1887 .] Prof. C. Ewart on Bacteria in Lymph , &c. } of Fish. 265 
slender, sometimes short and thick, in the peritoneal fluid and blood 
of the whiting ( Gadus merlangus), haddock ( Gadus ceglefnus ), cod 
( Gadus morrhua), and herring ( Clupea harengus), and in the peri- 
toneal fluid only of the flounder ( Platessa flesus), plaice ( Platessa vul- 
garis ), and lumpsucker ( Cyclopterus lump us). I have not hitherto 
succeeded in demonstrating the existence of bacteria in either the 
peritoneal fluid or blood of the skate ( Raia balls), dog-fish ( Acan - 
thias vulgaris ), or fishing-frog ( Lopliius piscatorius). 
Perhaps the difference in the number and distribution of bacteria 
in the sea fish examined by Olivier and Eichet and those I have 
recently studied may be accounted for, either by a difference in the 
temperature of water from which the fish were taken, or by the fish 
having been longer under less favourable conditions in the one case 
than in the other. 
It is extremely desirable that a continuous series of observations 
should be carried on throughout the year, in order to ascertain 
whether bacteria are more abundant in summer than they are in 
winter, whether they increase or diminish before and during the 
spawning period, and whether the bacteria indirectly influence the 
migration and distribution of fish — the fish which readily suffer 
from an increase of the bacteria in the peritoneal cavity either 
remaining in comparatively cold seas or selecting cold currents when 
they migrate in search of food, or in obedience to their spawning 
instinct. 
There can he no doubt that the bacteria enter the body-cavity by 
penetrating the walls of the intestine, neither can there he any 
doubt that having once established themselves in the peritoneal 
fluid they do their utmost to find their way into the blood and 
tissues. It may he taken for granted that ordinary bacteria flourish 
in the intestinal canal of all vertebrates, and that they assist in 
digestion by helping to disintegrate the food particles. Notwith- 
standing the presence of active bacteria in the intestinal canal and 
the bile and pancreatic ducts, I have failed to discover either 
bacilli or micrococci in the body-cavity of either amphibia, reptiles, 
birds, or mammals when in a healthy condition. Hence it may be 
taken for granted — (1) that in the higher vertebrates under ordinary 
circumstances the walls of the intestine form an effective filter or 
screen which prevents the passage of the bacteria into the body- 
