ALIMENTARY CANAL, ETC., OF AMIURUS CATUS. 415 
A glycerine extract of the liver digests fibrin in a 0°5 solution 
of sodium bicarbonate, requiring but a few hours for a piece of 
moderate size. 
In young cat-fishes, of from one to two inches in length, and from 
which I made a series of sections in the neighbourhood of the liver and 
midgut, I was unable to find a trace of pancreas. This is possibly to 
be explained, as Bernard suggested, by the supposition that digestion 
by the stomach is quite sufficient for the food of young fishes. It is 
also to be observed that hepatic tissue does not penetrate between all 
the capillary vessels of the liver. It is quite safe to say that the 
pancreas is of later development, and is connected with the portal 
vein in some such way as to be dragged by it into the liver when the 
latter Increases in size. 
The fact discovered by Krukenberg that the extracts of the livers 
of different fishes accomplished a tryptic digestion may be explained 
by the possible distribution of the pancreas in the liver in the way 
that is describel above. Among those fishes studied by this physi- 
ologist, were Perca fluviatilis, Labrax lupus, Belone rostrata, Crenila- 
brus pavo, Dentex vulgaris, Trigla hirundo, Sargus Rondeletii, Gobius 
miger, &c. In Perca fluviatilis, according to Cajetan, the pancreatic 
ducts entwine about the portal branches till they sink into the liver. 
It may be added that it is possible in this fish, as well as in those given 
above, that the pancreas follows the portal vein as it does in the cat- 
fish. ,The organs so affected are, however, by no means to be denomi- 
nated a hepato-pancreas, as that name is understood in invertebrate 
anatomy. 
A more careful study of the pancreatic tubules in the cat-fish 
shows that it undergoes the ordinary changes effected during diges- 
tion. In a fasting condition the cells are filled with granules, the 
round nucleus situated near the outer part of the cell, and the whole 
stains feebly in carmine. When the liver is cut out four or five 
hours after the fish has been feeding, the granules are gathered into 
a region adjacent to the lumen of the gland, and this portion stains 
feebly, the rest of the cell strongly, in carmine. Fig. 11 gives a 
representation of this stage. 
I could observe a membrana propria for these gland tubules as 
little as in those of the gastric glands. The fibres of the connective 
tissue surrounding them are arranged in a dense sheath which serves 
all the purposes of membrane. 
