1885.] NATURAL SCIENCES OF PHILADELPHIA. 131 



as air-bladder. This difference in organization was probably the 

 result of some marked difference in their life habits. The Ganoids 

 may, in their original state, have inhabited poorly aerated waters 

 or waters otherwise ill-adapted to breathing, while the Elasmo- 

 branchs may have had their primordial habitat in clearer and 

 purer waters. 



But there were other conditions which may have been the main 

 influencing causes in the development of an organ for air-breathing. 

 We know that the land was habitable during long ages ere it 

 gained any vertebrate inhabitants. The presence of insects in 

 Devonian and Silurian strata proves this. It must have possessed 

 much food material, both vegetable and animal, and it is hard Im- 

 probable that the active fish forms of the earty seas made no 

 effort to obtain a share of this food. Long ages passed during 

 which we have no evidence of land animals higher than insects 

 or snails. It is highl}' probable that many fishes gained the habit 

 of leaving the water temporarily for the land in search of food 

 during this period. We know that many fishes do so now, and that 

 some even climb trees, in spite of the rnany dangerous foes that 

 now exist on land. In the era referred to there were no such 

 dangerous foes. Such fishes as left the sea for the land would find 

 only food to repay their enterprise. Thus there must have been 

 a powerful inducement for fishes to assume this habit. 



The indications, however, do not lead to the idea that the 

 original development of an air-breathing organ was due to occa- 

 sional visits from sea to shore. Such an organ must have slowly 

 developed under the pressure of less extreme changes of condi- 

 tions. It probably arose through the effect of such influences as 

 still act upon fish, and force them to occasionally breathe air ; 

 such as foul or muddy water, or a lack of proper aeration arising 

 from any cause. Another important influence is the drying-out 

 of pools, by which fish are left in the moist mud until the recur- 

 rence of rains, or are even buried in the dried mud for the six 

 months of the dry season. Such is the case with Lepidosiren, 

 which uses its lungs during this period. In certain other fresh- 

 water fishes, of the family Ophiocephalidae, air is breathed while 

 the mud continues soft enough for the fish to come to the surface, 

 but during the remainder of the dry period it remains in a torpid 

 state. In these fishes the air is breathed into a simple cavity in 

 the pharynx, whose opening is partly closed by a fold of the 



