14 ARKIV FÖR ZOOLOGI. BAND 7. NIO 30. 



verse muscles.^ When the air-sac is filled with air in Diodon- 

 tids and Tetrodontids the belly is blown up like a ball, and 

 that the above mentioned muscles are of great importance 

 for the contraction thereof needs no further explanation. 

 We find the most specialized form of the air-sac, as I have 

 shown, in Spheroides, in which the sac is distinctly differen- 

 tiated from the rest of the stomach by a circiilar fold. This 

 fold contains a sphincter muscle, but whether this is able to 

 close the opening of the air-sac perfectly I cannot state with 

 certainty. In all the specimens I have dissected or made 

 microscopical sections through this opening has not been closed, 

 but that does not pro^ve that it is impossible. 



We may consider the original function of the air-sac to be 

 a reservoir for air, which is by and by forced into the gill- 

 chambers in order to oxygenate the ivater. To this function is 

 added in the Diodontids and Tetrodontids that of defence, by 

 the extraordinary development af the air-sac. ivhereby the body 

 may be inflated like a ball and the spines in the integnment 

 erected. No direct respiration takes place in the air-sac. The 

 air is taken in by swallowing movements and kept in the sac 

 by means of the circidar muscles in the wall of the oesophagus, 

 and by the pyloric valve. In Spheroides (and perhaps all Tetro- 

 dontids) the air-sac may be closed perfectly or atleast to a certain 

 extent by a special sphincter muscle. The sac is emptied by 

 relaxing of the respective muscles and at Icast in the Diodontids 

 and the Tetrodontids even by the action of the ventral body 

 muscles, which are adapted to this purpose. 



III. Modifications relating to the deyelopment of the Air-sac. 



We have every reason to expect, that the development 

 of a part of the stomach into an air-sac, through which the 

 belly may be inflated and thus the form of the body highly 

 altered, has caused several modifications in the structure or 

 shape of several organs. We must, however, be very careful 

 in regarding all the alterations that at a hasty glance seem 

 to be related to this power, to be directly caused by this. 

 As will be shown below, some conditions in the anatomical 

 structure have originated in species in which an air-sac has 



^ These muscles will be described in a special section of this series. 



