The purpose of this study was to examine the extent and manner in which teachers use mathematics textbooks in their daily teaching. Specific attention was given to textbook implementation differences related to two types of mathematics textbooks where the mathematical content is organized differently. Only schools where a dual curricular option program in which students had a free choice between enrolling in an integrated mathematics program or enrolling in subject-specific courses (Algebra 1, Geometry, and Algebra 2) were included in the study. In order to measure implementation of both curriculum types, each teacher completed a Textbook Use Diary and a Table of Contents Record. Data from these instruments were compared to recommendations offered by authors in the teacher editions of the textbooks used in the respective classrooms. Additional implementation data were gathered via Classroom Visit Protocols, which were using during classroom visits to document implementation with regard to content fidelity and presentation fidelity dimensions. Results from three consecutive years of data collection indicate that teacher use of specific lessons in textbooks varies greatly from textbook author recommendations. Teachers using subject-specific curricula spent fewer than the recommended number of days on lessons reported on in their diaries and often assigned fewer homework problems than the textbook authors consider optimal. Similarly, teachers using integrated curricula also assigned less homework than recommended, but conversely spend more than the recommended amount of time on lessons. With regard to teachers' textbook use across the entire textbook, teachers on average covered approximately 70% of the content in the textbook, but content covered often included the use of supplementary materials. Teachers of integrated textbooks tended to cover a smaller proportion of their textbooks than teachers of subject-specific textbooks. Teachers of integrated curriculum vary more in their coverage of the content of the textbook than do subject-specific teachers, but when they do cover the content, they tend to follow the book more closely than those teaching subject-specific curricula. Classroom visits revealed that the content of the lessons was primarily attributable to the textbook; however, the manner in which lessons were taught was less consistent with the author's expectation than was the content of the lessons taught. Given that during three successive school years as much as one-third of the mathematical content addressed included the use of supplementary materials, measures of the impact of particular curricula on student learning should take supplementation into account as part of data analysis. (Contains 1 footnote, 4 tables, and 7 figures.) [This paper is based on research conducted as part of the Comparing Options in Secondary Mathematics: Investigating Curriculum (COSMIC) project, a research study supported by the National Science Foundation.]