It would have been nice to omit the parenthetical cliché-"more research is needed"-from the title of this article, but it is difficult to do for most pesticides. Limited research on pesticides in humans is a major problem, particularly the lack of large cohort studies with valid exposure assessment. Research on glyphosate (the main active ingredient in Roundup ® , its best-known commercial formulation) is especially complicated. Regulatory and legal aspects specific to glyphosate have been a severe additional impediment for primary and postmarket research and control of exposure to this herbicide, which is the most commonly used globally (IARC 2015).In this issue of Environmental Health Perspectives (EHP), Silver et al. ( 2021) report an association of prenatal exposure to glyphosate and its environmental degradate [aminomethylphosphonic acid (AMPA)] around the 28th week of pregnancy, with a 30-70% increased risk of preterm birth. Exposures earlier in pregnancy were not associated with an increased preterm birth risk. Glyphosate urine concentrations were slightly lower in this population compared with those in other general population studies. This was a small case-control study nested within the Puerto Rico test site for Exploring Contamination Threats cohort of pregnant mothers, which evaluated exposures using biomarkers in urine. EHP and many other journals tend to publish larger studies that would have more precision. It happens, however, that this study of around 250 mothers is the largest available among studies using biomarkers for glyphosate exposure that evaluate a major health outcome (in this case, preterm birth).The authors evaluated preterm birth for two reasons. First, experimental evidence and animal studies indicate that this is a plausible adverse health outcome related to glyphosate exposure. Second, preterm birth is a major, but poorly understood, public health issue that contributes to potential long-term health effects. The study's strengths include its design (a nested case-control study within a general population cohort) and exposure assessment based on measurement of glyphosate and AMPA in urine. Its main limitation was size. The study is important, and the findings indicate potential reproductive health effects of low-level glyphosate exposures in the general population that are relevant globally. However, it certainly is far from conclusive. Perhaps the most important message from this study is that, for the most commonly used herbicide globally, the most conclusive