Readability estimates are usually based on measures of word difficulty and measures of sentence difficulty. Word difficulty is measured in two ways: by the structural size and complexity of words or by reference to phonomena of language use, such as word-list frequency or the regularity of spelling patterns. Sentence difficulty is measured only in terms of size or complexity, despite models of the reading process which suggest that readers could use a knowledge of recurring syntactic patterns to economize their scan of the text. This study identified syntactic patterns which occurred in samples of student writing and text and counted how often each pattern recurred. McCall-Crabbs test passages were classified for the commonness of their sentence patterns according to these counts. The commonness of the patterns was significantly associated with the readability of the passages, with other factors controlled. The results bear implications for measuring readability and designing instruction. (Author)