232 
W. F. Cole. 
Method of Separation . — In the prepa.ration of the material for X-ray 
analysis a nmnber of grades of particle size were separated from each sample, 
but only two, a coarse fraction of particle size 61 p and less and a fine fraction 
of partick^ size 2 p and less, were examined. 
The coarse fractions were obtained by sieving the three samples throngli a 
250-mesh Tyler standard sieve of screen opening 0*061 mm. 
The fine fractions were obtained by elutriation and settling. In the 
elutriation an Andrews Kinetic Elutriator was used to disperse the particles. 
Material of |)article size 61 p and less, obtained by sieving, was fed into the 
elutriator together with the washings off that material which did not pass 
through the 250-mesli sieve. Tlie run off from the elutriator Avhich carried 
partic^les of size 10 p and less was collected in cylinders and tlie particles 
allow('d to settle. Depending upon the mean temperature of the laboratory, the 
settling time was adjusted so that particles of ('ffective diameter 2 p and* less, 
would fall a distance of 28 cms. At the end of the apjwopriato time the top 
28 cm. was siplunied off and the clay particles removed from suspension bv 
flocculation with calcium chloride, filtration, washing with absolute alcohol 
and air drying. Xo attempt was made to remove the organic matter present, 
Tlic coarse fractions of the soil, subsoil and greensand were dark brown 
brown and dark grtHm in colour, whereas iit the same order the fine fractions 
were cliocolate brown, light bi'own anfl liglit green. 
The amount of clay material collected in this mannei* was small. I'his, 
however, is in keeping with the meclianical analyses of soils overlying tlie Gin- 
gin gre(msand imblished b\' Hosking and Greaves (7). They note that tlie 
Wakea sand, which covers both the upper and lower ferruginous sandstones and 
grits from A\ hich it is foi-nu'd in situ, rarely contains more than 10 p(u‘ cent. clay. 
h:XPERIMFNTAL DETAILS. 
A Hilger-Mueller impro\'(Hl X-ray goniometer spectrograph was used with 
certain modifications pr(‘\-iouHly referred to (4). Radiation was supplied by a 
Hilger all-steel water cook'd gas X-ray tube. The tube was energised by a high 
tension unit arranged as in Fig. 3. This unit was modified from a Watson 
200 K.V. Gonstant Potential Deep Therapy set. 
Phologra]dis wen^ taken witii both Vo and Cu radiation. It was found 
that films taken with ( 'u ratliation sliowed a pronounced general scattering 
which was greatly impr«)\ ed by the use of Fe radiation. This can be explained 
by the facts that the sam|)les contain minerals of liigh iron content and that 
Cu characteristic radiation is capable of exciting Fe characteristic radiation. 
The F(^ radiation was iMuulered monochromatic by the use of a manganese 
filter which served as a window for the X-ray tube. The action of such a 
filter is to su]3])ress the radiation and transmit Ka radiation. The advan- 
tage of using filtered radiation was realized early in the work wIkui. in exam- 
ining a film of the gj-eensand (61 p and less) taken with Fo radiation, it was 
found that many lines in the central region could b(^ possibly contributed to by 
the Kj8 compoiumt. 'fhe manganese was elcctro-de}>osited on aluminium foil 
of thickness 25 p. The nuTliod of [)re])aration used is descrilaxl by Wood (8). 
With the filh'r us('d in the cours<^ of this work no trace of the component 
appeared even after tlu' longest (exposures. 
Suitable jjliotographs with Fe Ko were obtained afh'r 16 hours exposure 
Avith 4-5 mA tube current at 45-50 K.Y. Films Avith unfiltered Cu radiation 
could be obtained AA'ith shortfn- exposures. When Cu radiation Avas used the 
manganese foil Avas still retained as a A\ indoAV for the X-ray tube. 
