368 Tnmm ctiom. — Geo logy . 



Tarata and Otukapuarangi ; here also were geysers, ngawhas, 

 mud-springs, steam-holes, solfataras and fumaroles, each and all 

 pouring out in larger or smaller quantities its volume of heated 

 water until the lake itself was fully deserving of its name, 

 " Rotomahana," (warm lake,) and its effluent Kaiwaka was 

 worthy of a similar distinction. Now, the body of water 

 debouching from this lake was large and continuous, and many 

 millions of gallons were daily discharged into the Tarawera 

 Lake. If now we turn to Rotomahana, and witness the effects 

 of these hot springs and geysers, we find an amount of sinter 

 deposited which is surprising, for though we have been used to 

 speak of the two terraces, there were several others in a state of 

 decadence or fragmentary condition, while lavishly around us 

 were the evidences of sinter deposit. Year after year, probably 

 for centuries, had this deposition gone on, though only a tithe 

 of the silica which rose in solution had been arrested. For- 

 tunately these waters have been analysed, the results of Mr. 

 Skey's examination showing the water from the White Terrace 

 to be charged with mineral matter to the extent of 144 grains to 

 the gallon, and from the Pink Terrace 154 grains. Accepting 

 this as equal for all the springs so constantly at work, we shall 

 have in the course of years a very large amount of rock material 

 withdrawn from the earth, most probably leaving cavernous 

 spaces, and a weakening of the earth's crust locally. It required 

 then only some local disturbance of the earth's crust to precipi- 

 tate the falhng-in of these spaces, which would have occurred 

 sooner or later without such disturbance. 



There can be no question that the first outbreak came from 

 the Tarawera Mountain, caused probably by some slow-moviug 

 earth-wave, evidences of which we have already adduced. This 

 in itself was sufficient to cause a precipitation of the weakened 

 honeycombed rocks through which the waters of the Rotomaka- 

 riri Lake would make their way into the chasm, and, coming 

 into contact with a large surface of the molten rock, would be 

 followed by a terrible convulsion, the escaping steam ripping 

 up the side of the mountain in the manner already described. 

 Water rushed down on the heated rocks only to be driven 

 back and di-^sipated into the surrounding space, together with 

 the fragmentary matter and dust resulting from the shock. 

 The water from the Rotomahana Lake would then be liriven up, 

 together with the steam and debvitic mass, to fall over kaig dis- 

 tances in the form of mud, as we now see it, until the water had 

 been repelled from the lake, and with it the solid material of its 

 bed. By this action the bed of the lake has been lowered, and 

 its sides greatly extended, while there can be but little doubt 

 that the whole of the terrace formation has been swept away. 



That the long dormant mass of molten lava underlying it 

 extended no further, is very questionable, and the evidences of 



