The paper reports the results of a test involving grammaticality judgments of English sentences given to native speakers of Spanish and Mandarin Chinese studying English as a second language. A detailed account is presented of aspects of Spanish and Chinese syntax that might explain the performance of the subjects, and an appeal is made for a more thorough search for and description of the areas of syntax in which language transfer is most likely to take place, based upon the study of acquisition data involving many different L1s and L2s. The results suggest that interference can come from sources not specified in traditional contrastive analyses and that researchers therefore must not restrict themselves to testing these analyses.