634 U. S. BUREAU OF FISHERIES 
The habits, in so far as the life in fresh water and the life in the ocean are 
concerned, are quite different in the various species. The simplest type of life 
history is that of the pink salmon. The fish spawn in fresh water, the young 
go out to the ocean as soon as they emerge from the gravel and remain out 
there until they are 2 years old, when they return to fresh water to spawn. 
We have an invariable habit here—the fish always being 2 years old and having 
spent the greater portion of that time in the ocean. 
The chum has, perhaps, the next most complicated life history. It has the 
same habit as the pink in going down to the ocean as soon as it emerges from 
the gravel, but it does not invariably return at 2 years of age, but at anywhere 
from 2 to 5, or possibly 6 years. 
The silver salmon presents a still different variation. It remains, almost 
invariably, in fresh water for one year and then goes out to the ocean. It 
remains there for two years, returning as a 38-year-old fish. There are some 
yariations from that, but the silver salmon, on the whole, is a 3-year fish. 
There remain the red salmon and the chinook. The red salmon, with very 
few exceptions, remains in fresh water for at least 1 year but anywhere from 
1 to 4, and then migrates to the ocean. It remains in the ocean for from 1 to 3 
or 4 years and then returns. It varies in age, at the time of its return from 3 
years (or possibly 2) to 7 years. The predominating age varies in different 
streams. In the Fraser River the predominating age is 4 years. In the Karluk 
River the predominating age is 5 years. 
The chinook salmon presents other complications and ranges in age, at the 
time of maturity, from 2 to 7 or 8 years, the same as the red salmon, but, 
especially toward the southern portion of its range, has a marked tendency to 
migrate some time during its first year. It may migrate seaward immediately 
after emerging from the gravel, or, on the other hand, it may remain in until 
it is 1 year old or more. Spawning takes place in the fall. Many of the salmon 
do not migrate until the second spring following spawning, when they are ap- 
proximately 18 months old, counting from the time the eggs were laid. In the 
Columbia River, particularly, which is the most important chinook-salmon 
stream on the west coast, the time of seaward migration is very variable and 
one may find seaward migrants in the lower portion of the Columbia River 
in almost every month of the year. 
There are two well-known features of the life history of the salmon that have 
molded our conception of the need of conservation measures to a very marked 
extent. One of these is the fact that the salmon invariably die after spawn- 
ing. That is a matter beyond question. The second, and most important, is 
the fact that the fish almost invariably return to the stream from which they 
came as little fish. In other words, what is known as the “ parent-stream 
theory ’”’ is more than a theory; it is unquestionably a fair statement of fact. 
As a result of this habit, there has become established in all of the streams, 
and even in the tributaries of the larger streams of the coast, races of salmon. 
These may not always be distinguishable by physical characteristics (in fact, 
they resemble one another in everything, so far as any structural peculiarities 
or general habits are concerned), but we have every reason to believe that even 
in those cases where there is no possibility of distinguishing there are races, 
or incipient races, or something of that sort—self-perpetuating colonies of sal- 
mon in all of the separate streams and presumably in all of the main tribu- 
taries. We have considerable experimental evidence to support this, resulting 
from an extensive series of marking experiments and the evidence derived from 
scale examinations; and there is evidence, which has never been very carefully 
sought but which could undoubtedly be worked out in considerable detail, of 
structural differences. ~ 
The present problems of the salmon fishery are rather different from those of 
many of the other fisheries of which we have heard during the past two days. 
We know fairly well the general features of the life history of the salmon. 
We have methods whereby we can get any particular detail in the life history 
of any of the races or groups of races of salmon, if we want it—-such as the 
age at maturity, the migration routes, and things of that sort. We have many 
data and methods well worked out whereby we can secure any particular item 
that we may happen to need. Furthermore, we have fairly reliable statistical 
records of the catches over rather long periods. In some places the statistical 
records are not as reliable as they should be; but in other places, particularly 
in some of the more important salmon-spawning areas in Alaska, the records 
are fairly good, and we feel that we are now in a position where we can begin 
