446 Growth, Variability and Correlation in Young Trout 
I have thus had at my disposal nearly one thousand young trout, all born of 
the same parents, and killed at five successive fortnightly stages after hatching, 
the period of growth examined thus comprising ten weeks. Mr Severn has been 
kind enough to furnish me with the following particulars of the mortality that 
occurred in this particular lot. Five per cent, died after hatching during the first 
fortnight, that is before any were sent to me. When the fish were about three 
weeks old (that is, between Stages I and II) a trifling number died of dropsy and 
at the same time about 5°/ 0 died which were deformed. Otherwise there was 
scarcely any loss at all. 
The dead individuals were not sent to me and I only measured the survivors. 
Whether the death-rate is high enough of itself to bring about the diminution of 
variability which, as we shall see, actually occurs, is a question to be discussed 
later. But whether that is so or not I think it may be in any case worth while to 
describe briefly tire growth and the change that occurs in the growth-rate, and in 
the variability both of the whole body and of different parts, both absolute and 
relative, as well as the correlations that exist between these parts. 
I have measured the following — total length from the tip of the nose to the 
extremity of the tail, diameter of the eye, length of head as determined by the 
posterior margin of the operculum, length of the bases (properly speaking the 
projections of these bases on the longitudinal axis of the body) of the median fins 
— anterior dorsal, posterior dorsal, ventral and caudal — -breadth of the body at the 
level of the anterior end of the anterior dorsal fin, breadth of the tail, and position 
of the pectoral, pelvic, anterior dorsal, posterior dorsal and ventral fins as deter- 
mined by the anterior end of the base of attachment in each case. The ventral fin 
lies just behind the anus as a rule, so that the position of the former can be taken 
usually as marking the position of the latter, but there may be a gap between the 
two. In these cases the posterior border of the anus has been separately measured. 
Lastly the number of bars or spots in the middle row on the animal's side has 
been counted. In all cases the measurements were made on the right-hand side 
of the animal. 
I begin with the absolute measurements. The mean values are given in 
millimetres in Table I, and in the same table the growth-rate at each successive 
stage, as expressed by the percentage increment of growth. 
The general increase of all parts needs of course no comment. The only exception 
is the number of spots which is established at once in Stage II and undergoes 
practically no increase. What is of interest is the diminution of the growth-rate. 
This is evident everywhere. Generally speaking the decrease is more rapid in the 
early stages than in the later — which is in harmony with the results obtained by 
other observers — but in some cases there is irregularity. In the eye the growth- 
rate sinks at Stage III, to rise again at Stage IV and sink once more in the final 
stage. In the posterior dorsal fin there is a marked increase at Stage III, followed 
by a great decrease, and this by a slight rise. A slight, perhaps hardly significant 
rise, can be detected in the final stage of the total length. 
