PATTERN-DISCRIMINATION IN VERTEBRATES 215 



dog's retina would suddenly appear blurred under those condi- 

 tions. He was unable to establish, however, whether these 

 effects were the result of accommodation or were due to astig- 

 matic refraction in the dog's eye. 12 



Dr. Cobb, in a later and more extensive skiascopic examina- 

 tion of Dog 1, without mydriatic, found occasional fluctuations 

 of refraction, varying from 0.25 to 0.75 D. These fluctuations 

 were seldom sustained for any considerable interval, and could 

 not be elicited by moving the experimenter's fingers, or even 

 food, to and fro before the dog's eye. While the most natural 

 inference is that the fluctuations were due to accommodation, 

 it would also be possible to account for them on other supposi- 

 tions. It was impossible to control the dog's fixation, and rota- 

 tion of the eyeball changed the length of the optical path. (This 

 extended from the light-source to the mirror, thence through the 

 comparison lens into the dog's eye and to the retina, thence 

 back through the comparison lens to the experimenter's eye.) 

 This change might be made by difference in angle of incidence 

 at the dog's cornea, by differences in refractive index of the 

 different layers of the dog's lens, and also by unevenness of con- 

 tour of the dog's retina. It seemed clearly established, how- 

 ever, that parallel rays, or rays proceeding from an object 20 

 feet or more from the eye, are sharply focused on the retina of 



12 1 have elsewhere asserted that the dog's mechanism of accommodation is use- 

 less unless it serves in some way to change the radius of curvature of the cornea, 

 as is the case in some birds, but, according to Hess, is not true of the mammals. 

 This assertion was based on my acceptance of a reference by an American writer 

 to Frey tag's work (Die Brechungsindices der Linse und der fliissigen Augenme- 

 dien bei der Katze und beim Kaninchen. Arch. f. verg. Ophth., vol. I, 1909-10). 

 In this reference Freytag is cited as saying that the refractive indices of the 

 lens and of the fluid media of the dog's eye are practically identical, and that several 

 other mammals suffer under the same condition. Since I made the above remark, 

 I have procured a copy of Freytag 's original article which before had been inac- 

 cessible to me. It appears that Freytag was incorrectly quoted by his reviewer. 

 He actually gives as mean values of the refractive indices in young and old dogs: 

 for the aqueous humor, 1.3349; for the vitreous humor, 1.33483; and for the lens, 

 values ranging between 1.4498 and 1.4666, depending on age. He reports a com- 

 parable difference between the refractive indices of the lens and fluid media in all 

 the mammals which he studied. 



I am satisfied that some individual dogs make very little use of the mechanism 

 of accommodation. If such is generally true, it would seem that the defect is 

 retinal, rather than in the accommodatory apparatus itself. If it may be assumed 

 that the stimulus to accommodation is indistinctness of the retinal image, it is 

 evident that an animal whose retina is relatively insensitive to detail would have 

 relatively slight stimulus to accommodation. 



Incidentally, the solution of a number of extremely interesting problems in 

 the relation between vision and field-behavior must await the determination of 

 the range of accommodation in representative infra-primate mammals. 



