NOTES ON A COLLFXTION OF TIDE-POOL FISHES EROM KADIAK ISLAND 
IN ALASKA. 
By CLOUDSLEY RUTTER, 
Assistant^ United States Fish Conunission. 
The collection on which the following report is based was made by the writer in 
May, 1897, mainly at Karlnk, Kadiak Island, Alaska. The beach there is eomi)osed 
of bowlders more or less overgrown with fiicus and kelp, or of coarse gravel, and is 
exposed to the open sea, there being no harbor. In some places the tide recedes as 
inucli as 100 yards, in which case there are a few true tide pools. Most of the tishes, 
however, were obtained among the bowlders. Just west of Karluk Head are a few 
good tide pools, and there a new species of JSfeoliparis was found. 
Uyak Bay contains a number of good collecting-i>laces, one near the Pacific Steam 
Whaling Company’s cannery and another near the mouth of Larsen Inlet. At both 
places the rocks are cavernous, and good pools, with seaweed, are left at low tide. 
Many specimens were obtained at the foot of an island on the east side of the bay, 
Avhere a broad rocky beach is exposed at low tide. 
Some collecting was done in Alitak Bay, at the southern end of Kadiak Island, 
but it is a very unsatisfactory place. 
The collection is now in the museum of Leland Stanford Junior University, where 
it was studied. 
The descriptions of the new genus and species, which this collection contains, 
were first published in Jordan & Evermaiin’s Fishes of JSiorth and Middle America. 
Notes on the fresh-water Coitus and the sticklebacks found in Karluk and Alitak 
lakes are included. 
1. Pygosteu.s pungitius (Liiintens). Nine-S2)ined Stlcklehack. 
Two specimens from <a brook near the moutli of Afitak Bay agree witli the variety hraolni- 
poda ill having tiie dorsal spines x-i and in having the ventral spines one-third length of 
liead, but it may be that the same tendency to variation exists in this s])ecie8 as in the 
following species of Gasterosteus, in Avhich case the shortening of the S 2 )ine 8 in fresh rvater 
will hardly justify systematic distinction. 
2. Gasterosteus cataphractus (Pallas). SUcklehack. 
Comnnm in the lakes and in Karluk estuary. From Karluk estuary 157 specimens Avere 
obtained, which shoAv all possible gradations between the tyjiical catapliractiis, with its long, 
slender, and fully .armed body, slender and acute spines, narrow pubic jilate. and thoracic 
area, and the form heretofore known as microcephaius, with its heavy and but i)a.rtly armed 
body, short, he.avy spines, and broad jnrldc plate and thoracic .area. Of the 157 specimens 35 
have the sides but jiartly plated, and only about as many more can bo distinctlj' classed as 
typical cataphractus. Those of the lakes are all of the form microcephalns. 
The two southern forms will have to be knoAvn as Gasterosteus cataphractus microcephalns and 
Gasterosteus cataphractus williamsoiii. It seems evident that specimens typic.al of microcephulus 
are iudeiiendently derived from the iiarent species in dift'erent localities. It seems more 
probable that the parent species h.as entered the fresh water in California and at Kadiak 
Island and iude 2 )eudentlj' changed into the fresh- water form in the two jjlaces than that the 
variety has kept distinct from tlui i^arent species while migrating from one pl.ace to the other. 
It is even more 2 >robablc that the i)arent si>ecios hiis entered the two lakes, K<arluk and Alitak, 
on Kadiak Island, and 2 )roduced the fresh-water form indeiiendently, than that the A'ariotv 
189 
