Syntactic and semantic cueing were examined as used in oral reading by 30 college freshmen, half of whom were higher in vocabulary on the Nelson-Denny Reading Test and half of whom were higher in comprehension scores on the same test. The subjects were selected from 168 freshmen, all of whom were poor readers and spoke predominantly English. All subjects read 150 randomly ordered expressions of three types: sentences, semigrammatical strings (semantically anomalous), and ungrammatical strings. Scoring was based on the number of miscues made on content words and on the number of corrections made in each category. Miscue analysis indicated that higher comprehension subjects used both semantics and syntax in oral reading, while higher vocabulary subjects used syntax but did not seem to use semantics. Corrections analysis yielded no differences in use of syntax and semantics. Higher comprehension subjects had more miscues than the higher vocabulary group. Possible interpretations of these findings are discussed with reference to Goodman's model of reading. Tables and references are included. (MS)