252 ‘TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 
by taking the difference between the mean increase in 
length of any one month and that of the preceding 
month. | 
It will be seen that the rate of increase for some 
a somewhat paradoxical result. 

months is negative 
This appears particularly with regard to the month 
August, and when I first observed it I ascribed the result 
to the under-measturement of the marked fish recovered 
during the August of the year in question. But the 
same anomalous result was observed in other years, and 
in different experiments considered separately; and it 
has also been observed by Mr. Farran,* so that one must 
regard it as susceptible of some other explanation. It is 
obvious that we have to deal with a _ differential 
migration in or about those months when a negative 
increment of growth is observed; that is, the movements 
of the plaice are such that the more rapidly growing fish 
are migrating to grounds where they are not caught, or 
are caught in far fewer numbers. On the other hand, 
the more slowly growing plaice have remained on the 
grounds where the fishery was relatively intense. This 
is, indeed, a very probable explanation. There is no 
sense in a negative growth. The value in column (3) 
cannot be less than that of the preceding month, and if 
one were drawing a smoothed curve through the points 
given in the column, he would be justified in inter- 
‘polating a value for the month exhibiting a negative 
growth. But the figures can be better dealt with 
otherwise. 
The rates of increase given in column (4) have been 
considered as values lying about a probability curve, 
and have been dealt with by Pearson’s method of 
* «« Plaice-Marking Experiments,’’ Fisheries, Ireland, Sci. 
Invest., 1907, III, [1909]. 
