1909.] . OUTLOOK OF SEISMIC GEOLOGY. 267 



is the simplest to work with, as also the most probable, and it is that which 

 has been adopted in the following pages. 



As we shall see, the fundamental difficulty which stood in the 

 way of the acceptance of the Schollen idea at the time Oldham was 

 writing, has since been removed by the " distant " studies of earth- 

 quakes (see below, p. 285), and the theory of a thrust-plane, which 

 he chose to adopt, has remained without any support in later work. 



Additional and important contributions toward the fault-block 

 theory of earthquakes have crowded about the beginning of the 

 twentieth century. In the year 1900, Yamasaki, in describing the 

 great earthquake of northern Honshu, which occurred in 1896, gave 

 as its cause the movement on two visible displacements which 

 opened on opposite sides of the mountain mass.^ 



Two long lines of fracture were discovered by me to be the cause of the 

 Riku-U. earthquake. . . . They lie on the two sides of the mountain 

 axis of the Central chain, and so this earthquake offers an example of the 

 longitudinal quakes (Liingsbeben) which but seldom occur. 



Thoroddsen, in a report which reached the scientific world first 

 through a German abstract of the year 1901,^ was able to show that 

 during each of the five heavy shocks of the South Icelandic earth- 

 quakes of 1896, a separate block of country had been shaken. These 

 several areas were all included in a low plain walled in by a ram- 

 part of mountains, and with a single exception they were contiguous 

 areas which did not overlap. 



Each of the heavy shocks was limited to a circumscribed area which was 

 made evident by a mass of collapsed houses, and from this the earthquake 

 waves were propagated outward in all directions. 



The ground beneath the low plain is probably separated into individual 

 parts and the continued movement on these cross lines [across the main 

 fissures on which the volcanoes of the island are ranged. — W. H. H.], as 

 well as the faults between the individual parts, appear to be the causes of the 

 many earthquakes of this district. If one studies the statistical tables of 

 the ruined houses from each shock [given in Icelandic report. — W. H. H.] 

 it is seen that the individual areas are somewhat sharply delimited ; while 

 upon them nearly everything was destroyed, the damage outside was rela- 

 tively small. 



* N. Yamasaki, Pet. Mitt., Vol. 46, 1900, pp. 249-255, map. 

 ° Pet. Mitt., Vol. 47, 1901, pp. 53-56. The full report had appeared in the 

 Icelandic language two years earlier. 



