Harmonization of terminology in developmental toxicology is a prerequisite to ensure a better risk assessment of chemicals. As part of an international effort of the International Programme on Chemical Safety (IPCS) to harmonize terminology in developmental toxicology, workshops have taken place in Berlin since 1995. This publication reports the main outcomes of the Fifth and Sixth Berlin Workshops held in 2005 and 2007, respectively. The objective of the Fifth workshop was to discuss a draft international proposal for updating the glossary of descriptive terms for fetal abnormalities put forward by Wise et al. [Wise LD, et al. Terminology of developmental abnormalities in common laboratory mammals (version 1). Teratology 1997;55:249-92]. The participants were asked to classify the new external, visceral and skeletal observations included within this new version 2 of Terminology of Developmental Abnormalities in common Laboratory Mammals according to the two-category scheme (malformation and variation) agreed at previous Berlin workshops. The discussions held during the Sixth Workshop were mainly focused on the causes of uncertainty and low agreement regarding classification of some fetal observations as malformations or variations. Lack of precision in descriptive terms and insufficient knowledge of the postnatal consequences of fetal observations had been identified as major causes of uncertainty and lower agreement among evaluators regarding the classification of "grey zone anomalies", i.e. abnormalities that do not fit readily into one of the two categories (malformation or variation). Imprecise anatomical terms, observation terms that are too broad, lack of information on severity and the use of different terms for the same change or different severities of the same change, were found to be the main reasons that descriptive terms are often not sufficiently precise to allow accurate classification of findings. It was agreed that provision of additional information, including sub-location within the affected structure, more detailed description of the nature of the change, in conjunction with presentation of photographs wherever possible, and a grading for severity would make descriptive terms more precise, thereby reducing misclassifications. A better knowledge of the adversity and postnatal consequences of fetal observations was considered as the key issue for achieving a substantial reduction in the number of misclassifications and grey zone anomalies. The urgent need for additional research along this line as a prerequisite for a better risk assessment was emphasized by the participants.