ACQUIRED CHARACTERS 37 



theory, or with the known stubborn and persistent nature of 

 heritable variations, when once they have arisen. 



Kammerer of Vienna has pubHshed in the last five years 

 the results of a long series of experiments with salamanders 

 and lizards designed to show the inheritance of acquired 

 characters. In this connection we will consider his experi- 

 ments with temperature. The coloration of several species of 

 lizard, with which Kammerer experimented, changes with 

 changes of temperature. Kammerer kept lizards at abnor- 

 mally high or abnormally low temperatures, and found that 

 the induced changes of coloration persisted to some extent 

 even after the animals were returned to normal conditions. 

 Further, while they were thus altered, the offspring which 

 they produced, inherited in some degree the supposedly in- 

 duced changes. The evidence for this case, as for many 

 similar cases which might be cited, is quite insufficient. Un- 

 doubtedly individual differences in coloration occur among 

 the lizards quite independently of external temperatures. 

 Further some probably change more readily and extensively 

 than do others in consequence of changed temperatures. A 

 corresponding variation among the offspring, plus and minus, 

 as compared with their parents, would then account for such 

 plus variations in pigmentation as Kammerer observed 

 among the offspring and which he ascribes to inheritance of 

 changes induced in the parents. 



Sumner (1915) kept white mice, some in a cold room, some 

 in a warm room, where they multiplied. The mice which 

 grew up in the cold room had shorter tails and feet than those 

 which grew up in the warm room. Animals reared in each 

 room were now transferred to a common room of ordinary 

 temperature and allowed to produce offspring there. In three 

 out of four such lots of offspring studied, the cold -room 

 parents had young with shorter tails and feet, but in a fourth 

 lot these relations were reversed. It seems doubtful, there- 

 fore, whether the agreement between parents and offspring 

 in three of the four cases studied is anything but a coinci- 



