of the Fishery Board for Scotland. 123 
first and the last meet but do not overlap ; they each overlap the range of 
variation of this character in minutus. In the distance of the opercular 
cleft from the anterior end of the body we obtain the same average in 
each species, and in respect to the hind edge of the operculum, while 
the averages differ to the extent of 2 per cent. between minutus and 
esmarkii, each of these differs from /wscws by | per cent. only. 
The position of the pectoral fin is practically the same in all three 
species in respect to the distance of the base of the first fin-ray trom the 
anterior end of the fish. The distance of the anus from the snout is a 
diagnostic difference between duscus and minutus. In the average, the 
position is at 26 per cent. in the former and at 32 at the latter; the 
ranges of variation do not overlap. In esmarki the anus is on the 
the whole slightly farther back than in minutus, with, of course, a corres- 
pondingly greater divergence from Juscus. Between esmarkit and 
minutus the ranges of variation overlap. 
In the position of the first dorsal fin, there is very close agreement, and 
overlapping. 
In the second dorsal fin all three overlap, both in respect to the point 
at which the fin begins, and also in the position of the end of the fin, 
Luscus and minutus have a longer fin-base than esmarkit. 
In respect to the beginning of the third dorsal fin, they are all equal, 
but as Juscus has a shorter fin-base than the two others—the fin ends in it 
a little farther anteriorly than in the other two. 
The first anal fin commences close behind the anus; it ends first in 
esmarkii, and then a little distance farther posteriorly in minutus and 
luscus. 
In the matter of the second anal all three species agree, both in its 
commencement and its end, 
The caudal peduncle is on the average smaller in Juseus than in 
esmarkit and minutus, but the ranges of variation overlap... 
The bend in the lateral line rises farther back in dwscuws than in the 
other two. It is usually just behind the middle of the length of the 
body ; in no case in Juscws did it commence on the anterior half of the 
body ; in minutus and esmarkii it did in some cases begin just on the 
anterior side of the middle of the body. ‘The lateral line rises more 
rapidly in duscus than in minutus and esmarkit, 
The general relation between the three forms in respect to the different 
characters selected has been briefly discussed, but these relationships 
expressed in this form are of themselves of no value from the point of 
view of specific description. These characters may and do indicate 
where specific characters may be found, Fora specific description it is 
necessary that the selected characters be tested on one fish. A specific 
character which infers comparison with a fish of another species is of 
only secondary value. ach character must be expressed in terms of the 
individual fish. The size of the eye, for example, may be of specific 
value when it is stated in terms of the length of the snout, or of any 
other part of the fish, while it is of no value when compared to the size 
of the eye in another species. 
The position of the anus is a very good guide in diagnosing certain 
species, and it is important in the present case. The form in which 
it is represented in the Tables, as being situated at a certain per- 
centage of the length of the fish from the tip of the jaws, is not a useful 
one for a specific description. It should be expressed ina relationship 
that is more readily measured, aud for this the relation between the posi- 
tion of the anus and the situation of the first dorsal fin. Now, an 
examination of the measurements of the distance of the two points shows 
