426 J. BARRELL INFLUENCE OF CLIMATES ON VERTEBRATES 



monary arteries consequently originate anterior to the two dorsal aortse. 

 This is true also of the amphibians. 



Thus considerable variation in the place of origin of the pulmonary 

 artery is exhibited, but it is in closest relation to the arterial system of 

 the stomach and intestines in those fishes, to whom in their evolutionary 

 history air-breathing by means of the air-bladder has never been more 

 than a secondary mode of oxygenation. The most forward connection is 

 found in the amphibians, but even here the pulmonary artery is a branch 

 from the fourth efferent artery and not in any sense an organ between 

 the body and the gills. 



Another line of evidence is found in the embryonic development of the 

 air-bladder. In all those fishes which have an air-bladder it opens, dur- 

 ing the larval stage, by means of a duct into the cesophagus distinctly 

 lower than the pharyngeal cavity and in some cases near the stomach. In 

 the dipnoans and ganoids this duct is permanently retained and serves as 

 a respiratory passage. The development of the air-bladder shows thus a 

 relationship to the cesophagus and not to the gill arches. 



From these lines of evidence it is to be inferred that the reorganization 

 of the heart and arterial systems which has had to take place to produce 

 the efficient circulatory and respiratory systems of birds and mammals out 

 of the two-chambered heart and gill -respiration of the primitive selachian 

 was necessitated by the original location of the accessory respiratory sur- 

 face as the forward part of the intestinal canal. 



Originally was the air used for oxygenation swallowed, or was it ex- 

 pelled from the mouth or gills? It appears probable from the oesoph- 

 ageal connection that it was originally swallowed, but on the develop- 

 ment of a larger respiratory surface, with the necessity for a rapid re- 

 newal of the air within the respiratory organs, the line of least resistance 

 became that of a forward expulsion, a rhythmic regurgitation. 



Even in those modern fishes which breathe air by means of accessory 

 organs in the pharyngeal chamber the gills are probably not the typical 

 channels for exhalation. It would appear reasonable, however, that if 

 such fishes came to rely more largely on air, some, at least, might adopt 

 the habit of inhalation through the mouth, exhalation through the gill 

 openings. Rhythmic movements in the gill covers similar to those in 

 lampreys, which are attached by their mouths, would apparently, in fishes 

 coming to breathe by a pharyngeal chamber, be a quite possible and a 

 most efficient means of maintaining respiration. The gill openings in 

 that case would remain permanent features of the anatomy of the result- 

 ant air-breathing animals. Expulsion of air forward from the oesophagus 

 would, on the contrary, naturally find its exit through the mouth or nos- 



