CADY AND MCFARLAND.| Composition of Natural Gas. 245 
Six samples, Nos. a to f were taken from the same well on 
the 18th, 25th, 28th and 29th of October, the 24th of November 
and the 4th of December, 1884. The analyses are supposed to 
show the variation in composition of the gas from this well: 
TABLE No. 10. 
a, b. Cc. d. e f. h. 
Marsh-gas.........- SACD NN ONLOM| dll Onl mGonZzon mOON(OMmt49soSnEOus00 
Elvdnromenry a. .sc... - 9.64 | 14.45 | 20.02 | 26.16 | 2903 | 35.92 | 22.00 
IDNA. oc oma Cee eone 5.20 A 80 3.60 5 50 7:92) 1230) Co 
Olefiant gas......... 0.80 0.60 0.70 0 80 0.98 0.60 1.00 
Oxygen es. 2. 2.10 1.20 1.10 0 80 0 78 080} 0.80 
Carbon monoxid.... 1.00 0.30 1.00 0.80 0.58 0.40 | 0.60 
Carbon dioxid........ 0.00 0 30 0.80 0.60 0 00 0.40} 060 
INiGhogeneers |. 23.41 2.89 0.00 0.00 0.00 0.00 | 8.00 
The variations above shown are truly remarkable, but since 
the data from which the hydrogen, methane and ethane were 
calculated were without much doubt obtained from explosion, 
the differences are probably apparent rather than real. (A 
little potassium hydroxid working into the burette from the 
carbon dioxid absorption pipette with each analysis would 
nearly, if not quite, produce all of these extraordinary results. ) 
What purports to be an average of the above six analyses is 
given in the annual report above referred to, page 676, and 
this has been rather widely quoted, but the figures cannot be 
obtained from those in the table by any ordinary method of 
averaging, and look like “happy means.” They are given in 
the table under h. 
Along this same line are the two following analyses of Grape- 
ville, Pa., dry gas made by Mr. Morrell, chief chemist for the 
Cambria Iron Company, February, 1886 :1°4 
TABLE No. 11. ANALYSES OF GRAPEVILLE DRY GAS. 
HebGae hebslis: 
IMIG NE AVEY" Gata G OCB oe Se ORI IDEA enc an 35.08 14.93 
OH MENTG). Gage Glee ocean nnn 28.87 39.64 
INGIERO SMa ee nie sere e cea cae tee oom eer 6 27.87 18.69 
lnlsnobaayere ay San So cee oe Byavetiste veut emotes 7.05 24.56 
Qe AMES AST. ste cre sev seeieieiece evel oo 0.17 0.96 
Oar ETI RPT ro eI seas or cud She amasnnemers ela ve 0.16 1.22 
CEMA NGI) THOMODdGl Kenic éeup E66 oacboo DOG 0.22 trace 
Cam DOMECIOXId aan. «teenie since cietisicuesale 0.58 trace 
100.00 100.00 
What has been said above concerning the variation appar- 
wmtly found by Mr. Ford applies here as well. In a foot-note 
134. Pa. Geol. Surv., Ann. Rep. 1885, p. 677, published in 1886. 
