Scholars have stated the particularities of the language used in specialized discourse but little attention has been so far paid to the role derivational morphology may play in register variation. The present research makes a contribution to the study of word-formation in scientific registers by means of a corpus-based approach to the productivity of 14 suffixes in two scientific English registers, i.e., computer science and medicine. In order to empirically examine the productivity of the suffixes in each register, types, tokens and hapaxes ratio were used. Results obtained were then contrasted with the presence of the same suffixes in the written language wordlist of the British National Corpus (BNC). The study shows that suffixes are a productive word-formation resource in scientific registers and that their productivity differs in the registers under study. Findings ranked higher productivity of abstract noun-forming suffixes such as -ity, -ion and -ness in scientific registers than in the BNC. The suffix –ize reached values in the scientific corpora highly over the ranking drawn from the BNC. On the contrary, the BNC yielded an outstanding productivity rate of –free and -like, suffixes which proved to be fully unproductive in the scientific registers under study. Montero Fleta, MB. (2011). Suffixes in word-formation processes in scientific English. LSP Journal - Language for special purposes, professional communication, knowledge management and cognition. 2(2):4-14. http://hdl.handle.net/10251/46766