1s BULLETIN OF THE BUREAU OF FISHERIES. 
way there was formed a pulverized mass of a dark brown or dark red color, which 
approximated the granular mass found in the intestines of the chitin-swallowing fish. 
Pieces of chitin were placed in acid of strengths varying from 0.4 to 35 per 
cent hydrochloric acid. Carbon dioxide was set free in each case, but in greater 
quantities with the stronger acids. -The chitin became softened, pliable, but did not 
dissolve. 
6. Crabs were fed to dogfish confined in small aquaria. The excrement of the 
fish was carefully watched, and in this excrement, known by its shape and color, 
some pieces of the softened but otherwise unchanged chitin could be found. No 
evidence was gathered that the fish ever regurgitated any of the chitin, though Yung 
believes that regurgitation might take place. 
Conclusion: The conclusion to be drawn from the experiments is that J/ustelus 
canis and other chitin-swallowing fishes do not digest the chitin. The frequent 
change of the gastric juice, combined with the movements of the stomach, dissolves 
out the salts, softens the shell, and breaks it up into a fine mass, such as may be found 
in the spiral valve. The chitin is not regurgitated, but on the contrary is excreted 
in a finely divided mass. 
MIDDLE INTESTINE. 
Extracts of the middle intestine or duodenum of the various elasmobranchs show 
no digestive activity. Whether or not the cylindrical cells and goblet cells lining 
the mucous membrane of the duodenum play any part in activating the pancreatic 
juice will be discussed under pancreatic digestion. 
SPIRAL VALVE. 
Extracts of the mucous membrane of the spiral valve showed no digestive action 
on starches, fats, or proteids; nor indeed was it possible to demonstrate any invert- 
ing power, though we may presume that such power may exist in thismucus. The 
main function of the spiral valve is absorptive. Its further action in pancreatic 
digestion will be taken up in the discussion of the function of the pancreas. 
PANCREAS. 
As may be seen by reference to page 9, there has been but little experimenta- 
tion on the pancreas of fishes, which fact is warrant for discussing the physiology 
of the pancreas of elasmobranchs. In order to get at the problem with the greatest 
clearness, it would perhaps be well for us to outline the growth of knowledge 
concerning pancreatic digestion, from the first experiments on the pancreas to the 
later researches leading to our present-day understanding of the action of this organ. 
The pancreatic ducts, according to Haller (1764), were discovered by Wirsung in 
1642. Although Wirsung appears to have observed the pancreatic juice, he did not 
pursue the subject further, and little was made of his discovery until Regner De 
Graaf took up the matter. In 1664, De Graaf, according to Foster (1901), made the 
first successful pancreatic fistula and collected the juice from a dog. But De Graaf 
did net obtain a definite grasp of the function of the pancreas. Eberle (1834), 
however, announced that a watery infusion of the pancreas when shaken with 
oil emulsifies it. In 1836, according to Corvisart (1857), Purkinje and Pappenheim 
discovered the proteolytic power of the pancreas, and nine years later Bouchardat 
and Sandras (1845) discovered and established with precision, by means of observation 
