310 Instrument for exhibiting^ a certain Optical deception. 



cord. This done, the most difficult part was yet to be performed, 

 and this was to swing him clear from the shelf. To do this, they 

 procured a large hlue fly, (2,) which they made fast to the web or 

 cord, and by continuing to roll the fly over in the web or cord, and 

 making it fast at each turn, they were enabled to swing the snake 

 about three inches clear from the lower shelf. The snake was alive 

 and active when discovered, and five days after when I was there 

 last, he was still living suspended in the manner above described. 

 During the day no spiders were visible in or about the web, but at 

 night there were three, much smaller than the common fly, seen 

 feeding upon the body of the snake. 



Very respectfully your obedient Servant, 

 ■' '■ D. Ltman Beecher. 



The enquiry has been made, whether the snake might not have 

 fallen by accident into a web previously made, and thus have become 

 inextricably entangled. — Ed. 



Art. XVII. — Description of an Instrument for exhibiting a cer- 

 tain Optical deception; by Prof. E. S. Snell. 



In the London Philosophical Transactions of 1825, Dr. Roget 

 describes and explains a curious optical deception, which presents 

 itself, when a person looks through a series of narrow vertical aper- 

 tures at a carriage wheel, as it rolls along on the ground. The 

 wheel appears to have simply a progressive, without a rotary motion, 

 as though it were sliding, and the spokes seem to be more or less 

 curved. The two spokes, however, which are in a vertical line, and 

 therefore coincident in direction with the apertures through which 

 the observer looks, remain in appearance, straight, while all the oth- 

 ers, both on the right and left, have considerable curvature, present- 

 ing their concavity upward, or towards the highest spoke. 



The illusion is the same, if the wheel revolve on a fixed axis, while 

 the vertical bars have a progressive motion between it and the ob- 

 server. The direction in which the spokes appear to be bent may, 

 however, be reversed in this form of the experiment. If the bars 

 move in the same direction as the lower part of the revolving wheel, 

 the concavity of the spokes will be upward ; if in the same direction 

 as the top of the wheel, the concavity will be downward. The rea- 



