No. 6.] DAWSON — E\RTHRUAKES. 343 



Science, by Professor Rockwood, of Princeton, he defines the 

 area in question as that of " an irregular trapezium whose angles 

 are marked by Pembroke, Ont., Three Rivers, P.Q., Hartford, 

 Conn., and Auburn, N. Y., and which is some 200 miles on its 

 northern and southern sides, about 300 miles on the eastern side, 

 and 175 on the western." So far as can be learned from the 

 reports, the shock seems to have been most severely felt on the 

 north side of the valley of the St. Lawrence and about Lake 

 Champlain, or may be said to have had its centre in the Adiron- 

 dack and Green Mountain region. 



In the notice of Canadian earthquakes in 1860, I mentioned 

 that it had been observed that the greatest and most fre- 

 quent shocks have occurred a little after the middle and toward 

 the close of each century. We are now approaching the latter 

 period, so that possibly the last shock may be the beginning of 

 a series of similar phenomena. Since, however, there is no 

 known reason for this periodicity, it may be a merely accidental 

 coincidence, or may depend on some cycle of about half a 

 eentury. 



If we add to the table of earthquakes in Eastern America, 

 given in Vol. V. of the Naturalist, the more recent earthquakes 

 observed in Canada, the proportion for the several months will 

 stand as follows : — 



January, 9 earthquakes; February, 4 ; March, 5; April, 5; 

 May, 7; June, 3; July, 4; August, 6; September, 4; October, 

 8; November, 15; December, 8. Total, 78. 



Thus of seventy-eight recorded Canadian and New England 

 earthquakes, fifteen, or nearly one-fifth, occcured in November; 

 forty, or more than half of the total number, in the third of the 

 year, extending from October to January inclusive. The pub- 

 lished catalogues show that similar ratios have been observed 

 elsewhere, at least in the Northern hemisphere. 



In some earthquakes a low state of the barometer has been 

 observed, as if a diminution of atmospheric pressure was con- 

 nected with the movements of the crust producing seismic vib- 

 rations. This we can readily understand if a low state of the 

 barometer should prevail over an area of the crust tending to 

 rise, simultaneously with a high pressure over a sinking area. In 

 this case a state of previous tension might terminate in a rent of 

 the crust causing vibration. In the present case no very decided 

 indication of such a cause appears, at least in so far as this part 



