As material objects, texts are as much the product of visual semiotic codes as of linguistic ones. And throughout history, verbal texts have been combined with nonverbal, visual modes of presenting information, taking a stance toward information and readers, and organizing parts into wholes. The major challenge to creating multimodal texts in the near future will be a lack of multimedial literacy. A more fundamental understanding of existing cultural conventions in communities for combining verbal and nonverbal elements in multimedial texts is needed. To understand how meaning is made simultaneously in several semiotic modalities, common features of all semiotic systems must be identified, i.e., the presentational, the orientational, and the organizational features. Scientific and technical texts have long preserved a tradition of incorporating nonverbal visual-graphic elements as integral and normal parts of their genres. What it means to "read" a text of this kind depends on the literacy practice involved; that is, on the cultural activity as part of which meaning is being made with this text. (Contains 31 references.) (NH)