THE LUNGFISH 79 



tion, or the accumulation of metabolic waste products- 

 all can be excluded by controlled observations; perhaps 

 it is an active inhibition induced by prolonged postural 

 fixation, a sort of autohypnotic ecstasy; or perhaps it is 

 just a negative state resulting from the absence of any 

 nervous activity other than that required for respiration, 

 and engendered by long insulation from any sensory 

 stimuli. 



When the crisis comes, it is the threat of death that 

 wakens it. Removed from the cocoon and placed in 

 water, the inactive blob floats imtil the air that is pres- 

 ently in its lungs is expelled and then it sinks, rolling 

 inertly to one side or the other on the bottom of the 

 aquarium. When next it opens its mouth to breathe, the 

 influx of water shuts the jaws by a reflex that must be a 

 imique patent of the Devonian air-breathing fishes. After 

 it has made several futile attempts to breathe under 

 water (which, of course, it cannot do), asphyxia begins 

 its work and convulsive movements appear that twitch 

 and rock the body violently, until, in a culminating par- 

 oxysm of effort, the animal partially uncurls and strug- 

 gles with grotesque movements to the surface. We sup- 

 pose that in nature, when the flood waters cover the 

 nest, a similar asphyxial convulsion wakens the fish so 

 that it pushes to the top of its burrow and, breaking the 

 soft edges of the blowhole, swims free as soon as the 

 water attains any depth. After its first victorious ascent, 

 aerial respiration is resimied (we never saw a lungfish 

 drov^ni if it was not mechanically imprisoned below the 

 surface) but for several days the stiffened and twisted 

 body may look very much Hke an animated horseshoe 

 trying to stick one end of itself out of the water, and for 

 days the swimming movements may be erratic and 

 unco-ordinated, as though the animal had forgotten 

 some of the most elementary motor acts. If the lungfish 

 is capable of learning by experience, it seems reasonable 

 to suppose that this knowledge is all erased during its 

 long sleep; it can be suspected (though our laboratory 



