40 



METHODS OF RECORDING 



The traditional physiological recording device is, of course, the smoked- 

 drum kymograph. The writing points may be operated mechanically by 

 cords, pneumatically by tambours, hydrostatically by floats, or electrically 

 by a variety of electromagnetic devices. Advantages of the kymograph are 

 that it is simple, the recording medium is cheap, and the results of extremely 

 long experiments can be displayed in records of manageable size. By the 

 same token, however, it is inappropriate for use in the field for which 

 electronic techniques are so admirably suited — the display of rapid pheno- 

 mena. In general the recording of quantities electronically derived is 

 achieved by other means, though Keynes^ has employed a kymograph to 

 register counts from radioactive isotopes via an electronic scaler. 



Within the frequency range 0-100 c/s the electromagnetic penwriter is 

 satisfactory. Penwriters may be of the moving-coil or moving-iron variety, 

 and the record may be made in a number of ways. In the most usual, ink is 

 fed under hydrostatic pressure along a thin tube to a nib at the end of the 

 writing arm. In this case the writing medium is ordinary paper and is 

 therefore cheap, but difficulty may be experienced from clogging of the 

 supply pipes, and an accidental overdriving of the recorder can cause 

 widespread distribution of ink. In another method the writing point is a 

 hot stylus which discolours a special paper as it passes over it. A third is 

 entirely electrical and uses 'Teledeltos' paper; a current of about 50 mA is 

 passed from the writing point into the paper and away via a metal table 

 which supports it. At the point of entry of the current the paper is burned, 

 leaving as a trace a fine black line. The latter two methods are extremely 

 reliable, but the special papers required are naturally expensive. 



The upper limit of frequency response of penwriters is set by the moment 

 of inertia of the mechanical system, particularly of the writing arm, which 

 has to be long if the machine is to write in anything like rectangular co- 

 ordinates. In one ingenious commercial recorder the writing arm is elimi- 

 nated and is replaced by a jet of ink which is sprayed from a nozzle which 

 rotates with the mechanical system. In this manner the upper frequency 

 limit has been raised to 500 c/s. 



Another commercial recorder has a frequency range which — for biological 

 work — is for practical purposes unrestricted, but this is obtained at the cost 

 of considerable uncertainty in the accuracy of any particular fragment of 

 the record. The input signal is used to control the frequency of a high- 

 frequency oscillator, whose output feeds a battery of band-pass filters to 

 each of which is connected one of a battery of fixed styli, which mark the 

 paper when excited {Figure 40.1). Thus any particular instantaneous 

 signal voltage causes oscillations of a definite frequency, and these are 

 passed by one of the filters and cause the appropriate stylus to mark the 

 paper. By choosing a sufficiently high mean oscillator frequency, the 



638 



