ROSÉN, STUDIES ON THE PLECTOGNATHS. 11 



ready at the beginning of tbis inquiry considered the air- 

 sac to be a dilated stomach. Thilo and otbers have put 

 fortb tbe same opinion, without giving any proofs of it. 

 Edinger's researches bave given us good data for fixing the 

 boundaries between the different parts of the alimentary canal. 

 The question will best be settled by stndying the conditions in 

 Spheroides, on the intestine of which I have made microsco- 

 pical researches. The oesophagus and the stomach are well 

 defined by the form of the epithelial layer, which in the for- 

 mer consists of goblet cells, in the låter of high cylindrical 

 ones. As glands are wanting in the stomach, a condition 

 which is found also in other Teleosts, these cannot be looked 

 upon as a character. That the part that follows on the 

 stomach represents the duodenum is shown by the histology 

 and by the opening of the hepatic duct into it. When 

 it has been proved, that the part of the intestine which forms 

 the air-sac in Spheroides represents the stomach, it is quite 

 evident that the dilatation of the alimentary canal that is 

 present in all the other species has the same origin. The same 

 situation and the continuous series of different stages are 

 sufficient proofs. Another view of which part of the intestine 

 that the air-sac represents, is, however, maintained by some 

 authors. Bridge writes in ^Cambrigde Natural History» 



(Vol. Fishes, p. 256) on the oesophagus as follows: » 



in the Plectognath Teleosts it gives off a large sac-like out- 

 growth (»air-sac»), which extends anteriorly as far as the head 

 and posteriorly to the beginning of the tail, and communi- 

 cates with the oesophagus by two apertures», a description 

 which has no support from the fact shown by the species 

 hitherto examined. 



What has been said above about the air-sac of the Plec- 

 tognaths may be summarized in the following lines: An airsac 

 is wanting in the Ostraciontidae, the Molidae and Triacanthidae. 

 Within the family Balistidae the air-sac takes its origin as a 

 dilatation of the stomach, which increases in size. In the Dio- 

 dontidae the air-sac is very large and hy means of solid chords 

 of connective tissue lined hy a peritoneal layer united with the 

 ventral hody tvalL In the Tetrodoyitidae the specialization has pro- 

 ceeded furtJiest. The air-sac is in this family {in all species?) mar- 

 ked off from the other part of the stomach hy a fold coniaining a 

 sphincter muscle and perfectly coalescent ivith the ventral hody wall. 



