Chap. IV. ] 
HOLLANDITK 
89 
to cavities. In most cases the difficulty is only one of deteraiining 
whether the mineral is to be called manganite or pyrolusite ; but in a 
few cases, when the crystals are small, or when they pass into more or 
less massive crystalline material too hard for manganite, it is necessary 
to recognise the possibiUty that hoUandite may be present ; this point can 
only be definitely settled by isolating some of the mineral and taking 
its specific gravity, and then, if this be not sufficient, making a chemical 
analysi? of i^". 
i In colour hoUandite varies from a light almost silvery grey through 
_, . , , ^ gre-sdsh-black to quite black. On both crystal faces 
Physical chaiactoif . ^ ■' -i . n- i 
and fracture surfaces it has a shumig metalhc lustre. 
In the Kajlidongri specimens, the only ones that exhibit definite crystal 
faces, the hardness is found to be 6 on the crystal faces and only 4 on 
the fracture surfaces. Owing to the infrequency of crystal faces the 
hardness test must as a rule be made on fracture surfaces, and therefore 
the latter figure may be taken as the more general value for the hardness 
of the mineral. This difference of hardness on the crystal and fracture 
surfaces is probably due to the fact that whilst the crystal faces are often 
smooth the fracture surfaces are almost invariably very much striated, 
so that testing the hardness on a fracture surface consists not in testing 
the real hardness of the mineral, but in tearing away the fibres of the 
mineral one from another ; consequently the hardness found by testing 
the crystal faces must be taken as the true one. As this value is 6, it will 
be seen that this brings the mineral into line, as far as this property 
goes, with the hard varieties of the amorphous manganates grouped 
as psilomelane. The v<a,y in which hoUandite splits into prismatic chips 
very much striated on every fracture in a direction parallel to the ver- 
tical crystallographic axis of the mineral is one of its most charac- 
teristic properties, although, of course, when the mineral does not occur 
in definite crystals it is not evident that the direction of sphtting is 
this axis. The streak of the mineral is black. 
The crystals that occur in the quartz-veins at Kajlidongri are some- 
^ ^ ^ ^. times of large size. Since my visit Mr. Winch has 
chwacters^^'^*^'^*'' kindly collected considerable quantities of this mine- 
ral as it became exposed in the course of quarrying 
the ore. A common size for the crystals is a diameter of ^ to | inch, 
but some of the specimens obtained by Mr. Winch are as large 
as 1 inch across. In their simplest form the crystals show what 
looks like a tetragonal prism surmounted by a flat tetragonal 
