the toruloid form of the organism were observed to sprout into 
beautitul filamentous fungi, and these aguin, as the fliid became 
vitiated by the growing fungus, were found to reproduce as 
coni ‘ia the spherical torulvid cells. 
Among other media inoculated with this Oidium was a solution 
of albumen obtained by treating a fresh-laid ezg with a solution 
of carhbolic acid, to destroy any organisms adhering to the shell, 
and then breaking it with carbolised fingers into a ‘‘ heated” 
vessel containing water that had been boiled and allowed to cool 
protected from dust, the solution being afterwards cleared by 
pfgsing it through a boiled filter in a ‘‘ heated” funnel with 
**heated” cover. This fluid had remained during the half-year 
which had since elapsed, free from putrefaction or any other 
change, except where organisms had been introduced, although 
the air had free access to it;* a fict which indicates pretty 
clearly that the putrefaction of eggs, which has been regarded as 
a stumbling-block in the way of the germ theory, must somehow 
or other be brought about by the penetration of ferments through 
the shell and membrane. This, indeed, becomes intelligible 
enough if we admit that Bacteria may be formed from fungi, and 
remember how the filaments ofsome parasitic fungi perforate the 
epidermis of leaves.t In a glass of this albuminous fluid the 
Oidium grew very slowly and feebly, but its development was 
accompanied by a remarkable alteration in the liquid, which, in 
the course of six weeks, changed from the colourless purity of 
spring water to the dark brown, almost black, appearance of 
porter. Yet the dark brown liquid remained perfectly free from 
smell, proving, what the author had long suspected as the result 
of experience in antiseptic surgery, that an albuminous fluid 
may undergo fermentation with odourless products. 
Another experiment given in full detail was performed with milk 
upon the same principle as those with urine and albumen, in 
the hope of removing another stumbling-block in the path of 
the germ theory. For, according to the high authority of Pas- 
teur, milk forms an exception to organic liquids in general, in the 
circumstance that a greater elevation of temperature than the 
boiling-point of water is required to kill Bacteria contained in 
it.{ But the advocates of the theory of spontaneous generation 
reply that any Bacteria present would be certainly killed by 
boiling, and therefore the subsequent appearance of living Bac- 
teria in the boiled milk in Pasteur’s experiments is proof of their 
spontaneous evolution from the chemical constituents of the 
liquid. If, however, by the use of antiseptic means, milk could 
be obtained uncontaminated from the cow, there being no orga- 
nisms to kill, boiling might be dispensed with, and the milk, 
like the unboiled urine, should remain free from organic develop- 
ment or fermentative change, if kept protected in ‘‘heated” 
vessels. Accordingly, five flasks with glass caps, and six test- 
tubes with wider test-tubes to cover them, having been heated, 
and allowed to cool under glass shades in the stable where the 
experiment was performed, the udder and adjacent skin of a 
cow were well washed with a strong watery solution of carbolic 
acid, which was also applied with a small syringe to the outlets 
of the milk ducts, the teat being held in the finger and thumb 
to prevent the entrance of the solution into the udder, and 
a milkman with his sleeves tucked up, and his hands and arms 
washed with the antiseptic lotion, was directed to milk into the 
glasses as their covers were successively raised. The cow did 
not give milk at all freely, and a considerable time was occupied 
in charging the flasks, but the small quantity required for each 
test-tube was got by asingle squirt from the teat, with almost 
momentary exposure, Yet not only in all the flasks but in all 
but one of the test-tubes organisms made their appearance. In 
one of the test-tubes, however, the unboiled milk had hitherto 
(for a quarter of a year) remained entirely unaltered. One such 
success was as clear evidence against the hypothesis of sponta- 
neous evolution of organisms as if all the glasses had remained 
free from them, and their occurrence in the other ten proved a 
most foriunate circumstance. For no two of them were alike in 
the organisms they contained, and in several instances there was 
apparently only some one species unmixed with others, so that 
the opportunity was afforded of studying various different orga- 
nismsjas modified by other media, and as regards any fermen‘a- 
tive influence which they might exert upon those media, Among 
the organisms in the milk glasses were Bacteria of different 
species, to judge from their size and other appearances, as well 
* Dr. Burdon-Sanderson had previously preserved unboiled white of egg 
epcanaed for six months ina ‘“‘ heated” tuve containing air, hermetically 
sealec. 
+ See De Bary, op. cit., page 216. 
t See Annales de Chimie et de Physique, 1862, p. 60, 
NATURE 213 
as numerous kinds of fungi; and when they were introduced 
into a series of glasses of the albuminous liquid before described, 
it was found that while some of the fungi grew in it others re- 
fused to do so, and while Bacteria obtained by adding a drop of 
water to urine throve in the alnuminous fluid, not one of four 
inoculations of Bacteria from four milk glasses was followed by 
any result. Thus was afforded, it is believed, for the first time, 
distinct phy-iological proof of real differences among Bacteria. 
But what was still more unexpected was the fact that when the 
inoculation was practised in a series of glasses of urine, two of 
the Bacteria refused to grow even in that liquid, which had been 
previously regarded as a peculiarly favourable nidus for Bacteric 
development. This fact, besides serving still more clearly to 
differenti ite the various species of Bacteria, suggested a possible 
explanation of the failure of experiments with milk in the hands 
of others, For if organisms thrive in milk which cannot grow 
in urine at all, milk must be a more difficult fluid to work with 
in experiments which aim at excluding organic development. 
Hence it seemed worth while to try again the effect of boiling 
milk, but in doing so to adopt more rigorous precautions 
against the entrance of organic germs. There could be 
little doubt that the organisms which appeared in the various 
milk glasses cf the experiment above related entered during 
the cooling, which though it took place just as in the success- 
ful experiments with urine, led to failure fin the case of 
the milk, partly from the favourable nature of that‘ liquid for 
organic development, and partly no doubt from the atmosphere 
of the stable being much more loaded with organic germs than 
that of the author’s study. The new precautions adopted were 
in the main these. The small wine glasses (liqueur g/asses) into 
which the fluid was to be decanted were covered, together with 
their glass caps, while still very hot, with cotton wool secured 
by fine iron wire tied tightly round below the cap, so as effec- 
tually to filter the air that entered during cooling; after which 
the cotton was carefully removed and the glass placed under a 
small glass shade on a separate piece of plate glass. For heat- 
ing the flask in which the milk was to be boiled a very high 
temperature was requisite to ensure destruction of all life in the 
considerable volume of air which it contained ; and this was 
arranged for by binding asbestos with wire round the junction of 
the neck of the flask and the glass cap, and then roasting the 
flask over a large Bunsen’s burner. The asbestos, which proved 
as good a filter as cotton wool, was removed after cooling, and 
the cap being lifted, a long ‘‘ heated ” funnel was passed quickly 
into the flask and the milk poured in through it after wrapping 
a piece of carbolised rag round the funnel and neck of the flask 
to exclude septic dust : scrupulous care being taken to avoid 
touching the neck of the flask with the moist end of the funnel 
as it was withdrawn. By this means security was obtained 
against the presence of any living organism inside the flask 
except in the fluid at the bottom of the vessel. The cap was 
then re-applied and carbolised cotton wool tied over it to filter 
regurgitant air during the boiling. The necessity for the air- 
filter was made very manifest during ebullition from the great 
tendency of milk to froth, involving the necessity of frequently 
removing the flame, fresh air entering on every such occasion : 
another peculiarity of milk which served further to explain the 
failure of previous experiments. But the efficiency of the means 
employed was shown by the appearance of the flask as exhibited 
to the Society. For although seven weeks had passed since it 
was filled, the milk was seen to be perfectly fluid and with no 
appearance of alteration. 
All trouble occasioned by frothing, involving constant watch- 
ing to prevent the froth from wetting the cotton, was afterwards 
avoided by acting on the suggestion of Mr. Godlee, of University 
College, London, who happened to beassisting the author at the 
time, and immersing the flask in boiling water above the level of 
the liquid, instead of applying the flame directly. This method 
had the further advantage of avoiding any risk of ‘‘ burning ” the 
milk, and also any loss by evaporation, A second flask ‘‘ heated” 
and charged with milk like tne other and similarly covered with 
cotton wool, was kept in this way at 212° F. for an hour, and, 
after cooling, its contents were decanted off into twelve ‘* heated” 
liqueur glasses, and in these it had remained during the seven 
wecks that had since passed perfectly free from change except 
when organisms had been intentionally introduced. To illustrate 
this the author drank, before the Society, the contents of one of 
the uninoculated glasses, which proved perfectly sweet and 
‘ood. 
° It was a curious circumstance that on the morning following 
