A trial of a fingerprint recording system (FPRS) to precisely record class attendance was carried out with third-year students at our medical school. Along with this new system, a conventional card method was used to confirm attendance. The FPRS consists of a fingerprint recording unit, a communication control unit, and a personal computer. Each student was required to register a fingerprint before the trial started. Attendance was recorded by placing the registered finger on the fingerprint recording unit. The recording, comprising each student's name and the time, was stored in the communication control unit and transferred to the personal computer through the RS-232 cable. The RS-232 cable can be replaced with an Internet environment. In the trial performed in a pharmacology course, 95 medical students were enrolled, and the total student-lecture number was 3008. On 149 occasions students forgot to record their fingerprints. One student reported that his fingerprint was not recognized by the recording unit. However, the problem was solved when he registered his fingerprint again. The attendance recorded by the conventional method confirmed that recorded by the FPRS. After completion of the trial, 26 students volunteered to give their impressions about the FPRS. They were divided on the basis of their impressions ( “positive”, “negative”, and “conditionally positive” ) into three groups of approximately equal number. The observations suggest that the impression of the FPRS varies among students and that this variation is unrelated to either student performance or the attendance rate. This trial found no major technical problems in recording attendance with the FPRS.