Defective Friend spleen focus-forming virus (SFFV) is able to interfere with the ability of its naturally occurring leukemia-inducing helper virus (LLV-F) to induce XC plaque formation in several different strains of mouse embryo cells. This interference has been observed by using two different SFFV preparations, one contained in an NB-tropic stock of Friend virus (FV) complex, and the second present in a C57BL-adapted strain of FV complex containing an associated B-tropic LLV-F helper. The LLV-F in NB-tropic FV complex effectively induced XC plaques in C57BL/6 (Fv-1(bb); Fv-2(rr)) mouse embryo fibroblasts (MEF) only in the absence of coinfecting SFFV, indicating that Fv-2-associated resistance to SFFV-induced focus formation in vivo does not necessarily extend to the restriction of SFFV function(s) in vitro (i.e., in Fv-2(rr) C57BL MEF). SFFV interference appears to be an intracellular event since LLV-F can adsorb onto, penetrate, and rescue defective murine sarcoma virus (MSV) from transformed 3T3FL S(+)L(-) cells with equal efficiency in the presence and absence of SFFV. However, significantly fewer LLV-infected S(+)L(-) cells released LLV-F progeny if SFFV was present. These observations suggest that Friend SFFV may be classified as a defective, interfering (DI) particle. Further support for this conclusion has come from studies designed to investigate two physical properties of defective SFFV particles. SFFV layered onto a 0 to 20% sucrose sedimentation gradient was recovered as a symmetrical band of virus that sedimented more slowly than standard LLV-F particles. Pooled SFFV-containing gradient samples contained visualizable type C virus particles and occasionally small amounts of detectable LLV-F. In an attempt to determine the buoyant density of sedimentation gradient-purified SFFV, pooled SFFV samples were layered onto a 25 to 50% sucrose equilibrium density gradient and were centrifuged to equilibrium. Greater than 50% of the infectious SFFV originally layered onto this gradient was recovered and seen as a narrow symmetrical band with peak SFFV infectivity at a sucrose density of 1.14 g/ml. The observed difference between SFFV and LLV-F buoyant densities appears to be related to an inherent physical property of each virus. Mixtures of these two viruses express the buoyant density of that virus population which is in excess in fabricated FV complexes probably due to the formation of SFFV-LLV aggregates. Finally, gradient-purified SFFV failed to induce XC plaques in MEF and did not function to rescue MSV as expected since SFFV itself is replication defective.