844 THE AMERICAN NATURALIST.  [Vou. XXXII. 
brates are like those of invertebrates, and that the retinal 
cones in man have a structure something like that shown in 
the accompanying diagrams (Fig. 9). In Fig. 9, a cone is seen 
in longitudinal section with the radiating fibrils on the right of 
the axial fibre; on the left are projected three color curves 
with maxima nearly opposite RX, Gr, and V. 
Let us assume that the longest visible ether waves produce 
the greatest stimulation when vibrating at right angles to the 
longest retinidial fibrils, and the shortest visible waves produce 
the greatest stimulation when vibrating at right angles to the 
shortest fibrils. All the fibrils should be stimulated to some 
extent by all visible ether waves, and there should be a gradual 
diminution from the maximum to the minimum stimulation of 
any fibril according as it becomes either too short or too long 
to give the maximum response to an ether wave of a certain 
length ; or provided the fibril has the optimum length, accord- 
ing as it diverges from a position at right angles to the plane 
of vibration and to the line of propagation of the wave, toward 
one parallel with the plane of vibration and at right angles to 
the line of propagation. Now let us see what will happen if 
we use the simplest possible stimulant, namely, a ray of light 
consisting of one wave length only and vibrating in one plane, 
that is, a polarized ray of monochromatic light. Then let us 
imagine, merely for economy of words, that the fibrils most 
stimulated by certain ether waves become luminous with the 
corresponding colors. Then (1) if a ray of polarized red light 
passes lengthwise of the cone, it should stimulate most the 
longest fibrils and those most nearly at right angles to its plane 
of vibration and to the long axis of the cone. This should pro- 
duce a band of red fibrils, brightest opposite R, as indicated by 
the thickness of the lines, and fading out above and below that 
level, where the fibrils become either too short or too long to 
make the maximum response. In a cross-section of the cone 
opposite Æ (Fig. 9, B), the fibrils at right angles to the plane of 
vibration, say at a, 4, should be the brightest, fading out on 
either side towards the fibrils at c, d, which, being parallel with 
the plane of vibration, should be stimulated but very little, if 
at all. 
