Jaituabt 22, 1909] 



SCIENCE 



137 



ness.^' This is nearly twice as large as a 

 similar figure expressing the fire risk for 

 the United States, as based on the accumu- 

 lated experience of underwriters. Just 

 as in the case of fire insurance, the premium 

 on earthquake insurance would be ad- 

 justed to the local conditions; it would be 

 higher for houses on soft ground than for 

 those on bed rock, relatively high for 

 houses near known earthquake foci, and 

 verjr low for houses classed as earthquake- 

 proof. 



In making this estimate the fire damage 

 occasioned by the earthquake damage of 

 1906 was treated as part of the earthquake 

 damage. Had the direct earthquake 

 damage alone been considered, the eompu- 

 "* In assembling data for this estimate I was 

 greatly assisted by Professors C. C. Plehn and A. 

 W. Whitney, of the University of California, but 

 these gentlemen are not to be held responsible for 

 the estimate itself. The estimate involves, among 

 others, the following assumptions : ( 1 ) in that 

 part of San Francisco burned over in April, 1906, 

 the loss from destruction and injury of buildings 

 amounted to one third the entire loss; (2) the 

 ratio of sound value to assessed value of buildings 

 in San Francisco in 1905 was 1.7; (3) the similar 

 ratio for the entire state was 2.0; (4) the average 

 value of buildings per capita in California was 

 the same for the entire period 1800-1908 as for 

 the single year 1905. Some of the elements of 

 the estimate are as follows: 

 Damage to buildings in burned dis- 

 trict of San Francisco in 1906 



{= J X $350,000,000) $117,000,000 



Damage to buildings in San Fran- 

 cisco, outside burned district, 1906 7,000,000 

 Damage to bmldings outside of San 



Francisco, 1906 15,000,000 



Damage to buildings in California, 



1800-1905 20,000,000 



Total earthquake damage to build- 



ings in California, 1800-1908 $159,000,000 



Total corresponding " exposure " 

 (=sum of annual value of 

 buildings in California 1800- 



1908) $22,000,000,000 



Basal insurance factor (=: ratio of total 



loss to total exposure) 0.00723 



Eisk on policy of $1,000 $7.23 



tation would have yielded a figure materi- 

 ally smaller, though still comparable with 

 the basal fire insurance factor. But there 

 seems no good reason for excluding the 

 fire damage, for not only was the San 

 Francisco conflagration caused wholly by 

 the earthquake, but fire is a frequent sequel 

 of the wrecking of buildings by seismic 

 shocks. Nearly all our appliances for 

 heating, cooking and lighting are sources 

 of fire danger when deranged by violence 

 to the containing buildings, and if the 

 agent of violence affects a large area, as in 

 the ease of earthquakes, the appliances for 

 extinguishing fires are apt to be disabled 

 at the same time. 



It is possible that the estimate of the 

 building risk is exaggerated by reason of 

 its having been made just after the great 

 disaster of 1906. It certainly would have- 

 been materially smaller if made by the 

 same method just before that disaster. 

 But this qualifying circumstance is largely 

 if not wholly offset by the fact that various 

 shocks of the same physical rank as that 

 of 1906 wrought comparatively little 

 havoc because at the time of their occur- 

 rence the areas shaken were sparsely 

 populated— at least by house-building 

 races. The Inyo earthquake of 1872, 

 having its origin in Owens Valley, de- 

 molished the village of Lone Pine with a 

 completeness not paralleled in 1906, and the 

 falling walls crushed to death a tenth part 

 of the village population. The shocks of 

 1812, affecting a tract on which Los 

 Angeles, Santa Barbara and other large 

 towns are now built, were limited in their 

 destructive effect to the Spanish Missions 

 because those were then the only houses; 

 but the mission buildings fared badly, and 

 it is related that thirty or forty mission 

 Indians lost their lives. The earthquake 

 hazard indicated by these occurrences was 

 certainly not less than that emphasized by 

 the recent disaster in a populous district,. 



