NO. 14-13. CAVE SALAMANDERS— BANTA AND McATEE. 69 



the caverns S. longicaudiis was much more common, while in the woods the two 

 species occurred in about equal numl)ers. 



Blatchley records tiiiding- two specimens beneath logs in Monroe 

 County, Indiana, a mile or more from any known cave and half that 

 distance from springs or streams. Its occurrence at Brookville and 

 Vincennes, Indiana, and in Decatur County', localities not in charac- 

 teristic cave regions, further indicates that it is more or less independ- 

 ent of caves. 



This account, however, has to do with the species particularly as a 

 cave inhabitant. In Maytield's Cave it is doubtless fairly common, 

 but is onl}" occasionally seen and then usually not in remote parts of 

 the cavern. Nearly all of the adults seen have occurred between 00 

 and 150 feet from the mouth. Four, however, were observed in a 

 passage 572 feet back, another at the edge of a shallow pool at 1,200 

 feet, and a recently matured specimen was collected 1,170 feet from 

 the mouth. 



The latter occurrence is ]-eadily to be understood because the larvse 

 are hatched and develop in these deeper fastnesses. Indeed, the whole 

 character of the salamander's distribution in the cave results from this 

 habit. The younger larva^ occur in the remoter parts. After, and 

 even before transformation there is a gradual movement toward the 

 mouth; here the adults — splendid fellows — are most numerous. Dep- 

 redations of enemies and escape to the outer world regulate their 

 number. That adults are found in the deeper portions of the cave is 

 probabh' due in most cases to the necessity of la3dng their eggs in 

 water, of which there is usuall}^ none near the entrance. When larva> 

 are found at the mouth or even outside of the cave it is probable that 

 in most cases they were carried there by freshets. 



Within the cave the adults are generally found in a crevice or upon 

 a shelf of the wall. Onl}' three have been observed on the floor. 

 Their farorite resting- places, therefore, are to be reached onl}" by 

 climbing. But this is an easy feat for S. ^rtaciilicaudus^ as the follow- 

 ing will illustrate: At a point 572 feet back two were seen near each 

 other in a cranny near the roof above a 15-foot embankment of earth, 

 while on the opposite side of the passage one had ascended more than 10 

 feet on a perpendicular surface of stone. In scaling such places they 

 are probably materially assisted by their tails. These are prehensile 

 to quite a degree, enabling the salamander to support itself by the tail 

 alone. The ability of the animals to climb is more severely' tested 

 when the}' are confined in glass jars. Yet they ascend vertical and 

 even overhanging surfaces and usuall}' remain clinging at the highest 

 point. 



When thus settled in a comfortable position the salamander is not 

 easily induced to mcfv^e. In the cave it is not readily disturbed b}" a 

 light or by an object moving near. While light of itself will rarely 



