The results of the Mathematics Skills Improvement Project used with about 2,000 low-achieving students in grades three through six are reported. The services of 35 mathematics consultants were provided for 36 schools. Data obtained (using the California Test of Basic Skills) on pupil achievement for one year in the project indicated that project participants achieved significantly greater gains than did comparable pupils not participating in the project. However, data obtained on pupil achievement for two years of service in the project indicated no significant differences in achievement. Implementation of process objectives of the project in relation to the extent of services to pupils and communication with classroom teachers and pupils' families are discussed briefly. (DT)