122 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
variations,” the number of rays in these streams decreases with the altitude and 
swiftness of the stream. Moreover, the Pacific streams of South America have still 
shorter and presumably still swifter streams, and no such modifications are seen in 
the fishes inhabiting these waters. 
The most striking case, that of Lenciscm (Richar(lso}dus) is explained more fully 
in the cliapter on local variations. In the vSubgenus Richardsonius, confined to the 
Columbia and to the Fraser systems, the number of anal rays varies from 12 to 25, 
an increase of from 2 to 15 rays over LeuGlseus, some of whose species have also 
reached the headwaters of the Columbia, but whose usual habitat is the Atlantic 
slope. The genus Oiicorhiinchm has a similar increase of anal rays over Salmo and 
8((lveU)ius, which are genera of wider distribution, some of the species being found 
on the Atlantic, some on the Pacific, and some on both slopes. On the other hand 
Thiimallus has a larger number of dorsal rays than any Pacific slope species. 
The change from rays to spines is seen in Archoplitcs^ Meda, etc. It is most 
strikingly marked in the change from Fercopsis to Cohmhia, the only known genera 
of the PercopHidK. The former is confined to the Atlantic, the latter to the Pacific 
slope. In the former, feelile unsegmented rays at the beginning of the dorsal and of 
the anal are developed into strong spines in the latter. Long ago Prof. Cojie* 
noticed a similar modification as to spines in Meda. Prof. Cope says: 
As one of the most valuable results derived from a study of the collections, it a])pears that the 
basin of the Colorado River is the habitat of a small group of fishes of the family which may 
be called the Plarjojjterin c, which embraces three genera — Placjoptevns Cope, Lepidomeda Cope, and Meda 
Girard. The group differs from others of the family in the possession of two strong osseous rays of 
the dorsal fin, the posterior of which is let into a groove in the hinder face of the anterior without 
being coossified with it, thus constituting a compound defensive s]>ine. The rays of the ventral fin, 
excepting the first anil second, are similarly modified. The greater part of their length consists of 
an osseoTis dagger-shaped spine, with grooved posterior edge, which overlaps the border of the succeed- 
ing ray, when the tin, like a fan, is closed up. The articulated jmrtion of the ray either emerges from 
the groove below the free acute apex of the spine, or appears as a continuation of the apex itself. 
* « Interest attaches to the Plar/opiteriiiw as the only type of fishes not known from other waters 
than those of the Colorado and San Luis basins. 
An iutei'estiiig condition is seen in Jli/steroccopus, the only fresb-water genus of 
the Emhiotocidce. It is confined to the Sacramento Basin and has 10 to 18 dorsal spines, 
as compared with 8 to 11 in the many marine genera. Unfortunately this is the only 
available example of the change from salt to fresh water. 
I give here a detailed comparison of the rays of the Pacific fishes as compared 
with their Atlantic relatives, from which it will be seen, as stated above, that in every 
family the modification is noticeable, although in maii}^ cases it is minute. As far as 
X>ossible the western and eastern rexu’esentatives of the same forms are x)laced ox)X)osite 
to each other. 
AClPENSEKlUm. 
Species 
Pacific slope. 
Species. 
. 
Atlantic slope. 
Dorsal. 
Anal. 
Dorsal. | 
Anal. 
Acipciiser transmontamis. . . 
44-48 
28-30 
• 
Acipenser sturio 
38 
27 
Acipeuser metlirostris 
33-35 
22-28 
Acipeuser rubicnudiis 
35 i 
26 
Acipenser brevirostris 
41 
1 
22 
* Cope & Yarrow, Wheeler’s Surveys, chaj^ter vi. Report upon the Collections of Fishes made 
in portions of Nevada, Utah, California, Colorado, New Mexico, and Arizona. 
