Notes on Luvarus imperialis Raf., a fish new to the fauna of Norway. 25 
being 1:5, but this proves nothing; the figure in other respects 
is not sufficiently correct, for instance the shape and size of the 
tail, and the body is furnished with large scales, in contradiction, 
hewever, to the printed description. The ventral is represented as 
consisting of a scallopped spinous ray, while in the formula for 
the fins we find V I, 4. If the latter be correct, Ausonia would form 
a connecting link between Luvarus and Astrodermus, the vertical 
fins corresponding to those of the former species, the ventrals to 
those of the latter. Notwithstanding that as yet no specimens 
have been found showing stages between Astrodermus and Lu- 
varus with regard to the reduction in the number of the rays in 
the vertical fins, I am bound to conclude with Gioglioli, who first 
made the suggestion, that the Astrodermus forms are stages of 
development of Luvarus. I base my conclusion on the comparisons 
I have been able to make between my specimen of Luvarus imperi- 
alis and the two specimens of Astrodermus elegans. Risso’s state- 
ments must be further confirmed before one can place any reliance 
upon them. 
The Systematic Position of Luvarus. 
Luvarus imperialis, with regard to system, is an interesting 
form, about whose systematic position there has been much diver- 
gence of opinion. It has been classified as belonging to the 
Scombridae and the Coryphaenidae, but is now referred to a separate 
family Luvaridae Gill, of which it is the only species, and it is 
generally considered to be nearly related to the group of mackerels. 
Regan (1902) has pointed out great similarities in the skeleton 
and the visceral anatomy of Luvarus and the fam. Acanthuridae, 
and building on these points of resemblance, he has suggested 
that Luvarus should be considered to be an ultra-specialized 
Acanthuride; its position in the system would then be altogether 
different viz. among the Squamipinnes, which only consists of 
tropical fish. Regan has, however, later (1903) given up this 
idea, after the publication (1902) of Waite’s account of ihe 
skeleton of Luvarus and Regan himself had the opportunity of 
examining a perfect skeleton of this species. On comparing it with 
the skeleton of Thynnus, Re gan comes to the result that Luvarus, 
