No dilemma in bioethics is a simple one. The issue of pesticide research readily demonstrates that even characterizing a debate as “two-sided” is often an oversimplification. Today the EPA announced that it will bar data from pesticide studies involving children and pregnant women. Protecting a vulnerable population from risk seems like a winning position. However, the impact of such a decision may, in the end, put these populations at even greater risk. As “safe” limits of pesticide exposure are studied and regulated, lack of data regarding the effects on children and pregnant women may mean that the legally acceptable levels are not really safe for them at all. But deciding to accept research using children and pregnant women is also not a simple matter. One recently proposed study that would look at the effect of pesticide exposure on young children had to be shut down when claims were made that it was targeting low-income, less-educated families and that the incentives were so high as to be considered coercive. Another criticism of the study was that the two year research time frame was not sufficient to find negative long term health effects such as developmental delays, cancers and puberty/hormonal disruptions. The history of pesticide testing reveals other thorny issues including problems involved in finding control groups when examining data obtained from observing the health outcomes of children who were exposed to pesticides in their natural environments.