Barrett—On Entoptic Vision. 55 
perceptible without using homocentric light. They can, if present, 
readily be seen upon looking at the sky or any illuminated surface, 
such as a sheet of white paper placed at a convenient distance 
from the eye. Keeping the head and eyeball as far as possible 
motionless, the observer can mark with a pencil on the sheet of 
paper the length and breadth of the projected image of one of 
these musce. For this purpose the paper can be about one or 
two feet (30 or 60 em.) from the eye, so as to be within easy 
reach of a pencil held by the observer. If the image is too 
small at this distance, the paper can be placed further off, and the 
size of the projected image judged by marks previously made on 
the paper. Knowing the distance of the paper, the length and 
breadth of the retinal shadow is to that of the projected image as 
the distance of the nodal point of the eye from the retina (viz. 
16 mm.) to the distance of the sheet of paper from the nodal 
point, or practically from the eye. Thus, in the case of a particularly 
persistent and large musca I recently noticed in my right eye, 
the length of the projected image of the filament was about 150 
mm., and its breadth 2 mm., the sheet of paper being 60 cm. 
from the eye. Accordingly the retinal shadow was 
16 
600 150 = 4 mm. long, 
and Le 2 = 0:053 broad 
600 27 3 mm. broad. 
As this musca was very near the retina, its actual size was practically 
the same as the above. 
In making this experiment one eye must be shut, and the eye 
used should be partly closed so as to give a sharper and darker 
shadow. Looking through an aperture made in a piece of card 
with a stout needle, gives a much better result, as already explained : 
here, however, the magnification of the image follows another law, 
as will be explained below. In this case the head can be rested 
on a support—such as the left hand, with elbow on the table—the 
eye being a foot or so from the sheet of paper, and the perforated 
card held or fixed close to the eye; excellent drawings and measure- 
ments of the reticulated and cellular structure of the musce, with 
the knots and loops of their filaments, can thus be made. Their 
