SECT. 5] MICROSEISMS 713 



The writer successfully tracked a storm in 1951 by using this technique 

 (1954). The correlation coefficients between the components were found by an 

 analogue computer, described by Tucker (1952), the two records being con- 

 verted into a black and white form and placed one above the other along the 

 surface of a perspex semi-cylinder. Two strips of light scan both records at 

 speed, one strip to each record, the distance between the strips being adjustable. 

 The whole arrangement is viewed by three photo-electric cells so that variations 

 in height on the records are converted into an electrical output. This is squared 

 and integrated. The final result gives the correlation and this varies as the 

 distance between the strips is varied. The maximum value is taken and the 

 strip separation then gives the phase difference between the two records. This 

 computer has since been improved by Iyer by incorporating band-pass filters 

 in it so that the correlating can be confined to waves contained within a narrow 

 period range such as 5-6 sec, 6-7 sec, etc. This proved to be extremely useful 

 when there w^as more than one source of microseisms, as the effect of the two 

 sources could then be distinguished if the two sets of waves generated had 

 different period ranges. 



Iyer altered the original assumption that Rayleigh waves and Love waves 

 come from the same source to that of assuming the Love waves to come equally 

 from all directions. With this assumption, equation (1) becomes 



tan e = (l/r^,2_ i)y./(i/^^^2_ 1)%. (3) 



Iyer tracked several storms by this method, allowing for the effect of refraction 

 by using the diagram shown in Fig. 7. He found that the estimated bearings of 

 the source consistently lagged behind the known bearing of the storm centre so 

 that, in the case of a north-moving storm, the generating area would be to the 

 south of the centre. This would be expected if the wave interference theory is 

 correct. Iyer was also able to calculate R{t)IL{t) for the various examples. His 

 results are shown diagrammatically in Fig. 8. It can be seen that the ratio 

 decreases as the distance of the source from the seismograph station is in- 

 creased. This implies that energy is progressively converted from the Rayleigh- 

 wave type to the Love-wave type as the microseisms travel on from the source. 

 He also found the mean value of the Rayleigh constant to be 0.73, independent of 

 the period and very near the theoretical value of 0.68 for simple Rayleigh waves. 

 Iyer's conclusions that Love waves emanate from all directions do not agree 

 with those of Gutenberg and Benioff (1956) who, by using strain seismographs, 

 could distinguish directly between the two kinds. Gutenberg concluded that 

 Rayleigh and Love waves come from the same direction. In this case the 

 microseisms, however, had almost certainly been caused by the reflection of 

 swell off the Pacific Coast of America so that the microseisms had not travelled 

 over a long ocean path. 



6. Instruments 



Analysis of microseisms is a lengthy procedure even with an analogue com- 

 puter and it is advantageous to be able to analyse the waves as they are 



