RETINA OF LATERAL EYES OF SPRENODON PUNCTATUS. 319 
of this theory. The spiral has been regarded by some as 
continuous with longitudinal fibrils in the inner segment of 
the visual cell (Ritte, 1891), by others as distinct from the 
inner segments and having no connection with it (Krause, 
1895). Hesse himself writes of a double spiral twisting in 
bony fishes, amphibians and reptiles, but could not trace the 
end of it. Howard ( 18 ) describes a well marked axial core 
in the outer segment of each rod in the retina of various 
vertebrates. These rods in the fresh state showed transverse 
banding. In a later paper ( 19 ) he figures a spiral appearance 
of the outer segments of certain cones in Necturus, and 
regards it as possibly due to the irregular separation of the 
transverse discs by vacuoles. 
Bernard ( 2 , 3 , 4 , 5 ) holds an entirely different view as to 
the structure of these outer segments. They are, according to 
him, irregular masses of protoplasm, squeezed out from the 
inner segments of the rod or cone-cells, and passing between 
the processes of the pigment epithelium. He only gets these 
results in certain preparations, and his figures suggest that 
they are due to imperfect fixation of his material. 
Between the outer and inner segments of the ordinary 
single cones in Sphenodon is a large globule of a fatty or 
oily nature which turns black with osmic acid, i.e. in Flemming 
preserved material (figs. 4, 6, 7, 8, o.g.). Such oil-globules 
are well known to be characteristic of the cone-cells of birds, 
reptiles and amphibians (Schafer, 33 ). After fixing with 
Zenker's fluid or acetic bichromate the oil-globule remains 
unstained with all the stains I have used (fig. 5, o.g.). 
The inner segment of the cone is somewhat cylindrical in 
shape, and contains the ellipsoid and the paraboloid. The 
former lies next to the oil-globule, and the latter nearer the 
membrane limitans externa. The ellipsoid differs much in 
detailed appearance according to the method of preparation. 
In material fixed in acetic bichromate and stained with picro- 
nigrosine it has a slightly granular appearance, and is pos- 
sibly vacuolated (fig. 5, ell. 1 ). In osmic acid (Flemming) 
material it has a denser and much more uniform appearance 
