REACTION OF TUNAS
AND OTHER FISHES
TO STIMULI

I Marine

Biological Laboratory!

APR 21 1C53
WOODS HOLE, MASS.

SPECIAL SCIENTIFIC REPORT: FISHERIES No. 91

UNITED STATES DEPARTMENT OF THE INTERIOR
FISH AND WILDLIFE SERVICE

Explanatory Note

The series embodies results of investif^ations, usually of restricted
scope, intended to aid or direct management or utilization practices and
as guides for administrative or legislative action. It is issued in
limited quantities for the official use of Federal, State or cooperating
agencies and in processed form for economy and to avoid delay in publica-
tion.

Washington, D. G.
November 19^2

United States Department of the Interior, Oscar Lo Chapmanj Secretary-
Fish and Wildlife Service^ Albert M, Day-j Director

REACTION OF TUNA AND OIHER FISH TO SnMJLI-1951

PART Is BACK (HOUND AND SUMMARY OF RESULTS, o o<. o «„o <, oby Albert Lo Tester
PART lit OBSERVAnONS ON THE GHEMORECEPTION OF TUNAo . oby Po B» van Weel

PART nil OBSERVATIONS ON THE REACTION OF TUNA TO

ARTIFICIAL LIGHTo o >.. o o . o o « o „.•.,. o o o .<> o -.o, .by Sidney C » Hsiao

PART IV s OBSmVATIONS ON SOUND PRODUCTION AND

RESPONSE IN TUNAo » o o » « » . o o » o o. „ <, , » <> » o o , • o . . « . oby Iwao Miyake

PART Vs NOTES ON THE RESPONSE OF A TROPICAL FISH
CKUHLIA SANDVICENSIS) to INTERRUPTED
DIRECT CUHRSNTc .,„..., o..- . = ..... o..«. o.. c by Albert Lo Tester

Special Scientific Reports Fisheries No, 91

Note,=-This report is also Contributions Nos„ 22-26^
of the Hawaii Marine Laboratory;, University of Hawaii

WASHINGTON s NOVEMBER 19^2

CONTENTS

Page

Part I s Background and summary of results, o c <. o o a <> . « . . <> <. o o « o 1

xnX'roQUcX'Xon jt, oo-ocs^^cci ..ocuwcoooitQoooootJooooo 0000000000* J.

i2w>p orxrnsn uax xxsno <.'■■- ^ - t. . . ^ooooooooooooooooooooooooooe ^

OUnUllciJ/j' 01 re SUX Tj S CoOOotvOyC&OOJOOO^dCOOOOOOOOO^OOOOOOOOOOO ^

rt" XorUilt^COocioococj 000 ooooocoooooouooooooc* 000000000000 0000000 J

Part lis Observations on the chemoreception of tuna,, . „ o « o o o oo o . . 8

Xii Ui vJQUO UX Olio OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO U

MaT«6rxarX a me o no qs anci uecnnx c[ue oooooooooooooooooooooooooooo o

rCeSUX uO ooooooooooouooooooo. 0*0000000000000000000000000000000 X^

X o i^ax uixsn waxier o oooooooooooooooconoo 0000000 000 0000 j-^

XX o xjaxvXx sn preparaT»xons 0000000 00000000000000 0000000 xi}.

XXX o Qtjuxci prepar'aTx ons fcooooooooooooooooooooooooooooo- xo

iVg Tuna flesh preparations o » o o o » o » o o » » „ o o . o » <. « » . « o » « 18

Vo Asparagxne and dol-=asparagine. o » o <. » » » o o » .> o o o <. o . . . 30

vX u uopper av,-© ua we ^000000^000000000000000000000000000 j^j

Discussion and Gonclusions 00000000000000000000000000^000000 31

itex er ences 0000000000 000000000000 00 00000 00000 000000000 00 00 ©• j?^

Part Ills Observations on the reaction of tuna to artificial

XX gn iL-.> 300000000000000 00000000000 00000000000000 OOOOOOO J^

Xntroctu ccxon coo j-o ^oooooooooooo 00000000000 <i 0000000000000000 ^o

Materx ax and metnous ooooooooooooooooooooouoooaooooooooooo«o ^ (

Xb3SUXT>S 00000000 n 000 oooooooooooooooooooooooooooo J OOOOOOO 0000 M-^

i„ Reactions to white and colored beams of light

from a carbon arc lan^) oo^oooooooocooooooooooo.o h3

iio Reactions to radiating light from bulbSooo o <. o o , <, . h^

iiio Reactions to radiating light from buIbs„o c o o „ » » . . hi
iVo Reactions to radiating light from white and

colored bulbs of low intensityc o o o » , o >. o o o » » » o » o u7
Vo Reactions to tiie horizontal beam of a projector
and to radiating white and colored light

from bulbs O OO0O0O0O00^OCiJO0UUUO1)0C>JOOOO0000000O PU

vio Reactions to white and colored beams of light

from a projector ooooooooo^ooooooooooooooooooooo pt

DiSCUSSiOno ooaOOOOOOOoooooooooc.oooooovoo<.oooooooooo»o>>oo.oo PU

Reaction:^' to continuous white lighto « o » o o <. oo oo ooo o o o o.. pU

Reactions to intermittent white lighto <, . « ^ ^ o o o o » « <. o o o o 5o

Reactions to continuous colored light c „ o .= o « o . o » , . o o » » » 5o

References ., , „ o o o o o o o o ^ . » .. , o ^ » o » - <> <> » « o o o o = » o u o o <> o o » o o « o , « » <, » 5o

fage

Part IVs Observations on sbimd production and response in tuna„ , 59

XiJ-^X^OCLU. O vXUn Bovvou OO'^oboOvOoOOOAOOOOOOOOOOOOGOOOOOOOOOOOi^O'- ^7

M3'T>6X^X^~L oXlQ 0.0^0.]/ 3>ULIo OOOOOOOOOOooOOOOOOOOuOOoOOOOoOOOOQOOO ^7

ooiinci proctUCwXon oy biiii3.o 0000000*0000 ooooooo© 00000000000000 ou

Jrl^O ». ■©QlUt© 000000000000000000000000000000000000000 oooo 00 DU

rCQ S uJ- wS OOOOOCOOOoOOOOOOOOOOOOOOOOOOOOOOOOOOOOoOOOOOOO-j DU'

O UiniTlSjr Jf OOOOOOOOOOOOOOOOOOOOOOOOOOOOnOOOOOOOOOOOOOOOOOo Oil

Response of tuna to sound stimuli ooooooo^ooooxooooooooooooo 63

X"^rO CQQ LUTS OOOOO OOOOjj OOOOOOOOUOOOOOOOOOOOOOOOOOOOOOOOOOCi D^

iV3SU.J- uS OOOOOOOOOOOOOOOOOOOOOOOOOOOHOOOOOOOOOOOOOOOOOOO O^

7=70 kilocycles per second,, o 00 o o o 00 » 0000 00 00 00000 6I4

500=5 :iOOO cycles per secondo » o o 000 000 000 ., o » 00 000 . 6U

100 and 200 cycles per second, 000 000 » o 000 »» 00 o uo » 61+

Q XjIUnlQjry OOOOOOOOOOOOOOOOOOOCOOOOOOOOOOOOOOOWOOOOOOOOOO o U

RS X Sr ©n CSoOO 0000000000000*000000000000000000000000000000000 '*-"-'

Part 1% Notes on the response of a tropical fish (Kuhlia

sandvicensis) to interrupted direct currento o « ., o = . » o » » 0 o « . o 69

Introduction o 000000»000000000000000000000000000000000000000 O^

Source voltage and electrode size 000. ...0000000000000000000 70

Reduction of "on=fraction" of a cycleoo o o o o » . . » 0 » 0 0 o 000 00 00 7u

Revolving disk interruptero 000000000000000000000000000 1

Plunging-electrode interruptero 00000000000000.00000000 ('5
Discussiono 000000000000000000000000000000000.0.00000000000.

References OOOOOOOOOOOOOOOOOOOOOOO.OOOOOOOOO., OOOOOOOOOO.OOOO

81
83

ILLUSIRAIIONS

Figure Page

lo Concrete tank in -v^iich yellowfin and tuimy 7/ere confined, oo 3

2» Plan of the concrete tank in which the tuna used for

experiments in chemoreception were kept„ ., . , . o o ,000000000 10

3o Graphs illustrating the reaction of one yellowfin to

baitfish water introduced at A., and at G^ and to seawater

Uo Graphs illustrating the reaction of tv/o yellowfin to

"clear" and- "murky" extracts of baitfisho o « o o » »» >, o o » » o 0 0 » » 15

5o Graphs illustrating the reaction of two yellowfin to

Preparation A (squid extract plus octopus ink) and of
two yellowfin and one tunny to Preparation B (squid

65 Graphs illustrating the reaction of two yellowfin and

five tunny to "whole" preparation of skipjack flesh,, 00000 19

7„ Graphs illustrating the reaction of two yellowfin and

five tunny to "clear" and "murky" portions of skipjack

X XSSil P rSpdJ/S. uXOn COOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO ^\J

80 Graphs illustrating the reaction of one yellowfin and five
tunny to "protein" and "fat" fractions of clear extract

of yellowfin flesh preparationo , , , , <> » <> o o o » o ,. o , » 0 o o o , . . » » o 0 22

9o Graphs illustrating the reaction of one yellowfin to
"protein" and "fat" fractions of clear extract of

yellowfin flesh preparationo co^ 00 0000.. 0,000000000000000000 2h

10 o Graphs illustrating the reaction of five tunny to "protein"
and "fat" fractions of clear extract of yellowfin flesh

Pir'6p3X'3l1il.OIlo oo»oooooooooooooooao 0000000000000000000000000 ^p

11 o Graphs illustrating the reaction of one yellowfin to

increasing concentrations (1 to h) of "whole" yellowfin

flesh preparationo 000,000,, 0,0,00000^,0^-0, ,00 « 00000,00000,0 <:o

12 o Graphs illustrating the reaction of five tunny to

increasing concentrations (1 to U) of "whole" yellowfin
flesh preparationo ,000,0,0,0000000000,0,00,0000,0,00000000

13, Graphs illustrating the reaction of on© yellowfin to

"whole" mar lin flesh preparationo ., , » o <, o o 0 0 , o o » , .» o o 0 , , o , , o

2?
29

Figure iMi

lUo Plan of the tank in which the tuna, used in experiments

with artificial lightj were kepto oo o _. o c ooo oo « oo oo o ooo oo 38

l5o Plan of instrument box housing light-producing and

optical ef^Uipmento OOO.-OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO ,57

l6o Transmission characteristics of Kodak Wratten filters^ooo hX

17 o Diagram of the tank (A) and pond (B) showing the
position of the transmitter and receiver used in
experiments with soundo oo oooooonooooooooooooooooooooooooo ol

l8o Circuit and apparatus used in 1he electrical experimentSoo 71

19 o The "revolving disk" mechanical interruptero oooooooooooooo 76

20o The "plunging electrode" mechanical inter rupter. oo «»

21 o Trend of average current (and current density) with
change in "on-fraction" of a cycleoooo

I O O O o o o

too OOOOOO O o ooooooooo

80
82

PART I

BACKGROUND, AND SUM^JARY OF RESULTsi/

by

Albert L, Tester
Professor of Zoology
University of Hawaii

INTRODUCTION

At the instigation of Co E, Sette, Director j, Pacific Oceanic
Fishery Investigations^ an agreement concerning a study of the reac-
tions of tuna to stimuli was completed between the United States
Department of the Interior^ Fish and Wildlife Service, and the Univer-
sity of Hawaii^ Honolulu j, To Ho^ on January 19 j 1951 (Contract No.
I6fw-1333l)„ The agreement called for "(a) the search of the litera-
ture on tuna reactions^ (b) the development of methods for holding
and caring for the various species in captivityj and (c) the study of
reaction of the fish, individually and in groups or schools, to stimu-
li such as lightj sound, chemicals and electricitys with emphasis on
the study of stimuli and reactions which promise to have application
to fishing operations,"

A sear^ch of the literature failed to reveal any papers dealing
directly with the reactions of tuna to stimuli. However^ many refer-
ences to the reactions of both freshwater and marine fish to stimuli
of various kinds were obtained. These are on file at the Department
of Zoology and Entomology 5, University of Hawaii o

Moderate success was encountered in establishing tuna and other
fish in captivity. This has been dealt vath m a separate publica-
tion (Tester 1952),

The papers which follow in this report deal wxth the congjletion
of the third part of the contract, insofar as tinie and fxmds were
available during the period of one yearo

We wish to express our sincere thanks to Oo Eo Sette, Director,
and to Dr, J, L, Kask and Dro W, F, Royce, Pacific Oceanic Fishery
Investigations, for their helpful suggestions and assistanceo We are
also indebted to Dr<, R, Wc Hiatt, Director^ Hawaii Marine Laboratory,
for general advice and assistance.

V Contribution No, 22 of the Hawaii Marine Laboratory^ University of
Hawaii

EXPERIMENTAL FISH

The experiments on chemical, light j and sound reactions were
conducted on two species of tuna? the yellowfin or ahi (Neothunnus
macropterus) and the little tunny (henceforth called "tunny") or
kawakawa (Euthynnus yaito), which were established in captivity in a
concrete tank at the Hawaii Marine Laboratory, Coconut Island^ Oahu^
over a 7-nionth periodo The experiments on electrical reaction were
performed on the "mountain bass" or aholehole (Kuhlia sandvicensis)^,
for reasons which /vvill be discussed elsewherso

The concrete tank in, which the tuna were confined (fig.. 1, repro-
duced, from figo ^s Tester 1952) is partially sunk m the ground^ it
has smooth 6-inch concrete walls and bottomj it is 3hc7 feet longj
10,8 feet widej and 3<,8 feet deep at the north end and UoO feet deep
at the south end^ its volume is 10 j 663 gallons;^ the rate of flow of
the saltwater supply is about 25 gallons per minute^ the inletj near
the northwest corner, is directed horizontally (towards the south)
at a depth of about 2^ feet^ the outlet is a notch cut in the top of
the south wall at its center o Baffles consisting of 3 x U-foot galva-
nized iron sheets^ painted white, are placed across three comers^ a
larger baffle, 3 x 10 feetj also painted white is placed across the
fourth or northeast comer for the purpose of housing equipment for
light reaction studies^ A ^=foot fence, with ipright posts and
horizontal, spaced plank bars, surrounds the tank at its upper edgeo
Towards the end of the summer the fence was lined with chicken wire
to keep the fish from jumping through, and the top was similarly
covered to keep visitors from throwing stones at the fisho Two 60-
watt bulbs were suspended above the tank and were lit from dusk to
daybreak o

During the experimental work varying numbers of tuna (from one
to seven) were present in the tank, as shown in tdole 1= As their
reactions to stimuli were dependent on their I'state of healths"
their history is briefly reviewedo

Yellowfin No, 1, the subject of most of the experiments, was
introduced to the tank on June 20^ 1951 .. started feeding on July 2<,
195ls and was in excellent condition until about October 31 j 1951 o
Following that date it took less and less food.^ and finally ceased
feedingo The normally shiny, black skin became whitish and dis-
tended, as if the body were swolleno The swelling around the eyes
made the eyeballs appear to sink within the sockets, and interfered
with the tuna*s vision » During late December and early January the
yellowfin resumed a desire to feed^ although it would snap at the
food, it would invariably misso It died on January 13,, 1952. and
in addition to being puffy and swollen, was found to have been blind
in one eyeo Its initial weight was estimated at about 5 pounds, its
weight at death was 11 pounds o

ik

FIG. I CONCRETE TANK IN WHICH YELLOWFIN AND TUNNY WERE CONFINED.

<H

li

0

<:

5

C

.0

rH

^

h

o

•\

OflS

«

o

rH

c^

rH -P

0

O,

«w

iH

in.

«^. G

rH

ft

.y^

*v^

60

^v. ©

rH

«

rH G

^v.

nD

^

'^■H

®

O

1 ^

cv

'^rH

l>»rH

rH

<;

iH

T)

H

0\ 41)

r-i

O OJ

c~

CO

>.<^.

<S>

^"^

O

"73 £

O

a S

\„

<D

^v^

•^g

c e

o 'k:

e.

taO

w

rH

V4

r^

CC

s a)

G 6

a

c

^\

*^^

e

^ M

(D

tH <k-(

•H

•H

J^

TJ

o

-P 0

b c

a ♦h

*-(

o

ta

T3I

<D

•H

)3 -H

«ri

? -p

i

0

P

o -c

C T3

3 «

^^

o o

H £3

•

«-i

Vl

•H

«

•rl

u ?i

-P 0)

O

rH C

-p

'O

-P

•O

tn fi

, %H

H 00

a) «

tj

•o

<D

ra

®

^

r-l C

>> o

o

0}

«>

<r>

0)

t:! to

0) P

■J-' °<^

■p

-p

-P

M

•<!•

M

® C rH

XI o

>,-p

o„. ^

«

u

H

fi,.H kn.

H^ a

ft

1) Jq

h

fi

«

TD

TJ

TJ

a Tj^^

o

c:», B

O

+3

P

©

c

©

3 ID r-f

T>

TJ ®

■^ ^

(3

to

«i

P

a

-P

"-^ <D cr\

oc, Q

0) +J

13 P>

5

U

u

V)^,

13 BJ

»H P

^ u

o

t-t

cv

a

■:;■.

n

rv o

0)

Tl »

>» iao

■^

r^

-p

X! r-i

«H SO

G G

3 ^ «r\i

c

o

o

(n

03

o m

"cl ra

>%^1

^

o

o

o.

O -H O

0)

G G

49

B

s

H

C\i

»r4

3 *H -P

3 =irl

G «
3 .

-p >/>,

O

c

c

0

o

o

G Shi C

^^^

bflrH C^

4S

°H

>H

o

o

o

•H (D O

.s^<

ca

•^

^

is;

s

a

<U Xi <»-t

+3 r^

^ tlO

tJ

Ai

S -P -P

0) G

,5 S '-^

«

T,

o

o

>s

>»

>»

O O <H

® rH

Si d

§

(H

C

rH rH C

£ o

■p ^

o

la's

•CJ

V

0)

<D

3

3

3

« rH O

si o

■§ rH

0) ft"H

p

BJ

H

tH

H

6->

t-«

!>« «3 O

E-t &5

CtJ «0 Tl

+3

a

G

x;

O

>.

fH
O

a

u
e

P G

5

XJ

(0

*H Vll

O P

c

© rH!
J3 «[

H

CV

Oi

Oi

»A

iTi

CNi

sH

^^

H

V

rH

rH

r-{

H

pH

rH

cv

t-\

■iC'

J^

m

fH

>n

^^

m

ir.

Xi

\.

»n,

H

'v.

vO

■\

=»i^

a

CNi

c-

CT^

-^

«>

<ri

"■^-s*^

.to

r^'

EH

■vi

^

■^

<

5'

<

^

C\J

<.

to

t30

00

O-

V^

H

rH

«H

[1

3

g

1]

O

rH

D

Ij

0

0

li

<-H .

t-\

r-{

rH

r-^

V\

H

rH

•O

y\

kr>.

-iC'

V\

rH

rH

■^N.

\r

»/\

O

\.

^v.

V^

m

C\i

" "v.

■■^^

«H

O

en

CO

O

\.

\

rH

E^

5^

^

o^

.fv

^

S'^

fn

ITk

^..

"\

'^,

©

">.

•\

"\

^^,

\,

\

!-i

OJ

CV

Ph

vO

to

to

Co""

Ov

o

r-\

rH

rH

Yellowfin NOo 2 was i^x..., odu^ea on August 22s 1951* started
feeding 3 days later, but Jumped from the tank and died on Septem-
ber ks 1951 » It we-; in excellent condition during its short life
in captivityo Its estimated weight was about 6 roo^ids.

The five tunny were introduced between August 27}i 1951 ^"^^o.
September 2, 1951 j and started feeding witiiin 1 to 3 days, t^ey
were in excellent condition until the end of October j fcilowing
which thej"- gradually became listless and fed only occasionally«
Three died on or about November 12, one i-n Dacamber 6, apd the
last one on December 8, l?>.o During tb.sir moribund condition they
lost their bright coloration but otherwise (apart from their beha-
"jlor) seemed normal in appaaran.ce,. Tb-e reason for the sickness and
subsequent mortality is unknown^ it followed a period of cold, wet
weather o During their period of confinement, the tunny increased
from an initial weight of about 2 pou3.ds to a final weight of 3 to
k pounds o

Except prior to experiments en chemoreception, the tuna ware fed
regularly once a dayo Notes on their feeding aR.d schooling beha^/ior
are included In the reports which folloWo

It was planned to study the reaction of tuna in large Pond Koo 5
(Tester 195'2) to light and sound stim^ili during J?;.nuary 1952. using
one yellowfin and one tunny. These were the sur-i-jors of a mortality
which occujred, as in the concrete tank, during Kcrembero Ihe sur%'-5~
vors were f eliding ,. and were apparently in good condition during
December and the first part of January. A few experiments with sound
were ner^ormed but, unfortunately, before the series could be complet,;:
the tunny disappeared (about January 22, 1952) and the yellowfin
died (about January 30, 1952), again following a period of cold, wet
weather. No experiments with light were conducted in Pond No. 5=

SUMI«IARY OF RESULTS

Dro P. Bo van Weel, studying cnenorecption in tuna, found uiat
both tl-i.e yellowfin and tunny have a well -developed sense of Suiell or
taste whereby they may be attracted to certain food substancea. Tr -^
were strongly attracted to clear, colorless extracts of tuna flesh «
Moreover, it was fcund that the attractant was contained in the
"protein" ratter than in the "fat" fraction of the clear extracts In
general,, the reactions of the . t-.mny were t^re pronounced than those of
the yellowfino Cn the other hand, "r.here was no positive reaction of
either species to "conditioned" water in rhich baitfish had been living,
nor to extracts of either baitfish or squido Tvtc chemicrl-s, otl';sr
than food substances, were tried — asparagine, a possible^ at-':.r£-jtant9
and copper acetate, a known shark repellent o ^ne fcrmsr did not
prove to be an attractant. The latter was a repellent to tuna, al-
thou'^h its effect was not as pronounced as on fish of other species

which were also present in the tank„ Future research should be
directed at identifying the particxLar "protein" substance in the
clear extract of tune, flesh which ^cts as the attractanto If this
can be isolated an . prepared in large quantities, it oould be used
in attempting to attract tuna to the stern of a fishing boat at sea.

Dr. Sidney Co Hsiao studied the reaction of the tuna ti -.rtificial
light generated from an arc larap;, a prcjsotion lantern^ and electric
light bulbs o liis experiments were performed after dark, wi r.h the
tank illuminated constantly by tirfo 60-watt bulbs o He found that
both yellcwfin and t'onny were attracted t,o continuous: white light
over a range of moderate intensity (about 70 to UBC foot candies )o
Hov/everj they were not attracted by a light of weaker intensity, and
tiiey were repelled by a light of stronger inter.oit;-o Both species
vrf-ere attracted to colored lights of moderate intensity, but to no
greater extent than to white light,, Similar results were obtained
with interrupted white light o There appeared to be no relationship
between the strength of the reaction ^zid the fre'^uercy c:l ini.9rr--i.'-
tion of the lighto It was noted that although tLa tana apprc-f-ched an
interrupted light of moderate intensity, they were repelled frcrr. tlie
near vicinity at the instant tlrie light flashed either on or ofl »
Future research might be directed profitably at determining the react-
ion of tuna to reflected light of different quantity and quality
originating from moving objects during daylight hours o

Professor Iwao liiyake attempted xo discover (l) if tuna proiuced
any sound, and (2) if they could be attracted or repelled by sounds
of various frequencies. Using a listening frequencj'' which ranged
from about 100 cycles to 70 kilocycles per second, he vras able to
identify low frequency sounds produced by the sudden movement of the
tail of the yellowfin in the tanko This might hve some significance
in respect to the mechanism of school forraationo No sounds produced
by the tuna at moderate, high, and supersonic frequencies were detect-
ed. In attempting to attract or repel tuna by continuous sound stimu-
li, sounds were produced at many freqiiencies within the IOC c^ole to
70 kilocycle ra-:ii3. No positive res^olts were obtainedo Ho^e-rer,
there were seve. .J: indications tJ-^^t ths "uuna might react pocitiveiy
to complex sou-ids :n* lov,- frequency. These, and also interrupts'
sound stimuli over the entire frequency range, might be invs.:- 'i-^T' ted
in the future,.

The writer attempted to extend the observations: made by Morgan
(1951) on the reacticn of the aholeh.^e to interrupted direct current
in a small wooden tank of s'jawiixero It ^f^s found t.hat by progressive
shortening of the on-fraction of ." cycle at h frequency of 15 cycles
per second the downyfai'd trend in average current necessary to attract
the fish, demonstrated by Morgan, was continued. Ths relaticnsni-
betv/een source voltage and electrode siae was also clarifiedo .n.ddi-
tional vj-ork on the reaction of this species in a small tank of yeawater

can be undertaken profitably in an attempt to determine the optimum
on-fraction and minimum c\irrent density for positive attraction o
However,, this would require a more satisfactory current interrupter
than that presently available „ After the optimum on-fraction has
been determined, the ej^jerimental work could be extended to include
a study of the reactions of tuna to interrupted di.rect current in a
large volume of seawater such as that of the concrete tank in which
the tuna were confinedo

REFERENCES

Morgan, Mo Eo

I95I0 The response of a tropical fish to interrupted direct cur-
rent and its application to the problems of electrofishing in
seawatero MoSo ThesiSj, University of Hawaiij June 19^8 68 ppo

Tester, Ac Lo

1952 o Establishing tuna and other pelagic fishes in ponds and
tankso UoSo Fish and Wildlife SerVo, Spo Scio ReptoS Fisheries
Noo 71g 20 pp.

PART II

OBSERVATIONS ON THE CHEMORECEPTION OF TUNAi/

by

Po Bo van Weel
Professor of Zoology
University of Hawaii

INmODUCTION

The ejqjeriments to be described in this paper were undertaken
during the summer and autumn of 19^1 at the Hawaii Marine Laboratoryo
The author undertook to determine v^hether tuna have a sense of taste
or smell whereby they might be attracted (or repelled) by food sub-
stances in suspension or solution o It was hoped to find some con5)on"
ent of the food of the fish which acted as an attractantj and which
might possibly be prepared in large quantity from some cheap sourcGo
If soj it might replace or supplement the live bait which is present-
ly needed for pole and line fishing (June 1951) and which is in short
supply in the Central Pacific area.

Originally it was not planned to investigate the reaction of
tuna to selected chendcalso However, two such substances were tried
at the suggestion of others g asparagine., a possible attractantj and
copper acetate s a known shark repellent o

I am very much indebted to DTo Ao Lo Tester for his critique and
kind help in preparing this paper,,

MATERIAL, IffilHODS AND TECHNI(|JE

The experiments were conducted on one or more of two yellowfin
(Neothunnus macropterus) and five little timny (Euthynnus yaito) which
were established in a concrete tank (figo I) at the Hawaii Marine
Laboratoryo Normally the tuna were fed a daily ration of tuna flesh
from skipjack (Katsuwonus pelamis), yellowfin^ or tunny, which they
accepted greedily,, They also accepted live baitfish (Pranesus insula-
rum), the heads of which had been pinched so that they floundered in
the water, Ihey were not observed to feed on a school of baitfish
Tirtiich was present in the tank, probably because they could not develop

£/ Contribution No„ 23 of the Hawaii Marine Laboratory, University
of Hawaii

siifficient speed in the corxfmed quarters to catch them. The tunas
accepted dead baitfishj, but only when very hungry^ and then_, with
apparent reluctance,, They also fedj with apparent reluctancsj on
marlin (Makaira laazara) flesh after the supply of tuna flesh had been
exhausted., The fish in the tank were not fed squid (used in the exper-
iments) although this is one of the food items which they will eat
when in their normal, habitat (Welsh 1950) o

The following substances were prepared and introduced to the tank
in a manner v\rhich will be described later s I— -baitfish watery 11—
baitfish preparations 5, III— squid preparations . IV-=-tuna flesh prepa-
rationS;j V—aoparagine and dol^asparagine soluticris,, and VI— =copper
acetate solutions o

The baitfish water consisted of 3 liters of standing seawater in
which 50 baitfish had been living for 3 hours,. This "conditioned"
water vras used as a test substance o

The baitfish, squid, and tuna flesh preparations were all made
in a similar manner o A quantity (to be reported under each experi-
ment) of the substance was quickly mashed m a blender and the mash
was extracted m the refrigerator for 3 hours with twice its weight
of distilled watero This "whole" preparation was used in some exper-
iments „ In others, the preparation was first fractioned into "clear"
and "murky" extracts before being usedj the preparation was centri-
fuged and the supernatant, conparatively clear fltiid was diluted to
3 liters with seawater (the clear extract) s the remaining debris was
suspended in 3 liters of seawater (the murky extract) 0 Variations
of or extensions to the above procedure are described under indivi-
dual experiments o The baitfish were freshly caught? the squid were
purchased from the fish market m a frozen conditiCii and thawed
just before using^' the tuna flesh, from skipjack and yellowfin^, was
in some cases from freshly caught and in other cases from frozen
fisho Marlin flesh, from frozen fish,, was also used in a few experi-
ments included under IVo

The asparagine and copper acetate were pure chemicals which were
dissolved in seawater m various concentrations as indicated in the
individual experiments o

The above substances were introduced to the tank for the most
part on the west side at Point A (figo 2)o To eliminate any reactions
based on hearingj they were introduced by means of a siphon from a
height of about 1 foot, with the rubber tube inlet about 6 inches
below the water surface a

Quantitative raeasiirement of the reaction of the fish posed a
difficult problemo Normally the tuna would cruise leisurely around
the tank in one direction for a long timeo At first it was thought

Hinos

ir>-i

lu

u

«J-

U-

1

*o-

rw-

UJ

-J

<

— -

o

(/]

o-*

t-
z

UJ

S

(E
UJ
Q.
X
UJ

cr
o

a

UJ

v>

3

UJ

X

X

o

X

5

z

UJ

I-

UJ

a:
o

z
o
o

UJ

X

u.
o

UJ

cr

UJ

Q.
UJ
O
UJ
(T
O

< UJ
_J X

5: <~>

CVJ

HIHON

10

that an attractive substance would lure the fish to the rubber inlet
and hold them there,, but this proved to be the case only with highly-
attractive substances o With more weakly attractive substances there
appeared tc' be an increase m cruising speed and a tendency to circle
closer to the inlet o Accordingly., the reactions were measured in two
ways; (l) by determining the time required for the fish to make 10
passes across an imaginar-y line (AC in fig, 2) drawn across the tank
from the inlet, witli the fish travelling in either direction (when
the fish is circling normally this vd.ll be equivalent to the time of
5 complete revolutions) ^ and (2) by counting the number of passes
(out of 10) made across the line AB in the neer or west half of the
tank (whea the fish is circling normally there ild be 5 passes
across AB and 5 across BC)o 'Therefore j perception of the stimulus
would be indicated by a decrease in the time of 10 passes from that
of normalj representing an increase in cruising speed and/or a de-
crease in size of the svamming circleo Attraction would be indica-
ted by an increase beyond 5 in the number of passes across ABj
representing one or more circles completely within the near (west)
half of the tank,. In one case, as indicated laterj the substance
was siphoned in at C and the number of passes across BC was countedo

It was assuraedj and later established; that the reactions would
vary with the state of hunger of the fisho To minimize variation
from this factor j the tuna were starved for 2ii hours before the start
of an experiments However j, the state of hunger induced a factor of
alertness which could easily result in grave error ss at the approach
of an observer the fish became excited, expecting food, and circled
close to the observer at increased speed for a considerable periodo
Tests were not started until the fish became accustomed to the pre-
sence of the observer,, and a "normal" cniising speed was resumedo As
the so-called "normal" cruising speed varied from day to day, it was
necessary tc establish its value before each experiments Itiis was
done by repeating the timing and counting of the passes until approxi-
mately constant values were obtained, vdth a half -minute interval
between successive tests,. The substance would then be introduced and
the timing test would be continued, still with half -minute intervals
between thera_, until the normal values Mere again approximated o

As for a long time only one yellowfin was present., many of the
results were obtained on this one fisho liVhen more tunas were intro-
duced the schooling instinct became apparent. When tv/o yellowfin and
five tunny were present, the two species tended to school separately
and to exhibit different reaction patterns » After one yellowfin died,
the other joined the tunny, and although a slower swimmer, it attenpt-
ed to keep up with the schoolo It showed, therefore, an increased
speed of reaction as compar-ed with the results obtained when it was
the sole resident of the tank. After four of the five tunny had
died, the single tunny schooled with the single, larger yellowfin

11

(which took the lead) and its reactions were slower than that of
the tunny school „ These differences in behavior make it difficult
to draw general conclusions as to the relative strength of the react-
ion in the two species o

Tll[hen the fish schooled, only one member of the school was tiiuedo
■When both yellcwfm and tunny were present j the timing tests were con-
ducted alternately on each species o

Finally it ^3hould be mentioned that the tuna tended to favor the
shady side of the tank, eogo> the east wall during the hours before
noono Before this was realized, many "positive" results were erron-
eously recorded in morning experiraentSj with ohe substance siphoned in
at C (figo 2)0 To avoid this difficulty, most of the experiments were
conducted between 11 aomo and 12s 30 pomo, with t he substance siphoned
in at Ao

RESULTS

The results ar'e shown gr-aphically in figSo 3 to 13^ The upper
panels show variation m the time in seconds required for 10 passes
in either direction across the line AC (figo 2)o The ordinate scale
has been reversed in direction so that increase in height of the
plotted points indicates increase in cruising speed and/or decrease
in the size of the swimming circles, ioeo, perception of the stimuluSo
In the acco\mt which follows, the word "cruising speed" has been used
to cover the complex behavior pattern measured by the time of 10 pass-
es,, The lower panels of figures 3-13 show variation in the number of
passes across the line AB| an increase beyond 5 shows that the fish
describes one or more complete circles in the near (west) half of the
tank, ioeo, it indicates attraction to the point of stimulation;,

io Baitfish water

Five experiments were performed with similai' resuitSo Only one
yellowfin was presento The results of one experiment are recorded in
figure 3o

During the preliminary timing^, the yellowfin' s cruising speed
gradually decreased to an approximately constant valueo When the
baitfish, or "conditioned," water was siphoned in at A> there was no
apparent change in either cruising speed cr in the number of passes
across AB- Similarly, when the baitfish water was siphoned in at C^
there was no apparent change in cruising speed or in the number of
passes across CC The experiment does not demonstrate the false
"positive" attraction which may result from the tuna's preference for
the shady side of the tank, although it was shown in other preliminary

12

o

> :

\

O

z

s S

o

UJ
CO

C\J

UJ

co

in

S3SSVd 01 JO 3yMl

S3SSVd JO

a3akNnN

q:

UJ

X
CO

<
m

2 O

UJ

>-

UJ

z
o

<

UJ

q:

UJ

X

o

<
cr

I-

3

X
Q.
<

q:

u.

O

cc

UJ

I

UJ

Q

z
<

o

z
<

o
to W

o

ID

o

o

13

experiments o It may be concluded trom. this and the four other
experiments that baitfish water has no apparent attractive or re-
pellent effect o

Figure 3 also shows the results of a control experiment in
which seawater was siphoned in at Ao Again^ there was no notice-
able or measurable reaction of significanceo

iio Baitfish preparations

In several experiments j, one of two schooling yellowfin was
timed in preliminary tests j, on introduction of the clear extract
of the baitfish preparationj and on introduction of the marky
extract of the baitfish preparationo The results of only one experi-
ment are recorded in figure h» in which 1^1 grams of baitfish (wet
weight) was usedo

On introduction of the clear extract^ there appeared to be an
increase in cruising speed and an increase in the number of passes g
particularly in Tests 10 and lit, The mean decrease in time for 10
passes was from 112 o 3 seconds for the preliminary tests to 960 3
seconds for the tests with the clear extracts The difference is not
statistically significant,,^

On introduction of the murky extract there was at first a slight
increase and then a considerable decrease in cruising speedo The
initial increase might be construed as a reaction to the substance $,
but not necessarily as an attraction^ as the nmaber of passes across
AB did not increase.,

"Hie other experiments yi&lded even less evidence of attraction
to the baitfish preparations o It may be concluded that this sub-
staace has either very slight attractive properties or none at alio

3/ A sin5)le test of the significance of the difference between the
~ mean times is not necessarily informative in experiments of this
nature as (i.) during the preliminary timing there may be a gradual
decrease in cruising speed as the fish become accustomed to the
* presence of the observer ^ and (h) during siphoning j there may be
at first no change, then an increase,, and finally a decrease in
cruising speed (and number of passes; during the times in irtiich
the substance is entering ^he tankg spreading over a small vol-
ume of sea water near the inlet, and gradually dispersing in
smaller concentration farther and farther from the iaJeto Un-
fortunately the data are not sufficiently extensive for more
detailed statistical analysis even if a procedure could be designed
to handle this con?)lex situationo

Hi

CM

~ LU
O
2
UJ

5 8

UJ
(A

o

z
<

'o:

<
ui

i

o

S

<

UJ X

(— <

'S .

.2 fe

§

<
q:

= X

UJ
t/5
X :

9; >
< 5C

cc a:
o o

s

sassvd 01 do 3WII

S3SSVd JO
b39WnN

15

iiio Squid preparations

Ebcperiments with squid preparations^, either whole or divided
into clear and murky portions., yielded no evidence of the presence of
attractive substances even when comparatively concentrated prepara-
tions were usedo

As the squid is a normal food of the tunas and as it will expel
ink when pursued^ the possibility exists that the combined stimuli of
sight and smell might evoke a response whereas either one, acting
alone, would not. To test this possibility., Preparation A was made in
the following manner o Firesh (thawed) squid C20J4. grams) was extracted
without centrifuging, and to this was added the ink of a freshly
speared octopus (Polypus mar^moratus ) ^ The material^, when siphoned
into the tankj formed a dar-kish cloud which was visible from abovoo
Two schooling yellowfin were present, one of which was timedo The
results are shown in figure 5o

In preliminary Tests 1 to ?;, the tuna were attracted by the pre-
sence of the observer but settled dovm to normal activity during
Tests 8 to 13o Ihile introducing the preparation^ there was an in-
crease in cruising speed during Tests 16 and 17o This, howevery can-
not be regarded as a reaction to the substance as the behavior was not
repeated in a second experimento Moreover, the number of passes
across AB remained constant at 5 throughout Tests lU to 21 _« indicat-
ing no attractiono

Although the cloud formed by Preparation A was plainly visible,
it was not particularly dark when the preparation became diluted in
the tanko As the walls of the tank were also dark, there resmained
the possibility that the contrast was not sufficiently sharpo
Accordingly, Preparation B was made in a similar manner (3^5 grams
of squid) except that India ink was added in such quantity as to pro-=
duce a ''pitch" black cloud in the tanko Two yellowfin and one tunny
were ^^.^sent, all schooling togethero The results are included in
figure 5o

As indicated in the preliminary tests, for' some unknown reason
the fish were restless and cruised around at a relatively high speedo
When Preparation B was siphoned into the tank,, the tuna avoided the
black cloudg remaining close to the south wall (figo 2), and not
crossing the line^ ACo The doud diffused across the tank and gradual-
ly drifted towards the south wallo As it apprca':Aedj the yellowfin
became more snd more excited, and finally darted through it., there-
after remaining in the up=stream portion of the tank until practically
all of the cloud had dispersed through the overflow„ Ihe tunny showed
an entirely different behavior., appearing to be undisturbed by the
presence of the black clcudo It cruised into and cut of the cloudj,
maintaining an approximately constant speedo

16

'y

I r

to

>:

K

1

to

^

n

P

O

O

W

3

_l

0.

1-

—

o

V

<

01

2

X

<

iLl

a

3

(D

O

->

tn

.J

^

a.

<

t-

z^

ey

o

h-

h-

X

<

LU

cr

<

Q

S:

-)

LlJ

a:

o

o

LjJ

K

OJ

O

z

J,

z

Ll.

o

LU

1-

3

Q

<

o

_l

IC

LlJ

LU

Si

</)

>

LU

q:

Q

CL

1-

o

to

(-

UJ

Ll.

t-

O

7-

2

z

O

_)

(-

1-

o

<

lU

S3SSVd 01 dO 3WII

S3SSVd dO

a38i(NnN

cc o

z u.

^1
cr _i
I- _i

1/5 lij

=) >-

X
CL
<
CE
O

o

5

17

It may be concluded ...e squid preparation ei trier of

itself or in combination -witu squid ink or India ink had no attract-
ive effect on tlie tuna,, " ■ "''"j, vrhan concentratedj appeared to
have a repellent effect ■-:. , olloi'tfinj but thlsvas probably a

visual reaction. It is imiikeiy that tlie repellent effect could
have been produced by sone chemical in the India ink as the carbon
particle^ are usually stabilized in suspension by gelatin, aid gelatin
has no effect on tlie behavior of the fish as shown by a. control
experiment ■'•rith a loO percent solutions * ''

iv.

During most of the experiments 'iirLth tuna flesh preparationsj
two yellowfin and five tunny were present in the tankj- forming trro
distinct schools vdth markedly different crioisins speeds,, Only one
individual (not necessarilj the same fish throughout the experiment)
of each school vfas timedo The timing tests on each species v/ere
conducted alternately o

A. In the e^qperiment portrayed in fi.giia'e 6 the ".vhols extract
of skipjack flesh (200 grams) was used^ vfithout centrifugingo The
tunny, in particular, '.vere attracted by the obser^^er^s presence, and
did not slow down, and extend their course ever the whole tank for
a considerable period of tiraeo '•ftiey appeared to have an exciting
effect on the yelloiTfin, the cruising speed of wtdch increased
during the preliminary trialso '^Yliile the preparation was being
siphoned into the tankj ttiere v^as no reaction during the first fev/
seconds until the tunny naared the inleto Then^ however, the turjiy
started railling around the inlet excitedly« ?he yellowfin, also^
soon showed this same reactiono One of the latter actually snaoped
at the rubber inlet. All fish took shreds and small pieces of the
extracted flesh into tJieir mouths, but invariably spat them out
again„ Apparently this material, after extractionj is tasteless
and -unpalatable to the fish, even though they normally were fed
txna flesh vmile in captivityo Kovrever, other fish in the tank,
(a Duffer, Tetrodcn hispidus (?) and a few manini , (Acanthurus
sandvi censi ij" di d eat the extracted materislo Although it
required only 3 to 5 JRxnutes to siphcn the 3 liter preparation ^
the general excitement of tlie tuna lasted for seme 10 tc 1^ minutes o
The timing (Tests 17 to 2$ for yellcmin and 1,8 to 26 for tunny)
was then resumed as liie fish grew calmer and started cruising
again.

Bo A series of experiments -.fas next conducted in -which the
tuna flesh preparation vfas fractioned into clear' and rnxn-ky portions „
As these all gave similar results, only one is recorded in detail
in figure 7, in v;hich 130 grams of skipjack flesh was usedo

During the preliminary tests, the speed of the tunny school
j^radually decreased, vriiereas tliat of the yellowfin remained fairly
constant, V^ien th« clear extract was introduced, there was a difference
in behavior of the two species. The tu:nny started milling around the

10

c>

'N\\N Sl-Ol

- 00

o

X

s

o

t-

z
z

3

>-
z
z

13

>-
<

Z

LU

a.

UJ

a

z
<

5
o

_l

o

■NIW Sl-Ol
ONHIIW

O

I-

LU

' Q.

I

t

>-

a:
<

UJ

a.

o

O

< X

tt: lij

_i

LU U.

X

o

er trt

I-

Crt u.

=) o

CO

X
Q.
<

a:
o

<
q:

UJ
X

a.

S3SSVd 01 dO 3Wli

S3SSVd dO

d3awnN

19

r

-i r

I r

5

ONnniiM

or
<

_i
o

<

q:

CL

3

<
LiJ

O

<

LU
CL

_ o

o

z

UJ
- CJ D

J

8

I L

S

<>■ en

CO

_ f^

S3SSVd 01 JO '3Wli

O in o

S3SSVd JO

<

CO

o

in
z
o

h-
cc
o
a.

>-

IT

3

O

z
<

_o
o

z

Q

Z
<

o

o

5

z
o

^

tr g

UJ

CO Q:

I Q-

CL

< X

a: in

C3 lU

C3

20

inlet after about 30 seconds > They became greatly excited and often
bit into the rubber tube. Some 10 minutes after siphoning started^
the excitement subsided and some 5 minutes later they resumed their
cruisingc The yellovrfin were markedly excited and were definitely
attracted to the inlet 5 iflhich they snapped at occasionally. However j
they did not mill around the inlet^ bat rather increased their cruis-
ing speed once they sensed the clear extract „ It should be emphasized
that this clear extract was invisible in the tanko

After h^ minutes, during which period the fish became calm and
resumed their normal cruising speed^^ the murky portion was introducedo
In the experiment recorded in figure Tj the tunny's speed was high
during the first minute of siphoning but decreased thereafter o Fr-om
this and other experiments it was concluded that the murky portion
had no apparent effect on the cruising speed of either the yellowfin
or the tunr^o Both species^ however^ were attracted by the whitish
color of the murky extract and by the shreds of flesh contained in it.
They snapped at the shreds but did not swallcvj them This attract-
ion is indicated by the increased number of passes across AB in Tests
36 to Uo„

Co One experiment using clear and murky extracts of skipjack
flesh (120 grams) was performed after the fish had been recently fedo
Neither the cruising speed nor the number cf passes gave any indica-
tion of a positive attraction = This was expected from observations
of their feeding activity in the tanko On thromng food to -^hem, at
first they take it greedily ^ milling around at increased speed,, soon
they r'eact more slowly as their hunger is satisfied^ finally they
ignore tlie foodo

Do Since it v/as established that the clear extract contains the
attractive factor(s)j an attempt was made to determine whether this
was contained in the "fat" (petrol ether soluble) or "protein"
(residual) partso The fat fraction was obtained by shaking the clear
extract with petrol ether and separating the latter fiom the residue
or protein fraction,. After evaporation of the petrol ether in the
refrigerator (where the protein fraction was also stored during the
period of evaporation) the fat extract was suspended in the same
amount of distilled water as that of the protein fraction,. Both frac-
tions v/ere diluted with seawater to 3 liters 0 In the experiments
which are discussed below^, one yellowfin and five tunny were present.
All fish tended to school together, with the yellovrfin trailing o

In one experiments the results of which are shonrn in figure 8,
yellowfin flesh (200 grams) was used.^ yielding 350 cubic centimeters
of the clear portion. This was shaken with 75 cubic centimeters of
petrol ether o During the preliminary trials^, the fish appeared to be
h\xngry, and were excited by the presence of the observer » The protein
fraction was siphoned in firsto D^iring the first 2 minutes there was

21

o

z

UJ

o

UJ

en

(O

CO

z
g

I-
o
<
q:

Q

Z
<

0.

z
z

3

UJ

>

o

z
<

z
u.

o

_J
_J

UJ

>■

UJ

z
o

I-
o
<

<

IT
UJ

q:
a.

< UJ

tr _i

(- u.

lO

3 z

s

(/) o

I _j

Q. _J

< UJ

tr >-

03

6

S3SSVd 01 dO 3mi

S3SS\fcl JO

22

no reactiorij, but then the fish swam rapidly in small circles
around the inlet „ Ihe reaction j particularly of the tunny, was
clearly positive, although not as violent as iivith the normal
clear extract o Tlie yellowfin behaved similarly to the tunny j
but might have been influenced by their behavior. None of the
fish snapped at the xnleto After normal cruising of the tank was
resumed, the fat fraction was introduced;, Ihere was no detectable
reactions

A second experiment illustrates the behavior somewhat better j,
as the reaction was somewhat less pronounced and the timing tests
could be continued throughouto In this<, yellowfin flesh (180 grams)
was again used_, yielding 225 cubic centimeters of the clear solution
which was shaken with $0 cubic centimeters of petrol ethero Both
species schooled together^ mth the yellowfin trailing. The results
are shown in figures 9 and 10 o As soon as the protein fraction was
introduced, the fish were attracted and cruised in small circles at
the inlet o The yellowfin "sniffed" at the tube during Tests l5 and
19s ^^^ "the tunny both "sniffed" and snapped at the inlet during
Test 18 0 After waiting 2$ minutes, the fat fraction was siphoned
ino Unfortunately when siphoning was started^, two onlookers
appeared whose presence attracted and excited the fish (Tests 33 to
14.0)0 Ullhen they went away^ the normal cruising speed of the fish was
resumedo Except for the disturbing presence of the onlookers^ the
results are similar tc those m the previous experimento It appears
that the attractivB substance is located in the protein rather than
the fat fraction of the clear extract of tuna fLesho

Eo A series of experiments was next performed to determine the
extent to which the tuna flesh preparation could be diluted and still
retain its quality of attractiono In these the whole preparation was
used J but it was made up in various dilutions o It should be kept in
mind that the concentrations given are those siphoned into the tank^
and not those in the tanko "he concentration in the tank would be
difficult to determine as it does not remain constant because of
diffusion,,

In the first experiments one yellowfin and five tunny were
present schooling togethero The whole preparation of yellowfin flesh
(156 grams ^ yielding 300 cubic centimeters of whole extract) was
divided into the following quantities and each was diluted to 3 liters
with seawaters (l) 10;, (2) 50-, (3) 90o and (ii) 150 cubic centimeterso
The results are shown in figures 11 and 12 0 When (1) was siphoned in,
there was an increase in cruising speed but no definite attraction to
the inlet o When (2) was siphoned in, there appeared to be a definite
attraction as both the yellowfin and tunny swam in small circles near
the inlet during the siphoning processo TShen (3) was siphoned in,
the reaction was similar, but for some unknown reason it was less
pronounced than in (2)o lilfhen (h) was siphoned in^ the reaction soon
became very pronounced., both species milling around near the inlet
at high speed for about 5 minutes »

23

o

<

O (-
? <

Q- a:

O IJJ
Q.

5
o

LlI

>-

5
o

< <

UJ cc

X

UJ LU

X

I-
3

to

X
Q.
<

o

<

_l
o

o

CO

<
u.

S3SSVd 01 dO 3Wli

S3SSVd JO
H3ekNnN

24

I r

o

L

ai
o

z

UJ

O
UJ

en

I-
5 UJ

- «o

sassvd 01 do 3WI1

O "' o

S3SSVd JO

aaawnN

a

z
<

z

LlI

a.

O

P s

I

O LlI

^

<

o

ir

I-

13

X z

< p

o ^
a:

o

25

ONIIIIW

>:

o

2
liJ

o

in

o

a:

I-

UJ

o
o

O

<

q:
o

o

>-

F :'

Q

<

^ k ^

-h — h — ^ — * — ^

o6 OT o — f^

S3SSVd 01 JO 3Wli

sassvd JO

2 2

tr

UJ

cc

Q.

X
CO

<
tr

o

_i -1

— Ul

>-
in
X =

Q- UJ

< Ij
cc o

5

26

}':

I

i

o

CO

o

<

IT

o

z
o
o

o

z

<

UJ
(T
O

g

>

o <

LlJ

a:

q:

Q.

UJ

— UJ

>-

CO

X :

Q. UJ

< -J

a: O

o X

cvi

S^SSWd 01 dO 3Wli

SBSSVd AO

27

It was observed in thisj, as in many other experiments^ that
before the substance was introduced^ the two species formed one
compact schoolj, with the yellowfin trailing, but when an attract-
ive substance was introduced, and the fish became excited^ each
species reacted at different cruising speeds^ thus breaking up the
compact school formationo

In a second experiment,, one yellowfin and two tunny were pres-
ent, the yellowfin leading the school « The whole preparation of
yellowfin flesh (150 grams ^ yielding 300 cubic centimeters of ^^hole
extract) was divided into the fcllowing quantities^ each of which
was diluted to 3 liters with seawaters (1) Sg (2) 10 j (3) 50, and
CU) 23$ cubic centimeters „ Three of the tunny had &ed and the
remaining two seemed listless when this experiment, waa performedo
Consequently J, strong reactions were not expectedo The results
were similar to those of the previous experiment although the
reactions were not as pronounced o There was an indication that
ths tunny sensed the most dilute (1) solution in that they often
hesitated at the inlet but then quickly increased their speed so
that they caught up with the yellowfin,, Both the tunny and the
yellowfin made more passes across AB than BC^ indicating attract-
ion,, With (2) J the stimulus was apparently strong enough to cause
the tunny tc leave the yellowfin and react independently o Again j
an increase in cruising time and number of passes for both species
indicated attraction. With (3), the tunny frequently remained near
the inletj and orOy occasionally joined the cruising yellowfino
Again attraction was indicated in both species. With (k) both
yellowfin and tunny showed a pronounced increase in cruising speed
and in number of passes across AC However the fish did not mill
around the inlet as in the previous experiment^ even when the re-
mainder of the whole preparation was poured into the tanko

Fo The objection might be raised that attraction to the txma
flesh prepai'ation occurs because the t-una were conditioned to this
kind of foodj and that a similar reaction would not necessarily be
obtained with "wild" fisho This possibility cannot be denied^ al-
though it is shown in the experiment to follow that the reaction
was also obtained with marlin flesh. The only survivor^ the yellow-
fin, had not yet been fed this material so it coul.d not have been
conditioned to ito

Unfortunately, the yellowfin was in goor condition and did not
show pronounced reactions to its food. Accordingly it was considered
that even a weak reaction would be evidence of an attractive sub-
stance. Several experiments were performed vvith similar results o
Only one is recorded in figure 13 in which 260 grams of marlin flesh
was used as a whole preparation.

28

-co

o

Z
UJ

o

LU
CO

UJ

8 d^ o _ _

S3SSVd 01 dO 3Wli

o

X

5

o

o

z
g

I-

UJ

f 2

2

cc

<

2 UJ

-J X
-I tf)

"" UJ

in uJ

^ ?

<
■ s

S3SSVd JO
d3QWnN

29

During preliminary Tests 1 to ±0 the cruising speed was erratic
and the fish often approached the observer^ apparently expecting
foodo The cruising speed became more regular during Tests 11 to l6o
The marlin flesh preparation was then siphoned into the tanko There
was an initial increase m cruising speedy followed by a decrease.,
during which time (Trials 18 and 19) the yellcwfm hesitated near the
inlet apparently sensing the substance. Following thiSj the cruising
speed increased againj and then gradually decreased^ The number of
passes across AB also increased during the siphoning period, indicat-
ing attraction. From these, and similar results in other experiments^
it is concluded that the yellowfin, even though in poor condition^ was
attracted by the mar'lin flesh preparationo

V, Asparagine and dol-Asparagine

One yellowfin and five tunny were present in the tank during
several experiments with these chemicals o Solutions ranging from
Ool to loO percent vv-ere usedo As no reaction? whatever were observed^
the results are not included here. It may be concluded that these
substances were not noticed by the tunao

vi. Copper acetate

The question arose as to whether copper acetate might be used to
repel sharks without repelling tuna during long-line fishing opera-
tionso It was suggested that some information on this point mi^t be
obtained by determining the reaction of tuna in captivity to this
che.Tiical, While it was not possible to compare the reactions of the
tuna to that of sharks^, it was possible to compare them with the react-
ions of a few manini and baitfish which were present in the tank.

The following concentrations of copper acetate were used: Ool>
0,2^ 0„5s and loO percenta Fcr each expsrimentj 3 liters of these
solutions were siphoned into the tank at A, at Cj or at both A aid C
together i, with the solutions flowing down-streamo A part of the tank
was thus kept clear of the solutiono Stronger concentrations were
not used for fear of injuring cr killing the tunao liVhen siphoned in-
to the tankj the solutions vfere clearly visible from above as a bluish
cloud.. The actual concentration of copper acetate in this cloud was
not determined; it changed rapidly as the cloud diffused,,

The behavior of the tuna was similar at all concentrations of the
solution and the reactions differed only in degree » The first fish to
exhibit reactions were the manini , They swam i;5)-stream and remained
near the seawater inlet for the duration of each experiment. The
baitfish also avoided the down-stream part of the tank into which the
solution was being siphoned,, In generalj the yellowfin and tunny
cruised around the cloud of copper acetate solution but did not enter
ito As the o'loud diffused^ both species kept to the up-stream part

30

of the tanko After the cloud had diffused until it was barely visi-
ble, the tuna eventually entered ito With the stronger solutions
it took a proportionately longer period of time for the tuna to re-
turn to the down-stream part of the tanko Apparently there is a
critical concentration above which there is a repellent effect,,
This J howe%'"er5 is not too marked as in one experiment (the only one
with this reaction) both the yellowfm aiid the tunny swam right
through the cloud (Oel percent, or weakest solution) from the start
of siphoningo Despite this instance, it may be concluded that cop-
per acetate solution has a repellent effect on tuniy although its
action is not as pronounced as in the case of manini and baitfisho

DISCUSSION AND CONCLUSIONS

As the tuna are predacious fish, it might be expected that the
most important sense involved in feeding would be the eyesight, fol-
lowed perhaps by hearing o However other sense 3 ^ as for example the
chemical sense in its widest meaning^ cannot be excluded on an a
priori basis o The experiments reported in this paper show that~a
chemical sense is present, and this in&cates that it may play some
part in feedingo It is impossible to say whether the chemical sense
is smell or taste or^ in other words, whether sense organs in the
mouth or the nose are being stimulatedo However j, both senses are
usually well-developed in fish as other investigators of fresh-
water and marine fishes have shown (Adrian and Ludwig I9385 Berghe
1929j Copeland 1912, von Frisch 19l|l> Greene 1925, gEz 19hl, Hasler
and V/isby 19U9| Huttel 19kli Klenk 1930| Naurath 19U9; Par-ker 1910.
1911, 1913, 1922| Scharrer, Staith, and Palay 19k7j Sheldon 1911?
Strick 192i+, 1925| Trudel 1929, Walker and Hasler 19ii9^ Wrede 1932) c

The experiments of g8z (I9i4.1) and Wred© (1932) showed that the
skin of fish secretes a substance which can be perceived chemically
not only by fish of the same species but also by fish of other spe-
cies, and that the substance could be recognized by smell o It was
thought that the same might be true for tuna^, ioCo, that they could
smell the presence of other fishes^ However, this was not the case,
as shown by the negative eaqjeriments with "conditioned" water in
which baitfish had been livingo

Von Frisch (19Ul) found that the skin of injured minnows
(Phoxinus laevis) gives off odoriferous substances, probably purin-
or pterin-like which cause alarm reactions in the same and related
species o It was thought that an alarm or repellent stimulus to the
prey might be an attractant stimulus to tte predator,, ioeo, that
baitfish or squid preparations might attract the tunao The experi-
ments reported here brought quite unexpected results to lights the
tuna reacted not at all, or only weakly,, to extracts of squid (a nor-
mal foodO and of baitfish (a food which they will take when thrown

31

from a iisliing boat), whereas they did react quite -strongly to ex-
tracts of tuna flesh. This attraction was based on chemical stimu-
lation rather than on sighto as proven by experiments with the clear,
centrifuged portion of the tuna flesh preparation in which the invisi-
ble fluid attracted the fisho 'fhe tuna were attracted to the debris
by sight, but they did not accept the particles as foodj, even when
taken into the mouths It was shown that an extract equivalent to 5
grams of tuna ilesh m J liters of seawater^ siphoned into the tankj
was sensed by the tunny j, and that in one experiment^ at least^ an
extract equivalent to 25 grams of tuna flesh in 3 liters of seawater
was positively attractive to both yellowfin and tumiyo Extracts of
marlin flesh also gave positive reactions^ showing that the response
in the tuna flesh experiments was not conditioned by the fcodo

Experiments with "fat" and "protein" fractions of the aqueous
extract of tuna flesh showed that the attractive substance was in the
lattero This is in contradiction to the results of Allison and Cole
Cl93U) who found that fatty acids had an effect on both freshwater
and marine fisheSo It must be left for future research to determine
what part of the protein-containing fraction is the actual attract-
anto Von Frisch's (19U1) experiments might be recalled in this con-
nectiont, suggesting that such substances might be purm- or pterin-
likeo

The tuna displayed no positive reaction to either asparagine or
dol-asparagine solutions o

Experiments with copper acetate, a well-known shark repellent
(V/hitley and Payne 19hl) showed that this substance has a repellent
action on tunas o However, they were not as sensitive to this chemi-
cal as other fish (manini and baitfish) which were also present in
the tanko Only a few experiments with relatively weak solutions
were conducted because of danger of harming the tunao

It should be emphasized that the experiments were conducted with
tuna in captivity^ rather than in their normal habitat, and that the
reactions in the latter might be different „ It should also be empha-
sized that when both yellowfin and tunny are present in the tank,
there is an interaction which affects both cruising speed and school-
ing pattemo For this reason, caution must be exercised in comparing
the intensity of the reactions of the two specieso In general,, how-
ever, the reactions of the tunny seemed to be considerably more pro-
nounced than those of the yellowfin, indicating a greater sensitivity
to the attractive substances o

32

REFERENCES

Alii son J J. Be and Wo H„ Cole

1.934« Stimulaticn by hydrochloric and fatty acids in Trssh water
and by fatty acids, mineral acids 5 and the sodium salts of
mineral acids in seawater. JouTo Geno Physiol. 17§ S03-6l6o

Adrian^ E„ Do and Co Ludwig

19^0 Nervous discharges of olfactory organso JouTo Physiolo
9ki UUI-I46O.

Berghe,, Lo van den,

1929, Observations sur 1 'olfaction et sur le raechanisme des

3 8 ^

c-ODirants olfactifs chez quelques Teleosteenso Acado Royale
Scio Belgiqu8o Clo Scio Bullo 1$% 2T8-3C5o

Copelandj Mo

1912, The olfactory reactions of the puffer or ^-Jirellfishj,
Sphercides maculatuSo Jouto Exper, Zool„ 1?^ 363.

Frischj Ko von

19l4.1<. Die Bedeutujjg des Geruchsinns in Leben der Fische.
Naturwisso 29s 321=333o

19l;l IJber einen Schrsckstoff der "Fischhaut urid seine biclogische
Bedeutungo Zeitschro verglo Physiolo 29 1 Ii6-lU5o

Greene., Co ]'L

19250 Notes en the olfactory and other physiological re^.f'tions
of the Californian hagflsh. Science 6I3 68=70.

Gozj, Ho ,j

19i;l. Ubei" den Art-= und Individuellgeruoh bei Fischeno Zeitschro
verglo PI" 3iol. 29s l-u5o

Haslarj, Ao Do asid Yfo Jo TiYisby

19i49o Use of fish for the olfactory assay of pollutants (phenclr^)
in wa'cer, Amer. Fisho Soc, Trans, 79s 6I4.-7C0

Huttelj, Ro
■• 19lilo Die chemische Untersuchung des Schreckstoffes aus Elritzenhau''- „
Naturwisso 29 § 333=331'. o

Klenko F„

t! IB

I93O0 Untersuchungen uber die Gesohmackswirkung "''•on Sauren bei
Fischen mit einem Vergleich am Menscheno Zeitschro -verglo
Physiolo 13s 3^9-3960

33

June, Fo Co

19^1o Preliminary fisheries survey of the Hawaiian-Line Islands
areae Part III - The live-bait skipjack fishery of the
Hawaiian Islandso Uo S„ Fish & Wildlife Serv„ Commo Fish„
Rev„ 13(2 )s 1-17

Parker, G,, Ho

1910 o Olfactory reactions in fishes » Jour, Expo Zoolo 10 g 535

I9II0 The olfactory reaction of the common killifish, Fundulus
heteroclitiSo JouTo Exper^ Zoolo lis 1-5 ^

1915 o The directive influence of the sense of smell in the dog-
fisho Bullo Uo So BuTc Fish„ 33(1913) t 6I-68.

1922., Smell, taste J, and allied senses in the Vertebrates o
Jo Po Lippincott, Philadelphia „

and Ro E» Sheldon o 1913 <> Sense of smell in fishes o Bullo Uo So
BuTo FLsho 32, (1912) s 35-i;6.

Scharrer, Eo, So Wo Smith, and S„ Lo Palayo

19lt7o Chemical sense and taste in the fishes Prionotus and
Trichogastero Jour, Compo Physiol „ 86: 15 3- 19 ST"

Sheldon^; R. Eo

I9II0 The sense of smell in Selachians » JouTo Expero Zoolo
lis 51 o

1911 o The reactions of the dogfish to chemical stimuli » JouTc
Compo Neurol o 19 s 273-311-.

Strieck, Fo

192iio Untersuchungen liber den Geruch- und Geschmacksinn der
Elritze (Phoxinu-s laevis Ao)o Arch, Mecklenburg » Nato
Rostock., "H I^/o

1925o Untersuchungen liber den Geruchsinn und Geschmacksinn
der Elritze (Phoxinus laevis Ae)o Zeitschro vergl, Physiolo
2s I22-I5U0

Trudel, Pc Jo

1929o Untersuchungen liber Geschmacksreaktionen- der Fische auf
slisse Stoffeo Zeitschro vergl, Physiolo 10s 367-1^09.

Walker, T, J, and Ao D, Haslero

19U9o Detection and discrimination of odors of aquatic plants
by Hyborhynchus notatus (Raf,)o Physiolo Zoolo 22 § U5-63o

3U

Welsh, J, Po

19$0o A trolling survey of Hawaiian waters » Hawaii (Terr,)
Bdo Comm, Agric, and Foro, PLVc of Fish and Gamej SpeCo
Bullo 2, 30 pp.

Whitley, Go P= and G„ H: Payne

19h7o Testing a shark repellent Australiar Zool lis 151-1^7

Wrede, Wo Co

1932 „ Versuche uber den Artduft der ELritzen > Zeitschr^ verglc
Physiol o 17s 510-519

35

PART III
OBSERVATIONS ON THE REACTION OF mi A ID ARTIFICIAL LIGHlV

by-
Sidney Co Hsiao
Associate Professor of Zoology-
University of Hawaii

INTRODUCTION

The jj'esponse of fish to visual stimuli -was studied as early as
1880 by Kuhne and Sewall, Since that early date^ a great deal of
information has been accumulated on the structure and function of
the piscian -visual organ^ and on the question of color vision in fishj
a subject of discussion and dispute amoung physiologists o The liter-
ature on color vision in fishes has been critically reviewed by
Warner (1931) » Wall's (19U2) monograph dealing with the adaptive
radiation of the vertebrate eye brings the literature review up to
19l;lo No further re-view has appeared over the past 10 years.
Nearly all the experiments on piscian vision^ both achromatic and
chromatiCj were doie on favorable laboratory specimens „ which were
hardy and of suitable size for indoor tanks „ As far as can be
ascertained^, no experimental work on the physiology of vision in
tuna has been reported^ none is included in a recently published
bibliography on the biology of the Pacific species (Shimada 19^1) »
Field observations made by amateurs and professional fishermen have
been accumulated for some time^ but conclusions based on them are
badly in need of verificationo

It is the purpose of these studies to discover the pattern of
reaction of tuna, established in captivity,, to different quantities
and qualities of light stimuli » The former is concerned with tb&
intensity and duration of light stimulation, and the latter with the
frequencies of the light usedj that is^ the different portions of
the -visible spectrum -Jrhich are involvedo The success of Tester
(19^2) in capturing several species of Pacific tuna^ transporting
them to shore, and keeping them alive in the confined space of an
outdoor pond and tankj makes it possible to experiment with these
oceanic species under controlled conditionSo It was hoped that
attraction or repulsion stimuli would be discovered which might have
some use in explaining the behavior of tuna in their natural habitat,
and perhaps in suggesting new or improved methods of captureo

The work was undertaken during the summer of 19^1 at the Hawaii
Marine Laboratory under the general direction of DTo Ao Lo Tester,
University of Hawaiij, whose assistance iff gratefully acknowledgedo

^/"contribution Noo 2U of the Hawaii Marine Laboratory, University
~ of Hawaii J Honoluluj To Ho

36

MATERIAL AND ME'IHODS

The fish used in these studies were one or two yellowfin
(Heothunnus macropterus) and five little tunny (Buthyiinus yaito)^
v,'hich were confined in a concrete tank (Tester 1952 )o

A general plan of the tank is shown in figure li^o In this Aj
Be, C, and D are four metal baffles, painted v/hite| the longer baffle
D houses optical equipment Eo The seawater inlet is indicated by
G and its exit by H, while L and U show the positions of two 60-watt
electric liglit bulbs which were lit each night from dusk tc dawn<>
The dotted line NO indicates the position of a cord placed across the
tank about 1 foot above the water surface to mark off the northern
one-quarter of the tanko It was fastened to the slat railing which
surrcunded the tanko The feeding station is shovm by Fo

The instrument box E^ shcrm in detail in figure l^j is designed
to carry a source of light^ a set of light filters , and shutters
with variable speedy submerged so as to send a beam of light of
desired frequency horizontally under the surface of the water from
one end of the tank to the other^ and to prevent seawater from coming
into contact with the instrumentSo The box is 24 inches long, 18
inches wide, and 2k inches deep., made of galvanized iron sheets sol-
dered at the joints^ and held rigid by a wooden frameo At first a
carbon arc lamp with a series of optical lenses was used as the
source of lighto However^ the inconvenience of having to change
the carbon pencil every half-hour or so led to its replacement by
a projection lantern,, The lantern (L in the diagram) consists of a
^00-watt Mazda incandescent lamp placed in front of a concave mirror
and behind a series of optical lenses which concentrate the light
into a narrow beam of about if inches in diameter at the glass win-
dow Wo The lantern is connected to the 110- volt, 60 cycle power
line through a powerstat (variable transformer) which has the follow-
ing specifications; output range^, 0-135 volts^ maximum amperes^ 7o5i
output KVAj 1;: frequency, 50/60 cycle A metal frame 1 inch larger
all around than the 3 x U inch glass plate used as a window Vif is sol-
dered on the outside of "the box over the 3~inch square opening tc
hold the glass in place, vfhile a patented caulking compound forced
into the space between glass and frame successfully makes the window
water tight. On its way to the window the light beam passes through
a shutter S which is a circular disc of aluminum carrying four open-
ings, each ij Inches in diameter and placed equidistally one in each
quadrant. The axis of the disc is mounted agairst a friction disc
F connected with a 3~inch pulley P, By changing the point of contact
between ths friction wheel and a fixed ring on the axis of the shut-
ter the speed of rotation of the shutter can be changed at will from
no motion/ when the point of contact is at the center of the friction
wheel, to maximum speed, when the contact is at its rimo By select-
ive use of either one, or two, or all four openings m the shutter,
the rate of interruption of light can be increased Uuree-fold, A

37

Hinos

1-

lO-

UJ

*-

1

lO-

CJ-

<

LU

_,

Ld

o

</)

o-l

o

irn

Q.

O

1
1

1
1

1

CO

1

0-1

(/)

H-
Z
Ui

s
a:

UJ
Q.
X

UJ

o

(O

3

.

<

Z

3

1-

1-

cn

LU

UJ

X

3f

1-
X

o

1-

X

QL

5

UJ

z

UJ

a:

ic

UI

z

^

<

p

K

LU

X

X

O

1-

_J

o

-I
<

2

CJ

<

ul

_J

)

Q.

t-
q:
<

5;

O

u.

HldON

38

FIG. 15. PLAN OF INSTRUMENT BOX HOUSING LIGHT- PRODUCING
AND OPTICAL EQUIPMENT.

39

secondary shutter E, placed across the lens system of the project-
ion lamp, can be manipulated by hand to produce variable and
irregular flickeringo The pulley P is driven by a 1750 RPM, J-HP
motor through a series of reducing pulleys and V-belts not shown
in the diagramo In front of the shutter Sj behind window Wj a
Masonite light block M and 3 Kodak light filters A^, B^ and C are
mounted in such a way that each can be easily pushed into the
path of the light beamo

The three Kodak filters are listed as "Series VII Wratten
filter A, ffl^) B., i58j and 05, j^7s" respectively„ Their trans-
mission characteristics were examined with a Beckman Model B
spectrophotometer with the results shown in figure 16, The fflS
red filter transmits a band of light with wave lengths between
about ^90 and 700 mju j the $xl blue filter between about 36C and
and 530 m/( , with maximum transmission at 14+0 m^ , and the #58
green filter between about l|.70 and 6l5 m/U j with maximum trans-
mission at 525 myu „5/

The intensity of illumination was controlled by use of the
powerstat when using the lantern, or by changing electric bulbs
from one wattage to anotner by methods to be described belowo
As lights of different ftrequencies have different intensity and
different penetrating power through water the intensity was
measured in situ by means of a photoelectric cell and an attached
microammeter;r The microariimeter was calibrated under standard
conditions 2! in terms of foot-candles and the calibration curve
was used in determining the intensities of the different lights
used as stimuli »

A piece of heavy canvas was dr-aped over the instrument box
E to protect the apparatus against the weather and to serve as a
blind for the experimenter. An opening about 2 inches square was
cut in the canvas as an observation "peek hole,"

To contrast the -effect of a single beam of light with that of
night lights used in certain commercial fisheries^ incandescent
electric light bulbs were adapted for submergence under water and
for carrying a frame for glass filters of different colors » To
produce different quantities of light for stimulation,, UO-^, 60-j
IOO-5 and 200-watt bulbs were used, while to produce intermittent
light a practically noiseless mercury switch was introduced into
the line. Commercial colored electric bulbs (similar to those

V The Kodak catalog lists this filter as transmitting two portions

from the spectral bands a major portion between UBO and 6L1.O ma

and a minor one between 670 and ?00 m/ui , with maximum transmis-
sion at 520 m/U

zJ It is a pleasure to thank Mr. R.. Oberdorfer of the Physics Depart-
ment for his help in the calibrationo

i;0

-v^

:^^

10-

20-

30-

^ 40h

in

<

a:

H

50h

60-

1

X /

/
^ ■

I

/

I

V

440

X'

525

70-

80-

^ A r

NO. 47 FILTER
NO. 58 FILTER

NO. 25 FILTER

901-

400

500

FREQUENCY IN m/J

600

700

FIG. 16 TRANSMISSION CHARACTERISTICS OF KODAK WRATTEN FILTERS

41

used for decoration) were also employedo In all experiments^
the bulbs were hung on the tank side of baffle D (figo Ih) and
were practically or completely submerged o

Preliminary studies showed that during daylight hours ohe
artificial light^, in contrast to sunlight,^ was too weak to
elicit definite response from the fisho Furthermorej even at
late afternoons the reactions were irregular because of various
distiirbances about the tanko Thereforej all the experiments
were performed after dusk or at night and at times when the
tide was sufficiently low t- insure clear water in the tank for
conparable light penetration.

As the tuna fed regularly from the western side of the
middle of the southern half of the tank (at F in figo lii)s all
light stimuli were applied from the northern edge of the tank
and only when a tuna entered into a specifically selected area
or field in the northern one-quarter of the tank would its
reaction be recordedo 'Ibis field is defined by the following
methods by applying one's eye to the observation opening in
the canvas described above, the observer can bring the top
edge of this 2-inch square hole to coincide with the twine NO
(figo li;)> and the middle point of its right side with the
shadow of the electric light Lo With a fixed line and fixed
point so selected^, a square shown by the area PQRS in figure
lU is determinedo This area vms used as the field of observa=
tiono Since this area is immediately in front of the source
of light and away from the feeding site, and traversed by the
light beam^ entrance by tuna into this area as they travel
toward the light was taken as a criterion of reactiono Rec .rds
were made of the number of times during an experimental period
that the tuna swam into this fieldj and also of the pattern
of movemento An experimental period^, the time during which a
particular stimulus was applied ^^ was arbitrarily fixed at 5
minutes »

Before starting a series of experiments;, preliminary obser-
vations were made of the behavior of the fisho liVhen more than
one was present they tended to form a school and cruLsed slowly
about the tank, occasionally passing between tlie observation
area and the baffle D (figo lU)s but more often circling in the
southern three-quarters of the tanko Only occasionally during
a period of about ^ minutes would tVey enter the observation
areas the number of entrances was usually Oj, occasionally Ij,
and more rarely 2 or 3o This was taken as control condition,
and care was taken that this condition was present before the
start of each experiment, although the number of entrances was
not always recorded nor included in the tables which follow.
At times the fish were unduly disturbed by extraneous factors

U2

such as the presence of visitors or an urge to chase a school
of baitfish wrdch was present in the tanko Experiments which
were attempted under these conditions gave erratic results and
were rejected o

All expieriments were performed while the tank was illumi-
nated by two 60-watt bulbs (L and M in figo lii) which were
suspended about 6 feet above the water's siirface . These lights
were not extinguished during the experiments for fear of fright-
ening the fish and causing them to ram the walls » They served
to illuminate the tank sufficiently for observations to be
made of the behavior of the fish in waiting for the establish-
ment of control conditions and during the experiments o

RESULTS

Several series of experiments were conducted between
August 20 and September 18, 195l> each on a different night
and therefore each under slightly different environmental con-
ditions „ They are discussed below in the order in which they
were performed c

Series i— Reactions to white?/and colored beams of light from
a carbon arc Ijaj^ "" — — _

The results of experiments with & continuous horizontal
beam from a carbon arc lamp are sho^vn in table 2o The estimated
intensity of the beam, using white light, was about k50 foot
candles (the apparatus for measuring the intensity was not avail=
able at the tine of the experiments)., Only one yellowfin was
present in the tanko

After control conditions were established^ the Masonite
shutter was slid from the window as quietly as possible and the
beam was allowed to penetrate the tank for 5 minutes » During
this period, the tuna entered the field llj times m Experiment
1 and 18 times in Experiment 2o The pattern of movement was
fairly regularo The fish swam across the beam of light at the
south end of the tank, then turned around = swimming parallel to
but outside of the beam towards the source „ After entering the
field at the north end of the tankj it again turned around,
passing through the light beam with its one side and eye illu-
minated, Tnis is considered to be a definite tropistic react-
iono At times, attraction was also indicated when the fishj on
leaving the fields described a small circle in the northern half
of the tank, and re-entered the field, l/ilhen the light was
turned off, the swimming pattern was maintained for but a minute

1/ The word "White" is used in the sense that no light filters
"ere employed to regulate Iha emittincr light beam's freauency,

43

-p

-a

•^

Si

e

>
O

s

O

a>

■*^
+»

03
O.

CO
©

CO

o

CO

o

(H
•rH
S

c

•p
-p

OS

m

d)

4*

-p

C!

a

<D

a

a

«rl

®

a

>

o

fV

a

-p

^

Oi

o

U

=H

c

^

u

€t

a

+5

»H

43

CB

(G

CU

fl)

O

fM

a

CO

as

•5

-p

bo

^

(D

O <B TJ C

f^ c <D e

0) CO »H

5 c o cl

S «D P "HI:

to

-*
rH

cn

H
o

+3

-P

13

00

u

a

V4

u

u

^

u

O o

<D

(D

©

<a

Pu

-p

P

+3

■p

to a

iH

H

iH

iH

C CO

•H

•H

»H

?H

O cM

<^A

<M

<M

«iH

+5 o

JS

X!

Si

^

O U

-p

P

-P

■p

-P

m CO

■g

•rt

•g

^

11

0

Pi c
1 o

:3

ti

o

O

o

o

O

0 ^

IH

u

u

u

^

»4

■hT,

CO

CO

ta

<a

<a

CO

CQ O

c

fl

a

c

c

3

<D

s

o

o

o

o

O

°ri a

^

■P

^

X)

JQ

U O

u

^

tu

u

»H

8^

W U

as

(0

«

a

<0

CO

CO «H

0
0

o

o

o

c:>

o

o

o

0

o

rH

0)

c

•P

<D

-p

l-l

CVi

c-^

-4-

w>

•^

•s

&

a

§

E-<

w

a

i

O

u

or sOo ITius, in the contrxjl condition, listed in table 2, the
fish entered the field 3 times diiring the first 2 minutes, but
not thereafter.

In Experiments 3 to 6^ colored filters were used. With the
red filter, the tuna entered the field 13? lU, and 17 times in
three separate experiments o With the green filter j, the tuna
entered tlie field 13 times in one experimento Under control
conditions, as before, there were only 3 entrances « Ihe pattern
of movement was similar to that with white light o

Apparently both white and colored continuous light from
the horizontally projected arc-light beam caused a similar
tropistic movements Unfortunately, experiments with a blue fil^
ter were interrupted by noise emanating from nearby buildingso

Series ii— Reactions to radiating light from bulbs

The results of experiments with continuous and interrupted
radiating light from a plain 200-watt (2l5 foot candles) and a
painted 200-watt (106 foot candles) electric light bulb are
shovm in table 3» "H^e painted bulb, with a slight orange hue,
was used because it happened to have been equipped with a water-
proof socket, it had been used previously in night lighting for
fish larvae,, Tv\ro yellowfin and five tunny were present.

Experiments 1 to 5 were performed with continuous light
from the two bulbs. In Experiments 1 and 2, which were made at
dusk (6sitO and 6s5'5 Ponio) there was no positive reaction, the
tuna entering the field fewer times during the experiment than
during control conditions, Ihe negative results may have been
due to the fact that it was not quite dark when the experiments
were made, A half-hour later, positive tropism was shown in
Experiments 3 and 1; with lit and 12 entrances into the field as
compared with h under control conditions <> At times the yellow-
fin entered the iield by themselves^ at other times they joined
the school of tunny, which entered as a unito In Experiment ^g
with the plain bulbj there were only 6 entrances, but this may
have been due to interference caused by the approach of on=
lookers to the south wall of the tank towards the end of the
5-minute test period. It is concluded that, after dark^ the
tuna show tropistic response to radiating light from bulbs with
intensities of 21^ and 10 6 foot candles.

In Experiments 6 to 10, the plain and painted bulbs were
interrupted at a rate between 60 and 75 times per minute, with
the light and dark period of the same durations The number of
entrances with the painted bulb (11, llj and 7) was greater
than with the plain bulb (0 and U) . The latter did not differ
appreciably from control conditions. However, observations

h5

•p
bO

•p

•H
•O

c
•p

CS
•H
•O

CO

4>
■P

§•

u

(D

-P
C

13

G

O

r3

a
■p

o

+5

O

in
C
o

•H
■P

o
a

to

(D
t

I

«

O

i-l

(0
6h

I

^ m r

o <D -a ci

c 01 a

03 «H.

■•^ :

c o c

a> -p -h!

o

o 0)!

mh p;

^ p s

o P. fl

-p ^j

0} O Ml

« p ©I

c o^

•H !

c
o

r-\
<D

«

c

s

p

0)

0)

c
o

c
o

c

•H

o

iH

(D O

O Q

C X!

CO O

f-i m
p

c >»

(1> c

(D p3

C! P

o

C\i

to

o

o

Si

o
'Ji

TO

o
c

p
c

w

■a

K

a

>»
x>

tX)

■O

-O

0)

d)
O
C

CO (H
^1 O

P o
c ja

(0

CV

Xi

^

■p

o

3

o

s

n

bo
S

g

i

h

o

«n

tn

93

fH

X5

s=^

4ri

©

•P

O

c
a;

i

-g

O

J3

p

c

O

CG

M

a

TS

®

0)

a

"S

g

XI

^

tH

<>

J

CO

>»

^

>»

o

r~i

H

M

•P

p

P

ft

m

<c

^

a.

O

O

(0

6

>»

S

(D

«e

s

-,

M

o

aj

0)

i>

<'

C(

JsJ

c

Tl

e

o

<a

s

P

CO

o

u

n

j3

u

iH

p

u

tSO

-p

c

a

0

^

d

o

w

H

rH

»

a

4)

-p

I

p

ID
?:

a

3

>»

rH
C
O

«

c

OS

u

+>

vO

>» 'H

V CD
^ 1 ©
ffl P rH

lA

sO

[

IT.

vO

C O T3'

rH

O

,]

rH

O

(DOC

CNJ

rH

CNJ

t-i

-P «H OJ

O^ O

rH

sO

«"-i

vO

sO

o

H

o

O

t~i

rv

rH

rH

0

9

0

0

i

0

0

s

o

o

<n

lo

0

0

0

1

n

B

0

c-

E^

o

E^

x>

X>

X>

^

XS

Xi

•3

x>

■3

x>

^3

J3

■3

•3

X»

■3

^

rH

3

XJ

Xt

v^

X>

■^

9

^

p
■p

X)

p
p

■°

J3

P

p
p

J3

p

s

-P

u

•p

«

P

S

53

•P

p"

%

+5

p

^

P

^

?

-P

S

n

o

0

i

3

o

0

11

1

0

o

0

o

6

f

(31

0

o

0

o

0

o

o

0

0

•H

o

Oi

o

<\J

o

cv

o

OJ

CNi

0

iJ

o

o

o

o

0

Oi

cv

CSi

Oi

©

C\l

c

•P

c

P

a

-P

a

P

p

c

«iH

C

•H

C

Ti

n

•H

S

Q

•H

<a

•H

a

•H

^

=H

«B

<H

«H

<0

rH

CO

rH

00

«

H

a

«8

rH

pL.

Oi

PL4

C^

Pn

PL4

Pl.

PL<

Pk

Pk

0

3 O

.H

rH

tH c

P

o

Fh

1-4

^

(D P

1 rH

CM

p

C^

-<t

4^

«~>.

vO

!>

to

0

0

S'?

t

O

o

H

u e

'

o

o

^6

of the behavior of the fish showed that with the plain bulb
the fish were attracted to the source, but with each abr'jpt
change between light and darkness, they were frightened
away if near the source, e„g,j near the south edge of the
fieldo Cn turning away from the light source, they swam
faster than usual and broke the school formation » Thu% the
tuna appeared to be attracted to lights of 106 and 2l5-can-
dle power when interrupted about once per second^ but were
repelled irvhen close to the stronger source^ by sudden inter-
nation or application of stimuli o

Series iii— Reactions to radiating light from bulbs

In these experiments^ summarized m table kg an attempt
was made to repeat the results of Series ii, using the
painted 200-watt bulb and a 100-watt bulb of the same lumi-
nous intensity (10^ foot candles), and a 60-watt bulb with a
lower intensity (7C foot candles) » One yellowfin and five
tunny were present.

Experiments 1 to U demonstrate a positive tropism to
continuous light at both intensities, thus confirming simi-
lar resiilts in Series ii„ In every case, the number of
entrances under experimental conditions was considerably
greater than under control conditions o

In Experiments 5 to 9, the light was interrupted regu-
larly at various rates, and also irregular lyo In the latter,
the time of interruption was varied during the course of the
experiment to observe the effect of the change from light to
dark (and vice versa) on the pattern of movementc Fith the
100-watt bulb there was positive tropism at regular rates of
70 and 100-120 times per minute with the on and off periods
equal, and also with irregular rates j, with no appreciable
difference in the strength of the response between the regular
and irregular rates o Similarly there was positive tropism at
a regular rate o"" 6 times per minute, with the on period 0,5
and the off period 9o5 seconds c. These results confirm and
extend the obser'/ations included in Series ii, Mth the 60-
watt bulbj however, there was no definite attraction to irreg-
ular interrupted light, the tuna entering the field only twieeo
This lack of response is possibly associated with the low
intensity of the bulb (70 foot candles).

Series iv— Reactions to radiating light from white and colored
bulbs of low intensityo

In these experiments sources of light of smaller luminous
intensity than that of the overhead lights were used with con-
tinuous and intermittent stimulation,, One yellowfin and five
tunny were present. The results are given in table 5o

t3

U

0

■p

0

•p
o

1

rH

-g

(D

a

CO

1^4

Q)

V

<D

<D

(D

=H

O

<D

-U

U

U

+3

<i-(

^

-P

e

<D

S

-P

(D

0)

0)

•p

C

-P

-P

C

>

!>

si

<H

CO

°d

»H

■P

•H

•rf

X!

to

a

P^

■P

■P

CO

•H

O

c

•H

*

•H

"ri

•ri

rH

H

'H

«t-l

■p

o

to

93

«H

<D

01

o>

^

m

s

O
P.

O
ft

a,'

+»

M

U

n

T*

G

O

•H

t;

0)

rH

o*»

«M

ct.

o^

Ci

o

H

X!

•73

rH

m

G

m

<a

cc

2

^

Q

^

<D

rH

o ,

bO

H

rH

pH

>»

a

0)

O

>»

O

•H

a

o

O

o

rH

o*

13

(£<

x>

O

-P

•H

o

•3

o

C

®

a>

>>

X!

O

^

X!

0

Si

D

X!

0

0

O

■p

•p

J2

O

CO

s

c>

0

o

0

o

0

B

s

0)

00

CO

m

0)

•n

01

c

in

CO

W

Sh

1-1

<D

M

•H

0)

£

•o

m

w

a

CO

C

•e

e

c

(»

^

OS

®

c

'H

o

d)

«H

=H

-H

CO

1^1

o

CO

hH

o

o

o

<n

tJ

e

CO

-P

*

t''*

■P

m

C

CO

0)

d

0)

•s

0)

Sh

c

a

•H

I-i

G

9

a

s

S tJ

•H

-p

P

S

CO

(X

-P

«

t>»

CO

CO

O rH

U

pt<

C

S-i

o

a

^

f^ c

^^

s

m <D

m

g

O

0)

-p

^

<»

p

l>>

■" o

■p

o

"H

a.

.§

c

-p

G ^

e -H

a

'H

C ^

»4

-4-

»

r^

W

M

»

M

Cm

<D

<iH

-P

U

(Q

O

O

c

o

(D

ts c

^

c

CO

0) a

•H

CO

t3

■^

rH

<^

<^

if\

c^

m,

o

«^

<^

CV

o

rv

«n

C

J2

M «H uN

(H

rH

H

rH

H

H

rH

fH

m

+3

o c

G

to

te^

0)

P "H

d

»

o

-P

SI

t»»

^— .

G

c

aJ

05

Q)

•H

•H

D <U

S

•P

01

+> r-l

sO

0

vO

O

0

vO

D

•^o

B

vO

o

0

vO

sO

•H

(3

C

S-B

O

B

O

r-

0

O

D

o

0

o

t-

0

O

O

U

5

(D

rH

i-t

rH

rH

(H

rH

rH

«•

o

43 Q-J aj

^

C

v^ O

o

M

(D

+>

M

«o

a

<D

o ®

^ HP

Si
•P

-p

Vt -P 0

O

o

0

O

O

P< C

cv

()0

bo

tn

5^ -a

0

e

0

0

0

0

0

o

0

rH

<D

11

o

*^

H

o

•p

l:^

B

n

8

0

0

0

0

o

J

0

o

I:

0

»4

vO

CO

^

0) Sm^

o

"H

»H

t4

-P ID

rH

a

S

C P

M

•H

•H

-P

O

o

cs

?-•

(D

>a

^

^^

t

^

^

Xi

x>

^

^

o

o

•H

1

•p

•3

•P

%

J3

.£)

i

■a

f-l

•p

J3

•P

jOt

o

°H

-P

«

(0

-p

-P

P

•p

o

•P

»

+3

0

»

0

■p

0

P

-p

0

P

■p

0)

in

<D

-P

g

0

0

o

-P

9

o

0

%

1)

n

•p

"

i

i

CO

•H

QI

0

fl

o

o

0

0

D

Q

in

TI

"H

o

w

B

Oi

o

o

0

O

o

o

0)

iJ

o

s

o

o

2

o

o

o

CO

rH

•m

•X3

H

iH

rH

rH

11
1

a

s

a

0)

a

a

a

G

G

g"

o

°H

n

-H

a

Ti

-H

•H

-H

"tH

o

■>*

eO

rH

01

^

•rl
03

^

CO

r^

(0

rH

CO

t-{

00

rH

aa

<D

PL.

(Ih,

&«

CI.

(U

PL,

P-.

CL,

Pk,

'2

H

G

^

o

(0

o

o

E-H

0 O

rH
O

p

rH
p

fH

o

rH
O

(D

m

u

U

»H

u

u

O) -P

f-H

p

<N

<»>

P

•-*

+>

l/\

-p

NO

C^

^3

to

(Tv

la

p. 53

c

g

c

c

a

G

H 0)

o

O

o

o

o

^

(3 a

o

o

o

o

o

48

c

CO
+>

cm

0)

O
H
O
O

■H

CO

<D
+»
•H

■§

13
(D

•P
O

■H

XS

s

a

•H

+5

C
O

o

o

+3

-p

•P

■P S

O

03

c
o

•H
-P
O
(0

(D
•H

7

0)

-P

CO

•H
T3
(0
»4

T3
<D

-P

?

<D
■P

a

o
■p

•^1

a
a

(D

p

•P

>» o
<i-t

&•

OS
c

2
p
c

(\J

CO

•H

o

o

<D

to

■si-

+3

^

P-i

O

•p

-P

-P

-p

•P

^

-g

g

I g

»H

D

11

•J

O

o

■^

^

OS

^

10

Eh

0

0 O

H

H

t""

2

g

<D P

r-\

p

cv

P

a. c!

c

a

X (D

o

o

m B

o

o

CO

•p

83

<H

iH

rH

O

O

O

D]

X!

o

>»

CO

^

t>»

o ^

o

(3

<D

o

O

m

t

w

-P

o

c

Cik

<a ss

u

to

p

"H

C! en

W

<r\

■P

p

p

CO

&

G
O

o
cv

c

•H

r^

?J

CO

o

NO

0)

■p
6

-p
p

i

0

o

a
CO

iH

CV

0)

p

■P

+3

0

o

c

•H
(0
rH

0-1

s

o
o

03

-p

<D

U

o

O

vO

^

nO

^

6
o

rH

-P
■P

u

8

o

o

13

(D

O

rH
O
O

U

cm

-p
-p

0

O

rH
O

o

CO

a

CO

-P
-P

0

8

<D
U
O

fH
O

O

C^

A9

Ihroughout the series there was no definite positxve tropism
to radiating white light from a 1+0 -watt (1+7 foot candles) bulb
with either continuous or interrupted lighto Although in Experi-
ment 1 there were 2 entrances j in Experiments 2,, !;« and 5 there
were noneo With continuous light from green^ red^, amber j, and
yellow bulbs of intensities ranging from 6 to 62 foot candles ,
there were no entrances into the field q even though the 100-watt
yellow "insect repellent* light had an intensity (62 foot candles)
only slightly less than that of the 60=watt white bulb (70 foot
candles) which attracted the fish in Experiment 3 of Series iii
(table ij,)o

Lack of response may have been due to the low intensities
usedj to a lack of responsive condition in the f ishj, or to a
combination of botho Lack of responsive condition is indicated
in Experiment 3 using continuouii white light from ' 20D-watt
bulb in which only 3 entrances were made as compared with li;
in Experiment 3^ Series ii (table 3) o There was no apparent
reason for the difference in behavioro As all of the experi-
ments were low intensity sources except one (Experiment 1), and
also the controls j, produced 0 entrances ^ it is believed that
the lack of response was due partly at least to the low intensity
of the various white and colored bulbs «

Series v-— Reactions to the horizontal beam of a projector and
to radiating white and colored light from bulbs o ~ ~

These experiments were performed to study the reaction of
txma to a strong (530 foot candles) beam of white lights and
to repeat amd extend the observations of Series iv on the
reaction to colored light of low intensityo One yellowfin and
five tunny were present o The results are shown in table 6o

The experiments were conducted with the projection lanterno
In Experiment 1 there was one entrance by the yellowfin j, but
on four occasions the mixed school approached the border of the
field and then turned baoko In Eiq^eriment' ? there was no
reaction, the fish behaving as under control conditions o That
the fish were in a moderately responsive mood Is shown by the
6 entrances under stimulation of a 200-watt bulb in Experiment
3o It would seem that the tuna were not attracted,, to the field
at least, by the strong continuous light bearao

In experiments h to 6^ light bulbs fitted with rede, greeny
and blue filters were used, giving intensities somewhat higher
than those of the bulbs of comparable color used in Series iv,
but less than those of the ai'c lamp beam of comparable color
used in Series lo The green light appeared to attract the tuna^,
but it is doubtful if the results are of general significanceo
Neither red nor blue light seemed positively attractiveo

50

fH

g

^

g

43
0

CO

43

1

43

on,

•P

oo^

CO XI

C >!

C

"iH

0

u

■H

Od

a

^

CO

o

<iH

<D

0

XI

•p

0!i

^

S

u

m

0

o

^

0

vH

«H

r-l

rH tj

(D

r-{

4^

0

rH

(D

•r-»

CO

r-{

0

0

(D

G

O

0

m ->*

0

X!

>»

1^

U

cdI

>>

G

0

J3

<x

orfl

^

a?

a

>»43

f>>

7\

43

.Q

e

X>

M

0}

C

to

o

,0

•H

-H

OJ H

14

0)

-d

«D

0

Ch

0)

+3

0
G

0
X3

^

^

g

0

OS

0

CO

bO to

1

43

Xi

P

p

"H X>

d

4^

m

G

G

>> 0

TJ X)

1

P4

M

W

G XJ

«M

to

o

01

O

0)

"Ci

c

+■' B

o

rH

-H

P. O

U

c

0)

s

S k

0)

CO

■H

U ch

,42

S-t

^

IT;

rH

0

0

vO 0

c^.

0

0

00000

0

0

T^O

Fh

-p

0) -P

^

c

O

c

•p x:

S

0)

+3

=rt!

5^

iH

>»

tJ

■p

CO

e -o

•H

B

a>

-<rco to r^fH

CO »»

03

+5

rH

00000

-(J

C

o

tS

0

0

8

m vO to

0

B

0 •< Cvi 0 0

8

0

«n 0

CQ SS

(D

o

c

cn

rr.

B

H c^CNi

fl

fl

f>^

H 0

rJ -H

•P

^

CO

>^

vo,

rv

ir\

CM

o -^

G

*-*-«'

o

:3 (0

M

C t4

•H

4^ TJ

Ci

a Q)

o

(D

o s-

■H

+5

o o

^ -P

S

H

O

&. fl

-P o

(D

s

-H

a

0

0

0

0 B

0

0

0

0 0 B B 0

D

0

0 0

43

M

0

B

B

B B

0

B

0

8 D fl fl B

fl

C--

C^ B

S-H

^

(1)

43

»4

0)

3 to

fi

P

4^

iH

0)
ti-l +:

S „ ^

O 'H

t^

0)

0)

0)

G

0 ^i G

(D

<D

s

o

43

+>

43

a>

<D

rH 0) (D 0)

P

P

to ^

fH

»H

Ti

0

-H t;

(D

d

B

TS rH X3 0) d

fl

"H

:d§

j3

O

X!

J3

B

xi 0

U

iH

0

Q) © B ^ rH

fl

^

o m

o

»

&

& U

bO

XI

u >» m MXI

^

B:

"H 3

+3 O

O 3

^

03 C

bulb
bul
b-olb
bulb
bulb

Q." .H

0

0

^

0

5

o

p

43

XI u

k

^

43

P

XI

> o

0
«

0
0

P P

^

P

43 43 4J 43 45
4J gj +J ,^3 43

0
0)

43

CO o

'r-3

1-3

+3 H

H

rH

*C ^ © ^ ^

S 0 & ^ S

"r-s

43

di -p

43

0

0

gslsl

•H

0

g 0

^

J3

S^

u

9

<*H

0

0 0 B 0 B

3

tH

0) pj

o.

a.

"

B
0 XJ

X!

XJ

fl

Q 0 0 Q Q
<0 rH vO vO NO

fl

P.

0 fl

0

CO £0

3

p

43

0 43

+3

43

43

0

0

4^

a

4^

CVJ -g -H

°g

t3 tJ T3 'O 'O
(D 0 Q) Q) <D

P

B^

^

^

CO

a

M »H ^ ^ ^

»

m a

vD

n

<x>

0 g]

•H X>

XI

X)

00000

0

d) -rl

■3)

0 J3

0

8-"

^'^'^

•^

rH rH rH rH rH
00000

0
0

Xi a

,-H

rH

4f>.

vn

cu m

03

m

0 0 0 0 CUf

ir.

CL<

Xi

<a

EH

0

0

H
0
U

(H

E

s

O 43

rH

fV

P

f^ -><-«A

^043r-t»OOrHP

C\J

sf^ 43

&

c

c

c

rH rH

G ^

pH c

0

0

0

0

0

»

s

0

0

0

0

51

In Experiments 7 tc 11^ colored bulbs were usedj, the
intensities of which were reduced to Oo1-Ilo8 foot candles
by placing a resistance in the circuito No attraction what-
ever was noted at these low intensities ^,

In Sxperiments 12 and 13 j, the reaction of the fish to
intermittent light was compared at high and medium intensi-
ties o With 70 interruptions (on and off periods equal) per
minute, 530 foot candles-, there was no positive responseo
With 30 interruptions per minute (on and off periods equal) ^
the tuna entered the field 3 times, whereas under control
conditions it did not enter at alio

Series vi— Reactions to white and colored beams of light from

proj

ector

As the results of the experiments in Series v suggested
that light over 500 or below ^0 foot candles is ineffective in
attracting the tunaj the present series was conducted with
white and colored light of moderate intensity (70 to 321; foot
candles) from the projector lanterno The intensity of the
beam was reduced by placing a resistance in the circuito One
yellowfin and five tunny were presento The results are shown
in table 7o. As indicated by the relatively large number of
entrances (6j 8^^ and 5) under control conditions j, the fish were
more active and restless than usual for some unknown reasono

In Experiments 1 and 2, using continuous white light j
positive tropism was observed, as indicated by 9 and lU entran-
ces compared vn.th 6 under control conditions „

In Sxperiments 3 to 6, colored filters were placed across
the projector beanio Although Experiments 3 and k ware not con-
clusive, there seemed to be a definite tropistic reaction in
the duplicate Experiments $ and 6 with red and green filters,
as shown by 17 'd 10 entrances as compared with 8 under control
conditions o

In Experiment 7, white light from the projector (32i4 foot
candles) was interrupted 72 times per minute (equal on and off
periods). The tuna entered the field 3 times., and were turned
away at the border ih times when the light changed. In Experi-
ment 8 with v^hita light from a 200-watt bulbj, similarly inter-
ruptedj the tuna entered the field 8 times as compared with 5
times under control conditions o

52

43

0

o
o
»n

a

(D

t

OS

3

o
ss

c

•H

o

e?

o
■p

CO

a

3
•P

o

a
o

=H

4^
O
CO

«

I'

0 o

°> o

-p

in o

CO P<

9
B
o

C^

0)

iH
X»
CD
Eh

CQ

O ® TJ S,

^ fl 0) e

<D CO "H

a -P

3 C o c

a 0) +> oH

■p aa

m -p H

(D o pi

-P Vl CB

(3 !

O Ol

=H -P
■P ;S
P. G

4h
O

(D

-P

■J3\

O

:S

0 o
0) -Pi

e
o

•H
+3

O
00
0)

o

•H
-P

ft
O

u

-p

<D

+5

•H

OQ

O
ft,

^0

O
■P

«H

■g

0
-P
(D

"H
CQ
<D
»H

J3
-P

t

U
O

■P
o

(D <t>

o a

U CO

o
u
■p

o

o

Tl

TJ

»tf

rH

H

H

c

(B

<U

a>

o

■c

5^

•ii

43

•P

(Q

CO

O

<^ «

<D

0}

<D

(0

t3 fel

X( ^

o

■o

CO

?

■ri 0)

•H a>

S3

cH

(D

h

§^

s-g

g

a

o

•H O

•H O

53

S3

+»

-H

43

»H

XI

^

O

O

a

-P

CQ

<Q

=r-t

•H

<D

OJ -«*

D3

<D +■>

0) -P

-P

■P

<D

rH

a

of^

S 00

oS''

O

O

tH

a

0

P<

•H

CD

CD

O

•H

>9

<H

o

•P CQ

■P CO

0)

(B

■P

g

+J

u

<C

0)

U

fn

U

0

■p

^a

o a

<D

r^

CO

CO

•H

O

O

^

<D

(D

® +>

(1) 4J

Ti

»H

<»

T3

u

^

P>

'-» ^

o

+3

•P

3

c>

«r

®

•H

a vo

C ■sj-

01

»

s

C

q

'^

t3

-P

CO

<0

"H

"rt

OD

(D

•H

^ -o

?H TJ

P,

O.

^

(h

0

)h

CR

■P c

•i- c

O

O

t)0

p

+3

ja

CO

O

fl CO

i:^ (Q

JH

Ih

•H

C

0

ft.

w

»

E-<

e-i

w

jS

CO

O

CO

o

CO

r^

to

+3

Si

0
4>

m

CO
0)

X!
43

o

0) U

••-a C
O CO
M

ft<

Cs!

4^

X!
43
»H
&

Li
O

4J

o

0)

°r^
o
u

ft.

r^

C

<D

(D
X!

O

o

<D
•>-s
O
U
ft.

a

+3 +3

•H
<1H

X5
43

•^ t

u
o
+>
o

ftl

o
+>

o

o
ft.

vO

iH

o

43

a
o
o

0)

+■>

X!

'H ®

O +»

+3 (0

O oH

<D 0}

O U

S->

ft4

(A

-4-

B

■-4-

0

0

CJ^

0

0

■^

irii

^

0

cv

E--

C-

C-

c-

0

OJ

H

c^

0^

en

CM

^ ^

»H 0

4»
4-''

0

o
o

(0

«0

o

43

a
o
o

53

DISCUSSION

It should be anphasi'?. ed that the experiments which have
been conducted were exploratory in nature^ and lead only to
tentative conclusions o If some of these appear to have practi-
cal application they should be checked by further studies con-
ducted according to planned experimental designs^ and with
suitable statistical analysis.

It should also be emphasized that the strength of the re-
sponse to light stimuli varied to some extent from night to
night (apart from obvious extraneous sour'ces of interference)
indicating a variation in the responsive condition of the fisho
Therefore the results of individual experiments are not strict-
ly comparable from series to series o

It should further be en5)hasized that the e}q)eriraents ware
conducted in a relatively small tank and hence under highly
artificial conditxonsr Although certain suggestions may be made
as to the response of tuna in the open sea to light stiraulij
these should be accepted with great cautionc

Reactions to continuous white light

The reactions of the tuna to continuous -vi^ite lights are
summarized in table 8^ omitting only those experiments which
were disturbed by known extraneous factor So From the many expe-
riments which showed positive results^ it is evident that the
tuna undergo a change in behavior when exposed to a continuous
white light of source intensity between 70 and k50 foot candles.
This change in behavior is judged to be a tropistic response to
the stimulus^ in that the fish approached closer to the source
under experLmental conditions as compared vvith control condi-
tions,, Moreover, with a horizontal beam, as opposed to light
radiated in all directions from a bulb 5 they approached the
source in a lin-. parallel to the beam and turned away only after
enterirxg the fifcji.d and approaching the baffle. There was no
conclusive difference in the strength of the reaction between
intensities of 70 and about i;^0 foot candles, although apart
from the results with the arc lamp (the higher estimated inten-
sity) there seems to be a tendency for greatest response (exper-
imental minus control entrances) at an intensity of I06 foot
candles o On the other handj vreak or no response was obtained
with light of low intensity (Ii7 foct candles) and also with
light of high intensity (530 foot cand]_es)o

xt seems possible that tuna cculd be attracted to a light
at seao However,, this would have to be a very high intensity
to penetrate the water for any great distance , Although the
tuna might be attracted to a light of high intensity., it is
doubtful if they would approach very close tc the sourceo

$h

CQ

(0

n

(D
-P

C
•H

■P
C

en

o
■p
■§,

•H
H

0)
+»

I

CQ

3
O

c

•H
-P
C

o
o

o

•H
-P
O
03
0

u

Cm
O

>»

as o
o +3

c c

CO O

u ol

-p
a

<D

<M -PI

o a

^

w

0 t D

> > M

D 0 0 0
M M M M

c^ ir\ (^ f^N

J 0 0 0
M M > >

9 Q

M M
> >

0 0 0 II

M 1-1 > >

O O

^H o8 O

•s^~^t o o

sO sO

r^ f^

o o

0)

CV

H i-i fH

^\0 «*N^O

r-i r-i

H O

crj

E^

-<i-

g

O

■^

o

o

CNi

iTl

'^

f^

->*

ITi

CO
(0

55

Reactions to intermittent white light

The response of the tuna to intermittent light of various
intensities and rates of interruption are s-ommarized in table
^, At lovf and high intensities CU? and 530 foot candles) there
was no positive response at any rate of interruption which -was
triedc At an intensity of 106 foot candles, there was positive
response but there was no clear relationship between its
strength and the rate of interruptiono Ihere is, perhaps, a
suggestion that the slower rates produced stronger reaction,
but this cannot be proven with the present data. Certainly,
however, the reactions with interrupted light were no more
pronounced than with continuous light „ At 21^ and 32I4 foot
candles, the results were erratiCo Although the fish were
attracted towards the sourcej, they were scared av/ay at or near
the border of the field when the light was switched on or offo
This "scared reaction" was not so pronounced with the weaker
light 0

The results indicate that in attempting to attract tuna at
sea, interrupted white light would give no better response than
continuous white light, and moreover, that the fish would prob-
ably approach less closely to the source o

Reactions to continuous colored light

With one exception C-^periment 5y Series v), continuous
colored lights of intensity of about I4C foot-candles or less,
did not induce a positive reaction from the tuna,, This was also
the case with a yellow "insect repellent" light of somewhat
higher intensity (6? foot candles) o

On the other hand, continuous colored lights of higher
intensity (about 7C foot candles) evoked about the same response
as continuous white light of moderate intensity (70 to about
it^C foot candles) o The response was evidently to the intensity
rather than to the color 0 There was no evidence that light of
any one color was a stronger stimulus than a light of any other
color for lights of approximately the same intensityo

There is no indignation that colored lights of the wave
lengths which were used would be of any advantage over white
light in attempting to attract tuna at sea^

56

•s

a
m

(D

43

<n
C

(D

-p
c

G

U
<D

*^
o

00

<D
-P

(D

+»

-P

"H O

•H

O -P

+-■> fX

«i
O f^

•H 0)

-P +3

O S3

Ri -H

(D

O

o tn
t>»-P

^ OS

53 u

i TJ

m OB

0

0

o

^

H

m
8>» a)

ra c
c <D

-p y

a -^^

M O
O

if% ^ a^ ^o to «^ o ^

0 a ! g 0 0 0 0

>> MMMMMM

(^ O

CO rH t" H

M > M M
> MM

^

M

(I

>

O O

C --*•<?■ li'^ •-* m M^ O -•* ■-t

o

o o

f<^ M M o t- tn

M M M M M

to 'fn o -^ f^

A/

o

■■o

vD O O O «^.
^£> C- E^ C~-

O

M

G

o
o

h|oj

O O T-.

<T-'. c- e^

E^

o

o

cv

-t o

M
«8
3

<y

I

m
o

u

®

<i-t
o

+5

a

Pi
®

a,

«!)
X!
-P

O

fH
M
«0

O

i«

■o

c
o

"J
<i>

o
O

c
o

o

57

REFSREircES

Kuhne, Wo and Ho Sewall

1380 o Zur Physiologie des SehepithelSj, insbesondere der
Fische, Heidelbergo UniVo Physiolo Insto Unterso Bd;. 3j
pp, 22-277,

Shimadaj Bo M:

1951 o An annotated bibliography on the biology of Pacific
tunas c UoSo Fish and '/Tilcdife Services Fishery Bullo V„
52 Cnoo 58), ppo 1-58

Tester, Ao L„

1952 o Establishing tuna and other pelagic fishes in ponds
and tankSo UoS. Fish and Wildlife SerVo, Speco Scio Reptc
Fisheries Noo 71s 20 ppo

Wallj Go Lo

19il2o "The Vertebrate Eye and Its Adaptive Radiation,"
Cranbrook Institute of Scio Bull, NOo 19o Augc 1942o

Warner, Lo Ho

I93I0 Ihe problem of color vision in fishes, Quartc Rev,,
Biolo, Volo 6, ppo 329-3U80

58

PART IV

OBSERVATIONS ON SOUND PRODUCHON AND RESPONSE IN TUNA^/

b7

Iwao Miyake
Associate Professor of Physics
University of Hawaii

INTRODUCTION

At the outset of this investigation, two questions were asked,
Cl) do tuna produce sound, and (2) do tuna respond to sound stimuli?
If tuna produce sound, there was the possibility that their presence
in the open ocean naght be detected by the use of listening deviceso
If tuna responded to sound stimuli ^ there was the possibility that
the emission of sound of a certain quality and quantity might attract
them to a fishing vesselj or that the emission of sound of another
quality and quantity might repel them to practical advantage in fish=
ingj eogoj upwards from the lead line of a purse seine during pursing
operationso

The exploratory observations reported below were conducted at
intervals from August 1951 to January 1952s on tuna confined in a
tank and a pond (Tester 1952) at the Hawaii Marine Laboratory 3
Coconut Island, Oahu, To Ho

MATERIAL AND APPARATUS

When the experiments were first conducted m the concjrete tankj
it contained but one yellowfin (Neothunnus macropterus) and a few
manini (Acanthurus sandvicensis)o Later j another yellowfin and a
small tunny CE'-^thynnus yaitojlyere present o When the experiments
were conducted in the large pond, one yellowfin and one tunny were
present, along with several small reef fishes of various ^ecieSo

Three dLfferenx types of borrowed eqiiipment were used during the
course of the worko These consisted of Model OAJ iioimd Measuring
Equipment (reception range^ 1 to 10 kilocycles per second) and Model
OCP-1 Sonar Test Equipment' (reception range 7 to 70 kilocycles per
secondi transmission range, 5 to 88 kilocycles per second) ^ both
loaned through courtesy of local representatives of the U.. S,, Navy<,
and NEL Underwater Sound Monitoring Equipment C^eception range, about
100 cycles to 10 kilocycles per second),; loaned through courtesy of
the Navy Electronics Laboratory, San Franciscoo In addition,, a
sound generator consisting essentially of a P-H (Packard-Hewlett)
audio oscillator (transmission range, 20 cycles to 30 kilocycles per
second) was also usedo

Contribution Noo 25 of the Hawaii Marine Laboratory., University
of Hawaii. J Honolulu, Tc Ho

59

During most of the work the apparatus was installed as
indicated in figure 17o In attempting to discover if tuna pro-
duced any soundj the hydrophones of either the OAY, CXJP-l, or
NEL apparatus were suspended 18 inches below the surface of the
water near the middle of the tank (about 17 feet from the end),
or 17 feet from the inner (unscreened) gate ports at the seaward
end of the pondo Sounds in the water were thus detected, anpli-
fied, and heard using either earphones or a loud speaker. In
attempting to discover if tuna responded to sound stimuli, the
transducer of the OCP-1 (a hydrophone used as a transmitter) and
of the P-H apparatus were suspended 18 inches below the siurface
of the water at the south end of the tank, or at the inner gate
ports of the pond. The reactions of tima were noted and recorded
under control and experimental conditions o

SOUND PRODUCnON BY TUNA

Procedure

Using one or another of the three hydrophone"-amplifier sys-
tems, over 100 hours were spent in listening to noises emanating
from the tank and pond, and in attempting to ascertain their
causes 0 Because of the possibility of sound production from tuna
occurring at some hours and not at others, the observations were
spaced to cover all hours of the day and nighto

Results

In the tank the hydrophones of the NEL and OAT equipment
picked up many sounds, but two were distinctly noticeable over
the background noise,, One sounded like the snapping of a dry
twig and the other was complex^ somewhat like that coming from a
beaker of violently boiling water.

The snapping sound was identified as that of shrimpso This
was established Ly bringing the hydrophone near a group of shriisps
hiding behind one of the comer baffieso The conplex sound was
for a time thought to be coming from either the tuna or liie
maninio It was later determined that this sound,, and the other
background noises, were caused by the water pun?)ing and tank
overflow systems.

During one of the night observations, when one yellowfm was
present in the tank, three distinctly different sounds were
heard. One was a very low frequency so\and of very short duration
sounding somewhat like a window rattling in the windo The second
sounded like water suddenly stirred with a stick The third
sounded like a stick being dragged over a piece of sheet metal „

60

HYDROPHONE

NEL AMPLIFIER
EARPHONES
BATTERY BOX

TRANSDUCER
AUDIO OSCILLATOR
AUDIO AMPLIFIER

B

INNER GATE PORTS

/

^q <!>,

Q

DEEP
SHALLOW

>r

TRANSDUCER ^ HYDROPHONE

..^ L

FILL

-T-T

FILL

^f:

TV^

FIG. 17 DIAGRAM OF THE TANK (A) AND POND(B) SHOWING THE POSITION
OF THE TRANSMITTER AND RECEIVER USED IN EXPERIMENTS WITH
SOUND.

61

Later, day and night observations showed that these sounds
were caused by tuna in the following manners

(1) The low frequency sound of very short duration was
caused by the sudden movement of the tail of the yellcwfino
\Bien getting underway^ the first one or two movemsits of the
tail created enough pressure in the water to produce sound
waves which were picked up by the hydrophone,, Vfhether this has
any biological sigiificance from the point of view7schooling is
a matter of conjecture,, Certainly^ in the large pondj the
speed undergoes sudden chai.ges while the tuna is swimming its
lengthy and this is also likely to occur in the open seao It
further suggests the possibility that tuna may react to sound of
very low frequencyo

(2) The second sound was produced when part of the tuna's
tail came above the siorface of the water^ As occasionally tuna
are observed to jump and play at the surface, this sound may also
be produced in the natural habitats

C3) The third sound was produced when the tuna accidentally
rubbed against the hydrophone o

No additional sounds were detected when two yellowfin and
one tunny were present in the tanko

During one of the tests, one of the yellowfin suddenly went
beserk, and began bumping and scraping itself against the walls
of the tanko Ylhen it died, its skin was almost conpletely eroded^
However, even during this period of frantic acti\'ity, no sound
of moderate or high frequency range (1 to 10 kilocycles) was
heard with the hydrophone in use at the time (OAY) o

It was decided that a test should be conducted to explore
the possibility that tuna might be emitting sound in the super=
sonic range o A o-hour trial with the 0CP=1 apparatus gave nega-
tive resultso Further trials were not conducted because the
apparatus was no longer available o

The NEL equipment was also used to investigate sound product-
ion by tuna in the pondo Many different sounds were heard,, but
none could be identified with either the yellowfin or the tunnyo

Summary

Over 100 hours were spent in listening to the sounds picked
up by the hydrophone placed in the tank and the pond contair mg
tunao The listening periods were staggered so that all hours of
the day and night were covered o

62

Certain low frequency sounds were produced by the sudden
movement of the yellcwfin's tail below and at the surface of
the waterc No sounds emanating from either yellowfin or
tunny were detected over a frequency range of 1 to 70 kilo-
cycles per secondo

RESPONSE OF TUNA TO SOUND STIMULI

Procedure

It was planned to investigate the reaction of tuna to
Cl) steady sound of various frequencies^ (2) interrupted sound
of various frequencies j, and (3) complex sounds of short dura-
tion. Unfortunately J difficulties with the equipment and the
eventual death of the experimental fish^ prevented the program
ftom being corapleted„ Such observations as were made pertain
to Cl)? onlyo

The experiments were started in the concrete tank, and
were later removed to the pond„ For stimuli over the super-
sonic range g the OCP-1 equipment was used in the concrete
tanko For stimuli at lower frequencies, the P-H equipment was
used in both the concrete tank and the pondo

In the latter experiments., the N3L hydrophone was also
used at a fixed distance (17 feet) from the transducer o This
distance 3, half the length of the tank, was maintained when
the experiments were performed in the pondo The 60-watt
anplifier of the P-H equipment was adjusted until the sound
became audible at this distance^

To enable any reactions of the tuna to be measured in a
roughly quantitative manner, the time which the fish spent in
Areas S (sound) and Q (quiet) was recorded under control and
experimental couditionso The areas are indicated m figure 17o
The sound stimulus was audible (through the hydrophone -ampli-
fier system) in all parts of Area S, but not in Area Q, except
near the boundary o It was assumed that the tuna„ if attracted
or repelled by a sound, would spend relatively more or less
time in area S than when there was no sound stimulus.

In the tank, under control conditions, the times spent by
the tuna in Areas S and (J should be equal „ However, in the
pond under control conditions, the time spent in Area Q would
be much longer than that spent in Area S because of the rela-
tively greater size of the former area and because, normally
the tuna circles the pond turning at the west end within Area
S, and turning at the east end at varying distances within
Area (Jo

63

Results

7 to 70 kilocycles per second

Experiments at supersonic frequencies were performed in
the tankj in which one yellowfin was presento Under control
conditions J the yellowfin spent about the same period of
time in Area S as in Area Qo

Supersonic sounds of frequencies of 7$ 10,, and in steps
of 5 kilocycles thereafter up to 70 kilocycles were generated^
each for a period of one-half hour, and the reactions of the
yellowfin were observed and recorded o As the results vfere
negative 5, the data are not included in this reporto In each
case 3 the time spent in Area S was about the same as that
spent in Area Qo

In these experiments there was no way of knowing whether
or not the transducer of the OCP-1 equipment was generating
a signal, because at the time, a separate hydrophone was not
available. It seems reasonable to assume that it was func-
tioning^ as it had been used successfully as a hydrophone in
a previous experiment » The resiilts indicate^ therefore ^ that
the tuna was not affected by a steady sound of frequency
between 7 and 70 kilocycles per second 3

$00 to 5y000 cycles per second

These experiments were also performed in the tank^ using
the P-H sound generator and the NEL receiver,. Tests:, each of
20 minutes duration^ were performed using 500^, 1^000, 2,000,
3j,000, li.OOO, and 5,i000 cycleso The series was repeated three
times 0 In none of the experiments was there any definite
reaction o

To deterraiae whether the yellowfin reacted to a sound fre-
quency other than, that used above, the audio oscillator was
varied continuously between 100 cycles and 10 kilocycles.
There was no noticeable reaction o

100 and 200 cycles per second

As there was the possibility that the tuna in the tank may
have become acciistoraed to strange noises^ such as those emanat-
ing from the pumps, the apparatus was moved to the large pond
where a yellowfin and tunny were presento After a few hoiirs of
work, the test came to an abrupt end when saltwater leaked into
the transducer, short-circuiting it, and burning cut the power
output transformer in the amplifier,, The fish in the pond
died before a new transformer could be procured^ thus terminat-
ing the experimental work.

6U

Such results as were obtained are discussed briefly^ as
they indicate a possible reaction to sound by the yellowfino
The reactions of the tunny were also observed^ but were
rarely recorded because of the difficulty of timing two fish
at onceo The tunny's behavior vfas independent of that of the
yellowfin^ and it tended to remain in Area (^,

Under control conditions j with the apparatus in position
but not in operation, the yellowfin spent about 6 minutes out
of an hour in Area Sj passing into the area about 30 times o
Successive periods of time spent in Area Q varied ftom 20
seconds to 3 minutes o

Ihe effect of continuous sound of 100 cycles per second
was first tested. The yellowfin spent 10 minutes out of the
hour in Area S^ entering it 26 times o

Daring the hom-, the noise of the exhaust of a boat was
picked up fairly loudly by the hydrophone^ Most of the sound
doubtless entered the area from the seaward or west end through
the screened gates o Ihe fish remained in Area S for a much
longer time during this period of disturbanceo It also re=
entered the area after being away for but 20 to U5 seconds o

If the period of disturbance is discounted, the fish
remained in Area S for about 6 minutes out of 55 minutes^ a
result which does not differ greatly from that of control
conditions,, Although there was no good evidence that the
yellowfin reacted to sound of 100 cycles per second, there is
the sijggestion that it might have been attracted by the complex
sound coming from the exhaust of the boato

The effect of continuous sound of 200 cycles per second
was next tested^ Ihe tuna spent about 17 out of hi minutes in
Area S, and entered the area 18 times,, Ihe results are included
in table 10 j alu:.fi with those under control conditions ^ both
to illustrate the reactions, and to show the type of data which
were taken o

For some unknown reason, the yellowfin became interested
in the hydrophone^ it swam up to it six times and appesn-ed to
exounine it closelyo It also circled in the area between the
transducer and the seaward end of the pondo Its sudden inter-
est m the hydrophone was peculiar, as it- had been swimming
past it, without any reaction, for the previous 3 hours,,
There is the possibility that the yellowfin might have been
attr'acted by a sound generated by the rubber cable rubbing on
itself or its support,, A strong gusty wind prevailed during
this part of the experiment, and although the cable was not
observed to move with the wind, it may have done sOo On the

65

other hand, the fish may have been attracted by the 200-cycle
sound, and may have been attempting to find its source o

Summary

Ihe yellowfin in the concrete tank showed no reaction to
sound frequencies over a range of ^00 cycles to 70 kilocycle?
per seconds nor to a sound with varying frequency from 100
cycles to 10 kilocycles per secondo

The yellowfin in the pond seemed to respond to certain
sound stimuli s but nothing definite may be stated on the basis
of the data now available,,

66

Table 10,, —Time spent by yellowfin tuna in Areas S and $ in
successive circuits of the pond during control
(no sound transmission) and experimental (sound
transmission at 200 cycles per second) conditions o

Control

Experimental

Area S

Area ^

Area S

Area Q

Circuit

(seconds)

(seconds)

Circui

.t (seconds)

(seconds)

1

20

125

1

3

127

2

2

US

2

30

200

3

25

85

3

2

128

h

3

152

h

15

lUo

5

i;

131

5

3

122

6

3

102

6

3

72

7

3

107

7

115

60

8

10

20

8

300#

30

9

25

175

9

180-*

20

10

15

185

10

20

110

11

2

108

11

3$^^

80

12

2

178

12

3S

120

13

3

57

13

60*

120

Hi

3

37

Ih

20

120

15

2

33

15

2

103

16

3

82

16

55*

110

17

30

30

17

70*

115

18

ko

30

18

50*

h^

19

k

81

19

25

190

20

3

52

21

3

hi

22

10

ll+O

23

2

138

2li

35

3^

25

3

87

26

2

58

27

2

58

28

20

50

29

i;

26

30

3

hi

31

2

128

32

3

152

33

2

163

^

15

165

35

30

100

36

30

60

368

3272

1023

2012

In area between transducer and gates
At and around hydrophone

67

REFEEiENCE

Tester, Ao Lo

1952. Establishing tuna and other pelagic fishes in ponds
and tankSo UoSo Fish and Wildlife Servicej Speco Scio
Repto Fisheries Nob 71s 20pp„

68

PART V

NOTES ON THE RESPONSE OF A TROPICAL FISH (KUHLIA
SANDVICBNSIS) TO INl^ERRUPTED DIRECT CURRENTy/'"

by

Albert Lo Tester
Professor of Zoology
University of Hawaii

INTRODUCTION

A study of the reaction of tuna to electrical stimuli
was not attempted for the following reasons ^ (1) because of
the expense of purchasing a generator of sufficient power to
produce a reasonably high current density Csay^ 0,002 amps
per square centimeter) in the large (10,663 gallon) concrete
tank in which the tuna vrere kept^ (2) because of the danger
of harming the tuna, which were being used for other experi-
mental purposes^ and mainly (3) because there were indications
from the work of Morgan (19^1) that further research could be
undertaken profitably with the aholehole or "mountain bass"
(Kuhlia sandvicensis)^ using a tank and generator which were
already available., Consequently, with the small amount of
time that was available, the author attempted to duplicate
and extend Morgan's experiments in an effort to discover the
optimum pulse duration for minimum power output to attract
aholehole in an interrupted^ direct current system., It should
be pointed out that before seriously considering the practica-
bility of catching tuna or any other fish by electro-fishing
methods nn the high seas,, still more efficient use of the
available power than that achieved either by Morgan (195l) or
by the Cooperative California Sardine Research Program (anonc,
1950) must be made. Otherwise j, the power plant required by a
fishing vessel weald be eoctremely large, expensive, and there-
fore probably impracticable »

As the present study represents an extension of Morgan's
(1951) work, and as his data are not readily available, his
results may be summarized here to advantage o He atten^jted to
discover the minimum current which would lead or force
aholehole to the positive pole in a column of salt-water
(wooderTTank) measuring 12 x 2 x 1 feet, using a source E,MoF<.
of 220-230 volts, Do C<, In some experiments thie current was

1/ Contribution Noo 26 of the Hawaii Marine Laboratory j
University of Hawaii^ Honolulu, To Ho

69

used in a continuous flow, in others it was interrupted by a
specially-desxgned interrupter consisting essentially of
revolving disks with different proportions of brass and
bakelite, the current being "on" when two brushes were in
contact with the brass sector j and "off" when the two brushes
were in contact with the bakelite sectoro Series of experi-
ments were conducted with various currents and with vaiaous
frequencies of interruption at each currento In other series
of experiments the "on -off" ratio was changed^, using disks
which allowed the current to flow for Oo75s Oo50, and Oo25
of one comjjlete revolution. To maintain a source voltage of
220-230 volts, he found it necessary to design a special type
of electrode— a carbon rod enclosed in a plastic tube with
open ends and with holes bored through the sideSo By adjust-
ing the exposure of the carbon rods to the seawaterj, it was
possible to vary the current in the system^, but at the same
time to maintain the voltage at the soui'ceo

Morgan's results showed (1) that frequency of interrupt-
ion was not critical for the species and size-range of fish
used — -about the same response was evoked for frequencies
between l5 and 2$ ropoSo, and (2) that, within the limits of
his experiments, the shorter the pulse duration the smaller
the average (interrupted) current required to give the
desired response— decisive and rapid movement from the center
of the tank to the positive poleo To illustrate, positive
response (an entire time period of 60 seconds spent by. all
fish tested in the half of the tank adjacent to the positive
pole) was obtained v/ith an average current of 8 amps and an
on-fraction of Oo75ii with an average current of 5 amps and an
on -fraction of Oo^O, and with an average current of 3 to U
anps and an on-fraction of Oo25o The corresponding current
densities may be calculated at OoOOii?? Oo0029, and Oo002i;-
O0OOI8 amps/cm^, respectivelyo Ihus^ as the on-fraction of
a pulse was reduced, there was a decrease in the average
current require., to attract the fish and a corresponding sav-
ing in power o

In extending Morgan's workj two lines of investigation
suggested themselves, (1) to determine the relationship betwee::
source voltage and electrode size^ and (2) to further decrea^:e
the on-fraction to determine the minimal current which would
evoke positive response o

SOURCE V0LIAG3 AND ELSGTRODE SIZE

The wiring hook-up of the apparatus is shown in figure'
18 The source of power is an Onanj, ^-cylinder^ gas-driven,
air-cooled, direct current generator unit with a maximum out-
put of 5000 watts C2I08 amps at 230 volts) o The source

70

Q
O

cr
h- «^

u a:

UJ

_j

UJ

en

tu

Q.
X

UJ

<

o

UJ

UJ

UJ

X

Q
UJ
C/)

3

CO

Q.
<

Q
<

o
q:

c3

00

d
u_

71

voltage (maintained by Morgan at 220 volts) may be varied
between limits (about ^0 to 250 vclts) by means of a rheostat
in series with the generator field coilso The voltmeter (V-j_)
measures the source voltage „ The voltmeter (V^) measures
the voltage after interruption, an average value which is a
fraction (approximately the on-fraction) of the source or
peak voltage 0 It also measures the voltage at, but not between.,
the electrodes if it may be assumed that the resistance of the
rever sing-polarity switch and the lead wires to the electrodes
is negligibleo The ammeter (A), which may be placed anywhere
in the circuit, measures the average interrupted current (I),
which again is a fraction (approximately the on-fraction) jf
the source or peak current,, Ihe interrupter is a motcr-driven
disk which may be changed to give the desired "on-off" ratio o
The electrodes have unknown resistances (R^ and Rp) wluchj,
with tube-encased carbon electrodes, vary with the extent oi
exposure to seavrater and the extent of polarizing by gas
bubbles which are generated during a current flow, Ihe
electrodes are immersed at either end of the tank of seawater,
Ihe seawater has a resistance (R3) which may be calculated
roughly as ,p

R3 ^ k ~-^ Oo0^23 -^Y" " 0,3138 ohms J

where k is the specific resistance of seawater at a chlorinity
of 19 PoPofflo and a temperature of 2.5° Co, L is the length of
the water column (feet) and Ar., is its area (square feet).
The apparatus and hook-up described above are identical with
those used by Morgan^ except for the insertion of the voltmeter
(V2) across the line after interruption o

Of the various voltage measurements which might be made^,
the average voltage (V^) across the electrodes is the only one
of importance from the point of view of the reaction of the
fisho This cannot be measured directly without a special volt-
meter, but it may be calculated according to Ohm's Laws

V3 ^ I'>R3

knowing the average current (l) and the resistance of the
column of seawater (Rj) = "^^^s the following values may be
obtaineds

I (amps) V3 (volts)

1 o7^

2 O06J

3 0.9h
h lo26
5 lo57

etco

7?

The average current may be varied by changing the resist-
ance of the electrodes, by varying the source voltage and thus
the average interrupted voltage ^ or by both„ Morgan was able
to maintain the source CV]_) at 220 volts with increasing
average current (l) by decreasing the resistance of the elec-
trodeso This is illustrated by the following empirical deter-
minations wMch were made with his apparatus (delivering only
210 volts at the time)^ using tube-encased carbon electrodes,
the exposed surface of which was varied^ and using an on-
fraction of Oo25 at 15 r»p.SoS

V-L(volts) iCamps) V2Cvolts) R - Ir-Jl (ohms)

210

1

U8

US

210

2.5

50

20

210

h

52

13

The increase m average current was accomplished by
decreasing the resistance of the electrodes (increasing
their area of exposure), without materially changing either
the source or the interrupted voltage, Ihe calculated resist-
ance R is mainly that of the electrodes R-, and Rp,

To demonstrate that the same result could be obtained by
varying both the source voltage and the resistance of the
electrodes, empirical determinations of V]_ and V^ were made
at a constant average current (I) of 3 ampSs and therefore
at a constant voltage (Vj) across the electrodes of 0o9ii
volts o The resistance or the electrodes was gradually-
decreased by changing from small tube-encased carbon rods,
to small carbon rods not encased, to large carbon rods
wholly immersed, and finally to plates of galvanized iron
partially and wholly immersed. In each casej, the source
voltage (Vj.) was adjusted to give an average current (l) of
3 amps with an on-fraction of 0,25 at l5 ropoS.s

V^Cvolts) VgCvolts) R ^ Vg/I (ohms)

210 52 17,3

160 38 16

110 21 7

95 16.5 5o5

85 12 k

Since in each of the above determinations j the current
(3 am.ps) and the calculated voltage across the electrodes
(0,91; volts) remained the same^ the response of fish in the
tank should be identicalo If low source voltage is a
desirable feature, attempts should be made to procure non-

73

polarizing electrodes of low resistance^ in nearly perfect

electrodes R )■ Ri-f R2 — ^'O" In the present trials^ minimum

resistance was obtained with galvanized iron electrodes the
area of v*iich nearly equalled that of the cross section of
the tanko Other types of electrodes were not triedo It maj
be noted that in the above table j, ^2/^1 pi"ogi"es3ively departs
from an initial Oo25/l ratio| the reason is not clear-o

A series of experiments were planned with a source poten-
tial of 95 voltsj currents of 2 3 3^ and I4 amps^ and on-frac<=
tion Oo25s ^'t 15 ropoSoj to demonstrate that the reaction of
the fish was the same as at the higher source voltage (220)
used by Morgan, Unfortunately only two fish were available
at the time the apparatus was adjusted for this experimento
These were used in replicate trials (reversing the polarity
of the electrodes) at 2 ampso but they died before further
experiments could be performed due to failure of the aquarium
water supplyo Of the four trials total positive response
was evoked in three and partial positive response on oneo
Although the data are meager and therefore non-conclusive j
they are comparable with those of Morgan at the same currentj
on-fraction, and frequency but at 220 volts source j and indi-
cate the general validity of the reasoning in the preceding
paragraphs o

REDUCTION OF ON-FRACnON OF A CYCLE

IVro sets of experiments were tried with a very limited
available supply of large aholehole^ similar in size (9 to
13 centimeters) to those used by Morgan, In one/ Morgan's
revolving disk interrupter was used) in the other j, a specially-
designed plunging electrode interrupter was used. The gal-
vanized iron electrodes in the 12 x 2 x 1-foot tank almost
covered the cross-section of the water column and had a
water-exposed area of about 1700 square centimeters o The
fish were confined by two gates at the center of the tank.
When the current was turned on, both gates were removed
simultaneously so that the fish was free to move to either
the positive or the negative electrodeo

The results are included in table 11 0 For conparison
some of Morgan's results are included in table 12 o As
total response- — rapid and decisive movement to the positive
pole from which there is no withdrawal— was desired as the
criterion of attraction^ both Morgan's and the writer's
results are recorded as the number of positive trials an^.
number of non-positive trials j, rather than as time spent in
the positive half of the tank out of a total test period of
one minute.

7U

Revolving Disk I^terri^ter

By using a half brass., half bakelite disk and staggering
the brushes (fig, 19) j it was possible to attain an on-frac-
tion of Ool^l (off "Inaction, Oo8U8)o The on-fraction was
measured as the length of the disk circumference corresponding
to the '"on" position as determined with an open^closed circuit
indicator (ohmmeter),, divided hy the total disK circiimferenceo

For each of five specimens, two replicate trials (revers-
ing the polarity of the electrodes between trials) were made
at an average current of lo5 amps (Experiment A), and again
at an average current of 2 anps (Experiment B) with a frequency
in each case of 1^,8 ropoS„ (table 11), At 1,5 ampsp 8 out of
10 trials vrere positive. At 2 amps, all 10 trials were posi-
tive <= It may be seen by comparison of the results in table 11
and table 12 ^ tliat by reducing the on-fraction from Oo25 to
Ool^lj the current for positive response has been reduced frcm
U to 2 anpSj vathi a corresponding reduction in current density
from O0OO2U to 0,0012 amps/cm^o

It was impossible to further reduce the on-fraction
because of the thickness of the brusheso Moreover, even with
a relatively weak average current of 2 amps.^ there was consi-
derable arcing at the make and breaks wluch tended to bum
the bakelite and corrode the brass portion of the dLsko Fur-
ther experiments with the disk-type interrupter were not
undertaken.

Plunging-electrode Interrupter

To attain finer adjustment of the "on-off" ratio and in
the hope of reducing the g)ark at *-he make and break, a new
interrupter was designed (figo 20) <, This consisted essentially
of a revolvi.ng wheel, an eccentrically connected rods and an
electrode which iips momentarily into a glass jar containing
oil floating on mercuryo The jar sits on a platform which
may be raised or lowered, Ihus permitting fine adjustment of
the duration of contacto The wheel was driven at approximately

15 ToPoSo

The apparatus was not satisfactory. It vibrated consi-
derably, causing mechanical and electrical connections to
break, and more serious, causing irregular waves on the sur-
face of the mercury. Despite the use of transformer oil, and
later of pure mineral oil^ arcing occurred at the make and
break. Ihis gradually burned the tip of the electrode, chang-
ing the on-fraction. The oil and mercury soon tended to form
an "emulsion", which further changed the on-fraction and
caused arcing at the surface,

75

CARBON BRUSH

CARBON BRUSH

FIG. 19. THE REVOLVING DISK' MECHANICAL INTERRUPTER

76

m

o
o

u

CO

e

(D

u

o

-p
o

0)
•H

T) m

<D C
-P O
PU-H

g^
^ CO
0) u
+3 Cm
C a

•H C

o

o

^^

0 CO

iH
O

es

cn
O

m

■P
C
<D

U

o

CO

o

O (0

p. >

09

« c8

9

iH 031

Hi

en

o

u ,

9 -Pf

S -He

°

-H

J3-
00 -P

•P ►J

CO r

ra ^

si

o

o;

■P
cO|

O O O rH iHJCV O O O O O O rH iT^ C\J vD iH CvJ CNj|j»<>

CN2 C\i C\J H rHlCO

33

o o <>

H CM f^ -4- lA

»r»,

ni CNi oi cv oi o

o o
O O O CNi o
r-i rH ft r-{

rH Oi r^ vj- ir\

if\ »A irt

C\i (^ Oi
H iH iH

rH CM r^

^f^tO tOlrH
CM

W> lA lf\

^

rH r-i

T-i Oi cr\

c«^

c^

00

t»

o

o

o

00

»r\

o

rH
rH

CM

o

o

s

0)

<J

n

77

»n

o

».iH

o ^^^

m

" a

ft to

0 bo

u u

o

«^S

r-i

c% O

*> u

a vi

o

u Td

S-i (D

0 CO

o c

<D

•P TJ

© o

•H

Vi O

(0

•H

TJ

T3

fl

o

(0

TS W

p

<D a

w

-P o

a..H

g^

J< CO

<D »^

-P ^

C 11

•H C

O

O

P TS

c

(D CO

H

O

CO

s:

p

0)

a

iH

CD

O

u

CO

S

O

^

o m

3

0) o

CQ .H

C U

O CO

p, >

m

<D +»

CE!

OJ

^1

H »

«> Si

t ?'

O

o

u ,

(D a>

3 Pi

-P or

m X)
°H a

o)

0 -ri
00

-^
^

t?'^

>

OHOOOH OOOOOO OOOOOfO

OVJrHCVfVOiO est Ol <M Ol Olio CNJCVCNJCNJCVO

»f\ «> «> m

o o o o
CV CV I
rH H r

>0

a^i^P-^

H Oi r>^ -sj- iTk

O

rH -H

iH CV !^ -4- W\

o

o

rH -^rH H

3

rH CV f^ ■^'A

O

o

o

o

0)
r-{
X>

CO

(D +3'
W SI

to

>A

OQ

78

QJ o

». jH

o U\

o H

O

CO

»n no

O

+»
fl a

0) o

u u

U 4-1

o no

0}

■P ta
o C
p ffi

•H O

T3 O

O

T)

a>

+» o
Q, m

? O
h O

© +»

Q 0
-P C

o

O

(1)

t3

a
as

0)

-p
B
0)

3 -O
O 0)

a> d

(0 CO c;

_ o +J
a<-H c
m u o

© C8 O

I

o

(0

o

o

a

00

o
-p

CO

Jl3 o

00 e
c o

1-3

X 0);
ra /3

Si

o

1 +>

G O

o a

';;;f

iH r-l
> O,

3i

OOOiOOCM OiOCVOO-^-

CV CV O ?N cvlto O CV O C\i OifO

o
Cvi (H O
rH H iH

lA kA

ITV

in «>

a

0^

^

.-H H rH

rH CV <^ -^ VN.

O
lA

rH C\J r^ -sj- kO,

O

cv

O
Oi

rr\

o

cv

<V

S
H

Q) P
W Si

w

79

V BELT

CONNECTING
RODs.

ELECTRIC MOTOR

PLUNGING
^^' ELECTRODE

m[i

2z3

GUIDE
-RUBBER STOPPER
FIXED ELECTRODE
OIL

MERCURY
ADJUSTABLE PLATFORM

SCREW ADJUSTMENT

FIG. 20. THE " PLUNGING ELECTRODE MECHANICAL INTERRUPTER.

80

Two experiments (C and D of table 11) were performed using
fresh mineral oil floating on the mercuryj and an on-fraction
of O„083o At Oo5 amps, 12 out of 18 trials with three ahole-
hole were positiveo At 1 anps 21 out of 26 trials (81 percent)
were positive « The latter results are comparable with those
obtained by Morgan at 3 asaps with an on -fraction of Oo25
Ctable 12 -D), Unfortunately m the present experiments there
is no guarantee that the on-^fraction was maintained exactly
at O0O835 nor that the current was maintained exactly at loO
anpsc However, the results indicate an additional saving of
power by further reduction of the on-fractiono

DISCUSSION

The regression of minimal current for total (or near-
total) response on on-fraction is shown in figure 21 for both
Morgan's results and those of the writero The present results j
although meager^ indicate the continuance of the downward
trendo They shoiLLd be checked with more fish and better equip-
ment;, and extended to include still smaller on-fractionso

It may be noted that at 2 anps average current and IS
ropoSoj the on-period is Ool^l/l^ " OoOl seconds^ or 10 milli-
secondso It is interesting to note that according to Houston
(19U9)j DTo Konrad Kreutzer of Germany apparently used an on-
period of about 2 milliseconds, with a frequency which varied
(2 to 20 r,p,s„) depending on the natural wriggling frequency
of the particular fish. The present results are therefore in
agreement with those of Kreutzer in indicating the desirability
of using a short on-fraction„ It will be interesting to deter-
mine if 2 milliseconds is the optimum value for positive
attraction at minimal current o The present results do not
agree with those of the Cooperative California Sardine Research
program (Anon 19$0)j in which greater response was found with
a longer on-fractiono

It seems useless to conduct further experiments with the
present apparatus., Rather j efforts should be made to invent
or perfect an interrupter which is rapid in action^ free from
arcings and provides an adjustment for varying the "on~off"
ratio. This might be electronic or mechanicals, Both have
been used by Kreutzer (Anono 1951) hut in his work the elec-
tronic switch was to be replaced by an inpulse switchj developed
by Siemens Werke of Germanyj wPd-ch "is quite compact^ is com~
pletely mechanicalj and is said to have an accuracy of l/lOOjOOO
of a secondo"

81

/

in
a.

<

ui
a:

o

<

q:

LiJ

>

<

6-

5-

3-

/

/

/i

/

/

/

/

/

/

/

/

A

.005

O - MORGAN
• - TESTER

X)04

S
o

en
<

.003

z
o

z

LU
4002 QC

q:
o

.OOt

"ON -FRACTION"

FIG. 21. TREND OF AVERAGE CURRENT ( AND CURRENT DENSITY) WITH
CHANGE IN "ON -FRACTION" OF A CYCLE.

82

REFERENCES

Anon,

19$0. Electrical stimulation of fish in sea water „ Cal
Coopo Sardine Res„ Program, Prog, Rept. 1950s i;6-ii7o

Anon,

I95I0 Vessel equipped with deep-^sea electrical fishing
device, U, S„ Fish and Wildlife Serv, Comm„ Fisheries
Rev. 13(1)? 5>51i»

Houston, R, B., Jr,

19h9o German commercial electrical fishing deviceo U„ S,
Fish and midlife Serv„, Fishery Leaflet 3k8s I6 pp<,

Morgan, M, E„

1951 » The response of a tropical fish to interrupted
direct current and its application to the problems of
electro fishing in sea water, M, So thesis ^ University
of Hawaii s June 19^1 s 68 pp.

ii3

i11?'rmKLi-''?r'"v ■ se,

^ JIMiillliii
5 WHSE 01073