Fig. 5. Monthly Distribution of Earthquakes. Pacific Coast Statistics. 

 McAclie. Total 676. 



stice (figure 5) are exceptional, and might become confusing 

 were it not for the special justification therefor. 



We now venture upon the most extensive and most 

 crucial test of all. From Mallet's "Catalogue of Earth- 

 quakes," published in the British Association Reports for 

 1852-3-4, I have collated monthly data of 5155 earthquakes, 

 which occurred from 1600 to 1842 inclusive, the catalogue 

 being based on authentic reports from every known field on 

 the earth's surface. From these exhaustive data we obtain 

 the following diagram. 



This curve, figure 6, is strongly emphasized, but does not 

 depart to such an extent from the characteristic line already 

 developed as to form a contradiction. The minor rise in the 

 summer solstice previously developed disappears, and is re- 

 placed by a decided minimum at that point. For this we had 

 a precedent in figure 4. 



In fact this latest development, figure 6, corresponds so 

 closely in essential features with those previously secured 

 that I feel warranted in making a composite of the whole 

 series. 



Figure 7 includes all of the statistics embraced in figures 

 1, 2, 3, 4, 5, 6. 



Here is the summary of our study — the conclusion which 

 must be relied upon to give the characteristic curve. If this 

 curve has any significance, it is to show that there is some 



43 



