95 



by the " Discovery" would be accompanied by corresponding perturbations in the niagnetogranis. That a 

 slight relationship of this description exists has already been noticed by Mr. BERNACCiu, lint now that the 

 register of the " Discovery " has been extended this may be more clearly established. 



When making this enquiry, large earthquakes which for various reasons were not recorded by the 

 " Discovery" should not be overlooked. A list of these is given (p. 87). Also that the time at which 

 disturbances of magnetic needles might be expected would probably correspond with the arrival of phase 

 PS must be kept in mind. 



To strengthen the assumption that " pulsations " are actual movements of the earth's surface, it would 

 be of interest to compare the times when these were frequent with the periods when magnetic needles 

 were unsteady or showed oscillatory movements. 



The fact that the magnetic rocks on Eoss Island have a high density is one reason which would lead us 

 to expect a marked difference between the observed and calculated values for (j. 



