Suspensions of fertilized eggs of Toxocara canis were mixed with 2% neutral formalin and preserved at 4 degrees C. When, after storage for 0, 12, 18, 21 and 24 months, samples of the eggs were incubated at 30 degrees C for 12 days, 96.8%, 92.6%, 74.1%, 51.0% and 19.3% of the eggs in the samples were found to embryonate. The embryonated eggs produced from the fertilized eggs preserved (in 2% neutral formalin at 4 degrees C) for 0, 12, 18 and 21 months were then tested for their infectivity to BALB/c mice, each mouse being given 800 embryonated eggs. The numbers of larvae recovered from the mice and the sites from which they were recovered, 2 or 14 days post-infection, appeared unaffected by the length of storage of the eggs. The infected mice all had similar eosinophil counts in their peripheral blood and similar serum titres of Toxocara-specific IgM and IgG antibodies, and cultures of their spleen cells produced similar amounts of interleukin-4, interleukin-5 and interferon-gamma when stimulated with concanavalin A. The results of SDS-PAGE indicated that egg preservation for at least 21 months had no effect on the excretory-secretory antigens in samples of medium from cultures of infective larvae released from the eggs. In summary, at least 50% of the fertilized eggs preserved in 2% neutral formalin at 4 degrees C for 21 months could fully embryonate and then had the same infectivity and antigenicity as embryonated fresh eggs.