The purpose of this study is to examine the interactions among measurement theories, writing theories, and writing assessments in the United States from an historical perspective. The assessment of writing provides a useful framework for examining how theories influence, and in some cases fail to influence actual practice. Two research traditions are described to classify measurement theories (test-score and scaling), and three research traditions are proposed for classifying writing theories (form, idea and content, and sociocultural context). The results of this study trace the impact of measurement and writing traditions on writing assessment practices within selected time periods during the 20th century in the United States. One of the major findings of this historical analysis is that measurement theory has had a strong influence on writing assessments, while writing theory has had minimal influence on writing assessments. We also found support for the idea that a new discipline of writing assessment has emerged. This new discipline combines multiple fields including the writing, composition, and measurement communities of scholars, and it has the potential to set the stage for the future of writing assessment in the 21st century. (Contains 2 tables, 2 figures, and 2 displays.)