Fluorescent in situ hybridization using 12 chromosome enumeration probes (for chromosomes 4, 6, 7, 8, 9, 10, 11, 12, 17, 18, X and Y) was used to evaluate fresh tumor touch preparations from 40 randomly selected radical prostatectomy specimens. Of the tumors 16 (40%) contained chromosomal aneusomies. Chromosome 8 was aneusomic in 9 tumors (23%). Gain of chromosome 7 was observed in 8 tumors (20%). Chromosome 17 was aneusomic in 4 cases, and chromosomes 10, 11, 12, 18 and Y were each aneusomic twice. Loss of chromosome 9 was observed in 1 tumor. Chromosomes 4, 6, and X were never aneusomic. The percentage of monosomy 17 nuclei was 2 to 4 times the amount noted with the other autosomes for tumor and benign tissue. Computer analysis demonstrated that these signals contained twice the signal density and were significantly different (p < 0.0001) than the single diploid chromosome 17 signals. This result is consistent with homologous pairing of chromosome 17 in benign and neoplastic prostate tissue. Anomalies of chromosomes 8 and/or 7 were present in 14 of the 16 cases (88%) aneusomic by fluorescent in situ hybridization. High grade tumors were more likely to be aneuploid on fluorescent in situ hybridization (p < 0.001). Tumors with chromosome 8 aneusomies were of higher stage (p < 0.05). Fluorescent in situ hybridization is more sensitive than flow cytometry for the detection of aneusomy/aneuploidy. The prognostic relevance of these findings will require further investigation.