Metazoa Rana. 293 



monary artery already mentioned in considering the branches 

 of the truncus arteriosus here subdivides again and again, 

 and its- ultimate capillaries ramify in the lung-wall, separated 

 from the air which normally fills the lung by a single layer 

 of squamous epithelium. It is manifest that a gaseous ex- 

 change between the atmosphere and the blood circulating in 

 the capillary terminations of the pulmonary artery is thus 

 made possible. (It must be noted that the pulmonary 

 artery, despite its name, contains venous blood.) The wall 

 of the limg is exceedingly elastic, and readily collapses if the 

 glottis be kept open or the wall be punctured. 



In principle the lung of the frog does not differ from the 

 gill of the fish, for in both the end to be gained is the 

 exposure of a maximum amount of blood to the atmosphere. 

 In the fish the blood is carried outside the body in capillaries, 

 which ramify in processes supported by a connective tissue 

 framework, and thus meets with the oxygen dissolved in the 

 water, while in the frog the air is sucked into the interior of 

 the body to the blood, a method by which risk of injury to 

 so delicate and important a system is reduced to a minimum, 

 If space permitted, it would be interesting to trace the 

 manner in which the lungs have arisen as a modification of 

 a curious organ -the swim-bladderdeveloped in many 

 fish. By altering the quantity of air contained in the swim- 

 bladder, fish are able to increase or decrease their buoy- 

 ancy. In higher animals, such as the rabbit or the dog, the 

 lung, instead of being a single hollow sac, is composed of an 

 enormous number of extremely minute sacs closely packed 

 together (fig. 155), and communicating with the external 

 world by means of a series of branched tubes (bronchi), 

 which ultimately unite to form one large tube (trachea), 

 of which the sole representative in the frog is the small 

 cylindrical chamber into which the glottis opens. The gain 

 in the extent of respiratory surface in the mammalian lung 

 by this arrangement must be at once evident, whilst at the 

 same time every particle of air inhaled is made use of. In 



