LITHOLOGIC CORRELATION IN 'BEND SERIES, TEXAS. 



The reader is urged to refer constantly to 

 the graphic logs and to make his own detailed 

 comparison of the units in the two wells, espe- 

 cially as they are represented in the percent- 

 age logs. To make such a detailed comparison 

 in this paper, with the constant references to 

 depth that would be necessary, would make the 

 discussion too cumbersome, and as the facts 

 are clearly brought out by the graphic logs it 

 seems superfluous. 



There are some differences between the two 

 percentage logs that can not be readily ex- 

 plained. They may be due to defects in the 

 observations, especially to the fact that prob- 

 ably too few thin sections of samples from the 

 Seaman well were examined, or to facts that 

 are not in agreement with the general principle 

 on which the correlation is based. Besides, 

 minor variations that are disclosed in the per- 

 centage log of the Seaman well mav be con- 

 cealed in the percentage log of the Rudd well 

 owing to the greater thinness of the equivalent 

 section in the Rudd well, so that of two samples 

 representing any given thickness of beds that 

 from the Rudd well generally represents a 

 greater time interval than that from the Sea- 

 man well. 



The comparison of the units in the two wells 

 will be based on the correlation I have arrived 

 at without at first proving this correlation in 

 each case. This is a necessary result of tlie 

 principle that the correlation is based not on 

 direct comparison of the lithologic character 

 of two units but on the relation of that char- 

 acter to the character of formations above and 

 below. 



UNIT A. 



The examination of the material from the 

 Seaman well was begun with material obtained 

 just below the lowest sample that consisted 

 mainly of yellow sand; of that from the Rudd 

 well with the lowest distinctly sandy sample. 

 Th,e percentage logs bring out clearly the sandy 

 character of unit A, which differentiates it 

 unmistakably from the material that lies below 

 it. In and near the basal samples of this unit 

 from both wells were found certain fragments 

 of white sandstone which differed from others 

 in the same samples partly in being more silici- 

 fied but especially in that one surface, instead 

 of appearing freshly broken like the rest, was 

 dull, gray stained, smoothed, and slightly 



pitted, suggesting the surface of a small pebble 

 (half an inch or so in diameter) in a conglom- 

 erate. They have been represented in the 

 synthetic log by a conglomerate at the base of 

 unit A. 



UNIT B. 



In the samples from both wells a difference 

 is immediately noticeable in the character of 

 the shales that directly underlie unit A. 

 They are blacker, less blue-gray than the shales 

 of unit A, much less sandy, more calcareous, 

 and more sulphidic. In shales below unit A 

 sulphides are almost invariably present though 

 in varying abimdance. 



This unit is too thhi iii the Rudd well to per- 

 mit much generalization about its representa- 

 tion in the ]iercentage log, though compared 

 with unit A in that log it is clearly much 

 more calcareous and much less sandy. The 

 much smaller ])ro])ortion of sand in the vSea- 

 man well is also obvious in the percentage log. 

 but the increase of lime is slighter below the 

 contact. 



The character of the basal boundary of imit 

 B is very different in the two wells. In the 

 Seaman well it has a feature of special interest 

 brought out in the percentage log — a pro- 

 nounced, fairly regular increase in sandiness 

 in the upper ])art of unit C, which reaches a 

 maximum at a dei)th of 3,045 feet, above which 

 there is a decrease m the amount of sand, also 

 rather regular but more gradual than the in- 

 crease. The sand appears in sandstones and 

 shales, as represented in a generalized way iJi 

 the snithetic log. The increase in sandiness 

 of the up]ier part of miit may be oidy appa- 

 rent, being due to caving of sandy material 

 from unit B in the drill hole, though the large 

 proportion of the sandy material in that part of 

 unit C is very much against this assumption. 

 If the facts are about as represented they may 

 record an ideal l)oundary between two units. 

 Wlierever there is a sharp contact representing 

 a bi-eak l)etween two units the sharpness of the 

 contact is due to the fact that the beds repre- 

 senting the time interval between the two imits 

 have not been deposited there. Every such 

 break is due to a rise of base-level (the term 

 base-level being used m the sense given to it by 

 BarrelP-) . The true boundary betw^een two for- 

 mations is the sedimentary siu-face coinciding 



12 Op. cit., pp. 778, 783. 



