308 



SCIENCE. 



[N. S. Vol. XII. No. 295. 



1900, Eigenmann and Shafer described the 

 different patterns of twin and single cones 

 found by them in the retina of several dif- 

 ferent species of fishes. No attempt was 

 made by them to determine the modifica- 

 tion of any of the patterns in different 

 parts of the same eye of any species. 



The present paper deals with the modifi- 

 cation of the pattern in the large- mouthed 

 black bass, Mieropterus salmoides. The ques- 

 tions more particularly dealt with are : 



I. Is the pattern of the twin and single 

 cones the same over the entire retina ? 



II. What relation does the direction of 

 the rows of cones which go to make up the 

 pattern bear to the surface of the eye ? 



III. What is the difference between the 

 number and size of the elements in the 

 young and old fish ? 



Several series of tangential sections were 

 cut from a band passing from the anterior 

 edge of the cornea around the back of the 

 eye to the posterior edge of the cornea ; 

 other series were secured from a band pass- 

 ing from the upper edge around the lower 

 edge of the cornea. 



I. The variation of the pattern. 



The general variation in the twin and 

 single cones in this species is that of Eigen- 

 mann and Shafer's pattern D. In this pat- 

 tern the twin cones are so arranged that if 

 the lines joining the centers of the compo- 

 nents of a twin (i. e., the axes) were con- 

 tinued they would form a square ; a single 

 cone is placed in the center of this square. 

 The division lines separating the compo- 

 nents of the twin cones thus point toward 

 the single cone. That is, the division lines 

 form I'ight angles with the sides of the 

 square. This ideal pattern for this species 

 is most nearly approached over the an- 

 terior and posterior surfaces of the eye. 

 As we go from the anterior and posterior 

 edges of the cornea toward the wider parts 

 of the eye, the pattern changes from a 

 square to a rhombus. Its area at the same 



time increases until we approach the back 

 of the eye itself, where the patterns are 

 again smaller and closer together ; even 

 crowded. 



Immediately at the upper and lower 

 edges of the cornea the division lines sepa- 

 rating the two parts of the twin cones in- 

 stead of pointing toward the single cones 

 are turned until they point almost directly 

 toward each other. At these points, the 

 square has varied to a rhombus of which 

 the two obtuse angles are almost one hun- 

 dred and eighty degrees. The single cone 

 remains in the center of this modification 

 of the square. As we go from the upper 

 and lower edges of the cornea toward the 

 back of the eye, the rhombus is quickly 

 changed into a square again. In other 

 words the double cones soon have their di- 

 vision lines turned again in the direction 

 of the single cones. Except very near the 

 cornea, the patterns in the band from the 

 upper to the lower edges of the cornea are 

 much more crowded than in other parts of 

 the eye. 



II. The relation of the pattern to the eye. 

 A study of the modification of the pattern 

 as described in the first section shows that 

 such a modification is brought about on the 

 surface of the eye if the axes of the twin 

 cones lie on two series of circles. The cen- 

 ter of one of these series of parallel circles 

 lies approximately at the upper edge of the 

 iris, the center of the other approximately 

 at the lower edge. These two series of cir- 

 cles cut each other at right angles near the 

 anterior and posterior edges of the iris and 

 cut each other at more and more acute 

 angles at the top and bottom of the iris. 

 The extreme modification that would be 

 brought about by the close adherence of the 

 twins on these lines is relieved by the inter- 

 polations of additional double rows of single 

 and twin cones. 



III. The patterns in the young and old 

 fish. A comparison of the eye of a young 



