560 SUMMARY OF CURRENT RESEARCHES RELATING TO 



eggs in these two cases — which are quite normal — are deposited in the 

 water, and metamorphosis takes place after some months into land 

 salamanders 45-5G mm. in length. In the case of S. atra, the two 

 larvae have very long gills in utero and no fin to the tail ; when they 

 are born they are 38—40 mm. in length. 



AVhat Kammerer sought to do, was to simulate alpine conditions in 

 the case of S. maculosa, l>y keeping the animals at a low temperature 

 and without water. The latter condition was sufficient by itself. He 

 found that the action of the changed surroundings was cumulative. 

 After a few pregnancies, 2-7 young are produced, 39— 43 mm. in length, 

 with short gills or without gills, and, to begin with, black in colour. 

 Finally, a stage is reached where only two young ones are produced, as 

 in S. atra. 



Secondly, Kammerer sought to simulate for S. atra the conditions 

 normal to S. maculosa. He kept the salamanders at a higher tempera- 

 ture (25°-30° C), and in shallow water or with abundance of water 

 beside them. When the full result was reached, 3-9 larvae, 35-45 mm. 

 long, were produced, with gills at most 8 mm. in length (in contrast to 

 the very long normal gills of S. atra), with a fin 2-3 mm. in breadth, 

 and of a coffee-brown colour instead of the usual black. These larvae were 

 more adapted to the water than larvae at the corresponding stage cut out 

 from the uterus of an ordinary pregnant S. atra. 



The offspring of the salamanders experimented with were kept for 

 2h years in vivaria indoors, but did not become sexually mature until 

 they were put in large vivaria in the open air where conditions were 

 more normal. They became mature when 3 J years old. 



1. The offspring of those specimens of S. maculosa that had been 

 subjected to cold and want of water gave birth to (a) very advanced 

 large-headed larvse, 45 mm. long, with much reduced gills, which under- 

 went metamorphosis several days after, or moderately advanced larvae, 

 40 mm. long, with large gills (both sets laid in water), or (b) to small 

 larvae, 26 mm. long, with rudimentary gills. The latter were laid on 

 land ; they were unable to live in water ; they underwent metamorphosis 

 after 4 weeks, and were then 29 mm. long. 



2. The specimens of S. atra which were born as larvse, bore in the 

 water 3-5 larvse of two sizes, 33-40 mm. or 21-23 mm., light in colour, 

 with gills 8 mm. long, and with a fin 3 mm. broad. 



In the case of (1) it should be observed that the conditions were 

 normal for S. maculosa ; in the case of (2) the conditions were to a 

 certain extent a continuation of the experimental conditions under which 

 the parents were born, for, as is well-known, S. atra lives at high 

 altitudes, where the temperature is low and where water is scarce. 



3. Specimens of S. maculosa born as salamanders, showed under 

 repetition of the experimental conditions an augmentation of the effects. 

 Their offspring were very like normal offspring of S. atra. 



As to the general interpretation of these very interesting results, 

 Kammerer admits that there is possibly a direct action of the environ- 

 mental conditions on the germ-plasm, but he considers that there is, 

 nevertheless, a true transmission of an acquired mode of reproduction. 



He does not accept Weismann's limitation of the concept " acquired 



