326 Dr. G. J. Stoney on the Limits of Vision : 



increased without diminishing too much the aperture of the 

 little lenses. 



With a mosaic such as is described above, the diameter of 

 each patch subtends about a degree and eight-tenths (1°'8) at 

 the centre of the hemisphere. Accordingly, the interval be- 

 tween two objects in nature would need to subtend an angle 

 of about a degree and three-quarters at the insect's eye, to be 

 distinguishable as two by the insect. Hence, if as far off as 

 ten inches, the distance at which we see most distinctly, they 

 would need to be separated by nearly the third of an inch to be 

 seen by the insect as more than one object ; while, if close to 

 the insect, only one-tenth of an inch of!', the separation would 

 need to be about the same as that which the human eye is 

 capable of distinguishing at a distance of ten inches. Thus, 

 the insect cannot see more detail upon its own antennae, close 

 as they are to it, than we can with our naked eye. We 

 must, therefore, dismiss from our thoughts the mistaken im- 

 pression that insects see very minute objects far beyond 

 human vision. On the contrary, their vision is imperfect 

 compared with ours. Still, it is evidently quite enough to 

 enable a bee to be guided in its search after honey by the 

 markings upon a flower, or effectually to assist a fly in its 

 wanderings about the room, or in sopping up its food. 



We have next to consider how this mosaic is formed. For 

 this purpose let us again turn to our model. Suppose 6400 

 hollow conical funnels to be pro- 

 vided, each one inch long. Let Fig . 2> — The Funnels of an 

 them be slightly more than the Insect's Eye (diagrammatic), 

 thickness of a lead-pencil at their 



larger end, and tapering from this ^ « 



down to a diameter of a sixteenth / J \S<^\\ 



of an inch at Jieir smaller end. AA\ ^>' h 



Let the insides of these funnels be /Ov^X 



blackened so as to stifle any light AZ^o*''' 



that falls on them. Fit a small A?Q^y 



lens of one-inch focus into the f/^ f ' 



larger end of each, and then pack ' / 



the funnels somewhat like the cells a. Outer surface, b. Primary 



of a honeycomb, over the hemi- surface. 



sphere spoken of above, the larger 



ends outwards, and the smaller planted on the middles of the 



little patches that were marked out on the hemisphere. The 



little lenses will then lie on an outer hemispherical sheet 



eighteen inches in diameter. 



Let us fix our attention upon one of the little lenses, and 

 consider how it operates. The light from distant objects in 



