264 Proceedings of Royal Society of Edinburgh. [june g, 
the number in the body-cavity and the number in the intestine, 
and they were most abundant in fish which had lived for some 
time in aquaria without food ; but in trout which had been fasting 
for at least ten days no bacilli could be observed in the peritoneal 
fluid. The carp which had bacilli in their blood had been living 
for some months in a small glass aquarium. 
The difference between the roach first examined and those 
examined subsequently led me to endeavour to ascertain whether a 
sudden change of temperature would produce any influence in the 
number and distribution of the bacilli. As I anticipated, a rapid 
change from a spring to a summer temperature (from 48° to 65° F.) 
greatly diminished the vitality of all the fish experimented with, 
except the carp. As the fish became more and more exhausted, the 
bacilli gradually increased, and when the temperature was raised from 
48° F. to 65° F. in two hours, the bacilli of the peritoneal fluid not 
only increased in the roach, perch, carp, and eel, but they made 
their appearance in considerable numbers in the body-cavity of the 
trout, and on one occasion, a number of small bacilli were found in 
the blood of a trout. Although the carp seemed to enjoy the rise 
of temperature, they were not exempt from the increase of the 
bacteria in the blood as well as in the peritoneal fluid. In some 
specimens of blood as many as eight short slender bacilli were 
visible in the field of the microscope at one time, and the peritoneal 
fluid, in some instances, swarmed with long and short bacilli, some 
of which were mobile. 
In some of the roach, in which no organisms could be detected in 
the blood, bacilli were found in the muscles immediately external 
to the peritoneal cavity. Further, bacilli were always abundant in 
the muscles of roach which had suffered from a sudden rise of 
temperature. The above observations were confirmed by cultiva- 
tions in gelatine, agar-agar, and infusions of fish muscles. In healthy 
active specimens of the roach and perch cultivations were easily 
obtained of the peritoneal bacilli, and generally also from the mus- 
cular fibres lying near the peritoneum, but in no instance did I 
succeed in obtaining cultivations when the blood, or the muscles 
from immediately under the skin, were used for infecting the culture- 
media. 
Of the sea fish examined, I have found bacilli, sometimes long and 
