THRISSO MORPHS. 
827 
the series extremely difficult. Within the Scandinavian 
fauna the series contains the following families. 
I: Hind part of the back furnished with an adipose fin. 
A: Ovaries without special oviduct com- 
municating directly with their cavity : 
the ripe eggs fall into the abdominal 
cavity and pass thence through a geni- 
tal pore. No phosphorescent spots on 
the sides of the body Farm Salnionida • 
B: Ovaries furnished with a complete ovi- 
duct. Sides of the body furnished in 
all Scandinavian forms with phosphore- 
scent spots'* Fam. Scopelidce. 
II: No adipose fin Fam. CAupeidae. 
Fam. S A L 31 0 X I I) E. 
Body of the typical Salmon form or more compressed , approximated to the Boacli or the Herring form. Scales 
middle-sized or small, rather firmly attached 1 '. No phosphorescent spots on the sides of the body. Dorsal margin 
of the tail furnished with an adipose fin. Margin of the upper jaw formed in front by the intermaxillaries 
and behind by the maxillaries. No barbels. Air-bladder simple and not connected with the cranial cavity. 
Branchial cavity furnished with pseudobranchice. Ovaries open on the outside, without oviducts or with oviducts 
opening inwards ( into the abdominal cavity). 
The family of the Salmons and Gwyniads is well- 
known everywhere in the frigid and temperate regions 
of the northern hemisphere. In the southern hemi- 
sphere, on the other hand, only two genera occur, a 
Capelin form ( Betropinna Richardsonii ) belonging to the 
fresh water of New Zealand, and two (?) Argentine 
species, marine iishes of the same locality. The family 
possesses more than ordinary interest both from an eco- 
nomical and a scientific point of view, to the economist 
because the flesh of these fishes is one of the most 
highly esteemed and most remunerative foods, to the 
scientist because there is hardly any other part of the 
system where he is confronted with such difficulties in 
defining the limits of the species. This is due in most 
cases to a plasticity and inconstancy of form, caused 
beyond doubt by the periodic migrations of the Salmo- 
noicls from the sea or the great lakes to running fresh 
water. The economical value of these fishes has also 
contributed to the said difficulties. Where they appeared 
at certain seasons and certain places in enormous mass- 
es, with one or other of their various dresses predomi- 
nant, the fishermen gave these dresses different names 
which the systematist had to explain, often merely after 
examining solitary specimens that were supposed to 
represent distinct species. “Kart barn har manga namn"% 
says the Swedish proverb; and this rule has repeatedly 
asserted itself in the nomenclature of the Salmonoids. 
The investigations of recent years, however, have elu- 
cidated the significance of the variation of form within 
this family and the relationship to each other of the 
more or less constant forms, arid have thereby thrown 
light upon the comprehensive question of the origin of 
the species. 
The geological researches of Agassiz d have traced 
the Salmonoid family back to the Cretaceous Period. At 
that time there lived, where the green sand now lies in 
modern Westphalia, and where the chalk subsided over 
Southern England, several forms, which Agassiz referred 
to the genera Osmerus and Osmeroidesj the latter fur- 
nished ■with scales resembling those of the modern ge- 
nus Argentina. The evolution of the Salmonoid family, 
we might hence conclude, began with marine fishes of 
the Smelt and Capelin type, with few pyloric appen- 
dages or none at all, and its development advanced in 
two directions, 1) towards the Salmon and Charr, fur- 
nished with teeth and with comparatively small scales, 
2) towards the toothless and comparatively large-scaled 
genus Coregonus. In both these directions of evolution 
the pyloric appendages were more and more developed, 
to assist these fishes in storing up a reserve supply of 
a An extremely rare form, Sadis atlanticus , is an exception. 
b The genus Salanx ( Albula chinensis , Osbeck, Ostind. Resa, p. 237), however, is scaleless or has small, extremely thin, and deci- 
duous scales (Gunther). 
0 “A dear child has many names.” 
d Rech. Poiss. Foss., vo). V, p. 103. 
e Gunther, however, considers that this genus perhaps belonged to the Scopeloid family. Introd. Study Fish., p. 582; Handb. 
Iclithyol., p. 417. 
