EXPERIMENTS ON KAOLINIZATION. 293 



in Fig. 1 9. As will be seen, tlie well-known contrivance for determining the 

 boiling point of tlierniometers wns made the pattern of construction. Steam 

 is generated in the interior conical compartment ahecdoi heavy tinned sheet 

 iron, 1 8 inches in diameter at the bottom and 12 inches at the top, and between 

 18 and 20 inches high. The top, dec, also conical,^ and provided with a 

 hole at c for the escape of steam, can be removed, and fits like a lid over 

 the walls of this compartment. The whole is surrounded by the cylindrical 

 mantle, /^ ?■ A-/', of the same material. The top, g ik, of this can also be 

 removed, has the form of an ordinar}' lid, and is provided with tubulures 

 for the insertion of corks, etc., at h and i. The exterior compartnient com- 

 municates with the air by the tubulures /and/'. 



In the interior of the inner compartment, and held in position by a 

 suitable tripod (not shown in the cut), is the cylindrical chamber r s u t,\\ 

 inches in diameter and 12 inches high, and provided — like a sieve — with a 

 bottom of wire gauze strengthened by radial supports of thick brass wire. 

 P q, finally, is a feed pipe for resupplying the boiler with water lost by 

 evaporation. 



The rock to be tested was broken into small fragments, from the size 

 of a hazel-nut down to that of a pin-head, but excluding dust, and placed 

 in the chamber r s u t. Previously, however, the thermo-element x y z 

 (described on the next page) had been fixed in position, supported by suita- 

 ble croSs-bars of wood covered with thick sheet rubber. In putting the 

 rock into the chamber care was taken to pack it sufficient!}^ tight to prevent 

 currents of steam from possibly passing through the mass. Steam reached 

 the interior by a process of diff'usion, thoroughly saturating the whole. Of 

 this 1 had frequent occasion to convince m}'self Water having been poured 

 into the boiler to a level I I, approximately, and heated to ebullition, the 

 steam completely enveloped the rock chamber, permeating the material in 

 its interior. Passing through the hole c, and again around the greater part 

 of the apparatus, it finally escaped at /and/ into the air. 



As a source of heat, two small kerosene stoves were found excellent. 

 By means of the four broad flames thus obtained, the heat could be regu- 



'Tbns serving a second purpose, namely, to prevent steam condensed on the top of the boiler 

 from dripping into the rock helow. 



