The detection of t(14;18)(q32;q21) is advisable for the diagnosis of follicular lymphoma (FL). In 51 patients with FL, we evaluated the applicability and sensitivity of interphase fluorescence in situ hybridization (FISH) and polymerase chain reaction (PCR) using commercially available reagents. In 23 patients, only a formalin-fixed lymph node was available. In 28 patients, both frozen and formalin-fixed lymph nodes were evaluated. Fluorescence in situ hybridization was found to be 100% applicable whatever the material type. With the use of IGH-BCL2 dual-fusion, dual-color probes, t(14;18) translocation was detected in 47 (92%) of 51 FL cases with concordant results between isolated nuclei (n = 41) and frozen cytologic imprints (n = 28). Twenty-two IGH-BCL2-positive cases were also studied on fixed sections with BCL2 split signal probes showing a BCL2 split in all. Conversely, no BCL2 split was observed in IGH-BCL2-negative cases (n = 4). Owing to DNA degradation as assessed by the failure of control genes amplification, the applicability of PCR was found to be 76% in fixed lymph nodes (n = 51). After exclusion of the 12 noninformative cases, the BIOMED-2 protocol allowed the detection of an IGH-BCL2 fusion in 25 (64%) of 39 fixed specimens with 11 PCR-negative (31%) of 36 FISH-positive cases. Even on frozen material with 100% applicability, the amplification of a BCL2-JH breakpoint was achieved in only 20 (71%) of 28 cases with 5 PCR-negative (20%) out of 25 FISH-positive cases. Therefore, FISH was found superior to PCR (using BIOMED-2 protocol) in detecting IGH-BCL2 fusion. Finally, FISH individualized 4 IGH-BCL2-negative FL cases without specific histopathologic features. With the use of split signal DNA probes, 1 case showed a trisomy of the BCL2 locus and another displayed BCL6 and IGH breakpoints that would suggest a t(3;14). Whether such IGH-BCL2-negative cases are characterized by alternative oncogenetic pathways remains to be determined.