Abstract
A common assumption in the selection of nonhuman animal subjects for research and the approval of research is that, if the risks of a procedure are too great for humans, and if there is a so-called scientific necessity, then it is permissible to use nonhuman animal subjects. I reject the common assumption as neglecting the central ethical issue of the permissibility of using nonhuman animal subjects and as being inconsistent with the principle of justice used in human subjects research ethics. This principle requires that certain classes of individuals not be subjected to a disproportionate share of the burdens or risks of research. I argue for an extension of this principle to nonhuman animal research and show that a prima facie violation of the principle occurs because nonhuman animals bear an overwhelmingly disproportionate share of the risks of research without sufficient justification or reciprocal benefit.