Loss of telomere repeat sequences occurs after each cell division and telomere shortening has been implicated in cellular senescence. The measurement of telomere length might therefore assess the lifespan of a cell. The aim of this study was to set up and validate a technique enabling the assessment of telomere length on tissue sections. Quantitative fluorescence in situ hybridization (Q-FISH) with telomeric probes was performed on smears and sections from cell preparations or human tissues. The mean fluorescence intensity of telomere spots (FI/spot) was automatically quantified by image analysis. Telomeric restriction fragment (TRF) length was assessed by Southern blotting. There was a positive significant correlation between telomere length, as assessed by Q-FISH, and TRF length determined by Southern blotting in corresponding samples (p < 0.01, r = 0.6 for tissue and p < 0.01, r = 0.79 for cells). FI/spot was higher on smears than on sections, but pairwise comparison showed a significant correlation both for cells and for tissues (r = 0.77, p < 0.001 for cells and p ≤ 0.01, r = 0.64 for tissue). Finally, since telomere length is expected to shorten with age, FI/spot was assessed in liver samples according to the age of patients: a negative correlation was demonstrated (r = 0.76, p < 0.01). Inter-assay variation was 7% for Q-FISH performed on tissue sections and 12% on touch preparations. This study shows that Q-FISH can be performed with confidence on fixed frozen tissue sections in order to assess telomere length. It is an easy, accurate, and reproducible in situ method for assessing telomeres in the context of cell type and tissue architecture. Copyright © 2003 John Wiley & Sons, Ltd.