136 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



Bay, and (2) from Luce Bay. Now the latter area is 

 strictly preserved against trawling, while the sea round 

 the Isle of Man is certainly fished for all it is worth. 

 Further, there is a very abundant plaice population in 

 Luce Bay, while that round the South end of the Isle of 

 Man is very sparse. It is, therefore, quite possible that the 

 local plaice were different from those of Luce Buy, and 

 had a more rapid rate of growth than the fish of the 

 latter region. This difference in the spawning fish would 

 result in a difference of rate of growth in the larvae and 

 young fish. The double mode, in fact, may be taken to 

 indicate two strains of fish. 



On the other hand, it may simply indicate that the 

 rale of growth of the plaice is such that the deviations 

 from the modal character for a homogeneous group form 

 a skew curve. It is not possible to settle this point in 

 the meantime. I have, however, attempted to make a 

 correspondence between the deviations exhibited by the 

 Port Erin plaice and the normal curve of frequency error. 

 Taking the mode at 2' 625 inches, the part of the series 

 as far above the mode, as the smallest fish is below it, is 

 neglected. The standard deviation and modulus have, 

 however, been calculated for the entire series ; but in 

 constructing the algebraic curve the modulus has been 

 taken as 0'8 instead of the calculated value, 0712. 

 Points on the normal curve are then found by a method 

 suggested by Mr. A. L. Bowley. ; The correspondence 

 is not good, and though it is not quite justifiable to con- 

 clude that the observed measurements are those that would 

 result, with least improbability, from the normal curve, 

 it seems to me that the distribution is almost as likely 

 to follow the latter law, as that expressed by the equation 

 of one of I he asymmetrical frequency curves. The 



* This differs slightly froxo that indicated in Mr. Bowley's Elt?nents 

 <>j Statistics. 



