214 The N.Z. Journal of Science and Technology. [May 
(1.) To determine the first under the conditions of the analytical process 
a sample of clay No. 7 was heated with a 10-per-cent, solution of hydro¬ 
chloric acid in successive quantities of 100 c.c. for an hour each on the 
steam bath. The weight of dissolved matter in each extraction was 
determined by evaporating a fraction of the supernatant clear solution 
on a watch-glass, and weighing the residue. In order to differentiate 
between the solvent action of the acid on the actual clay and on the 
rhyolitic material respectively a sample of the pumice rhyolite from the 
clay beds at New Lynn was examined. It was ground and elutriated, 
and that portion was selected which came over at the same rate of 
water-elevation as does the “ clay fraction ” in mechanical analysis. It 
was extracted with 10 per cent, acid exactly as the clay above. The 
result is shown in Table III. 
Table III. 
Extraction. 
Percentage of Substance dissolved. 
Pumice. 
Clay No. 7. 
First 
4-15 
3-85 
Second 
1-30 
1-20 
Third 
1-00 
1-65 
Fourth 
0-35 
1-90 
Fifth 
• • 
1-55 
Sixth 
1-55 
It is seen that the clay is continuously soluble in the acid employed, 
while the pumice loses all its soluble material in four extractions. It 
must therefore be the actual clay which is dissolving in the acid. The 
amount of solubility (2 per cent.) is not enough to prevent the use of the 
process in combined analyses, provided that the acid is used only for the 
minimum time necessary to dissolve the carbonates, &c. This does not 
exceed thirty minutes. 
(2.) Some of the pumice sand which had been extracted with 10-per-cent. 
HC1 in investigation (1) was heated to redness in a platinum crucible and 
weighed. It was then treated successively with sulphuric acid and with 
a solution of sodium carbonate exactly as for combined analysis. At the 
end of the process it was found to have lost 7-6 per cent, of its weight. 
The magnitude of the error thus introduced into the estimation of the 
actual clay in the combined analysis depends upon the proportion of the 
rhyolite and on its relative fineness : e.g ., in clay No. 7 the actual clay 
may be shown 2 per cent, too great for this reason. In most clays the 
error will not exceed 1 to 1*5 per cent., because of the relatively small 
quantity and greater coarseness of the pumice material. The conclusion 
reached is that the process gives results of sufficient accuracy for the 
requirements of the analysis. 
Fire Tests of Clays and Synthetic Mixtures. 
In order to limit the scope of the investigation two properties only 
were selected for observation : (1) The shrinkage of moulded samples on 
drying and on firing respectively, and (2) the porosity of the fixed ware. 
A series of samples of varying proportions of actual clay to sand fraction 
was prepared from each clay and investigated. The method of preparation 
