RESU LTS

3. 1 . Circular synmetr iness of the receptive fie ld of hor izontal cells

In the carp retina, two different L-type hor izontal cells can bee seen. Photop ic L­

type rece ives input mainly from red cones having spectral maximum at around 620 nm in wave length, while scotopic L-type rece ives input exclus ive ly from rods having spectral maximum at around 523 nm . �5)

Sample records of responses of photopic L-type and scotopic L-type hor izontal cells to the sequential flash of LED· components are . shown in Fig. 2. From the response am­

plitude for each LED component of the display, we can . see the spatial distr ibution of the decayed response. F ig. 3 shows the typical distridutions of 25 photopic L-type and 9 scotopic L-type in contour plot where distr ibuted numbers are response amplitude to

the flash of L E D components 'lt the po s ition, m units of - 0.02 mV fo r the photop ic L­

type (e . g. 70 denote - 1 .4 mV) and in units of - 0.05 mV fo r the s cotop ic L - type ₍e . g.

45 denote - 2. 25 mV)^. Contou r l ines we re ploted by increments of 0.3 mV fo r the phot­

opic L - type and 0.5 mV for the s cotop ic L - type, respective ly. The mark (x) ind icate s the pos i tion of record ing e lec trode t ip. The resu lts ind icate that the receptive field of L- type ho r i zontal c e l l s i s c i rcularly s ynme try a round the record ing s ite.

3. 2 . E ffe c ts of l ight intens ity on spatial decay of ho r i zontal c e l l re spon s e s

To know the who le profi le of the receptive field requ i r e s much t ime i n expe r iments even though we s t i mulate some d i s c re te s amp le po ints.

A

Photop i c L

Scotop i c L

Photop i c

·3

• 4

• 3 .,

__j

^{2 mV}

20 5

F i g . 2 . Samp l e record s o f L - type hor i zontal c e l l r e s p o n s e s t o t he seque n t i a l f l a s h o f l.E D component s .

fted l i ght and green l i ght we re used f o r the pho topic L- typc and s c o top i c L-type. r e s pe c t ive l y .

L B. Scot op i c L

• 3 • 2

• 3 • 2 • 3 • 3

• 3 . '

. ' .. • 4

·• • 9 . '

• 1

• 2 • 1

·3 ^·3

• 4 • 3

• 5 • 4

• 1

• 3

• 3 • 2

• 3 ^·5 • 5

• 4 • 6 •7 •5 • 6

•3 •6 •9 • 6 • 5

• 1 0 o 7 •5 • 5 • 9 • 6 • 7 • J • 4 • 6 • 6 o 6 o7 o 4 o 2 o 2

l mm · 2

F ig. 3 . Spat i a l d i s t r i bu t i on o f L- type ho r i zontal ce l l rdspo n s e s . R e s p o n s e amp l i tude for t h e LED d i s p l ay we re p l ot t ed i n contour p l o t ( s e e Text ) .

---• 2 l mm

Howeve r, the profile of the receptive fie ld in the radial direc tion i s suffic ient fo r the who le d i s t r ibution because of c i rcular s ynme tr ine s s of the receptive field. We u s ed a s mall te s t spot moving with a cons tant speed ac ro s s the receptive field of ho r izontal c e l l s pas s ing the tip of record ing e lec trode. Obta ined respon s e s we re s l ightly deviated

70

-S pectral and -S pat i a l C oding of H o l i zontal C e l l Responses i n Carp Ret i na Kenk i chi F U K U R OTANI and Junsaku Y O S HI D A

from the real profile by time factors

Photop i c ^L involved in response: hyperpolarizing

peak appeared after the spot passed the r ecording s ite ; decay of the hype rpola­

r iz ing phase before the peak was s teeper than that of repolarizing phase after the

M o v i ng spot Spot d iameter, 0.1 1 4 mm Moving speed. 0.50 mm/sec

peak; and often a rebound could be seen at the repolariz ing phase. However, it was a fas t, direct method to know the continuous profile of the receptive filed and made pos s ible to record responses

from a cell for stimul i of diffrent condi - ^{0 Log} tions.

Fig. 4 shows sample records of photo­

pic L-type hor izontal cell responses to

a spot of 0. 1 14 mm in diameter, 620 nm

I

^{0.5 m V}

in wave length, and 0.41 mm sec in movi­

ng speed. The upper record is for the l ight intens ity of - 1 Log, and the lower record is for 0 Log. To compare the receptive fields for di ffe rent l ight in­

tens ities, the responses were digitalized by an AID-converter, smoothed by a digit­

al filter, and normali zed as the peak

hy-Fig. 4 . Samp l e records of photop ic L-type hor i zontal ce l l responses to a moving spot.

perpolar ization reduced to -L F ig. 5 shows the normal ized responses of L-type hor izont al cells to the spot of diffe rent intens ities of monochromatic l ight . 620 nm was used for the photopic L-type and 520 nm was for scotopic L-type. From the results of F ig. 5, it is conc luded that the receptive field of L-type horizontal ce lls became broader as the br ighter light was used.

A P h o t o p i c L B. Scotopi c L

a. E

<(

Loc a t i o n of t h e> s p o t ( mm ) -3 -2 - ]

-²Log

;;!

"C ..

::J 50

a. E

<(

F ig. 5 . Normal i zed response s of L- type hor i zontal c e l l s .

Locat i on of the> spot ( mm )

Light spot was 0. 1 14 mm i n d i ameter and moved with a cons tant speed of 0.41 mm/sec at diame t r i c d i rect ion of the recept ive fie ld.

The same properties of the receptive field could be obse rved for C-type hori­

zontal cells. For example, Fig. 6 shows the normal ized re sponses of a RIG-type horizontal ce l l · to the spot of 500 nm in wave length.

Hartl ine reported the s imilar proper­

ties of gengl ion cells; receptive fie ld of gangl ion cells in the frog retina became broade r us ing br ighter l ight for the test

1221

flash .

3. 3 . Effects of l ight wave length on spatial decay of hor izontal cell re sponses F ig. 7 shows the normalized respons ­ es of L- type hor izontal cells for the l ight

R i G

a. E

4:

Loca t i on of t he spot ( mm )

-3 ^{- 2} - 1 ⁰ 1 2 3

I I I ^I I I I I I I I I I ^I

F ig. 6 . No rma l i zed responses of a RIG- type ho r i zontal ce l l .

L i ght spot was 0. 1 14 mm in d i ame t e r and moved w i t h a cons tant speed of 0 . 4 1 mml s e c .

spot of diffe rent wave lengthes. The l ight intens ity was 0 Log for the photopic L-type and - 1 Log for the scotopic L- type, respective ly. The results of F ig. 7 indicate that the receptive fie ld of L-type hor izontal cells became the broader as the more sensitive wave length was used for the l ight spot.

Fig. 8 shows responses of a RIG- type ho rizontal cell for the l ight spot of 700, 660, 620, 600, and 500 nm in wave length. The l ight intens ity of the monochromatic l ights was identica l ly 0 Log. For 700 nm, the response is depolariz ing all ove r the ve rtical sec­

tion of the receptive fie ld. For 660 nm, the depolarization is extending but a hype rpo­

lariz ing inflection can be seen near the recording s ite. The hype rpolar iz ing inflection grows up rap idly and ove rcomes the depolariz ing distribution, with dec rements of the wave ­ length from 620 to 500 nm. Because of the var iety of spectral re sponse prope rtie s amo­

ng RIG- type hor izontal cells, we could see apparent inflection for some RIG-type hori­

zontal ce l ls even in the re sponse to the l ight spot of 700 nm, while we could not any inflection in the re sponse, for some cells, to the l ight spot of 660 nm. W ith dec rements

A. P^{h o t}o^pic ^L B. Scotop i c L

0

0

<1>

"0 -::: :J 50

a. E

4:

1 00

L o c a t i o n of t h e spot ( mm ) Locat i on of t h e spot ( m m )

-3 -2 -1 ^{0 2} 3 -3 ^-2 -1 ^{0 1} 2 3

�

^\

I

�

_I ^/

I

I 500 n m

I ·. _{I I I}

/''"'

^{6 20 nm}

I I I 1

0

\

.. \

\�

Cl>

"0 :J 50 �

\'

· -i:i

�

4: E I

1 00

F ig. 7 . Norma l i zed respon s e s of L- type ho r i zontal c e l l s to a m ov i ng spot o f d i fferent wave l engthe s.

The spot d i ame t e r was 0 . 1 14 mm and the mov ing speed was 0 .4 1 mml s e c .

-72

-r)

-?--­

(J

I /I

,^; '/

I ·---__

!f---.--___ ^{6 60 n m} f/

�

^{620 n m}

/,�

^{570 n m}

'// ^{520 nm}

1/

Spect ral and Spat i a l C od i ng of Hol i zontal C e l l Responses m C a rp Ret i na Kenk i chi F U K U R OTANI and Junsaku Y O SH I D A

R IG

Mov i ng spot Spot d iameter. Mov i ng speed . 0.1 1 4 0 41

mm/sec mm

700nm �

�losmv

I

^{0.5 mV}

660nm

/\_

_,

^�

I

^{0.5 mV}

6 20nm� V

600nm

I

^{0.5 mV}

500n

m

^-�-�

₁

_{1 mV}

I I I I I I I l I I I l I I I I I I I

-4 -3 -2 -1 0 1 2 3 4 Locat i on of the spot (mm )

Fig. 8 . Spectral responses of a RIG-type ho r i zontal c e l l to a moving spot.

The responses were witten by a pen recorder.

V / RB

Mov i ng s pot Spot d i ameter, 0.1 14 mm Mov i ng speed, 0.4 1 mm/sec

620 nm ^{__ ___.}

600 nm

l

^{z m}

^v

570 nm

1

^{2 m V}

530nm

! zmv

500 nm

1

^{2 mV}

4GO "m

� ^J

^{2 mV}

I I I I I I I I I I I I I I I I I I I

-4 -3 -2 -1 0 1 2 3 4 Locat i on of the spot ( mm )

Fig. 9 . Spectral response s of a Y /RB-type hor i zontal c e l l to a mv ing spot.

of the wave length of the l ight spot, howeve r, the growing up of the hype rpolariz ing m­

flection in the response was commonly obse rved for all the R/G-type hor izontal cells m ­ vestiga ted i n this study.

S imilar spatial properties were also observed in Y /RB-type hor izontal cell respons­

es. F ig. 9 shows responses of a Y /RB-type horizontal cell to the moving l ight spot of 0 Log intens ity, and 620, 600, 570, 530, 500, 460 nm in wave length. The responses for o 620, 600, and 570 nm are depolariz ing; however, a s l ight inflection can be seen near the recording s ite. As the shorter wave length, 530, 500, 460 nm, was used for the light spot, the hype rpolariz ing inflection grows up and hides the depolariz ing response all over the receptive field.

These spatial properties of C-type horizontal cells were not inherent to the respon­

se only for the moving stimulus, but we could find them in the response to the fixed sti­

mulus. It is demonstrated in Fig. 10 that C-type hor izontal cell responses change the ir polarity for the LED images at the center and surround of the receptive fie ld. Re

spon-s e spon-s of the R I G - type ho r i zontal c e l l or the f i r spon-s t, 6 - th, 7-th and 14-th red L E D image spon-s wh ich i l luminate some parts of central reg ion of the receptive fie ld are hype rpo l a r i z ing, wh i l e r e s pons e s fo r othe r LED image s wh ich i l luminate some parts of sur round reg ion of the receptive fie ld are depo lar i.z ing. Respon s e s of the Y I R B - type ho r i zontal c e l l fo r the fi r s t and 6 - th green LED image s wh ich i l luminate some parts of central region of the receptive fie ld are hype rpo l a r i z ing, wh i l e re spon s e s fo r othe r L E D image s wh ich i l luminate some parts of sur round reg ion of the receptive fie ld are depo lar i z ing.

5 1 0

I I I I I I I RIG

Y/ RB

Spot numbe r

1 5 20 25 30

I I I I I I I I I I I I I I I I I ^{I 35} I ^I I ⁴⁰ I I I I ⁴⁵ I I

F i g . 10 . Respo n s e s of C - type ho r i zontal c e l l s to the sequen t i a l f l a s h of L E D componen ts ( s e e F i g. 1) . Red l ight and green l i ght was used for the R I G - t ype and Y / R B - type, respe c t ive ly.

C - type ho r i zontal c e l l respon s e s are compo sed of hype rpo lar i z ing and depo la ­ r i z ing components. The obta ined resu ­ lts of F igs. 8 and 9 indicate a spatial mode of inte ract ions between R and G components of R I G - type ho r i zontal c e l l respon s e s , and be tween Y and B compo­

nents o f Y I R B - type ho r i zontal c e l l r e s ­ pon s e s . A d i ffe rent mode of inte rac ­ tions we re found be tween R and Y com­

ponents of Y I R B - type ho r i zontal ce l l respons e s . F ig. 1 1 shows re spon s e s o f a Y I R B - type ho r i zontal ce l l to the mo ­ ving l ight spot of - 1 Log intens ity.

The re sponse i s hype rpo lar i z ing a l l ove r the receptive fie ld fo r 700 nm as we I I

as 460 nm. I n the contrary with the monopha s ic d i s tr ibut ion, the d i s t r ibut ion fo r 660 nm i s biphas ic and depo l a r i z ing inflection can be seen near the r e c o rd­

ing s i te . The depo l a r i z ing infle c t ion grows up and then ove rcome s the hype r ­ polar i z ing R component, with dec re men ­ ts of the wave length from 660 to 570 nm.

Y / R B

Mov i ng spot S po t d i ameter. 0.1 1 4 mm M o v i ng sp eed . 0.45 mm/sec

700nm�

�

^{1 m V}

660 nm�^----

�

^{1 mV}

57 0 nm

�

...__

�

^{2 m V}

460nm

I ^I I ^I I ^I I ^I I ^I I ^I I ^I I ^I I - 4 -3 -2 ^{- 1 0 1} 2 3 4

Loca t i on of the spc t ( mm )

F i g. _{1 1 .} Spe c t r a l responses o f a Y / R B- type ho r i zontal ce II to a mov i ng spot.

-74

-Spectral and Spat i al C od i ng of Hol i zontal C e l l Re sponses _m C arp Ret i na Kenki chi FUKU R O T A NI and Junsaku Y O SH I D A

It is interesting that the receptive fie ld of C-type hor izontal cells has center-surro­

und antagonism for some monochromatic l ights though it is not complete as that of bipo­

lar cells o r gangl ion cells. Furthe rmore, Rl G-type hor izontal cells show colour oppon­

ency only in the surround-response, while Y IRB-type hor izontal cells show colour oppon­

ency in both center-response and surround-response.

4. DI SCUSSION 4. 1 . Spatial spread of hor izontal cell re sponses

The spatial distr ibution of hype rpolariz ing response of L-type hor izontal ce lls was monophas ic and be ll-formed. Any inflections could not be found near the recording s ite.

The spatial distr ibution was c ircular synmetry around the recording s ite. Morphologi ­ cally, it is known that hor izontal cells have a long dendr ite extending to inner neuc lear

(33)

layer; the dendr itic field is apparently out of c ircular synmetriness . The obtained resu-lts of this s tudy indicate that the e lectrical coupling among adjacent hor izontal cells is so tight that the ir dendritic figure is hidden, and the e lectr ical coupl ing has no directio­

nal se lectivity. It was reported that the e lectr ical coupl ing among hor izontal cells of the tige r salamander was very tight in spite of the ir course distr ibution in the retina.

4. 2 . Spatial spread of input from s ingle type receptor

Photopic L-type hor izontal cells rece ive inputs mainly from red cones, and scotopic L-type hor izontal cells rece ive input exclus ive ly from rods in the carp. Therefore, the spat ial properties of L-type horizontal cells reflect the bas ic propertie s of the spread of input from s ingle type receptor among the e lectrically coupled hor izontal cells. The receptive field became brorder as stimulus intens ity increased or the more sens itive wave length was used. The results suggest that the spatial property of hor izontal cells rece iving input · from s ingle type receptor depends only upon the effective ly absorbed photons in the receptor; a princ iple of univar iance is also appl icable to the spatial pro­

perty.

4. 3 . Spectral and spatial interactions m C-type horizontal cell responses

C-type hor izontal cell responses are composes of depolariz ing and hype rpolarizing components from different type cones. RIG-type hor izontal ce lls of the carp rece ive hyperpolariz ing input from green cones and blue cone s, as we ll as depolariz ing input from red cones1181• As we saw in Fig. 8, hyperpolariz ing inflection which was the spread of green cone input grew up and cove red the spread of depolariz ing input from red cones, with dec rements of the stimulus wave length from 700 to 500 nm. If RIG-type hor izon­

tal cells directly rece ived depolariz ing input from red cone s as we ll as hyperpolariz ing input from green cones, and both inputs transmitted with the same attenuation in the row of RIG-type hor izontal cells, the spatial distr ibution of the RIG-type hor izontal cell response must be monophas ic having a maximal peak at the recording s ite irre spective of polarity of the response. The obtained results indicate that, howeve r, the spatial distr ibution of depolariz ing input from red cones diffe red from that of hype rpolariz ing input from green cones. The inputs independently obeyed to the spread prope rty: the receptive field of each cone input became broader as the corresponding cones absorbed

mo re photons and the inputs became la rge r m amp l i tude . F rom the se resu lts, it is con­

c luded that the inputs from red cone s and green cone s brung out pas s ing through d i ffe ­ rent pathways.

The inte rac t ions lead ing to the spectral re sponse prope r t i e s of ho r i zontal c e l l s pro­

bab ly mod i fy the spatial d i s tr ibution o f cone inputs. R I G - type ho r i zonta l c e l l s o f the tu rtle rece ive hype rpo l a r i z ing input d i rec tly from green cone s but depo l a r i z ing input from L- type ho r i zontal c e l l s via green cone s(211• A d i fferent conc lus ion was obta ined fo r R I G ­ type ho r i zontal c e l l s of the p ikeperch; the i r spec tral response p rope r t i e s we re p roduced

( l l

only by pos tsynaptic inte ract ions . We have no evidenc e s to d i fine the true me chan i s ms re spons ible fo r the spectral re spons e p rope r t i e s . Howeve r, the results of th i s s tudy ind i cate at least that R I G - type ho r i zontal c e l l s of the carp do not rece ive d i re c t inputs from both green cones and red cone s .

F rom the results of F igs. 9 and 1 1 , it 1 s a l s o sugge s ted that the three cone inputs brought in Y I R B - type ho r i zontal c e l l s pa s s ing through d i ffe rent pathways each othe r.

Fu rthe rmore, the spatial prope r t i e s o f C - type ho r i zontal c e l l s are we l l expla ined a s s u ­ ming that R I G - type hor i zontal c e l l s rece ive depolar i z ing input from photop ic L-type ho r i zontal c e l l s and Y I R B - type ho r i zontal c e l l s rece ive pola r i ty- inve r ting input from Rl G - type ho r i zontal c e l l s . Mo rpho log i c a l ly, ho r i zontal c e l l s are s trat i f i ed in some laye r s , photop i c L type, R I G type and Y I R B type from recepto r te rminal t o the affe rent d i re c

-• Cfl

twn . The a s s umed info rmation flow in the s e ho r i zontal c e l l s ag ree w i th the mo rpho lo-gical ar range ment.

S ome que s t ions s t i l l rema in though the spatial d i s tr ibution of C - type ho r i zontal c e l l re spon s e s a r e we l l expla ined b y the compo s i tion of the spat ial spread of d i ffe rent cone inputs. We could not any depo l a r i z ing infle c t ion in the spatial d i s tr ibut ions of R I G - type ho r i zontal c e l l respon s e s fo r green l ight ( F ig. 8) and Y I R B - type ho r i zontal c e l l re spon­

ses fo r blue l ight ( F igs. 9, 1 1 ) . If the cone inputs we re l inearly summed in the C - type ho r i zontal c e l l s , a depo lar i z ing infl e c t ion should appear in the spatial d i s tr ibut ion of the re spon s e s . The refore, some non l ine ar summat ion seems to mod i fy the spatial p rope r ­ t i e s of C - type ho r i zontal ce l l respon s e s .

4. CONCLU S I ONS

In the d i s c u s s ions about the re s u l ts' ,of th i s s tudy, the fo l low ing conc lus ions we re obta ined: 1 ) input from a s ingle type photo receptor spread late r a l ly in the row of ho r i ­ zontal c e l l s attenuat ing i n ampl itude w i th d i s tanc e ; 2) the spatial sp read of the recepto r input became broade r as the mo re photons abso rbed b y the photorec epto r; 3) the spatial d i s t r ibut ion of C - type ho r i zontal c e l l respons e s was compo sed of some components from d i ffe rent type photo recepto r s that spread independently in the receptive fie ld; 4) the spa­

tial spread of the photo receptor inputs into C - type ho r i zontal c e l l s wa s mod i f i Pd by some non ] inear inte ractions.

(2) -(20,>

In the p revious s tud i e s we examined the spec tral and tempo ral p rope r t i e s o f ho r i -zontal c e l l re spon s e s . Pu tting toge the r the r e s u lts of the p revious s tud i e s and th i s s tu­

dy, the ove rall p rope rties of ho r i zontal cell re spon s e s we re e luc idated. The role s of

-

76-S pectral and 76-S pat i a l C oding of Hol i zontal C e l l Re sponses i n Carp Ret i na Kenk i chi F U K U R O T A N I and Junsaku Y O S HI D A

hor izontal cells for spatial and temporal information process ing are low-pass filtering in the space and time dimens ions, respective ly. On the othe·r hand, the i r roles for spect­

ral information proces s ing are the colour coding in the mutual ly orthogonal tri-chroma­

tic components.

ACKNOWLEDGEMENT

The authors wish to expre s s the ir thanks to Mr. H. Hase for his he lp in the experi­

ments. One of the authors ₍K. Fukurotani₎ is also indebted to Prof. K. Hara of Yamagata University for his guidance and encouragements.

REFERENCES

1) Burkhardt, D. A. and Has s in, G. : J. Phys iol. 28 1 , 125 (1978) .

2) Fukurotani, K., Hara, K. and Oomura, Y. : Vis ion Res., 1 5, 1403 ( 1975) . 3) Fukurotani, K. and Hara, K. : Trans. IECE. Japan, 58 - C, 282 ( 1 975) . 4) Fukurotani, K. and Hara, K. : Trans. IECE. Japan, 58 - C, 632 ( 1 975) . 5) Fukurotani, K. and Hara, K. : Bioi. Cybernetics, 20, 1 ( 1 975) .

6) Fukurotani, K. and Hara, K. : Bioi. Cybernetics, 2 1 , 221 ( 1 976) . 7) Fukurotani, K. and Hara, K. : Trans. IECE. Japan, 59 - C, 347 ( 1976) . 8) Fukurotani, K. and Hara, K. : Trans. IECE. Japan, 59 - C, 396 ( 1 976) . 9) Fukurotani, K. and Hara, K. : Bioi. Cybe rnetics, 23, 149 ( 1 976) 10) Fukurotani, K. and Hara, K. : Trans. IECE. Japan, 6 1 - A, 794 ( 1978) . 1 1 ) Fukurotani, K. and Hara, K. : J. Phis iol. Soc. Japan, 40, 274 ( 1 978) .

12) Fukurotani, K., Hase, H., Umino, _0. and Hara, K. : Trans. IECE. Japan, 6 2 - A, 317 ( 1 979) . 13) Fukurotani, K. and Hara, K. : Trans. IECE. Japan, 59 - C, 818 ( 1 976) .

14) Fukurotani, K., Hase, H. and Hara, K. : Trans. IECE. Japan, 59 - C, 820 ( 1 976) . 15) Fukurotani, K., Hase, H. and Hara, K. : Trans. IECE. Japan, 59 - C, 590 ( 1 976) . 16) Fukurotani, K., Hase, H. and Hara, K. : Trans. IECE. Japan, 59 - C, 684 ( 1 976) . 17) Fukurotani, K., Hase, H. and Hara, K. : Trans. IECE. Japan, 60 - C, 445 ( 1 977) . 18) Fukurotani, K., Hase, H. and Hara, K. : Proc. International Union of Phys iological

S iences, 1 3, 247 ( 1 977) .

19) Fukurotani, Hase, H. and Hara, K. : Abstr. S ixth International Biophys ics Congress, 326 ( 1 978) .

20) Hara, K. Fukurotani K. and Oomura Y. : Proc. First Sympos ium on Testing and Identification of Nonl inear Systems, Calif. Inst. of Technology, 275 ( 1 975) . 21) Fuortes, M. G. F. and S imon, E. J. : J. Phys iol. 240, 177 ( 1974) .

22) Hartl ine, H. K. : Amr. J. Phys iol., 130, 690 ( 1 940) .

23) Hashimoto, Y., Kato, A., Inokuchi, M., and Watanabe, K. : Vis ion Re s. 1 6 , 25 ( 1 976) . 24) Kaneko, A. : J. Phys iol. 2 1 3, 95 ( 1971) .

25 ) Kaneko, A. and Yamada, M. : J. Phys iol., 227, 261 ( 1972 ) . 26 ) Lamb, T . D . : J . Phys iol., 263, 239 ( 1976 ) ,

27) Marmare l is, P. Z. and Naka, K. - I. : Biophys. J., 1 2, 1515 ( 1973 ) . 28 ) Marshall, L . M . and Werbl in, F . S . : J. Phys iol., 279, 321 ( 1978 ) .

ドキュメント内 富山大学工学部紀要 (ページ 73-98)