3 i4 SCOTT'S LAST EXPEDITION 



invariable constant. Thus a piece of lead or other substance 

 weighed at sea level on a spring balance would be heavier at the 

 Pole than at the Equator by about 5 parts in 1000. This differ- 

 ence may be considered as partly due to the weakening, as one 

 goes towards the Pole, of the centrifugal force of the earth's 

 rotation, and partly due to the increased force of attraction by 

 virtue of the flattening of the earth at the poles, and consequent 

 shorter distance from the centre of attraction. 



Dealing with such small differences in value of the gravity 

 constant ' g,' it becomes essential, if any theories are to be tested, 

 that observations should be carried out with the most extreme 

 accuracy. 



The universal method of measuring ' g ' is by noting the 

 time of swing of a pendulum, and as absolute measurements are 

 of the utmost refinement and delicacy, comparative measurements 

 are nearly always undertaken, and referred to similar observa- 

 tions at the standard station in Potsdam. 



The pendulums used were of Col. von Sterneck's pattern and 

 are gilt, three in number, swinging from agate planes in two 

 directions at right angles to one another. The pendulums are 

 swung at atmospheric temperature and pressure, and corrections 

 are applied to reduce to normal. 



The time of swing is measured by the method of coincidences, 

 with reference to a special clock making an electric contact each 

 second. This clock is used as the standard and rated (if possi- 

 ble) to Hoth second in the day, by observation of stars at 

 meridian transit. 



With care the value of ' g ' should be accurate, after all cor- 

 rections are applied, to one part in a million, but under the par- 

 ticularly unfavourable conditions in the Antarctic it is doubtful 

 if a much higher accuracy than i in 250,000 is obtainable. 



The difficulties experienced with the instrument in the Ant- 

 arctic, though apparently trifling at this distance, were very real 

 at the time. For instance, observation on stars for clock rate 

 was usually complicated by a temperature of 40 with a slight 

 wind, quite sufficient to keep one nursing his nose, and to be very 

 careful not to put one's eye to the telescope lest it freeze and 

 remain there. Other little troubles, such as the stopping of the 

 clock by a bodily shift of the wall of the hut upon which it was 

 hung, also tended to reduce the accuracy of the observations. 



