In her introduction to this conversation on hormones, Sandra Bärnreuther conceives of hormones as a set of social relations rather than as things in and of themselves. Nayantara Sheoran Appleton supports this argument in her discussion of birth control pills. As a scientist, I could not agree more. I study plastic pollution, which involves industrial chemicals that either exist within or are attracted to plastics called endocrine-disrupting compounds (EDCs). EDCs include Bisphenol A (BPA) and Polychlorinated Biphenyls (PCBs), as well as hundreds of other pesticides, plasticizers, and industrial chemicals, most of which come to us through consumer goods. EDCs’ relation to bodies is not like that of poisons, which break down cellular functions, but like that of relatives that act akin to hormones. EDCs make things work in the body though similar mechanisms as, but in very different ways than, hormones that bodies produce.

The controversies that Appleton outlines when discussing contraceptive pills in terms of sex and gender, geography, and bodily control also crop up in scientific work on EDCs. EDCs impact the genetic materials being laid down during fetal development, and as many EDCs mimic estrogens, they also tend to disproportionately affect females. The case has also been made that EDCs affect people in different geographies unevenly. The World Health Organization’s “state of the science” report shows that diseases linked to EDCs such as endometrial, ovarian, testicular, and prostate cancer are higher in high-GDP nations flush with consumer goods that leach and offgas EDCs than in low-GDP nations, even when correcting for differential access to health care.

These geographic differences led an American colleague to ask me to partner on a project that compared rates of EDC exposure of populations in the northeastern United States to those of populations in Newfoundland and Labrador, in northeastern Canada where I live and work. Her hypothesis is that Newfoundlanders would be in a more “pristine” exposure state than residents of the northeastern United States, as Newfoundland is consistently ranked as one of the poorest provinces in Canada, has very little manufacturing activity, and is geographically remote.

I brought this proposal to my introductory undergraduate statistics class last semester, a class in which most students are from Newfoundland and Labrador. “But what about the mines?” they asked. “And the methylmercury poisoning from damming on Lake Melville?” “And isn’t Labrador a hot spot for PCBs, and that’s why we can’t eat certain seal organs anymore?” Valid points. The “pristine” Arctic, after all, is one of the most contaminated areas on Earth—this because of a phenomenon called the Arctic Paradox, whereby global air currents push atmospheric and water-borne contaminants to the north. Arctic ecosystems are some of the most contaminated on Earth, particularly in terms of marine food webs, since many of the EDCs that make their way north can accumulate in animals and magnify up the food web (see Cone 2005). This means that people who rely on wild-caught food in the North carry some of the highest EDC body burdens in the world. Not so pristine after all.

My students and I set out to do our own study. There is a noninvasive way to monitor EDC exposure in a population: when a male fetus is exposed to EDCs that mimic estrogen, the ratio between the pointer finger (2D) and ring finger (4D) increases. People whose sex assigned at birth is female tend to have an average 2D:4D digit ratio of 1.00 mm. Those assigned male tend to have an average ratio of 0.98 mm. As male fetuses are exposed to estrogen mimics, the average 2D:4D ratio on their right hand increases to closer to 1.0mm. The phenomenon is called feminization (at least they did not call it hysteria). The final class project was to investigate whether the population of Newfoundland had been exposed to EDCs by scanning, measuring, and comparing a few hundred 2D:4D ratios of people who were carried to term in the region.

The study is designed to compare digit ratios in both sexes before and after a specific point in Newfoundland and Labrador’s history to see if there is a “feminizing” trend. Newfoundland did not join Canada until 1949 and lacked provincial-scale infrastructure until around this time. Most students chose 1965 as the date they thought signified an increased exposure to EDCs, as this was when the Trans-Canada Highway was completed in the province, allowing goods to travel easily for the first time. It was also the end date for the massive (forced) relocation of small, remote rural communities into so-called hub communities, which allowed the provincial government to centralize health care, education, groceries, and other services. Another group of students chose 1992, the date when the cod fishery collapsed, which meant that people had to transition from depending on local, wild-caught fish to packaged, store-bought food. Over the course of the research, students had no problem identifying populations that were most at risk: Inuit people, remote rural Newfoundlanders, people with low or no incomes, military, and mining workers.

Hormone disruption is always uneven, and that unevenness is always based on difference, as Appleton shows in her piece. My students found myriad ways in which location, class, diet, occupation, and Indigenous or settler group identification meant differential exposure to EDCs. The class’s final project was inconclusive in terms of identifying a digit ratio trend over time because we used convenience sampling and our sample had a massive number of eighteen- to twenty-two-year-olds that skewed the data beyond the realm of validity. But the lessons of difference and justice (and statistical analysis, one hopes) stayed with them.

Not all effects of EDCs are harmful. After all, we were measuring changes in 2D:4D digit ratios, which is not a form of harm. Some sharp students in the class generalized this lesson: “Doesn't feminization imply a negative change if it is happening to male digit ratios? That's misleading.” Indeed it is. It echoes Appleton’s finding that discourses around hormones focus disproportionately on controlling the alleged wildness of feminine bodies (whether male or female). The next time I teach the course, therefore, we might come up with different descriptions for the effects of EDCs on the body following a discussion about the gender bias of science (our readings might include Martin 1991, Di Chiro 2010, Haraway 2013, and Allen 2007).

The last section of our class ethics discussion was about consent and the ability for participants to withdraw from the study if they so choose. One student raised her hand: “We can get their consent for having their hand scanned, but what do we say if they don’t want to consent to exposure to EDCs?” Students in Newfoundland are brilliant.