TRANSFORMATIONS OF BUFFALO GNATS. 29 
In 1905 Dr. Louis Sambon advanced the theory that pellagra might 
be transmitted by a blood-sucking fly of the genus Simulium. An 
accurate knowledge of the biology of the genus is very necessary in 
order to prove or disprove such a theory. It must be proved that 
Simulium not only bites in very large numbers, in order that a suffi- 
cient percentage of specimens will become infected which will live 
and become capable of transmitting pellagra, but it must also be 
proved that they will bite man after they have become infected, 
since hereditary transmission is probably entirely out of the question. 
Simulium exists in many places in large numbers, and if it should be 
proved that it normally requires a meal of the blood of vertebrates 
before it can fully develop the ovaries, then the chances that it may 
become infective are greatly increased. 
Evidence on these points may be obtained in the following three 
ways: 
(1) By rearing adult females from the immature stages, and then 
allowing them to engorge, oviposit, and reengorge. This method 
was tried a number of times, but with no success, for the fly will not 
readily engorge while in captivity, as was noted in the discussion of 
the feeding habits of the adult. 
(2) By the capture of adults actually feeding on mammals, and by 
inducing them to oviposit and then to reengorge. This method also 
was unsuccessful for the same reasons. 
(3) By dissections. If it could be proved that when a Simulium, 
which has once engorged on blood and oviposited, is still in a condi- 
tion favorable to a second oviposition — that is, if rudimentary eggs 
are present in the ovaries awaiting only a second blood meal for their 
successful development — then there would be a more definite basis 
for a theory of disease transmission as opposed to the theory of merely 
a close coincidence in the distribution of pellagra and Simulium. 
This is the method adopted by the writer and is the one discussed in 
the following pages. The experiments were concluded in Spartan- 
burg County, S. C, a country very favorable for the production of 
Simulium, being hilly and with a network of small streams, rapid 
and especially suitable for the development of the immature stages. 
The experiments on the actual biting and feeding habits were carried 
on more or less continuously from June 13 to September 16. The ma- 
terial for these experiments consisted of specimens of 8. venustum; 
a good series of reared specimens (about 30) ; a number of females 
found engaged in oviposition; about 300 females all of which were 
taken in the act of engorging on mammals. Of the last named only 
about 90 were successfully dissected, as the engorged adults after 
death became internally disintegrated and hardened after an hour 
or so, despite every precaution. Many also died between midnight 
and 7 a. m., or while they were being transported from the place of 
