29G VARIATIONS AND RELATIONSHIP OF 



divided presumably from the epithelium, travels forwards over the 

 surface to its anterior edge, and spreads over part of the under surface 

 posteriorly. When this process is complete the scale is said to have 

 reached the Dor n- Stadium,. 



In the next chapter the author deals with the different local forms 

 of the plaice and flounder from different localities in the North Sea 

 and the Baltic. We do not propose to give a detailed statement of his 

 conclusions, because it appears to us that the evidence on which these 

 are based is quite insufficient. Taking the first locality dealt with as 

 an example, we may note that the author diagnoses 20 male and 8 

 female flounders. He then takes the most common measure (in 28 

 individuals) for each character considered, and writes a formula which 

 he calls Die Mittelformel fur die Konigsberge Form. Further, a table is 

 appended in which the frequencies of the several variations of each 

 character in the individuals examined is expressed in percentages of 

 the total number of individuals. Thus we are informed that 55 'per cent, 

 of the W male flounders from Konigsberg have 36 vertebrae. 



It appears to us that little reliance can be placed on conclusions which 

 are drawn from an examination of so small a number of individuals : 

 and it is simply misleading to express the results of such examination 

 in percentages, when fewer even than a hundred individuals have been 

 examined. 



The author proceeds to summarise his results by giving two formulte, 

 expressing the characters of Baltic and North Sea flounders. We 

 are not altogether sure what meaning the author attaches to these 

 generalised formuhie. The formulae, assuming the results obtained for 

 the separate localities to be accurate, express the most common measure- 

 ments of the several characters in all the individuals examined from the 

 Baltic and the North Sea, and they may be said to show how in a 

 general way the flounders from these two regions differ from one another 

 in respect to each of the several characters considered. But it would, 

 we think, be a mistake to take these formulaj as expressing the combined 

 characters of the ideal form which the environmental forces were tend- 

 ing to produce in these seas. If it is desired to show the direction in 

 which evolution is tending to transform these populations, account must 

 be taken of the facts of correlation. In a paper on certain correlated 

 variations in Carcinus mcenns* it has been pointed out by Weldon that 

 before we can estimate the changes at present going on in a race 

 or species, we must know, among other things, (a) the percentage of 

 animals which exhibit a given amount of abnormality with regard to a 

 particular character, and (&) the degree of abnormality of other organs, 

 which accompanies a given abnormality of one. The ideal form of which 



* Weldon. I'roc. Roy. Soc, vol 54., p. 318. 



