84 



RECORDS OF THE AT'STRATJAN MUSEUM. 



marked, on the opposite edge. In direction it is appi-oximatelv 

 parallel to the plane of two opposite edges, nameh' the iilanc 

 bisecting the angles bq and pd (tig. 4). There is no sign of a second 

 cleavage perpendicular to this plane. Three systems of striations 

 are j^resent, but they will hv more fully described below. 



As regards their composition, the pseudomoi'phs consist of 

 precious opal of inferior quality and prevailing green or bluish 

 tints, in other cases of common milky opal, or of clear glassy 

 livalite, with patches of the rarer black opal. 



Measurements were made as a rule by each of us independently, 

 but the agreement being close, only the mean values are given in 

 the subjoined table. Provisional letters are assigned to the four 



pyramidal faces, the crystal 

 being oriented by means of the 

 edge on which the cleavage 

 appears (fig. 4). The mean 

 normal angles obtained plainly 

 indicate that we are dealing with 

 a monoclinic mineral having a 

 plane of symmetry bisecting the 

 angles hd and pq. 



h'i^. 4. 

 So.lieraatic projection on a plane at 

 right angles to the axis of elonga- 

 tion, with cleavage traces on edge h. 



Conchisioiis. — The problem 

 now is to find a mineral, mono- 

 clinic in symmetry, having a 

 prominent cleavage perpendicular to the plane of symmetry, 

 with angles approaching the values found, and the mode of 

 occurrence of which is compatible with the geological conditions 

 of the AVhite Cliflf's Upper Cretaceous beds. Obviously the facts 

 of form ali-eady l)rought out dispose of the claims of gypsum, 

 anhvdrite, celestite and sulphur, while the angles do not even 

 approximate to those of gay-lussite. After passing in review 

 all the likely minerals that suggest themselves, we huxe come to 

 the conclusion that the species most nearly fulfilling the requireri 

 conditions is glauberite, sulphate of soda and lime, which is mono- 

 clinic in crystallisation, and has a perfect basal cleavage. In 

 accordance with this theory, we have incorporated in the table 

 the theoretical angles of glauberite, which we regard as cor- 

 responding to the measured angles of the pseudomorphs, 



