ATTAINMENT OF VERY LOW TEMPERATURES. 23 



The hydrogen enters the apparatus from the compressor at a dis- 

 tributing cross piece below the gauge, as in the last model, and passes 

 first to the bottom of the regenerator coil Z. This consists of two 

 pipes of 3.2 mm. external and 2 mm. internal diameter, wound coaxially 

 to form a coil 300 mm. long and 100 mm. in diameter. The coil is 

 covered with a layer of flannel and fits tightly into a brass case com- 

 municating above with the top of the annular space F and at the 

 bottom with the escape pipe G. The upper end of the coil communi- 

 cates with the coil in the liquid air chamber B. 



The structure of the remaining parts of the apparatus is identical 

 with that of the last model and needs no further description ; though 

 there is a slight difference in the arrangement of the pipes / and G 

 and of the cocks t, V (and zv), their uses are identical and a description 

 of the method of manipulation of either apparatus will serve for both. 



Olszewski* has described a liquefier somewhat similar to mine, 

 though at the time at which his paper was published the apparatus had 

 not been tested. He states that it is possible to liquefy hydrogen by 

 cooling it to the temperature of liquid air boiling under normal pressure 

 before allowing it to enter the regenerator coil {D, figs. 4, 6). It 

 appears to me, however, that as the regenerator coil is of the same 

 dimension as in my machine, Olszewski's apparatus will be found to be 

 considerably less efficient. 



Before describing my latest researches on the liquefaction of hydro- 

 gen it will be convenient to consider the method of compressing the 

 hydrogen, and of manipulating the liquefier. 



X. Compression of the Hydrogen. 



The Whitehead torpedo compressor, as employed in connection 

 with the Hampson air liquefier, cannot be used for the compression of 

 hydrogen without modification. In the usual form of this machine the 

 packings of the pistons in both cylinders consist of cup-fibers carried 

 on the ends of the pistons and moving with them. In fig. 7, A, the low 

 pressure cylinder is shown in section. Air is drawn into the cylinder 

 through the valve a and expelled through the valve 6; c is the piston, 

 which carries the cup fiber d, and is actuated by means of the connect- 

 ing rod e. The piston and cross head f are one solid piece of metal. 



It will easily be seen that when gas is being compressed the contact 

 between the edge of the cup-fiber and the walls of the cylinder is perfect, 

 but that when gas is being drawn into the cylinder, the pressure in it 

 falls below that of the atmosphere, and there is a tendency for air to 

 enter it between the metal and the fiber. The extent to which this 

 would take place would be considerably increased if the cross head 

 became slightly worn and acquired a little side play. 



* Acad, des Sci. de Cracowie, December, 1902. 



