TEANSACTIONS OF SECTION A. 39 



in April, with another (nearly as low) in November, and a 

 lesser fall in February. 



As the number of observations on which Mallet's curve for 

 the Southern Hemisphere was based is only 223, it is obvious 

 that it might be considerably modified by the inclusion of the 

 remainder of the 775 earthquakes in the New Zealand record. 

 Of course, to form a revised curve we should need also the 

 records for Australia, South America, and South Africa. 



D [exhibited] is simply an enlargement of C. I cannot 

 perceive any law of connection between the season of the year 

 and the number of earthquakes- -at least, as far as New Zea- 

 land is concerned. 



II. DiSTEIBUTION IN PlACE. 



E [exhibited] is a map of New Zealand shaded to show 

 the regions of greater or less earthquake-frequency. Itiis 

 constructed in the following manner : Out of the 775 shocks, 

 116 were felt (or recorded) at three or more places. Each 

 earthquake is plotted on the map by joining the outermost 

 places at which the shock was felt, and the included ai'ea re- 

 ceives one wash. No attempt is made by difference m the 

 washes to mark differences in the intensity of the several 

 shocks ; but all the washes are of the same depth. The re- 

 maining earthquakes are marked by similarly shading a cir- 

 cular area of ten miles radius round the place at which each 

 shock was recorded, omitting those places at which the total 

 number of such isolated shocks since 1818 has been less than 

 six. 



F is a photograph of the same map. It will be seen at once 

 that the darkest part of the map is that part of Cook Strait''= 

 included in the triangle Wellington, Blenheim, Wanganui : 

 in other words, wherever the origin may be, the earthquakes 

 felt at these three places are commoner than any other earth- 

 quakes except, perhaps, those that are local in a still stricter 

 sense and are felt at one place only. The next shade of fre- 

 quency includes Christchurch ; the next. Nelson. Then we 

 must extend the area to include New Plymouth, Masterton, 

 Eeefton ; next, Hokitika with Greymouth and Westport ; then 

 Napier ; and so on. There is an isolated district of local earth- 

 quakes round Eotorua and Tarawera ; and a larger district 

 extending eastwards (and probably westwards, too, if we had 

 records) from Queenstown. Two large tracts of country are not 

 shaded — the extreme north and south-west of the colony. In 

 the latter this is due to lack of records, but that is certainly not 

 the case in regard to the country north of Auckland. 



* I use " Cook Strait" in this paper in the widest sense — that is, to 

 inchtdo the region from Cape Campbell to Cape Farewell and Taranaki. 



