JoHS. Schmidt 183 



bottom near each other, projecting above the water so that feeding 

 could easily be carried out. The pairing and the development of the 

 embryos thus took place in captivity and under the same external con- 

 ditions for both population-samples. In December 1916, i.e. at a season 

 when Zoarces contains large embryos, the boxes were lifted out of the 

 water and their contents preserved. In the population-sample from 

 Station 31 was found 27 females with embryos, and in that from 

 Station 34 a number of 11. 



Under natural conditions the average number of vertebrae in the 

 population at Station 31 is about 113*2 (see Table I), while in the 

 population at Station 34 it is about lOS'O (see my first paper in this 

 Journal, Vol. vii, 1918, the graphs p. 110 (Station 31) and the graph to 

 the right at the foot of the table inserted between pp. 110 and 111 

 (Station 34, average 107'98)). The values for the two populations thus 

 differ by about 5 vertebrae, a difference which the annual investigations 

 since 1914 have proved constant. In spite of the small distance of 

 about 50 kilometers between the two stations, their natural conditions 

 differ widely, e.g. at the first of them the salinity of the water is about 

 21 7oo» *^d ^^ t'h*^ other only about 12 7oo- I^ might therefore be 

 conceivable that the cause of the difference in the number of vertebrae 

 would be found in the different environmental conditions. If this was the 

 case, it was to be expected that the difference would disappear when the 

 two populations were brought under similar external conditions. This 

 was done in our transplantation experiment, and it is the question now, 

 whether the difference of 5 vertebrae, appearing under normal con- 

 ditions, has disappeared in the offspring, all of which have developed in 

 our experiment under the same external conditions. 



The result of the experiment is given in Table V (p. 190). For 

 comparison it may be stated that the average number of vertebrae in 

 offspring developed under natural conditions in 1916 for Station 31 was 

 113-101 ± 0-4471 ^125 decades), and for Station 34 was 107-768 ± 0-468 

 (116 offspring-samples each consisting of 5 specimens), showing as usual 

 a difference of about 5 vertebrae. 



It is seen from the table that the offspring of the transplanted popu- 

 lation sample from Station 31 had on an average 114-819 + 1*059 verte- 

 brae, while the corresponding value for the sample from Station 34 was 

 108*283 + 1-363. In other words the difference between *the two popu- 



1 The uncertainty of the average is here and in the following indicated by the fluctua- 

 tion equal to 5 times the probable error (see my first paper in this Journal, Vol. vii. 

 No. 2, p. 118 note, 1918), 



