266 Proceedings of Royal Society of Edinburgh. [june 6, 
cavity, or (2) that the living cells of the mucous and other layers 
so act on the bacteria that they are destroyed before they reach the 
body-cavity, or (3) that the cells of the peritoneal fluid effectively 
sterilise the bacteria which succeed in entering, or (4) that the 
bacteria are destroyed as they pass along the lymphatics towards 
the general circulation. The results which follow the injection of 
septic and other solutions into the body-cavity of rabbits are con- 
sidered at length in the Lumleian Lecture given by Dr Burden 
Sanderson in March 1882. From the experiments referred to, it 
was made clear that whenever the solution could not be at once ab- 
sorbed without any irritation being set up, bacteria rapidly appeared 
in the body-cavity, and caused death by producing poisonous bye- 
products. In many fish, on the other hand, bacteria not only reach 
uninjured the body-cavity, but continue to live there in considerable 
numbers without disturbing seriously, if at all, the vital processes 
of their host, — in other words, most fish seem capable of tolerating 
the presence of one or more kinds of bacteria in the peritoneal fluid, 
whilst others can even tolerate considerable numbers in their blood. 
It seems, however, that there is a limit to this toleration, for when 
the equilibrium is disturbed, when by a change of the surroundings 
the vitality of the tissues is diminished, the bacteria rapidly in- 
crease, and unless the tissues recover the position they have lost, 
the bacteria may directly or indirectly cause death. It has been 
suggested by Metsclinikoff and others, that bacteria are kept in 
subjection chiefly through the influence of the colourless blood 
corpuscles. This may be so in some cases, but it may be taken 
for granted that the living tissues as a whole repel the advance of 
the destructive organisms, and that some bacteria are arrested and 
destroyed by one tissue, while other bacteria are sterilised by another. 
A very small swing of the balance may determine whether a given 
bacterium will develop or not. This may be inferred from the 
behaviour of culture-media, e.g ., whether gelatine will act as a 
suitable medium for a given bacterium may depend on its reaction 
or on the amount of moisture it contains. In the same way, whether 
a given bacterium is able to disintegrate a piece of muscle may 
depend on the reaction or rigidity of the muscle. 
The distribution of bacilli in the tissues of fish, in which the 
conditions were favourable for their growth, is somewhat remark- 
