379 
W. to cause both to turn, unless held back in some way. The key of an 
alarm clock (21) was fastened to a train of gears (20), and these in 
turn to the stem of roller (18), so that the rate of motion was held back 
to a speed governed by the running of the clock. In this case the speed 
was 30” an hour. The rollers were set in such a way as to cause the 
paper to pass about 4” under the end of the orifice (16) in pipe (15). 
The strip of paper was marked at the beginning and end of the test, so 
that the exact time when any point on the paper was under the orifice 
could be determined. Thus a continuous record was kept of the amount 
of tar in the gas. 
It might be well to state here that all of the tar is not taken out of 
the gas by this means. The tar particles in the cooled gas are very fine 
and light, and many of them are cushioned off from the paper, and never 
touch it to stick. Consequently the tar deposited on the paper does not 
represent all of the tar content of the coal. On the other hand, the slightest 
trace of tar in the gas will quickly black the paper, and the deposit is 
probably very nearly proportional to the entire tar content of the gas. 
6. To prevent re-distillation of tar once deposited in exit pipe. The 
remaining problem was that of making sure that no tar could be deposited 
in pipe (15) at a low temperature and be later re-distilled. to show up 
on the paper at too high a temperature. To accomplish this a large 
number of duplicate pipes were made and carefully fitted into the threaded 
bushing (14). By changing these pipes (15) every five or ten minutes, 
the effect of such a tendency was quite completely eliminated. 
The coals tested were chosen to represent the different American 
grades. They were first ground and screened over a mesh of 20 wires 
per inch and through one of 10 wires per inch. About 315 grams of coal 
will fill the cartridge, and it was put in in layers 34” thick between iron 
disks. The cartridge was then placed in the middle of the furnace, the 
thermo couple put in place, and the asbestos packed in against both ends, 
cutting off the radiation here, and Causing the ends and middle, all to 
keep at the same temperature. A current of 6 amperes was then passed 
through the resistance wire of the furnace. While the latter was coming 
up to the temperature where the light oils start to come off, the thermo 
couples were connected up, the flame placed under the gas pre-heating 
coil, the gas from the storage tank turned on, the paper rolls connected 
up to the clock and the clock started. The pipes (15) were cleaned 
and prepared for immediate use, and the data sheet prepared. As soon as 
