IIO THE AMERICAN NATURALIST. [Vor. XXXIV. 
Coregonus the outer segments of the single cones have a cen- 
tral deeper staining core. In this respect they agree with the 
rods. One of us was in doubt concerning these single cones, 
inclining to the opinion that they were rods (Eigenmann ('99), 
p. 548). 
The rods are not taken into consideration in this paper. It 
may be said, however, that generally the number of rods is 
inversely proportional to the number of single cones. 
The eyes examined were taken, for the most part, from 
museum specimens preserved in alcohol, and originally not 
intended for anatomical or histological purposes. The eyes 
were sectioned in paraffine, and the sections stained with 
hemalum and eosine, or the Biondi-Ehrlich three-color mix- 
ture, or Weigert’s haematoxylin; the latter gave good results 
even with the alcoholic material. No attempt has been made 
in this paper to describe the extent of variation in the patterns 
noted in different parts of the same retina. 
The earliest notice that may have a possible bearing on the 
present subject is that of Guennelon (1686). According to 
Ziegenhagen, he noted that the retina of Gadus eglifinus is 
striped. We are not at all sure that this observation related 
to the arrangement of the twin and single cones. 
The first undoubted notice that the twin and single cones 
may form a definite pattern was that of Hannover (43), who 
figured the pattern of the *'Ródspetten" and of the pike. 
Both patterns figured are those of our diagram D. We find 
the pattern of the pickerel to be that of diagram Æ. 
Much later, Müller (72, p. 59) says that in Perca ffuvia- 
tilis the twins are in excess, with such an arrangement that 
each single cone is separated from its neighbor by the twins, 
not counting the rods. He also stated that in many fishes 
only single cones exist. | 
Nunneley (58) and Krause ('68) described cones in the 
eel, and here, according to Krause, only single cones are found. 
Friis (79) examined a large series of fishes and found that in 
Accipenser and Anguilla only single cones exist. He found 
twin cones only in a large number of species, but states that in 
all of these, single cones are always to be found about the ora 
