Vol. 50.] STRUCTURE OF CARBONIFEROUS DOLERITES AND TUFFS. 619 



or zinc sulphide was obtained. A few drops of ammonium car- 

 bonate were then added, which produced no precipitate, neither did 

 ammonium oxalate, after waiting long enough for any precipitate to 

 appear ; tbe absence of lime was therefore inferred. On adding 

 ammonium phosphate, a small precipitate of ammonium-magnesium 

 phosphate was obtained, which was filtered off, and after removing 

 phosphoric acid from the filtrate by lead, and the excess of lead by 

 sulphuretted hydrogen, and then evaporating to dryness and heating 

 to expel ammonium salts, the residue gave a decided reaction for 

 potassium when tested by platinum chloride, and a very decided 

 reaction for sodium by the flame-test. Another 5 milligrammes 

 of the mineral were rubbed to a powder with 10 milligrammes of 

 fluorspar free from silica, and gently warmed in a very small platinum 

 crucible. A drop of water in a loop of platinum wire was supported 

 within the crucible a little above the surface of the fluid mixture, 

 and an unmistakable precipitate of silica was formed in the drop. 

 The whole analysis was carried out in platinum, and on account of 

 the minute quantity submitted to examination, every care was taken 

 to use pure reagents and those in the minimum proportion -necessary. 

 The oxides found, besides silica, were iron oxide in comparatively 

 large quantity, a fair amount of alumina, a small quantity of magnesia, 

 and small quantities of soda and potash. The estimates of relative 

 proportion can of course only be regarded as rough guesses." 



These results differentiate it from iddingsite, which, according to 

 Lawson, is non-aluminous, and contains lime, whilst the Potluck 

 mineral contains potash and alumina, but no lime. The qualitative 

 analysis points rather to a biotite ; the optical properties also agree 

 with those of an almost uniaxial mica. Only its brittleness and 

 want of elasticity make it differ physically from biotite, though 

 Eosenbusch l says that the elasticity of cleavage-flakes becomes less, 

 even to brittleness, in phlogopites and biotites, and is often quickly 

 lost when biotite begins to alter to chloritic aggregates. The optical 

 axial angle is too small, and the pleochroism is too great, for bastite 

 or for antigorite. 



Sprodglimmer and chlorite are the only minerals which have a 

 like development of cleavage in kind and degree. 2 Of the former 

 group xanthophyllite and clintonite, which have negative double 

 refraction, have only moderate pleochroism and are hardly attacked 

 by acids. The double refraction of chlorite is too feeble. 



The Peak Forest pseudomorph, though differing in appearance 

 from the Potluck one, has the same optical properties. The only 

 difference is in the colour and small degree of cleavage. It un- 

 doubtedly replaces olivine. The colour is green and yellow in the 

 same crystal, while that from Potluck is generally red or green or 

 yellow throughout. In two outcrops we have the two kinds of 

 pseudomorphs, though not together in the same thin section. I am 

 inclined to think that the Peak Forest pseudomorph is a transition- 

 stage towards the formation of the Potluck pseudomorph. "We 



1 Mikr. Phvsiogr. vol. i. (1885) p. 476. 



2 Ibid. 



