( 499 ) 
to be considered in connection with a winter period passed on the 
upper course of the river? It is impossible to answer these questions 
with the material at present at my disposal. 
1 think my description shows that, whereas the application of the 
method has already given some interesting results with regard to 
the growth and the age of the Rhine-salmon in the first years of 
their life, for the age and the growth of older fish several obscure 
points are still waiting for elucidation 1 ). To begin with n will be 
very important to study the scales of the kelts descending to the 
sea in early spring and especially, to compare their structure with 
those of the so-called winter-salmon, the name we are accustomed 
to give to big clean fish, the largest of all ascending the river. 
As I have already said for that investigation no material was at my 
disposal, but I intend to continue these researches and will with 
pleasure report again on them later. Here a few more remarks may 
find a place regarding the scales, drawings of which are reproduced 
in the figures 21—23. 
Fig. 21 was drawn from the scale of a salmon of a length of 20.4 
cm. and shows a somewhat abnormal central part. Such scales are 
by no means rare; they have instead of the usual centre beginning 
as a very minute ring and narrow rings encircling it, a considerably 
larger irregularly defined central part of an oval or more rounded 
shape. Sometimes several of these scales are observed in close proxi¬ 
mity, normal ones lying between the abnormal scales. It is obvious, 
I think, that these abnormal scales are such, the development of 
which was retarded, so that a centre of much greater dimension was 
to be formed at once. It is also possible that they occupy, like so 
many new formations, the places of other scales which the fish lost 
by accident. The earlier investigators of fish-scales (Stuart Thomson 
and Johnston) also mention these abnormal scales but neither of 
them pronounce themselves in a decided way on their origin. 
Figure 22 represents the scale of a brooktrout (Salmo fario ) and 
figure 23 one of a seatrout (Salmo trutta). To distinguish smaller 
specimens of these fishes from young salmon (Salmo salar) of about 
the same size is even now a somewhat difficult affair. Therefore 
I think it is important, that the structure of the scale in the first 
x ) By saying so 1 do not mean to be prejudicial to the work of my Scotch 
predecessor (Mr. Johnston). The salmon of the Tay formed the main object of his 
investigations and he pointed out himself, that it was not probable that the statis¬ 
tical particulars of Tay scales would apply even to those of all other rivers (of 
Scotland) without any modification. 
