112 
Rex T. PiiiDER. 
1922, }>. 82). The tliet that the boulder beds are eonlined to a eoinpura- 
tivdy small arey stuoiis indicative of either a lliiviatile orig-in or as an 
aceumiilalion of beach l)onl{lers in a small bay. From the degree of 
rounding of tin* saml grains in the associated sandstones (to be described 
})resently) the possibility of Iluviatile origin seems remote. It appears 
most probable tiiat these i)oulders were derived by direct marine abrasion 
of the neai'by coast which was made uj) of these Fre-Cambrian rocks since 
the rock types noted amongst the boulders can Ijc matched with the rocks 
in the Hidge Hill area. 
(ii) Tin' sandstones form the bulk of the ext>osure of this series. 
They are I'eddish in colour, have no bedding, ai’e unfossiliferous and no i 
certain means exist of accni’ately determining their geological age. They 
contain occasu)nal water-worn cjuartz pebbles which arc well-rounded and ' 
in >omc instances highly j>(.»lished. The sand grains are almost entirely ] 
(juai-tz and two types can be distinguished: — (a) grains with a rough j 
irrcuular surl'ace whicli nevertheless shows signs of considerable a))rasion ; 
and {!)) smootli-hiirfaced rounde<l gi'ains with dull to polishe<l surface ; 
textures which a])])eai’ to he the I’esult of a [)olish .su])erimposed on earlier 
froslmg^ t '('(‘sc'entic j)ereussion marks are generally well developed on i 
the larger grains, iV s])ecimen (22798) of this ferruginous sandstone from : 
three feel htdow the Iow-lev(*i laterife cajjping on the Hidge Hill Shelf j 
was disintegrated by boiling in IK'l and meehanical-, heavy mineral-, and ' 
shape- a7ia!yse>> of the insolul)le residue (79 per cent, of the sample) were j 
made, ^fhe results of these analyses are sol down in Table III and in j 
(Johuiin I) of the histograms of tigiires five and six. The results of a j 
chemical analysis of this specimen are recorded in Table I and the heavy 
minoi'al analyses are shown in Table V. ( 
The main features disclosed by these analyses are:^ — 
(a) The mechanical analysis (by lunid sieving with Tyler screens) indi- 
cates that the insoluble material is fairly well graded, 48 [)er cent, lying lx*- | 
tween Yj and Vi nun. <liameter and 99.5 per cent, lying between Vi and j 
Vs mm. diameter. No significance should he attached to the relatively high 
(11 per cent.) proportion whicli passed 259 mosh (i.e. less than OlKil mm. 
<!iamclcr) as microscopic (‘xamination shows that it consists largely of Iiroken 
tuberose i'raguumts of while material which, because of its irregular and 
iii’amliing forms, appears to he an autliigeuic mineral uni'elate<l to the ^ 
original detrilal sand grains. This material is isotropic with refractive index 
varying ))etween 1.52 and 1.54 and the refi'activc indices do not vary after 
ignition. It is insoluble in IICI and stains readily with malacliile green. It 
therefore ap]>ears to he either monlmorilkmite or a dehydraterl alumina-silica 
gel with ALO.. : SiO^— 1 : 2.8 (vSpiidial, 1922, p. 288) and in view of its tube- 
rose nature more pi-obalily the latte?'. *■ The ALO.. : Si(L ratio of approximately 
I : 3 is conlirmed by tlu' chemical analysis of the rock (Table 1, column 1) 
which shows that the rock contains alumina and combined silica in the mole- 
cular ratios 31 : 90. This material forms practically all of the minus 250 mesh 
fraction (11.2 per cent.) and ai)proximately half the 115-250 mesh fi'action 
(3.9 per cent.) but very little was present in the coai'ser grades and if this 
*Mr. A, J. Gaskin has recently (1947) made a thirmal examination of the clayey frac- 
tion of two soil samples from over the ferruginous sandstones and low'-level lateritVs and 
finds from the thermal data that they contain liimmite and kaolinite (much of which 
\- srmi-amorphous) with a possibility that gibbsite is also present. 
