The following is a commentary on the recently published guidelines for breastfeeding in women with substance use disorders. Delta-9-tetrahydrocannabinol (DTHC) is the major compound in marijuana, but there are many other toxic compounds in marijuana, and it may be laced with harmful adulterants. THC is psychoactive, is distributed into breastmilk and to the brain, and is lipophilic/stored in body fat. Therefore, active, passive and cumulative exposures to the breastfed infant all must be considered. THC can be concentrated in breastmilk up to eight times the level in plasma in heavy users and is absorbed and metabolized by the breastfed infant; metabolites are found in infant stool. Physiologic effects including sedation and poor feeding have been described in infants receiving breastmilk from marijuana-using mothers. Explorations of the effects of marijuana exposure via breastmilk on longer-term child development have led to conflicting results; however, this research was done in a time period in which the potency of marijuana was approximately fourfold less than it is today. Cannabinoid CB1 receptors emerge early in brain development and are the major targets for the action of cannabinoids. They are abundantly expressed in certain brain regions that play key roles in neural development. THC delivered via lactation to the infant may therefore affect the ontogeny of various neurotransmitter systems, leading to changes in neurobiological functioning. THC is present in exhaled breath for 2–4 hours after a single marijuana cigarette, and indirect passive smoking exposures can be significant. Prenatal marijuana exposures should not be discounted in this discussion, as it is important to consider that, for the mother using marijuana during lactation, exposures throughout gestation are likely. Gestational marijuana use is not uncommon: among pregnant women, nearly 4% used marijuana in the past month. THC readily crosses the placenta. Infant neurobehavior is altered in prenatally marijuana-exposed infants, and prenatal cannabis exposure is associated with adolescent behavioral abnormalities as well as increased vulnerability to neuropsychiatric disorders in adulthood, indicating an ‘‘unmasking’’ of earlier deficits. Additional postnatal exposures to the rapidly developing infant brain may compound prenatally acquired deficits. Finally, the identification of the lactating mother as a chronic marijuana user and potentially a person with a substance use disorder in the postnatal period presents additional difficulties. Maternal self-report of marijuana use in the perinatal period is inaccurate. Urine toxicology screening, used by many institutions to identify substance use among women in the perinatal period, is not a useful tool for detecting marijuana use due to its long half-life in urine and relays no information regarding quantification or patterns of use. Chronic marijuana use and dependence are associated with a wide range of psychiatric concerns, making identification important for the provision of services for at-risk dyads. Although there are no data to point to use by women in the perinatal period as indicative of chronic or heavy use, we do know that use of marijuana declines over the course of gestation, making users of marijuana at term likely to be chronic or dependent users. In a large U.S. sample between 2007 and 2012, among past-year marijuana users, 12.8% of nonpregnant women used nearly daily, and 11.4% met criteria for abuse/dependence. These figures jump to 16.2% and 18.1% for pregnant women. The breastfed infant must necessarily be in the company of his or her mother, and people with substance use disorders can have periodically altered sensoriums, chaotic environments, and poor judgment regarding their own safety and the safety of their children, portending immediate physical harm for the infant and potentially long-term developmental harm as well. It is necessary to consider that it may represent a missed opportunity to assist a newly postpartum woman with finding acceptable treatment for a substance use and/or an underlying psychiatric disorder, to provide intensive postnatal services to at-risk dyads, or to detect an infant at risk for physical or developmental harm by minimizing marijuana use detected at the time of delivery. In conclusion, the cumulative evidence above supports that when an empirically derived approach to the issue of maternal marijuana use and lactation is taken, the recent new recommendation by the Academy of Breastfeeding Medicine is both erroneous and disappointing. It is unclear why a recommendation would err on the side of breastfeeding with potentially toxic exposures and other risk factors that could portend shortand long-term infant harm. Perhaps this view reflects a lack of understanding of substance use disorders in general. What is clear is the need for future research to fully establish the risk–benefit of breastfeeding with active