284 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1916. 



predict just when this period is reached, although usually a great 

 divergence of opinion prevails, due to whether the prophet is a pro- 

 ducer or a consumer. As a rule, the crest of the production curve 

 is not a sharp peak, but is represented by a more or less wavy dome, 

 showing that the production remains near the maximum for several 

 years. In the case of the Pennsylvania-New York curve, the period 

 of high production extended over about 10 years; that of Ohio and 

 West Virginia over 8 years ; and that of Illinois over 2 years. Fol- 

 lowing this period the production is more or less irregular, but in 

 general decreases at a fairly regular rate, the rate of decrease, based 

 on the previous year's production, becoming gTadually less and less 

 as is indicated in the theoretical curve in the diagram. In figures 

 this decrease may be tabulated thus, the basis of computation being 

 the maximum yearly production for the field : 



Per cent of maximum production. 



At end of first 10 years 50 



At end of second 10 years 30 



At end of third 10 years 20 



At end of fourth 10 years 15 



At end of fifth 10 years 12. 5 



At end of sixth 10 years 10 



Average for 10 years , 75 



Average for 10 years 40 



Average for 10 years 25 



Average for 10 years 17. 5 



Average for 10 years 13. 75 



Average for 10 years 11.25 



The usual development history in the period of high production 

 is, first, a decrease of yield followed by increase in price, then re- 

 newed development activity, with a resultant increase in yield, a 

 fall in price, and so on until the development reaches a point where 

 the production of new wells fails to make up the decrease of the old, 

 when the final period of decrease begins. 



ESTIMATED FUTURE PRODUCTION. 



The following table gives the estimated future production of 

 petroleum in the United States, together with the approximate 

 figures as to the proven and prospective oil-bearing areas, and a 

 summary of the principal points regarding the occurrence and 

 character of the oil. 



The figures of future supply take into account a certain per cent 

 of the prospective oil area, as the curve on which they are based 

 pertains to an area where new fields have been added from time to 

 time as development progressed. In the case of Texas, Wyoming, 

 etc., where the ratio of prospective area to proven area is high, the 

 future supply may be considerably greater than that predicted if 

 the bulk of the prospective land proves productive. 



COMPARISON WITH DR. DAy's FIGURES FOR 10 08. 



Table IV is a comparison of the estimates given b}^ Dr. Day for 

 1908 with those bj^ the writer for 1915. They are here presented by 

 fields in order to correspond with Dr. Daj'^s divisions. 



