very Small Portions of Time. 319 



as the former. As the laws of the motion of such a magnet are 

 accurately known, it may be calculated with precision how much 

 the velocity of the magnet must have been altered by the current 

 in order to produce the observed change in the oscillations, and 

 from this again may be* determined how long the force must have 

 acted in order to produce this effect. The best mode of obser- 

 vation is to permit the current to act when the magnet is passing 

 the meridian, and when the direction of its motion coincides with 

 that produced by the electro-magnetic force. In this case the 

 calculation of the time is very simple ; it is only necessary to 

 multiply the difference between the arcs of oscillation before and 

 after the operation of the electro-magnet with a constant factor. 

 The magnitude of the latter depends only upon the strength of 

 the current and the time of oscillation of the magnet, As the 

 electro-magnetic force may be increased at pleasure by increasing 

 the number of coils and of voltaic elements, it is possible in any 

 time, however small, to produce a sensible effect upon the magnet. 



In applying this method, it is necessary so to arrange matters 

 that the commencement and the end of the galvanic current 

 mentioned above shall exactly coincide with the beginning and 

 end of the process whose duration is to be measured, which of 

 course may be effected in different ways, dependent upon the 

 special object of the measurement. This procedure possesses 

 the great advantage, that it renders the clockwork with constant 

 rotation unnecessary. Up to the present time, indeed, the pro- 

 blem of constructing such instruments is only approximately 

 solved, and all of them require constant control. In short, 

 simpler and more easily managed apparatus are necessary here. 

 The first invention of such is due to Pouillet, in the year 1844; 

 he made a proposition for artillery purposes which was applied 

 practically in some cases, ~but has not been used further, on 

 account of certain specialities which detract considerably from 

 the accuracy of the instrument. After him I have been the first 

 to make use of the method for physiological purposes. By 

 observing the magnet in the highly convenient and delicate 

 manner introduced by Gauss and Weber, which consists in 

 attaching a mirror to the magnet, and determining the constant 

 factor necessary to convert the differences of oscillation into dif- 

 ferences of time, in a more accurate manner than Pouillet, I have 

 been able with comparatively simple apparatus to make accurate 

 determinations up to j^^dth. part of a second. To extend the 

 delicacy of the measurement beyond this was of no interest to 

 me, and would simply have unnecessarily increased the difficulty. 



I now come to my measurements of the physiological processes*. 



* Completely described in Muller'is Archives, 1850. 



