376 Messrs. Milne and Gray's Experiments 



the final yielding, and the subsequent shocks the gradual 

 settling down of the ground into a state of equilibrium. 



Effect of Earthquakes on Buildings. 



So far as we have been able to observe, the effect of 

 an earthquake on a building depends on the following cir- 

 cumstances : — 



Whether the different parts of the building are so propor- 

 tioned that they are able to vibrate in unison without producing 

 excessive stresses at any part. We have observed, for instance, 

 that brick chimneys passing through the roofs of wooden 

 houses, and chimney-stacks built outside a house and sup- 

 ported by tie-rods from the house, are very subject to damage 

 by earthquakes. We have observed, also, that cracks pro- 

 duced by an earthquake generally open and close during the 

 vibrations of succeeding earthquakes, thus giving evidence of 

 a want of agreement in the vibrational period of the portions 

 of the building on opposite sides of the crack. 



The absence of arches having sharp angles at the crown 

 or at the buttresses is also of considerable importance. An 

 examination of a great number of brick arches in similar 

 buildings, some of which had sharp angles, as here described, 

 while others were curved at these points, showed a Yerj 

 marked difference in the effects which had been produced in 

 the two cases. The rounded arches were seldom at all 

 damaged, while the others were nearly all more or less 

 cracked. 



The nature of the foundation on which a house is built may 

 also influence the effect considerably.' Probably the safest 

 foundation is a soft one, the building standing on a broad 

 concrete base. We have been led to this conclusion from 

 observations on the relative effects of the same earthquake on 

 buildings, some of which were built on rock, others on piles 

 driven to considerable depths in soft materials, and others on 

 concrete. The buildings on rock are generally more shat- 

 tered than those in the same locality but on a softer material, 

 such as hardened mud. Again, those on piles appear to suffer 

 more than those on concrete. Whether this may be due to 

 the piles causing the shock to be more severely given to the 

 building, or due to the piles sinking and hence causing 

 stresses in a more indirect way, we are not yet able to say. 



The configuration of the ground in the vicinity of the 

 building is a very important factor. The greatest damage 

 dependent on this element is done to buildings which are 

 placed near the edge of bluffs, or near the junction of a plane 

 with a steep hill. The effect of a bluff is probably due to the 

 free surface causing a greater amplitude of motion, or possibly 

 a more one-sided motion than takes place in an open plane. 



