12 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
Shtimagin Bank. — This hank lies to the south and southeast of the Shumagin 
Islands, and its outer margin follows approximately the trend of the coast line formed 
by the adjacent islands. It has been traced westward to about longitude 159° 52' W., 
but probably extends farther in that direction. East of the Shumagin Islands it 
reaches north to the latitude of Big Koniushi Island. Its width inside of the 100- 
fathom curve varies from 15 to 35 miles, while its area has been estimated at about 
1,800 square miles. The depths over a large part of the bank are less than 50 fathoms, 
and the bank is not separated from the islands by deep water. 
Shumagin Islands to Kadiak Island. — Only a single series of soundings was carried 
across this wide area to the eastward of Shumagin Bank, with a double line extend- 
ing from the neighborhood of Light house Bocks to Mitrofania Bay. These sound- 
ings were insufficient to demonstrate the full value of this region, but they indicated 
the existence of several fishing banks the outlines and characteristics of which must 
be left for future investigations. The extent of the area thus partly developed was 
estimated at about 4,400 square miles. 
Albatross Bank. — This bank lies off the southeastern side of Kadiak Island and 
extends the entire length of that island and also in front of the Trinity Islands. At 
the eastern end it is practically continuous with Poi tlock Bank. Along some por- 
tions of the coast, as in the neighborhood of Sitkalidak Island, the bank is separated 
from the land by comparatively deep water, while in other places shoal water inter- 
venes. The 100-fathom curve is distant 25 to 45 miles from the land, inside of which 
limit there is an estimated area of 3,700 square miles. The existence of this bank was 
liredicted by Prof' George Davidson upon the evidence of a few isolated soundings, 
which were the only ones that had been made previous to the investigations of the 
steamer Albatross, from which it has derived its name. 
Portlock Bank. — This is the largest single bank that has yet been discovered on 
the Alaskan coast, its area inside of the lOO fatbom curve being about 6,800 square 
miles, or only 1,600 square miles less than that of George’s Bank, the second largest 
of the great banks of the western Atlantic. It extends northeastward from Kadiak 
Island, in the direction of Middleton Island, a distance of about 120 miles, and is irreg- 
ular in shape. Isolated soundings of 68 to 81 fathoms occur near Kadiak Island, at 
the western end of the bank, but there are no indications of a marked or extensive 
depression between the bank and the land. 
From Portlock Bank the soundings were carried to Middleton Island, the position 
of which was ascertained by a careful series of observations on a clear day. The 
reported position of Pami)lona Bocks, in latitude 59° 03' N., longitude 142° 40' W., was 
next visited, but only deep water was found within a radius of 20 miles of this locality. 
The Albatross then proceeded to Washington Territory. 
Character of the bottom on the fcanZcs.— Sand was the predominant material com- 
posing the bottom on these several banks, a gray sand being the most common. This 
was combined in many places with pebbles, gravel, or broken shells, which were also 
recorded separately in some localities. Mud rarely occurred upon the banks or any- 
where inside of the lOO-fathom line. Bocks were not found upon Davidson Bank, but 
on Sannakh Bank they compose a large iiart of the bottom, even in the deeper sound- 
ings. Bocky patches are numerous on Shumagin and Albatross Banks, but were 
observed only at the extreme western end of Portlock Bank, near Kadiak Island. In 
