294 Mr. R. II. M. Bosanquet on the Present 



hand-blown wind ; and where these investigations are to be 

 made on any large scale it- will be preferable to employ a small 

 engine, of a kind affording uniform rates of motion, fortius 

 as well as for many other acoustic experiments. 



This suggestion forms the key to a conception of an acoustic 

 laboratory, which is, I believe, new, and would, it seems to 

 me, constitute a considerable step from an experimental point 

 of view. This is, that, in an acoustic laboratory, power should 

 be employed to produce all the effects required, in a continuous 

 manner where continuity is suitable, and without effort or 

 attention on the part of the investigator when the experiment 

 is once arranged. It is easily seen that determinations of all 

 sorts, which now present almost insuperable difficulties, would 

 become perfectly easy under such conditions ; and accurate 

 knowledge would soon take the place of many of our guesses 

 of today, through numberless extensions of the work now pos- 

 sible. 



I proceed to mention the principal other 'subjects which 

 suggest themselves as suitable for research under these con- 

 ditions. 



Aerial Mechanics generally. 



So far as our knowledge of the mechanics of fluids, and 

 especially of the air, has progressed beyond its elements of late 

 years, the progress has been mainly in the mathematics of the 

 subject. No doubt the deficiency of experiments arises mainly 

 from the difficult nature of the subject matter; but it is also 

 certainly due to the small extent to which real effort has been 

 made to devise experiments of easy and certain execution, which 

 shall supplement and check the mathematical investigations. 

 No doubt a beginning has been made, and the experiments 

 which are known have been made to tell their tale with admi- 

 rable perseverance and ingenuity ; but it is my opinion that 

 the mathematical structure has, in some respects, been built 

 up too rapidly. 



The questions which arise are of great difficulty ; and I do 

 not purpose here to enter on any of them. It is sufficient to 

 say that I believe there will be no difficulty in devising a 

 number of experiments by which the simpler cases of aerial 

 motion may be examined in detail. Until the whole circum- 

 stances of crucial experiments of this kind have been made cut 

 experimentally, I shall continue to feel grave doubts as to the 

 stability of certain portions of the mathematical edifice. 



I may allude to a few experiments of importance. Much 

 has been built on a mathematical solution of the following- 

 problem: — If a circular disk, closing a circular hole in an in- 

 finite plane, execute oscillations at right angles to the plane, 



