644 PROCEEDINGS OF THE ACADEMY OF [Scpt.,, 



the violin instead of from a metal bar. Nevertheless we got no reac- 

 tion. Weld does not make clear that his ants were always isolated 

 from all except aerial vibrations, nor that their reactions were constant 

 under repeated stimulation. It seems to us possible that his ants may 

 have reacted at times to vibrations of the solid base upon which they 

 rested and to which, as we shall show presently, they are very sensitive, 

 or their supposed reactions may have been accidental. Certainly our 

 own experimental evidence gives us not the least reason to suspect 

 that ants are stimulated by sound waves in air. 



Having reached this conclusion we next endeavored to ascertain 

 whether ants would respond to vibrations of the solid base upon which 

 they stood. Wlien a nest containing Stenanima fulvum was held in 

 the air within a centimetre or so of the woodwork of a piano, and the C, 

 giving 261 vibrations per second, w^as struck, no response followed. 

 When, however, the nest was allowed to rest on the woodwork and the 

 note was again sounded, almost all the ants started forward simul- 

 taneously. Thus a vibration that comes to an ant through the air 

 is not necessarily followed by a reaction, though the same vibration 

 when it reaches the ant through a solid may be very stimulating. All 

 the eight species of ants with which we experimented were thus stimu- 

 lated, though they failed to react to the same vibrations in the air. 

 The range of the different species was by no means uniform. All 

 reacted to the 27 vibratioas per second and to higher notes up to a 

 certain pitch characteristic for each species. Crcmastogastcr reacted 

 at 522, but to no higher note. The superior limit for Formica fusca, 

 var. suhsericea was 1,044, and for Lasius latipes and Stigmatomma 2,088. 

 Stenamma always reacted at 2,088, usually at 3,915, but failed at 4,176. 

 Camponotus regularly reacted at 3,480, but failed at 4,176. Formica 

 sanguinea, which invariably responded at 2,088, occasionally did so 

 at 4,176, a pitch regularly reacted to by Lasius umhratus. Thus each 

 species seemed to have a characteristic superior limit for stimulating 

 vibrations received through solids. 



Ants are not only sensitive to the tones of a piano transmitted 

 through a solid, but they are also sensitive to vibrations from other 

 sources similarly transmitted . This is well seen in the following experi- 

 ment on Stenavima. When the edges of two Petri dishes were 

 rubbed against each other in the air, the ants did not respond; l^ut 

 when the edge of the dish in which the ants were held was rubbed even 

 lightly by the edge of another dish, they reacted with great precision. 

 These reactions occurred even when the Petri dish containing the ants 

 was floating on water and the edge of the vessel containing the water 



