44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 134 
and Palmer (1955) observed seasonal fluctuations in populations of 
two species of Periplaneta in privies in southwestern Georgia. 
The evidence for the transmission of infectious agents by both flies 
and cockroaches is largely circumstantial because most of the evidence 
is indirect. Rarely is either insect the only means of transmitting a 
disease agent, except when serving as an obligate intermediate host 
for an endoparasite (e.g., Pycnoscelus surinamensis as host of the 
chicken eyeworm). When other means of transmission are present 
it is difficult, if not impossible, to prove conclusively that specific out- 
breaks of disease were caused by pathogens transmitted by insects. 
Lindsay and Scudder (1956) cited only a few instances of proven 
correlations between disease morbidity and fly populations, but the 
experiments they discussed are highly convincing. It is obvious that 
closely controlled experiments are needed to prove the relationship 
between mechanical vectors and the morbidity of specific diseases. 
It is not our thesis that filth-bearing insects are the most important 
means of disseminating enteric disease organisms. We recognize that 
all sources of pathogens, including insects, play greater or lesser roles 
in epidemiology. Sabin (1951) has stressed that “. . . one cannot 
apply “unitarian’ epidemiological hypotheses and concepts in dealing 
with an infectious agent that is predominantly stool-borne.” This 
statement applies with equal validity to the mechanical transmission 
of other infectious agents as well. 
There is sufficient direct and indirect evidence to warrant fur- 
ther study of the cockroach as a vector of disease and to initiate 
stronger measures designed to exterminate this insect, especially 
in hospitals and public eating establishments. 
JANSSEN and WEDBERG (1952). 
XV. CONCLUSIONS 
The existing evidence, which is presented in detail in the following 
appendices, should be sufficient to convince all but the most skeptical 
that cockroaches are highly dangerous, potential vectors of disease 
agents. At least 18 species of domiciliary cockroaches have been in- 
criminated, naturally or experimentally, in the transmission of in- 
fectious agents or have been claimed to bite man. Most of these 
species will eat the feces of humans and domestic and exhibition ani- 
mals. Several of the commonest species (Periplaneta americana, P. 
australasiae, P. brunnea, P. fuliginosa, Blatta orientalis, and Blattella 
germanica) have been captured repeatedly in sewers, cesspools, septic 
