180 Racial Studies in Fishes 



In addition to the ordinary population-analyses individual offspring 

 analyses have been carried out each autumn since 1914, the offspring of 

 each gravid female being separately investigated. From each female, 

 10 embryos (a "decade") were taken at random and their number of 

 vertebrae etc. determined \ 



For the sake of brevity I shall restrict myself to mentioning here a 

 single character, viz. the number of vertebrae. This character has proved 

 to be of great importance in distinguishing local races in Zoarces, where 

 the number has been found to vary from 99 to 126. A large mass of 

 material of many thousands of specimens was analysed and statistically 

 treated. The .countings were made by myself (1914-1917) and by 

 Mr Vilhege, M.Sc. (1917-1919). 



II. Comparison of the Different Annual Classes within 



A Population. 



For this purpose the decade investigations, consisting in an analysis 

 of ten embryos taken at random from each gravid female, are very well 

 suited. By examining the embryos instead of the adult specimens all 

 doubt is of course excluded that the individuals belong to the same 

 annual (;lass. 



Table I (p. 188) shows the results of offspring-analyses (decade 

 investigations) within the population from Station 81 during the years 

 1914-1919. Altogether 871 decades or 8710 individuals rather evenly 

 distributed between the six years have been examined. 



It is seen from this table that the population-averages for the 

 different annual classes agree very well. By means of the present great 

 mass of observations it has thus been possible to point out that there is 

 no difference between the various annual classes with regard to the 

 average number of vertebrae. In other characters such a difference can, 

 however, be pointed out, e.g. for the number of pectoral rays^. 



In Table II (p. 189) the same material is arranged in another way. 

 If the figures are set out as a curve it will be seen that this is very 

 regularly shaped, as might be expected where the material is so large. 



Table III (p. 189) also deals with the same material and shows the 

 average values of ten decades, each figure representing a hundred 

 individuals. It will be seen that the highest value is 114"17, the lowest 



^ See Johs. Schmidt : " Zoarces viviparus L. and local races of same" (C. E. Labora- 

 toire Carlsherg, Vol. xiii, p. 389, Copenhagen, 1917). 



2 See my paper in Comptes-Rendus du Laboratoire Carlsberg, Vol. xiv. No. 1, pp. 2-11, 

 1917. 



