26 SECTIONAL ADDRESSES 



the earth's rotation, the apparent value of the gravitational intensity 

 increases in passing from equator to pole. The total change is about 

 5 cm. /sec.-, and the maximum rate of horizontal variation is at latitude 45". 

 In this region the change of g for a step of one metre northv^'ards is 

 8 X 10"^ cm. /sec. 2, or, approximately, only one thousand millionth of 

 the gravitational acceleration. This the Eotvos torsion balance, even 

 in its early forms,^ was capable of indicating definitely, being several 

 times as large as the limit to which the instrument would respond. And 

 the measurement could be made with the instrument occupying a single 

 position in a space of less than a square metre, simply by making observa- 

 tions with the apparatus as a whole in a number of different azimuths. 

 When we compare this with what is possible in the measurement of 

 gravity by ordinary pendulum methods, we see how great a step Eotvos 

 made. By timing a standard portable pendulum, with all the precautions 

 and corrections usually employed, variations of g of, perhaps, one part 

 in a million can be detected. Eotvos, in effect, multiplied by a thousand 

 the accuracy of measurement of terrestrial gravity variations. 



This remarkable sensitivity was secured by deliberately excluding 

 gravity itself from exercising any control in the instrument, which was 

 constructed so as to respond only to variations of the gravitational field. 

 The same is true with regard to Boys' apparatus, where the small forces 

 of gravitational attraction between lead and gold spheres are balanced, 

 not against any component of terrestrial gravity, but against the elastic 

 torsion in the suspending fibre of quartz. The principle in both was, 

 of course, not new ; and we ought accordingly to spare some of the 

 credit for the Rev. John Michell, who, towards the end of the eighteenth 

 century, proposed and began to construct the first apparatus of the torsion 

 balance type. He died before being able to carry out his plans, but his 

 apparatus, with certain modifications and improvements, became the 

 instrument with which Cavendish made the first laboratory measurement 

 of the constant of gravitation. 



Eotvos adopted the same principle in his torsion balance. By the use 

 of suitable suspending wires he obtained the necessary sensitivity, and 

 he secured protection against the spurious influences of air convection by 

 proper design of the enclosure. He also extended the functions of the 

 apparatus by arranging in an appropriate manner the distribution of the 

 masses in the suspended beam. It would take too long to describe the 

 instrument, and at the same time do justice to those used in other branches 

 of geophysical sui-veying. It must suffice here to indicate that the 

 Eotvos torsion balance provides means of measuring, normally by 

 observations of the changes of torsion accompanying changes of azimuth 

 of the instrument as a whole, two properties of the local gravitational field, 

 each having magnitude and direction. The magnitude of the first, for 

 which a satisfactory name has not yet been devised — the ' horizontale 



^ There have been many newer forms designed and constructed by various 

 inventors since this date. For example, photographic and automatic recording 

 has been successfully introduced, and there have been improvements in the 

 details of construction. These modifications cannot be discussed here. It is 

 worthy of note that, among modern instruments, those produced by Suss of 

 Budapest, the firm which made the earliest models, are still in the first Tank. 



