(20's) 
‘ Untersuchungen tiber das Sehorgan der Arthropoden,’ from 
which it will be gathered that Miiller’s theory of mosaic 
vision is the one that has proved most generally acceptable. 
The earlier entomologists appear to have been contented with 
the simple statement that insects possess the sense of sight, 
or that they see; but Johannes Miiller, about sixty years ago, 
propounded a theory as to the nature of their vision, and this 
has been known since as the mosaic theory. It may be 
briefly described by saying that he thought an insect saw by 
means of a picture, as our own eyes do; and that the picture 
in the insect eye was composed of a large number of separate 
pieces, each facet contributing a small piece. 
This is evidently the simplest view that can be at all main- 
tained, and we may tale it for granted that it was pretty sure 
to be the one that would first occur. There is no way of 
explaining insect-vision so easy as that of saying it is 
like our own, and then adding that it is broken up somewhat 
because of the numerous facets of the insect-eye; and it is 
probably owing to this primitive simplicity that the mosaic 
theory has had so long a career; for though it has been 
frequently opposed and even altogether buried, it has always 
recurred again when rival theories proved unsatisfactory ; and 
I am of opinion that the vitality of the theory is chiefly due to 
its ease of comprehension by making the vision of the insect- 
eye so very similar to that of our own eye. But... 
I believe that my much-gifted and ever-to-be-remembered 
friend, the late W. K. Clifford, once stated that the human 
or vertebrate eye is a very imperfect optical apparatus; 
and he was undoubtedly correct in this, as he also was 
in explaining that the defects of the apparatus itself are in 
part removed, and its deficiencies in part supplemented, by 
secondary means. The vision of the human eye is by picture 
cast on the retina, which is practically a highly sensitive 
recipient screen, placed at the back of a camera obscura; but 
the picture so limned is, as a representation of external 
nature, very imperfect: for instance, everything is flat, and 
it is indistinct except in its centre. Now if picture-vision 
is to be of much use to the insect it should have the 
secondary means of making it useful. What are they? 
