Eyes of a Lower Vertebrate Are Susceptible to the Visual Environment

Abstract

purpose. Recently, it has been found that form deprivation myopia can be induced in fish (tilapia). This study examined the sensitivity of the tilapia eye to positive and negative lenses. It further investigated the sensitivity of the fish eye to form deprivation by examining the effect of fish weight.

methods. Twenty-five Nile tilapia (Oreochromis niloticus; group 1) were weighed (range, 26–101 g) and killed, and their eyes were measured to provide normative data regarding fish eye size, body weight, and refractive state. Goggles with lenses of refractive powers in water of either +15 D (group 2, n = 7) or −12 D (group 3, n = 7) were sutured over the right eye of for 2 weeks to induce hyperopia or myopia. The untreated contralateral eye served as a control. An additional six fish (group 4), each wearing a goggle with an open central area, were used to evaluate the effect of the goggle itself. Refractive measurements for these 20 fish were made before and after treatment, after which the fish were killed, the eyes were removed, and axial lengths were measured from frozen sections. Another 21 fish were treated with goggles with lenses for 2 weeks, after which the goggle was removed and the refractive states of both eyes were measured every day for 6 days (day 19) and then after 28 days. These fish were placed in one group (group 5) wearing negative (−12 D) lenses (n = 8; average weight, 25.5 g) and two groups (groups 6, 7) of different size (average weights, 13.9 g [n = 5] and 26.9 g [n = 8], respectively) wearing positive (+15 D) lenses during the treatment period. In addition, translucent goggles were applied for 2 weeks to induce form deprivation myopia in three groups of fish (groups 8, 9, 10) of different weights, averaging 16.0 g (n = 7), 57.4 g, (n = 8), and 98.4 g, (n = 7), to provide an evaluation of the effect of weight on the development of form deprivation myopia.

results. In untreated fish (group 1), the axial length of the eye, ranging from 5.86 mm to 7.16 mm, was proportional to weight (26.5–101 g), whereas refractive state shifted from hyperopia (+15D for 10-g fish) toward emmetropia. The +15D lens-treated fish (group 2) became hyperopic relative to the contralateral eye (+7.7 ± 1.6 D; mean ± SD), whereas the −12 D lenses (group 3) induced myopia relative to the control eye (−8.4 ± 0.8 D) within 2 weeks. Hyperopic eyes were shorter (4.16 ± 0.11 mm vs. 4.28 ± 0.06 mm) and myopic were eyes longer (3.96 ± 0.36 mm vs. 3.84 ± 0.27 mm) than their contralateral control eyes. There were no significant differences in eye size or refractive state between treated and untreated eyes of fish wearing open goggles. In the groups that were allowed to recover (groups 5, 6, 7), the fish treated with minus lenses developed an average of −9.8 ± 1.9 D myopia, whereas +15 D lenses induced average hyperopia amounts of +8.1 ± 1.4 D (group 6) and +6.2 5 ± 2.87 D (group 7). All these fish recovered completely within 2 weeks once the goggles with lenses were removed. Pretreatment and posttreatment refractive results indicated that the contralateral control eyes were affected by the positive and negative lens treatments, though to a lesser extent. Form deprivation myopia was induced in all three different weight groups, averaging −11.9 ± 2.9 D for group 8, 6.3 ± 2.5 D for group 9, and −2.3 ± 1.0 D for group 10. All form-deprived eyes and those treated with positive and negative lenses recovered—i.e., little or no difference resulted in refractive state or dimensions between the treated and untreated eyes—to pretreatment levels within 1 week of goggle removal.

conclusions. Tilapia, a lower vertebrate species, exhibits positive and negative lens-induced refractive change, as is the case for higher vertebrates. In addition, the level of sensitivity to form deprivation is weight dependent.

Myopia is a common ocular refractive condition found worldwide that is of increasing concern because of its high prevalence in youth. In Singapore, for example, the myopia incidence in high school students is greater than 70%. ¹ Intensive research during the past few decades has revealed a relationship between the visual environment and heredity and the development of myopia. ^{2 3 4 5} However, the precise cause of myopia is still unknown. 

Experiments dealing with refractive error development are frequently based on animal models such as chick, ² monkey, ^{6 7} and tree shrew. ⁸ In fact, various species of animal—including gray squirrel, ⁹ mouse, ¹⁰ guinea pig, ^{11 12} cat (Sommers D, et al. IOVS 1978;17:ARVO Abstract 295), ^{13 14} and American kestrel ¹⁵ —have been used to induce form deprivation myopia. All are higher vertebrates. Typically, experimental myopia is induced in young animals; often the younger the animal, the higher the induced refractive error. ^{16 17} In addition to form deprivation myopia, positive and negative lenses have been used to induce myopia and hyperopia, ¹⁸ and efforts to induce astigmatism with cylindrical lenses have been carried out. ^{19 20 21} Experiments such as those in which the chick optic nerve is cut show that refractive error development is determined by a local retinal mechanism. ²² Somehow the eye distinguishes whether the input visual signal is overfocused or underfocused ¹⁸ and, in chicks at least, rapidly adjusts retinal position and focal length of the eye by thinning or thickening its choroid (Wallman J, et al. IOVS 1992;33:ARVO Abstract 1053), ^{23 24} Eye growth then accelerates or slows down for a more permanent change in refractive state resulting from an eye that is either too long or too short. The growth factor TGFβ may play a key role in regulating scleral synthesis and in reshaping the globe in this process. ^{25 26}  

Almost all experimental animals are able to recover from the induced refractive errors and the contralateral differences in eye length after removal of the cause of the visual interference during the early stage of ocular development. However, it appears that monkeys are not always capable of recovering fully from form deprivation myopia. ²⁷  

Recently, it was found that form deprivation myopia can also be induced in fish, the largest group of vertebrates, and that complete recovery is possible. ²⁸ This common susceptibility to the visual environment in the vertebrate world suggests that a basic or universal mechanism controls refractive error development in nature, regardless of differences in habitats, genetics, and ocular anatomy and physiology. In addition, because fish can continue to grow and develop through life, the eye remains in a lifelong plastic state in terms of dimension and retinal development. ²⁹ This suggests that the fish eye may represent an interesting new animal model for studying the effect of the visual environment on the refractive state of the eye. This study investigated the susceptibility of the fish eye to the effect of positive and negative lenses and to the effect of fish weight on form deprivation myopia. 

Materials and Methods

Fish

Nile tilapia (Oreochromis niloticus), a commonly cultured species of cichlids, were obtained from a local fish farm (Northern Tilapia Inc., Bondhead, ON, Canada), kept in aquariums with cycled and filtered water at 28°C, and fed with tilapia fish food (3PT Regular; Martin Mills Inc., Elmira, ON, Canada). The size of the fish used was chosen for ease of handling; most weighed between 15 and 30 g. However, in some experiments, fish of weights ranging from 9.4 g to 154.2 g were selected on the basis of availability or to study the effect of weight on form deprivation myopia. Fluorescent lighting in the aquarium room was set to a diurnal (12-hour light/12-hour dark) schedule. All fish were cared for according to the Guidelines of the Canadian Council on Animal Care and in accordance with the policies of the University of Waterloo and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Table 1summarizes treatments and numbers of the 10 groups of fish studied in this research. 

Twenty-five untreated fish (group 1) ranging in weight from 26 to 101 g and in eye axial diameter from 4.96 to 7.34 mm were weighed and killed, and their eyes were measured to provide normative data regarding the relationship between fish eye size and body weight and between refractive state and fish weight. Another 64 fish were used in the convex and concave lens and form deprivation studies that followed. All treated fish were weighed at the start of treatment and at the end of treatment 2 weeks later. Pretreatment measurements of weight and refractive state of these fish were added to those of the untreated fish of group 1 to generate the weight-refractive state relationship (see 1 2 Fig. 3 ). Fish were weighed in a beaker, with water, of known weight, and the additional weight, measured to the nearest 0.1 g, represented that of the fish. The weights refer to weight at the start of treatment. Fish were killed with overdoses of 2-phenoxythanol anesthetic (3 mL/L) and then decapitated, and the eyes were enucleated. 

Experimental Protocols

Lens-Induced Myopia and Hyperopia.

Specially designed goggles with central convex (positive) and concave (negative) lens inserts were used to induce myopia or hyperopia (Fig. 1) . The inserted lenses consisted of custom-produced intraocular lenses 6.0 mm in diameter with powers, in water, of −12 D and +15 D. The lenses were plastic (polymethyl methacrylate) of 1.49 refractive index, with equal anterior and posterior radii of curvature consisting of 26.175 mm for the negative lenses and 20.801 mm for the positive lenses. The lightweight plastic goggles were directly sutured over one eye, the treated eye, for 2 weeks (Fig. 1) . The right eye was always the treated eye, and the left eye served as a contralateral control. Sutures were sewn through the skin and soft bones around the orbit of anesthetized fish using nylon stitches through holes in the edge of the plastic goggle. The fish were removed from water for approximately 1 minute during the suturing process. 

To control for the possible mechanical effect of the goggle, a ring-shaped plastic goggle with the central area open was sutured onto the right eye of a group of six fish (group 2, average weight 33.3 g) to evaluate the effect of the suturing and the goggle itself. Two additional groups, each consisting of seven fish, wore goggles with positive (group 3) or negative (group 4) lenses for 2 weeks. The average weights of the fish treated with positive lenses and negative lenses were 30.4 and 15.8 g, respectively. Refractive measurements for these 20 fish were made before and after treatment, after which the fish were killed, the eyes were removed, and the intraocular dimensions were measured from frozen sections. 

The globes, both treated and untreated, were enucleated, immersed in a freeze section medium (Stephens Scientific, Little Rock, AR), and frozen by placement in dry ice. Axial lengths were measured by freeze sectioning. Marks made on the eyes with indelible ink when the fish were first killed identified the nasal and temporal limbi, the apex of the cornea, and the posterior scleral exit point of the optic nerve. The eyes were freeze sectioned on a cryostat microtome until a horizontal section through the geometric axis of the eye (premarked with indelible ink) was apparent. A loupe with a scale (±0.10 mm) was placed on the section, and a photograph was taken with a digital camera. Later, the digital image of the hemisectioned eye was transferred to a computer and processed using Image J software (National Institutes of Health, Bethesda, MD). Axial length was measured by placing a curser on the anterior corneal apex and at the retina–choroid border. An effort to estimate the repeatability of axial dimensions was made by repeating the measurement for several images several times each. Results indicated that the mean difference in repeated measurements of axial diameter amounted to 0.01 mm with an SD of 0.02 mm. 

Twenty-one additional fish were treated with goggles for 2 weeks, after which the goggle was removed and the refractive states of both eyes were measured every day for 6 days (day 19) and then after 28 days. These fish were divided into one group (group 5), wearing negative (−12 D) lenses (average weight, 25.5 g; n = 8) and two groups (groups 6 and 7) of different weights (average weights, 13.9 g [n = 5] and 26.9 g [n = 8]) wearing +15 D lenses during the treatment period. 

Although the untreated contralateral eye in this study and in others involving animal models of refractive error development was considered a control eye, it has been pointed out that the contralateral eye can be affected by treatment as well. For example, Wildsoet and Wallman. ³⁰ reported that the refractive states of the untreated eyes of chicks in which one eye is treated with positive or negative lenses show small shifts in the hyperopic or the myopic directions. In the present study, we examined this by pooling refractive states before and after treatment for the groups treated with positive lenses (groups 3, 6, 7) and then with negative lenses (groups 4, 5). 

Form Deprivation.

Lightweight translucent plastic goggles were directly sutured over the treated eye (the right eye) for 2 weeks (Fig. 1) . The fish were divided into three groups: group 8, consisting of fish (n = 7) weighing from 12 g to 20.5 g and averaging 16.0 g (approximately 4 months old); group 9, consisting of fish (n = 8) weighing from 51.7 g to 61.6 g and averaging 57.4 g (approximately 7 months old); and group 10, consisting of fish (n = 8) ranging in weight from 60.2 g to 154.2 g and averaging 98.4 g (approximately 10 months old). These ages are estimates based on information provided by fish farm personnel when the fish were obtained added to the time the fish were maintained in holding aquaria during experimentation. All fish were treated with goggles for 2 weeks, after which the goggle was removed and the refractive states of both eyes were measured each day for 6 days (to day 19) and then after 28 days. 

Ocular Measurements.

Refractive states were measured with a streak retinoscope and trial lenses at a working distance of 25 cm through the glass wall of a specially designed narrow aquarium while the fish were anesthetized with 0.6 mL 2-phenoxythanol/L to fix the direction of gaze and to minimize accommodation. The results are an overestimation of the refractive error resulting from the difference in refractive index of water and glass. True refractive error is obtained by dividing by 1.33, and all values reported here were corrected in this manner. ³¹ Retinoscopic values were estimated to be accurate to within ±0.50 D. Means and standard deviations of the results are given to one decimal. Retinoscopic measurements were made along the direction perpendicular to the plane of the pupil, and the results are expressed as the difference between the refractive state, in diopters, of the treated and untreated eyes. Although the values are given as diopters of myopia or hyperopia, they represent shifts in refractive state in the myopic or hyperopic directions. 

The retinoscopic measurements were carried by an experienced refractionist (WS). These were verified periodically and independently on a limited number of fish by a second experienced investigator using an alternative instrument, a photorefractor (PowerRefractor; Multi-channel Systems Co., Reutlingen, Germany) with which the retinoscopic reflection was neutralized with trial lenses, as in retinoscopy. The results were never different by more than 0.50 D. Given that the most findings were made by retinoscopy, the results presented are retinoscopic measurements. 

Results

Normal Fish Eye Development (Group 1)

It was found that the axial length of the tilapia eye is proportional to weight, showing a strong correlation (R = 0.91; n = 25; P < 0.001) within the weight range (26.5–101 g) measured. Axial length ranged from 5.86 mm to 7.16 mm (Fig. 2) . At the same time, the refractive state of the eye gradually shifted from hyperopia of approximately +15 D for fish approximately 10 g in weight to emmetropia and leveling off at approximately 4 D of hyperopia (Fig. 3) . 

Lens-Induced Myopia and Hyperopia

In the fish with goggles with no central lens inserts (group 2), no significant change in refractive state (refractive state change [±SD], 0.2 ± 0.2 D before treatment and 0.6 ± 0.2 D after treatment) was noted (Fig. 4) . Similarly, the axial dimensions of the treated and control eyes were not significantly different (average axial diameters, 4.33 mm and 4.34 mm, respectively). The +15 D lens-treated fish eye (group 3) became significantly hyperopic compared with the control eye (+7.7 ± 1.6 D; n = 7; paired t-test, P < 0.0001), whereas the −12 D lenses (group 4) induced significant myopia (−8.4 ± 0.8 D; n = 7; P < 0.0001) within 2 weeks (Fig. 4) . The hyperopic eyes were shorter than the contralateral eye (treated eye, 4.16 ± 0.11 mm; untreated eye, 4.28 ± 0.06 mm; paired t-test, P < 0.05), and the myopic eyes were longer (treated eye, 3.96 ± 0.36 mm; untreated eye, 3.84 ± 0.27 mm; P < 0.01). There were no significant differences in nasal-temporal dimensions between the eyes of the two groups. In group 2, these dimensions were 7.41 ± 0.11 mm versus 7.41 ± 0.0.08 mm (P = 0.95). In group 3, they were 6.39 ± 0.49 mm. versus 6.41 ± 0.51 mm (P = 0.73). 

As noted earlier, treated fish were weighed before and 2 weeks after treatment. A small increase in weight, generally 1 to 3 g, was noted. However, there was no correlation between weight change and induced ametropia, a point also noted in our earlier study. ²⁸ In the additional groups of fish treated with lenses, positive and negative lenses were worn for 2 weeks, after which the goggle was removed and recovery from induced refractive errors was observed. The fish treated with negative lenses (group 5) developed −9.8 ± 1.9 D of myopia (n = 8; P < 0.001; Fig. 5 ). After 2 weeks of treatment with a +15 D lens, the treated eyes developed hyperopia (treated eye compared with control eye) of 8.05 ± 1.35 D (group 6; n = 5; P < 0.001) and 6.25 ± 2.87 D (group 7; n = 8; P < 0.001; Fig. 5 ). In addition, Table 2shows the change of refractive states of individual fish in group 5 before and after treatment with −12 D lenses, whereas Table 3shows results for individual fish wearing +15 D lenses (groups 6 and 7). After lens removal, the induced refractive errors decreased gradually (P < 0.001, ANOVA; Fig. 5 ). Refractive states measured 6 days after recovery were close to the baseline in all groups. All three groups showed no statistical difference between the two eyes 2 weeks after lens removal. 

Form Deprivation Myopia and Weight

After 2 weeks of treatment with translucent goggles, form deprivation myopia was produced in all three groups (groups 8, 9, 10) to significantly varying degrees (P < 0.001, ANOVA; Fig. 6 ). Expressed as average differences in refractive state between the two eyes, the highest amount was 11.9 ± 2.9 D and was found for group 8. In group 9, the myopia averaged 6.3 ± 2.5 D, whereas in group 10, the heaviest fish, the average was 2.3 ± 1.0 D. The treated eyes completely recovered to pretreatment refractive state levels within 2 weeks of goggle removal and showed no differences among the three groups. 

Effect on Contralateral Control Eyes

The results (Fig. 7)indicate a small but insignificant contralateral eye effect, with the fellow eyes of the negative lens groups dropping slightly below the horizontal while those of the positive groups remain more or less horizontal. This is also seen in the means of the individual data of Tables 2 and 3 . It is possible that these changes were caused in part by the normal change in refractive state as a function of weight, indicated by the graph shown in Figure 3 , particularly for the negative lens results. However, we think this is an unlikely explanation given that fish weights increased marginally (1–3 g, as noted earlier) during the 2-week treatment period and given the difference between the positive and negative lens treatments. Small contralateral effects were also noted in form deprivation results. For example, though the treated eyes of the lightest fish (group 8) showed an average myopic refractive state change of almost 12 D (Fig. 6) , the contralateral eyes changed approximately 1 D (from an average of +8.1 D to +7.0 D) in the same direction. These contralateral changes may be another example of the agreement between results for higher species such as chicks and lower ones such as tilapia. 

Discussion

This study demonstrates for the first time that in addition to form deprivation myopia, fish eye refractive development can be manipulated toward myopia and hyperopia with the use of positive and negative lenses; the refractive changes are a result of change in the axial length of the eye. Thus, early eye growth and refractive development of the eyes of lower and higher vertebrates, including primates, ³¹ appear to be guided by the visual environment. We suggest that the same mechanism(s) are involved and likely evolved at an early point in vertebrate evolutionary history. 

Although the results for the groups of fish treated with positive and negative lenses showed significant change in refractive state in the hyperopic and myopic directions, the changes do not fully compensate for the +15 and −12 D lenses used. It may be that the lens powers used exceeded the eye’s compensatory ability and that further testing with other lens powers is needed to clarify this point. However, we think it is more likely that a longer treatment period would produce greater compensation, a point also made by Wildsoet and Wallman ³² with respect to chicks. In fact, a recent study by Kisilak et al. ³³ demonstrates that chicks treated with −30 D lenses can fully compensate for the lens treatment if the lens is applied to the eye for a longer period. Unfortunately, because of reflections from the goggle surface, it was not possible to measure refractive state consistently with the lenses on the eye to monitor the level of compensation before the end of each experiment. The 2-week period of treatment was based on concern for the ability of the sutures to keep the goggles on the fish. 

The normal refractive state of tilapia in the 14 g group is approximately +11 D (10.9 ± 0.8 D) along the axis perpendicular to the iris plane. In fact, refractive state often varies with the axis of measurement in fish, and it is illusory to refer to a specific refractive state. ³⁴ The direction perpendicular to the iris plane is generally the most hyperopic. Moreover, fish often inhabit relatively monochromatic (blue) environments; thus, chromatic aberration is a factor to consider. A refractive state measured under white light conditions may be considerably less hyperopic at wavelengths that are closer to the blue end of the visible spectrum. ³⁵ Nevertheless, in this study, and with respect to the axis perpendicular to the iris plane, the hyperopia measured decreases with age and fish weight. 

It is possible that the small eye artifact of retinoscopy may be a significant factor in the change of refractive state measured during fish eye growth. ³⁶ The rationale for this point is that the retinoscopic reflection used in measuring refractive state emanates from the retina–vitreous border rather than from the receptor plane of the retina, thus biasing the finding in the hyperopic direction by an amount equal to the dioptric value of the thickness of the retina. Because retinal thickness varies little with eye size (the tilapia retina is approximately 200-μm thick), ²⁸ the smaller the eye is, the larger the artifact. Small eye artifact in relation to retinoscopy has been studied mainly in relation to small-eyed mammals, ^{35 37} but there is no reason to rule out the possibility that it is a factor in the study of the fish eye as well. Thus, the refractive states of the untreated heavier fish of group 1, which appear to level off at approximately 4 D of hyperopia (Fig. 3) , may in fact represent emmetropia or near emmetropia in the direction perpendicular to the plane of the pupil. For the 25- to 30-g fish studied with eye diameters of 7 to 8 mm, the artifact could amount to approximately +5 D, whereas for the smaller fish (approximately 14 g) with smaller eyes (approximately 5 mm in diameter), the artifact can be substantially greater. ³⁵ Thus, fish in group 6 that were treated with a +15 D lens showed +11 D of hyperopia on average before treatment. In fact, it has been noted in other species that the response to positive defocus is variable, sometimes even producing a change in the opposite direction with an elongated eye after treatment. ³⁸ The tilapia model may be a useful one for further research into the accuracy of retinoscopic and other refractive measurements of small-eyed species. 

Although fish eye size keeps increasing throughout life, the susceptibility of the tilapia eye to manipulation of the visual environment is strongly related to weight. Thus, the heavier the fish, the lower the amount of form deprivation myopia produced. When fish weight was approximately 100 g (group 10), it was more difficult to induce form deprivation myopia. Only −2.3 ± 1.0 D is induced on average in this group, compared with −11.9 ± 2.9 D in the lightest fish (group 8). In addition, a goggle with positive power induced higher levels of hyperopia in smaller fish (groups 6, 7). There may exist a critical period of eye development in fish, as in other animals such as tree shrews, in which induced myopia is not strictly dependent on the growth rate of the eye. ¹⁶  

It is still unknown how susceptible the fish eye is to visual manipulation during the earliest period, from hatching to approximately 10 g, and an additional study of this period is warranted. Furthermore, sexual maturation takes place earlier in aquaculture conditions when the fish are in tanks than when they are in ponds and lakes. ³⁹ Growth rate and therefore body weight can vary greatly, depending on conditions such as water temperature and availability of food. ⁴⁰ In fact, in our earlier study, no correlation was found between the increased weight of the fish during a 4-week treatment period and the amount of form deprivation-induced myopia. ¹⁹ However, this may not be the case during earlier periods of development. 

In the recovery process, the refractive error of the treated eye always stops changing when there is no difference between it and the untreated contralateral eye, regardless of the starting point. Fish are not different from other animal models in this regard. ^{16 17 38 41}  

As noted, the tilapia eye recovers completely, not only from form deprivation myopia but also from lens-induced myopia and hyperopia, regardless of the weight of the fish. Recovery from form deprivation myopia resembles that found in a guinea pig study, in which approximately half of myopia disappeared in the first day. ¹¹ However, generally, young animals completely recover in approximately 1 week. ^{11 17 38 42} By comparison, primates (monkeys) only recover to a limited extent from diffuser-induced myopia. ^{27 31} Though recovery varies with species, it still may be part of visually guided regulation of eye growth. The choroid of the fish eye may also play a role in this process, as found in other higher vertebrates. ^{23 24 43 44} In fact, the choroidal stroma of the teleost fish eye is rich in blood vessels, mainly veins, which make up the choroidal gland and the lentiform body. The role of these structures in short-term changes in ocular focal length remains to be investigated. 

 Supported by a grant from the Natural Sciences and Engineering Research Council of Canada (JGS).

 Submitted for publication October 23, 2006; revised January 16, March 1, May 30, and June 25, 2007; accepted August 14, 2007.

 Disclosure: W. Shen, None; J.G. Sivak, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Summary of the Numbers, Weights, and Treatments Received by the Groups of Tilapia Used in This Research

Group	Fish (n)	Treatment	Average Weight (g)
1	25	Untreated, freeze sectioned	26–101
2	6	Open goggle	33.3
3	7	+15 D, sacrificed + freeze sectioned	30.4
4	7	−12 D, sacrificed + freeze sectioned	15.8
5	8	−12 D lens-goggle + recovery	25.5
6	5	+15 D lens-goggle + recovery	13.9
7	8	+15 D lens-goggle + recovery	26.9
8	7	Form deprivation goggle + recovery	16.0
9	8	Form deprivation goggle + recovery	57.4
10	8	Form deprivation goggle + recovery	98.4

Figure 1.

 

Tilapia showing goggle/lens combination (top) and form deprivation goggle (bottom). Overall fish lengths were approximately 10 cm.

Figure 2.

 

The correlation between the fish eye size (mm) and body weight (g). Fish eye size is measured from the cornea to the retina (mm). There is a significant linear correlation between fish eye size and body weight during the early growth stage.

Figure 3.

 

The relationship between refractive state and weight of the fish indicates that the tilapia eye becomes less hyperopic with growth. Data were collected from measurements made on the group of 25 untreated fish (group 1) and from pretreatment measurements made for fish of the remaining groups and were fitted to the inverse exponential function providing the best correlation.

Figure 4.

 

Fish eyes were treated with lenses of positive or negative power over one eye. In addition, a ring-shaped goggle with an absent central lens was used as a control for the mechanical effect of the goggle (group 2, n = 6). In group 3 (n = 7), +15 D lenses induced +7.7 D of hyperopia, whereas in group 4 (n = 7) the −12 D goggle induced −8.5 D of myopia. Values given are in diopters and represent the differences between refractive states of treated eyes and untreated contralateral eyes. Error bars in this figure and in the remaining ones represent the SD.

Figure 5.

 

One group of tilapia (group 5, n = 8) was treated with −12 D lenses over one eye for 2 weeks and was allowed to recover. Two additional groups of tilapia of average weight 13.9 g (group 6) and 26.9 g (group 7) were treated with +15 D lenses for 2 weeks and were allowed to recover.

Table 2.

 

Refractive Data for Treated and Contralateral Untreated Eyes for Fish in Group 5 and Mean ± SD before and after 2 Weeks of Treatment with −12 D Lens

Fish	Pretreatment (D)			Posttreatment (D)			Interocular Difference after Treatment (D)
		Treated	Control Eye	Treated	Control Eye
1	7.5	6.8	−3.8	4.5	8.3
2	7.5	7.1	−4.1	5.3	9.4
3	9.0	9.0	−0.8	5.3	6.0
4	5.3	3.8	−7.1	3.8	10.9
5	6.4	6.0	−6.8	4.5	11.3
6	7.1	6.0	−6.8	4.5	11.3
7	5.3	4.9	−7.1	4.1	11.3
8	9.4	9.4	−3.8	6.4	10.2
Mean	7.2 ± 1.5	6.6 ± 1.9	−5.0 ± 2.3	4.8 ± 0.8	9.8 ± 1.9

 Refractive measurements were corrected for the effect of the air/water interface but not for the small eye artifact. Refractive errors are for hyperopic unless preceded with a minus (−) sign.

Table 3.

 

Refractive Data for Treated and Contralateral Control Eyes for Fish in Groups 6 and 7 and Mean ± SD before and after Treatment of with +15 D Lens

Fish	Pretreatment (D)			Posttreatment (D)			Interocular Difference after Treatment (D)
		Treated	Control	Treated	Control
1	11.3	10.9	18.8	10.5	8.3
2	9.8	9.8	20.3	10.5	9.8
3	10.9	10.5	18.8	10.5	8.3
4	10.5	11.3	19.6	11.7	7.9
5	12.0	11.3	18.8	12.8	6.0
6	8.3	8.3	14.3	4.9	9.4
7	6.8	6.0	14.3	6.0	8.3
8	5.3	4.9	11.3	5.3	6.0
9	8.3	6.8	15.8	7.52	8.3
10	5.6	5.6	7.1	4.9	2.3
11	6.4	5.6	7.5	2.6	4.9
12	6.8	6.8	13.5	4.9	8.7
13	5.6	4.9	7.5	5.3	2.3
Mean	8.3 ± 2.4	7.9 ± 2.5	14.4 ± 4.8	7.5 ± 3.3	6.9 ± 2.5

 Refractive measurements were corrected for the effect of the air/water interface but not for the small eye artifact and indicate hyperopia.

Figure 6.

 

Form deprivation myopia and recovery for three different weight (age) groups of tilapia. The smallest fish (group 8, n = 7; average weight, 16.2 g) were estimated to be 4 months of age, whereas the fish labeled group 9 (n = 9; average weight, 57.4 g) and group 10 (n = 8; average weight, 98.4 g) were estimated to be 7 months and 10 months of age, respectively. All tilapia were treated with a translucent goggle over one eye for 2 weeks. The results are given in diopters for the average difference in refractive state between the treated and the untreated contralateral eyes.

Figure 7.

 

The two graphs show the change in refractive states in diopters of positive and negative lens-treated eyes and the untreated contralateral eyes before and after treatment to induce hyperopia and myopia.

The authors thank Nancy Gibson, Robin Jones, Kelley Moran, Vladimir Bantseev, and Raj Suryakumar for their assistance.