native Carbonate of Alumina and Lime. 465 



We are disposed, therefore, to regard this mineral as colly rite 

 mixed with a varying amount of a hydrated double carbonate of 

 alumina and lime. If it should bear a distinct name, it may be 

 termed Hovite, from the place where it was first recognized. 



On looking over published analyses of silicates, it did not 

 appear that a carbonate had often been found entering into the 

 composition of such minerals, yet there is a perfectly analogous 

 instance on a smaller scale. Allophane is another hydrated 

 silicate of alumina, and it occurs in a chalk-pit at Charlton, 

 also in the upper chalk, and under circumstances almost iden- 

 tical with those under which we found the collyrite at Hove. 

 In Mr. Dick's analysis of this*, and in the analyses of four differ- 

 ent specimens by Mr. Northcotef, there was always found more 

 carbonic acid than was required to saturate the lime. In none 

 of these instances, however, did it exceed 1*31 per cent., and 

 Mr. Northcote propounded no other view than that it existed as 

 a bicarbonate. 



On inquiring about the quarry at Charlton from which this 

 mineral was obtained, we were informed by Mr. Church that 

 the allophane was accompanied by a substance resembling our 

 collyrite. Some of this was obtained, and the two following 

 analyses were made of portions having rather different physical 

 characters. 



The first was more compact and more vitreous in its fracture 

 than the specimens from Hove, and not so perfectly white. It 

 more closely resembled the specimens of collyrite in the British 

 Museum. On analysis it was found to be a silicate of alumina 

 soluble in acids, with a little carbonate of lime. A portion 

 which had been pounded and allowed to stand two days over 

 sulphuric acid, lost 7*12 per cent, of water when heated to 100°C. 



The following was the analysis made : — 



Silicic acid 16*58 



Alumina 44*51 



Water expelled above 100° C. 



and some Carbonic acid . . 34*17 



Lime and loss 4*74 



100*00 



This shows a somewhat larger proportion of silicic acid in 

 comparison with alumina than any of the preceding analyses, 

 and indicates that the mineral was intermediate between allo- 

 phane and collyrite, in chemical composition as in physical pro- 



* See Professor Morris's paper, Quart. Journ. Geol. Soc. vol. xiii. p. 13. 

 t Phil. Mag. May 1857- 



Phil Mag, S. 4. Vol. 23. No. 156. June 1862. 2 I 



