258 THE EVOLUTION OF LIFE 
originate from no Torulz corpuscle. They seem to 
develop from some minute primordial particles. 
After two samples of the flakes had been abstracted 
with a sterilised pipette, this tube was closed—now, 
of course, containing air—and was put aside in a 
box for three weeks. When examined again after 
this interval, some of the flakes of silica showed 
numerous groups of rather large and very typical 
Micrococci, such as are shown in Fig. 16, though 
the fluid above the deposit remained, as usual, 
perfectly clear. 
Three other experiments previously made with 
the AA solution in airless retorts, which had been 
heated to 100 C. for fifteen minutes and subse- 
quently exposed to light for about two months, 
also showed either Bacteria alone or these mixed 
with a few Torule, in the flakes of silica. The 
experiments made in 1871, in which organisms 
were found, as recorded on an earlier page of this 
Chapter (p. 241), were also made with airless flasks ; 
so that none of the organisms found in either of these 
vessels could have derived carbon from the carbonic 
acid existing in the air, seeing that the great bulk 
of this had been expelled from the tubes, and that 
the organisms were, moreover, found only in the 
silica lying at the bottom of the tubes, and there- 
fore far away from the surface of the fluid. 
