Talk:Glyphosate-based herbicides/Archive 2

Revert
Regarding this revert [//en.wikipedia.org/w/index.php?title=Glyphosate-based_herbicides&diff=857108317&oldid=857108056] it is partially supported only by the 2012 source which says "formulated products containing POEA, such as Roundup, may be even more toxic". But the article content was written based on the 2004 source which is discussing surfactants generally - "Although surfactants probably contribute to the acute toxicity of glyphosate formulations...". The second issue is that the 2004 source says "there is insufficient evidence to conclude that glyphosate preparations containing POEA are more toxic than those containing alternate surfactants" - this is not a problem with the version I restored, because the content in the article is not discussing the relative toxicity of different formulations, so I think the change to can increase the relative acute toxicity of the formulation is less accurate. It might be better to add the quote to the citations when making "tweaks" like this in the future. I will do that now. ( Seraphim System ( talk ) 16:48, 29 August 2018 (UTC)
 * You're mixing up concepts here. The sentence in question is talking about relative toxicity of each ingredient to each other and potential interactions between them. It isn't talking about differences with other formulations in either version. It's not looking like there was any particular problem with my edit based on these comments. Kingofaces43 (talk) 17:07, 29 August 2018 (UTC)
 * I don't think I am mixing anything up. I understand what you were trying to say, but the way it was worded in the edit made it more confusing, especially for readers who are not familiar with the source material. The source also says "probably" - we can't change that just because you say "can" is more accurate. Anyway, I think the original version was fine but I added the direct quotes from the article in the interests of avoiding a long debate about this. Seraphim System ( talk ) 17:12, 29 August 2018 (UTC)
 * Please remember that this is significantly changing previously agreed upon language and is introducing unneeded quotation when a much more plain language version was already present. That will need to be undone. It's concerning you call the phrase that I added relative to redundant. I can wait for other editors to chime in if they see any legitimate problems with my tweak though, but I can also add in The surfactants in glyphosate formulations generally do not increase the toxicity of glyphosate itself towards humans, but the combined formulation toxicity can be higher than glyphosate alone due to higher relative toxicity of the surfactant. That should take care of any future confounding over synergism, additive effect, etc., and the following sentence continues the train of thought. Otherwise, better to go back to the status quo for how simple of a tweak it was intended to be. It's also completely silly to make the revert about can vs. probably. That was just a more concise term and had nothing to do about the accuracy comment. Kingofaces43 (talk) 18:03, 29 August 2018 (UTC)
 * Why is it concerning? The content that you changed was an accurate paraphrase of the content in the cited source. It covered synergism. This "tweak" was not supported by the source. Changing probably to can is not just a more concise term, it changes the meaning of the source. Seraphim System ( talk ) 20:49, 29 August 2018 (UTC)

I hope the two of you realize the extent to which you both are reverting to the kind of discussions-without-resolution that have been happening before. I've taken some time to look at the sources, and I'll try again here to break down the issues.

Over the past day or so, there have basically been four versions of the content. The first was as of this version:, before the recent changes was made, which I'll call Version 1. The second was the change by KofA as of this version:, which I'll call Version 2.

Version 3 is after SS's revert:, and Version 4 is after SS's subsequent edits: , which is close to what is on the page now.

I'll go straight to my own opinion, and then after that I'll explain why. Bottom line: I think that Version 2 is the best, and we should either (a) go with Version 2, or (b) modify Version 2 by: The surfactants in glyphosate formulations generally do not increase the toxicity of glyphosate itself towards humans, but can increase the relative acute toxicity of the formulation.

I looked at two sources. The first is Bradberry, where I read the abstract. They say: It is difficult to separate the toxicity of glyphosate from that of the formulation as a whole or to determine the contribution of surfactants to overall toxicity. Experimental studies suggest that the toxicity of the surfactant, polyoxyethyleneamine (POEA), is greater than the toxicity of glyphosate alone and commercial formulations alone. There is insufficient evidence to conclude that glyphosate preparations containing POEA are more toxic than those containing alternative surfactants. Although surfactants probably contribute to the acute toxicity of glyphosate formulations, the weight of evidence is against surfactants potentiating the toxicity of glyphosate. That contains a lot to parse, but they are basically saying that, complexities of the isolated toxicities of the individual chemicals set aside, glyphosate preparations containing POEA appear not to be more toxic than those with other surfactants, in the sense that they have not been shown to be so, but there appears to be some increase in toxicity of glyphosate plus surfactant preparations relative to glyphosate preparations with no surfactant.

The other source is SERA dated 2003. I focused on the Executive Summary and the Human Health Risk Assessment sections, looking for where they talked about POEA and toxicity. (I was struck by their statement on p. xvi that Glyphosate with the POEA surfactant is about as irritating as standard dish washing detergents, all purpose cleaners, and baby shampoos.) Anyway, on p. 3-3 they say While surfactants are typically classified as "inert" ingredients in herbicides, these compounds are not toxicologically inert and some surfactants may be more toxic than the herbicides with which they are used. Although surfactants may play a substantial role in the interpretation of a large number of suicides and attempted suicides involving the ingestion of glyphosate formulations, primarily Roundup, the acute mammalian toxicity of different glyphosate formulations do not appear to differ substantially. This is in contrast to the available data on the toxicity of various formulations to aquatic species, as detailed in the ecological risk assessment. So: surfactants add significant toxicity to aquatic organisms, but for acute mammalian toxicity, they can add some toxicity but not enough to make for meaningful differences between one formulation and another. On p. 3-7, talking about deliberate ingestion of concentrated products, the POEA surfactant used in glyphosate formulations (e.g., various formulations of Roundup) is a factor, and probably the dominant factor, in some of the effects seen in humans in cases of the suicidal ingestion of glyphosate formulations. On p. 3-21: Thus, in repeated dosing, the NOAEL for glyphosate of 1000 mg/kg/day was substantially higher than the NOAEL for either POEA (15 mg/kg/day) or neutralized POEA (50 mg/kg/day). That's in a lab, with giving very big doses to rats; a higher NOAEL means less toxic. Same page: Although there is evidence that POEA is more toxic than glyphosate to aquatic species (Section 4), the acute oral toxicity of Roundup (glyphosate and surfactant, LD50 in rats of 5400 mg/kg) is almost the same as that of glyphosate (LD50 in rats of 5,600 mg/kg). Meaning: for rats, the acute toxicity of glyphosate plus POEA, and that of glyphosate alone, are about the same. Taking all of that together: yes, POEA can add some measurable toxicity to glyphosate, but ultimately the change in toxicity for real-world human exposure isn't a particularly big change. And that's quite consistent with Bradberry.

OK then, so here is why I like Version 2. First, it's more precise to say "concentrated", as opposed to Version 1. More centrally, if we say "probably do", that makes it sound like people are probably in more danger when the herbicide contains POEA. And that does not reflect the sources. If we say POEA "can increase", that correctly reports that it can add more toxicity but it does not mean that "POEA: you're in more danger". In other words, in a lab setting it's possible to measure an increase in acute toxicity due to POEA, so that's what it can do, but that lab result does not translate into a measurable difference in what happens to people when herbicides are used in real life. And I don't think that the quotes in Version 4 do much to make things any clearer, and they do sort of sound like they contradict each other.

Now I also recommend deleting "generally do not increase the toxicity of glyphosate itself towards humans, but", because I think it's too technical and confusing for general readers. (It doesn't make glyphosate more toxic but it makes herbicides containing glyphosate more toxic – one has to be a chemist or toxicologist to make sense of that. Readers are unlikely to care about the mechanism of additivity or synergism.) If we just say The surfactants in glyphosate formulations can increase the relative acute toxicity of the formulation we're saying what matters: yes, they can cause more toxicity in a relative comparison – but we don't need to belabor the technical aspects.

I hope that helps. --Tryptofish (talk) 21:02, 29 August 2018 (UTC)
 * I prefer Tryptofish 's suggestion to the lengthier versions, as long as the SERA source is cited. My comments were based on the sources that were actually cited when the edits were made. Seraphim System ( talk ) 21:24, 29 August 2018 (UTC)


 * Yeah, I'll admit I'm disappointed that a rather simple change became suddenly controversial. Your summary and version 2 more or less gets across what I was working towards, so I'm good with that. Your sentence POEA can add some measurable toxicity to glyphosate, but ultimately the change in toxicity for real-world human exposure isn't a particularly big change. is more or less the take home message we want when reading this sentence and the rest of the paragraph. I don't have a super strong preference on dose makes the poison or Dose–response relationship, though the latter would need some text addition.


 * The problem with removing the text related to synergism is that it addresses a noteworthy fringe claim that POEA makes glyphosate much more toxic. Obviously we don't want to bend over backwards addressing it, but keeping it to a one-liner helps a lot. We've talked about it before in the lead, but here we can have a little more detail beyond just overall safety of the formulation and try to take the details up just one step in detail (ideally not more than that). It's a problem trying to distill synergism, etc. for encyclopedic text, which is why even though I proposed a lengthier more explicit version above, I much prefer the shorter version 2 way of clarifying it. It's a "happy" medium of sorts balancing a few different things. If there's a better way of summarizing what you said about technically increasing formulation toxicity in the lab, but not really having much of a real-world effect, I'm all for replacing the "generally does not increase the toxicity" language with something better. There's a lot of potential moving parts here as you definitely dug into in your analysis, so that's why I was mostly focusing the slight changes in version 2. Kingofaces43 (talk) 00:08, 30 August 2018 (UTC)
 * Since it looks like everyone was fine with version 2, I've gone ahead and reinserted it. We can move forward from there as a starting point for whether we want to alter the brief comment on lack of synergism, but I'm fine as is if no one else feels strongly. Kingofaces43 (talk) 18:49, 30 August 2018 (UTC)
 * I'm not sure whether everyone was fine with the longer form of Version 2, but we will see. Speaking only for myself, I'm OK with it, although I'm not convinced that the synergism thing will be accessible to readers. My only other issue is with the blue link to The dose makes the poison, because we already have the exact same link in the lead section. I changed it back to Dose–response relationship (count it as my revert for the day), but I also made the other link a hatnote to the section. --Tryptofish (talk) 20:26, 30 August 2018 (UTC)
 * I would have changed the blue link if you asked since I wasn't super picky on which as I mentioned above (I did add a little text at the dose-response article). I wasn't sure on what was preferred, so no big deal there for now.
 * As for the longer form, I wasn't entirely sure if people were referring to my lengthier version I proposed before you posted the four previous versions, but if it's in reference to the phrase in version 2 addressing synergism, it seems like the current version is a baseline to work forward on for refining if anyone feels strongly. I'm OK with it since it gives appropriate weight while not getting too far into the weeds of jargon that such text easily could. If we're going to remove it, we'll need some sort of more concise replacement text to get a similar idea across. As I've said before though, I think we're at a happy medium to at least have a stable version. Kingofaces43 (talk) 20:53, 30 August 2018 (UTC)
 * When I've referred to a version, I mean the ones I numbered in the four boxes above, not anything earlier or later than that. And the shorter version of Version 2 is the one I discussed with SS just above, leaving out the synergy. --Tryptofish (talk) 20:58, 30 August 2018 (UTC)
 * I'm definitely not fine with Version 2, I supported Tryptofish's proposed change. I don't know how I could have made it clearer that I object to Version 2. Seraphim System ( talk ) 21:05, 30 August 2018 (UTC)
 * I figured I would just wait to hear from you, but that was certainly clear to me. So I just shortened it (which I do not consider to be a revert, in case anyone is thinking that way). If KofA would like to include some explanation of the synergy issue, let's do that in a separate sentence. --Tryptofish (talk) 21:11, 30 August 2018 (UTC)
 * Yes, I'm not even sure it belongs in the acute toxicity section. The FRINGE synergy stuff KoA is worried about is not about acute oral toxicity. Seraphim System  ( talk ) 21:21, 30 August 2018 (UTC)
 * Tryptofish, just a procedural point since you may have lost track of it in the shuffling content, but this content is the rough status quo that has been removed a time or two without consensus gained for removing it after a bit of talk page discussion, so that needs to remain at this point until we get consensus on something else. I'm all for swapping it out if we get something nailed down though, but it's a bit early for removal. Kingofaces43 (talk) 21:38, 30 August 2018 (UTC)
 * I've said already I'm mostly fine with keeping that text as-is as it balances most of the concerns brought up in discussion pretty well. What are editors suggesting as an improvement though? Kingofaces43 (talk) 21:38, 30 August 2018 (UTC)
 * That's not my understanding of the procedural issue, but I'm just trying my hardest to keep a lid on the arguments here. I think there was a pretty thorough discussion of why it would be good to shorten that sentence, and you kind of ignored it when you implemented the longer form of V2. SS said very clearly that they accepted my overall analysis and supported the shorter version that I proposed. Anyway, I don't want to argue about this, but I have no objection to what is on the page now, and I have no objection if you want to add a separate sentence or two that explains the synergy issue more clearly. So my suggestion for improvement is that you cover the synergy in a new sentence or two, not in the existing sentence. --Tryptofish (talk) 21:46, 30 August 2018 (UTC)
 * Generally, when someone removes text and that removal is undone, we don't go removing it again unless there is consensus for it in terms of 1RR procedure here.
 * Either way, I'm a bit confused now. Originally, we were talking about not doing something much more involved for the synergy stuff because it gets to be too much detail, and the removed phrase was in the "just right" territory for the body. Are you suggesting we shift more in the former direction a little now? As for the previous conversation on it, I was doing the opposite of ignoring it, but addressing why we wanted to address the synergism stuff in that sentence concisely. In part, the sentence is including components of individual toxicity and combined toxicity. We didn't have any sort of consensus to exclude the phrase though. If you can clarify what you're thinking for direction, I can look at potential new sentence ideas, but we might be better off just sticking with the status quo for that phrase too. Kingofaces43 (talk) 22:34, 30 August 2018 (UTC)
 * The status quo would be version 1, before the changes you made that resulted in the compromise version. I think we should keep the current compromise version, I don't think the nuances of the longer sentence are going to be obvious to a reader without substantial background knowledge. That is why I supported the current version. Seraphim System  ( talk ) 00:08, 31 August 2018 (UTC)
 * Correct. Everyone has agreed on the changes to the first sentence and second half of the second sentence in version 1, so we wouldn't go back to version 1 entirely. That only leaves the bit about not increasing the toxicity of glyphosate as the remaining status quo since that's the only thing under dispute. Either way, I want to be sure of what Tryptofish's intent with my previous question, so that could be moot too. Kingofaces43 (talk) 00:41, 31 August 2018 (UTC)
 * Actually, I think version 2 is worse then version 1, but I think the shortened version is preferable to the original version 1. I don't really want to keep dragging this out but I'm very worried you are going to revert again and claim that I agreed to the changes., so I want to make sure there is no continued misunderstanding about this. There is currently no support for your proposal to reinstate your preferred version. Tryptofish has floated the possibility of adding this in a separate sentence. Seraphim System ( talk ) 00:56, 31 August 2018 (UTC)
 * KofA, the short version is that I see it the same way as SS here. To go into more detail, I don't know if there was earlier discussion about not having anything longer about synergy, because it must have happened before I got closely involved. I don't have a problem with omitting synergy entirely. I also don't have a problem with covering it. We could do it under acute toxicity, or under long-term effects, or even part and part. I don't care about that either. If we add a single, well-constructed sentence, that would be much clearer than the within-sentence phrase that I removed. The one thing I do care about here is that it needs to be clear to readers. --Tryptofish (talk) 15:59, 31 August 2018 (UTC)
 * I went ahead an added a separate sentence, so hopefully that moves things towards being resolved. If that change needs to be undone, we'll need to go back to this version. I'm content with what we have right now for satisfying NPOV, but we could add more language that the formulations are not a significant risk to mirror the introductory paragraph of the Toxicity section to be more concrete if needed. Does this move towards what you were thinking without getting too technical? Kingofaces43 (talk) 23:09, 4 September 2018 (UTC)
 * I tweaked it a bit further, but yes, it looks good to me. Thanks. I'm satisfied with it, the way it is now. --Tryptofish (talk) 23:23, 4 September 2018 (UTC)
 * Just to clarify, Tryptofish was supporting version 2 overall, but also considering if it should be shortened. You said you were fine with their suggestions, so I assumed you were fine with version 2 while looking at further refining the interaction content in this discussion. Kingofaces43 (talk) 21:38, 30 August 2018 (UTC)
 * I don't think this section needs anything about synergy. The disputed studies about synergy were about long-term effects, I don't think the article should be bloated to "disprove" something that is not a major issue with respect to acute toxicity. Seraphim System  ( talk ) 21:50, 30 August 2018 (UTC)
 * I'm not sure what you're referring to here considering I've never claimed this was just for long-term effects. These claims come up on multiple fronts, including acute toxicity, but more importantly, MEDRS sources make a specific point of talking about this with respect to acute toxicity. Kingofaces43 (talk) 22:34, 30 August 2018 (UTC)

Is there Roundup in your breakfast cereal?
What can we do with this?— Vchimpanzee  •  talk  •  contributions  •  20:00, 29 August 2018 (UTC)
 * I'd rather have a better source than People's Pharmacy. Can we go back to the actual scientific studies? And we really need to have a WP:MEDRS secondary source to evaluate whether or not the amounts that are in the food are meaningful. --Tryptofish (talk) 21:05, 29 August 2018 (UTC)
 * At best, this would go on the Environmental Working Group's page as part of the their many controversies related to non-scientific "reports". What I've seen so far in sources really wouldn't warrant mention here or most other articles in terms of MEDRS though. Kingofaces43 (talk) 00:31, 30 August 2018 (UTC)
 * I was really just pointing out the topic. I wouldn't even know how to find the proper sources.— Vchimpanzee  •  talk  •  contributions  •  14:02, 30 August 2018 (UTC)
 * This would belong in the main article really, but here are some potential sources. It doesn't look as if there are any good reviews, but at least IMO, this content wouldn't come under MEDRS if we are just reporting the encountered residues and relating them to MRLs/ADIs, and not making any statements as to whether they are safe or not.
 * https://www.tandfonline.com/doi/abs/10.1080/02652030119594
 * https://www.tandfonline.com/doi/abs/10.1080/02652030412331282385
 * https://enveurope.springeropen.com/articles/10.1186/s12302-015-0052-7
 * https://onlinelibrary.wiley.com/doi/abs/10.1002/ps.775
 * https://pubs.acs.org/doi/abs/10.1021/jf025908i
 * https://www.sciencedirect.com/science/article/pii/S1438463911002379
 * https://www.sciencedirect.com/science/article/pii/S0021967300001035 (mainly of use for finding more primary research)
 * https://www.tandfonline.com/doi/abs/10.1080/0265203031000109477
 * https://link.springer.com/article/10.1007/s11356-016-7425-3
 * https://onlinelibrary.wiley.com/doi/abs/10.1002/ps.2128
 * http://www.food.gov.uk/sites/default/files/multimedia/pdfs/cropguidecerealsdec06.pdf
 * http://cereals.ahdb.org.uk/media/308647/pr464-final-project-report.pdf
 * SmartSE (talk) 15:15, 30 August 2018 (UTC)

Surfactants - today's edit uses a 14 year old review to claim safety
This edit is concerning. We need to use the latest scientific findings, not dig up ones from 2004. What was the purpose in making this edit today?

This is a 2016 review PMC 4947579


 * 'Glyphosate can display endocrine-disrupting activity (80, 82), affect human erythrocytes in vitro (83), and promote carcinogenicity in mouse skin (84). Furthermore, it is considered to cause extreme disruption in shikimate pathway, which is a pathway found in plants and bacteria as well as in human gut bacteria. This disruption may affect the supply of human organism with essential amino acids (85). Commercial glyphosate formulations are considered to be more toxic than the active substance alone (80, 83, 86, 87). Glyphosate-based herbicides, such as the well-known “Roundup,” can cause DNA damages and act as endocrine disruptors in human cell lines (60) and in rat testicular cells (88), cause damages to cultured human cutaneous cells (89), and promote cell death in the testicular cells of experimental animals (88, 90). There is evidence also for their possible ability to affect cytoskeleton and intracellular transport' (91).

And this from The Intercept with new insights gleaned from the Roundup Cancer Trial discovery phase:


 * 'Until recently, the fight over Roundup has mostly focused on its active ingredient, glyphosate. But mounting evidence, including one study published in February, shows it’s not only glyphosate that’s dangerous, but also chemicals listed as “inert ingredients” in some formulations of Roundup and other glyphosate-based weed killers'.

For the encyclopedia it is best to use more recent data.  petrarchan47  คุ  ก   00:07, 30 August 2018 (UTC)
 * I didn't dig it up, it was added by KoA (who copied it from the main Glyphosate article). However, the conclusion that acute oral toxicity for mammals is low is repeated in the 2018 review, which I added, so I did not challenge the 2004 Bradberry source because I was able to confirm that it has not been superseded.. The review quoted above, however, seems to be discussing something different: Therefore, the determination of “safe” levels of exposure to single pesticides may underestimate the real health effects, ignoring also the chronic exposure to multiple chemical substances. Seraphim System  ( talk ) 00:23, 30 August 2018 (UTC)
 * I do think it would be a good idea to replace older sources with newer ones. We need to be thoughtful, however, about the kind of context that I discussed above in, in terms of animal toxicology relative to human risk: I'm not sure that anything in the scientific consensus has really changed. --Tryptofish (talk) 20:31, 30 August 2018 (UTC)
 * This is why I'm ok for now at least on focusing on these older sources that newer reviews just tend to cite as (look at this 2000 summary for more details). I've been actively looking for new sources, but most of them aren't really saying anything different as you're saying. It might just be a matter of adding newer references to what we already have without changing content much if at all. Kingofaces43 (talk) 20:44, 30 August 2018 (UTC)
 * Also, I've been getting the impression that with all the back-and-forth changes, there might be some cites that have ended up after the wrong sentences, so it would be worth taking a close look at all of that. --Tryptofish (talk) 20:55, 30 August 2018 (UTC)

Regarding this change made today and the use of updated (WP:MEDRS compliant) sources, have you (or others active on this thread) done a literature search for updated material before continuing to utilize this 14 year old study?

A review of WP:MEDDATE may be in order.  petrarchan47  คุ  ก   07:33, 5 September 2018 (UTC)


 * I think that editors in the discussion above have already indicated a receptiveness to finding and using new sources. There was nothing final about the source choice for the edit in the diff you cite, so I think that providing us with a more recent source would be welcome. It simply needs to be a MEDRS-reliable source that is directly about the material in that sentence, and it has to be cited accurately. --Tryptofish (talk) 15:42, 5 September 2018 (UTC)


 * Apologies for my lack of clarity. I did not mean to ask if editors were receptive to updated material. I am asking whether a search for new material was done or not, prior to these recent edits.   petrarchan47  คุ  ก   20:00, 5 September 2018 (UTC)


 * I just did a quick search. I found two recent sources that look to me to be relevant: and . There is also this:, but it's by Seralini. --Tryptofish (talk) 20:56, 5 September 2018 (UTC)
 * Those first two, however, are primary sources. I did some more searching, and when I searched PubMed for "POEA" and set it to "review", I only got the Bradberry and Williams reviews, nothing more recent. But looking for "herbicide surfactants" and "review", I got these, which are I think the best out there for this purpose: and, as well as one about aquatic applications: . --Tryptofish (talk) 21:11, 5 September 2018 (UTC)
 * According to Tarazona: "In fact, the UN and EU guidance recommends carcinogenicity and genotoxicity studies to be conducted on individual chemicals, limiting testing of mixtures/formulations to cases where synergistic effects are expected (United Nations 2015)." - I think if we are going to mention synergy, it would have to be in the carcinogenicity section, not acute oral toxicity, and we should follow the most up to date source which seems to be Tarazona (2017). Seraphim System ( talk ) 04:13, 15 September 2018 (UTC)
 * Mention of synergy for carcinogenicity is not the same for acute toxicity, so I don’t see why you bring up acute toxicity. Someone could mention it for carcinogenicity in addition if they want to craft something obviously. Kingofaces43 (talk) 04:49, 15 September 2018 (UTC)
 * I'm sure SS is referring to the sentence added in the "Acute toxicity#Human" section. I just modified that sentence to make it more specific: . But if surfactants are different with respect to their non-acute effects on carcinogenicity, then the "Carcinogenicity of active ingredient" section should be expanded beyond the active ingredient, and it should be covered there. --Tryptofish (talk) 18:34, 15 September 2018 (UTC)

Please update this page re surfactants
Sera 2011 *:


 * ''Many glyphosate formulations include surfactants, and the toxicity of these surfactants is of equal or greater concern to the risk assessment than is the toxicity of technical grade glyphosate ... surfactants appear to be agents of concern

We also have PUBMED 24434723 (2014)
 * These results confirm that G formulations have adjuvants working together with the active ingredient and causing toxic effects that are not seen with acid glyphosate

And PMC 5756058 (2018)
 * G being tested alone in chronic regulatory experiments to establish the ADI (RfD in USA) appears inappropriate, in light of these results. As a matter of fact, synergistic toxic effects undoubtedly occur 

And PUBMED 24999230 (2014)
 * In conclusion, our results demonstrated that the addition of adjuvant to glyphosate formulation increase the toxicity of the mixture in cell culture. Furthermore, cell culture exposed to agrochemical mixture showed an increased ROS production and antioxidant defenses

Even after being shown the updated Sera 2011 language, KoA43 has reinserted the following using a source from 2004:
 * Surfactants generally do not cause synergistic effects that increase the toxicity of glyphosate within a formulation.

I would encourage editors to look at WP:MEDRS, WP:NPOV and WP:NOTHERE before continuing to edit in this topic area.  petrarchan47  คุ  ก   19:00, 15 September 2018 (UTC)
 * The text in question currently says: "The surfactants in glyphosate formulations can increase the relative acute toxicity of the formulation. Surfactants generally do not cause synergistic effects that increase the acute toxicity of glyphosate within a formulation." Those sources are all supportive of the first of those two sentences, and at least based on what you quoted, they are not about the second sentence (with the exception of the one by Séralini). There is a difference between "additive" and "synergistic". --Tryptofish (talk) 19:09, 15 September 2018 (UTC)
 * And I linked to and commented on the Séralini paper in the talk section just above. --Tryptofish (talk) 19:16, 15 September 2018 (UTC)
 * "But it's by Seralini" - is this supposed to discount the paper that included his name? Here are the facts:
 * The Seralini smear campaign
 * Seralini studied the effect of Roundup formulation on rats, using the same design as the one Monsanto uses to claim safety, except that he extended the study to include the rat's full lives, as opposed to the 90 day study Monsanto prefers. Wikipedia took part in the smear campaign that followed the release of his findings. It is time to stop that nonsense. He is widely cited in literature and remains well respected.   petrarchan47  คุ  ก   19:55, 15 September 2018 (UTC)
 * I just carefully re-read the recent Séralini paper that includes the quote about synergy, and taken in context, the comment about synergy is referring to heavy metals ("As, Cr, and Ni. Pb and Co") rather than to surfactants. It might be worth adding something about metallic cations to this page if there are other independent sources that confirm that, but we should be clear about whether we are talking about surfactants or about metals. --Tryptofish (talk) 20:16, 15 September 2018 (UTC)
 * I did a search about heavy metals in glyphosate, and here is what I found. There is one group that has reported a possible human toxicity of glyphosate herbicides interacting with heavy metals that are present in the environment:, . There are two other groups who reported, contradicting Séralini, that glyphosate (and in one case, RoundUp) actually reduces the toxicity of heavy metals: , . --Tryptofish (talk) 20:43, 15 September 2018 (UTC)
 * These are all non-WP:MEDRS sources aside from SERA, and we more or less already reflect what SERA is saying in other sources. Surfactants can have relative higher toxicity than the active ingredient (that is considered to have very low toxicity in general), but the absolute toxicity of the combined ingredients (i.e., the formulations) is considered low or not a significant risk. That's except for directly spraying POEA=based formulations in water because aquatic organisms tend not to do well when you put things like dish soap in water. Kingofaces43 (talk) 05:29, 16 September 2018 (UTC)
 * These do seem to be primary, but this 2018 review from Mesnage does seem to pass MEDRS  Seraphim System  ( talk ) 19:12, 16 September 2018 (UTC)
 * Mesnage is part of Séralini's group. --Tryptofish (talk) 19:35, 16 September 2018 (UTC)
 * Seralini does not seem to be involved with the group currently and is not one of the authors about the 2018 review. Regarding Seralini, there is one study that received a lot of attention, and was retracted and republished. It is a primary study, so this is not really a FRINGE argument. It's not MEDRS anyway. The comments about FRINGE should be directed to publications, not persons (this includes language like fringe scientist, which only happened once in earlier discussions on this talk page.) Exclusion of a particular scientist is not the same as asking whether a particular source is reliable for a particular statement. MEDRS is intended to make medical content more reliable, so I appreciate and support any concerns about inclusion of a primary study that has been retracted and has not been included in any of the most recent risk assessments (IARC and EU both excluded)) - but we would need a lot more then the publication of a controversial study for an outright ban of multiple recognized scientists.  Seraphim System  ( talk ) 20:36, 16 September 2018 (UTC)
 * Well, sort of. Anyone who has published multiple times with Séralini can still be expected to have similar preconceptions, and that's what we have with the 2018 review. Where the review talks specifically about glyphosate-based herbicides and their effects, it pretty much only cites papers by Séralini, so I do not consider it a reliable source for that specific question. (There are other lab groups who get opposite results, and the review does not acknowledge them.) But a much larger part of the review is about analyzing and making recommendations about how pesticide toxicity should be measured and how resulting regulation should be determined: that the formulation and not just the active ingredient needs to be examined. I'm fine with citing it for the latter, especially since that's really what most of the review is about. I just want to be careful about the parts that are dubious. And after all, a big part of having this page, separate from the one about glyphosate, is that we should be addressing the aspects that relate to the formulations and not just the active ingredient. --Tryptofish (talk) 21:04, 16 September 2018 (UTC)
 * It's fine to have a personal aversion to Seralini, but to try and insert it here by arguing that all of his work should be rejected is WP:OR. Don't we look at impact factors anymore? If the editorial board sees fit to publish, a Wikipedia editor cannot override that decision, in my understanding.   petrarchan47  คุ  ก   21:39, 16 September 2018 (UTC)
 * I hear you about the editorial board, but this is why we treat independent secondary sources differently than primary sources or non-independent secondary sources. --Tryptofish (talk) 21:45, 16 September 2018 (UTC)
 * I don't think we have enough to say anything along the lines of "there are synergistic effects" and I think we have to consider the 2012 Long term toxicity primary study was excluded by both the IARC and the EU in the most recent risk assessments. I agree with what Tryptofish is saying here - we can't use a primary study to dispute the conclusions of secondary sources. But I do think the article would benefit from clarifying a few points. For example a separate 2012 source by Mesnage and Seralini includes important content about the composition of the formulations, not only medical content. The conclusion of the study is that formulations should be studies as mixtures for toxic effects. I don't see anything fringe or even controversial about that. Thus, I don't think the author of the article alone is enough to justify removal of sourced content without further explanation. Seraphim System  ( talk ) 21:52, 16 September 2018 (UTC)
 * The issue that Tryptofish is getting at (I think) is that Mesnange is not WP:INDEPENDENT of Seralini or the controversies around his lab. It's similar to how we are wary of reviews citing their own primary research. The more important thing is that the journal is not reliable though being a Frontiers journal. Typically those get removed pretty quickly by editors on the look out for predatory journals. Kingofaces43 (talk) 21:59, 16 September 2018 (UTC)
 * Can you link to the RS/N discussion about Frontiers? I found the 2013 study from Elsevier, I don't think there's anything wrong with the journal . It's not just one study, but if there is a problem with this journal I would like to review previous discussions about it. Seraphim System ( talk ) 22:07, 16 September 2018 (UTC)
 * Oh, I hadn't realized that it was a predatory journal. That's a very serious problem with using it as a source. It's possible that no editorial board really made a decision to publish, other than making sure that the check cleared (slight hyperbole on my part). Frontiers Media, but cf:, . Added after ec: I agree with SS that it would be a good idea to check RSN. --Tryptofish (talk) 22:11, 16 September 2018 (UTC)
 * I found it at Reliable sources/Noticeboard/Archive 205. Looks like consensus was not-RS. --Tryptofish (talk) 22:17, 16 September 2018 (UTC)
 * I would also like to repose my question: what ever happened to using impact factors to determine reliability?   petrarchan47  คุ  ก   22:23, 16 September 2018 (UTC)
 * I don't think that WP:RS says anything about impact factors, although maybe I missed something. WP:PROF does, but that's for determining notability, not reliability. --Tryptofish (talk) 22:36, 16 September 2018 (UTC)

It seems in this case we would treat this as SPS since the journal is non-RS. The authors are RS but since this is pay-to-publish this is in practice no different from any other SPS. Attributed use of SPS is allowed for established scholars of this class. Seraphim System ( talk ) 23:29, 18 September 2018 (UTC)
 * I'm assuming that SPS is "self-published source". As such, it depends a lot on how we use the source. If we quote it for an opinion, not a statement of fact, I guess it can be used that way, but if we only use an opinion, there are due weight concerns. --Tryptofish (talk) 23:36, 18 September 2018 (UTC)
 * WP:RS/SPS makes it clear that these kinds of source are of very limited use. --Tryptofish (talk) 23:39, 18 September 2018 (UTC)
 * I agree re due weight it would have to be kept brief in proportion to more authoritative sources, but is a significant viewpoint that should be covered briefly per NPOV. Seraphim System  ( talk ) 23:42, 18 September 2018 (UTC)
 * I'm having a hard time picturing what the content would actually be. We could attribute to them the opinion that most regulatory mechanisms are not rigorous enough, but then we would have to balance it with the majority view, and that would leave us with "almost everyone says X, but a few people say Y". --Tryptofish (talk) 23:48, 18 September 2018 (UTC)

Endocrine disruption
I'm noticing a lot about endocrine disruption in the sources. Maybe that's something that needs to be added to the page, along with carcinogenicity. --Tryptofish (talk) 19:38, 16 September 2018 (UTC)

Seralini
Just a note about rejecting Seralini, not in the above case, but it general. Unfortunately a smear campaign was successful in causing bias against Seralini (see KoA43 comment just above re "controversies around his lab"); this is similar to concerns about Monsanto's ghostwritten science: it isn't always clear what we can trust. About Seralini's retraction:
 * A publication on the harmful effects that Roundup has on rats has been discredited, perhaps thanks to the relationship between a Monsanto employee and the Journal of Food and Chemical Toxicology. Monsanto employee David Saltimiras also successfully lobbied in 2012 for various people to write letters to the journal, which were then also published - which accused the study of deficiencies, selective statistics and bias WZ   petrarchan47  คุ  ก   22:23, 16 September 2018 (UTC)
 * I actually think that there has been a lot of fishiness around Monsanto, you may perhaps be surprised to know. But two wrongs do not make a right. A key fact about Séralini is that most of what he has published has turned out not to be reproducible when tested by independent labs (academic, not company labs). --Tryptofish (talk) 22:34, 16 September 2018 (UTC)
 * It would be helpful if you would source your statement. "most of what he has published has turned out not to be reproducible when tested by independent labs" does not appear to be factual.
 * Here is a bit more detail about the (one and only, AFIK) Seralini study that has come under attack:
 * One of the victims of their campaign was French toxicologist Gilles-Éric Séralini. He did exactly what Monsanto should have done. For two years, Séralini dripped Roundup into the drinking water of laboratory rats and fed them glyphosate-laden, genetically modified corn. What he found was alarming: Some of the animals developed kidney damage, while the females developed breast cancer at remarkably high rates.


 * When the journal Food and Chemical Toxicology published the study in September 2012, all hell broke loose in Séralini's life. Hundreds of researchers protested and Séralini was accused of making "false statements" and "using animals for propaganda purposes." The journal withdrew the publication in November 2013. It may have been a coincidence, but the magazine had appointed a former Monsanto employee to its advisory board six months earlier.


 * The internal memos also confirm how Monsanto exerted pressure. David Saltmiras, a Monsanto expert at the time, boasted that he had "successfully facilitated numerous third party letters to the editor." He described his actions as being "in our own best interest" and as "the last rites for Séralini's few remaining shreds of credibility." Monsanto Faces Blowback Over Cancer Cover-Up    petrarchan47  คุ  ก   04:56, 18 September 2018 (UTC)
 * It's tricky to use a commentary by a journalist for drawing such conclusions. Here is an independent scientific review that describes Séralini's reputation pretty well: . "The study appeared to sweep aside all known benchmarks of scientific good practice and, more importantly, to ignore the minimal standards of scientific and ethical conduct in particular concerning the humane treatment of experimental animals." It would be pretty odd for one study by an investigator to do that, and then all the subsequent work be just fine. Thus, the paper that we discussed above, from a pay-to-publish journal. Also above, I pointed to recent papers that got results that were opposite to what that paper concluded. I don't know if that response is sufficiently detailed, but I've thought about it carefully, and maybe this is something that could be better dealt with via an RfC or at WP:RSN. --Tryptofish (talk) 21:31, 18 September 2018 (UTC)

Roundup page
Editors here should be aware that Roundup (herbicide) is starting to be recreated in mainspace. --Tryptofish (talk) 21:38, 16 September 2018 (UTC)
 * Yep. Five years after it was decided it should be. related RfC   petrarchan47  คุ  ก   21:42, 16 September 2018 (UTC)

RfC: Coverage of Roundup Cancer Case
RfC is here.  petrarchan47  คุ  ก   04:11, 17 September 2018 (UTC)