Abstract
It is shown that pion electroproduction experiments can, in principle, distinguish between a pseudoscalar and a pseudovector pion-nucleon coupling. It is further shown that this is not the case for the corresponding photoproduction experiments.