887 
increased by substituting a compound engine; but the fact must not be 
overlooked that there is even a greater chance of increasing the efficiency 
by improving the gas engine, which is sti]l new and unperfected. 
Some of the lignites of the Philippines contain sulphur in sufficient 
quantity to injure a boiler in the course of time. The United States 
Geological Survey has shown that similar coals from the United States can 
be used satisfactorily when they are converted into producer gas, and it is 
thought that all native coals can be utilized in that way. The results 
given by the producer gas plant on Kentucky coals were satisfactory, 
although the report says that the percentage of sulphur was very near 
the limit allowable for the purifying apparatus. The amount of sulphur 
contained in these coals was over 4 per cent, whereas of all the Philippine 
coals analyzed in this laboratory (over 100), none have contained such 
a high percentage of sulphur. 
The majority of our coals have been classed as non-coking. This 
Bureau has made several attempts to obtain a satisfactory coke, but has 
not yet succeeded. If we are successful in securing such a one from 
the Philippine coals, it may have too high a sulphur content to be used 
in an iron furnace, but it could be utilized in other ways. However, 
although the sulphur might interfere in some respects, the coals appear 
to be in every way suitable for use in a producer gas plant. The fact 
that they are non-coking has an advantage in this method of their utiliza- 
tion. The ash can be discharged easily and there will be no clogging of 
the furnace. 
To assist in the demonstration of the efficiency of the coals of the Phil- 
ippine Islands in the manufacture of producer gas, their gas-producing 
power upon dry distillation has been investigated. A proximate or an 
elementary analysis of a coal is not by.any means a good criterion as to 
the amount and quality of gas which can be obtained from it, for one 
must know certain physical properties which can not be determined by 
an ordinary analysis. Attention has already been called to the fact! 
that, in consulting most of the written works on gas manufacture, the 
chemist finds only a description of a rather large amount of special 
apparatus which is designed for use in the gas works. However ingenious 
the apparatus may be, it is generally too expensive and too cumbersome 
for the purposes of the scientific chemist. The instructions accompany- 
ing the apparatus Jeave nothing to the skill and judgment of the analyst. 
One publication,'* after mentioning the size of a cast-iron D-shaped retort 
to be used, says: 
On starting a test of a sample of coal it is of course very essential that an 
average sample is taken and small pieces, about the size of a walnut, used. Three 
quantities of 2.24 pounds each will be found sufficient to give a reliable opinion 
* Rhodin, J. G. A.: J. Soc. Chem. Ind. (1902), 19, 12. 
“Phillips: Engineering Chemistry, London (1902). 
46941——__7 
