PACIFIC SALMON

Hatchery Propagation and Its Role
In Fishery Management

CIRCULAR 24

FISH AND WILDLIFE SERVICE
UNITED STATES DEPARTMENT OF THE INTERIOR

Abstract

Population growth and industrial in-
crease have intensified the problems of
salmon-fishery maintenance. Natural
propagation has been interfered with by
pollution and by dams that cut off the
salmon from their natural spawning
grounds. Hatchery propagation helps
maintain the fishery by offsetting the loss
in natural spawning. This review of
salmon-hatchery operations describes the
life history of the Pacific salmon and ex-
plains the equipment used and the methods
followed in hatchery propagation.

PACIFIC SALMON

Hatchery Propagation and Its Role
In Fishery Management

By William Hagen, Jr.

Circular 24

United States Department of the Interior, Douglas McKay, Secretary
Fish and Wildlife Service, John L. Farley, Director

UNITED STATES GOVERNMENTiPRINTING OFFICE : 1953

For sale by the Superintendent of Documents, U. S. Government Printing Office, Washington 25, D. C.

Price 30 cents

Contents

Page

Tlu> Pacific- saliiK.n 3

Life hist ory 3

("iiinook 8

Rod 9

Silver 10

Pink 11

Chum 11

Steelhead trout 12

The liomiug instinct 12

Economic value 14

Hatchery propagation of Pacific salmon 16

Early history and objectives 16

Circumstances reciuiring hatchery propagation 16

Methods 22

Spawn taking 25

Incubation of eggs 28

Rearing 30

Diseases 32

Foods for hatchery salmon 33

The hatchery and fishery management 37

The salmon hatchery 39

Hatchery water supply 39

The hatchery building 42

Troughs 42

Food preparation 44

Food storage 45

Rearing ponds 46

Trapping adult salmon 48

Adult holding ponds 51

Personnel, equipment, and facilities 53

Literature cited 54

Photographs: Fig. 1 by Dick Black; fig.2by \V. F. Kuhichek (FWS); flg.Sby G. Kelez (FWS);
figs. 4, 9, 10, by H. F. Kelly (FWS); figs. 5, 13, 14, 20, 21, 22, 26, 27, by E. B. Horn (FWS); figs 6,
11, 23, 25, by Fish and Wildlife Service; figs. 7, 8, by Corps of Engineers; figs. 12, 15, 16, 17, by
Bureau of Reclamation; figs. 18, 19, by Z. Parkhurst (FWS); fig. 24 by W. Hagen, Jr. (FWS).

^

H

mm

.-J

Figure 1. — Coleman (Calif.) Salmon Hatchery of the Fish and Wildlife Service. In
foreground, diversion dam in stream, fish ladder to adult-salmon ripening ponds,
fingerling rearing ponds. Buildings from left to right : equipment and shop, cold
storage and food preparation, and the hatching building. The waste-water channel
back to the stream serves also for passage of young fish released from the rearing
ponds.

IV

PACIFIC SALMON

Hatchery Propagation and Its Role
In Fishery Management

The salmon of the Pacific coast
have alwaj^s been an important food
source for man. Craig and Hacker
(1940) give an excellent account of
the dependence of the Indians upon
the salmon of the Columbia River,
and estimate that before the arrival
of the white man the catch was
about 18 million pounds annually.
Eostlund (1952) also indicates the
importance of the salmon to all the
Indian tribes of the western shores
of North America. The early set-
tlers relied upon the resource for
food, and later exploiters took the
salmon in great numbers for com-
merce.

Throughout the development of
the Pacific Northwest and Alaska,
salmon have played an important
part. Today, the salmon fisheries
are essential in the economy of
these two regions. The resource
contributes a substantial bulk of
protein food.

Population growth and indus-
trial increase have intensified the
problems of salmon-fishery mainte-
nance. Streams required for the
natural reproduction of the species
have been polluted and dammed for
irrigation and the production of
electric power; watersheds have

been denuded of forest cover result-
ing in floods ; and the rivers gener-
ally have been so utilized as to deny
the original users — the salmon —
adequate spawning area for self-
perpetuation. Overexploitation,
too, has depleted the stocks.

Man's inevitable progress has had
a damaging impact upon the salmon
resource. There have been rela-
tively minor attempts to counteract
or mitigate the damage.

Artificial or hatchery propaga-
tion of salmon on the Pacific coast
was first undertaken in the ISTO's,
when efforts were made to establish
runs of various species in other
areas of the United States and in
foreign countries. In later years,
salmon-hatchery operations have
had as their objective the mainte-
nance of salmon runs in streams
where overfishing, pollution, dams,
and other factors have decimated
populations. The operation of a
present-day hatchery is considered
successful only if its output of
young fish contributes to the man-
agement of a fishery and results,
either directly or indirectly, in the
maintenance or increase of catch by
sport or commercial interests.

On the Pacific coast, industrial
advancement and its demands upon
water for power lias discouraged
and will further discourage the nat-
ural reproduction of salmon. The
need for the hatcheiT-reared sal-
mon to overcome, at least in part,
the manmade deterrents to natural
reproduction is becoming ever more
apparent.

The conservation of the salmon
resource is a public responsibility
and, as such, is of concern to the
agencies of the State and Federal
govermnents. In the Territory of
Alaska, tlie United States Fish and
AVikllife Service is charged with
jurisdiction over the fishery; in the
coastal States, the salmon fisheries
are regulated by the States and con-
servation activities are undertaken
by both the Service and the States.
It is not possible for a private indi-

vidual or a group to control the
catch of salmon resulting from nat-
ural or artificial reproduction in a
given stream. This is true because
the catch is from public waters, ei-
ther the open ocean or the mouth of
the stream. However, in years past
some packers of salmon have under-
taken to improve salmon runs by as-
sistance to natural reproduction and
through artificial propagation.
Probably the earliest such eftort was
by Hume (1893).

This publication describes briefly
the species of Pacific salmon and
their characteristics, reviews jjres-
ent-day methods of the artificial
propagation of the Pacific salmon,
outlines the human factors endan-
gering the maintenance of the fish-
ery, and explains the role of the
hatchery in the attempt to maintain
the fisheries at their highest possible
level.

FiGUlil

Indiana aL Culilo Falls uii llie Columbia Jtivt'r lisliiu.^- i\ir .salmon. TLl
Dulles Dam Keservoir will eliminate this ancestral fisliing site.

THE PACIFIC SALMON

Five species of salmon tire native
to botli North Pacific Ocean coasts,
from Monterey Bay in California
and from northern Japan on the
Asiatic coast, northward to and
somewhat beyond Bering Strait be-
t ween Alaska and the Soviet Union,
One additional species that is not
native to the North American coast
is found in northern Japan.

Attempts have been made to in-
troduce various species of Pacific
salmon into waters of the eastern
coast of the United States, as well
as into waters of many foreign
countries and the Hawaiian Islands.
Chinook and red (sockeye) salmon
have been planted in waters of 17
Gulf and Atlantic Coast States.
Introductions of the species have
been successful only in southern
Chile and in southern New Zealand,
although temporary success has at-
tended the efforts to establish chi-
nook, sockeye, and silver salmon in
the coastal streams of eastern Can-
ada and northeastern United States.
These efforts and successes, as well
as the factors apparently limiting
the habitat of Pacific salmon, are re-
ported by Davidson and Hutchinson
(1938). No substantial and suc-
cessful efforts to transfer species
have been undertaken since their
report.

]\Iany lakes in Canada and in
northern United States contain
landlocked red or sockeye salmon.
Within the natural range of sockeye
salmon, many individuals remain in
the lakes by choice; whereas else-
where the species has been stocked
in lakes not accessible from the

ocean. Tliese lake-dwelling forms
generally have the same life cycle
as their ocean-reared brothers but
usually attain not more than half
their size.

Life History

All the species of Pacific salmon
are anadromous, that is, they spend
a portion of their lives in the ocean
and at maturity ascend fresh- water
streams to spawn. All have the
same general pattern of life. Early
life is in fresh water ; growth to ma-
turity is in the ocean ; and maturity-
is followed by the return to the orig-
inal stream to spawn and die. There
is no authenticated record of a
Pacific salmon surviving the first
spawning and returning to the
ocean. There are variations among
species and within species with re-
gard to the length of time spent in
fresh-water and in salt-water habi-
tats. Man-made conditions have
resulted in considerable deviation
from the natural schedules of salm-
on, often with disastrous conse-
quences to the population of salmon.

"When salmon enter fresh water
OTi their spawning migrations, they
cease feeding. Pink and chum
salmon, being "short-run" species,
may stop taking food while still
in salt water and some distance
from their home streams. Chinook
and silver salmon, the principal
sport species, strike at lures prob-
ably in annoyance rather than as
possible food after entering the
brackish and fresh waters of their
streams. Sustenance for the fish
during the journeys up the river

is iDrovided by the body fats and
tissues, which also provide ma-
terials for the development of
rej)roductive products — the eggs or
ova of the female and the
spermatozoa of the male. The lat-
ter, however, are usually fully de-
veloped when the fish enters fresh
water, whereas the eggs of the fe-
male become "ripe" or mature at
about the time the fish arrives at the
spawning area. The flesh of the
adult salmon deteriorates rapidly
after the fish enters fresh water, as
the body is being drained of its sus-
taining materials for the survival of
the fish and the development of re-
productive products, while there is
no intake of food for conversion to
energy. Thus, the salmon taken at
the mouths of the rivers, or offshore
in the ocean, are much more desir-
able for human consumption than
are those taken uj)stream in the
rivers. Fish taken offshore or at

the mouths of rivers have firm flesh
of good color but the fish taken up-
stream have flesh that is soft and off
color. There is variation among
and within the species in this
respect.

Durmg the spawning migration
in fresh water, the adult salmon will
proceed quite rapidly upstream.
The rate of progress usually is from
3 to 10 miles or more a day. Chi-
nook and red salmon in the larger
streams of Alaska may spawn at the
extreme headwaters 1,500 or more
miles from the ocean, and the same
was true in the Columbia River be-
fore the construction of dams.
Chum and pink salmon usually
spawn within a few miles of the
ocean and often just above the reach
of the tides. Here, too, there is con-
siderable variation among and
within species.

Despite the much-publicized abil-
ity of adult salmon to leap vertical

"\

Figure 3. — Salmon leaping at falLs.

distances of 10 or more feet, it has
been established that waterfalls of
this height will quite effectively
block the upstream migration of
most salmon. Even a waterfall of
5 or 6 feet will stop the majority of
salmon unless water-flow conditions
are ideal. Fish ladders for salmon
at dams rarely require the fish to
leap a vertical distance of more than
2 feet, and most ladders are so ad-
justed as to permit the fish to swim,
rather than jump, a rise of about
1 foot.

Upon arrival at the spawning
area the female deposits her eggs in
a nest, or redd, which she digs in
the gravel of the stream or, in some
instances, in shallow lake-shore wa-
ters. Burner (1951) describes the
characteristics of the redds, as well
as their preparation and the activi-
ties of the fish :

During the prespawning state, the fe-
male salmon is green, that is, the eggs are
neither ripe nor loose in the ovaries.
Males are seldom in attendance, and are
frightened away by the female, who re-
pels all intruders of either sex. The fe-
male digs the redd as she turns on either
side, at an angle of about 45° to the cur-
rent, head upstream, body arched, and
makes a series of violent flexions with
body and tail. The tail strilies gravel oc-
casionally and the strong-boiling current
created carries gravel and silt a short
distance downstream. This material
spreads out in a flat semicircle at first;
then, as the digging upstream proceeds,
it collects into a loose pile called the tail-
spill. With more digging, the redd as-
sumes a long oval shape about twice the
length of the salmon and several inches
deep. The prespawning digging of the
redd may go on for as many as 5 days.

At the beginning of the spawning stage,
the nest is ready for the eggs. All loose
gravel and fine material have been re-
moved from the pot, or center of the redd,

the shape of which is such that any cur-
rent in the bottom flows upstream then
upward and outward. Usually there re-
main in the pot large stones too heavy
for the fish to move far, and the crevices
between these rocks provide excellent
lodgement for the eggs. Males are con-
stantly present now. The female alter-
nately digs at the redd and settles back
into the depression to release eggs. A
male then moves quickly alongside the
resting female, curves his body against
hers, and releases sperm in a small milky
cloud that settles briefly in the bottom
of the redd where the eggs are lodged.
The newly deposited eggs are thus sur-
rounded by sperm and eventually fer-
tilized. Excess sperm is carried slightly
upstream along the bottom of the redd
and gradually carried away by the cur-
rent. During the spawning stage the redd
inci-eases considerably in length and
depth, and appears to move upstream as
a result of the continued digging at the
upstream wall and the filling in of the
tailspill area.

The postspawning stage begins after
the female finishes depositing her eggs.
Males are no longer attentive. The fe-
male is gaunt and spent, but she
continues to dig at the gravel with ever-
weakening efforts until she dies. This
postspawning digging, which may con-
tinue for 10 days, becomes shallow, off
center, and ineffective. The area of the
nest is increased without (after the first
day at least) adding to the protection
of the eggs.

Each salmon female produces
from 2,000 to 5,000 eggs, the number
depending upon the s^^ecies and size
of the fish. The time required for
the eggs to hatch is regulated by the
temperature of the water. One
measure that was devised by Seth
Green for brook trout was 50 days
at 50° F., and plus or minus 5 days
for each degree less or greater than
this average. This formula can be
applied only as a general guide for
salmon. Newly hatched fish, called

259307—53-

alevins or fry, live in the gravel of
the redd, where they are somewhat
protected from their larger enemies.
They gradually absorb the food in
the attached abdominal yolk or
umbilical sac. Upon almost com-
plete absorption of the yolk sac, the
young fish emerge from the gravel,
usually in late winter or early
spring, and seek food. The young
of some species almost immediately
start downstream toward the ocean,
but others remain in fresh water for
a year or more. The young of pink
and chum salmon usually enter salt
water soon after emerging from the
gravel. The young of the fall
chinook salmon may also start down
very soon after emergence. The
fingerlings of spring chinook, red,
and especially silver salmon may
remain in fresh water a year or
longer.

Why some species of salmon enter
salt water soon after emergence
from the gravel and why other
species remain in fresh water for
extended periods, is imperfectly
known. Clemens (1952) believes
that the behavior of the fish at all
ages is the result of a series of inter-
actions between the fish and the en-
vironment. Seaward migration of
red (sockeye) salmon fingerlings
from the Fraser River (British Co-
lumbia) is described from this view-
point. Then, too, there must be
some physiological changes in the
organism. According to Black
(1951), "Young salmon entering
the sea must have adequately de-
veloped chloride-secreting cells in
the gills to survive."

Studies indicate the heavy mor-
tality among young salmon during

their periods in stream and ocean.
Neave (1953) shows that, of young
pink salmon spawned, 13.1 to l.G
percent enter the ocean, and of
those entering the ocean about 2
percent survive to return as adults
to their home stream. The mortal-
ity of other species seems to be about
the same. Under completely bal-
anced conditions, two fish of each
brood, a male and a female, would
have to survive to spawn^ — repre-
senting 100-percent effective main-
tenance of the species.

There are years of too few fish and
years of too many fish for the
spawning areas available, but with-
out outside influence, the runs
would tend to stabilize themselves.
As in the maintenance of all living
things, one species lives upon, and
is controlled by, other species to pro-
vide a balance throughout the natu-
ral environment.

It is well that in nature's scheme
of survival the salmon produces so
many eggs and resulting offspring,
for through the stages of its life
the salmon is preyed upon by in-
sects, birds, reptiles, mammals, and
fishes. The eggs in the nest provide
food for bottom organisms such as
crayfish and the water forms of
some insects. The fry are taken by
birds, other fishes, crayfish, flies,
and even by snakes. The young fin-
gerling on its way to the ocean runs
a gantlet of larger fish and birds.
In the ocean, other fishes and mam-
mals such as seals and sea lions prey
upon the salmon. When returning
to the river, the adult fish is sub-
jected to attacks by birds and bears
in the shallow rapids and on the

6

'^

•

Figure 4. — Stages of development of silver salmon {Oncorliynchus kisutch ) from eyed
egg to feeding fingerling. Approximately twice normal size.

spawning beds. In addition, death
from other natural causes, inchid-
ing- disease, reduces the brood. Top-
ping all of these hazards are man's
nets and hooks, and his dams and
pollution, which take a large toll of
the remnant of the original brood.
Fortunately, nature was so very

generous in the stock of fish surviv-
ing to enter the rivers to spawn that
in most instances a spawning stock
was left even after man's take and
destruction, although man has been
(horougli in his harvesting or de-
struction of entire runs in some

In general the order in which the
species are discussed here is the se-
quence of their time of entering the
streams to spawn, as well as the dis-
tance they ascend the streams.
Those which run first go farthest.
The economic value of the individ-
ual fish of each species rates in the
order given, but the magnitude and
the overall value of the species differ
from the order of listing.

The identifying characteristics
for the systematic classification of
the adults of the various species of
salmon are reported by Jordan
(1925), Schultz (1936), Clemens
and Wilby (1946), and many
others ; characteristics for the iden-
tification of young salmon, by For-
ester and Pritchard (1944) , Schultz
and Hanson (1935), and others.
The identifying characteristics are
not repeated here.

Oncorhy^7ichu.s tshawytscha: Chi-
nook (Columbia Kiver and south),
spring (British Columbia), king
(Alaska), quinnat, tyee, Columbia
or Sacramento salmon.

Range : Bering Sea to Monterey
Bay, but predominant in the Colum-
bia and Sacramento Rivers. Of the
1951 United States and Alaska pack
of canned chinook salmon, 61 per-
cent was from the Columbia River,
though many of the fish originating
in this river were taken at sea or
landed elsewhere.

Weight: Average at maturity,
from 12 to 40 pounds. Fish of
greater average weight are found
in Ahiska, and fish of progressively
smaller weights are taken farther
south. The average for the Cohun-
bia River is 20 to 22 pounds; and
for the Sacramento River, about 16

pounds. The maximum weight of
chinook salmon is about 120 pounds,
but 50- to 80-pound fish are quite
common in Alaskan waters.

The species is the first of the sal-
mon to enter the rivers in the early
spring, but the spawning migra-
tions are distinctly separated into
spring and fall runs, and often a
summer run is considered to exist.
The spring run consists of the most
desirable fish, entering the rivers in
the spring of the year and ascend-
ing farthest upstream. The fall
run enters the rivers in August and
September. The fall fish are some-
what heavier than the spring fish
and are more nearly mature upon
entering the rivers ; the flesh of the
fall fish is soft and pale when in the
river, and the exterior has the dis-
coloration typical of salmon ap-
proaching the time of spawning.
The fall chinook do not seek the
upper reaches of the streams to
s})awn. On the Columbia River the
bulk of the fall chinooks spawn
within 200 miles of the ocean,
mostly within a few miles of salt
water.

Spawning of spring-run chinooks
may occur as early as mid- July;
that of the fall run in August and
September, and often into October.
In some streams south of Sacra-
mento, spawning may take place in
midwinter. In one small stream
tributary to the Columbia River,
fall chinooks of large size have been
observed spawning in December,
considerably later than is usual in
the watershed.

Chinooks usually are 4 years of
age at the time of spawning, but a
few may be younger and substantial

8

nunilHM's may be 5 years old, Avliilo
some are several years older.

Yoinig of the spring diiiioolv may
remain in fresh "vvater for a 3^ear or
more after emerging from the
gravel, or thej^ may enter the ocean
^vithin weeks of hatching. The latr
ter is generally true of the fall chi-
nooks. Studies indicate that the
fish that migrate to the sea at the
earliest time attain the greatest
mature size (Van Hyning 1951),
but it is probable that the smaller
the fish upon entrance into the
ocean, the less the chance of
survival.

Chinook salmon are most abun-
dant in the Columbia River, and the
spring run used to migrate hundreds
of miles to the headwaters. The
power and irrigation dams erected
in the upper river, therefore, have
been particularly harmful to this
species. The blocking of the spring
chinooks from major spawning
areas, together with pollution and
overfishing, has resulted in a major
reduction of spring-chinook popu-
lations in the Columbia River, as
well as in other streams. The fall
chinook has, in most watersheds,
been able to reproduce satisfactorily
in the lower tributaries, but water-
use projects also threaten these fish.

Of interest to this discussion is
the appearance of "grilse," "jacks,"
or precocious chinook males in the
spawning migrations. These males
usually are 3 years old, whereas the
average age of chinooks is 4 years.
These smaller fish are fully mature
but are not utilized in hatchery
spawning operations and are at a
disadvantage in natural spawning.
It is reported that the numbers of

"jacks" or rod salmon enicring the
Fraser Rivei- aic indicative of the
size of the run (o be expected the
following year (Gilbert 1931-34,
Rounsefell and Kelez 1938). The
writer was unsuccessful in attempts
to correlate the returns of "jack"
fall chinooks to specific streams in
the Columbia River Basin wdth the
returns of average-age fish the fol-
lowing year.

OncorhynchuH nerlu: Red (Alas-
ka), sockeye (British Columbia and
Piiget Sound), blueback (Columbia
River), Fraser River salmon.
Landlocked forms are known as
kokanee, silver trout, yank, or little
redfish.

Range : Bering Sea to the Colum-
bia River, predominating in suit-
able streams north of Puget Sound ;
rarely found south of the Columbia
River, although in 1953 five "strays"
entered the traps at the Coleman
Hatchery (California) of the Fish
and Wildlife Service in Battle
Creek, tributary to the upper Sacra-
mento River.

Weight : Columbia River average,
3 pounds; British Columbia and
Alaskan waters, up to 7 pounds,
with maximum about 16 pounds.

The red salmon, like the spring
chinook, often ascends the rivers for
great distances. It spawns only in
streams having lakes in their head-
waters. (Rare exceptions to this
rule have been reported. ) The adult
red salmon remains in the head-
water lake until the reproductive
products are almost fully developed
and then ascends the smaller
streams tributary to the lake to
spawn. A few fish may spawn in
gravel in shallower lake areas

9

where springs issue from the l)Ot-
tom, or in the gravel of the outlet
from the lake. Red salmon gen-
erally are 4 years of age at spawn-
ing; a few may be younger and
many may be 5 or more years old.

The young red salmon, after
their emergence from the gravel,
descend to the lake and remain
there for a year or more before fol-
lowing the river to the ocean. As
red salmon feed upon plankton
(minute organisms in the water),
the length of residence in the lake
may be determined by the abund-
ance of plankton. Some of the
young fish may remain in the lake
throughout their lives, spaw^ning
and dying at 4 years of age like
their sea-grown brothers but usually
attaining only about one-half their
size. The progeny of these lake-
dwelling forms may descend to the
ocean and return as normal sea-run
specimens.

The run of red salmon into the
rivers begins in June and continues
tiiroughout August. Spawning
may start as early as July in north -
erii streams and be concluded in
September, Jbut in the upper Colum-
bia spawning is complete early in
October.

Oncorhynchiis hisiifch: Silver,
coho, silversides.

Range : Sacramento River to Ber-
ing Strait, but most abundant in
Puget Sound and British Columbia
waters. Present in most streams of
the Pacific coast.

Weight: In Alaskan waters the
species attains an average weight of
about 14 pounds, but in the streams
to the south the weight is consider-

ably less. Maximum weight is
about 30 pounds.

Most mature silver salmon enter
fresh water from late September to
early November, rJthough some may
appear earlier or much later. Of
the five species of salmon, the silver
is best adapted to diverse condi-
tions. It spawns in the very head-
waters of streams as well as very
close to salt water. The wide vari-
ations in time and place of spawn-
ing result in differences in time of
hatching and in development of the
fry and fingerlings.

Silver eggs usually hatch during
the early spring, and a few of the
young fish migrate soon after to the
ocean. Most of the young fish re-
main in fresh water throughout the
summer and the following winter,
usually migrating to the ocean early
in their second year. Because of
this tendency to remain in fresh wa-
ter for a year or more, the silver,
like the red salmon young, must de-
l)end upon the food supply of a lim-
ited fresh-water area, and survival
probably is jeopardized during
years of heavy spawning and sub-
sequent overutilization of available
food.

Investigations have indicated that
young silver salmon in the ocean
may remain relatively close to their
home stream, often remaining in
estuaries and inside passages while
making remarkable growth. The
majority return to their respective
streams to spawn during the fall of
their third year.

The silver salmon is taken, along
with the Chinook, in the offshore
commercial troll fishery and is

10

caught by sportsnioii throughout tlie
year in some bays and estuaries.

Oncorhynchm yorhuHcha : l*iuk
or lumii)back.

llange: AVashingtou State to
northwest Ahiska, but most abun-
(hint northward from Puget Sound.
Kare appearances in California
(Smedley 1952).

Weight: Smallest of the Pacific
salmon, averaging only about 5
pounds and rarely attaining a
weight of y pounds.

The pink salmon usually repro-
duces in the smaller streams a short
distance from the sea and often just
above tidewater. With the ap-
proach of the spawning season in
August and early September, tlie
fish develop a prominent hump on
the back (from which is derived one
of the common names of the species)
and a distortion of the jaw. Each
female deposits about 2,000 eggs.

The fry, upon emergence from tlie
giavel, migrate at once to the sea.

'J'he pink salmon invariably have
a life cycle of 2 years. In some
areas there is a large run every
second year.

OncorhyncJivs Jxcta: Chum or
clog.

Range: Columbia River north-
Avard, especially abundant in Alas-
ka. Occasionally taken in the Sac-
ramento River.

Weight: Average about 10
pounds; maximum, 30 pounds.

Chum are the latest of the sal-
mon to run, usually reaching the
streams and spawning from Octo-
ber through December. The ma-
jority of the migrating adults are
4 3 ears old, but a large proportion
of the run are 3-year fish. Spawn-
ing is in the lower tributaries of the
main rivers and in a great many of
the smaller streams. The species
does not ascend the streams for anv

FiGtiBE 5. — Chum salmon (0/

■hyncluis keta) at spawning time. Male upper,
female lower.

11

great distance, usually spawning
very close to salt water. Upon
emergence from the gravel, the fry
migrate to salt w^ater.

The chum is the least valuable of
the Pacific salmon. The flesh de-
teriorates rapidly after the fish en-
ter fresh water.

Salmo gairdneri: Steelhead trout.

Range: Southern California to
Avestern Alaska.

Weight: Average is about 12
pounds, but individuals may exceed
35 pounds.

Steelhead are included in this ac-
count because of their close associa-
tion to salmon in habits, their en-
trance into commercial and sport
fisheries, and their inclusion in
hatchery operations along with the
salmon.

The steelhead trout is a rainbow
trout that has spent part of its life
in the ocean and ascends the coastal
streams to spawn. It does not stop
feeding upon entrance into fresh
water, nor does it necessarily die
after the first spawning.

Steelhead trout apparently range
widely in the ocean along the coasts,
and it is believed that they return to
their native streams to spawn. The
steelhead may enter fresh water in
almost any month, although they do
not spawn until late winter or
s]jring

Steelhead adults may enter prac-
tically all streams. They may as-
cend to the extreme headwaters or
sj)awn very close to salt water. In
this respect they are similar to silver
salmon.

The young steelhead trout spend
1 or 2 years in fresh water and 2 or
more summers in salt water. Adults

may enter the rivers in their third,
fourth, or fifth years. Like the At-
lantic salmon, they may spawn more
than once, returning to the sea after
each spawning, but it is reported
that less than 15 percent survive to
spawn a second time.

Steelhead enter into the commer-
cial fishery for salmon to a consider-
able extent and are much sought
[ifter by sport fishermen.

The Homing Instinct

Jordan (1925), discussing the
theory that Pacific salmon return to
their home streams to spawn, reit-
erated his 1880 statement that "we
fail to find any evidence of this
[homing] in the case of the Pacific-
coast salmon, and we do not believe
it to be true. It seems more prob-
able that the young salmon hatched
in any river mostly remain in the
ocean within a radius of 20, 30 or
40 miles of its mouth." He believed
that the salmon return to their
home stream by chance rather than
through instinct. Nevertheless,
there is little doubt that salmon do
possess an instinct that leads them
from the ocean to their home stream
and hundreds of miles upstream to
the particular tributary of their
birth.

iVn interesting experiment was
undertaken at the Spring Creek
hatchery of the Fish and Wildlife
Service at Underwood, Wash., some
200 miles upstream from the mouth
of the Columbia Eiver. At this
hatchery a run of 10,000 to 18,000
fall Chinook salmon has been estab-
lished: the adults return to the
small stream where they were re-

12

Siiliiioii li;i(l iippi'iirt'd hcl'orc llu'
estaltlishiiu'iit of (lie lialcluTV in
is;)'.». In ID.M. I.", oftlu'si' i-ctnrninii-
iulults wt'iv marked, loaded into a
tank trnrk. and released IS miles
farther np the Columltia Kivei'. Of
these. H\t' I'etnrned downstream,
past tlu> mouth of the Bi<j;- AVhite
Salmon River (wliore thousands of
sahnon spawn) to tlie fish huhler at
the liatelierv. and ascended a<2;ain
to the liatehery spawning ponds.
Othei-s were caught in tlie fishery,
and a few were taken elsewhere.
AUhough the resnhs of this experi-
ment aie not conchisive, tliey indi-
cate that salmon do retnrn to their
own ])articnhu' streams whenever
possible. Eich (1939) states:

The evidence also shows clearly that the
Pacific salnioii return from their life in
the sea predoruinatel.v to their home
streams tlms justifying acceptance of
what is known as the "home stream
tlieory."'

In recent years a nunil)er of
studies of the ocean movements of
salmon, particnlarly chinook and
silver, have been reported (Van
Hyning 1951, Kanffman 1951, Neave
1951, Fry and Hughes 1951.)
These studies indicate tlie extensive
coastwise movements of some of the
salmon species. British Columbia
investigators (Pritchard 1931,
Mottley 1929) found that Avhen
young chinook salmon leave their
natal streams along the coasts of
Washington, Oregon, and Cali-
fornia, and particularly the Colum-
bia River, they disperse northwest-
erly along the United States and
British Columbia coasts. Some ap-
p arently go south w a rd . It h a s been
shown that many of the adult Pa-
cific salmon that have been captured

in the ocean, tagged and released,
then travel hundreds of miles in the
ocean bi'foi-e entei-ing streams to
spawn. Adult chinook salmon wei'e
tagged in 19-2,S by the iiurean of
Fisheries (now the Fish and Wild-
life Service) off l^aranof Island,
Alaska. Of the i'eco\-eries (lO ])er-
cent were taken in the Columl)ia
River, indicating the great distance
traveled by these salmon.

It may be concluded that young
salmon of some S})ecies do migrate
great distances in the ocean and
upon reaching maturity almost in-
variably return to their home stream
to spawn.

The impulses or guides that direct
the fish from the ocean to the natal
stream and to the proper tributary
of that stream are unknown. When
tlie young salmon enters the ocean,
the search for food causes him to
ti-avel many hundreds of miles and
often go great distances from his
home stream. At maturity, when he
is 1 year to more than 3 years old, he
starts for his particular stream, pro-
ceeding rapidly and in a direct
route. This is indicated by the in-
vestigations of Pritchard (1934),
Mottley (1929), Rich (1939), and
others. The salmon, hinidreds of
miles from his native stream, ])os-
sibly depends upon currents, vary-
ing salinities of the ocean, geo-
gi-aphical features, or the sun or
moon to direct him. Perhaps when
tlie salmon feels or tastes or smells
the flow of his particular stream in
the ocean, he recognizes it by senses
more accurate than the finest ana-
lytical equipment man has devised.
The delicate perceptiveness of the
salmon not only enables him to enter

259301

-3

13

the proper main stream but also
leads him upstream to the exact
tributary from which he emerged as
a small fish. In reaching this
stream, he probably will have
passed and ignored many other
streams flowing from watersheds
similar to the one for which he is
bound. Many such streams are
identical so far as man can deter-
mine, but the salmon senses a differ-
ence. Hoar (1951) and Black
(1951) are of the opinion that the
thyroid hormone indirectly influ-
ences the migration of fish.

It is believed that the entrance of
salmon into their natal stream to
begin the upstream journey to their
species' spawning areas is deter-
mined neither by chance nor by de-
gree of maturation of sexual prod-
ucts. Lloyd Eoyal (1951) of the
International Pacific Salmon Fish-
eries Commission presented inter-
esting facts and theories on this
f)oint. Years of investigation of the
Fraser River (Canada) stocks of
red (sockeye) salmon have revealed
that there are distinct races entering
the river and bound for spawning
areas at greatly varying distances
from the ocean. After "loitering"
in the general area at the mouth of
the Fraser for a time, each race will
quite suddenly move upstream to-
ward its tributary-stream spawning
area. It appears that the red
salmon of the various races time
their departure so as to arrive at the
spawning area, "whether 30 or 730
miles from the sea," at the period of
appro})riate water temperatures at
that point. Too-early arrival re-
sults in ineffectual spawning during
high water temperatures. Late ar-

rival results in nonproductive
spawning because of low tempera-
tures. Studies in other places have
not been so intensive as those on the
Fraser, and consequently neither
corroborative nor contradictory evi-
dence is available; but the Fraser
River study suggests the effect that
manmade obstructions have on the
races by delaying individuals on
their upstream migration.

Economic Value

The most recent complete records
compiled by the United States Fish
and Wildlife Service and by the De-
partment of Fisheries of Canada
reveal the magnitude and impor-
tance of the commercial salmon fish-
ery to the economy of the Pacific
coast.

In 1951 the commercial fishermen
operating in Alaskan waters landed
277 million pounds of salmon, for
which they received 32.4 million dol-
lars. Pink, red, and chum salmon
made up the bulk of the catch. The
salmon prepared for market, includ-
ing 165 million jDounds of canned
salmon valued at 79 million dollars,
totaled 189 million pounds valued at
86 million dollars.

The catch in the waters of Ore-
gon, Washington, and California
for 1951 totaled 98 million pounds
valued at 20 million dollars to the
commercial fishermen and at about
three times that value for the final
market product. Chinook, silver,
and chum made up the bulk of the
catch. The landings in Washing-
ton, Oregon, and California were 77,
14, and 7 millions of pounds, respec-
tively.

14

Figure 6. — Brailing salmon from a floating trap in Alasl-;a.

Statistics for the 1951 British Co-
himbia commercial catch are 202
million pounds landed with a value
of 28 million dollars. Sixty percent
of the poundage resulted from chum
and pink catches. Of the total
poundage, 44 percent was taken by
seine and 41 percent by gill net.

The 1951 total of the commercial
salmon fishery of the Pacific coast of
North America is 577 million
pounds with a value of 80 million
dollars to the fisheriyien.

In addition to the commercial
fishery, there is a tremendous sport
fishery from southeastern Alaska to
the Sacramento Kiver. It is almost
impossible to estimate the catch and
value of this fishery with any degree
of accuracy. An attempt has been
made to determine the sport catch
fi-om the Columbia River. The fig-

ure arrived at is a catch of 2 million
pounds annually. If all of a sport
fisherman's expenditures in connec-
\ ion with his seeking of salmon were
charged against his catch, the flesh
would certainly be valued at several
dollars a pound. In one of the best
studies of its type, Wallace (1952)
determined that in Washington
State the 400,000 fish-and-game li-
cense holders spent 36.7 million dol-
lars in 1950 for sport fishing. The
license holders who participated in
the survey reported an average ex-
penditure of $125 per capita for
fishing.

The exceptional nutritive value of
salmon flesh is well known. This
flesh is quite as valuable as beef and
other meat products and even ex-
ceeds beef in desirable components
such as iodine, phosphorus, and
fluorine.

15

HATCHERY PROPAGATION OF PACIFIC SALMON
Early History and Objectives

The modern program of aitificial
or hatchery propagation of Pacific
sahnon had its beginning at the
Baird Station (California), on a
tributary of the upper Sacramento
River (Stone 1878). This egg-
taking station was established by
the United States Commission of
Fish and Fisheries (now the Fish
and Wildlife Service) for the pur-
pose of collecting eggs of the
chinook or cjuinnat salmon for ship-
ment to foreign waters. Additional
egg-taking stations were later estab-
lished in Oregon and Washington.

As the salmon fishery expanded
and became more efficient, exploita-
tion reduced the runs of salmon to
dangerous lows. In many places
fishing became unprofitable, al-
though the production potential of
the spawning areas of the streams
remained constant. Attempts were
made to maintain the runs of
salmon or to restore them to former
levels of abundance through hatch-
ery propagation. Salmon hatch-
eries were established for the
purpose of taking and incubating
eggs and releasing the resulting
small fish or fry.

Salmon hatcheries are now recog-
nized as tools to assist in the man-
agement of fisheries. The product
of the hatchery, regardless of size
of the fingerlings, is not considered
an end in itself. There is a definite
need for the hatchery output of
young salmon to counteract the in-
crease of human population, the
resulting industrial devel()])ment,
and the intensified fishing.

In the early years of commercial
fishing for salmon, the fish had free
access to their original spawning
areas in the hundreds of coastal
rivers. The advance of civilization
gradually limited the areas avail-
able for spawning. The modern
salmon hatchery is in existence to
counteract so far as possible the
harmful influences of man. Major
dependence for maintenance of the
salmon runs must remain with nat-
ural reproduction. The primary
purpose of the hatcheries is to pro-
vide stock with which to restore or
develop runs, althouoh the hatch-
ery-reared sahnon contribute to a
considerable degree directly to the
catch. The maintenance of a fish-
ery, however, must depend largely
ui)on regulations and the removal
of blocks to migration, including
pollution, in order that salmon may
ascend to spawning areas. Stream
and lake areas must be available for
tlie early rearing of the salmon.

In the 1930's it was realized that,
in order to assure adequate return
for the fishery and for spawning,
some species of salmon must be
reared to large size before their re-
lease. This continues to be the gen-
eral hatchery practice.

Circumstances Requiring
Hatchery Propagation

Probably the best example of the
harmful effects of man's progress
upon anadromous fish is in the Co-
lumbia River Basin. It has been
established that at least half of the
original si)awniiig and ivaring area
of tlie Columbia Rivei- watershed
lias b(HMi ma(U' unsuilable or inac-

IG

oessiblo to salmon and steellioad
trout. Dams have been responsible
for this riMhiction of area availabh'
for 11 a t u r a 1 reproibict ion. The
liiiiuh-eds of sinall (hiiiis erected
many years a^'o in the C'obimbia
Basin in connection witii h)«:i:ino-
operations and for irrigation pnr-
[)oses (U'prived sahnon of headwater
spawnino- areas, as did Grand Con-
lee and other lar^e dams. Many
additional large dams are planned
for constrnction in the Colnmbia
River and these will fnrther limit
salmon migrations or make the re-
maining spawning areas more diffi-
cnlt to reach. Even if the adnlt
hsh surmonnt the dams and reach
the spawning gronnd, the seaward-
migrating fingerlings will snffer
heavy mortality in j)assage over the
dams and through the power tur-
bines.

Another factor that has inijiaii-ed
the salmon resources is ])olbition
from in(histrial sources. In tlie
Willamette Rivei", tiibiitary to the
Columbia, a polhition block de-
stroyed the substantial chinook runs
not only by blocking the upstream
juigrations but also by causing the
death of any fingerlings migrating
downstream during the summer and
fall. Fortunately, very few streams
of the Pacific coast are so polluted as
to desti'oy runs entirely. Pollution
^\■as one of the causes for the almost
complete loss of Atlantic salmon in
the New England States. Modern
regulations and laws have tended to
reduce this threat to salmon during
tiie relatively recent industrial de-
velopment of the West. In areas
liJve the Willamette River, where
pollution has actually blocked sal-
mon migration, substantial im-
pi ovement has been effected through

I'^IOrKE

-Fishways in fci

uul and
("oluinhi

.n f:
Kiv(

V shove at Bnnneville r>ii

17

the efforts of State and Federal
agencies. It is anticipated that the
Willamette block will be eliminated
by 1954. However, sawmill and
mining wastes, as well as warm wa-
ters from atomic-energy plants on
many streams have tended to de-
crease the suitability of the waters
for salmon, as well as for humans.

Early logging operations were
detrimental to salmon. All cover
was removed from the hills, logging
debris was dumped into streams,
and logging dams were constructed
wherever convenient. The net re-
sult was extremely harmful to anad-
romous fish. Kapid runoff waters
fi"om the bare hills silted streams
and spawning beds, high- and low-
water periods were accentuated, wa-
ters often became too warm for sal-
mon survival, and dams blocked up-
stream migration. Only in recent
years have our forests been managed
on the basis of sustained yield.
Many of the heavily logged water-
sheds have now developed good sec-
ond-growth cover, and present-day
logging practices and laws are more
conducive to preservation of salmon
habitat.

A large-scale salmon-salvage i)ro-
gram was undertaken on the upper
Columbia River in 1939, when
construction of Grand Coulee Dam
blocked salmon from hundreds of
miles of spawning area. This pro-
gram transferred the runs of salm-
on that formerly ascended to
the areas above Grand Coidee to
tributaries below the dam. As part
of the salvage program, several
salmon hatcheries were constructed.
The chinook and silver salmon as-
cending the river to Grand Coulee

Dam have been maintained in pre-
dam numbers, and the populations
of red salmon have shown tremen-
dous increases (Fish 1948). The
hatchery production and release of
fingerling red salmon can definitely
be correlated with the return of the
adult fish to the fishery 3 years later.
On the Sacramento River in Cali-
fornia, the construction of Shasta
Dam stopped salmon from ascend-
ing to the major spawning areas of
the watershed. A hatchery (fig. 1)
was constructed on Battle Creek, a
tributary below the dam, and adult
salmon ascending to Keswick Dam
immediately below Shasta were
trapped and transferred to Battle
Creek. This operation has resulted
in maintenance of the salmon in the
Sacramento River — ^liatchery oper-
ations have provided necessary
stock, and cooler waters issuing
from the reservoir have made the
river below the dam more suitable
for salmon spawning. The hatch-
ery operation has resulted in defi-
iiite and substantial contributions
to the fishery in the San Francisco
Bay area, as well as in the coastal
waters as far north as Vancouver
Island. The improvement of con-
ditions for spawning of salmon be-
low Shasta Dam has not been dupli-
cated in connection with the con-
struction of any other major dam.

To mitigate the losses of salmon
expected to result from the con-
struction of additional dams in the
Columbia Basin, the Fish and
Wildlife Service, together with the
fishery agencies of Idaho, Oregon,
and Washington, evolved in 1947 a
plan known as the lower Columbia
liivei- tislicries develoi^ment pro-

18

.5 o

5

19

gram. This had as its objective the
maximum development of popiihi-
tions of sahnon in the streams trib-
utary to the lower Columbia River.
Under this program, log dams and
debris were removed from streams,
and natural waterfalls were blasted
or laddered to permit free access for
salmon to all possible spawning-
areas. A large number of hatch-
eries are producing fingerling sal-
mon that are stocked into barren
streams or used to develop as rap-
idly as possible the maximum popu-
lations of salmon in all suitable
streams. Most of the dams to be
constructed will include fish-pas-
sage facilities to permit the mainte-
nance of at least a part of the upper-
river runs of salmon. As a result

of these activities, it is anticipated
that a substantial portion of the
average (1940-49) sport-connner-
cial-Indian catch of 31 million
pounds of salmon a year attributa-
ble to production from the Colum-
bia River Basin can be maintained.
Irrigation developments have
most consistently been destroyers of
}()ung salmon. On their migration
lowarcl the ocean, large numbers of
(he young fish using any major
downstream How enter irrigation
ditches and merely provide fertilizer
for the farm fields. During the
])ast 20 years, primarily in the Co-
lumbia River watershed but also in
other streams of the Pacific coast,
the Fish and Wildlife Service and
the fisherv aaencies of Washington,

20

FiGi'KE 10. — Revolving fish screens in irrigation canal. Waterflow from right to left.

Oreo-on, and California have in-
stalled or have re({iiired the instal-
lation of hundreds of screens to pre-
vent entrance of the j^oung fish into
the irrigation ditches. On some of
the very large Federal irrigation de-
velopments such as the Delta-Men-
dota diversion, at Tracy, Calif.,
complicated fish-screen installations
are required. Such screens save
many thousands of young fish an-
nually. In many streams, however,
the diversion of water for irrigation
has resulted in the complete loss of
spawning areas, as the water re-
maining has been insufficient to sup-
port fish life. Often artificial prop-
agation must be relied upon to re-
])lace these losses.

Probably the greatest threat to
the resource is in the contemplated
construction of power dams in the
streams of Alaska and British Co-
lumbia, where tremendous hydro-

power potentials exist. Piecemeal
reduction of available spawning
areas has been most harmful. The
construction of a dam on a relatively
small stream usually is not consid-
ered particularly harmful to the
overall salmon production, but in
the aggregate these projects destroy
major portions of the resource. It
is anticipated that large segments of
the remaining salmon populations
will be destroyed by the construc-
tion of dams, and that the salmon
hatchery will be necessary for par-
tial preservation of some of the af-
fected populations.

A comprehensive research pro-
gram is expected to determine
means by which adult salmon may
be passed over dams without delay
and means by which the young sal-
mon may proceed downstream with-
out suffering substantial mortalities
in passage through turbines or over

21

spillways. A popular belief is that
if fish ladders or fishways are pro-
vided in a structure for the up-
stream passage of adult salmon, the
problem is solved. This is far from
the truth. A salmon, as jDreviously
stated, has stored energy which is
sufficient only to maintain himself
and develop reproductive products
during the diffcult journey from the
ocean to the spawning area. Fur-
thermore, he follows a time sched-
ule and must arrive at the
spawning area during the brief
period when water temperatures
are suitable for successful spawn-
ing. His time schedule does not
take into consideration one or more
dams at each of which he may be
delayed several days before finding
and ascending the fish-passage fa-
cility.

Losses of young salmon when
passing a dam while on their way
to the ocean are substantial. Stud-
ies at 62-foot Bonneville Dam on
the Columbia River indicate sub-
stantial loss among young salmon,
and current investigations of the
Washington Department of Fish-
eries and of the International Pa-
cific Salmon Fisheries Commission
at 3 dams over 100 feet high reveal
losses of 35 percent or more among
downstream migrant salmon.

It is apparent from the foregoing
that fish ladders alone will not pre-
serve salmon runs. The adults must
quickly find and easily ascend the
ladders, and the young salmon must
be led to safe passageways by some
as-yet-undeveloped device.

There are still other undesirable
features that develop upon the con-
struction of many dams. Species of

fish that prey upon young salmon
and are not themselves desirable for
sport or food find reservoirs behind
some dams conducive to their exist-
ence and increase. In a single
reservoir, and particularly in a
series of impoundments such as is
planned for the Columbia River,
predatory species will cause major
losses among young salmon. Also,
reservoirs often inundate spawning
areas, making them unattractive to
spawning salmon because of water
depth and absence of quite rapid
flow over and through the gravel.

Hatchery propagation is not ex-
pected to be able to compensate for
or to mitigate satisfactorily the
losses caused by dams and pollu-
tion, but hatchery-reared salmon
can help maintain populations at
the highest possible level of abun-
dance by supplementing natural
reproduction.

Methods

The salmon hatcheries operated
by the fishery agencies of Califor-
nia, Oregon, and Washington, and
by the United States Fish and Wild-
life Service in the three States,
propagate, collectively, chinook,
red, silver, chum, and pink salmon,
and steelhead trout. There are dif-
ferences in techniques and proce-
dures employed by different hatch-
eries, usually because of the species
propagated and the physical char-
acteristics of the watershed and the
fish-cultural station, but generally
the methods of salmon propagation
at hatcheries on the Pacific coast
are quite similar.

The species of salmon differ con-
siderably in characteristics, and

22

tlioso (lifftM-ences complioatc liatcli-
t'ly [)r()paiiation. The adults of one
s[)ecies may be diiru-ult lo \\-a\) and
liold to spa^Ynino•, and t he youno- of
another species may be dillieult to
rear. Differences in these and
other respects are found within
species in tlie same Avaterslied, al-
though the hatchei'y water supply
or physical characteristics may de-
termine the extent of deviation
within species.

.Vhhou^h more detailed descrip-
tions of procedures and equipment
will be given, it is well to outline
briefly the essential stei)s taken to
propag-ate salmon at a hatchery.
The methods are quite similar to
those employed at trout hatcheries.

The first requirement is adult
salmon from which eggs and sperm
may be secured. These brood fish
are trapped by placing a rack or
racks across a stream up which the
mature fish are migrating. ''Ripe''
females are selected and killed, and
the eggs are removed from the body
cavity. The milt or sjDerm of males
is expressed onto and mixed with
the eggs. After a few^ minutes the
surplus sperm is washed from the
eggs and the latter are taken into
tlie hatchery, wdiere they are placed
in baskets or on stacked trays so
arranged in the hatchery troughs as
to permit circulation of water
among the eggs. At the end of the
incubation period, which varies ac-
cording to the water temperature
(50 days at 50° F.), the small fish
or fry emerges from the egg shell
with an attached umbilical or yolk
sac that supplies sustenance until
the fish is capable of taking food

hy mouth, a])pro.\iinately '■] week-
later.

When the young fish are free-
swimniing, they are fed in the
hatcheiy troughs for a time and
then transferred to outside rearing
ponds. There the fish continue to
receive food and grow to the ap-
])i-opriate size for release. In the
early years of salmon culture, prac-
tically all young fish were released
as unfed fry. In ij^cent years, as
many have been held for pond rear-
ing as facilities wall permit. Ex-
ceptions to this are the chum and
pink fry, which, under natural con-
ditions, immediately seek salt water
and so are released from the hatch-
ery. Many of the fall chinook
salmon also are released as fry or
after a few^ weeks of rearing, prin-
cipally because of a lack of rearing
space. On the other hand, wdiere it
is possible to do so, red, silver,
spring chinook, and steelhead trout
are held in rearing ponds for a year
or more.

Some comments regarding the in-
dividual species of salmon and of
steelhead trout are of interest and
pertinent to a discussion of salmon
culture but need not be included in
the detailed descriptions that fol-
low.

As has been mentioned, the chi-
nook-salmon runs into the rivers are
quite arbitrarily divided into two
groups. The spring chinook adults
enter the rivers in the spring and
early summer and usually ascend
the rivers farthest, spawning from
midsummer to about October. The
fall chinooks enter the rivers from
midsummer to September, spawn-
ing from August to October in

23

relatively lower reaches of the rivers
and often close to tidewater. The
spring chinook is in prime condition
when it enters the rivers, having
stored fats for sustenance on a long
journey inland; and usually the
female contains immature or green
eggs that will not be developed for
spawning for many weeks or per-
haps several months. The fall chi-
nook, on the other hand, usually
is not in prime^condition when it
enters the river; the flesh deterio-
rates rapidly, as the spawning time
is often a matter of only days or
possibly 2 or 3 weeks.

It is evident that to trap the
spring chinooks for spawning at the
mouth of a long river would require
that these fish be held for many
weeks or several months before the
sexual products would be "ripe"
and suitable for artificial spawning.
Successful holding of adult spring
chinook, even in carefully con-
structed "natural" ponds or streams
sections, has proved very difficult.
Experience at several hatcheries has
revealed the difficulties to be en-
countered in such holding. At a
few points spring chinook adults
have been retained successfully
until mature; the success appeared
to be due to the Avater supply.
There may be also a psychological
as well as a physiological effect upon
the fish because they are in foreign
waters and not en route to their
home streams. Theoretically, it
would be possible to trap these
spring-run fish immediately below
their spawning areas, but the cost
in relation to the relatively few fish
that w^ould be taken in the small
tributaries would be excessive.

Even here, too, many quite green
fish would be taken.

The fall chinook adults are quite
readily trapped and spawned in the
tributaries of the lower reaches of
the various watersheds. There the
fish are almost ripe and usually need
be held for only a few days before
spawning. The fall fish do not
"resent" being held in ponds or be-
tween racks as do the springs. The
fall fish are much more docile and
are quite readily "herded" or led
where most convenient for accom-
plishment of spawning.

After the eggs of spring or fall
fish are taken and placed in the
hatchery, there is no appreciable
difference in the incubation. Maxi-
mum returns to the fishery are se-
cured by rearing the young spring
chinooks for about 1 year before re-
leasing them into the streams. Such
rearing should be wnth as nearly
natural water temperatures as pos-
sible. Experience has shown that
when the spring chinook fingerlings
attain a size of 3 to 5 inches they ex-
hibit a decided desire to migrate
downstream. In hatcheries where
warm spring water is available, the
tendency is to provide warmer than
natural waters for rearing, resulting
in unnaturally great food intake
and growth of the fish. Thus, the
migration urge may come upon the
young fish, requiring their release
during periods of severe cold when
there is anchor ice in the sti-eams or
other factors unfavorable for stream
survival. The fall chinook finger-
lings do not generally exhibit such
migration urges.

The adults of silver and red sal-
mon trapped long before their

24

spawnin<»: i)eri()(l also are difficult to
hold in ])()nds or .streams without
mortality, but experience with these
species has not b?en so extensive as
with the chiuooks. (uMierally arti-
ficial spawniiit:' of" silver salmon
takes place neai- the spawnina- areas.
and the red salmon are tiap[)ed and
spawned as they ascend streams
from lakes to spawn. Steelhead
trout adults, too, wlien trapped
along with salmon in the fall, are
difficult to hold over the winter for
spring spawning.

Studies by the Washington De-
partment of Fisheries have shown
that silver-sahnon young should be
reared in the hatchery ponds mitil
their second spring and then re-
leased. The same appears true of
steelheacl-trout young. Yearling
red-sahnon fingerlings have pro-
duced excellent returns when re-
leased into their "home" lake in the
fall of the year.

Chum salmon usually are almost
mature sexually when they enter
fresh water, and the adults can be
readily trapped and spawned.
Under natural conditions the young
fish proceed to salt water immedi-
ately after emergence from the
gravel, and the usual hatchery prac-
tice is to release the young fish at
that time.

The fall chinook salmon is the
species most extensively propagated
at most of the salmon hatcheries.
It is probably the easiest to secure
and handle in the hatchery. Fur-
thermore, the fall chinook is native
to the lower reaches of the water-
sheds, and therefore usually below
major dams, and can be maintained
and developed most readily. There

is economic justification, too, for
concentration upon the propagation
of fall chinook in those areas wiiere
mamnade conditions req.uire the
assistance of the hatchery for
the maintenance of the salmon re-
source. All of the species require
about the same capital and annual
investment for hatchery production
to secure like returns in numbers of
adult fish, but the fall chinook, at-
tains the greatest adult size and
thus contributes more pounds of fish
per dollar invested.

The following descriptions of
methods, procedures, and equipment
are of those generally employed in
the propagation of fall chinook
salmon. The few deviations from
this pattern to propagate other spe-
cies will be noted if pertinent.

Spawn taking

Wlien a group of adults is avail-
able in the racked stream area or in
the holding pond, and the majority
are mature, the fish are brought to-
gether in a small area for sorting
and spawning. The usual spawn-
ing run consists of both sexes in
about a 50-50 ratio.

Experienced men sort the salmon,
placing males and ripe females in
separate enclosures. Jacks or pre-
cocious males, as well as surplus
males, are killed. Green females
are returned to the pond for further
maturation.

The indication of ripeness in the
female is the separation of the eggs
in the ovaries, but it is not usually
possible to determine this by an ex-
terior examination. Experienced
spawntakers rely mainly upon the
general appearance of the female

25

and upon the looseness and position
of the eggs as determined by gently
passing the hand over the abdomen.
Picking the fish up and gently
pressing on the abdomen will result
in the extrusion of eggs from the
vent, but this method is not always
indicative of the maturity of all
eggs, as those near the vent may be
ready for spawning while others
further forward may not have sepa-
rated. Under natural conditions,
spawning by each female is under
way for several days, the eggs being
released as they become mature. It
is impractical in artificial spawning
operations to extend the spawning
of each fish over several days, so
quite often the spawning of appar-
ently ripe females is delayed for an
additional day in order to be certain
that all eggs are separated, which
results in improved fertilization of

eggs. However, where large num-
bers of adults are handled, as at
some of the lower Columbia River
stations, space does not permit this
practice.

Usually all males attending the
females are ripe and will produce
adequate sperm. Only 1 male is
considered necessary for the fertili-
zation of the eggs of 3 or 4 females
when artificially spawned. As the
sexes appear in approximately equal
numbers the surplus males usually
are removed, j^articularly when
large numbers of fish must be han-
dled and space is at a premium. In
the event that there is a shortage of
males, those present are utilized sev-
eral times, as they create new sperm
to replace that taken.

When the sorting of fish is com-
pleted and all ripe females and a

FitiUKK 11

(t'iiiii- i»rt'i):ire(l

qiawniii

2S

sufficient luiiiiluM- of luiiles are eoi-
ralled. workers kill (he lish b v sharp
blows on the top of the head. The
salmon are hiid in a row, three fe-
males and a male, and so on. The
tail of each tish is cut to drain blood
from the body so that the blood will
not later become mixed with (he
eggs and interfere with fertiliza-
tion.

The spawntake • holds a female
salmon vertically by inserting one
hand in the gills, with the back of
the fish hanging between his knees.
The vent is held just above a 10-
to 14-quart pail or bucket. A knife,
usually one specially made for this
operation is inserted in the vent and

drawn upward toward the head, the
body wall being cut to one side of
the ventral fins and to (he top of the
body cavity (fig. \-^}. If the eggs
are fully se})arated. (liey will poui'
into the bucket.

After the eggs have been stripped
from about three females, which is
done rapidly, the sperm of a male
is extruded into the i)ail and thor-
oughly hand-mixed with the eggs.
( )ften a small amount of sperm will
be placed in the bottom of the bucket
before eggs are placed therein, and
sperm may be introduced after eggs
have been taken from each female.

The foregoing process may be re-
peated until the bucket is about two-

FiGiRE 12. — Spawning cliinook salmon at the Coleman (Calif.), Hatchery of the
Fish and Wildlife Service : Left — eggs being removed from female ; right — extrud-
ing spermatozoa from male onto eggs.

27

thirds full of eggs. Within 2 min-
utes, if the eggs and sperm have
been thoroughly mixed, all eggs
should be fertilized, each egg hav-
ing been entered by one sperm only.
The bucket of eggs is then gently
immersed in clear water to permit
the flow of water into and around
the eggs so that they harden. Ex-
cess sperm which may adhere to the
eggs and encourage the growth of
fungus is washed from the eggs.
Recently a device for better washing
of eggs has been developed by the
Fish and Wildlife Service, 'xhis
device consists of a perforated inner
bucket or pail which may be re-
moved and set in running water to
wash all the surplus sperm from the
eggs. (See fig. 12.)

Each pail of washed eggs is car-
ried into the hatchery building,
where the eggs are deposited in
trays or baskets for the incubation
period.

Incubation of eggs

When the freshly fertilized and
washed eggs are taken into the
hatchery, they are placed on trays
or in baskets in troughs through
wdiich water is circulated at all
times.

The temperature of the water de-
termines the rate of development of
the salmonoid eggs and therefore
the number of days required for
hatching. The highest sustained
temperatures should be less than 60°
F. Slightly warmer waters may re-
sult in excessive mortalities. On the
other hand, prolonged ])eriods of
water temperatures below approxi-
mately 34:° F. result in a very long-
incubation period and often in the
production of abnormal frv in

which the yolk material is not ab-
sorbed. In waters having an aver-
age temperature of 50° F., it can be
expected that the salmon (or trout)
eggs will hatch in about 50 days, and
that the fry will have absorbed the
yolk sac in about 3 weeks and will
then start taking food by mouth.
The rate of development of the eggs
and fry is important to the routine
fish-cultural operations.

Throughout the various stages of
development the salmon egg is sub-
ject to mechanical injury. There
are some stages in which the egg is
much more susceptible than in
others, a condition that influences
hatchery procedures. Two princi-
pal phases of development of the
salmonoid egg are commonly recog-
nized by the fish culturist. The
"green" state is that period of devel-
opment from the fertilization and
water hardening of the egg to the
closing of the blastopore. During
the first 24 to 48 hours of this phase,
the eggs are quite resistant to me-
chanical injury, and it is during this
period that the eggs must be placed
on the trays or in baskets to remain
undisturbed until the green stage is
complete. Usually the eggs are so
placed witliin a few hours of spawn-
ing. The remainder of the period
of egg development to hatching is
called the ''eyed'' stage and is char-
acterized by the appearance of the
eyes of tlie embryo through the
shell of the egg. After all eggs are
eyed, they can be handled gently.
It is common practice in salmon
and trout hatcheries to shock the
eggs at this time to rui)ture the
vitelline membrane of undeveloped
eggs. This is done by dropping the

28

Figure 13. — Saliuon eggs in basket raised

eggs several inches into water, with
the result that infertile or undevel-
oped eggs turn white. These eggs,
which would eventually become
dead tissue and encourage fungus
growth, are removed.

Shocked eggs are replaced on the
stacked trays in the water or, if they
were in baskets, they may now (or
later) be placed on trays, to re-
main there until hatching is com-
plete and the fry have absorbed
most of the yolk sac. The young
fish are then placed directly in the
troughs, where they are fed finely
ground meats. In some salmon
hatcheries the eggs are permitted
to hatch in the baskets, and the fry
drop through the wire mesh to the
trough bottom.

Dead eggs provide the tissue for
the growth of fungus that, if not
controlled or removed, will spread
over and smother surrounding eggs.
In view of the delicacy of the eggs
during the green stage, it is not de-
sirable to attempt to remove the
dead eggs because adjacent eggs
might be injured. It is possible to
prevent or inhibit growth of fungus
by routine chemical treatments of
the eggs, and during the green stage
such treatments are a weekly rou-
tine at many hatcheries. Burrows
(1949) describes this treatment.

Silt carried in the water supply
can also coat and smother eggs.
For this reason and others a clean
water supply for the hatchery is
desirable. Silt can be removed by
passing the water through a filter

29

^ (lend (wliite) eggs from tray,
cell and the darker colored eggs a

A

•e al

.. .. \\y hatclied fry
lut to liateh.

before the water enters the liatch-
ery. Silt also may be drawn from
the stacks of eggs by raising dam
boards between stacks and drawing
the water down the entire trongh.
This should not be done during the
critical development stages when
the eggs are green.

Rearing

Most hatcheries have a number of
ponds in which the young salmon
are reared to larger sizes. It is a
generally accepted principle that
the larger the fish are when released,
the better the chance of survival to
maturity. Chum and pink salmon
fry normally migrate to salt water
inmiediately upon emerging from
the gravel of the stream bed, so
these species are released from the
hatchery as fry. Fall chinook
salmon fry usually ;tre reared in

hatchery ponds for a few weeks to
several months, the time depending
upon the capacity of rearing facili-
ties. For the best results with
spring chinook, I'ed, and silver
salmon, and steelhead trout, the
young hsh nuist be reared for a
year or more. Silver and spring
chinook fingerlings released in the
second spring about II/2 years after
spawning have produced good re-
turns, whereas the release of silvers
at small sizes has given unsatisfac-
tory returns.

It is apparent that the rearing of
some species of salmon for varying
periods is necessary for adequate re-
turns. It is equally true that such
rearing can be justified on a basis of
economics, the hatchery costs being
calculated against the returns of
adult Hsh to the commercial and

30

sport fishermen. This subject is dis-
cussed in a following section.

Upon absorption of the yolk sac
(see fig. 4), fry usually are fed in
the troughs for a week or more be-
fore being placed in the rearing
ponds. To secure maximum pro-
duction, the ponds are stocked to a
safe carrying capacity and some of
the growing fish are removed at in-
tervals and stocked into native wa-
ters. By this practice the quanti-
ties of salmon being reared may be
near the safe carrying capacity at
all times. If adequate rearing space
is available, or if all of the fish must
be reared to a specific size or for a
definite period before releasing, the
initial stocking of the ponds is lim-
ited to the numbers of salmon young
that can safely be held in the ponds
until released.

It is the practice at all Fish and
Wildlife Service salmon hatcheries
to use pounds as the measure for
calculating the carrying capacities
of ponds, troughs, or fish-distribu-
tion tanks, and the quantities of
food to be presented to the fish. At-
tempting to determine carrying ca-
pacities and quanities of food to be
fed when only numbers of fish are
known is inconvenient and inaccu-
rate.

The objective of Fish and Wild-
life Service hatcheries is to produce
annually 1 pound of salmon for each
cubic foot of water available in
troughs and ponds for rearing. Oc-
casionally, at some hatcheries, this
goal is exceeded, but the average is
still below the objective. When at-
tempting to achieve an annual pro-
duction of 1 pound per cubic foot
of water, it is necessary to carry

near the maximum poundage in all
rearing facilities throughout the
year.

One important factor that may
limit the efficient utilization of rear-
ing equipment is the incidence of
disease. Some hatcheries may ex-
perience persistent disease among
the stocks of fish held, usually at-
tributable to the water supply.
Although the usual diseases at the
modern hatchery need not cause ex-
cessive mortalities if prophylactic
measures are routine procedure, the
methods of applying treatments
often require that the rearing facili-
ties be stocked with fish at levels
considerably below capacities.
Such routine control measures also
require labor. Diseases and their
control wdll be discussed in a follow-
ing section.

Obviously the foods presented to
young salmon in ponds are of ex-
treme importance, not only for ade-
quate growth but also to maintain
the fish in a healthy condition.
This subject is discussed separately.

Cleanliness in the troughs and
ponds in which salmon or trout are
being reared is desirable. Al-
though some hatcheries appear to
be able to produce fish successfully
and efficiently with but a minimum
of labor expended on the cleaning
of ponds, experience has shown the
advantages of cleanliness. Only
the concrete ponds in general use
for the rearing of salmon can be
cleaned satisfactorily. At some
liutcheries the growth of algae on
pond walls and floors is removed by
weekly brushing. Another proce-
dure is to apply special paint or a
solution of copper sulfate and salt

31

Figure 15. — Cleaning a raceway p

to tlie dry pond, and then tiush the
pond tlioroughly before the intro-
duction of fish. This inliil)it8 or
prevents growth of algae during the
rearing season.

Algae are not harmful to salmon
in ponds and usually (as in the
wild) harbor small animal organ-
isms upon which the fish can feed,
but the large numbers of fish held
in a rearing pond minimize the con-
tribution of this food source. Also,
the algae growth is a lodging place
for fish excrement, unconsumed
food, and disease organisms.

With water adequate in tempera-
ture, quality, and quantity, with dis-
ease control, proper diets, and exper-
ienced care, the maximum produc-
tion of healthy salmon may be
expected.

Diseases

Throughout their lives all fishes
are subject to attack by a great

tingerlii

variety of disease organisms. In
tlie wild, diseases rarely become epi-
demic, because of tlie Avide disper-
sion of the fishes, which greatl}'
decreases the probability of direct
or indirect transmission of disease
organisms.

In the hatchery pond or trough,
the fish usually are exposed to the
same diseases as are hsh in the wild,
})articularly if— common in the
l)ropagation of salmon — the hatch-
ery water supply is taken from a
stream. Disease organisms among
hatcher}^ fish have an ideal oppor-
tunity to spread because crowded
conditions permit easy transmission
l)etween lish. To combat disease
epidemics or even relatively minor
losses of hsh, the fish-culturist elim-
inates the possibility of disease
breeding or being retained in the
algae and excrement in ])onds and
troughs. He has as a tool an ade-
(jiiate flow of water tliat washes

32

oi'gaiiisins from the pond as I'apidly
us ])()ssib]e. Furllu'r. and most im-
portant, lu' nst's propri' diets to Ivccp
his lish in healthy condition ami
thus less snsct'pt ihle to disease. lie
has. too. the i-ecommendations of
researchers who ha\ c identified dis-
ease ortjanisnis and methods of
control. If unable to cope with a
<lisease problem, tlie liatcherynnin
usually can call upon a ])atliol()f»;ist
for assistance.

In some salmon and trout hatch-
eries the control of organisms that
nuiy destroy eggs or fish is a con-
tinuous routine, and at all stations
constant vigilance is required to rec-
ognize and control diseases at first
;ippearance.

The most serious losses among
eggs are occasioned by fungus
growths. Other diseases of eggs,
such as white spot and soft eo:g, are
less well known, although white spot
is believed to be caused by mechani-
cal injury to the egg.

One group of plant organisms
and four groups of animals com-
])rise the principal organisms caus-
ing the diseases of salmon and trout.
The bacteria, considered in the plant
group, ai'e the most important in
connection with fish diseases, al-
though the protozoans, trematodes,
cestodes, and crustaceans — all ani-
mals— are represented in the fish-
disease groups. Recently a virus
infection has caused excessive mor-
talities among red-salmon finger-
lings, and a similar infection may
have been responsible for losses
among other species in hatcheries
throughout the country.

Parasitic infections may ap})ear
on the external surface of the fish.

or they may be internal. Those on
the exterior surfaces usually can be
controlled or eliminated by chemi-
cal (reatment. Intei'nal j)arasites
ai-e much mort' diflicidt to recognize
and coiUi-oI. The incorpoi'ation of
api)ropriate chemical disinfectants
in the food is elfective with some in-
teiiial parasites. Davis (1947)
gives detailed descriptions of vari-
ous disease organisms and their
control, and includes an excellent
list of references. Since the publi-
cation of Davis' work, many other
researchers have provided addi-
tiomd information and references
including Rucker and associates
(1952) on gill disease, and on kid-
ney disease (1951).

Too much stress cannot be i)laced
upon the necessity for adequate
diets for the production of disease-
resistant fish. Nutritional deficien-
cies open the door to infection.
When fish become weak from infec-
tion, they usually stop feeding and
thus become weaker and less able
to resist disease.

Foods for Hatchery Salmon

Until comparatively recent years
the requirements of salmon and
trout for the diet components neces-
sary to life — carbohydrates, pro-
teins, fats, minerals, and vitamins —
were practically unknown. Re-
search has not provided much spe-
cific information regarding the
needs of fish, but the work of re-
searchers on other animals, a lim-
ited amount of fisheries work, and
trial-and-error methods, have im-
proved tremendously the informa-
tion available to the fish culturist.
A review of recent publications will

33

show the information available on
fish nutrition. Although such a re-
view is not included here, it is sug-
gested that interested persons con-
sult the issues of The Progressive
Fish Culturist (Fish and Wildlife
Service publication available in
many libraries) for the past few
years. This periodical contains
much information of value to the
salmon and trout culturist.

Experience and research indicate
that the various species of salmon,
and of trout, do not thrive upon the
same diet. In fact, a diet of appar-
ently the same components fed to a
species of fish at one hatchery may
not be adequate for the same species
at another hatchery. Basic diets
have been developed, but these
usually must be altered to meet the
reauirements at different hatcheries.

In past years it was possible for
each hatchery to secure its fish food
locally. Beef, hog, and sheep
livers, spleen, lungs, tripe, and
other products not commonly used
by humans were readily and cheap-
ly available. In recent years the in-
creased consumption of liver by hu-
mans and the marketing of dog and
cat foods utilizing the other prod-
ucts have greatly reduced the sup-
ply and increased the costs. During
this same period the requirements
for salmon and trout fingerlings to
maintain fisheries increased tre-
mendously. The result has been a
search for substitute foods readily
available in quantity and at low
cost. Many waste products have
been tried, but only salmon vis-
cera— eggs, testes, and ovaries — are
available in large quantities, es-
pecially in Alaska . Salmon viscera,

when combined with some fresh
meats, have proved nutritionally
equivalent to foods previously fed.
It is probable that salmon viscera
from Pacific Northwest canneries as
well as large quantities from Alaska
must be utilized, as the annual fish-
food requirement of Pacific-coast
salmon hatcheries alone is expected
to exceed 10 million i)()uiids within
a few years.

A practical hatchery diet must be
made up of components available in
quantity at low cost and providing
the nutrients necessary to produce
desired growth with minimum mor-
talities. Whether raw or frozen,
the food must be fresh when re-
ceived and must be retained un-
spoiled until fed to the fish. Meat
or fish products that have been
frozen should not be permitted to
thaw until required for the fish, and
thawed products or diets should not
be refrozen. The materials to be
incorporated in the diet should be
such as will combine to be presented
to the fish in the most desirable con-
sistency and form.

Dry animal meals have been used
extensively in both trout and
salmon diets. Unfortunate results
have attended some of these efforts,
but satisfactory results may be se-
cured by preparing the meal at low
temperatures and considering the
percentages to be included in the
diet, the water temperatures, and
the species to be fed. Good results
have been achieved at the Grand
Coulee stations of the Fish and
Wildlife Service, where Burrows
(19-19) determined that—

the Leavenworth production diet more
closely fulfills the various requirements

34

of a practical diet for the blviebjuk [red |
salmon than any other developed to date,
niis diet at tenii)eratures above 50° F.
consists of 20 i>erceut each of beef livei-,
hoR liver, and hog spleen^ SO percent
salmon viscera, and 10 percent flame-dried
siilmon-offal meal. At temperatures be
l(i\v r>()° the meal is deleted from the die!
and tlio other ingredients increased pro
Ijortionally.

At all salmon and trout hatcllel■ie^^
the objective is to produce healthy
fish at the minimum cost. One gen-
eral measure of success is the pounds
of food required to produce a pound
of salmon or trout. A few years
ago 4: to 5 or more pounds of food
were required to produce a pound of
Hsh. This ratio has been consider-
ably reduced through research ; now
it is not uncommon to average about
3 pounds of food to 1 pound of fish.
In general, costs, too, have been
reduced, but this does not neces-
sarily follow, as the foods in the
low-ratio diet may be more expen-
sive than those in other diets.

At salmon hatcheries of the Fish
and Wildlife Service, it is con-
sidered most desirable for the fish
to be presented suitably prepared
food that floats on the surface of the
water for a time. This is accom-
plished by preparing food that does
not easily separate in the water.
There is less leaching of the food,
and consequently almost all of the
food is available to the fish. Cer-
tain ingredients tend to bind foods
together. Salt in combination with
hog, beef, or horse liver will form
a rubberlike mass that is resistant
to water. Dry^ meals in a food will
absorb the nutritive juices of the
meats, and the binding action will

hold tlie food together so fish will
receive the full benelit.

Reconnnended hatchery practice
is to juvpare and feed a food in
the same (h>y or within -21 lioui's at
the uiost. l*repared foods should
not be ref rozen but shoidd be held at
low temperatures.

The preparation of food involves
grinding the components to a fine-
ness required for the fish size, mix-
ing, and presenting the prepared
food to the fish at regular inter-
vals— several times each day to
newly feeding fish and once or twice
daily to larger fish. For the small
fish the consistency of a diet may be
changed so that the food will break
up more readily in the water to pro-
vide particles fine enough for the
fish.

The quantity of food to be pre-
sented to the fish is governed by size,
species, and the temperatures of the
water. The water temperature de-
termines the activity of the fish and
of the digestive processes. The in-
take of food to meet nutritive re-
quirements increases as the water
temperature rises, and vice versa.
Growth of salmon in hatcheries is
extremely rapid during the first
year of life, when the body weight
increases many fold. The food re-
quirements of the various species of
salmon and trout are definitely dif-
ferent. As a general guide to the
feeding of salmonoids, charts have
been prepared by which the quan-
tities of food to be fed to a pond of
fish may be determined if the total
weight of the fish in the pond is
known. The first charts w^ere by
Tunison (1936) and by Deuel,
Haskell, and Tunison (1937, 1942).

35

Figure 16. — Weigliinii' a sample
weight of all tish in rh ejxi

)f a known nnml
(1 ti) ascertain qi

»f salmon finger
rities of food to

gs to determine
fed daily.

Modified feeding charts have been
prepared for the various species of
sahnon.

To use feeding charts it is neces-
sary lo know the weight of the fish
in each ])on(L By weigliing repre-
sentative samples of fish at intervals
and knowing the numbers of fish in
the pond, tlie weight of all fish can
be calculated. At large hatcheries,
where it would be impractical to
weigli numbers of fish in each pond
at semi weekly inter\als to de-
termine amounts of foods to ()e fed.
the Fi.-h and Wildlife Service has

developed the use of pilot lots of
fish. These lots are held in small
laidvs and are weighed to indicate,
with the application of factors to
compensate for differing conditions,
tlie weights of the fish in the many
large i)()nds (Palmer et ah. f^rtii).
The daily offering of food to the
lish is from 10 |)ercent to .") percent
of the total weight of the fish —
the smaller iisli receiving the greater
pei-centages.

A variety of methods ha\e been
employed to disti'ibute the food to
lish. 'I"lu' problem is one of [)re-

36

seiitini:' ;i di'siraWlc food in pfopcr
form and of (list I'ilxit iiiu' i( so as to
eliiuiiiate waste and assure an ade-
((uate daily ration for oaeh lish. A
converted potato ricer witli holes of
api)ropriate size lias been used suc-
ressfully wlien feedini>- small lish in
ti'ou<>hs. Food for the tish is i)laced
in the ricer and pressure is a])plied
by -liaiul to squeeze the food throuah
small lioles in the bottom of tlu>
ricer. which is moved l)ack and forth
in the water, breakino- oti' the
"worms"" of food. An extension of
this })rincii)le has been developed at
t he Leavenworth station of the Fish
and Wikllife Service, where a lai-<iv
hand-operated })ond ricer is used to
feed ponds. At the Coleman station
of the Service the same principle is
applied, but compressed air is used
to drive worms of food through the
ricer perforations. The salmon
liatcheries now being constructed
for the Service include permanent
air lines to ponds for pressure feed-
ing. These methods have resulted
in si.<>nihcant savinj^s in funds and

lanpow ('
a\(' |)i'o(
»i'm size.

d li

the s;
Ithv lis

Figure 17. — Placing Icod
pond with air-pressure

n a salmiin
•icer at the

Coleman (Calif.) Hatchery of the Fish
and ^^■iIdlitV Service.

THE HATCHERY AND FISHERY MANAGEMENT

The fingerling salmon prodnced
in the modern hatchery are utilized
to the best advantage in the mainte-
nance and development of popula-
tions of fish decimated or threat-
ened w^th extinction. A number of
hatcheries contribute to sport and
commercial fisheries by making pos-
sible the capture of substantial
numbers of adult salmon that w^ere
released as hatchery-reared finger-
lings.

Under natural conditions, that is.
in the absence of man-made deter-

rents to natui'al reproduction, the
tremendous salmon populations
were able to maintain themselves by
entering the many coastal streams
to reproduce. The early efforts of
Indians in capturing salmon were
relatively insignificant, for more-
than-adequate numbers of fish usu-
ally reached the spawning areas.
The fishing methods of the white
man were much more intensive and
efficient, but even then the tremen-
dous catch need not have been par-
ticularly harmful to the runs. Sci-

37

entific regulation of the fisheries
would have permitted the escape of
the necessary numbers of salmon
during the feak of the run, for these
were the fish having the greatest
successful s p a w n i n g potential.
However, biological data were not
available to indicate the desirability
of such escapement, and extreme
decimations of some stocks of
salmon resulted. Basically, then,
all that is necessary for the mainte-
nance of a fishery is regulation to
permit adequate escapement of the
best spawning stock.

Unnatural conditions exist today
in most of the salmon-spaw^ning
streams of the Pacific Coast States
and in many of the streams of Brit-
ish Columbia and Alaska. These
conditions have been discussed:
dams and pollution, denudation of
watersheds, irrigation diversions.
These and other conditions have
created blocks to upstream and
downstream migrations of salmon,
have destroyed or made unsuitable
and inaccessible hundreds of miles
of spawning area. These projects
have interrupted the finely balanced
timing of the salmon on their
spawning migrations, and, regard-
less of man's ingenuity in mitigat-
ing the harmful effects of a dam
(for instance, by the inclusion of
fish-passage facilities), the salmon
may still be delayed too long to ar-
rive at the spawning area during the
period of desirable water tempera-
tures.

The salmon hatchery often, but
not always, can substitute in part
for natural reproduction lost by rea-
son of man's activities. The pro-
gram on the Columbia River, for

example, includes salmon-salvage
projects of unprecedented scope in-
volving the transfer of very large
runs of salmon to other than their
natal tributaries; it also includes
the further development of salmon
populations in the lower-river trib-
utaries to compensate in part for the
anticipated loss of major portions
of up-river races by reason of mid-
dle-river dams constructed and
planned.

In the first instance, occasioned by
the construction of Grand Coulee
Dam, dependence was placed upon
the instinct of salmon to return to
the stream in which they were born
or in which they spent their early
months before migrating to the
ocean. This program, dependent
upon hatchery operations to a great
extent, has been particularly suc-
cessful. Hatchery production and
stocking of red (blueback) salmon
can be directly correlated with the
tremendously increased runs of this
species returning as adults.

The lower Columbia River pro-
gram has not yet proved successful,
but there is every indication that
the clearance of streams (to permit
greater utilization of spawning and
rearing areas) and hatchery stock-
ing will result in maximum produc-
tion in the lower river tributaries
that will largely replace loss of up-
stream production.

Studies of the results of hatchery
production on the Sacramento River
in California have revealed the
major contribution of the hatchery
jH-oduct directly to the fishermen.
(). B. Cope and D. W. Slater (un-
publislied report) state that hatch-
erv fish released from the Coleman

38

station of the Fish and Wildlife
Service contribute an aveia<»-e of
14.5 percent of the catch of tlie net
fisliery alone. Preliminary data in-
dicate the ini])()rtaiu'e of the salmon
from this hatchery in the offshore
troll fisliery northward from the
Sacramento Rivei' to Vancouver Is-
land, Caiuula.

These and other hatchery opera-
tions have been successful in par-
tially compensatino- for the salmon

losses caused by water-use projects.
At many localities the direct and
economical contributions of hatch-
eries to the salmon fisheries can be
demonstrated. It is not inconceiv-
able that hatchery propagation
alone can maintain a salmon fishery
of some magnitude in the event that
conditions preclude natural repro-
duction and if the requirements for
successful hatchery operations are
available.

THE SALMON HATCHERY

Hatchery propagation of salmon
is not recommended as a substitute
for natural propagation but may
be necessary to maintain salmon
population under certain circum-
stances. The construction of a
salmon hatchery is justifiable where
there is a particular need.

A successful installation requires
an appropriate site and adequate
water of suitable quality and tem-
perature. Plans for a hatchery
must take into consideration the
numbers and species of salmon to
be produced, and the sizes of fish
to be released. These factors will
determine the rearing space re-
quired, the number of troughs or
other facilities needed to incubate
the eggs, and the quantity of fish
food to be stored, prepared, and fed.

Hatchery Water Supply

Most important for the success of
a salmon hatchery is the water sup-
ply in which the eggs are incubated
and the young fish reared. The sup-
ply must be adequate in volume.
Each hatchery trough may require
5 g. p. m. (gallons per minute) or

more ; a rearing pond should receive
from 50 to 400 g. p. m., the amount
depending upon pond size and the
numbei-s of fish, as well as water
temperature and oxygen content.
A typical hatchery of 150 troughs
and 20 raceways, using water only
once through each trough or pond,
should have available not less than
7,000 g. p. m. or about 14 cubic feet
per second. If there are adult hold-
ing ponds, these can receive the
water from the ponds. Under
crow^ded conditions, when the num-
bers of adults approach the maxi-
mum capacity of the ponds, an ad-
ditional supply of fresh, well-
aerated water should be introduced.
The water to be used for salmon,
and trout, propagation must be of a
temperature within a definite range,
should be relatively free of silt and
debris, must be unpolluted, and
should not contain excessive quan-
tities of dissolved gases such as
nitrogen or of minerals such as cop-
per and iron. Within the range of
the Pacific salmon, hard water
rarely is encountered; most of the
coastal streams in which salmon are

39

found, and wliere hatcheries are
located, are rehxtively soft water.
The water should have a dissolved
oxygen content of not less than 5
parts per million wlien introduced
into the trough or pond, but greater
oxygen content may be required if
ponds or troughs are heavily
stocked, depending upon water tem-
peratures.

The numbers and size of fish in
a pond or trough, and the water
temperature, vrill determine the
quantity of water required to be
introduced. There is no rule by
whicli the number of fish to be
stocked may be determined, al-
though a pond may be expected to
carry up to 1 pound of fish per cubic
foot of water. Generally a greater
weight of larger fish than of small
salmon can safely be carried in a
pond or trough, but capacities vary
with the species. The maximum
carrying capacity of a trough or
{)()nd should be determined by trial
at each hatchery. The total pound-
age should be determined with ap-
propriate consideration for pro-
longed treatments for disease and
for water drawdown when cleaning
ponds or removing fish.

Generally the water supply for
the hatchery is drawn from a stream
draining an established watershed
within which are few if any farms
{ind livestock. Such a watershed
usually provides impolluted and
relatively silt-free water. At some
hatcheries it is necessary to provide
a sand-gravel filter to remove ob-
jectionable silt.

The availability of spring water
with a temperature range of 40° to
(^0° F. is desirable to warm winter

stream waters used in the hatchery.
Often warmer waters are needed to
promote more rapid growth in order
to meet certain planting require-
ments. Present knowledge indi-
cates, however, that the rearing of
some species, particularly spring
chinook, shoidd not permit too
ra})id growth. The young salmon,
if they are reared rapidly to large
size (6 to 8 inches) may exliibit a
desire to migrate to the ocean in
midwinter, when stream conditions
are most unfavorable in some locali-
ties.

For the most efficient hatchery
production of salmon, the water
temperature should range from
about 45° to 60° F. Incubation
temperatures should be 45° to 55°
F., and rearing temperatures of 50°
to 60° F. are desirable. Higher
temperatures are not to be souglit
because of disease development, and
lower temperatures retard growth.

The supply of water for the
luitchery may he carried from the
source in a pipe or flume ; an earthen
ditch is not recommended because
of the algae growth and the possi-
bility that disease-carrying fish may
Cnd the ditch a suitable habitat.
Flumes should be covered if leaves
and wind-borne debris are present.
The intake structure on a stream
usually includes a l)ari-ed grill or
grissly to exclude logs and large
debris, and a revolving or other type
screen to remove smaller debris and
to divert fish back to the stream.
Also at the intake is a \alve or stop-
log arrangement for control of the
'^ater.

Water may be introduced into
troughs or })onds tlirectly from a

40

Fi(.L];l lb. — ltc;\(jl\iii„ ^LiLL'U 111 wuLlt sui)i)l\ Hume diverts fish and debri.s baeli tu
stream at Little White Salmon Hatchery of the Fish and Wildlife Service.

]Mpe or from a head trough or flume,
lu the latter case, adjustable open-
ings or spigots control the flow of
water into each pond or trough.
The manner of introduction of the
water determines the extent of
increase of bound oxygen.

If circular ponds or other types
of j)onds in which the water is re-
circulated are in use, it is necessary
to have enough force or head behind
the entering water to circulate the
pond water properly. Here, for best
results, the water is introduced from
a pipe.

Water may be used more than
once, that is, it may be passed from
one pond or trough into a second
pond or trough and thence into
others. There should be sufficient
fall of the water from the foot of
one pond into the upper end of the
next pond to assure adequate dis-

solved oxygen. One serious diffi-
culty when passing the same water
through a series of ponds or troughs
is the possible spread of disease
throughout the series, whereas
l^onds supplied with individual wa-
ter would limit the disease to the
single pond except where the source
of water for the entire hatchery is
atfected, or tools and equipment are
contaminated. There apparently
are no advantages to reusing water
\7hen an adequate quantity is avail-
able.

Experience has shown the neces-
sity of selecting a water supply suit-
able for the propagation of salmon
a] id locating the hatchery at that
point. The disadvantages of many
such locations as to isolation are tol-
erable, but an inadequate or unsuit-
able water supply will prevent suc-
cessful production of fish.

41

Fj(;ri:K ID. — ]'.;ittery of .snljiK.n I'l-aiiiii; ponds, larcc sp.iw iiiiii; jioikU .md fishway
from stream to ponds at the Little White Salmon Hatchery of the Fish and Wildlife
Service.

The Hatchery Building

It lias become quite common for
modern liatclieries to include in the
hatchery building- not only the
troughs for the incubation of eggs
and the rearing of very small fish
but also the service facilities needed
in connection with artificial propa-
gation : the food-preparation room,
cold-storage facilities, garage, shop,
office, and often a small laboratory.
When the establishment is quite
large, it may be more economical to
erect a separate garage building.
To have all of the facilities under
one roof is convenient, usually re-
quires smaller initial cost, and re-
duces annual maintenance costs.

Trouf/hs. — Hatchery troughs in
the early days of salmon cultiu"e

were connnonly constructed of red-
wood or cedar with inside dimen-
sions of about 14 feet long by 16
inches wide by 8 inches deep.
Although some fish-culturists con-
tinue to prefer the shallow troughs,
most salmon hatcheries constructed
in recent years have been equipped
with deep-type troughs, of the same
dimensions except that the depth is
usually 16 inches. A more modern
innovation is a trough 3 feet deep
by about 8 feet long which has the
same egg capacity as the deep-type
trough but requires only about one-
half the floorsjDace. Troughs often
have been constructed of more mod-
ern materials, such as concrete, alu-
mininn, and enameled sheet steel.
Plastic or plastic-covered troughs

42

Figure 20. — Interior view of Quilcene, W;i8ii., Hatchery of the Fish ami Wihllife
Service. Double-deep-type troughs contain baslvets of salmon eggs.

have been tried, but the wood
troughs continue predominant,
])rincipally for economy.

Trays for the incubation of
sahnon eggs are of various sizes (for
the deep troughs, 14 inches by 16
inches) ; most have a wood frame
(11/2 inches by % inch) on the bot-
tom of which is fastened a screen
of suitable mesh. The eggs are
phiced on the trays in a single or
double layer, and the trays are
stacked and placed in the troughs.
An arrangement of metal dam
boards directs the flow of water up
through a stack of trays, back to
the bottom of the trough, and up
through the next stack of trays.
The trays stacked in a deep-type
trough have a capacity of about
200,000 chinook-salmon eggs, run-
ning 50 to 80 eggs per fluid ounce.

When the young fish are free
swimming and capable of taking-
food, they are taken from the trays
and i^laced directly in the troughs
(from which the dam boards have
been removed ) . A wire-mesh screen
or a perforated plate at the lower
end of the trough prevents the es-
cape of fish. The water level of the
tiough usually is maintained by a
standpipe inserted between the
screen and the lower end of the
trough. The pipe can be removed
readily to facilitate cleaning of the
trough.

Water for the trough is intro-
duced either from a head trough or
by pipe, and the water may be
passed from one to another of a
series of troughs to conserve water,
but with the disadvantages men-
tioned previously. Troughs usu-

43

Figure 21.

-Stack of trays containing salmon eggs being lowered into a deep-type
trough for the period of incubation.

ally are mounted on horses or pedes-
tals about waist high. Occasionally
troughs may be placed on or in con-
crete tanks which later are used for
rearing.

In some hatcheries the waste
water from the troughs may be
utilized in outside ponds. At the
Spring Creek (Wash.) Hatchery of
the Fish and Wildlife Service, it is
possible to drain the fish from the
troughs into the waste-water trough
in the floor and to divert known
numbers of these very small fish into
the various ponds for further rear-
ing, or directly to the fish ladder
down which they will descend to the
Columbia River and thence to the
ocean.

Food preparation room. — At
salmon hatcheries the young fish are

fed for different lengths of time be-
fore release. The diet received by
these fish is of major importance.
To prepare the foods properly for
presentation to the fish, a special
room and items of equipment are
required. The size of the room and
the size and variety of equipment
depend upon the quantities of food
to be prepared and the period of
t ime over which such quantities will
be required. A typical salmon-
liatchery feeding program starts
with the preparation of only a few
pounds of food for very small fish.
As the salmon grow, the quantities
of food required gradually increase
until a ton or more of prepared food
may be needed each day.

A feeding operation of this mag-
nitude is accomplished in a food-

44

FiGi Ki; 22. — Typical food preparation room at the Quilcene, Wash., Hatchery of the
Fish and Wildlife Service. Bandsaw, grinder and vertical mixer are standard
equipment. Note chutes from dry meal storage bins.

preparfition room with equipment
such as a large meat grinder, a
heavy-duty bandsaw, a dough mixer
and/or a Large vertical mixer, and
a bone-and-gristle chopper, all pow-
er driven, and scales and sink. Con-
struction of the room is such as to
provide adequate light, good drain-
age, easily cleaned walls, and pro-
duction-line arrangement of equip-
ment.

Food s t o r a g e. — Cold-storage
space is essential, as most of the
food for salmon is meat or fish
waste. A reasonable measure of
the space needed is enough room to
freeze and store one-half of the total
perishable food requirement for a
year. If all of the foods received

temperature of the cold room should
be maintained at about 0° F., with
the ability to freeze rapidly large
quantities of fresh meats received
in one shipment. If fish products
are to be held in cold storage, the
temperature should be maintained
near —10° F. or below, with a ca-
pacity of at least —15° F. to freeze
quickly fresh-fish products such as
salmon carcasses.

At most salmon hatcheries a cool
room is constructed as a first en-
trance into the cold room. In the
cool room are stored those prepared
foods that are to be kept cold for
the next day's feeding but that
should not be refrozen. The cool
room is entered directly from the
food-preparation room. The tem-

45

peratiire of the cool room may be
maintained by the effect of the ad-
joining cold room, but more often
a separate refrigeration unit is
necessary to maintain the cool-
room temperature just above freez-
ing.

Animal and fish meals often are
used in fish hatcheries. These
products are stored in rodentproof
rooms or bins, usually in the loft
over the food-preparation room.
Downspouts from the bins, with
controls, make the meals easily
available in the feed room. (See
fig. 22.)

Rearing Ponds

Eearing ponds specifically de-
signed for the most convenient and
efficient operation are required at
salmon hatcheries. These ponds
are of a variety of sizes and shapes,
but certain designs have proved
most successful.

The early ponds used for rearing
of young salmon, particularly chi-
nook and silver, were earthen, usu-
ally very long and narrow, and had
the water flowing from one pond
through a second and often through
several j^onds. The ponds at the
modern salmon hatcheries are of
concrete, and the tendency is to pro-
vide a good flow of water and use
it only once when an adequate water
supply is available. This practice
permits easier cleaning of the ponds.
In certain types of ponds the flow of
water over the concrete bottom, as
well as the movements of the fish,
tends to concentrate debris at the
outlet of the pond. Probably the
greatest advantage of the concrete
pond is in this cleanliness, there

being minimum lodgement of dis-
ease organisms; whereas the oppo-
site is true of the earthen ponds.
When disease must be treated with
chemicals, the concrete pond lends
itself to such treatment. In gen-
eral, despite the greater construc-
tion cost, the concrete pond is pref-
erable to earthen ponds.

Probably the most common type
of rearing pond at salmon hatch-
eries is the raceway, in which a sub-
stantial and uniformly dispersed
flow is maintained from the water-
intake end to the lower or screened
end. (See fig. 19.) Ponds of this
type are preferred at Fish and
Wildlife Service hatcheries. The
ponds are usually 8 feet wide and
80 feet long and have an average
water depth of 2.5 feet. From the
upper to the lower end, the bottom
has a slope of 1 inch in each 10 feet
of length. The inflow of water is
200 to 400 g. p. m., and spills from
a head trough into the pond in a
slieet as wide as the pond. This
manner of introducing water pro-
vides as nearly as possible a uni-
form movement of water down the
pond witliout dead-water areas.
The screen and dam boards also ex-
tend the full width of the pond.
Tlie removable screen (of appro-
priate mesh, galvanized after weav-
ing) or a perforated plate, placed
in slots or grooves in the concrete
at the lower end of the pond, pre-
vents the escape of fish. The dam
boards, also removable, are located
downstream IVoiu the screen and
maintain the (U'sired ])()nd level. At
some liatchei-ies (he lowei' end of the
l)()ii(l is seak'd by a concrete Avail
and the overflow i>oes down a stand-

46

pipe of appropriate height located
hotwoon tlio oiul Avail uiul the screen.

The r;uH'\va_v-t y|)c rcarino- pond
lias sonic disadvantages. A very
substantial supply of water is re-
quired, although this could be re-
used in other poiuls in a series.
Young fish show a tendency to accu-
mulate at the inflow end of the
pond, thus apparently not utilizing
clliciently the space pi-ovided.
However, most efficient utilization
of the pond space by stocking the
pond to near capacity will result
ill better dispersion of the fish. It
is believed that the foregoing, if a
disadvantage, is minor as compared
to the advantages of this type of
pond in disease inhibition, ease of
feeding:, treatment for disease, and
handling of fish.

The Washington State Depart-
ment of Game has used most suc-
cessfully a circular concrete pond,
usuallv 4:0 feet in diameter, about

24: inches in water dcptli, and hav-
ing a very slight bottom sh)pe to-
ward the center staiid-i)ipe outlet,
which is surrounded by a square
screen. Exceptional weights of
trout have been produced from
these ponds.

The circular pond i'e(|uires a
minimum of water, and its use is ad-
vantageous where water supplies
are limited. Because of the con-
tinuous circular How of water, how-
ever, disease organisms have greater
o})portunity to attack the fish and
disease treatments are more diffi-
cult.

The Washington State Depart-
ment of Fisheries and the Oregon
Fish Commission favor a rectan-
gular pond with a center partition
that stops short of the end walls to
permit circulation of the water.
(See fig-. 23.) In some of these
ponds the inflow and outflow are at
the same end of the pond. In later

Figure 23. — Salmon rearing ponds at the Green River Hatchery of the Wasliingtoii
Department of Fisheries.

47

years, the inflow and outflow have
been placed at opposite ends of the
pond, eliminating dead-water areas
to some extent.

Of the many types of rearing
ponds in use, the raceway pond is
believed best suited to mass produc-
tion of salmon fingerlings. Its
capacity, measured in production
per cubic foot of water, or per gal-
lon per minute, probably is not so
great as the circular pond; but its
disease-inhibiting characteristics,
as well as ease of cleaning, feeding,
and handling the fish make it desir-
able where ample water supplies are
available.

Trapping Adult Salmon

The location of traps for the tak-
ing of adult salmon for spawning
will depend upon the species of fish
sought. The most desirable trap-
ping locality is immediately below
the natural spawning area, where
the fish can be taken when fully ma-
ture, but this usually is not feasible.
The trapping for spawning of
spring chinook, silver salmon, and
steelhead trout offers the greatest
difficulties, for many adults of these
species may need to be held for
weeks or months before spawning.

Until recent years it had been
the general practice to place racks
or weirs across a stream at a suitable
point to enclose an area within
which the adult salmon were
trapped for spawning. The lower
rack was so constructed as to permit
the upstream passage of fish
through one or more V-openings
which could not easily be found
again for escape. The upper rack
permitted no passage. The mature

fish were seined from the area and
fipawned. Several variations of this
type of rack are sketched and de-
scribed by O'Malley (1920).

Although the trapping and
spawning of some species of salmon
in the stream continue to be neces-
sary where circumstances permit, a
general practice is to install a di-
version rack or dam in the stream
and thus to divert the adult fish into
a side channel or up a fish ladder
:and thence into concrete or earthen
holding and spawning ponds.

A rack placed across a stream con-
sists of weighted wood or steel tri-
pods between which stringers are
extended. Sections of suitably in-
terspaced strips laid vertically are
placed or leaned against the string-
ers on the upstream side of the tri-
pods. The interspaces in the rack
are of such width and number as
to permit the free flow of water
through the rack but are sufficiently
small or narrow to deny passage of
the salmon. The rack must be fish-
tight along the stream bottom.
O'Malley (1920) gives a detailed
description of rack construction.

The streamflow characteristics, as
well as the size and quantity of
debris, such as logs, brush, and
leaves, must be considered when de-
termining the size and strength of
the rack. The largest salmon rack
ever attempted was laid across the
main Sacramento River. The steel
tripods were 12 feet high, and each
of these weighed many tons when
loaded with rock (on a platform
within the three legs) to hold posi-
tion in the cuireut. The racks were
of steel or strap iron. Flash floods
in the river frequently shifted or

48

Figure 24. — Dam constructed across creek to divert adult salmon up fishway and
into holding ponds. (See also fig. 1.)

carried away sections of the instal-
lation. Identical tripods have been
utilized in the Big White Salmon
Eiver, tributary to the lower Co-
lumbia River.

Smaller rack installations are the
rule and usually can be held in the
stream with constant attention and
effort through the few weeks of the
spawning season. Under favorable
circumstances it is possible to rack
only a portion of a stream and thus
divert part of the run of salmon into
holding ponds. At other places,
such as Walcott Slough on Hood
Canal off Puget Sound, Wash., a
permanent trap is installed in a
small spring outlet within the reach
of tidal rise. Here chum salmon
are spawned with utmost conve-
nience, as the adults entering the
traps are fully mature. This run,
incidentally, is hatchery-created
and hatchery-maintained.

In certain streams where the bot-
tom and channel are stabilized, con-
crete aprons about 15 to 20 feet wide
often are placed across the bed of
the stream, with appropriate means
of anchoring tripods to the apron.
Often, too, concrete blocks have been
poured at intervals across the apron
to replace the tripods. The racks
are then laid against the sloping up-
stream face of the blocks. Where
floods and logs are encountered, such
permanent installations are undesir-
able because they form a solid foun-
dation for a debris dam and
resultant flooding of surrounding
areas. On the Klickitat River, un-
der the Federally financed fisheries
program, the Washington Depart-
nient of Fisheries has installed a
new kind of tripod that is most con-
venient and successful in this type
of stream with no heavy debris.
This installation consists of a con-

49

Figure 25. — Salmon trapinng and spawning installation by tlie Fish and Wildlife
Service on Wind River, Wasli. Adult salmon enter the enclosure through a small
opening in the far rack and are seined from the area.

ciete apron with permanently fixed
pipe-rack supports to replace the
tripods. At the conclusion of the
spawning period, the racks and the
plank walkway across the top are
removed and the individual pipe
supports, attached by swivel joints
at each end, are lowered to the apron
surface. For the next spawning pe-
riod these are caught with a hook
and raised to a vertical position, and
the walkway and rack sections are
placed. This is purely a rack to stop
salmon from further ascent of the
river and to divert them into pre-
pared holding ponds.

Another type of diversion struc-
ture is that placed permanently in

Battle Creek, tributary to the Sacra-
mento River, to divert adult salmon
into the holding ponds of the Cole-
man station of the Fish and Wild-
life Service. (See fig. 24.) This
structure is a low concrete dam with
a downstream sloping surface and
apron over which the water pours in
a sheet and in which salmon cannot
secure a "foothold" to attempt a
jump. During the salmon runs the
fish are diverted to a fish ladder
leading to large holding ponds.
Tliroughout the rest of the year, a
fish ladder in the dam itself is open
to permit ascent of trout and some
salmon to the stream above.

The type of rack installation for

50

I'iGUKK 2G.— Seining chum salmon from the holding area at the Walcott Slough,
Wash., spawning station of the Fish and Wildlife Service. Entrance V at lower
right. --■^M^'^^^ik

either trapping or diverting the
sahnon to holding ponds will de-
pend upon the character of the
stream and upon the species of fish
to be spaAvned. Also dependent
upon these factors will be the deter-
mination whether fish will be
trapped in the stream or diverted
to holding and spawning ponds.
The latter is most convenient and
successful for sjoawning operations
at a hatchery where water supplies
are readily available for the ponds.
In isolated areas or at points re-
moved from the hatchery, a tempo-
rary installation for trapping and
spawning directly in the stream is
most economical.

Adult Holding Ponds

Holding or retaining ponds for
adult salmon differ greatly in size,
depth, and general features. The
principal differences are found in
ponds designed for receiving and
holding fall chinook salmon and
those for spring chinooks. The
spring chinook is more vigorous and
aggressive in attempting to leave
the pond and move further up-
stream. He also must be held much
longer before maturing than the fall
fish.

The retaining ponds for spring
chinook salmon are quite compli-
cated and are carefully designed
and constructed to avoid death of

51

adults as a result of injury and
disease. At the Grand Coulee sta-
tions of the Fish and Wildlife Serv-
ice, nmch attention has been given
to the development of satisfactory
ponds. The most successful pond
yet devised, and one in which spring
chinooks have been held very satis-
factorily, is described by Koger
Burrows (unpublished) of the Fish
and Wildlife Service.

The retention of the fish until sexual
maturity must meet two main require-
ments for successful operations. First,
it must prevent mechanical injury to the
adult stock, and second, it must inhibit
the development of disease.

(1) It is characteristic of adult salmon
when barred from further upstream prog-
ress to circle the area of impoundment,
attempt to move upstream, and finally
return to the upstream barrier and at-
tempt to surmount it by jumping. Fish
thus trapped will circle and mill around
restlessly for 10 days to 2 weeks but
eventually will settle down in the deeper
portions of the area to await maturity.
By that time, however, the damage has
been done as far as mechanical injury
is concerned. Fish have jumped out on
the banks and died or returned to the
pond severely abraded, injured them-
selves by hitting obstructions, or worn
their nose fighting the upstream barrier.
Any traumatic condition resulting from
mechanical injury no matter how minute
opens an avenue for infection and the
death of the fish may result.

It is a relatively simple matter to pre-
vent fish from jumping in a retaining
area. Fish jump at the point of water
inflow but they also jump at vertical
banks. To prevent the latter all banks
in an artificial pond should have at least
a 45° slope. Salmon before making a
jump nose along at the surface of the
water. If the water surface is covered
the fish will not jump. Taking advan-
tage of this characteristic the water sur-
face in the area immediately below the
obstruction is completely covered by
floating strips of canvas.

By the use of a submerged culvert the
water may be introduced into the cen-
ter of the upper section of the pond so
that the fish can jump but still do them-
selves no harm by striking the banks.
As far as is know^n jumping by salmon
produces no ill effects as long as mechan-
ical injury is not incurred. This type
of barrier to the upstream migration has
two advantages over the usual weir, first,
the fish do not fight the upwelling water
sufficiently to cause abrasions and, sec-
ond, if properly placed or covered, jump-
ing will either do no harm or can be
prevented.

(2) The type of holding area and the
water temperature are the two major
factors concerned with the development
or inhibition of disease.

A holding area to be satisfactory must
have a continuing and fairly rapid inter-
change of water throughout the entire
impoundment. Natural pools fulfill this
requirement otherwise they would silt
up and no longer be pools. Salmon pre-
fer deep water in which to lie between
migrations but this preference should not
be satisfied at the sacrifice of water
interchange. If relatively stagnant
water is present in the deeper portions
of the area, disease organisms which are
waterborne have an opportunity to con-
centrate in these portions. As these
areas are the parts preferred by the fish
the infections become dominant and the
natural resistance of the fish is not sufii-
cient to cope with the disease. A good
holding area, therefore, should have a
bottom contour such as to avoid the
development of areas of semistagnant
water.

The temperature of the water is, also,
an important factor which controls the
development of disease. Both fungus
{Saprolegnia imrusitica) and columnaris
{Chondrococcus columnaris), the princi-
pal causes of mortality in adult fish, do
not reach their point of optimum develop-
ment at temperatures below 60° F. For
this reason holding areas in which the
water temperature does not rise above
60° may be operated with a smaller water
inflow and, consequently, a slower rate
of interchange than areas which have

FlOLKE 'Si

-Adult lipeiiiug and spawning (foreground) ponds to which adult sulmc
are diverted from the river in background.

temperatures in excess of 60° for ex-
tended periods. It is believed that an
increase in the rate of interchange of
the water in a holding area can be sub-
stituted for low water temperatures to
inhibit disease development."

The retention of adult fall chi-
nook salmon until ripe for spawning-
is not so difficult as the retention of
springs. The former fish usually
are almost ready for spawning when
trapped and need to be held in the
])onds for only a few days. These
fish, also, are not so aggressive and
can be retained more easily in the
ponds. For these reasons it is not
necessary to prepare such elaborate
holding ponds. This is true of chum
salmon, also. An adequate supply
of aerated water is the principal
requirement.

Personnel, Equipment, and
Facilities

The number of persons required
to staff a hatchery during the entire

year depends upon the type of oper-
ation, including species propagated,
climatic conditions, and the location
of the unit. In general, the normal
hatchery may be permanently
staffed with 1 man for each 10,000
pounds of salmon produced. This
rule has been applied to Fish and
Wildlife Service units on the Co-
lumbia River. The production fig-
ure per man is higher than is usually
attained at present. Additional
help required can be recruited lo-
cally when needed.

The salmon hatchery is equipped
with sufficient trucks to facilitate
operations and with a shop in which
repairs can be made to such equip-
ment as traj'S, nets, and racks. Ma-
jor vehicle repairs are most fre-
quently made in a commercial shop
unless a number of hatcheries in an
area can profitably maintain a cen-
tral repair shop with a mechanic in
charge.

53

Accurate records of hatchery op- order to calcidate the quantities of
erations are maintained to facilitate food to be presented to the finger-
activities and to provide a fund of lings in the ponds. Efficient opera-
basic information to evaluate, in the tion of each hatchery requires tabu-
future, the results of the stocking of lation of expenditures for labor,
salmon. Eoutinely, the poundage food, and other items, as Avell as the
of fish on hand must be known in production of fish.

LITERATURE CITED

Black, V. S.

1951. Osmotic regulation in teleost fislies. Part II of "Some aspects of the
physiology of fish," by W. S. Hoar, V. S. Black, and E. C. Black. Ontario Fish.
Res. I.ab., No. 71. Univ. Toronto I'ress.

Burner, C. J.

1051. Characteristics of spawning nests of Colunihia liiver salmon. V. S. Fish
and Wildlife Service, Fi.shery Bulletin, No. 61. vol. 52. pp. 'J7-110.
Burrows, R.

1949. A lu-ophylactic treatment for the control of fungus (Saprolegnia para-
sitica) on salmon eggs. U. S. Fish and Wildlife Service, Prog. Fish-CuL, vol.
II, No. 2.
Canada Department of Fisheries.

1952. British Columbia catch statistics (1951)
Clemens, W. A.

1952. On the migration of Pacific salmon {Oucorhiinchus). Trans. Roy. Soc.
Can. (3) 45, pp. 9-17 (1951)
Clemens, W. A., and G. V. Wilhy

194G. Fishes of the Pacific coast of Canada. Bull. Fisli. Res. Bd. Can., 68,
pp. 81-89.
Craig, .T. A., and R. L. Hacker

1940. The history and development of tiie fisliories of the Columbia River.
Bull. U. S. Bur. Fish., No. 32, vol. 49. i)p. 133-216.
Davidson, F. A., and S. .T. Hutchinson

1938. The geographical distriliution and environmental limitations of the Pacific
salmon (Genus Oin-orhinirhiis) . Bull. U. S. Bur. Fish.. No. 26, vol. 48, pp.
6G7-692.
Davis, H. S.

1947. Care and diseases of trout. F. S. t^ish and Wildlife Servico. Kosearcli
Report 12. 98 pp.

Deuel, C. li., D. C. Haskell, and A. V. Tun i son

1937. The New York State Fisli Hatchery Feeding Chart. N. Y. Conserv.

Dept. 29 pp.
1942. The New York State Fish Hatchery Feeding Chart. N. Y. ("onserv. Dept.,
Fisheries Res. Bull. No. 3. 38 pp.
Fish, F. F.

1948. The return of blueback salmon to the (V)lumbia River. Sci. Monthly,
vol. 46, No. 4. 52 pp.

FoRESTEii, R. E., and A. L. Pritchard

1944. The identification of the young of five species of Pacific salmon with notes
on the freshwater phase of tiieir life history. Rpt. B. C. Fish. I>('pt., pp.
86-97.

54

FUY, D. II.. .lu., Mild K. V. IlTHiHKS

1051. The Ciilit\inii:v salmon troll fislit'ry. Bull. I'ac. yUiw I'isli. Cuniiii., Xo. li,
pp. 7-42.

(JlUJEKT, C. H.

1914-25. ("oiitrihiitions to the life hi.stor.v of the sockeye salmon. ( 1-1(1) Kpts.
Brit. (\>1. Comni. of Fish.

IIOAB, W. S.

lOni. Hormones in lisli. I'art I of "Some aspects of the physioloj-y of lish,"
by W. S. Hoar, V. S. Black, and E. (". i'.lack. Ont. Fish. Kes. Lab., No. 71, Univ.
Toronto Press.
HUMK. R. D.

1893. Salmon of the I'acific coast. Schnudt Label and Lilt. Co., San Francisco.
Jordan, D. S.

1925. Fislies. pp. 292-312. D. Appleton & Co., N. Y.
Jordan, D. S., and B. W. Evermann

1896. Fishes of North and ^Middle America. Bull. U. S. Nat. Mus., No. 47,
part I, pp. 478-482.
Kauffman, D. E.

1951. Research report on the Washington State offshore troll fishery. Bull.
Pac. Mar. Fish Comm., No. 2, pp. 77-92.
MOTTLEY, C. Mc.

1929. Pacific salmon migration : Report on the study of scales of the spring
salmon tagged in 1926 and 1927 off the west coast of Vancouver Island. Contr.
Can. Biol. Fish. (N. S.), No. 4.
Neave, F.

1951. Observations on troll-caught salmon of the west coast of Vancouver
Island. Bull. Pac. Mar. Fish. Comm., pp. 93-101.

19,'j3. Principles affecting the size of pink and chum salmon populations iu
British Columbia. J. Fish. Res. Bd. Canada, IX, 9, pp. 450-491.
O'Malley, H.

1920. Artificial propagation of the salmons of the Pacific coast. Bur. Fish. Doc.
879, appendix II, Rpt. "U. S. Comm. of Fish, 1919, pp. 11-17.
Palmer, D. D., H. W. Newman, R. L. Azevedo and R. E. Burrows

1952. Comparison of the growth rates of chinook salmon fingerlings reared in
circular tanks and Foster-Lucas ponds. U. S. Fish and AVildlife Service,
Prog. Fish-Cult., July.

Pritchabd, a. L.

1934. Pacific salmon migration : The tagging of tlae spring salmon in British
Columbia waters in 1929 and 1930. Biol. Bd. of Can., Bull. No. 41, pp. 1-31.
Rich, W. H.

1939. Local populations and migration in relation to the conservation of Pacific
salmon iu the Western States and Alaska. Am. Assn. Adv. Sci., No. 8, pp.
45-50.

ROSTLXJND, E.

1952. Freshwater fish and fishing in native North America. Vol. 9, Univ. of
Cal. Pub. in Geography.
Rounsefell, G. a., and G. B. Kelez

1938. The salmon and salmon fisheries of Swiftsure Bank, Puget Sound, and the
Eraser River. Bull. U. S. Bur. Fish., No. 27, vol. 48, pp. 693-823.
Royal, L. A.

1951. The past, present, and future production of soclveye in the Eraser River.
Exhibit A, Minutes 46th Meeting, Columbia Interagency Committee.

55

RucKER, R. R., H. E. Johnson, and G. M. Kaydas

1952. An interim report on gill disease. U. S. Fish and Wildlife Service, Prog.
Fish-Cult., vol. 10, No. 1, pp. 19-22.
RucKER, R. R., A. F. Bebnier, W. W. Whipple, and R. E. Bureows

1951. Sulfadiazine for kidney disease. U. S. Fish and Wildlife Service, Prog.
Fish-Cult., vol. 13, No. 3, pp. 135-137.

SCHULTZ, L. P.

1936. Keys to the fishes of Washington, Oregon, and closely adjoining regions.
Univ. Wash. Pub. ZooL, No. 2, pp. 103-228.
ScHULTZ, L. p., and H. A. Hanson

1935. Salmonoid game fishes in the national forests of northwestern United
States. U. S. Forest Service.

Smedley, S. C.

1952. Pink salmon in Prairie Creek, California. Calif. Fish and Game, vol. 38,
No. 2.

Stone, L.

1880. Report of operations at the United States salmon hatching station on the
McCloud River, Calif., in 1878. App. 33, Rpt. U. S. Comm. Fish. 1878, part VI,
pp. 741-770.

TUNISON, A. V.

1936. Tables applicable to hatchery operations. New York State Conserv. Dept.,
Fish Culturist, vol. 2, No. 5, pp. 6-16.

United States Bureau of Fisheries

1928. Alaska salmon tagging operations. Fisheries Serv. Bull. 152, pp. 2-3.
Van Hyning, J. M.

1951. The ocean salmon troll fishery of Oregon. Bull. Pac. Mar. Fish Comm.,
No. 2, pp. 45-76.

Wallace, R. F.

1952. Economic aspects of wildlife resources of the State of Washington. Bull.
State Coll. Wash., No. 19, pp. 1-42.

O

56

5 WHSE 00087