California. Dept. of Fish and Game.

California fish and game (Volume 43, no. 4) online

. (page 2 of 9)
Online LibraryCalifornia. Dept. of Fish and GameCalifornia fish and game (Volume 43, no. 4) → online text (page 2 of 9)
Font size
QR-code for this ebook


Seven hundred and ninety-foui- noi'thern anchovy (Engraulis
mordax), appearing in 27.6 percent of the stomachs examined, consti-
tuted 29.1 percent of the total volume of food consumed. This was the
most important item in terms of volume and frequency of occurrence.
Both young and adult anchovies were eaten ; they ranged in length
from 2.0 to 6.5 inches.

Northern anchovies were utilized as food for the salmon during all 10
months in which stomachs were collected. Consumption was heaviest
in the late summer and fall months, when most of the salmon were
caught inside the 20-fathom curve. This item accounted for 60.4 percent
of the August diet, 77.9 percent of the September diet, 74.7 percent of



KING SAIiMOX FOOD HABITS



261



TABLE 4

Percentage Composition, by Volume, of the Food of King Salmon Taken During July and August,

According to Locality of Capture



Food item



Outside 20 —
fathom curve



July Augxist*



Pacific herring

(C'lupea pallasi)

Unidentified clupeids

Northern anchovy

{Engraulis mordax)

Pacific tomcod

{Microgadus proximus)

Unidentified cadids

Shiner seaperch

(.Cymatogaster aggregala)

Rockfishes

(Sebastodi's spp.)

Unidentified fish remains

Euphausiids

Crab megalops

Octopus

Squid

Unidentified cephalod remains

TotnU

Total volume of food m cubic centimeters
Number of stomachs



11.5



74.2
l.o
8.9
Tr.
0.2
2.1
Tr.



99.9



1,376.1
71



12.1



4.5



71.1
1.4
6.6

2.3
2.0



100.0



443.3
28



Inside 20 —
fathom curve



July



7.7



92.3



100.0



97.5
11



August



4.2
1.2

90.7

0.4
0.3

1.5

0.4
1.4



100.1



848.3
65



* Omitted were seven specimens taken almost on the 20-fathom curve.

the October diet, and 89.6 pfercent of the November diet. Northern
anchovies were of some importance in February and March, when most
of the salmon came from the Dnxbnry Point-San Francisco Lightship
region. Dnring the spring months, however, when the salmon were taken
chiefly beyond the 20-fathom curve, northern anchovies were seldom
encountered in the diet. This was the only item identified from the
stomachs of the salmon captured in San Francisco Bay during Septem-
ber, and the only item to occur more than once in the stomachs of
salmon taken in the Bay during April.

Rockfishes

One thousand, four hundred and fifty-nine rockfishes (Sehastodes
spp.) were found in 21.4 percent of the stomachs and comprised 22.5
percent of the total bulk. This item was second in importance from the
standpoints of volume and percentage of salmon stomachs containing
them. All of the rockfishes taken were small, varying between 1.0 and
7.5 inches in length. Few specimens exceeded a length of 3.5 inches.
Among 519 undigested or slightly digested specimens sent to Mr.
Phillips for identification, 19 species were represented (Table 5).

More than two-thirds of the rockfishes identified were shortbelly
rockfish {Sehastodes jordani). This species attains a maximum total
length of approximately 12 inches, and is of no commercial value.
According to Mr. Julius B. Phillips (personal communication), it ap-
pears to be an important forage fish, for it occurs in great numbers off
the California coast and is commonly eaten by a number of other



2(i2



CALIFORNIA FISH AND CAME



TABLE 5
List of Rockfishes (Sebastodes spp.) Identified



Species



Shortbelly rockfish (Sebastodes jordani)

Speckled rockfish (Sebastodes oralis)

Yellowtail rockfish (Sebastodes flavidus)

Widow rockfish (Sebastodes entomelas)

Whitebelly rockfish (Sebastodes vexillaris) or copper rockfish (.S. caurinus)

Squarespot rockfish (Sebastodes hopkinsi) .

Dark-blotched rockfish (Sebastodes era men)

Aurora rockfish (Sebastodes aurora)

Black rockfish (Sebastodes melano-ps)

Stripetail rockfish (Sebastodes saxicola)

Bocaccio (Sebastodes paucispinis)

Redstripe rockfish (Sebastodes proriger)

Rosy rockfish (Sebastodes rosaceus)

Sharpchin rockfish (Sebastodes zacentrus)

Brown rockfish (Sebastodes auriculatus)

Chilipepper (Sebastodes goodei)

Canary rockfish (Sebastodes pinniger)

Turkey-red rockfish (Sebastodes ruberrimus)

Greenstriped rockfish (Sebastodes elongatus)

Rockfishes (unidentifiable)

Total number of specimens



Number of
individuals



362

23

19

13

8

7

6

4

4

3

2

2

2

2

2

1

1

1

1

56



519



species, including the larger rockfishes. Most of the shortbelly rockfish
consumed by the king salmon were from 3.0 to 3.5 inches in length.

Despite their importance in the total diet, rockfishes appeared in
only six of the 10 monthly samples. Consumption of this item gradually
increased from 0.7 percent in April to 69.7 percent in July, and then
declined to 2.9 percent in September. Although rockfishes were found
mainly in the stomachs of salmon captured beyond the 20-fathom curve,
a few had been eaten by salmon taken inside the 20-fathom curve and
even inside the 10-fathom curve.

Euphausiids

These small shrimp-like creatures appeared in 13.6 percent of the
stomachs and amounted to 14.9 percent of the total volume. Over 24,000
euphausiids were found, more than all the other food organisms com-
bined. >Samples of euphausiids from 120 king salmon were identified.
Only two species were pre.sent — Thijsanoessa spinifera, which was rep-
resented in 118 of the samples, and Euphausia pacifica, w^hich occurred
in 13. One stomach was found with immature Thysanoessa spinifera in
it ; the rest contained only adults.

According to ^Ir. Edward Brinton (personal communication), great
shoals of Thysanoessa spinifera have been seen at the surface of the
water near the islands off southern California, and presumably may be
found in more northern waters also. He added that this seems to be
the onh- species occurring in California n waters which is conspicuous
by its faculty for swarming and shoaling. Regarding Euphausia pacifica,
Mr. Brinton stated that its heaviest concentrations occur farther off-
shore than the continental shelf region in which Thysanoessa spinifera
predominates.



KING SALMON FOOD HABITS 263

The.e small crustaceans, which averaged about one inch in length,
were often consumed in great numbers. More than one-quarter of the
137 stomachs with euphausiids in them contained from 200 to 1,000
individuals, and over 1,500 (170 cc.) were found in the stomach of a
salmon 27.5 inches long.

Euphausiids were taken exclusively from March to August. Con-
sumption was negligible in March, but increased to 33.0 percent and
47.7 percent, respectively, in April and May. They then gradually
diminished in importance until only 2.1 percent of the August diet
consisted of this item. Althougli a few euphausiids were found in the
stomachs of salmon taken just inside the 20-fathom curve, the great
majority of these organisms had been eaten by salmon which -were cap-
tured beyond the 20-fatliom curve.

Pacific Herring

One liundrcd and fifteen Pacific herring (Clupea pallasi), appearing
in 9.1 percent of the stomachs, made up 12.7 percent of the total food
consumption. This was the largest item commonly eaten by the salmon.
The specimens consumed ranged in length from 3.0 to 10.0 inches. How-
ever, the majority were between 7.0 and 9.0 inches. Although usually
there was only one large herring per stomach, on two occasions four
of these fish were found in the stomachs of large salmon.

Pacific herring entered the diet in all 10 months covered by this
study. They appeared in the stomachs of salmon taken at widely scat-
tered points, from the Farallon Islands to one-quarter of a mile off
the coast. This item was most heavily utilized during the months of
February (47.4 percent), March (51.2 percent), and April (25.9 per-
cent). The large herring were represented in all the monthly samples,
but those 5.0 inches or less were taken only in the late summer and fall
months.

Squid

Eighty-six squid accounted for 9.3 percent of the entire diet, and
6.6 percent of the stomachs Avere found to contain this item of food.
Most of these squid were between 3.0 and 6.0 inches in body length,
but a few were smaller. Only one species, Loligo opalescens, was identi-
fied. However, many partly digested specimens, which could not be
specifically identified, were encountered.

This item was observed in all of the monthly samples, although it
was never of primary importance. Squid were taken most frequently
during the spring months, forming 20.3 percent of the April food,
19.1 percent of the May food, and 13.5 percent of the June food. While
sc^uid occurred in the stomachs of salmon captured inside the 20- and
even the 10-fathom curves, the majority had been eaten by salmon
taken beyond the 20-fathom curve.

Crab Megalops
More than 10,800 crab megalops occurred in 10.7 percent of the
stomachs and comprised 4.0 percent of the total bulk. This was the
smallest item frequently taken by the salmon. Crab megalops were not
consumed in such great numbers as were euphausiids. Rarely were there
more than 500 individuals per stomach.



2G4 CALIFORNIA I'lSlI AND GAME

],;i('k of descriptions of the ineti'alops stag-es of llic different crabs
occnrrinpf in Califoi-nia waters made it iiiip()ssil)le to specifically
identify the specimens found in the stomaclis. The relatively large size
(average length about 0.8 inches) and general appearance seemed to
indicate tliat tlie species was Cancer magister, one of the commonest
forms in the study area. However, more than one species may have
been eaten.

Crab megalops were taken onl,y from March to July, ami consump-
tion of this item was greatest in March (13.3 percent), April (15.0
percent), and May (9.5 percent). They were consumed by salmon cap-
tured inside and outside the 20-fathom curve.

Other Fishes

Seven and three-tenths percent of the total food volume was made
up of fishes other than those already mentioned.

Twenty unidentified clupeids occurred in a total of 17 stomaclis.
jMany, if not all, of these were doubtless Pacific herring, since no in-
tact specimen of any other elupeid was found.

In all, 45 smelt (family Osmeridae) appeared in 24 stomachs. Three
forms were identified — Allosmerus elongafus, Spirinchus sp., and Hypo-
mesiis pretiosus. Young and adults, ranging from 2.5 to 7.5 inches in
length, were eaten. Smelts of one species or another were encountered
in all the monthly samples except those of ^lay, June, and July, but
were not found in the stomachs of salmon known to have been cap-
tured outside the 20-fathom curve. Consumption was heaviest in March,
when 10.6 percent of the diet consisted of this kind of food.

Seven stomachs, collected during October and November, contained
the remains of 10 Pacific saury {Cololabis saira). It may be signifi-
cant that the only food present in the stomachs of the five salmon cap-
tured well outside the 20-fathom curve during October was one Pacific
saury.

Gadids, including the Pacific tomcod (Microgadus proximus), oc-
curred 10 times, only one per stomach, during February and March
and from August to November. They were found solely in the stomachs
of salmon captured inside the 20-fathom curve.

A total of 16 young sanddabs {Cithorichfhys sp.) and young un-
identified flatfishes were eaten by seven salmon captured outside the
20-fathom curve during April and May. The remains of one unidenti-
fied flatfish, estimated to be about eight inches long, were found in
the October sample.

During October two stomachs each were found to contain an ather-
inid, which was probably a jacksmelt {Aiherinopds calif ornien sis).
These were the largest food organisms seen (11.0 and 12.0 inches long),
and because of their size the^^ accounted for 4.6 percent of the October
material.

A total of five embiotocids, including shiner seaperch {Cymatogaster
aggregata), appeared in the March, April, August, and October diets.
All were found in the stomachs of salmon caught inside the 20-faThom
curve. One of the shiner seaperch was eaten by a salmon taken in San
Francisco Bay.

Young lingcod {Ophiodon elongafus), 34 specimens in all, were pres-
ent in 18 stomachs collected from March to June. They were all quite



KING SALMON FOOD HABITS 265

small in size (2.5 to 4.5 inches long). This item was most important in
May, when it made np 2.3 percent of the diet of that month. Almost
all the lingcod appeared in the stomachs of salmon from beyond the
20-fathom curve.

Young sculpins (family Cottidae) were represented in the diet by
at least two species — cabezon {Scorpaeniclithys marmorafus) and Irish-
lord {Hemilepidotus sp.). The former ranged in length from 2.0 to
3.0 inches, and the latter were from 1.0 to 1.5 inches long. Thirty-four
salmon containing 66 sculpins were taken outside the 20-fathom curve
during April, May, and June.

Other Invertebrates

Although a variety of other invertebrates was encountered, only 0.1
percent of the total bulk consisted of items in this category.

Remains of polychaetes were found in three stomachs collected dur-
ing April. Some were identified as heteronereids, which are the free-
swimming sexual stages of otherwise benthic polychaetes.

Three mysids were found in the stomach of a salmon captured dur-
ing April.

Adult decapod crustaceans were eaten occasionally. One prawn
{Pandalus sp.) was taken in February. Two shrimp (Crarjo francis-
corum) and one ghost shrimp (Callianassa sp.) appeared in the October
diet. The remains of one adult crab were found in the stomach of a
salmon captured in San Francisco Bay during April.

Unidentified crustacean remains were found only in March, April,
and May. These remains may have been composed largely of euphau-
siids and crab megalops, since these organisms were most heavily uti-
lized at that time of year.

Small octopi were represented seven times from May to August in
the stomachs of salmon captured outside the 20-fathom curve.

Differences in Food Habits According to Size

Chapman (193(i) compared the stomach contents of three size groups
of king salmon examined at Neah Bay, Washington. He found that tbe
smallest sizes, 11 to 20 inches in length, contained no sardines or her-
ring, but lived almost entirely on euphausiids. Those intermediate in
size, 21 to 30 inches in length, still ate euphausiids but also took sardines
and herring, in the ratio of two grams of the former to one of the
latter. The largest sizes, 31 to 50 inches in length, subsisted mainh^ on
fish, taking about three grams of sardines to one of herring to 0.3 of
euphausiids.

Milne (1955) found that one-half of the stomachs of king salmon
over 26 inches in total length contained herring, whereas only one-fifth
of the stomachs of smaller specimens had this type of food. (These her-
rings ranged from 5.9 to 7.5 inches in standard length.)

In the present study the food habits of the large and small salmon
were found to be generally similar, even in February and March, when
large Pacific herring formed a substantial part of the diet. Frequently
salmon less than 22 inches in fork length had taken large herring. Tii
one instance the stomach of a 19-inch salmon contained a herring no
less than eight inches long, and there were other eases of salmon having
swallowed food fishes more than one-third their own length. Only among



266



CALIFORNIA FISTT AXD P.AMK



TABLE 6

Percentage Composition, by Volume, and Percentage Frequency of Occurrence of the Food Items Consumed
by Different Sizes of Salmon Captured Near the Farallon Islands During April



l-'ood ileiii



Pacific herring {Clupea pallasi)

Unidentified clupeids

Rockfishes (Sebastodes spp.)

Lingcod (Ophiodon elongatus)

Cabe?on (Scorpaenichthys marrncratus)

Inshlord (Hemilepidotus sp.)

Unidentified cottids

Unidentified fish remains

Euphausiids

Crab megalops

Unidentified crustacean remains

Squid

Totals

Total volume of food in cubic centimeters
Number of stomachs



I,' - - I li.in 2.") iiiclics
(.fork length)



Volume



11.8

0.8
1.1



0.2
0.1
0.2

53.8
1.2
0.5

30.4



100.1



Frequency



12.5

20.8
4.2

12.5
4.2
16.7
62.5
25.0
4.2
29.2



25 inches or more
(fork length)



Volume



45.6
0.2
0.7
0.4



0.3

0.2



0.2
0.3

29.6
4.1
0.1

18.4



100.1



Frequency



24.2
3.0

15.2
3.0
3.0

15.2
6.1

24.2

69.7

15.2
3.0

18.2



576 . 5
24



914.6
33



the salmon captured near the Farallon Islands during April was a
marked difference in the food habits of large and small salmon noted.
These salmon were arbitrarily separated into Iavo size groups — those
less than 25 inches in fork length and those 25 or more inches in fork
length — and their food habits compared (Table 6). Although euphau-
siids appeared in roughly the same percentage of stomachs in both
size groups (62.5 percent of the small salmon; 69.7 percent of the large
salmon), they formed 53.8 percent of the food consumption of the
former group and 29.6 percent of the food consumption of the latter
group. Pacific herring comprised 11.8 percent and 45.6 percent, re-
spectively, of tlic food of the small and large salmon.

Differences in the Amounts of Food Eaten
Silliman (1941) noted a positive correlation between the troll catch
of king and silver salmon and the quantity of fish found in their
stomachs. However, in the case of silver salmon captured during
October and November he observed an exception, which he attributed
to the approaching sexual maturity of this species. Of this he had the
following to say :

In seeming con trn diction to the correlation, a large catch of silver salmon was
taken in October and November, when the fish content of stomachs was small.
This may have resulted from a condition not present during the other months of
the study, namely, the sexual maturity of the silver salmon. The species is a "fall
spawner" . . . and the fishing in October and November is largely upon schools
starting their spawning migration to the streams and rivers. Since the silver
salmon spawn closer to the mouths of the streams and rivers than the chinooks
the gonads of members of the former species preparing to make the upstream migra-
tion must be further developed than those of the latter.

Now it is known that spawning fish do not feed normally after they have entered
fresh water. That this influence extends also to fish nearing sexual maturity that
are still in the sea, seems to be borne out by the extremely small amounts of food
of any kind found in the stomachs of silver salmon collected in October and No-



KIXG SALMON FOOD HABITS



267



^•emb(>^ . . . Nevertheless, the fish still strike at fishermen's lures, as evidenced by
the lai^e catch. There apparently persists in the rine fish a reflex which causes
them ti> strike at objects resembling fish, even though they do not swallow them . . .

In view of Silliman's observations on silver salmon, an attempt was
made to determine whether the king salmon utilized in the present
study showed any indication that they had ceased to feed, or at least
had taken less food, prior to entering fresh water to spawn.

According to Fry and Hughes (1951), most of the salmon tagged off
central California turned out to be from the Sacramento-San Joaquin
River system. The California Department of Fish and Game statistical
reports show that the heaviest commercial salmon catches in the Sac-
ramento-San Joaquin River system are made during the month of
September. Therefore, it would seem likely that if the salmon cease to
feed prior to entering fresh water to spawn, those in the vicinity of
San Francisco would do so during September.

A comparison of the percentage of empty stomachs occurring in each
montlily sample (Table 2) revealed that while a large portion of the
stomachs of salmon taken in September were empty, even larger per-
centages of specimens captured in February and March liad no food
in their stomachs. However, when the quantity of food per unit of body
weight was calculated for each monthly sample (Table 7), the data re-
vealed that during September there were fewer cubic centimeters of
food per pound of salmon than in any other month.

The September specimens were then grouped according to their de-
gree of sexual maturity. By plotting the fork lengths of the salmon
against the weights of their preserved gonads it was possible to dis-
tinguish salmon about to spawn that fall from those not ready to spawn
for at least another year, because of the enormous development of the
gonads of the salmon in the former group. (Specimens with gonads in-
termediate in size were disregarded.) Among the individuals with large
gonads a distinction was made as to whether they had been caught in



TABLE 7
Cubic Centimeters of Food per Pound of Salmon, Acording to Month



Month



Number

of

specimens



Total

volume

of food

consumed

(in cc.)



Average

volume of

food per

stomach

(in cc.)



Average

weight

salmon

of

(in pounds)*



Cc. of

food

per

pound

of salmonf



February.-

March

April

May

June

July

.\ugust

September

October

November



61
89
93
71

119
89

109
89

187
97



486.6
1,095.4
1,974.2
1,343.8
3,138.9
1,582.1
1,378.9

744.7
2,270.3

798.6



8.0
12.3
21.2
18.9
26.4
17.8
12.7

8.4
12.1

8.2



4.5
5.3
7.0
9.0
9.9
8.2
7.5
8.3
5.3
4.5



1.8
2.3
3.0
2.1
2.7
2.2
1.7
1.0
2.3
1.8



* To obtain these figures the average fork lengtlLs, as given in Table 1, column 6, were first converted to average
dressed weights, by the use of Fry and Hughes' (1951) length-weight relationship. The true average weights
were then determined by adding 12.8 percent to the average dressed weight in accordance with the findings
of Fry (1952) concerning cleaning losses in king salmon.

t These figures were calculated by dividing the average volume of food per stomach by the corresponding average
weight of the salmon.



268



CALIFORNIA FISH AND GAME






■2 X





O


^




u


rJ










<l







e:


^




o


c


oo




■r


LU


i>n


' '


— 1


..


,


CO


o


,-'


■ej




"


\—














=


-^




o


^




o.


o










»






o.


r— 1




o


c5




£


03






<K



iS - *-!































-r M o




Tf ;C —




-^ ci d




^2S


o


65-S


o — d




«5






O










c3






02












"o


^




•B


2 c






i- *


,


t^


SJ &


cc tc «


O


■o 5


CO cc O


a


5 »


odd


:.
S






■o


^ O




O


m
















*~










=






c3






-f-


00 — •*




c


1^ C-. o




e;






(U


<6 '^ d




§




*3








« -jt




^ ^ ^




•r 5 X
























Mi.


t- I- —






i. s




> '.C-




i ^




-^




- _r
















> 5 = r;


— o rc






5:1:^-"^ =


"" ^ "^


t' ^ - '-^




c^ i'








->




w.












O ^
















_2 X o


O 00 CO










> g c


— C^J cc

-^ o




o


5 "^








5




r™




'/.




:_ —
































~ —


"■


2 2.




1






1 1












w) ; ;










•■N




hi


S S rf




3




.^




E


5 = V




- S >>


"o


S§J


(S


cS . .


bO


0. a a




" S &


CO


o a 63




■^ a a




M « «








552




















*;. O





O ^ J2






z-^*!:









a =



u o
= o






KING SALMON FOOD HABITS 269

the ocean or in San Francisco Bay, the latter being regarded as fur-
ther along in their sexual development, inasmuch as they had already
left the ocean and presumably were on their way to the spawning beds
in the Sacramento-San Joaquin River system, which empties into San
Francisco Bay. (No specimens with undeveloped gonads were taken in
the Bay during September.) The (quantity of food per unit of body
weight was computed for each of these three groups of salmon (Table
8). The lowest figure was obtained for maturing salmon captured in
San Francisco Bay and the highest figure for maturing salmon taken
in the ocean. It seems that the king salmon which were taken in San
Francisco Bay and which presumably were moving into the Sacra-
mento-San Joaquin River system to spawn had almost ceased feeding,
whereas the ocean-caught salmon which were about to spawn had not.

SUMMARY

1. The stomach contents of 1,004 troll-caught king salmon from the
vicinity of San Francisco were analyzed.

2. Six items constituted 92.5 percent of the total food consumption.


2 4 5 6 7 8 9

Online LibraryCalifornia. Dept. of Fish and GameCalifornia fish and game (Volume 43, no. 4) → online text (page 2 of 9)