132 Fourth Supplement to Danas Mineralogy. 



limiting the essential and non-essential "water. But until this is clone, or some other 

 method for the purpose be ascertained and carried out, it is better to acknowledge 

 ignorance than to rely on mere suppositions. The formulas above deduced contain 



the improbable silicate 3:1 Si^ (a silicate least of all to be expected in a compound 

 containing so very little silica), and this gives them an additional hypothetical char- 

 acter. The other member, S:lfl^ (Gibbsite), is probable enough; and in fact allo- 

 phane is often mixed with Gibbsite at the Kichmond (Mass.) locality (see Silliman, 



Am. J.Sci., [2j,yii,411).^ 



As suppositions are easily made, we suggest two or three, to show farther what 

 hypothesis may do. The ratios deduced above, for a temperature of 100° C., are 

 quite nearly- 



(1)3:4:4 (2) 3 : 6 : 5 (3) 3 : 6 : 6 (4) 3 : 6 : 6 



M 



From these ratios we may have the following formulas : 



(1) SlSifi«+plfi' (2) 3tlSifi3_^|Slfi[3 ^3) (4) 5lSifi^+Slfi« 



in each of which, the first member 'is JIalloysite (a product, like allophane, of the 

 decomposition of feldspar), and the second Gibbsite. Adding nH after each of the 



above, any excess of water that the analyses may be regarded as affording might be 



indicated. 



Again, we may suppose that lOC^ C. may drive off part of the essential water, 

 as in Laumoatite (see above). Take the ratios, with a slight alteration as follows : 



(1) 3:4:5 (2) S : 5 : 6 (3) (4) 3:6:7 



"We then may have the formulas 8Sifl+4Slfi«, 3gifi+5*lfl:3, 3SiS+6Slfl^ 



each a compound of hydrated silica (opal) and gibbsite, both known resul^^ of feld- 

 spar decomposition. 



Or, supposing that about half the water driven off at 100° C, were previously in 

 chemical combination, we can make out the ratios — 



(1)3:4:7 (2) 3 : 5 : 8 (3) (4) 3 : 6 : 9 



in which, the oxygen ratio is 1 : 1 for Si-f-^tl and fi; and the general formula might 



then be (3tl, gi)fi3, or »iSlfi3-|,nSi£[^ the last ratio giving 2Xlfi3+si;a3, q^\i 

 lemin obtained 35*74 fi (as Mr. Northcote mentions) after drying at 100^ C., and his 

 analysis gives the ratio 6 : 9 : 16 (=2Si, 3Sl, ISS), in which there is the same oxy- 

 gen ratio, 1 : 1, for 3tl-|-Si and fi,— whence the formula, 33clfi^-(-2§iS^ or 3 of 

 Gibbsite and 2 of a supposable gelatinous silica. 



Or,includbg the whole of the water obtained, all the ratios but the first give for 



the oxygen of Si-j-Al to that of S, the ratio 3 : 4, whence the general formula (Si, 



Sl)S3+fi. Si and 3tl having the proportions 3 : 6 in 2, and 1 : 2 in 3 and 4. 



In this way many other suppositions might be made. But until we have a better 

 basis of facts, they avail little to the science, 100^ 0. is an arbitrary point in the 

 scale of temperature. That it is the point limiting stability in all hydrous com- 

 pounds is well known to be not the fact ; that it is so in any one case, is a question 

 for special research. Science will no doubt ultimately include all compounds, or- 

 ganic and inorganic, under a common system of formulas ; but the mineral allophane 

 will not take its true place in the series until it is better understood. — j, p. p.] 



. 



