The classical Nine Mile strain of Rickettsia burneti, which had been maintained in guinea pigs since first isolation, was adapted to growth in yolk sacs. Despite abundant rickettsiae, antigens prepared from each of the first seven yolk sac passages failed to react in complement fixation tests with homologous Q fever antisera. The antisera did react, however, with antigens prepared from the eighth and subsequent passages. This change in behavior during egg adaptation, which resembles that found previously with freshly isolated strains, is termed "phase variation". Rickettsiae in an early stage of egg adaptation which fail to fix complement with antisera are called Phase 1; after further egg adaptation and the development of ability to react with antiserum, they are termed Phase 2. Strains in Phase 2 rapidly reverted to Phase 1 after inoculation into guinea pigs, mice, or hamsters, and naturally occurring strains from a sheep's placenta, cows' milk, and ticks were also found to be in Phase 1. It is possible that all strains used for preparing stock complement fixing antigens are laboratory variants in Phase 2 obtained through multiple yolk sac passages.

To characterize IgM and IgG antibody responses in Rocky Mountain spotted fever (RMSF), a microtiter enzyme-linked immunosorbent assay (ELISA) using density gradient-purified Rickettsia rickettsii as antigen was developed. Sera of vaccinated individuals and patients with RMSF were tested by ELISA and by indirect fluorescent antibody (IFA) tests. Diagnostic agreement between ELISA and the IFA test was 76% and 52% for IgG and IgM antibody, respectively. Diagnostic agreement between the ELISA for IgG antibody and the IFA test for total immunoglobulins was 84%. The ELISAs for IgM and IgG antibody were as specific (100%) and as sensitive (100%) as the IFA test (83%-100%) in detecting antibody increases in paired sera from persons with RMSF and were superior to the IFA test in detecting seroconversions in vaccinees. The ELISA also detected antibodies in a single convalescent-phase serum with sensitivity and reliability. The ELISA for IgG antibody is appropriate for seroepidemiology and serodiagnosis since it permits measurement of antibody at a single dilution of serum up to a year after illness.

Further studies of the phase variation of Rickettsia burneti described by Stoker (12, 13) and Stoker and Fiset (14) are reported. Animals (guinea pigs and rabbits) infected or vaccinated (with killed purified suspensions) with either Phase 1 or Phase 2 of R. burneti (Nine Mile and Christie strains) develop antibodies that are detectable very early (within 21 days after inoculation) when their sera are tested with antigens in Phase 2. Except for the group of guinea pigs vaccinated with killed purified suspensions of Nine Mile Phase 2, all the animals also produce antibodies against the Phase 1 strains but these appear much later than the anti-Phase 2 antibodies. It is believed that R. burneti when reverting from the non-reactive Phase 1 to the reactive Phase 2 through egg adaptation loses some of its surface antigens.

Rickettsia mooseri infection has been studied in syngeneic guinea pigs inoculated intradermally with the objective of developing a model for the study of immune mechanisms. Characterization of infection included the following: a study of replication, dissemination, and clearance of rickettsiae; measurement of the antibody response with different rickettsial antigens and tests; and attempts to measure the cell-mediated immune response using the correlate of delayed-type hypersensitivity skin reactions. Following intradermal inoculation, rickettsiae replicate locally and then spread to the draining lymph nodes and subsequently cause systemic infection. Spread to draining lymph nodes occurred before the appearance of circulating antibody, whereas systemic infection occurred afterwards. Two distinct patterns of acquired resistance developed. The first was marked by a cessation of rickettsial growth within a given organ and the second by a clearance of rickettsiae. The duration of each of these phases differed markedly from one organ to another. Delayed-type hypersensitivity was not demonstrated by skin testing.