The Daily Rhythm of Shedding and Degradation of Rod and Cone Outer Segment Membranes in the Chick Retina

The daily rhythm of shedding and degradation of rod and cone outer segment membranes in the chick retina

Richard W. Young

Newly hatched chickens were maintained on a daily light cycle of 12 hr of light and 12 hr of darkness for 12 days. The pigment epithelium was then examined by electron microscopy at different times of day. Shortly after the beginning of the light period, the rods discarded groups of outer segment membranes. During the remainder of the light period, the membranes were degraded by the pigment epithelium. Early in the dark period, the cones shed membranes, which were digested by the pigment epithelial cells during the subsequent hours of darkness. Available evidence suggests that at least some of the chemical activities of the visual cells and pigment epithelium oscillate with a daily rhythm, tohich is synchronized with the daily fluctuation of light in the environment.

Key words: visual cells, daily rhythms, renewal, retina, membranes

The continual renewal of the light-sensitive steps of membrane assembly. Groups of old membranes of rod outer segments has been membranes are shed intermittently from the amply documented in a wide selection of ver- tips of the rods. This tends to occur early in tebrate animals. In fact, the process has been the morning, in animals maintained on a observed wherever it has been sought—in cycle of 12 hr of light and 12 hr of darkness.3 amphibia, reptiles, birds, fish, and mammals. The adjacent pigment epithelial cells, which It seems to be a fundamental and universal envelop the ends of the outer segments, in- characteristic of rod visual cells.1 gest and degrade the detached membranes.4 The basic mechanism of the renewal pro- In contrast, progress in our understanding cess in rods is "membrane replacement," of the mechanism by which the other class of which has two components: formation of new visual cells, the cones, renew their outer membranes and disposal of old ones.2 The segments had been frustratingly slow. This new membrane constituents are produced in deficiency is even more disturbing when we the inner segment of the cell, then trans- acknowledge that cones play a more impor- ported to the outer segment for the final tant role in meeting human visual needs than do rods. At one time, it seemed as if cones did not From the Department of Anatomy and Jules Stein Eye produce new membranes, but the au- Institute, University of California at Los Angeles. toradiographic evidence on which this con- This research was supported by grants EY 00095 and EY clusion was based later was found to have 0444 from the United States Public Health Service. been misinterpreted.1' 5 Similarly, it ap- Submitted for publication June 1, 1977. peared at first that cones did not shed mem- Reprint requests: Dr. Richard W. Young, Department of Anatomy, School of Medicine, University of Califor- branes. This too proved to be incorrect, nia, Los Angeles, California, 90024. when evidence of disc shedding by cones

Downloaded from iovs.arvojournals.org on 09/30/2021 Invest. Ophthalmol. Visual Sci. 106 Young February 1978

100 r

75 X < 50

25 A 10 12 14 16 18 20 22 24 TIME OF DAY

Fig. 1. Changes in the number of phagosomes in pigment epithelial cells of the chick retina at different times of day, expressed as a percentage of the maximum observed concentration. Time of day is given in hours, counting from the onset of the light period. Phagosomes recorded during the light period were derived from rod outer segments; those observed during the dark period were from cones. The shaded curves (filled circles and solid line) represent freshly detached membranes (stage I, Table I). The other curves (triangles and dashed lines) represent older phagosomes, showing signs of degradation (stages II to IV, Table I). Thus the shaded curve indicates the approximate time and duration of membrane detachment and phagocytosis, and the dashed line indicates the period of membrane digestion. Note that although a few cones discard membranes throughout the dark period, the major phase of membrane shedding by cones occurs at the beginning of the dark period, 12 hr after the phase of synchronized shedding of rod membranes.

was detected in humans and squirrels.6' 7 wise, their outer segments would soon disap- Nevertheless, no synthesis emerged to tie pear. the sparse bits of information into a coherent This suggested a simple hypothesis: Rods hypothesis of cone outer segment renewal. and cones both renew their outer segments Recently, in considering the implications by membrane replacement. The process fol- of LaVail's report3 that rat rods shed mem- lows a daily rhythm, and the rhythms of rods branes early in the daily light period, it oc- and cones are separated in time by approxi- curred to me that cones might also dispose of mately 12 hr. outer segment membranes at a certain time To learn if renewal by membrane replace- of day. Accordingly, O'Day and I8 looked for ment was unique to goldfish cones or was a evidence of this process in goldfish which more fundamental process which also occurs were on a daily cycle of 12 hr of light and 12 in other vertebrate animals, I repeated the hr of darkness. experiment in a diurnal lizard which con- Shortly after the beginning of the light tained only cones in its retina." The results period, the rods shed packets of membranes were unequivocal. Throughout the entire 12 from the ends of their outer segments, hr light period, there was not the slightest thereby confirming in a fish a rhythmic pro- evidence of any membrane detachment from cess previously documented in rats3 and the ends of the cones. However, less than an frogs.9' l0 Later in the day, soon after the hour after the beginning of the dark period, onset of the dark period, a comparable burst there was a brief phase of membrane shed- of membrane detachment from the ends of ding from the tips of the cone outer seg- the cones took place. If the cones dispose of ments. Before the dark period was over, the membranes on a regular basis, they must also membranes had been fully degraded by the repeatedly form new membranes. Other- pigment epithelium.

Downloaded from iovs.arvojournals.org on 09/30/2021 Volume 17 Number 2 Rhythmic shedding of rod and cone membranes 107

The highly reproducible demonstration of Table 1. Number of phagosomes/100 a rhythmic daily disposal of membranes by visual cells in the central region of the cones in a fish and a reptile strongly sup- retina at different times of day ported the proposal that this was a common Phagosome stage characteristic of vertebrate cones. However, Time of day there are some unusual features in these (hr:min) I II HI IV animals. Both are "cold-blooded" (ectother- Light period (0:00): mic). The lizards are unusual among verte- 0.15 0 0 0 0.5 brates in that they lack rods entirely. Fur- 0:30 5.9 8.1 0.8 0.6 0:45 2.3 8.8 2.3 1.9 thermore, the pigment epithelium in the 1:00 0.4 6.7 11.3 0.8 goldfish is atypical, being filled with tapetal 2:00 0 3.4 5.1 19.2 granules, which seem to limit its phagocytic 4:35 0 0 0 8.1 6:35 0 0 0.4 5.2 and degradative capacities, so that these are 8:35 0 0 0 0 supplemented by the acitivities of ameboid 10:10 0 0 0 0 cells. 11:10 0 0 0 0 11:40 0 0.5 0 2.0 Therefore I repeated the experiment in Dark period (12:00): another class of vertebrate animals—a class 12:10 4.8 0 0 0.4 which is homeothermic, with a duplex retina 12:20 7.2 0 0 0.7 and a more conventional pigment epithe- 12:30 4.4 4.1 2.2 0 12:40 12.8 8.4 2.7 0 lium. The animal selected was a bird, the 12:50 9.3 0.7 1.4 0.7 chicken. A report of the results of this exper- 13:00 0.6 6.6 5.9 5.0 iment is given below. 13:30 0.9 7.3 20.0 7.4 14:00 0 4.6 15.1 4.4 Methods 15:00 0.5 11.6 4.8 10.5 16:00 1.2 7.3 4.1 19.8 Twenty-six male chicks (Gallus doinesticus, 18:00 1.1 5.3 2.8 11.9 white Leghorn, XL-link variety) were obtained 20:00 0 4.3 10.0 12.5 22:00 0 1.6 2.0 4.0 from Pace/Setter Products, Inc., Cucamonga, 23:00 2.2 5.1 1.8 2.9 Calif., the clay after hatching. They were main- 23:55 0 0 0 0.8 tained in a photographic darkroom in three metal cages, each measuring 30 by 45 by 60 cm, with a gently removed from the cage without disturbing wire mesh floor. An electric heating pad, set at the remaining chicks. The heads were decapitated "medium,' was lodged against the back wall of into ice-cold fixative in the dark, then brought into each cage to raise the temperature slightly above the light for further dissection. The eyes were that of the darkroom, which was regulated at 24° to enucleated, and the front half of the globe 25° C. Each cage was illuminated with a 60-watt trimmed off. The eye cup was then transferred to a incandescent bulb mounted in a standard goose- specimen bottle containing ice-cold fixative, which neck desk lamp. The bulb was situated 10 cm was allowed to come to room temperature. After a above and 10 cm away from the front of the cage, few hours, the retina was trimmed into smaller 2 yielding 32 foot-candles (345 lu/m ) of light in the pieces, and fixation was continued overnight in the center of the cage. The chicks were placed on a refrigerator. The fixative solution was 0.5% form- cycle of 12 hr of light and 12 hr of darkness for 12 aldehyde and 0.5% glutaraldehyde in 0.08M clays. Chicken feed and water were available at all phosphate buffer, pH 7.1. The specimens were times. then fixed for 1 hr in 1% osmium tetroxide in the On the thirteenth day, the animals were sac- same buffer and embedded in Araldite. For the rificed by decapitation at the following times light microscope, sections were cut at a thickness (hours: minutes) after the beginning of the light of 0.5 fJim, the embedment was removed by incu- period: 0:15, 0:30, 0:45, 1, 2, 4:35, 6:35, 8:35, bation in 1/6-saturated sodium hydroxide in abso- 10:10, 11:10, 11:40 (lights off at 12:00), 12:10, lute ethanol, and the sections were stained with 12:20, 12:30, 12; 40, 12:50, 13, 13:30, 14, 15, 16, 1% toluidine blue in 1% sodium borate. For the 18, 22, 23, 23:55. Those taken during the dark electron microscope, ultrathin sections (silver period were quiet, apparently sleeping, and were interference color) were cut, then stained with

Downloaded from iovs.arvojournals.org on 09/30/2021 Invest. Ophtkalmol. Visual Sci. 108 Young February 1978

Fig. 2. Phagosomes within the cytoplasm of the chick pigment epithelium containing clusters of membranes detached from the tips of cone outer segments. A to C, "Fresh" phagosomes, which are not yet noticeably degraded (Stage I). D to F, Stage II phagosomes, in which signs of degradation are visible but the membranous structure still predominates. (A, x48,600; B, x39,000; C, x40,000; D, X38.000; E, x52,600; F, x45,000.)

Downloaded from iovs.arvojournals.org on 09/30/2021 Volume 17 Number 2 Rhythmic shedding of rod and cone membranes 109

Fig. 3. Phagosonies containing degraded membranes derived from cone outer segments. A to C, Stage HI phagosonies, in which there is only sparse evidence of the original membranous structure. D to F, Stage IV phagosomes. The outer segment membrane structure is no longer detectable, although the predominantly reticular residues may contain regions with a layered orientation. Frequently, parallel filaments which appear to have the same structure as those forming the framework of melanosomes (cf. Fig. 6, A and B) were deposited on the surface of stage IV phagosomes (arrows, D and E). Two partially melanized premelanosomes which seem to be attached to the surface of a phagosome (x) are visible in F. (A, X45,3OO; B, X35,8OO; C, X28.500; D, X35.600; E, x51,300j F, X37.000.)

Downloaded from iovs.arvojournals.org on 09/30/2021 Invest. Ophthalmol. Visual Sci. 110 Young Fchruary 1978

Fig. 4. For legend see opposite page.

Downloaded from iovs.arvojournals.org on 09/30/2021 Volume 17 Number 2 Rhythmic shedding of rod and cone membranes 111

uranyl acetate and lead citrate. Most of the blocks difference is greater in the periphery, where were oriented so that sections were cut parallel the rod outer segments are longer, bulkier, with the long axis of the visual cells. The analysis and more numerous. (Rods comprise 33% of was largely restricted to the central region of the all visual cells in the periphery compared to retina. However, a few transverse sections and a 14% in the center.13) Rod and cone mem- few specimens of the peripheral retina were also branes also reacted differently to the fixative. examined. The cone outer segments were not well Quantitative analysis. In specimens from the central area of the retina, sectioned longitudinally preserved, although other structures, includ- for electron microscope examination, I recorded ing rod outer segments and phagosomes de- the number of phagosomes in the pigment epi- rived from the cones were satisfactorily thelial cells (Fig. 1, Table I). The phagosomes stabilized. The rods underwent photo- were subdivided into four classes, which are de- mechanical movements, extending in the picted in Figs. 2 and 3. Stage I included those light and contracting in the dark, but the which were not perceptibly degraded, although cones were immobile. Each pigment epithe- they usually showed increased density. Stage II lial cell is associated with 16 to 18 visual cells comprised phagosomes which were partially de- in the center and 11 to 12 visual cells in the graded, but consisted primarily of identifiable periphery of the retina. outer segment membranes. Stage III phagosomes contained some recognizable outer segment The hexagonal pigment epithelial cells membranes, but most of the contents were di- have the usual features of this type of cell, gested beyond recognition. Stage IV included the including infoldings of the basal surface; final stages of degradation, in which outer segment lateral junctional complexes; and long, apical, membranes could no longer be identified. In each melanosome-containing processes which en- region of cell body analyzed, the number of asso- velop the visual cell outer segments. The cell ciated rod and cone outer segments was recorded, body contains small Golgi zones, multivesicu- so that the counts of phagosomes could be related lar bodies, autophagic vacuoles, small to a known area (defined by the number of associ- granules (like those described in the lizard"), ated visual cell outer segments). Whenever the melanosomes, mitochondria, myeloid bodies orientation and preservation of stage I phagosomes free polysomes, very sparse smooth and permitted, I counted the number of outer segment discs (doubled membranes) in the packet. rough endoplasmic reticulum, and phago- These quantitative assessments were made on the somes. The phagosomes vary in number fluorescent screen of the electron microscope. and structure, depending on the time of day, as will now be described. Results The light period. During the first few The chick retina contains one type of rod minutes of the light period, the rods began and several types of cones.12- I3 The outer their photomechanical movement, pushing segments of all the cones are similar but dif- their outer segments deeper into the pig- fer from those of the rods. The cone outer ment epithelium, until their tips were on a segment discs are continuous with the outer level with the ends of the stationary cone membrane and with one another, forming a outer segments, or slightly beyond. After 15 continuous membrane system. In rods, only min there was no evidence of membrane the discs at the base of the outer segment shedding, although a few phagosomes re- retain such continuities.14 The rod outer mained in the pigment epithelium in late segments are larger in diameter (about 2.0 stages of degradation from an earlier shed- than those of the cones (1.5 fim). This ding event (Table I). Fifteen minutes later,

Fig. 4. A, Small phagosome containing cone membranes (upper left) is shown next to the much larger phagosome containing rod membranes. A very small minority of the cones shed membranes early in the light period, during the brief interval when the rods discard membranes. An example of this rare event is recorded in this field. (x31,000.) B, Three phagosomes situated near the tip of a single rod outer segment (r). Rods occasionally shed membranes in more than one group, as appears to be the case in this example. (X 18,500.)

Downloaded from iovs.arvojournals.org on 09/30/2021 Invest. Ophthalmol. Visual Sci. 112 Young February 1978

Fig. 5. For legend see opposite page.

Downloaded from iovs.arvojournals.org on 09/30/2021 Volume 17 Number 2 Rhythmic shedding of rod and cone membranes 113

the displacement of the rods was complete, The range was 11 to 72, and the average was and the shedding of membranes was vigor- 37 discs/packet. The vast majority occurred ously underway (Figs. 4 and 5, A). All or singly, but in a few cases the membranes practically all of the rods appeared to discard were shed from one cone in two separate membranes. Rarely did a section through a clusters. rod fail to contain a large phagosome near the The structural changes which accompany tip. The vast majority of the membranes were the degradation of rod and cone phagosomes detached in a single packet. However, occa- were similar. In the latter stages of digestion, sionally two groups of discs appeared to have when it was no longer possible to discern vi- been derived from the same rod (Figs. 4, B, sual cell membranes in the digestive vacuole, and 5, A). A few new (stage I) phagosomes the contents commonly appeared reticulated containing cone membranes were detected. and were frequently associated with oriented These were rare, but unequivocal (Fig. 4, A). filaments which were in layers on the outer The number of membranous discs was re- surface of the phagosome. These filaments corded in 25 rod phagosomes. The average were very similar to those which form the was 68 discs/phagosome (range 52 to 86). The framework of melanosomes (Figs. 3, D to F, shedding of membranes was completed by 1 and 6, A to C). There is also a close associa- hr, and by the mid-point of the light period, tion between the residues of phagosome di- the degradation of the ingested membranes gestion and melanosome formation in the was in its terminal stages (Table I and Fig. 1). pigment epithelium of the lizard." Twenty minutes before the end of the light Changes in the appearance of mitochon- period, a few cone-derived phagosomes were dria in the pigment epithelium were noted detected. (Fig. 5, B and C). Increased polymorphism The dark period. Ten minutes into the seemed to predominate in the middle of the dark period, the rods had initiated their light period and again early in the dark photomechanical contraction which, when period. Sequestering and degradation of complete about 20 min later, would result in melanosomes along the apical edge of the their tips being on a level near the base of the pigment epithelial cell body (Fig. 6, D and E) cone outer segments. At 10 min, the cones appeared to reach a maximum early in the were already actively discarding membranes. light period, although it was not restricted to As had been the case with the rods during the that interval. light period, the detachment of membranes from the ends of the cones was largely com- Discussion pleted within an hour. However, a small The experimental animals formed a re- fraction of the cones continued to shed markably homogeneous group. They were all batches of membranes throughout the dark males from an inbred stock, hatched the period, and as noted above, a few continued same day, and raised together in the same to do so as late as 1 hr into the light period. environment. The experimental design was As a result, phagosomes in various stages of of utmost simplicity, since the only known degradation were present within the pigment variable was a light cycle of 12 hr of illumina- epithelial cells throughout the entire dark tion and 12 hr of darkness. Nevertheless, period (Table I and Fig, 1). dramatic changes were recorded in the pig- The number of discs contained in 32 ment epithelium and visual cells which were cone-derived phagosomes was determined. strictly correlated with time of day. These

Fig. 5. A, Three dense phagosomes slightly displaced from the ends of two rod outer segments are visible in this field. Essentially all the rods appear to shed membranes early in the light period, as indicated by the presence of phagosomes near their tips. The rods do not discard membranes at any other time of day. (X 7,400.) B and C, Fluctuation between the rounded, regular appearance of mitochondria in the pigment epithelium (B) and the polymorphic state (C) occurred at different times of day. (B, x 19,000; C, X23,5OO.)

Downloaded from iovs.arvojournals.org on 09/30/2021 Invest. Ophthalmol. Visual Sri. 114 Young February J97S

Fig. 6. Melanosomes are continually renewed in the chick pigment epithelium. A to C, Premelanosomes in three stages of development. D, At certain times of day—especially, early in the light period—groups of melanosomes are sequestered and degraded. E, This process takes place in the apical part of the cytoplasm, near the visual cell outer segments. The arrow indicates a residual body resulting from melanosome degradation. (A, X65.000; B, X65,600; C, x48,800; D, x 12,000; E, x 17,300.)

Downloaded from iovs.arvojournals.org on 09/30/2021 Volume 17 Number 2 Rhythmic shedding of rod and cone membranes 115

events must have been synchronized by the At 13 days after hatching, rod outer seg- light cycle—a conclusion fully consistent with ments contain about 575 discs and cones an established biological principle: Daily about 860 discs in the central retina.14 These rhythms are entrained by the daily fluctua- are based on daytime counts, when the rods tions in light energy.15"17 are somewhat shorter. If we use 600 discs as a In these chicks, the rods shed groups of rough estimate for the rods, a renewal time of membranes from the ends of their outer about 9 days (600 -^ 68/day) is obtained. For segments early in the light period. Then, cones, the estimated renewal time for the one-half day later, the cones discarded pack- outer segment would be on the order of 23 ets of membranes from their tips early in the days (860 -r- 37), on the assumption that they dark period. These results perfectly comple- all shed daily, and even longer if they don't. ment those obtained earlier in a fish8 and a These are only crude estimates, but they do reptile,11 thereby adding strong support to suggest that outer segment membrane re- the hypothesis stated in the introduction, placement occurs at a slower pace in chick that rods and cones renew their outer seg- cones than it does in chick rods. ments by membrane replacement according In Fig. 1, the undegraded (stage I) phago- to daily rhythms which are separated in time somes are separated from phagosomes in var- by approximately 12 hr. ious states of digestion. This distinguishes All the rods appear to shed membranes two different pigment epithelial cell activ- daily. This conclusion is based upon the ob- ities—phagocytosis of the membranes and servations that (1) 30 to 45 min into the light digestion of the phagocytized material. Un- period a phagosome was associated with the like the lizard, with its pure-cone retina, tip of practically every rod and (2) the sum of which manifests a single interval of phago- all stages of phagosomes was about 15 per 100 cytosis and digestion during the dark peri- visual cells in the central retina (Table I), od,11 there are two such daily events in the which is similar to the proportion of rods duplex retinas of the chick and goldfish.8 The (14%) in that part of the retina. It was difficult degradative phase overlaps the phagocytic to determine whether all the cones shed phase and lasts longer. In fact, degradation daily, since the event was less restricted in continues throughout most of the day, with time, occurring mainly at the beginning of two broad maxima. In the lizard pigment the dark period but taking place sporadically epithelium, in addition to the phase of phago- throughout the 12 hr of darkness. In none of cytosis and digestion, a phase of accu- the chicks did the sum of all phagosome mulation of small granules and a phase of au- stages exceed 36% (Table I). However, even tophagy have been described." These also if it is assumed that all the cones discarded overlap in time, and their maxima and membranes daily, the average number of minima do not coincide. The daily rhythm of discs per phagosome was only slightly more metabolism may therefore be conceived as a than half that recorded for rod phagosomes regulated succession of numerous overlap- (37 compared to 68). These chicks were ex- ping maxima and minima producing continu- amined 13 days after hatching, when outer ously changing chemical conditions in the segment growth is essentially complete.14 cell in a sequence which is cyclic; that is, it Therefore, in this steady-state condition, repeats itself. where disc formation is balanced by disc- In recent years, the awareness has grown shedding, the rate of membrane renewal is that the metabolism of the visual cells and less in cones than it is in rods. During the pigment epithelium is closely intertwined. 14 prehatching stage, Godfrey found that The study of daily rhythms adds another di- when the outer segments are still increasing mension to this interrelationship—the time in length, the rods produce about 120 discs/ dimension. At least some, and probably all,18 day and the cones about 90/day, adding sup- of the chemical activities of these cells pro- port to the conclusion that rods produce ceed according to daily rhythms which are membranes faster than cones in the chicken. synchronized by light.

Downloaded from iovs.arvojournals.org on 09/30/2021 Invest. Ophthalmol. Visual-Sci. 116 Young February 1978

The assistance of Mirdza Lasmanis, Zoja Trirogoff, receptor shedding is initiated by light in the frog Myunghee Chun, Margarita Woo, and Dr. William T. retina, Science 194:1074, 1976. O Day is gratefully acknowledged. 10. Hollyfield, J. C, Besharse, J. C, and Rayborn, M. E.: The effect of light on the quantity of phagosomes REFERENCES in the pigment epithelium, Exp. Eye Res. 23:623, 1. Young, R. W.: Visual cells and the concept of re- 1976. newal, INVEST. OPHTHALMOL. 15:700, 1976. 11. Young, R. W.: The daily rhythm of shedding and 2. Young, R.. W.: Biogenesis and renewal of visual cell degradation of cone outer segment membranes in outer segment membranes, Exp. Eye Res. 18:215, the lizard retina, J. Ultrastruct. Res. 61:172, 1977. 1974. 12. Morris, V. B., and Shorey, C. D.: An electron mi- 3. La Vail, M. M.: Rod outer segment disk shedding in croscope study of types of receptor in the chick ret- rat retina: relationship to cyclic lighting, Science ina, J. Comp. Neurol. 129:313, 1967. 194:1071, 1976. 13. Morris, V. B.: Symmetry in a receptor mosaic dem- 4. Young, R. W., and Bok, D.: Participation of the onstrated in the chick from the frequencies, spacing retinal pigment epithelium in the rod outer segment and arrangement of the types of retinal receptor, J. renewal process, J. Cell Biol. 42:392, 1969. Comp. Neurol. 140:359, 1970. 5. Ditto, M.: A difference between developing rods 14. Godfrey, A. J.: The photoreceptors of the chick. and cones in the formation of outer segment mem- Ph.D. Dissertation, Department of Anatomy, Uni- branes, Vision Res. 15:535, 1975. versity of California, Los Angeles, 1974. 6. Steinberg, R. H., Wood, I., and Hogan, M. J.: Pig- 15. Bunning, E.: The Physiological Clock, ed. 3, New ment epithelial ensheathment and phagocytosis of York, 1973, Springer-Verlag. extrafoveal cones in human retina, Philos. Trans. R. 16. Aschoff, J.: Exogenous and endogenous components Soc. London [Biol.] 277:459, 1977. in circadian rhythms, Cold Spring Harbor Symp. 7. Anderson, D. H., and Fisher, S. K.: The photo- Quant. Biol. 25:11, 1960. receptors of diurnal squirrels: outer segment struc- 17. Pittendrigh, C. S.: Circadian oscillations in cells and ture, disc shedding, and protein renewal, J. Ultra- the circadian organization of multicellular systems. struct. Res. 55:119, 1976. In Schmitt, F. O., and Worden, F. C, editors: The 8. O'Day, W. T., and Young, R. W.: Rhythmic daily Neurosciences. Third Study Program, Cambridge, shedding of outer segment membranes by visual Mass., 1974, MIT Press, p. 437. cells in the goldfish, J. Cell Biol. (in press, 1977). 18. Young, R. W.: Visual cells, daily rhythms, and vi- 9. Basinger, S., Hoffman, R., and Matthes, M.: Photo- sion research, Vision Res. (in press, 1977).

Erratum In the December issue of INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE, p. 1142, the name of the author in reference 15 and the first author in reference 23 should be Lovekin.

Downloaded from iovs.arvojournals.org on 09/30/2021