0042~6989190 $3.00+ 0.00
Viri~n Res. Vol. 30, No. 10, pp. 1517-1520, 1990 Printed in Great Britain. AI1rights resewed
Copyright 0 WJO Pe+mon
Press plc
RESEARCH NOTE BLUE LIGHT HAZARD IN RAT DIRK VAN NORREN’~* and PEIERSCHELLEKENS’ ‘F.C. Donders Institute of Ophthalmology, Utrecht Academic Hospital, P.O. Box 85500, 3508 GA Wtrecht and m0 Institute for Perception, P.O. Box 23, 3769 ZG Soesterberg, The Netherlands (Received 8 January 1990, in revised form 28 Fehaary 1990) Abstract-Rats have been extensively used in light damage studies. Retinal damage threshold for white light were found at I-10 J/cm2, and the action spectrum resembled the absorptionwrn of visual pigment. We wished to answer the question whether a different class of light damage, the “blue light hazard“, with white light damage thresholds at about 300 J/cm*, and an action spectrum peaking in the ultra-violet, could also be d~onst~t~ in rat, To that purpose 5 deg pat&a of retina were exposed to white xenon tight with exposure times between 10 se-c and 1 hr. We found that for funduscopic thrashold damage the product of irradiance and exposure time was constant at a level of 315 J/cm2. Thereafter. the action spectrum was measured by exposing rat eyes to narrow band spectml lights. Threshold irradiant dose ranged from 4 J/cm2 at 379 mn to 2000 J/cm2 at 559 nm. Thus, susceptibility for damage sharply increased towards the ultra-violet, just like in earlier monkey studies. We conclude that in similar experimental conditions susceptibility to photic injury in rat is comparable to that in primatea. Rat is the fvst species for which two different action spectra of phot~hem~l damage have been ~~bti~~. Light damage
Rat
Action spectrum
Funduscopy
Retina
has been extensively used as an animal model for the study of retinal light damage. In the first of such studies Noell, Walker, Kang and Berman (1366) established that continuous exposure for 1 or 2 days to moderate light levels could cause extensive damage to photoreceptors in albino rat. The irradiant dose (product of irradiance and exposure time) for threshold damage by broad band green light in Noell’s study was about 5 J/cm2 (estimate by Kremers & van Norren, 1988). White light (retinal) thresholds were later found in the range I-10 J/cm* for pigmented as well as albino rats (Rapp & Williams, 1980). The action spectrum of retinal sensitivity to threshold damage was similar to the absorption spectrum of the visual pigment rhodopsin (Noel1 et al., 1966; Williams & Howell, 1983). In monkeys a different pattern of light damage was found after brief, intense exposures. Histology showed first signs of damage in the retinal pigment epithelium, threshold damage The rat
dose for white light was about 300 J/cm2, and the retinal damage action spectrum increased towards the ultra-violet (u.v.) (Ham, Mueller & Sliney, 1976; Ham, Mueller, Ruffolo & DuPont Guerry, 1980, 1982). With the crystalline lens in place the action spectrum effectively peaks in the blue part of the spectrum, hence this type of damage is often called the “blue light hazard”. Thus, although mechanisms are complicated and far from fully understood (Noell, 1980; Lawwill, 1982) two main classes of light damage can be distinguished (Mainster, 1987; Kremers & van Norren, 1988). The validity of the class distinction across animal models, however, is hampered by the fact that with relatively few exceptions, rats were studied in conditions evoking damage of the Noel1 class, and primates in conditions of the Ham class. The obvious thing to do is to look whether rats may acquire Ham class damage, and whether primates may be subject to Noel1 class damage. In this study we will answer the first question in the affirmative.
METHODS *To whom all correspondence should be addressed at: IZF-TNO, P.O. Box 23, 3769 ZG Soesterberg, The Netherlands.
Pigmented rats of the WEK/U strain were obtained from the Utrecht University animal
1517
Research Note
1518
facilities and kept at 12 hr light-dark cycle (IO-50 Ix at cage). Animals were sedated with ether, and anaesthetized with a single dose of Hypnorm 0.5 ml/kg i.m. and atropine 0.1 mg/ kg; Valium was added 5 mg/kg i.p. Pupiis were dilated with a drop of cyclopentolate HCl O.S%, and phenylephrine 2.5%. To prevent drying, the cornea was regularly irrigated with saline. Body temperature was maintained between 37.538.O”C with a rectal thermometer controlled electrical blanket. A xenon coagulator, incorporating Maxwellian view output optics, was used for light exposures. The 5 mm output pupil always fitted in the animal’s dilated pupil. Retinal field width was 5 deg. Infra-red light was blocked by two 3 mm KG3 heat filters (Schott, Mainz, W. Germany). Output was calibrated with a radiant exposure meter with a spectrally flat measuring head. Retinal irradiance level was calculated (Calkins, Hochheimer & D’anna, 1980) with radius of the eye 0.4 cm, and index of refraction 1.33. With rat lenses virtually transparent (Alpern, Fulton & Baker, 1987) transmission was taken as 0.9 for all wavelengths. Light level was controlled with neutral density filters. Interference filters with 10 nm bandwidth were used in determining the damage action spectrum.
1,
102
,>
103 irrodionce lmW/cm’I
-1
104
Fig. 1. Time vs retinal irradiance results of funduswpic observations after irradiation of rat tina with white xenon light. A straight line with slope - 1 was&ted by eye through the data points optimally sepa&ag sub from suprathreshold data. The tine represents a dose of 315 J/c&.
narrow band spectral lights, each at the maximum available level. Threshold irradiant dose ranged from 4 J/cm2 at 379 nm to 2000 J/cm2 at 559 nm (Fig. 2). The action spectrum peaks in the near U.V. and therefore typically resembles more Ham et al.‘s (1982) monkey spectrum than Noel1 et al.‘s (1966) rat visual pigment type spectrum.
RJ!XJLTS
We first established the threshold irradiance for white light with exposures between 1Osec and 1 hr. Within one experimental session 2-4 spots, positioned around the optic disk, were irradiated, generally at a single light level with varying exposure times. Positions of the spots with respect to major blood vessels were noted. Two to three days after irradiation damage was assessed by funduscopy. Threshold damage was defined as a just visible whitish spot. In suprathreshold damage the change in fundus colour was easily visible. In Fig. 1 results obtained on 7 eyes of 5 rats are summarized. A straight line with slope - 1 could be drawn through the data points separating sub- from supra-threshold damage. We conclude that, for threshold damage, reciprocity between irradiancc and exposure time holds over two orders of magnitude. the irradiant threshold dose for white light is 315 J/cm*, close to the 230 J/cm2 found in monkey with same light source and damage criterion (Kremers & van Norren, 1989). Thereafter we measured the action spectrum by exposing the rat eyes to various durations of
DlscussrnN
Apparently, when rats are studied in exposure conditions mimicking those generally usud in monkeys and when the same criterion for damage is used, properties of photochemical damage in both species are similar with respect to action spectrum, threshold irradiant ~dose, and reciprocity of irradiance and exposure time. The notion that monkeys have much l&her thresholds to photic injury than rats (Tso, 1989) is thus not true in a general sense. Con&i&g literature data with the present results, rat is the first animal for which two action spectra for light damage are now available. Another indication that two classes of damw may be p-t in rat was recently obtained when exposittin to U.V. resulted in damage to the retinal piemeat epitheliuxn, and exposition to visible light in damage to the photoreceptors (Rapp, Dhindsa & Tolman, 1989). With regard to monkeys Sykes, Robinson, Waxier and Kubawara -(I98 1) demonstrated threshold damage in photoreceptors after long exposures to fluorescent
Research Note
1519
100
c-4,
-
0
S&b-thmhold
@
fhnahold
l
Swo-
X
Ham ll976,1982t
tlmshotd
lo-'
:
E ?
z 2 .a r
10-o
z :: d
40-'
10-4 300
400
500
600
Wavelength Inm) Fig. 2. Dose for fundumopic lesions as a function of wavelength obtained on 17 eyes of 10 rats. By eye a line was drawn through estimated threshold doses. Ham et al.% data were obtained with JO and 100 see exposures (Ham et al., I!?76 for wavele~hs >4UOnm; Ham et ai. 1982 for those ~400 nm).
light, indicative for injury of the Noel1 class. These findings emphasise the validity of the distinction between two classes of photic injury across many animal species. A visual pigmentlike action spectrum in primates is still missing, though. Although similarities between rats and monkeys stand out in ‘the present study, differences can be noted too. Rats are somewhat less sensitive to damage than monkeys at most wavelengths, except below 400 nm; the two curves thus seem to cross. If this could be confirmed in a more extensive experiment it would be indicative for a multi-pigment cause of the blue light hazard (Sliney, 1988). At short wavelengths the rat lens is more transparent than that of monkey (Alpern et al., 1987). This makes the rat retina rather vulnerable to damage by near U.V. light. The threshold dose for white light will depend on the spectral composition of the light source; with common laboratory sources as fluorescent and halogen light substantial differences may be VR30/I&-I
expected with the value reported here for xenon light (Sliney, 1984). Also, on funduscopy in monkeys U.V.damage had a character different from that at visible wavelengths in that it showed imm~iately after exposure, and on histology (of a single eye) photoreceptors seemed most susceptible (Ham et al., 1982). In rats we never observed damage directly after exposure. In conclusion, although many details await further elucidation, we have demonstrated that rat is vulnerable to the blue light hazard. Acknowledgements-We thank Drs Jan Kremers and Hans Vos for helpful discussions.
REFERENCES Alpem, M., Fulton, A. B. & Baker, B. N. (1987). “Selfscreening’* of rhodopsin in rod outer qments. Vision Research, 27, 1459-1470. Calkins, J. L., Hochhehner, B. F. & D’anna, S. A. (1980). Potential hazards from specitic oph~lm~ devices. F&ion Research, 20, 1039-1053.
1520
Research Note
Ham, W. T., Mueller, H. A. & Sliney, D. H. (1976). Retinal sensitivity to damage from short wavelength light. Nature, 260, 153-155. Ham, W. T., Mueller, H. A., Ruffolo, J. J. &DuPont Guerry, R. K. (1980). Solar retinopathy as a function of wavelength. In Williams, T. P. & Baker, B. N. (Eds.), The e&crs of constant light on visual processes (pp. 319-346). New York: Plenum Press. Ham, W. T., Mueller, H. A., Ruffolo, J. J. & Dupont Guerry, R. K. (1982). Action spectrum for retinal injury from near-ultraviolet radiation in the aphakic monkey. American Journal of Ophthalmology, 93, 299-306. Kremers, J. J. M. & van Norren, D. (1988). Two classes of photochemical damage of the retina. Lasers and Light in Ophthalmology, 2, 41-52. Kremers, J. J. M. & van Norren, D. (1989). Retinal damage in macaque after white light exposures lasting ten minutes to twelve hours. Investigative Ophthalmology and Visual Science, 30, 1032-1040. Lawwill, T. (1982). Three major pathological processes caused by light in the primate retina: A search for mechanisms. Transactions American Ophthalmological So ciety, 80, 517-559. Mainster, M. A. (1987). Light and macular degeneration: A biophysical and clinical perspective. Eye, I, 304-310. Noell, W. K. (1980). Possible mechanisms of photoreceptor damage by light in mammalian eyes. Vision Research, 20, 1163-I 171.
Noel], W. K., Walker, V. S., Kang, B. S. & Berman, S. (1966). Retinal damage by light in rats. Investigative Ophthalmology, 5, 450-473. Rapp, L. M. % Williams, T. P. (1980). A parametric study of retinal light damage in aJbino and pigmented rats. In Williams, T. P. & Baker, B. N. (Eds.), The effect of constant light on viwalprocesses (pp. 135-195). New York: Plenum Press. Rapp, L. M., Dhindsa, H. S. & Tolman, B. L. (1989). There are at least three different kinds of retinal light damage in rat. Investigative Ophthalmology and Visual Science (Suppl.), 30, 461. Sliney, D. H. (1984). Quantifying retinal irtadiance levels in light damage experiments. Current Eye Research, 3, 175-178. Sliney, D. H. (1988). New chromophores for ophthalmic laser surgery. Lasers and Light in Ophthalmalogy, 253-61. Sykes, S. M., Robinson, W. G., WaxJer, M. & Kubawara, T. (198 1). Damage to the monkey retina by broadspectrum fluorescent light. Investigative Ophthalmology and Visual Science, 20, 424-434. Tso, M. 0. (1989). Experiments on visual cells by nature and man: In search of treatment for photoreceptor degeneration. Investigative Ophthalmology and Visual Science, 30, 2430-2454. Williams, T. P. & Howell, W. L. (1983). Action spectrum of retinal light-damage. Investigative Ophthalmalagy and Visual Science, 24, 285-287.
Viri~n Res. Vol. 30, No. 10, pp. 1517-1520, 1990 Printed in Great Britain. AI1rights resewed
Copyright 0 WJO Pe+mon
Press plc
RESEARCH NOTE BLUE LIGHT HAZARD IN RAT DIRK VAN NORREN’~* and PEIERSCHELLEKENS’ ‘F.C. Donders Institute of Ophthalmology, Utrecht Academic Hospital, P.O. Box 85500, 3508 GA Wtrecht and m0 Institute for Perception, P.O. Box 23, 3769 ZG Soesterberg, The Netherlands (Received 8 January 1990, in revised form 28 Fehaary 1990) Abstract-Rats have been extensively used in light damage studies. Retinal damage threshold for white light were found at I-10 J/cm2, and the action spectrum resembled the absorptionwrn of visual pigment. We wished to answer the question whether a different class of light damage, the “blue light hazard“, with white light damage thresholds at about 300 J/cm*, and an action spectrum peaking in the ultra-violet, could also be d~onst~t~ in rat, To that purpose 5 deg pat&a of retina were exposed to white xenon tight with exposure times between 10 se-c and 1 hr. We found that for funduscopic thrashold damage the product of irradiance and exposure time was constant at a level of 315 J/cm2. Thereafter. the action spectrum was measured by exposing rat eyes to narrow band spectml lights. Threshold irradiant dose ranged from 4 J/cm2 at 379 mn to 2000 J/cm2 at 559 nm. Thus, susceptibility for damage sharply increased towards the ultra-violet, just like in earlier monkey studies. We conclude that in similar experimental conditions susceptibility to photic injury in rat is comparable to that in primatea. Rat is the fvst species for which two different action spectra of phot~hem~l damage have been ~~bti~~. Light damage
Rat
Action spectrum
Funduscopy
Retina
has been extensively used as an animal model for the study of retinal light damage. In the first of such studies Noell, Walker, Kang and Berman (1366) established that continuous exposure for 1 or 2 days to moderate light levels could cause extensive damage to photoreceptors in albino rat. The irradiant dose (product of irradiance and exposure time) for threshold damage by broad band green light in Noell’s study was about 5 J/cm2 (estimate by Kremers & van Norren, 1988). White light (retinal) thresholds were later found in the range I-10 J/cm* for pigmented as well as albino rats (Rapp & Williams, 1980). The action spectrum of retinal sensitivity to threshold damage was similar to the absorption spectrum of the visual pigment rhodopsin (Noel1 et al., 1966; Williams & Howell, 1983). In monkeys a different pattern of light damage was found after brief, intense exposures. Histology showed first signs of damage in the retinal pigment epithelium, threshold damage The rat
dose for white light was about 300 J/cm2, and the retinal damage action spectrum increased towards the ultra-violet (u.v.) (Ham, Mueller & Sliney, 1976; Ham, Mueller, Ruffolo & DuPont Guerry, 1980, 1982). With the crystalline lens in place the action spectrum effectively peaks in the blue part of the spectrum, hence this type of damage is often called the “blue light hazard”. Thus, although mechanisms are complicated and far from fully understood (Noell, 1980; Lawwill, 1982) two main classes of light damage can be distinguished (Mainster, 1987; Kremers & van Norren, 1988). The validity of the class distinction across animal models, however, is hampered by the fact that with relatively few exceptions, rats were studied in conditions evoking damage of the Noel1 class, and primates in conditions of the Ham class. The obvious thing to do is to look whether rats may acquire Ham class damage, and whether primates may be subject to Noel1 class damage. In this study we will answer the first question in the affirmative.
METHODS *To whom all correspondence should be addressed at: IZF-TNO, P.O. Box 23, 3769 ZG Soesterberg, The Netherlands.
Pigmented rats of the WEK/U strain were obtained from the Utrecht University animal
1517
Research Note
1518
facilities and kept at 12 hr light-dark cycle (IO-50 Ix at cage). Animals were sedated with ether, and anaesthetized with a single dose of Hypnorm 0.5 ml/kg i.m. and atropine 0.1 mg/ kg; Valium was added 5 mg/kg i.p. Pupiis were dilated with a drop of cyclopentolate HCl O.S%, and phenylephrine 2.5%. To prevent drying, the cornea was regularly irrigated with saline. Body temperature was maintained between 37.538.O”C with a rectal thermometer controlled electrical blanket. A xenon coagulator, incorporating Maxwellian view output optics, was used for light exposures. The 5 mm output pupil always fitted in the animal’s dilated pupil. Retinal field width was 5 deg. Infra-red light was blocked by two 3 mm KG3 heat filters (Schott, Mainz, W. Germany). Output was calibrated with a radiant exposure meter with a spectrally flat measuring head. Retinal irradiance level was calculated (Calkins, Hochheimer & D’anna, 1980) with radius of the eye 0.4 cm, and index of refraction 1.33. With rat lenses virtually transparent (Alpern, Fulton & Baker, 1987) transmission was taken as 0.9 for all wavelengths. Light level was controlled with neutral density filters. Interference filters with 10 nm bandwidth were used in determining the damage action spectrum.
1,
102
,>
103 irrodionce lmW/cm’I
-1
104
Fig. 1. Time vs retinal irradiance results of funduswpic observations after irradiation of rat tina with white xenon light. A straight line with slope - 1 was&ted by eye through the data points optimally sepa&ag sub from suprathreshold data. The tine represents a dose of 315 J/c&.
narrow band spectral lights, each at the maximum available level. Threshold irradiant dose ranged from 4 J/cm2 at 379 nm to 2000 J/cm2 at 559 nm (Fig. 2). The action spectrum peaks in the near U.V. and therefore typically resembles more Ham et al.‘s (1982) monkey spectrum than Noel1 et al.‘s (1966) rat visual pigment type spectrum.
RJ!XJLTS
We first established the threshold irradiance for white light with exposures between 1Osec and 1 hr. Within one experimental session 2-4 spots, positioned around the optic disk, were irradiated, generally at a single light level with varying exposure times. Positions of the spots with respect to major blood vessels were noted. Two to three days after irradiation damage was assessed by funduscopy. Threshold damage was defined as a just visible whitish spot. In suprathreshold damage the change in fundus colour was easily visible. In Fig. 1 results obtained on 7 eyes of 5 rats are summarized. A straight line with slope - 1 could be drawn through the data points separating sub- from supra-threshold damage. We conclude that, for threshold damage, reciprocity between irradiancc and exposure time holds over two orders of magnitude. the irradiant threshold dose for white light is 315 J/cm*, close to the 230 J/cm2 found in monkey with same light source and damage criterion (Kremers & van Norren, 1989). Thereafter we measured the action spectrum by exposing the rat eyes to various durations of
DlscussrnN
Apparently, when rats are studied in exposure conditions mimicking those generally usud in monkeys and when the same criterion for damage is used, properties of photochemical damage in both species are similar with respect to action spectrum, threshold irradiant ~dose, and reciprocity of irradiance and exposure time. The notion that monkeys have much l&her thresholds to photic injury than rats (Tso, 1989) is thus not true in a general sense. Con&i&g literature data with the present results, rat is the first animal for which two action spectra for light damage are now available. Another indication that two classes of damw may be p-t in rat was recently obtained when exposittin to U.V. resulted in damage to the retinal piemeat epitheliuxn, and exposition to visible light in damage to the photoreceptors (Rapp, Dhindsa & Tolman, 1989). With regard to monkeys Sykes, Robinson, Waxier and Kubawara -(I98 1) demonstrated threshold damage in photoreceptors after long exposures to fluorescent
Research Note
1519
100
c-4,
-
0
S&b-thmhold
@
fhnahold
l
Swo-
X
Ham ll976,1982t
tlmshotd
lo-'
:
E ?
z 2 .a r
10-o
z :: d
40-'
10-4 300
400
500
600
Wavelength Inm) Fig. 2. Dose for fundumopic lesions as a function of wavelength obtained on 17 eyes of 10 rats. By eye a line was drawn through estimated threshold doses. Ham et al.% data were obtained with JO and 100 see exposures (Ham et al., I!?76 for wavele~hs >4UOnm; Ham et ai. 1982 for those ~400 nm).
light, indicative for injury of the Noel1 class. These findings emphasise the validity of the distinction between two classes of photic injury across many animal species. A visual pigmentlike action spectrum in primates is still missing, though. Although similarities between rats and monkeys stand out in ‘the present study, differences can be noted too. Rats are somewhat less sensitive to damage than monkeys at most wavelengths, except below 400 nm; the two curves thus seem to cross. If this could be confirmed in a more extensive experiment it would be indicative for a multi-pigment cause of the blue light hazard (Sliney, 1988). At short wavelengths the rat lens is more transparent than that of monkey (Alpern et al., 1987). This makes the rat retina rather vulnerable to damage by near U.V. light. The threshold dose for white light will depend on the spectral composition of the light source; with common laboratory sources as fluorescent and halogen light substantial differences may be VR30/I&-I
expected with the value reported here for xenon light (Sliney, 1984). Also, on funduscopy in monkeys U.V.damage had a character different from that at visible wavelengths in that it showed imm~iately after exposure, and on histology (of a single eye) photoreceptors seemed most susceptible (Ham et al., 1982). In rats we never observed damage directly after exposure. In conclusion, although many details await further elucidation, we have demonstrated that rat is vulnerable to the blue light hazard. Acknowledgements-We thank Drs Jan Kremers and Hans Vos for helpful discussions.
REFERENCES Alpem, M., Fulton, A. B. & Baker, B. N. (1987). “Selfscreening’* of rhodopsin in rod outer qments. Vision Research, 27, 1459-1470. Calkins, J. L., Hochhehner, B. F. & D’anna, S. A. (1980). Potential hazards from specitic oph~lm~ devices. F&ion Research, 20, 1039-1053.
1520
Research Note
Ham, W. T., Mueller, H. A. & Sliney, D. H. (1976). Retinal sensitivity to damage from short wavelength light. Nature, 260, 153-155. Ham, W. T., Mueller, H. A., Ruffolo, J. J. &DuPont Guerry, R. K. (1980). Solar retinopathy as a function of wavelength. In Williams, T. P. & Baker, B. N. (Eds.), The e&crs of constant light on visual processes (pp. 319-346). New York: Plenum Press. Ham, W. T., Mueller, H. A., Ruffolo, J. J. & Dupont Guerry, R. K. (1982). Action spectrum for retinal injury from near-ultraviolet radiation in the aphakic monkey. American Journal of Ophthalmology, 93, 299-306. Kremers, J. J. M. & van Norren, D. (1988). Two classes of photochemical damage of the retina. Lasers and Light in Ophthalmology, 2, 41-52. Kremers, J. J. M. & van Norren, D. (1989). Retinal damage in macaque after white light exposures lasting ten minutes to twelve hours. Investigative Ophthalmology and Visual Science, 30, 1032-1040. Lawwill, T. (1982). Three major pathological processes caused by light in the primate retina: A search for mechanisms. Transactions American Ophthalmological So ciety, 80, 517-559. Mainster, M. A. (1987). Light and macular degeneration: A biophysical and clinical perspective. Eye, I, 304-310. Noell, W. K. (1980). Possible mechanisms of photoreceptor damage by light in mammalian eyes. Vision Research, 20, 1163-I 171.
Noel], W. K., Walker, V. S., Kang, B. S. & Berman, S. (1966). Retinal damage by light in rats. Investigative Ophthalmology, 5, 450-473. Rapp, L. M. % Williams, T. P. (1980). A parametric study of retinal light damage in aJbino and pigmented rats. In Williams, T. P. & Baker, B. N. (Eds.), The effect of constant light on viwalprocesses (pp. 135-195). New York: Plenum Press. Rapp, L. M., Dhindsa, H. S. & Tolman, B. L. (1989). There are at least three different kinds of retinal light damage in rat. Investigative Ophthalmology and Visual Science (Suppl.), 30, 461. Sliney, D. H. (1984). Quantifying retinal irtadiance levels in light damage experiments. Current Eye Research, 3, 175-178. Sliney, D. H. (1988). New chromophores for ophthalmic laser surgery. Lasers and Light in Ophthalmalogy, 253-61. Sykes, S. M., Robinson, W. G., WaxJer, M. & Kubawara, T. (198 1). Damage to the monkey retina by broadspectrum fluorescent light. Investigative Ophthalmology and Visual Science, 20, 424-434. Tso, M. 0. (1989). Experiments on visual cells by nature and man: In search of treatment for photoreceptor degeneration. Investigative Ophthalmology and Visual Science, 30, 2430-2454. Williams, T. P. & Howell, W. L. (1983). Action spectrum of retinal light-damage. Investigative Ophthalmalagy and Visual Science, 24, 285-287.
