50 



suspended from one pan of an analytical balance is just sufficient 

 to keep the balance in equilibrium. The adjustment of this 

 current is effected by the movement of a sliding- contact along 

 slide wires, and this movement is in its turn controlled by the 

 flight swing of the pointer attached to the balance beam, as the 

 latter moves from its equilibrium position. When the current 

 — and hence the weight on the second pan of the balance — 

 reaches a pre-arranged value, a subsidiary circuit is automatically 

 closed, and a small phosphor-bronze ball of known weight is 

 deposited on the pan above the magnet, the sliding contact is 

 drawn back to its initial position, and the cycle of operations 

 recommences. 



The arrangement of the circuits is such that the distance 

 of the sliding contact from its zero position is to a close approxi- 

 mation linearly related to the current, and hence a recording- 

 ammeter is not needed, as a record on a rotating drum of the 

 slider position is sufficient to give the required data. The records 

 consist of a series of stepped curves and a very open scale is 

 obtained. 



The apparatus can be used with no loss of sensitivity up to 

 the maximum load the balance is designed to carry. Further, 

 the sensitivity can be very simply adjusted, so that both rapid 

 and slow changes of weight can be recorded. 



The apparatus can be employed with advantage in experi- 

 ments involving a continuous measurement of increasing or 

 decreasing weight, and its application to the study of sedimenta- 

 tion and flocculation of soil particles, and the evaporation of 

 water from fibres is illustrated in the present paper. 



The earlier work was carried out with the assistance of the 

 Cambridge Instrument Company, 45, Grosvenor Place, London, 

 and the completed form of the instrument has been placed by 

 them on the market. 



XVI. (a). B. A. Keen. " The Oden-Keen Automatic 

 Balance." Proceedings of the Fourth International 

 Conference on Soil Science. Int. Inst, of Agric, 

 Rome, 1924. 



(vSee preceding paper for abstract.) 



XVII. R. A. Fisher and Sven Oden. " The Theory of 

 the Mechanical Analysis of Sediments by Means of 

 the Automatic Balance." Proceedings of the Royal 

 Society of Edinburgh, 1924. Vol. XLIV., pp. 98- 

 115. 



Ideally the mechanical analysis of a soil should enable us 

 to state what fraction of the soil consists of particles smaller 

 than any assigned size. In 1916 OdeVi showed that the distribu- 

 tion by size of the particles could be obtained by a sedimentation 

 process. The necessary sedimentation curve may most readily 

 be obtained by means of the Od^n-Keen automatic balance. 

 (Sec paper XVI.) The present paper consists of : — 



(i) A simplified mathematical statement of the theory of the 

 changes taking place in the fluid during sedimentation, showing- 



