMO product will cause harm. The bigger question is whether THE PROCESS that humanity is currently employing is likely to cause serious harm at some point. And common sense suggests the answer is yes.

    1. Thank you! So well articulated and aligns precisely with my thoughts on the topic.

      I invite you to share your viewpoint on other platforms like Instagram where you can find misinformation about glyphosate and GMOS by supposed scientists named, I’m not kidding, ‘Food Science Babe’ (eye roll) who runs that IG account playpen fast and furious pro GMO etc. She skirts the facts that don’t align with her pro GMO etc stance, and she will power play her title as scientist on her IG page by demanding a peer reviewed citation from anyone who comments there…even if you aren’t citing anything. She thinks doing so is her firewall, but she gets away with it as IG is low hanging fruit as far as audiences go; they won’t challenge her. It’s very disturbing, but there’s a small contingency who calls her out there…it needs fortification, though. ha

      I’m emboldened by your comment to continue my research. Have a good one!

  22. It is not clear to me if any of the people who have commented here are close to the soil or understand the difference between USDA certified Organic and true sustainable practices in harmony with Nature which was simply known as organic farming prior to the federal government commandeering the word organic.
    Find a thorough, documented analysis of the subject of the original article at the following link:

  23. This is a very unusual article for this website . . . it is fairly logical and scinetific. Although the ridiculus use of “Super Weed” puts it back in the propaganda category.

  24. Your blog has provided useful information for the work. All the tips in the post are great. Thanks for sharing. keep blogging~!

  25. I’m a contractor by trade and many of my neighbors and clients use round up and swear by it to maintain their properties and others wouldn’t touch it. Many of them end up asking me to pave patios over their yards so they do not have to deal with the weed maintenance. I tend to veer away from these pesticides personally. Thanks for providing this study to us all for a nuanced perspective.

  26. Indeed, numerous farmers employ glyphosate to minimize tillage by utilizing it as a weed control method rather than for mere pleasure. You referenced a “chemical merry go round,” but let’s also consider the potential challenges of the tillage merry go round associated with organic farming, where multiple trips per year may be necessary, potentially leading to soil structure degradation.

Leave a Reply

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