part 1] AISTNIVERSART ADBKESS OF THE PRESIDENT. Ixiii 



general increase of stress is 100 times this figure we get the result 

 that the rate of increase of strain is equivalent to that which 

 would be produced by an increase of downward pressure, or a 

 corresponding reduction in the support of the crust, at the rate of 

 1/94,000 of that due to gravity, in each period of six hours. 



It has long been established that the strength of the crust is far 

 from being able to withstand the crushing strains which would be 

 set up by removal of support from below, and the estimates, which 

 have been independently made by different investigators, concur in 

 putting the limit of the removal of support, which would result in 

 crushing, where an area comparable with Italy is concerned, at not 

 more than about 1/400 of the force of gravity. If this fraction 

 is divided by that obtained in the jDrevious paragraph we get the 

 result that, starting from a condition of no strain, fracture would 

 come about after an interval of 235 periods of six hours, or not 

 quite 59 dsijs. The calculation, therefore, indicates that the rate 

 of growth of strain in Italy has been, on the average, such that 

 the breaking point would be reached in about two months from 

 start, with a wide variation on either side. Some other relations 

 between the frequency of earthquakes and the diurnal variation 

 of the tidal stresses might be, and have been, investigated; none 

 of them seem so appropriate as that which has been detailed, and 

 all give fairly confirmatory results, the longest period indicated, 

 as required for reaching the breaking strain, being just about 

 a year. 



It must not be supposed that value can be attached to the 

 precise figures. As is invariably the case, in all calculations 

 regarding physics of the Earth, many considerations are involved of 

 a very uncertain nature ; but the reasoning does show that the 

 increase of strain must have taken place at such a rate that the 

 breaking point was reached in a period measurable at most by 

 months. They prove conclusively that the period could not have 

 been of such length as to be measurable by j^ears or decades, for, 

 had this been the case, the disparity dealt with would have been 

 much greater than that actually found. 



The same conclusion may be reached in another way. The 

 stress-difference required to produce fracture in average hard rocks, 

 as they are met with at the surface, is round about 1,000,000 

 grammes per centimetre square, and, allowing for the greater 

 strength at depth which is indicated by the experiments of Prof. 

 Adams and the computations of Prof. Burrell, we may put the 

 breaking strength of the Earth's crust at about double of this, so 



