The corpus was developed as a linguistic resource for Automatic Summarization research and his relation with different issues to engage studies on the discourse treatment. Summ-it consists of fifty texts from Science domain extracted from Science section of Brazilian daily newspaper Folha de Sã...