EARTH TREMORS. 185 



In the following bibliography, the first date is that of the year iu 

 which, so far as known, the instrument was originally constructed. 



1. 1832. L. Hengeller : ' Phil. Mag.,' vol. xlvi., 1873, pp. 412-416. 



2. 18.51. A. Gerard : ' Edinburgh, New Phil. Journ.,' vol. Iv., 1853, 



pp. 14-16 ; Kennedy, pp. 94-95. 



3. 1862. Perrot : 'Paris, Acad. Sci. Compt. Eend.,' vol. liv., 18G2, 



pp. 728-729, 851-852. 



4. 1869. Rev. M. H. Close : Barrett and Brown's 'Practical Physics,' 



1892, p. 241 ; Kennedy, p. 96. 



5. 1869. F. ZoUner : ' Phil. Mag.,' vol. xliii., 1872, pp. 491-496. 



6. 1871. C. Delaunay : 'Paris, Acad. Sci. Compt. Rend.,' vol. xcvii., 



1883, p. 230. 



7. 1879. Lord Kelvin : 1880-81, G. H. and H. Darwin : ' Brit. Assoc. 



Rep.,' 1881, pp. 93-112. 



8. 1887-88. E. von Rebeur-Paschwitz : 'Nova Acta der ksl. Leop. 



Carol. Deutschen Akademie der Naturforscher,' Bd. Ix., 1892, 

 pp. 1-216 ; 'Brit. Assoc. Rep.,' 1893, pp. 303-309. 



9. 1892. J. Milne: 'Brit. Assoc. Rep.,' 1892, pp. 107-109; 'Fed. 



Inst. Mining Eng. Trans.,' 1893, pp. 6-7; 'Seismol. Journ.,' 

 vol. i. 1893, pp. 88-90. 

 10. 189.3. H. Darwin : 'Brit. Assoc. Rep.,' 1893, pp. 291-299 ; 1894, 

 pp. 145-146, 158-160 ; 'Nature,' vol. 1., 1894, pp. 246-249 ; 

 'Seismol. Journ.,' vol. iii., 1894, pp. 61-63. 



In every case, I believe, except those numbered 8 and 10, the principle 

 of the instrument was discovered independently. The horizontal pendu- 

 lum has also been designed as a time-recorder for small disturbances by 

 Professor J. Milne (Japan, 'Seismol. Soc. Trans.,' vol. iii., 1881, pp. 61- 

 62; 'Nature,' vol. xlii., 1890, p. 347); Professor T. C. Mendenhall 

 (' Amer. Journ. Sci.,' vol. xxxv., 1888, p. 105) ; and Professor G. Grablo- 

 vitz ('Boll, della Soc. Seismol. Ital.,' vol. i., 1895, pp. 12-17). 



All the different forms of horizontal and bifilar pendulums agree in 

 one respect : the vertical distance between their points of support is very 

 great compared with the horizontal distance between them. In principle 

 they merely differ in the method of suspension ; and, according to this 

 method, they may be grouped in the following three classes : — 



1. The pendulum in which the rod or mirror is suspended by two 

 wires. These may be again subdivided : (a) The pendulums of Close and 

 H. Darwin, and practically also of Delaunay, and Lord Kelvin and the 

 Darwins, in which the centre of gravity of the rod or mirror lies between 

 the two points of attachment of the suspending wires, {b) The pendulums 

 of Hengeller, Perrot, and Zollner, in which it lies outside them. 



2. The pendulums of Gerard and Milne, on which the rod is supported 

 l)y one wire and on one steel point. 



3. The pendulum of von Rebeur-Paschwitz, which is supported on two 

 .steel points.' 



' In tliis clas-s should be included Professcr Ewing's horizontal pendulum seis- 

 mogri-nph, which, though designed for a different purpose, also records slow tilts of the 

 ground (Enri/cl. Brit. vol. x.^i. p. 628). 



