Thalidomide, mainly used for the treatment of leprosy, is a current teratogen in South America, and it is reasonable to assume that at present this situation is affecting many births in underdeveloped countries. Moreover, the potential re‐marketing of thalidomide for the treatment of a large variety of diseases may extend the problem to the developed world. When the drug is available, the control of its intake during early pregnancy is very difficult since most pregnancies are unintended. The ongoing occurrence of thalidomide embryopathy cases went undetected by the ECLAMC, due to several factors: (1) low populational coverage through this monitoring system; (2) pre‐existence of the teratogen with its effects present in both baseline (expected) and monitored (observed) materials; and (3) lack of a defined phenotype to be monitored. Thus, if thalidomide re‐enters the market throughout the world, due to the wide range of new applications, occurrence of phocomelia alone might not be sufficient to detect its effects. By a case‐reference approach, the ECLAMC registered 34 thalidomide embryopathy cases born in South America after 1965 whose birthplaces correspond to endemic areas for leprosy. Phocomelia was found in five of eleven fully described cases. Thus, phocomelia alone is neither specific nor sufficient to serve as a suitable phenotype to survey the teratogenic effects of thalidomide. Therefore, a thalidomide‐like phenotype, defined as any bilateral upper and/or lower limb reduction defect of the preaxial and/or phocomelia types, should be included in the routine surveillance of birth defects in all programmes. Teratology 54:273–277, 1996. © 1997 Wiley‐Liss, Inc.