- Email: [email protected]
Refractive state, depth of focus and accommodation of the eye of the California ground squirrel (Spermophilus Beecheyi)
V;~,on Res. Vol. 24. No. IO. pp. 1261-1266. Printed m Great Britain. All rights reserved
1984 Copyright
004?-6989:84 $3.00 + 0.00 >c‘s1984 Pergamon Press Ltd
STATE, DEPTH OF FOCUS AND REFRACTIVE ACCOMMODATION OF THE EYE OF THE CALIFORNIA GROUND SQUIRREL (SPERMOPHILUS BEECHEYI) MARK Department
E. MCCOURT* and GERALDH. JAcosst
of Psychology, University of California, Santa Barbara, CA 93106, U.S.A. (Receioed 22 September 1983; in revked form 16 March 1984)
Abstract-Retinoscopy and electrophysiological refraction were performed on 55 and 24 faraccommodated eyes of California ground squirrels (Spennophilus beecheyi), respectively. These two indices were highly correlated, revealing the eye of this animal to be roughly emmetropic (-0.25 to -0.13 D). Depth of focus was assessed by measuring the effect which defocusing produced on the spatial resolving power of 32 optic nerve fibers. Depth of focus of the ground squirrel eye for a pupil diameter of 2.5 mm is estimated to be + I .6 D, but will increase rapidly for smaller pupi!s. Accommodation in eleven ground squirrels ranged from 2 to 6 D, with a mean value of 3.9 D. Ground squirrel
Single-unit
Refractive state
Retinoscopy
INTRODUCTION
temperature
Hughes (1977) and Powers and Green (1978) demonstrated in electrophysiological experiments that the eye of the rat has tremendous depth of focus (approx. + / - 12 diopters) and suggested that it is, despite its large apparent hyperopic ametropia, effectively emmetropic for small ( < 3.0 mm) pupils. Gur and Sivak (1979) measured the refractive state of two species of diurnal rodent, the Mexican (Spermophifus mexicattus) and thirteen-lined (S. tridecemlineatus) ground squirrels. Based on results from both retinoscopic and electrophysiological measurements they concluded that these eyes were emmetropic (f0.8 D), and that their accommodative range was 2.0-2.5 D. Because a great deal of electrophysiological and behavioral work has been performed on the ground squirrel it is important to know the refractive and accommodative parameters of this animal. Consequently, the proposal that the eye of the ground squirrel is emmetropic and does not show the artifact of retinoscopy is addressed here. Our conclusions are based on experiments in which the eye of another species of ground squirrel (S. beecheyi) was examined with respect to refractive state, depth of focus and accommodative range, using retinoscopy and electrophysiological recording techniques.
Depth of focus
Accommodation
colony room on a 12 hr light: 12 hr dark
lighting regimen. Preparation of animals and recording
Details of the preparation are provided elsewhere (McCourt and Jacobs, in press). Briefly, animals were anesthetized with urethane (2.4g/kg, i.p.) and placed in a stereotaxic instrument. A small hole was drilled in the skull directly overlying the optic nerve of the test eye. The pupil was dilated and accommodation paralyzed by cornea1 application of atropine sulfate (0.04:/,) and phenylephrine HCl (10% ophthalmic). A 2.5 mm aperture contact lens was installed to prevent cornea1 drying and to provide a standard artificial pupil. The test eye was retinoscopically refracted to a distance of 33 cm through the use of trial lenses placed within 5mm of the cornea. A large sample of eyes was additionally refracted using a non-contact lens 2.5 mm artificial pupil to correct for any constant refractive error introduced by the departure of contact lens curvature from that of the cornea. Normal body temperature (37-38°C) was maintained through the use of a rectal thermometer and a circulating hot-water heating pad. Single units in the optic nerve were isolated and their activity recorded with glass-insulated tungsten microelectrodes.
METHODS Subjecrs
Stimuli
Adult (SOO-1000 g) California ground squirrels (Spermophilus beecheyi) of both sexes were trapped in
Stimuli were electronically generated on a visual display oscilloscope (Hewlett-Packard model 1332A, P4 phosphor) which subtended 21.5” in width by 16.8” in height at a distance of 33 cm. Display luminance was constant at 50 cd/m2. Modulation of the Z-axis (e.g. spatial frequency, contrast and phase of gratings) was provided by a digital function generator under the program control of a laboratory
Santa Barbara
county
and housed in a constant-
*Present address and address for correspondence: Department of Physiology, John Curtin School of Medical Research, The Australian National University, G.P.O. Box 334, Canberra A.C.T. 2601, Australia. tTo whom reprint requests should be addressed.
1261
1262
MARK E. MCCWRT and GERALD H. JACOBS
computer. The unit response measure used was spikes/set above spontaneous discharge rate.
Retinorcopic
lO-
Slit-lamp retinoscopic examination of the test eye of each animal was performed 20 min after the application of a&opine sulfate. Refractive states were determjne~ to the nearest 0.5 D, A subset of the retinoscopically refracted squirrels were subsequently examined in electrophysiological experiments, where units were first tested to determine the highest spatial frequency orating to which they could reliably respond. High contrast (80%) square wave grating stimuli were used. Grating spatial frequency was incremented under program control in steps of 0.1 c/d. Six seconds of unit activity were sampled at each spatial frequency, and collected by computer. Resolution limit was defined as the spatial frequency to which response fell below 0.5 spikes/set above spontaneous discharge rate. Drift rate was from 1 to 3 Hz depending upon the unit; 6 set of stimulus presentation thus indexed the response of a unit to between six and eighteen grating cycles. Temporal frequencies in this range did not differentially affect spatial resolution. Following the determination of a unit’s resolution at the refractive state initially determined by retinoscopic examinat~u~~ corrective tens power was successively incremented and/or decremented in one diopter steps. At each new refractive state the resolution limit and response (to suprathreshold contrast) of the unit under consideration was remeasured as described above. For some units reso~utio~~response was examined over a range of seven diopters, while for others the range explored was more limited, The refractive state (total lens power minus 3 D working distance) yielding the highest value of spatial resolution/response was taken as a measure of the optimal refractive state of the eye. Only units from the central 20” of the visual field whose spectral sensitivity reflected input solely from a 525 nm photopigment were tested. The latter restriction was employed to avoid problems associated with refractive state measurements using the short wavelength light to opt~rna~~y stimulate the (460 nm) required spectrally-opponent units of the ground squirrel optic nerve (Jacobs and Tootell, 1980, 1981; Jacobs el al., 1981). Accommodative range was measured in eleven animals by the method of retinoscopy. Atropine sulfate was administered to the cornea of one eye to paralyze accommodation at its far-point, while pilocarpine HCI was administered to the cornea of the opposite eye to paralyze accommodation at its nearpoint. Refractive state was measured for the atropinized eye through a 2.5 mm artificial pupil. The small pupil diameters (0.5-I .2 mm) of eyes administered p~~oca~pi~e obviated the need for arti~cjal pupils. Differences in refractive state observed between the near- and far-accommodated eyes were
5
0.
Far Point ~di~pters) Fig. I. ~istrjb~t~on of r~tinoscopicaIly and electrophysiologically determined far-accommodated refractive states of the California ground squirrel. Right panel. Distribution of refractive states obtained by retinoscopic measur~m~nt (33 cm working distance) for the far-
~cc~rn~~at~ eyes of 55 ground squirrels. Mean retinoscopic refractive state is -0.25 D (SD = 2.0). Left panel. Distribution of far-accommodated refractive states determined by the electrophysiological refraction of single optic nerve fibers in 24 squirrels. Distance of the eye from the oscilloscope display was 33 cm. Mean electrophysiological refractive state is -0.13 D (SD = 2.3). The means of these distributians are not significantly different.
taken as a measure of the accommodative each animal.
range of
RESULTS
Retinoscopy was performed on the atropinized eyes of 55 ground squirrels. Refractive states in these eyes ranged from myopia f -4 D) to bypero~~a (i- 5 D). The mean of the fifty-five values was -0.25 D (SD = 2.0). No degree of astigmatism was observed which exceeded the precision of the refraction estimates (i.e. >0.5 D). The distributjon of the retinoscopically measured refractive states is shown in Fig. 1. From the total of 55 retinoscopically refracted animals, 24 were subsequently refracted during si~gje-chit chording using e~ect~~pbys~ol~~ca~ criteria. The results of etectrophysiological refraction are also shown in Fig. 1 and indicate a similar distribution of refractive states, ranging from myopia ( - 4 D) to hyperopia ( + 4 D). The mean value in this case was -0.13 D (SD = 2.3). No significant difference was found between the means of these distributions (F = 0.04, n.s,, d.f. = 1,77). The degree of correlation between these two measures of refractive state appears in Fig. 2 which shows a scatterplot of the correction values for the 24 squirrels in which both retinoscopic and e!:ctrophysiological estimates of for-accammodated refractive state were determined. The coe~cie~t of correlation (r) is highly significant (r = 0.88, P < 0.01, N = 24) and the 99% confidence interval for the least-squares regression coefficient (0.87) is 90.34, hence spawning a slope of 1.0. It thus appears that retinoscopic methods for estimating the refractive state of the eye of the ground squirrel are as effective
Refractive state of the ground squirrel eye
1263
54.
N=U
3.
t-t’087
2I ’
-4
-3
~2
-I
0
I
2
3
4
-3
5
-2
-1
0
DioDtera Ret~norcop~c
For
Pant
(diopters)
Fig. 2. Scatter-plot of retinoscopic vs electrophysiological values of far-accommodated refractive state in 24 California ground squirrels in which both measures were obtained. The independent measures are significantly correlated (r = 0.88, P < 0.01, N = 24). and the regression parameter (B) is 0.87. The slope of the (dashed) least-squares regression line does not differ significantly from unity (solid line).
methods over the entire range
Depth offocus as indexed by optic nervejber
responses
Depth of focus refers to the maintenance of visual resolution despite defocusing. The effect which defocusing produces in the response or resolution of single-units in the visual system has been successfully used to measure depth of focus both in the rat (Hughes, 1977; Powers and Green, 1978) and rabbit (Meyer et al., 1972). Figure 3(a) illustrates the effect which defocusing has on the spatial resolution capacities of 32 single optic nerve fibers in the ground squirrel. The functions have been normalized to their optimal refractive states. The inverted “V” shaped functions show the degree to which resolution is degraded as the refractive error of the dioptric system is shifted in either direction (with positive or negative lens power) away from its optimal value. The slopes of the falloff in resolution for individual units ranged from 0.01 to 0.6 c/d per diopter of defocus. The average ascending (negative lens power) and descending (positive lens power) slopes are 0.253 and -0.238c/d per diopter of defocus, respectively. The absolute values of the slopes are not significantly different (F = 0.12, n.s., d.f. = 1,62). Additionally, while the data shown in Fig. 3(a) reflect spatial resolution measures, similar results were obtained where the dependent variable consisted of unit response to a constant contrast stimulus. Note that the falloff in resolution is sharp for units whose spatial resolution is high, and is quite shallow for the lowest resolution units. Figure 3(b) plots the slope of the resolution falloff (in c/d per diopter) as a function of maximum spatial resolution for these 32 units. Solid symbols represent the falloff with negative lens power, open symbols the falloff with positive
.2
.3
(4 0.7 $
N=6L r=067 rJ=l?10
0.6
:
0.5:
0
7
.O
. .
:
0.4.
0
.
.
,_,e
y_-
,__*
s-d .*
,/ . __-00 0. . d’ 0 p--’ . 0 q/5+y’8 8 l 0 0 .C’k 0 &o 0
$ 0.3 : = . 2s 0.2 .: ‘i: -‘o 0.1 E LT
0
1
P
8
as electrophysiological of refractive states.
.l
from oDti”W”l
. _,,..-‘6 OL
0 .
.
00. 0
0.5
1.0 Spatial
15
2.0
Resolution
2.5
3.0
-
3.5
4.0
(c/d)
(b)
Fig. 3.(a) Spatial resolution of 32 optic nerve fibres plotted as a function of defocusing, normalized to their optimal refractive state. Optimal resolution ranged from 0.4 to 3.8c/d. Average slopes of resolution falloff do not differ significantly with plus or minus lens power. Steeper falloffs are observed for units whose spatial resolution is higher. (b) Slopes of positive (open symbols) and negative (solid symbols) resolution falloffs with defocusing for the 32 units shown in Fig. 3(a), plotted against optimal spatial resolution. The dashed regression line, fitted by the method of least-squares, depicts a significant positive correlation between spatial resolution and its fallotT with defocusing.
lens power. The dashed regression line was fitted to the 64 data points by the method of least squares. The coefficient of correlation (r) is highly significant (r = 0.67, P < 0.01, N = 64), indicating that spatial resolution of units with acuities of 4.0 c/d, for example, falls off at a rate of 0.5 c/d per diopter of defocus. The range of values obtained from the measurement of accommodative capacity in eleven eyes is shown in Fig. 4. These eyes were observed to possess from 2 to 6 D of accommodative range, with a mean value of 3.9 D. DISCUSSION
The results of the present study confirm and extend those of Gur and Sivak (1979), who described the far-accommodated refractive state of the Mexican and thirteen-lined ground squirrel as emmetropic (+0.8 D). The California ground squirrel far-
MARKE. MCCOURTand GERALD H. JACOBS
1264
Accommodative
Amplitude
(II)
Fig. 4. Estimated amplitude of accommodation of the eye of the California ground squirrel. The histogram shows difference values obtained by retinoscopic examination of eleven ground squirrels following pharmacological manipulation of accommodative state with atropine sulfate and pilocarpine HCI. Accommodation ranged from 2 to 6 D; mean range = 3.9 D.
accommodated refractive state can also be described as emmetropic (-0.13 to -0.25 D)*. By contrast, retinoscopic measurements of a variety of rodent, lagomorph and other species whose eye size is similar to that of the California ground squirrel reveal that a variable degree of hyperopia (I-10 D) is common. Thus, the domestic rabbit shows a faraccommodated refractive state between +2 and +4 D (Barlow et al., 1964; Hughes and Vaney, 1978; Hughes, 1972; Nuboer and Genderen-Takken, 1978; Millodot and Sivak, 1978), the guinea pig is +4 D (Glickstein and Millodot, 1970), the tree shrew (Tupaia g/is) is reported to be approx. +4 D (Schafer, 1969), and the refractive state of the Siberian chipmunk (Eutumius sibiricus) is between + l-3 D (Polkoschnikov, 1980). A retinoscopic artifact of 8.9 D has been confirmed in the rat (Hughes, 1977), and in the South American opossum which is retinoscopically 2.3 D (Oswaldo-Cruz et al., 1979), but *We have not corrected our estimates of refractive state, depth of focus or accommodative capacity for the effects of either spectacle magnification or efirtive dioptric value of corrective lenses (Ogle, 1968; Hughes, 1977). Such effects contribute approx. 5% of our measured values, and their correction would exceed the precision of our measurements. tThe resolution limit of the California ground squirrel, measured in behavioral and visual evoked potential tests (Jacobs et al., 1980) and from single-unit studies (McCourt and Jacobs, in press) is 4.ic/d. The 80% value is thus 3.2 c/d. The falloff in resolution of fibers with 4.0 c/d resolution, given from the regression line of Fig. 3(b), is 0.5 c/d per diopter. Thus the range of defocusing permitted before these fibers fail to detect a 3.2c/d high-contrast grating is + 1.6 D. SCalculated from the geometric model of Green et al. (I 980). for a pupil diameter of 0.16 mm. Depth of focus (D) in this model is given by: D = 7.03/b * c), where t’ is the spatial frequency cut-off of the visual system in question and p is the pupil diameter in mm. Alternatively, using the same pupil diameter, and an estimated focal length of the ground squirrel eye of 5.0 mm. depth of focus corresponding to a 20% decrease in image intensity, computed from diffraction theory, is + 10.7 D (Born and Wolf, 1959).
electrophysiologically refracted at a value of -2.3 D (Picanco-Diniz et al., 1983). To the contrary a very strong positive correlation was revealed in the ground squirrel between two independent measures of refractive state: retinoscopy and electrophysiological refraction. This result is concordant with those of Gur and Sivak (1979). The hypothesis of Glickstein and Millodot (1970), that hyperopia is typically found in retinoscopic measurement of the eyes of small mammals and is an artifact of the method of testing, is thus in this instance not supported. Whether the absence of retinoscopic artifact observed here for ground squirrel is a consequence of its diurnal habits (Gur and Sivak, 1979) or not remains to be tested. It would be of interest, therefore, to compare electrophysiological and retinoscopic estimates of refractive state in other truly diurnal species, such as the tree shrew, or the marsupial numbat. The individual variations in refractive state found in the present sample of ground squirrel eyes are typical. In a sample of five squirrels, Gur and Sivak (1979) report a 5 D range of variation in resting refractive state. Schafer (1969) reported substantial variation (6 D range) in the refractive state of tree shrews. A seventeen diopter range of refractive state has been reported in a sample of 25 South American opossums (Picanco-Diniz et al., 1983). Whereas the rat (Hughes, 1977; Powers and Green, 1978) has been shown to possess substantial depth of focus (approx. +/- 12 D), the results of the defocusing experiments performed with a similar pupil size on single units of the ground squirrel suggest a heightened sensitivity to image blur. Adopting the criterion of Green et al. (1980), according to which depth of focus is defined as the degree of defocusing permitted before spatial resolution is reduced to 80% of its optimal value, the California ground squirrel is estimated to possess a depth of focus of f 1.6 Dt. This estimate is similar to that for the rabbit (Meyer et al., 1972), and to that which may be inferred from the data of Gur and Sivak (1979), who measured the effects of defocusing on contrast thesholds of optic nerve fibers in a related sciurid species. The estimates of both far-accommodated refractive state and depth of focus were obtained for a constant pupil size of 2.5 mm, and it is well known that small entrance apertures provide optical systems with an enlarged depth of focus. Ground squirrel spatial resolution (4.0 c/d) becomes diffraction-limited only for pupil diameters smaller than 0.16 mm, and the possession of small entrance pupils, in combination with the very high (and therefore compensatory) ambient daylight illumination in which the strongly diurnal ground squirrel is active (Linsdale, 1946; Fitch, 1948) suggests that the potential depth of focus of this eye may be as large as +/-- 1I D.: Estimates of the range of accommodation of other sciurid eyes have recently been reported (Gur and Sivak, 1979; Sivak, 1980). Retinoscopy performed on
Refractive state of the ground squirrel eye the eyes of 10 squirrels (S. r~j~~c~~~j~~u~~ and S. rttexicanus) revealed differences in refractive state, before and after the application of atropine and pilocarpine, of between 2-4 D. Similar techniques applied to the present sample of t 1 S. ~~e~~~y~ indicate a greater potential for accommodation (range = 2-6 D, mean = 3.9 D). In an animal of similar eye-size, the Indian mongoose (HerpeJles auro~~~c~u~~~), a considerably larger range of a~~mrnodation (I L-1 3.5 D) has recently been reported (Sivak, 1980). There are at least three possible sources of error in our estjmates of accommodative range. The first is that the initial refractive states of the two test eyes may not have been equivalent. In 6 animals for which the far-accommodated refractive state was measured in both eyes, in only one was there a discrepancy of over 1.O D. Furthc~ore~ while the range of values will be affected by this source of error, the average amplitude of accommodation shall not. The second is that drum-indu~d aeeomm~~ative responses may ~~ferest~rn~te the natural capacity of the animal. This is almost certainly the case, since cycloplegics inactive ciliary muscle tone as well as ocular accommodation, which enhances the apparent amplitude of accommodation by the equivalent of the tonic component. The effect in humans is, however, typically small, around 0.5 D. The third possibility is that reduced pupil site in p~l~carp~ne-administered eyes may produce myopjc shifts in refractive state independently from those due to ocular accommodation. This would occur if overcorrected (negative) spherical aberration of ground squirrel eyes were severe ~Camp~ll and Hughes, 1981). The ground squirrel lens has been shown to possess substantial overcorrected spherical aberration (Sivak et al., 1983). However, based on retinoscopy at multiple pupil sizes the intact ground squirrel eye appears free from spherical aberration (Cur and Sivak, 1979), and so it is unlikely that we have greatly overestimated accommodative capacity. Additional evidence for the presence of apprax. 4 D of accommodation in S. ~~ec~e~~comes from the finding that spatial resolution in behavioral tests of this species is constant at moderate light intensities over viewing distances ranging from 15 to 40 cm (Jacobs et a!., 1980). Other behavioral observations made under both laboratory and field conditions also suggest the utility of accommodative mechanisms for the ground squirrel. In particular, this species lives in close association with rattlesnakes (their chief predator), with which it frequently engages in aggressive interactions (Owings and Coss, 1977; Owings el a/,, 1977). Such encounters typically involve a close vigilant approach ~~5-5~cm), avoidance of strike and occasionally, attack. Such a high degree of dependence upon visually-guided behavior in clearly life-threatening situations might be argued to be an adequate selection pressure for the degree of accommodation reported here.
1265
Ac~no~~ed~ement~-We
thank John Foley, Jack Loomis and Barbara Btakeslee for comments on the manuscript in an earlier form, Horace Barlow for useful discussion, Melanie Campbell for much helpful and constructive criticism, and Jan Ljvingstone for secretarial assistance. This research was ~~~po~ed by Grant E~-~l~~ from the National Eye Institute.
Barlow H. B., Hill R, M. and Levick W_ R. (1964) Retinal
ganglion cells responding selectively to direction and speed of image motion in the rabbit. J. Ph)~iol. 173, 337-407.
Born M, and Wolf E. (1959) P~~~cjple~~~~pt~cs, Pergamoo
Press, London. Campbell M. W. C. and Hughes A. (1981) An analytic gradient index schematic lens and eye for the rat which predicts a~rra~io~s for finite pupils, Vision Res. 21, 1129-I 148. Fitch H, S. (1948) Ecology of the California ground squirrel on grazing land. Am. mid. Nat. 39, 513-596. Glickstein M. and Millodot M, (1970) Retinoscopy and eye size. Science 168, 605-606. Green IX G., Powers M. K. and Banks M. S. (19$~) Depth of focus, eye size and visual acuity. ??&?I Rex. 20, 827-835. Gur M. and Sivak J. G. (1979) Refractive state of the eye of a small diurnal mammal: the ground squirrel, ADI. .l. opr. Pbys. optics 56, 689-695. Hughes A. (1972) A schematic eye for Ihe rabbit. Vision Res. 12, 123-138. Hughes A. (1977) The refractive state of the rat eye. Vision Res. 17, 927-939. Hughes A. and Vaney D, I, {~97~) The refractive state of the rabbit eye: variation with eccentricity and correction for oblique astigmatism. Vision Res. 18, 1351-1355. Jacobs G. H, and Tootell R. 13. H. (1980) Spectrallyopponent responses in ground squirrel optic nerve. Vision Res. 20, 9-14.
Jacobs G. H. and Tootelf R. B. H. (1981) Spectral response properties of optic nerve fibers in the ground squirrel. J. Neuruphysiol. 45, 89 l-902.
Jacobs G. H., Blake&e B. and Tootell R. B. H. (i98 1) dolor-disc~m~na~ion tests on fibers in ground squirrel optic nerve. J. ~euruphy~io~. 45, 903-914. Jacobs G. H., Blakeslee B., McCourt M. E. and Tootell R. B. H. (1980) Visual sensitivity of ground squirrels to spatial and temporal luminance variations. J. camp. Physioi. 136, 291-299. L&dale J. M. (1946) The Cu~?~~~~~~Ground S4~~r~~~, University of California Press, Berkeley. McCourt M. E. and Jacobs G. H. Spatial filter characteristics of optic nerve fibers in the California ground squirrel ~~~@~~~~~i~~~ beec~l~~i). J. ~ez~ro~/~J,~~o~,III press.
Meyer D. L., Meyer-Hamme S. and Schaeffer K. P. (1972) Electrophysiological investigation of refractive state and accommodation in the rabbit’s eye. P’ugers Arch. ges. Physiik 332, 80-86. Millodot M. and Sfvak J. (1978) Hype~etropia of small eyes and chromatic aberration. Vision Res. 18, 125-126, Nuboer J. F. W. and Genderen-Takken H. van. (1978)The artifact of retinoscopy. Vision Res. 18, 1091-1096. Ogle K. N. (1968) @pi&, 2nd Edition. Charles C. Thomas, Illinois. Qswaldo-Cruz E., Hokoc J. N. and Sousa A. P. B. (1979) A schematic eye for the oppossum. Vision Res. 19, 263-278. Owiags D. H. and Goss R. G, (1977) Snake mobbing by California ground squirrels: adaptive vacation and ontogeny. Behnviour 62, 50-69.
1266
MARK
E.
MCibURf
Owings D. H., Borchert M. and Virginia R. (1977) The behavior of the California ground squirrel. Anim. B&v. 25, 221-230. P~~an~~~~~~~ C. W., Silveira L. C. L. and ~swaldo-~r~z E. (1983) ~lectrophysio~o~~a~ dete~~nation of the refractive state of the eye of the opossum. Vision Res. 23, 867-872. Palkoshnikov E. V. (1980) Selective sensitivity of visual cortex Negroes in the Siberian ~bipmu~k (~~~~~~~ sibi&us) to velocity and dilution of movement and orientation of stimuli. J. Ed. Biochem. Physid. 15, 475-477. Powers M. K. and Green D. G. (1978) Single retinal gangiion cell responses in the dark-mated rat: grating
and
GERALD
H.J~coes
acuity, contrast sensitivity, and defocusing. Vision Res. 18, 1533-1539. Schafer D. (1969) Utersuchungen aur Sehphysiologie des SPitzho~che~s T~pffi~ glis (Diard 182~~~Z. Flprgf. Physiol. 63, 204-225.
Sivak J. G. (1980) Accommodation in vertebrates: a comparative survey. In Current Topics in Eye Research (Edited by Zadunaisky J. A. and I-I. Davson H.), Vol. 3, Academic Press, New York. Sivak J. G., Gur M. and Dovrat A. (1983) Sphericat aberration of the lens of the ground squirrel (Spermophilus tridecemfineatus). 261-265.
Ophrhal. Physiol.
Opt. 3,
1984 Copyright
004?-6989:84 $3.00 + 0.00 >c‘s1984 Pergamon Press Ltd
STATE, DEPTH OF FOCUS AND REFRACTIVE ACCOMMODATION OF THE EYE OF THE CALIFORNIA GROUND SQUIRREL (SPERMOPHILUS BEECHEYI) MARK Department
E. MCCOURT* and GERALDH. JAcosst
of Psychology, University of California, Santa Barbara, CA 93106, U.S.A. (Receioed 22 September 1983; in revked form 16 March 1984)
Abstract-Retinoscopy and electrophysiological refraction were performed on 55 and 24 faraccommodated eyes of California ground squirrels (Spennophilus beecheyi), respectively. These two indices were highly correlated, revealing the eye of this animal to be roughly emmetropic (-0.25 to -0.13 D). Depth of focus was assessed by measuring the effect which defocusing produced on the spatial resolving power of 32 optic nerve fibers. Depth of focus of the ground squirrel eye for a pupil diameter of 2.5 mm is estimated to be + I .6 D, but will increase rapidly for smaller pupi!s. Accommodation in eleven ground squirrels ranged from 2 to 6 D, with a mean value of 3.9 D. Ground squirrel
Single-unit
Refractive state
Retinoscopy
INTRODUCTION
temperature
Hughes (1977) and Powers and Green (1978) demonstrated in electrophysiological experiments that the eye of the rat has tremendous depth of focus (approx. + / - 12 diopters) and suggested that it is, despite its large apparent hyperopic ametropia, effectively emmetropic for small ( < 3.0 mm) pupils. Gur and Sivak (1979) measured the refractive state of two species of diurnal rodent, the Mexican (Spermophifus mexicattus) and thirteen-lined (S. tridecemlineatus) ground squirrels. Based on results from both retinoscopic and electrophysiological measurements they concluded that these eyes were emmetropic (f0.8 D), and that their accommodative range was 2.0-2.5 D. Because a great deal of electrophysiological and behavioral work has been performed on the ground squirrel it is important to know the refractive and accommodative parameters of this animal. Consequently, the proposal that the eye of the ground squirrel is emmetropic and does not show the artifact of retinoscopy is addressed here. Our conclusions are based on experiments in which the eye of another species of ground squirrel (S. beecheyi) was examined with respect to refractive state, depth of focus and accommodative range, using retinoscopy and electrophysiological recording techniques.
Depth of focus
Accommodation
colony room on a 12 hr light: 12 hr dark
lighting regimen. Preparation of animals and recording
Details of the preparation are provided elsewhere (McCourt and Jacobs, in press). Briefly, animals were anesthetized with urethane (2.4g/kg, i.p.) and placed in a stereotaxic instrument. A small hole was drilled in the skull directly overlying the optic nerve of the test eye. The pupil was dilated and accommodation paralyzed by cornea1 application of atropine sulfate (0.04:/,) and phenylephrine HCl (10% ophthalmic). A 2.5 mm aperture contact lens was installed to prevent cornea1 drying and to provide a standard artificial pupil. The test eye was retinoscopically refracted to a distance of 33 cm through the use of trial lenses placed within 5mm of the cornea. A large sample of eyes was additionally refracted using a non-contact lens 2.5 mm artificial pupil to correct for any constant refractive error introduced by the departure of contact lens curvature from that of the cornea. Normal body temperature (37-38°C) was maintained through the use of a rectal thermometer and a circulating hot-water heating pad. Single units in the optic nerve were isolated and their activity recorded with glass-insulated tungsten microelectrodes.
METHODS Subjecrs
Stimuli
Adult (SOO-1000 g) California ground squirrels (Spermophilus beecheyi) of both sexes were trapped in
Stimuli were electronically generated on a visual display oscilloscope (Hewlett-Packard model 1332A, P4 phosphor) which subtended 21.5” in width by 16.8” in height at a distance of 33 cm. Display luminance was constant at 50 cd/m2. Modulation of the Z-axis (e.g. spatial frequency, contrast and phase of gratings) was provided by a digital function generator under the program control of a laboratory
Santa Barbara
county
and housed in a constant-
*Present address and address for correspondence: Department of Physiology, John Curtin School of Medical Research, The Australian National University, G.P.O. Box 334, Canberra A.C.T. 2601, Australia. tTo whom reprint requests should be addressed.
1261
1262
MARK E. MCCWRT and GERALD H. JACOBS
computer. The unit response measure used was spikes/set above spontaneous discharge rate.
Retinorcopic
lO-
Slit-lamp retinoscopic examination of the test eye of each animal was performed 20 min after the application of a&opine sulfate. Refractive states were determjne~ to the nearest 0.5 D, A subset of the retinoscopically refracted squirrels were subsequently examined in electrophysiological experiments, where units were first tested to determine the highest spatial frequency orating to which they could reliably respond. High contrast (80%) square wave grating stimuli were used. Grating spatial frequency was incremented under program control in steps of 0.1 c/d. Six seconds of unit activity were sampled at each spatial frequency, and collected by computer. Resolution limit was defined as the spatial frequency to which response fell below 0.5 spikes/set above spontaneous discharge rate. Drift rate was from 1 to 3 Hz depending upon the unit; 6 set of stimulus presentation thus indexed the response of a unit to between six and eighteen grating cycles. Temporal frequencies in this range did not differentially affect spatial resolution. Following the determination of a unit’s resolution at the refractive state initially determined by retinoscopic examinat~u~~ corrective tens power was successively incremented and/or decremented in one diopter steps. At each new refractive state the resolution limit and response (to suprathreshold contrast) of the unit under consideration was remeasured as described above. For some units reso~utio~~response was examined over a range of seven diopters, while for others the range explored was more limited, The refractive state (total lens power minus 3 D working distance) yielding the highest value of spatial resolution/response was taken as a measure of the optimal refractive state of the eye. Only units from the central 20” of the visual field whose spectral sensitivity reflected input solely from a 525 nm photopigment were tested. The latter restriction was employed to avoid problems associated with refractive state measurements using the short wavelength light to opt~rna~~y stimulate the (460 nm) required spectrally-opponent units of the ground squirrel optic nerve (Jacobs and Tootell, 1980, 1981; Jacobs el al., 1981). Accommodative range was measured in eleven animals by the method of retinoscopy. Atropine sulfate was administered to the cornea of one eye to paralyze accommodation at its far-point, while pilocarpine HCI was administered to the cornea of the opposite eye to paralyze accommodation at its nearpoint. Refractive state was measured for the atropinized eye through a 2.5 mm artificial pupil. The small pupil diameters (0.5-I .2 mm) of eyes administered p~~oca~pi~e obviated the need for arti~cjal pupils. Differences in refractive state observed between the near- and far-accommodated eyes were
5
0.
Far Point ~di~pters) Fig. I. ~istrjb~t~on of r~tinoscopicaIly and electrophysiologically determined far-accommodated refractive states of the California ground squirrel. Right panel. Distribution of refractive states obtained by retinoscopic measur~m~nt (33 cm working distance) for the far-
~cc~rn~~at~ eyes of 55 ground squirrels. Mean retinoscopic refractive state is -0.25 D (SD = 2.0). Left panel. Distribution of far-accommodated refractive states determined by the electrophysiological refraction of single optic nerve fibers in 24 squirrels. Distance of the eye from the oscilloscope display was 33 cm. Mean electrophysiological refractive state is -0.13 D (SD = 2.3). The means of these distributians are not significantly different.
taken as a measure of the accommodative each animal.
range of
RESULTS
Retinoscopy was performed on the atropinized eyes of 55 ground squirrels. Refractive states in these eyes ranged from myopia f -4 D) to bypero~~a (i- 5 D). The mean of the fifty-five values was -0.25 D (SD = 2.0). No degree of astigmatism was observed which exceeded the precision of the refraction estimates (i.e. >0.5 D). The distributjon of the retinoscopically measured refractive states is shown in Fig. 1. From the total of 55 retinoscopically refracted animals, 24 were subsequently refracted during si~gje-chit chording using e~ect~~pbys~ol~~ca~ criteria. The results of etectrophysiological refraction are also shown in Fig. 1 and indicate a similar distribution of refractive states, ranging from myopia ( - 4 D) to hyperopia ( + 4 D). The mean value in this case was -0.13 D (SD = 2.3). No significant difference was found between the means of these distributions (F = 0.04, n.s,, d.f. = 1,77). The degree of correlation between these two measures of refractive state appears in Fig. 2 which shows a scatterplot of the correction values for the 24 squirrels in which both retinoscopic and e!:ctrophysiological estimates of for-accammodated refractive state were determined. The coe~cie~t of correlation (r) is highly significant (r = 0.88, P < 0.01, N = 24) and the 99% confidence interval for the least-squares regression coefficient (0.87) is 90.34, hence spawning a slope of 1.0. It thus appears that retinoscopic methods for estimating the refractive state of the eye of the ground squirrel are as effective
Refractive state of the ground squirrel eye
1263
54.
N=U
3.
t-t’087
2I ’
-4
-3
~2
-I
0
I
2
3
4
-3
5
-2
-1
0
DioDtera Ret~norcop~c
For
Pant
(diopters)
Fig. 2. Scatter-plot of retinoscopic vs electrophysiological values of far-accommodated refractive state in 24 California ground squirrels in which both measures were obtained. The independent measures are significantly correlated (r = 0.88, P < 0.01, N = 24). and the regression parameter (B) is 0.87. The slope of the (dashed) least-squares regression line does not differ significantly from unity (solid line).
methods over the entire range
Depth offocus as indexed by optic nervejber
responses
Depth of focus refers to the maintenance of visual resolution despite defocusing. The effect which defocusing produces in the response or resolution of single-units in the visual system has been successfully used to measure depth of focus both in the rat (Hughes, 1977; Powers and Green, 1978) and rabbit (Meyer et al., 1972). Figure 3(a) illustrates the effect which defocusing has on the spatial resolution capacities of 32 single optic nerve fibers in the ground squirrel. The functions have been normalized to their optimal refractive states. The inverted “V” shaped functions show the degree to which resolution is degraded as the refractive error of the dioptric system is shifted in either direction (with positive or negative lens power) away from its optimal value. The slopes of the falloff in resolution for individual units ranged from 0.01 to 0.6 c/d per diopter of defocus. The average ascending (negative lens power) and descending (positive lens power) slopes are 0.253 and -0.238c/d per diopter of defocus, respectively. The absolute values of the slopes are not significantly different (F = 0.12, n.s., d.f. = 1,62). Additionally, while the data shown in Fig. 3(a) reflect spatial resolution measures, similar results were obtained where the dependent variable consisted of unit response to a constant contrast stimulus. Note that the falloff in resolution is sharp for units whose spatial resolution is high, and is quite shallow for the lowest resolution units. Figure 3(b) plots the slope of the resolution falloff (in c/d per diopter) as a function of maximum spatial resolution for these 32 units. Solid symbols represent the falloff with negative lens power, open symbols the falloff with positive
.2
.3
(4 0.7 $
N=6L r=067 rJ=l?10
0.6
:
0.5:
0
7
.O
. .
:
0.4.
0
.
.
,_,e
y_-
,__*
s-d .*
,/ . __-00 0. . d’ 0 p--’ . 0 q/5+y’8 8 l 0 0 .C’k 0 &o 0
$ 0.3 : = . 2s 0.2 .: ‘i: -‘o 0.1 E LT
0
1
P
8
as electrophysiological of refractive states.
.l
from oDti”W”l
. _,,..-‘6 OL
0 .
.
00. 0
0.5
1.0 Spatial
15
2.0
Resolution
2.5
3.0
-
3.5
4.0
(c/d)
(b)
Fig. 3.(a) Spatial resolution of 32 optic nerve fibres plotted as a function of defocusing, normalized to their optimal refractive state. Optimal resolution ranged from 0.4 to 3.8c/d. Average slopes of resolution falloff do not differ significantly with plus or minus lens power. Steeper falloffs are observed for units whose spatial resolution is higher. (b) Slopes of positive (open symbols) and negative (solid symbols) resolution falloffs with defocusing for the 32 units shown in Fig. 3(a), plotted against optimal spatial resolution. The dashed regression line, fitted by the method of least-squares, depicts a significant positive correlation between spatial resolution and its fallotT with defocusing.
lens power. The dashed regression line was fitted to the 64 data points by the method of least squares. The coefficient of correlation (r) is highly significant (r = 0.67, P < 0.01, N = 64), indicating that spatial resolution of units with acuities of 4.0 c/d, for example, falls off at a rate of 0.5 c/d per diopter of defocus. The range of values obtained from the measurement of accommodative capacity in eleven eyes is shown in Fig. 4. These eyes were observed to possess from 2 to 6 D of accommodative range, with a mean value of 3.9 D. DISCUSSION
The results of the present study confirm and extend those of Gur and Sivak (1979), who described the far-accommodated refractive state of the Mexican and thirteen-lined ground squirrel as emmetropic (+0.8 D). The California ground squirrel far-
MARKE. MCCOURTand GERALD H. JACOBS
1264
Accommodative
Amplitude
(II)
Fig. 4. Estimated amplitude of accommodation of the eye of the California ground squirrel. The histogram shows difference values obtained by retinoscopic examination of eleven ground squirrels following pharmacological manipulation of accommodative state with atropine sulfate and pilocarpine HCI. Accommodation ranged from 2 to 6 D; mean range = 3.9 D.
accommodated refractive state can also be described as emmetropic (-0.13 to -0.25 D)*. By contrast, retinoscopic measurements of a variety of rodent, lagomorph and other species whose eye size is similar to that of the California ground squirrel reveal that a variable degree of hyperopia (I-10 D) is common. Thus, the domestic rabbit shows a faraccommodated refractive state between +2 and +4 D (Barlow et al., 1964; Hughes and Vaney, 1978; Hughes, 1972; Nuboer and Genderen-Takken, 1978; Millodot and Sivak, 1978), the guinea pig is +4 D (Glickstein and Millodot, 1970), the tree shrew (Tupaia g/is) is reported to be approx. +4 D (Schafer, 1969), and the refractive state of the Siberian chipmunk (Eutumius sibiricus) is between + l-3 D (Polkoschnikov, 1980). A retinoscopic artifact of 8.9 D has been confirmed in the rat (Hughes, 1977), and in the South American opossum which is retinoscopically 2.3 D (Oswaldo-Cruz et al., 1979), but *We have not corrected our estimates of refractive state, depth of focus or accommodative capacity for the effects of either spectacle magnification or efirtive dioptric value of corrective lenses (Ogle, 1968; Hughes, 1977). Such effects contribute approx. 5% of our measured values, and their correction would exceed the precision of our measurements. tThe resolution limit of the California ground squirrel, measured in behavioral and visual evoked potential tests (Jacobs et al., 1980) and from single-unit studies (McCourt and Jacobs, in press) is 4.ic/d. The 80% value is thus 3.2 c/d. The falloff in resolution of fibers with 4.0 c/d resolution, given from the regression line of Fig. 3(b), is 0.5 c/d per diopter. Thus the range of defocusing permitted before these fibers fail to detect a 3.2c/d high-contrast grating is + 1.6 D. SCalculated from the geometric model of Green et al. (I 980). for a pupil diameter of 0.16 mm. Depth of focus (D) in this model is given by: D = 7.03/b * c), where t’ is the spatial frequency cut-off of the visual system in question and p is the pupil diameter in mm. Alternatively, using the same pupil diameter, and an estimated focal length of the ground squirrel eye of 5.0 mm. depth of focus corresponding to a 20% decrease in image intensity, computed from diffraction theory, is + 10.7 D (Born and Wolf, 1959).
electrophysiologically refracted at a value of -2.3 D (Picanco-Diniz et al., 1983). To the contrary a very strong positive correlation was revealed in the ground squirrel between two independent measures of refractive state: retinoscopy and electrophysiological refraction. This result is concordant with those of Gur and Sivak (1979). The hypothesis of Glickstein and Millodot (1970), that hyperopia is typically found in retinoscopic measurement of the eyes of small mammals and is an artifact of the method of testing, is thus in this instance not supported. Whether the absence of retinoscopic artifact observed here for ground squirrel is a consequence of its diurnal habits (Gur and Sivak, 1979) or not remains to be tested. It would be of interest, therefore, to compare electrophysiological and retinoscopic estimates of refractive state in other truly diurnal species, such as the tree shrew, or the marsupial numbat. The individual variations in refractive state found in the present sample of ground squirrel eyes are typical. In a sample of five squirrels, Gur and Sivak (1979) report a 5 D range of variation in resting refractive state. Schafer (1969) reported substantial variation (6 D range) in the refractive state of tree shrews. A seventeen diopter range of refractive state has been reported in a sample of 25 South American opossums (Picanco-Diniz et al., 1983). Whereas the rat (Hughes, 1977; Powers and Green, 1978) has been shown to possess substantial depth of focus (approx. +/- 12 D), the results of the defocusing experiments performed with a similar pupil size on single units of the ground squirrel suggest a heightened sensitivity to image blur. Adopting the criterion of Green et al. (1980), according to which depth of focus is defined as the degree of defocusing permitted before spatial resolution is reduced to 80% of its optimal value, the California ground squirrel is estimated to possess a depth of focus of f 1.6 Dt. This estimate is similar to that for the rabbit (Meyer et al., 1972), and to that which may be inferred from the data of Gur and Sivak (1979), who measured the effects of defocusing on contrast thesholds of optic nerve fibers in a related sciurid species. The estimates of both far-accommodated refractive state and depth of focus were obtained for a constant pupil size of 2.5 mm, and it is well known that small entrance apertures provide optical systems with an enlarged depth of focus. Ground squirrel spatial resolution (4.0 c/d) becomes diffraction-limited only for pupil diameters smaller than 0.16 mm, and the possession of small entrance pupils, in combination with the very high (and therefore compensatory) ambient daylight illumination in which the strongly diurnal ground squirrel is active (Linsdale, 1946; Fitch, 1948) suggests that the potential depth of focus of this eye may be as large as +/-- 1I D.: Estimates of the range of accommodation of other sciurid eyes have recently been reported (Gur and Sivak, 1979; Sivak, 1980). Retinoscopy performed on
Refractive state of the ground squirrel eye the eyes of 10 squirrels (S. r~j~~c~~~j~~u~~ and S. rttexicanus) revealed differences in refractive state, before and after the application of atropine and pilocarpine, of between 2-4 D. Similar techniques applied to the present sample of t 1 S. ~~e~~~y~ indicate a greater potential for accommodation (range = 2-6 D, mean = 3.9 D). In an animal of similar eye-size, the Indian mongoose (HerpeJles auro~~~c~u~~~), a considerably larger range of a~~mrnodation (I L-1 3.5 D) has recently been reported (Sivak, 1980). There are at least three possible sources of error in our estjmates of accommodative range. The first is that the initial refractive states of the two test eyes may not have been equivalent. In 6 animals for which the far-accommodated refractive state was measured in both eyes, in only one was there a discrepancy of over 1.O D. Furthc~ore~ while the range of values will be affected by this source of error, the average amplitude of accommodation shall not. The second is that drum-indu~d aeeomm~~ative responses may ~~ferest~rn~te the natural capacity of the animal. This is almost certainly the case, since cycloplegics inactive ciliary muscle tone as well as ocular accommodation, which enhances the apparent amplitude of accommodation by the equivalent of the tonic component. The effect in humans is, however, typically small, around 0.5 D. The third possibility is that reduced pupil site in p~l~carp~ne-administered eyes may produce myopjc shifts in refractive state independently from those due to ocular accommodation. This would occur if overcorrected (negative) spherical aberration of ground squirrel eyes were severe ~Camp~ll and Hughes, 1981). The ground squirrel lens has been shown to possess substantial overcorrected spherical aberration (Sivak et al., 1983). However, based on retinoscopy at multiple pupil sizes the intact ground squirrel eye appears free from spherical aberration (Cur and Sivak, 1979), and so it is unlikely that we have greatly overestimated accommodative capacity. Additional evidence for the presence of apprax. 4 D of accommodation in S. ~~ec~e~~comes from the finding that spatial resolution in behavioral tests of this species is constant at moderate light intensities over viewing distances ranging from 15 to 40 cm (Jacobs et a!., 1980). Other behavioral observations made under both laboratory and field conditions also suggest the utility of accommodative mechanisms for the ground squirrel. In particular, this species lives in close association with rattlesnakes (their chief predator), with which it frequently engages in aggressive interactions (Owings and Coss, 1977; Owings el a/,, 1977). Such encounters typically involve a close vigilant approach ~~5-5~cm), avoidance of strike and occasionally, attack. Such a high degree of dependence upon visually-guided behavior in clearly life-threatening situations might be argued to be an adequate selection pressure for the degree of accommodation reported here.
1265
Ac~no~~ed~ement~-We
thank John Foley, Jack Loomis and Barbara Btakeslee for comments on the manuscript in an earlier form, Horace Barlow for useful discussion, Melanie Campbell for much helpful and constructive criticism, and Jan Ljvingstone for secretarial assistance. This research was ~~~po~ed by Grant E~-~l~~ from the National Eye Institute.
Barlow H. B., Hill R, M. and Levick W_ R. (1964) Retinal
ganglion cells responding selectively to direction and speed of image motion in the rabbit. J. Ph)~iol. 173, 337-407.
Born M, and Wolf E. (1959) P~~~cjple~~~~pt~cs, Pergamoo
Press, London. Campbell M. W. C. and Hughes A. (1981) An analytic gradient index schematic lens and eye for the rat which predicts a~rra~io~s for finite pupils, Vision Res. 21, 1129-I 148. Fitch H, S. (1948) Ecology of the California ground squirrel on grazing land. Am. mid. Nat. 39, 513-596. Glickstein M. and Millodot M, (1970) Retinoscopy and eye size. Science 168, 605-606. Green IX G., Powers M. K. and Banks M. S. (19$~) Depth of focus, eye size and visual acuity. ??&?I Rex. 20, 827-835. Gur M. and Sivak J. G. (1979) Refractive state of the eye of a small diurnal mammal: the ground squirrel, ADI. .l. opr. Pbys. optics 56, 689-695. Hughes A. (1972) A schematic eye for Ihe rabbit. Vision Res. 12, 123-138. Hughes A. (1977) The refractive state of the rat eye. Vision Res. 17, 927-939. Hughes A. and Vaney D, I, {~97~) The refractive state of the rabbit eye: variation with eccentricity and correction for oblique astigmatism. Vision Res. 18, 1351-1355. Jacobs G. H, and Tootell R. 13. H. (1980) Spectrallyopponent responses in ground squirrel optic nerve. Vision Res. 20, 9-14.
Jacobs G. H. and Tootelf R. B. H. (1981) Spectral response properties of optic nerve fibers in the ground squirrel. J. Neuruphysiol. 45, 89 l-902.
Jacobs G. H., Blake&e B. and Tootell R. B. H. (i98 1) dolor-disc~m~na~ion tests on fibers in ground squirrel optic nerve. J. ~euruphy~io~. 45, 903-914. Jacobs G. H., Blakeslee B., McCourt M. E. and Tootell R. B. H. (1980) Visual sensitivity of ground squirrels to spatial and temporal luminance variations. J. camp. Physioi. 136, 291-299. L&dale J. M. (1946) The Cu~?~~~~~~Ground S4~~r~~~, University of California Press, Berkeley. McCourt M. E. and Jacobs G. H. Spatial filter characteristics of optic nerve fibers in the California ground squirrel ~~~@~~~~~i~~~ beec~l~~i). J. ~ez~ro~/~J,~~o~,III press.
Meyer D. L., Meyer-Hamme S. and Schaeffer K. P. (1972) Electrophysiological investigation of refractive state and accommodation in the rabbit’s eye. P’ugers Arch. ges. Physiik 332, 80-86. Millodot M. and Sfvak J. (1978) Hype~etropia of small eyes and chromatic aberration. Vision Res. 18, 125-126, Nuboer J. F. W. and Genderen-Takken H. van. (1978)The artifact of retinoscopy. Vision Res. 18, 1091-1096. Ogle K. N. (1968) @pi&, 2nd Edition. Charles C. Thomas, Illinois. Qswaldo-Cruz E., Hokoc J. N. and Sousa A. P. B. (1979) A schematic eye for the oppossum. Vision Res. 19, 263-278. Owiags D. H. and Goss R. G, (1977) Snake mobbing by California ground squirrels: adaptive vacation and ontogeny. Behnviour 62, 50-69.
1266
MARK
E.
MCibURf
Owings D. H., Borchert M. and Virginia R. (1977) The behavior of the California ground squirrel. Anim. B&v. 25, 221-230. P~~an~~~~~~~ C. W., Silveira L. C. L. and ~swaldo-~r~z E. (1983) ~lectrophysio~o~~a~ dete~~nation of the refractive state of the eye of the opossum. Vision Res. 23, 867-872. Palkoshnikov E. V. (1980) Selective sensitivity of visual cortex Negroes in the Siberian ~bipmu~k (~~~~~~~ sibi&us) to velocity and dilution of movement and orientation of stimuli. J. Ed. Biochem. Physid. 15, 475-477. Powers M. K. and Green D. G. (1978) Single retinal gangiion cell responses in the dark-mated rat: grating
and
GERALD
H.J~coes
acuity, contrast sensitivity, and defocusing. Vision Res. 18, 1533-1539. Schafer D. (1969) Utersuchungen aur Sehphysiologie des SPitzho~che~s T~pffi~ glis (Diard 182~~~Z. Flprgf. Physiol. 63, 204-225.
Sivak J. G. (1980) Accommodation in vertebrates: a comparative survey. In Current Topics in Eye Research (Edited by Zadunaisky J. A. and I-I. Davson H.), Vol. 3, Academic Press, New York. Sivak J. G., Gur M. and Dovrat A. (1983) Sphericat aberration of the lens of the ground squirrel (Spermophilus tridecemfineatus). 261-265.
Ophrhal. Physiol.
Opt. 3,