Objectives. To evaluate the pathologic characteristics of clinical Stage T1c (nonpalpable, prostate-specific antigen [PSA]-detected) prostate cancers detected in the 2.6 to 4.0-ng/mL PSA range and compare them with Stage T1c cancers concurrently detected in the 4.1 to 10.0-ng/mL PSA range. All cancers were detected in a prostate cancer screening study. Methods. We studied 94 patients with clinical Stage T1c prostate cancer diagnosed by four or six-sector ultrasound-guided needle biopsy who underwent radical prostatectomy between June 1995 and December 1996. We included all men whose prostatectomy specimens were processed with complete embedding of all prostatic tissue. Of these, 42 had a PSA level of 2.6 to 4.0 ng/mL and 52 a PSA level 4.1 to 10.0 ng/mL at the time of cancer detection. We determined the tumor volume by complete embedding and grid morphometry, pathologic stage, Gleason sum, and surgical margin status and compared the cancer volume and pathologic tumor stages for each group. Results. Men with cancer detected at the 2.6 to 4.0 ng/mL PSA range had significantly smaller cancer volumes (1.1 ± 1.1 cm3 versus 1.8 ± 1.5 cm3, P = 0.02); however, no difference was found in the proportion (11.9% versus 11.5%, P = 0.9, and 23.8% versus 26.9%, P = 0.7, respectively) of tumors that met previously published criteria of "clinically insignificant" (organ confined, less than 0.2 cm3 tumor volume, Gleason sum 6 or less) or "clinically unimportant" (organ confined, less than 0.5 cm3 tumor volume, and Gleason sum 6 or less) tumors. Using the lower PSA cutoff point resulted in the detection of a significantly higher percentage of organ-confined tumors (88% versus 63%, P = 0.01). Conclusions. The use of a 2.6-ng/mL PSA threshold for screening resulted in the more frequent detection of small, organ-confined tumors without overdetecting possibly clinically insignificant ones.