That's an interesting thought, but I haven't seen any support for this theory in the literature I've read (but then I'm an amateur lay person). There is a lot of research on the initial outbreaks in Africa caused, it seems, by environmental and transportation changes, particularly railroads, which accelerated the spread of tropical diseases as well as re-use of contaminated needles in poor areas. There is also a lot of research regarding the main "reservoirs" of the disease in Africa. If outbreaks occurred hundreds or thousands of years ago, they must have been limited to small villages that probably didn't communicate with neighboring villages, limiting transmission. So I find it rather doubtful there would be wide-spread immunity in a fraction of the population.
Filoviruses in Bats: Current Knowledge and Future Directions
By Kevin J. Olival 1 and David T. S. Hayman1 February 2014
Viruses 2014, 6(4), 1759-1788; doi:10.3390/v6041759
...Very diverse taxa have been suggested as potential reservoirs for filoviruses over the years, including bats, rodents, arthropods, and plants. In a massive field investigation to find the natural reservoir following the 1995 Kikwit, DRC outbreak over 3000 animals were collected primarily from forest areas near the home of the index case, but no evidence of Ebolavirus was found. The sampling included 78 mammal species, 51 bird species, and 22 reptiles and amphibians species were collected, and 18 species and approximately 1/5 of all the animals collected were bats. However sample sizes per species were low, with only 4 bat species having greater than 20 individuals collected. Swanepoel et al. demonstrated that plants, reptiles, invertebrates and some vertebrates were unlikely reservoirs, because experimentally they were refractory to infections. However the bats they tested were able to survive infection, support replication, and mount an adaptive immune response. Despite years of investigations, it took nearly forty years from the discovery of Marburgvirus in the late 1960s to identify fruit bats as (at least one of) the primary natural reservoir for this virus.
In just the past few years, antibody reactive with Reston ebolavirus and Zaire ebolavirus antigen have been detected in bats from the Philippines, China, Bangladesh, and orangutans from Indonesia (as previously mentioned). Though not conclusive evidence of the presence of these infections, the presence of these or related viruses are not entirely surprising considering the recent discoveries of Marburgvirus and Ebolavirus from congeneric species (Rousettus spp.) in Africa, and considering the large and overlapping geographic ranges for many of these bat species. Rousettus amplexicaudatus bats in the Philippines were found seropositive for Reston ebolavirus and implicating as the potential reservoir host for this virus in Asia.
An Introduction to Ebola: The Virus and the Disease
C. J. Peters and J. W. PetersJournal of Infectious Diseases. (1999) 179 (Supplement 1): ix-xvi. doi: 10.1086/514322
Filoviridae is the only known virus family about which we have such profound ignorance. We do not even understand the maintenance strategies employed in nature by the agents, and we know much less about the resulting diseases, their pathogenesis, and detailed virology.
Biomedical science first encountered the virus family Filoviridae when Marburg virus appeared in 1967. At that time, commercial laboratory workers with a severe and unusual disease were admitted to a hospital in Marburg, Germany.
In the late 1970s, the international community was again startled, this time by the discovery of Ebola virus as the causative agent of major outbreaks of hemorrhagic fever in the Democratic Republic of the Congo and Sudan. International scientific teams that arrived to deal with these highly virulent epidemics found that transmission had largely ceased; however, they could reconstruct considerable data from the survivors. Medical facilities had been closed because of the high death toll among the staff, thus eliminating major centers for dissemination of infection through the use of unsterilized needles and syringes and the lack of barrier-nursing techniques. In contrast, patients in the affected villages were segregated through traditional methods of quarantine, a step that controlled the situation outside the clinics.
After Ebola hemorrhagic fever (EHF) appeared in Africa in 1976–1979, it was not seen again until 1994. Was it “gone” during those 15 years? In one sense, certainly not—it was circulating in its natural reservoir. Was the virus causing sporadic human infections that remained undetected because the patients never contaminated hospitals to produce the savage nosocomial epidemics that brought Ebola virus to medical attention?
Of interest, there was an appreciable seroprevalence among the residents of Kikwit and those of surrounding villages, which was thought to represent temporally distant infections.