For the first months of life, a breastfed baby may receive almost everything it needs from one source: its mother’s milk.
We encourage breastfeeding because its benefits are well established. Breastmilk provides nutrition, antibodies and other biologically active components, and changes as a baby grows.
But there is another question that receives considerably less attention.
What else is getting into breastmilk?
More specifically, what happens when a mother is exposed to glyphosate — the active ingredient in the world’s most widely used herbicides?
Does glyphosate enter her breastmilk?
Can her baby ingest it?
And if the answer is yes, even occasionally and at very low concentrations, what do we actually know about exposing a newborn baby during one of the most rapid periods of development it will ever experience?
After more than 50 years of glyphosate use, those might seem like questions science would have answered comprehensively.
It hasn’t.
Instead, the relatively small number of human breastmilk studies conducted to date have produced remarkably different results.
Some researchers have found no detectable glyphosate at all.
Others have found very little.
And researchers in Brazil have reported glyphosate in every breastmilk sample they tested.
So what is going on?
The short answer is that the research is conflicting. Some highly sensitive studies have detected little or no glyphosate in human breastmilk, while studies in Brazil have reported glyphosate in every sample tested. Differences in analytical methods, populations, geography and potential exposure mean the results cannot simply be compared as though the studies were identical.
Does This Research Mean Mothers Should Stop Breastfeeding?
Before going any further, something needs to be made very clear.
Nothing discussed in this article should be interpreted as a reason for mothers to stop breastfeeding.
The benefits of breastfeeding are well established, and the research discussed here does not demonstrate that breastfeeding is unsafe.
If an environmental contaminant is finding its way into human breastmilk, breastmilk isn’t the problem.
The exposure is.
Nor should mothers be made to feel responsible for environmental exposures over which they may have little or no control.
A mother doesn’t determine agricultural spraying practices. She doesn’t establish pesticide residue limits. She doesn’t control what has been applied to every ingredient in the food she eats, what has been sprayed along roadsides or in public spaces, or what chemicals may be used around her workplace or neighbourhood.
The appropriate question isn’t whether mothers should breastfeed.
It is whether we are doing enough to prevent unnecessary chemical exposure to mothers and their babies in the first place.
Can Glyphosate Get Into Human Breastmilk?
This seemingly simple question has produced surprisingly complicated answers.
Glyphosate has chemical characteristics that make significant bioaccumulation in breastmilk unlikely.
Unlike persistent fat-soluble contaminants that can accumulate in body fat and concentrate in breastmilk, glyphosate is highly water-soluble and is generally considered to be eliminated relatively rapidly from the body.
That has formed an important part of the argument against claims that glyphosate might accumulate in human milk.
But there is an important distinction here.
“Does not bioaccumulate” does not mean “can never be present.”
A substance doesn’t necessarily have to accumulate in the body over months or years to be present following exposure.
And that brings us to what researchers have actually found.
Was Glyphosate Found in Breastmilk in the 2014 Moms Across America Testing?
International concern about glyphosate in breastmilk accelerated in 2014 after a small testing project commissioned by Moms Across America [PDF] reported glyphosate in three of ten breastmilk samples.
Reported concentrations ranged from 76 to 166 micrograms per litre.
The findings attracted considerable attention.
They also attracted considerable criticism.
This wasn’t a peer-reviewed population study, the sample size was tiny, and questions were subsequently raised about whether the analytical method being used was appropriate for accurately measuring glyphosate in human milk.
In 2015, toxicologist James Bus published an analysis arguing that the reported concentrations were inconsistent with glyphosate’s physicochemical properties, animal toxicokinetic evidence and available human biomonitoring data. The analysis was funded by the Joint Glyphosate Task Force, a group of companies holding technical registrations for glyphosate; members of the Task Force reviewed drafts, although Bus stated that the analysis and conclusions were his own.
Those criticisms matter.
Extraordinary findings require reliable analytical evidence.
And scientists subsequently went looking using more sophisticated analytical methods.
2016 US Research: Was Glyphosate Detected in Breastmilk?
In 2016, researchers published a study involving 41 lactating women living around Moscow, Idaho, and Pullman, Washington.
They tested breastmilk for glyphosate and its main breakdown product, aminomethylphosphonic acid — AMPA — using a validated analytical method capable of detecting both at low concentrations.
Neither glyphosate nor AMPA was detected in any of the 41 breastmilk samples.
Yet glyphosate was detected in many of the women’s urine samples.
That is an important finding.
It showed that, in this group of women, exposure could be occurring without glyphosate necessarily appearing at detectable concentrations in breastmilk.
Other analytical work published around the same period used LC-MS/MS and GC-MS/MS to investigate glyphosate in German breastmilk samples and similarly failed to confirm the high concentrations suggested by the original 2014 testing.
For anyone arguing that glyphosate routinely accumulates to high concentrations in human breastmilk, these studies present an obvious problem.
But the story didn’t end there.
2022 Brazil Study: Glyphosate Reported in 67 of 67 Breastmilk Samples
Researchers in Paraná, Brazil, collected breastmilk from 67 lactating women — 41 living in rural areas and 26 in urban areas.
The samples were collected during April and May, coinciding with a period of intensive glyphosate application to corn and soybean crops in the region.
The researchers reported glyphosate in every breastmilk sample analysed.
All 67.
The mean concentration reported was 1.45 micrograms per litre.
Perhaps just as interestingly, researchers did not find a statistically significant difference between the rural and urban mothers.
That doesn’t establish where the glyphosate came from.
It doesn’t prove that every mother in Brazil — let alone New Zealand — has glyphosate in her breastmilk.
And there is an important methodological qualification.
The researchers used a commercial enzyme-linked immunosorbent assay, or ELISA, rather than the more specific mass-spectrometry techniques used in some of the studies that found little or no glyphosate.
That makes analytical methodology an important part of this story.
But it doesn’t make the study disappear.
Instead, it raises another question.
Why did researchers studying one population report glyphosate in every sample, while researchers elsewhere using different analytical methods found none?
2024 US Study: Glyphosate Found in 1 of 74 Breastmilk Samples
In 2024, another US research team examined breastmilk from 74 women, including 26 vegans, 22 vegetarians and 26 nonvegetarians.
The researchers used liquid chromatography coupled with mass spectrometry, with a reported glyphosate detection limit of 1 microgram per litre.
Glyphosate was detected in just one of the 74 breastmilk samples.
That is a reassuring finding.
But it wasn’t quite zero.
AMPA — aminomethylphosphonic acid — was detected in an additional seven samples.
AMPA is a major environmental degradation product of glyphosate, although it can also have other environmental sources. Its detection therefore cannot automatically be attributed to glyphosate exposure.
The researchers concluded that, among their samples — collected mainly from women living in urban areas of the United States — glyphosate detection was rare and that breastfed infants were at low or minimal risk of glyphosate exposure through their mother’s milk.
But the authors also raised an interesting possibility.
They suggested that whether glyphosate is present in breastmilk — and at what concentration — may depend on where a mother lives and the timing of breastfeeding in relation to agricultural glyphosate application.
That matters.
Because if exposure varies with geography and timing, a study involving predominantly urban American mothers cannot necessarily tell us what might be found among mothers living in other environments, under different agricultural conditions, or at different times of the year.
And once again, we arrive at the same question.
Where is the larger research needed to find out?
2026 Brazil Study: Glyphosate Reported in 100 of 100 Breastmilk Samples
Then, in February 2026, researchers from São Paulo State University and collaborators published another Brazilian study.
They collected breastmilk from 100 postpartum women in Tupã, São Paulo — 90 living in urban areas and 10 in rural areas.
Using their HPLC screening method, the researchers reported detecting glyphosate in 100 out of 100 samples.
The reported concentrations were striking, ranging from 7.705 mg/L to 19.43 mg/L.
But those numbers require considerable caution.
The researchers used an HPLC method with UV-Vis detection following derivatisation. Their reported limit of detection was 0.14 mg/L and their limit of quantification was 0.43 mg/L.
More importantly, the researchers themselves describe their method as a screening tool, rather than a confirmatory method.
They explicitly state that positive findings obtained using the method should be confirmed using more sensitive and fully validated analytical techniques such as LC-MS/MS.
The authors also acknowledge that the method’s limit of quantification is higher than reference values used for related matrices, restricting its applicability for regulatory or confirmatory purposes.
In other words, this is not a study we should simply cite as proof that all 100 women had those concentrations of glyphosate in their breastmilk.
The authors themselves say positive findings require confirmation.
But that leaves us with another uncomfortable question.
Has anyone done it?
Have these samples — or comparable samples from the same population — subsequently been analysed using highly sensitive confirmatory LC-MS/MS?
Because a screening study reporting glyphosate in 100 out of 100 mothers should surely prompt further investigation.
Why Do Glyphosate Breastmilk Studies Produce Such Different Results?
Put these studies alongside one another and something becomes obvious.
They aren’t simply different groups of researchers producing different answers.
They have also used different analytical methods, with substantially different detection capabilities.
The 2024 US study, for example, reported a glyphosate detection limit of 0.001 mg/L.
The 2026 Brazilian screening study reported a detection limit of 0.14 mg/L.
That’s a 140-fold difference in the stated detection limits.
The more sensitive US study detected glyphosate in only one of 74 women.
The much less sensitive Brazilian screening method reported glyphosate in all 100.
That doesn’t establish that either study is wrong.
Indeed, the concentrations reported by the Brazilian researchers were high enough that analytical sensitivity alone cannot explain the difference.
But it does demonstrate why these studies cannot simply be counted as votes:
Two studies found glyphosate. Two studies didn’t. Therefore nobody knows.
Science doesn’t work like that.
The analytical method matters.
The population matters.
The timing matters.
The environment may matter.
And the reliability and reproducibility of the measurements matter.
Could Location, Timing and Exposure Explain the Different Results?
It would be tempting to choose whichever study supports the conclusion we already prefer.
If we want reassurance, we could point to studies that found virtually nothing.
If we want alarm, we could point to Brazilian studies reporting glyphosate in every sample.
Neither approach tells us what is actually happening.
The studies involved different populations, different countries, different agricultural environments, different analytical methods and different detection limits.
Potential sources and timing of exposure may also have been very different.
So what if the apparently contradictory findings are themselves important?
What if glyphosate doesn’t substantially bioaccumulate in breastmilk, but can nevertheless appear there following exposure?
What if timing matters?
What if geography matters?
What if living near agricultural spraying matters?
What if occupational exposure matters?
What if diet or drinking water contributes?
What if a partner who works with agricultural chemicals inadvertently brings residues home on clothing, footwear or equipment?
And what if mothers with no obvious connection to agriculture can also be exposed?
These are questions.
The existing breastmilk research isn’t sufficient to answer them confidently.
And that is precisely the point.
What Could Glyphosate in Breastmilk Mean for a Baby?
This is where the discussion needs particular care.
Detecting glyphosate does not prove that the concentration detected will harm a breastfed baby.
The human studies discussed above do not demonstrate that glyphosate in breastmilk causes disease in infants.
Regulatory assessments typically evaluate glyphosate exposure against established health-based guidance values such as acceptable daily intakes.
That deserves to be stated plainly.
But it doesn’t make every scientific question disappear.
A newborn isn’t simply a small adult.
During pregnancy, infancy and early childhood, organs, the brain, endocrine systems, immune systems and metabolic processes are undergoing extraordinary development.
Researchers have therefore investigated glyphosate and glyphosate-based herbicide exposure during pregnancy and lactation in animal models.
Some experiments have reported neurobehavioural, oxidative-stress, reproductive and other biological changes in offspring following developmental exposure.
Animal studies cannot tell us that a breastfed human baby exposed to trace concentrations will experience the same effects.
The doses, exposure routes, formulations and species can be very different.
But these studies establish something much more modest — and important:
Developmental exposure is a legitimate scientific question worth investigating.
So where are the large human studies?
Are Regulatory Glyphosate Exposure Limits Enough to Protect Babies?
Regulatory toxicology frequently evaluates exposure by comparing an estimated dose with an acceptable daily intake or another health-based guidance value.
That’s useful.
But consider what question it answers.
It asks, essentially:
Is this estimated exposure below a level regulators currently consider acceptable?
There is another question a mother might reasonably ask:
Why should my newborn baby be exposed to a herbicide at all if that exposure can reasonably be prevented?
Those are not the same question.
The first is a regulatory risk question.
The second is a public-health and precaution question.
Both deserve an answer.
How Much Research Has Been Done on Glyphosate in Human Breastmilk?
This may be the most uncomfortable question of all.
Glyphosate isn’t a newly discovered chemical.
It has been used for more than half a century.
It is used in agriculture, horticulture, forestry and weed control around public and private spaces.
Human exposure has repeatedly been demonstrated through biomonitoring studies, particularly through urine.
Yet when we ask one of the simplest possible questions —
Does glyphosate reach human breastmilk?
— we find studies involving 41 women, 67 women, 74 women, 100 women.
Where are the studies involving thousands?
Where are the repeated samples from the same women before, during and after known exposure?
Where are the studies comparing mothers living beside sprayed crops with mothers living far from agricultural areas?
Where are the studies examining spray operators and their families?
Where are the studies carefully comparing diet, occupation, season and location?
Where is the research establishing how quickly glyphosate might appear in breastmilk following exposure — and how quickly it disappears?
And where are the long-term studies following children whose mothers had measured glyphosate and AMPA concentrations during pregnancy and breastfeeding?
For a chemical used on such an enormous scale, why are we still asking such basic questions?
Has Glyphosate Been Tested in Breastmilk in New Zealand?
That brings the question home.
Glyphosate is widely used in New Zealand.
It is used on farms, orchards and vineyards, around infrastructure, and for weed management in public and private spaces.
We also know that glyphosate residues can occur in food. No More Glyphosate NZ’s independent laboratory testing has repeatedly demonstrated that reality in products sold here.
So what do we know about breastfeeding mothers in New Zealand?
Do mothers living in our agricultural and horticultural regions have detectable glyphosate in their breastmilk?
What about women working on farms, orchards, vineyards or market gardens where herbicides are routinely used?
What about the partners of farmers and spray contractors who may handle or wash work clothing?
What about mothers living beside land that is regularly sprayed?
And what about mothers living in Auckland, Wellington, Christchurch or Dunedin who have never knowingly handled a glyphosate product in their lives?
Do they have glyphosate in their breastmilk?
AMPA?
Neither?
We don’t appear to know.
And “we don’t know” should never quietly become “there isn’t a problem.”
They are two very different statements.
How Could Breastfeeding Mothers Be Exposed to Glyphosate?
There is something deeply unfair about where responsibility can end up in conversations about environmental contaminants.
A mother doesn’t manufacture glyphosate.
She doesn’t approve it.
She doesn’t determine how much may legally remain in food.
She doesn’t decide where councils spray it.
She doesn’t determine agricultural practices.
She may never have purchased or used a glyphosate-based herbicide herself.
Yet if dietary, occupational or environmental exposure results in glyphosate or its breakdown products entering her body — and potentially her breastmilk — she and her baby are the ones carrying that exposure.
Surely she deserves to know whether it is happening.
And surely society has an obligation to find out.
If Glyphosate Is Found in Breastmilk, Should Mothers Stop Breastfeeding?
If future research establishes that glyphosate or other agricultural chemicals are reaching breastmilk, the conclusion should not be:
Stop breastfeeding.
It should be:
Find the source of the exposure and reduce it.
Breastmilk is not responsible for environmental contamination.
Mothers aren’t responsible for environmental contamination.
And babies certainly aren’t.
The responsibility belongs upstream — with the systems, practices and regulatory decisions that determine what chemicals people are exposed to in the first place.
Glyphosate in New Zealand Breastmilk: If We Don’t Look, We Don’t Know
The international evidence on glyphosate in breastmilk is not neat.
Some of it is reassuring.
Some of it raises difficult questions.
Some studies have detected nothing.
Others have reported glyphosate in every mother tested.
And differences in analytical methodology mean those results cannot simply be compared as though all the studies were identical.
But perhaps the most remarkable finding isn’t that scientists have produced conflicting results about whether glyphosate is present in breastmilk.
It is how little research exists to settle the question.
After more than 50 years of glyphosate use, we should not need to speculate about whether breastfeeding mothers are being exposed, whether that exposure reaches their milk, which mothers are most exposed, or what that could mean for their babies.
We should have the data.
In New Zealand, we apparently don’t.
So perhaps the question isn’t:
“Can anyone prove glyphosate in breastmilk is harming New Zealand babies?”
Perhaps we should start one step earlier.
Is glyphosate getting into the breastmilk being fed to New Zealand babies?
Because before we can understand what an exposure might mean, we first have to establish whether that exposure is happening.
And if we don’t look, we don’t know.
Research on Glyphosate in Human Breastmilk
Brandão LE, Costa RPR, Marandola RF, et al. (2026).
Rapid and Efficient Detection of Glyphosate in Breast Milk Samples Using High-Performance Liquid Chromatography (HPLC). Processes, 14(4), 677. DOI: 10.3390/pr14040677.
The study analysed breastmilk from 100 postpartum women in Tupã, São Paulo, Brazil — 90 urban and 10 rural — and reported glyphosate in all samples using their HPLC screening method. The authors recommend that positive results be confirmed using more sensitive, fully validated techniques such as LC-MS/MS.
Pawlak R, Wooten A, Selim M, Kew KA. (2024).
Reassuring Quantitative Analysis of Glyphosate and Aminomethylphosphonic Acid Levels in Breast Milk Using Liquid Chromatography Mass Spectrometry.
Breastfeeding Medicine, 19(9).
DOI: 10.1089/bfm.2024.0118.
The study analysed breastmilk from 74 US women using liquid chromatography-mass spectrometry with a reported glyphosate detection limit of 1 ng/mL. Glyphosate was detected in one sample and AMPA in an additional seven. The authors concluded that glyphosate detection was rare in this predominantly urban US population and suggested that detection may vary according to place of residence and timing in relation to agricultural application.
Camiccia M, Candiotto LZP, Gaboardi SC, Panis C, Kottiwitz LBM. (2022).
Determination of Glyphosate in Breast Milk of Lactating Women in a Rural Area from Paraná State, Brazil.
Brazilian Journal of Medical and Biological Research, 55, e12194.
DOI: 10.1590/1414-431X2022e12194.
The study examined breastmilk from 67 lactating women — 41 rural and 26 urban — and reported glyphosate in all samples, with a mean concentration of 1.45 µg/L. The study used a commercial enzyme immunoassay.
McGuire MK, McGuire MA, Price WJ, et al. (2016).
Glyphosate and AMPA in Human Milk Are Not Detectable.
American Journal of Clinical Nutrition, 103(5), 1285–1290.
DOI: 10.3945/ajcn.115.126854.
The study analysed milk and urine from 41 lactating women in Idaho and Washington. Neither glyphosate nor AMPA was detected in the human milk samples, although glyphosate was detected in urine from many participants.
Steinborn A, Alder L, Michalski B, et al. (2016).
Determination of Glyphosate Levels in Breast Milk Samples from Germany by LC-MS/MS and GC-MS/MS.
Journal of Agricultural and Food Chemistry, 64(6), 1414–1421.
DOI: 10.1021/acs.jafc.5b05852.
The researchers developed and validated two independent mass-spectrometry methods for determining glyphosate in human breastmilk and did not confirm the high concentrations previously reported by non-peer-reviewed testing.
Bus JS. (2015).
Analysis of Moms Across America Report Suggesting Bioaccumulation of Glyphosate in U.S. Mother’s Breast Milk: Implausibility Based on Inconsistency with Available Body of Glyphosate Animal Toxicokinetic, Human Biomonitoring, and Physico-Chemical Data.
Regulatory Toxicology and Pharmacology, 73(3), 758–764.
DOI: 10.1016/j.yrtph.2015.10.022.
This paper critically examined the original Moms Across America findings and argued that substantial bioaccumulation of glyphosate in breastmilk was inconsistent with available toxicokinetic, biomonitoring and physicochemical evidence.
Gallegos CE, Bartos M, Bras C, Gumilar F, Antonelli MC, Minetti A. (2016).
Exposure to a Glyphosate-Based Herbicide During Pregnancy and Lactation Induces Neurobehavioral Alterations in Rat Offspring.
NeuroToxicology, 53, 20–28.
DOI: 10.1016/j.neuro.2015.11.015.
This animal study examined developmental exposure to a glyphosate-based herbicide during pregnancy and lactation and reported neurobehavioural alterations in offspring. Animal findings cannot be directly extrapolated to trace exposure through human breastmilk but contribute to the broader scientific investigation of developmental exposure.
Image Source & Attribution
We’re grateful to the talented photographers and designers whose work enhances our content. The feature image on this page is by Lopolo.
