FEMS a GMNED SEDIMEI‘US WETH THE AID OF THE SPECTROGWs-é AS APPLEED 'FO THE MISSISSIPPIAN = DEVGNEAN SEQUENCE EH OGfiMAW CQUNYY, MICE-{EGAN Thests for H19 Degree of M. 5. MZCE‘EEGEN SMTE E‘NES’EESETY John Hardin Hefner 1957 JHESIS [J LIBRAR Y Pwfiichigan State University A mm moo m m 0mm or mm smmms'wrm m: AID or m ammonium AS man no m mmmmm man In comm: comm, mm mmmm A mm submitted to the College of Science and Arts of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of mmwscmm Department at Geolog 1957 A RAPID ME'JHG) FOR 'IHE CORREIA'H'QV OF mammal) SEDIMEN'IB WIMEEADOF'IEESPE'JBWASAPPLEDWEE LEBSISSIPPTAN-DEVQIW EMINENCE m OGEMAW COW'H, MICHIGAN mmmmm ABS'JRAC'I' A review of methods for correlating stratigraphic units indicates that there is a paucity of ways to establish time-line horizons in the geologic column. In attempting to detemine another method of locating time-line horizons, the hypothesis was set forth that at certain times in geologic history, rare elements, or cannon elements in abnormal amounts, were deposited fran the sea over a wide sedimentation area simltaneously and thus would be time-line horizons. 'Ihe hypothesis was tested by analyzing the solubles of sedi- ments from the Mississippian-Devonian sequence in the Michigan Basin. A spectrograph was used to make the analyses, and since qualitative results were negative, a rapid method of making semi- quantitative determinations with this instrument was devised. The elements in the samples tested which seemed to show varia- tions in their relative quantities were calcium, magnesium, har- ium and strontium. Final tests were made concerning the relative quantities of barium and magnesium in samples from rotary wells in the Michigan Basin. The results Obtained by plotting the relative ratios of barium to magnesium.against depth were capable of being corre- lated, but these correlations are not believed to be time-line correlations. It appears that the samples were contaminated in such a.manner as to.make the results that were obtained a type of rock.unit correlation. Therefore, the tests of the hypothesis that thme-lines could be established by semi-quantitative analysis of the elements involved were inconclusive. The fact that results which could be correlated were obtained indicates that the spectrographic procedure which was developed here may'be worthy of use in other similar work. .Also, the hypothesis that time-lines can'be established by determining variations in.the amount or elements precipitated from.the sea has not been proven or disproven here. Perhaps this idea would merit further investigation. ACMW'IB A great deal of the work outlined in this paper was done in collaboration with Mr. David 8. Paige. 'lhe construction of the densitometer and the formulation of ideas by Mr. Paige greatly facilitated the completion of this project. he writer is indebted to Dr. B. '1'. Sandefur for his helpful suggestions and for correcting the manuscript. 'Jhe late Dr. S. G. Bergquist also displayed genuine interest in the project and offered considerable aid towards its completion. {the writer would also like to thank the Michigan Geological Survey and the members of that organization who were of assistance in obtaining the samples which were used. comma LISTOFJIHJS'JBATImS . . . . mmnm......... Statementocfproblem. . Section analyzed . . . . Areastudied...... General method of attack SPEEGCOPI......... Terminology. . . . . . . Originofspectra. . . . Qualitative spectrograph ic technique Quantitative spectrographic technique . Semi-quantitative spectrograph ic technique Selection of lines for measurement of optical density. W O O O O O O O O O O O O O O O O O O O 0 0 O O O DEVEIDHMT 0F mm MENIQUE Obtaining the samples. . Preparing the samples. . Making the electrodes. . Loading the electrodes . O Arcing the samples and developing the spectrogram. Reading Optical densities. . . Calculating relative ratios . Plotting the ratios. . . mmormas.. CWOUBICHS......... rd 5' (g (D \ooooooooxmmp-H l...- O 13 16 17 18 18 18 19 19 20 23 2h 26 27 29 31+ OFTEAL mam-mm INNS]!!! CHART. mm. 0 O O O O C O O O O O O 0 Well no. 1 Well no. 2 Well no. Well no . 313W . 3 1; P886 . .39 ..h7 Figure IIIIB'IBATIGS Page Mirical characteristic curve of photographic Wim. O O O O O O O O O O O O O O O 0 O O 0 O O 15 beatingthesannplesvithacid...........l8 loadingtheelectrodes...............20 Positive print of a typical group of sanmle spectra showing'lhelinesusedforanalysis . . . . . . . .21 Densitometer made‘ for use with celluloid film Ships. 0 O O O O O O O O I I O O O O O O O O O O O 23 Welllocationindexmap..............36 Curves shoving changes with depth of the relative ratiosorfbarimtomsgnesimn. . . . . . . . . . . 37 Curves showing trends of average relative ratios d buim m ”€81“. O O O O O O O O I O O O O O 38 iv A RAPID mason Fm 'IEE CORREIA‘I'ICN 0F FINE-m SEDIIM‘JB WI'E'IEEAJDOF'IEESPEC'JBWASAPPIJEDTO'IEE MISSISSIPPIAN-DEVCNIAN MOE IN 06524”! C(IN'H, MICHIGAN BYJ’OBNEARDEEEFNER mmonwm In recent years, stratigraphers have made many advances in solving the problems encountered in correlating stratigraphic units. One of the most important realizations has been that the environment in vhich a sediment is deposited has a great influ- ence on the resulting strata. With this realization, stratigraph- ers have recognized that former correlation methods are not all equivalent to one another, and therefore it has become necessary to review and classify methods of correlation. A good review of correlation problems and methods of correlation is given by Knmbem and Sloss (1951, Chapter 10). In general, It may be said that stratigraphic correlation is the demonstration of equivalency of stratigraphic units ,. and that this equivalency may be expressed in rock, biostratigraphic or time-rock units. In studies involving geologic history, paleogeography and facies mapping, great use is made of time-rock units, that is, strata that were deposited within a specific interval of geologic time. However, time-line horizons, which are the bomdaries for time-rock units are, as yet, established in only a fev portions of the geologic column. Rock units and biostratigraphic units are very common and correlations of these units are not too difficult,‘but they muy'not'be parallel to time-lines so they are not necessarily time-rock units. ‘Formations, members and other rock units are determined by the physical Characteristics of the strata, and since these phys- ical characteristics are easily observable, rock unit correlation is the most cannon method in use. Also, since nearly all geo- physical methods of prospecting are based upon physical char- acteristics of the strata such as ability to reflect or refract sound waves, spontaneous potential, resistivity, radioactivity and so forth, all correlations made‘by using these geophysical methods would be rock unit correlations. Rock units do not necessarily have any direct relationShip to time-lines because the type of sediment deposited is dependent upon the environment of deposition. Gradual tectonic movements may cause an environ- mental condition to change position as time passes. This causes strata to transgress time and is evidenced in.many geologic sections by overlap, offlap and other time transgressive strati- graphic relationships. .Also, lithologic facies changes can occur where two or'more different environments exist adjacent to each other making the determination of time-lines in such a situation extremely difficult. Rock units have been used as time-parallel horizons in many studies because of a.lack of any other horizon which will serve as a'better time marker. ‘However, stratigraph- ers realize that in doing this, they'may‘be introducing consider- able error into their studies, and thus they cannot place too much confidence in their results. Biostratigraphic units are strata.characterized by their fossil assemblage and are usually called biozones. Since the organisms comprising the fossil assemblage are dependent upon the environment in which they could live, biostratigraphic mits are subject to the same limitations as rock units in regard to using them as time-rock mits. Often, biozones will transgress time, and biofacies will result from two or more adjacent environmental conditions. However, many organisms living in the past were adaptable enough to be able to exist in many different environ- ments. 'lhus, biozones based upon these organisms can be con- sidered to be time-parallel, and are the best working tool that the stratigrapher has available for establishing time-line hor- izons in the geologic column. Using these time-parallel fossils, various stratigraphers and stratigraphic emissions have estab- lished fairly reliable time-rock divisions of systems and series in North America. However, for lesser time-rock divisions, fossils can be used only locally and their time-parallelism is more in doubt. Another difficulty in using fossils for estab- lishing time-rock mits is that great thicknesses of strata are often encomtered in thich no fossils can be fomd, or if fossils are scarce, the lowest and himest fossil in the section may not be discovered causing incorrect location of the boundaries. ‘lhis is especially true in subsurface work where macrofossils are seldom seen and almost all paleont010gical studies are limited to the use of microfossils. Other methods have been tried for establishing time-line horizons. 'Ihe radioactive ”time-clock” of uranium has been used to date certain time-rock units, but inaccuracy of determinations and nonoccurrence of uranium in some sediments makes this of little value to the stratigrqaher . Certain natural phenomena have occurred in the past to place some time-line horizons in the geologic section. Volcanic ash was, at times, blown over wide areas of ancient seas, and then settled to form bentonite layers which are time-parallel. Cyclical sedimentation and, on a smaller scale, varves afford time-line horizons which the stratigrapher may use. However, it can be seen that there are few usable time-line markers available to the stratigrapher and that any type of time-line horizon which could be added to this list would substantially aid future geologic studies. Statement of problem- Since there is an apparent lack of satisfactory methods by which time-rock correlations can be made at the present time, it seems evident that any research along this line would be useful. One of the newer approaches to the establishment of time-lines in ancient sediments has been the use of geochemistry. It has been hypothesized that during the deposition of sed- iments certain comparatively rare elements may become simultan- eously entrapped in all sediments being deposited in a basin. If such a horizon of an element could be found by some geochem- ical method, this would provide an excellent time-line marker Just as a bentonite Layer does. Since it is known that elastic sediments vary with environment of deposition, as is shown by lithofacies changes, it does not seem likely that examination of elastic components of sediments would yield any widespread time-line markers. On the other hand, if the water in a basin were of consistant composition throughout the basin, and if a rare element were introduced into the water, the mobility of the element in solution would cause the element to be spread over the entire area of the basin. Subsequent deposition of the ele- ment by chemical precipitation would provide a time—line marker. 'mus, since the element would be chemically precipitated into the sediment, it should be possible to extract it by treatment with chemicals so that it could be analyzed apart from the variable clastic portion of the sediments. Section M- 'lhe section chosen for this study is a thick sequence of Mississippian-Devonian strata in the Michigan Basin. this sequence is made up almost entirely of shales. sligztly calcareous, with small sandstone lenses appearing in the Berea and Bedford formations. The stratigraphic sequence stulied consists of the Antrim, Bedford, Berea, Smbm'y and Cold- water formations with a total thickness of approximately 1500 feet. lhis section was chosen primarily for develomnent of a technique and not because any special correlation. problems are involved. lhe sandier Berea-Bedford sediments can be different- iated fron the shales above and below and the Smbury shale is much blacker than the Antrim and Goldwater shales. The tap of the Smbury formation is probably the most distinctive horizon in the sequence. Although this study was made mostly on shales, it shoud be emphasized that the methods used can be applied to any rock type and not to shales alone. Area stxfiied- Since the main objective of this study was to locate a horizon which could be used for correlation, samples fran fairly closely spaced wells were used. Four wells were chosen in Foster, West Branch and Mills Townships, Ogemaw County, Michigan and sample cuttings from these wells were obtained from the Michigan Geological Survey. However, a similar study was being carried on at the same time in Grant and Sheridan Townships in Clare County, Michigan and data from this study were made available to the writer for use in longer range correlation. A map indicating the locations of the wells used in these studies appears on page 36. General method g attack- Since only the non-clastic frac- tion of the sediments contain the elements which were synchron- ously deposited by chemical precipitation from the sea, only the solubles of the sediments were analyzed. When analyzing the sol- ubles from elastic sediments, only small amounts will be avail- able, so it was decided that the analyses could best be made spectrogaphically because with this instrument large samples are not needed. Also, with the spectrogaph the presence of an element which may occur in a very small amount in the sample can be determined quickly and a record on film may be permanently retained. At first, it was hoped that time-line horizons could be located in the sequence of shales by means of trace element or minor element analyses. Spectrograms of samples from one well were analyzed qualitatively to determine whether or not there were any trace or minor element horizons in the stratigraphic column. In this examination, the elements calcium, barium, strontium, magnesium and iron were the only elements identified, and they were present in all of the samples. ‘Ihus, there were no trace or minor element horizons found, nor were there any horizons characterized by the absence of any of the above named elements. These results were much the same as the qualitative results obtained by 81088 and Cooke (19%) in a similar stuly of the IMadison.Limestone. In other words, qualitatively, all of the samples were remarkably similar in character. While making the qualitative studies, it was noticed that the various samples contained different amounts of the five elements found. This was evidenced by the varying density of the spectrographic lines of each of the elements in.different samples. For example, one sample would have dark'barium lines and light.magnesium lines while another sample would have light bariun lines and dark mamasiun lines. 'niis imlicated that perhaps a.quantitative study might provide time-line correlations which qualitative studies were unable to do. That is, a stratum» such as one of low“bariumrhigh magnesium content,:might‘be found which could be traced throughout the basin as a time-line marker. It seemed evident that the best.method of attacking this problem was to determine a.ratio of the concentrations of the two ele- ments for each sample, and then.this ratio, whiCh may'have sig- nificantly high or lovaalues, could.be used for correlation purposes. Thus, the problem at this point was essentially a two-fold.one. The first part was to develop a quantitative spectrographic procedure which was rapid and also retained enough accuracy so that results capable of interpretation could be Obtained. The second part of the problem.was to interpret the data obtained and determine whether or not any relationship for time-rock correlation existed in the sediments as hypothe- sized earlier. SPECTROSCOPY Many theories of light are involved in spectrographic work and their discussion would be such to complex to be included in this study. Further information along these lines may be obtained from any of the many books which have been written on the subject of spectroscopy. hree excellent books which the writer used for reference are those canpiled by Abrams (1950), Brode (19’43) and Harrison, Lord and loofbourow (19%). However, a resume of these theories is apprOpriate here because the entire study is based upon spectrographic determinations. Terminolgg- A spectroscope is an instrument which uses a refracting prism or a reflecting grating to disperse light into a spectrm of its component wave lengths. When light entering the spectroscOpe is passed through a slit and diffracted by a grating, the different wave lengths are completely separated and the spectrm consists of many lines which are made by each com- ponent wave length of light. A spectrograph is an instrment which records the spectrum on photographic emulsion, and the resulting picture of the spectrm is a spectrogram. 9_r_igi_n _oz_f_ sp_e_ctr - When atom are subjected to an excita- tion force such as an are or spark, they are known to emit light of definite wave length values. he excitation force pushes electrons into the outer levels of the atom where they are m- stable. he electrons fall back to their normal level, losing the energy that was imparted to them by the excitation force by emission of light. The frequency of the emitted light is directly proportional to the amount of energy lost by the falling elec- tron, which in turn is the same as the ammznt of energy the atom absorbed fran the excitation force. lheoretical. and experimental work has demonstrated that this absorption and emission of energy takes place in accordance with fixed and well-defined rules, and electrons which are forced away from the nucleus can ”junp" only to certain predetermined positions. Also, the electrons can return to their original positions only through a certain number of finite moves, either returning frcn their unstable positions to their normal positions in one move, or stopping at an inter- mediate level before returning to their original position. Each different electron movement in one atan involves a different amount of energy, and thus each emits light of its own unique wave length. All of the different wave lengths of light emitted by the electron movements within stuns of one element make up the spectrum of that element. Since atomic structures of the elements are all different, each element will produce its om characteristic spectrum. Qualitative spectromhic technique- Since each element gives its om identifying spectrum when arced, the problem of qualitative determination of elements in an unknown resolves into spotting each elemental spectrum on a spectrogram. {this problem may be approached in several ways. If certain elements are sus- pected of being present in the unknown, a pure sample of each of these elements can be arced so that the spectrogram of the known and unknown samples are in juxtaposition. If each major line of a known sample coincides with a line of the unknown sanzple, the suspected element is actually present. Other lines on an unknown spectrogram, which cannot be accowted for in the above manner, can be identified by measuring their wave length values and referring these to apprOpriate tables. The wave lengths are best measured by arcing an iron spectrogram adjacent to the wknown spectrogram. {then by referring to pub- lished charts of the iron spectrum, in which the wave lengths of the iron lines are given, the wave lengths of the wknown lines can be interpolated from the adjacent iron lines of known values. It will be fowd that a line whose wave length has been deter- mined could possibly have been caused by several elements. How ever, upon measuring more lines it will. turn out that all of the lines belong to the spectra of one or two elements. 'Jhe pres- ence of these elements can then be checked by identification of the principle line of the elements or by juxtaposition of spec- trograms of pure samples of these elements with spectrogram of the wknown as described above. {this process is carried out wtil all lines in the unknown spectrogram are identified as belonging to the spectra of one of the identified elements. Quantitative Whic technique- A line of a partic- ular element on a spectrogram may vary in density depending upon how brigat the light of that wave length is in the spectrum One of the factors controlling the intensity of the light, and thus the density of the line on the spectrogram, is the amowt of the element which is present in the sample. 'Ihus, if other factors affecting the density of a line on a spectrogram can be eliminated or be held constant, a quantitative determination of 10 the element can be made by measuring the density of the line. Elhe density of the line can be determined by using a densitome- ter to measure the amowt of light which the line can transmit. {me factors which effect the density of a line besides the amowt of element in the sample include such variables as the exposure time, character of the arc, time lapse between arcing and exposure, development time of the spectrogram and the char- acter of the photographic emulsion. Quantitative spec‘ln'ochemical procedm'es have been developed for elimination of these variables so that the exact percentage of an element in an wknown sample can be determined. 'nxese procedures include the use of internal standards, the use of known angles, and the calibration of the photographic emulsion. Elbe photographic enmlsion must be calibrated to determine how it reacts to different intensities of light. This is accanp- lished by exposing the plate to liwt of known relative intens- ities by use of a step density tablet or other means. 'lhen optical densities of the different relative intensities are meas- ured and plotted one against the other. 'lhese points are con- nected to produce the characteristic curve of the emulsiw fran which optical density readings of lines in the known and wlmown samples can be converted to their relative intensities. Samples with know concentrations of the elements being tested can be prepared and arced onto the same spectrogram as the wknown sample. 'lhen by comparing the relative intensities of the lmow and wknown samples, the quantity of the element in the wknown could be determined. However, since other factors besides quantity affect the intensities, these variables must be 11 eliminated too. 'lherefore, an internal standard must be used in the known and wknown samples. An internal standard is another element which is present in the samples in large or constant quantities and is spectrochemically related to the elements being analyzed. 'lhe internal stanlard may be introduced into the samples or it may be already present in the material being analyzed. 'Ihe optical densities of a line of the internal stan- dard are also measured and their relative intensities determined. then a ratio of the relative intensity of the element being meas- ured to the relative intensity of the internal standard is cal- culated, and this ratio is used in further work. The use of the internal standard will compensate for sample variations in ex- posure time, character of the arc, and development time. hat is, if a sample were exposed for a longer time, the relative intensities of both the element being measured and the internal standard would be increased, but the ratio between the two would remain the same. Care must be exercised in the selection of an element for an internal standard or this will not always be the case. [the previously mentioned references discuss the inportant spectrochemical factors which must be considered in selecting an internal standard. 'Ihe above described relative intensity ratio is computed for all samples on the spectrogram. 'nzen by plotting this ratio of the known samples against their concentrations, a "working curve" is established. 'lhe relative intensity ratio of the wknown sample can then be inserted into this "working curve" and the concentration of the element in the sample can be determined . In emery, quantitative spectrographic procedure involves the following steps: (1) Arcing the wknown and several know samples onto the spectrogram. (2) Development of the spectrogram. (3) Calibration of the photographic emulsion and construction of the characteristic curve. (h) Measurement of optical density of lines of the test materials and internal standard in all samples and determination of the backgrowd reading for each line meas- ured. (5) Conversion of optical density values to relative intensity values by use of the characteristic curve. (6) Cal- culation of ratios of relative intensities for all samples. (7) Construction of the "working curve" from known sample con- centrations and their relative intensity ratios. (8) Quantita- tive determination of the test material by insertion of relative intensity ratio of the wknown into the ”working curve'.‘. Semi-quantitative spectroggphic technique- As can be seen above, the determination of the absolute percentage value of a constituent in a sample by spectrochemical methods can be quite an arduous task. This is especially true if there are many sam- ples to be treated or if there are several elements in each sanple which need to be analyzed quantitatively. Since this study in- volves many samples with several elements to be determined in each, it was deemed necessary to develop a method of analysis which would speed Lp the process and would still obtain reliable results. 'Jhe nature of the desired results in this study as well as the similar spectrographic behavior of the elements involved allow several short-cut methods to be applied. As mentioned earlier, a ratio of the concentrations of two 13 of the elements in each sample would be the best values to use for correlation purposes. In using this ratio of the concentra- tions of two elements, it is not necessary to determine the abso- lute percentages of the elements in order to calculate the ratio. that is, the ratio can be calculated just as effectively from the ratio values of the relative intensities of the two elements. lhis refinement of normal spectrographic technique eliminates the necessity of using known solutions for construction of a "working curve", and thus reduces the amowt of work involved for each sample a great deal since arcing and reading of line densities of known smles is eliminated. Also, since ratios of element concentrations are to be used in the correlations, the use of an internal standard is not nec- essary with the elements analyzed. The reason for this is that the spectrochenical behavior of the elements analyzed are so similar that the same internal standard could be used for both elements, and thus, the intensity readings of the internal stan- dard cancel out in calculating the ratio used in correlation. 'mat is, using the elements barium and magnesiw for example, the ratio would be calculated in the following marmer: [Ba] ReIe Of 33 6.3. g = 3.1. of Internal 4s't'enderd""" x _(R.I. of Ba K [Mg] r ReIe Of k - ROI. Of "8 R.I. of Internal Standard 0.3. : Ratio used in Correlation; RJ. - Relative Intensity; K 2 cmmte The constant, K, in the above equation arises from the fact that even if the magnesiun and barium lines which are measured have the same density, they do not necessarily have the same concentration in the sample. 'me determination of this constant is not deemed 1h to be important in this study so that the correlation ratio used is just a ratio of the relative intensities of the chosen lines of the elements. One final short-cut was employed. By standardizing the procedures in developing the spectrogram, it was fowd that the characteristic curves of different spectrograms varied very little. ‘merefore, an empirical characteristic curve was estab- lished (figure 1) and used for all spectrograms. Tests indicated that any error introduced by this step would be less than other errors which are inherent in the sampling procedure. A table was made of the values from the characteristic curve so that easier conversion of optical densities to relative intensities could be accomplished. this table appears on page 39. ioo[._ % ":00 80y . p 1 ~20 s : M-uo ho” _ ' ‘ ”50 2? ” '7 d j; ;. ,, 3 "" 3° . ‘ j 70 a a “ «a :7: 20 “i ‘ I" 80 g :1 >_‘ ‘ _. 85 3 15 a . 3 1° : 1590 g 8 F :92 a-_ g 6 e -~-9h z, ’_‘-..--.- age 100 200 hoo 800 Relative Intensity Figure 1. Empirical characteristic curve of photographic Wimse 15 hus, with the above mentioned refinements, the semi—quanti- tative procedure involves only the following steps: (1) Arcing several wlmown samples onto each Spectrogram since known samples are not needed. (2) Development of the spectrogram. (3) Meas— urement of Optical densities of selected lines in the samples and determination of their backgrowd readings. (h) Conversion of the readings to relative intensities by use of the table. Selection gag lines for measurement 2; optical density- A line of the spectrum of each element to be determined must be chosen for measurement of its Optical density. he line thich is selected for a particular element will then represent the con- centration of that element in the sample. he lines chosen for each element should be of approximately the same intensity in most of the samples. his being the case, exposure time can be regulated so that most readings of optical density will be con- verted to their relative intensities using the straight portion of the characteristic curve, and result in more reliable corre- lation ratios being obtained. Another factor that must be con- sidered in choosing a line for measurement is that there must not be an interfering lines from spectra of other elements in the sample. Also, when samples are arced on carbon electrodes, carbon wites with nitrogen in the air to form cyanogen which produces a band spectrum. his cyanogen “backgromid” often covers a sizable portion of the spectrogram, depending upon how long the smle is arced. Obviously, the lines measured must not fall within the area covered by the cyanogen ”background". When these requirements in choosing a line are satisfied, any 16 further choice of available lines may be made for convenience or "film saving" reasons. WT he instrument used in this study was a Cenco grating spec- trograph equipped with a replica grating. he excitation source was a D. C. are and the spectrograms were produced on Ansco Isopan cut sheet film. Quarter inch standard spectrographic carbons manufactured by National Carbon Cmpany were cut into two inch lengths and used for electrodes. Strips of soft wood were cut and. drilled for use as electrode and sanple bottle holders. Standard sanple bottles, such as those used by the Michigan Geological Survey for storage of well sanples were used for preparation of the well cuttings. Chemically pure concen- trated hydrochloric acid was diluted to 6 N with distilled water and used for sample treatment. A separatory fwnel was used to introduce the acid to the samples. A hot plate in a fume hood was needed to sinner the samples in the acid, and a drying oven was used in preparing the electrodes. Standard darln‘oom equip- ment was needed for deve10pment of the spectrograms. he densitometer used. for the density determinations was constructed especially for this type of study. Paige (1952), who constructed the densitometer, gives the details for the construction of the instrwlent, in which an extra-sensitive light meter was used to obtain the optical density readings. Construction of the densitometer was necessary because no densi- tmeter equipped to handle celluloid film was available. 17 DEVEIDHMT 01'I IABORA'IW WIQIIE he development of a concise laboratory technique was an important part of this stuly. hereforc, a resume of the more significant technique develOpments which were made will be inclined here, while a complete step by step procedure will be outlined later. Obtaining the sagles- he samples used in this study were rotary well cuttings obtained from the Michigan Geological Survey. Approximately one gram of cuttings from each interval, which was usually ten feet, were placed in sample bottles and used for anal- ysis. Since ratios of element concentrations were to be used in the final calculations, the quantity of cuttings treated was not a critical factor in the research. Care was taken not to include am cuttings which appeared to be extraneous to the whole sample. 32993295 the safies- he sample bottles containing the cuttings were placed in the bottle holders, put on a hot plate ‘ -..I~- Fight: 2. Treating the sanples with acid. 18 in a fuse hood, and nearly filled with 6 n hydrochloric acid (figure 2). he hot plate was turned on and the cuttings sim- mered to hasten the action of the acid. his also disaggregated the sanple so that complete solution could take place. he bottles were then removed fran the hot plate and set aside to allow all insoluble particles to settle out of the solution. _Ma_lgi_ng the electrodes- Preliminary tests indicated that standard spectrographic carbons were sufficiently free of the elements being tested to make it wnecessary to assme the added expense of using spectroehemieally pure carbons for the electrodes. Tests were also made to determine what kind of electrode surface was most suitable for these experiments. It was decided that flat-topped electrodes worked the best, and that are wandering was reduced if the positive electrode was partially sharpened. Therefore, the carbons were cut into two inch lengths, and both ends leveled and polished on emery cloth and the positive elec- trodes were partially beveled with a small pencil sharpener.. he electrodes became contaminated upon firing a sample, but by using both ends, each electrode could be used for two samples. Loading the electrodes— he negative electrodes were placed in the electrode holders and heated in the drying oven to remove anymoisture andmake themmoreporous. henaneyedropperwas used to introduce a drOp of liquid from each sample onto the surfaces of the electrodes (time 3). he eyedropper was rinsed in two different beahers of distilled water after each use. Care wastakentoreeordwatsamplewasputoneacheleeu’ode sothat the data would not get mixed. he electrodes with the droplets 19 on them were left in the drying oven to evaporate the liquid and leave the salts of the samples baked onto the electrodes. he electrodes were kept in the drying oven until Just before they were to be arced because the salts would absorb moisture from the air readily and this increased the difficulty of maintaining a steady arc. Figure 3. Loading the electrodes. Arcing the smgles and developing the spectm- These two processes were so interwoven that the techniques involved are best described together. Since there are so many steps to be perfonnend in this portion of the work, it was found that two people performing the Operation together greatly facilitated the procedure. One person could be kept busy changing the electrodes, adjusting their heigit and arcing the sample while the other 20 Operated the shutter, moved the plate holder for each sample, and recorded the sample mmbers and their positions on the film. Also, while one person was developing the spectrograms, the other could be cutting film and loading the plate holder for the next samples. Sheet film, cut to size to include only the desired portion of the spectrogram, was used (figure 1+). Two plate holders were employed and the unused portions of the plate holders were blocked off by taping old film into them so the film for the spectro- grams could be fitted only into their proper positions. One plate holder was marked with tape, and the film placed in it was notched so that the spectrogams could be identified in the dark and not be transposed in the developing process. Figure h. Positive print of a typical group of sample spectra showing the lines used for analysis. In this portion of the work, it was necessary to develop a method of operation which would save time and would keep frm transposing the sample spectrogram. he individual sample spectra could not be marked until the spectrogram was developed and dried, but by organizing the procedure, spectrograms were 21 Ioept separated in the following way: he spectrogram develop- ing process provided three "resting places” for the spectro- grams. hesewereinthefix, inthefinalwashandonthe dryer. hus, it was possible to have two spectrogram (one notched)onthedryer, twointhewashandtwointhefixwhile two more were being prepared by arcing samples on the spectro- graph. A rotation between these resting places was maintained, marking the sample spectra with their code number as they came off the dryer, and thus, the time per sample analysis was greatly reduced. Preliminary testing also provided some technique improve- ments in the arcing of the samples. Ordinarily the time lapse between arcing the sample and exposure of the film is quite an important spectrochemical factor to be considered in performing an analysis of this type. However, tests showed that one sample could be exposed after any of several different time lapses and still obtain the same results. Also, a sample electrode could be arced twice with the same results being obtained. his was because the elements measm-ed in the sample were quite similar in their spectrographic behavior, and all elements were present in sufficient quantities so that they were not burned up readily. hus, knowing this, if a good steady arc was not obtained imed- iately, it was all rigit to wait until the arc stabalized before almosing the film. Also, an occasional spectrum would be unac- countably poor on the spectrogram. This would be noticed as soon as the spectrogram had been placed in the fix, so this samle could be re-run imediately frm the same electrode which 5 was used for that sample before. hese methods were important time-saving procedures. Ming ggtical densities- he densitiometer was so con- structed as to allow film strips the width of the spectrmn to pass by the slit as in a movie proJector (figure 5). herefore, ‘ ' U ‘ ' L O K t \ l . i \ . . . I ' O C Figure 5. Densitometer constructed for use with celluloid film strips. the spectrogram were cut into the individual spectra and spliced together to form the film strip. To do this, a razor blade and 81’.th edge were used to cut the spectrogam on the side oppos- ite the emulsion, after which bending along the cut would break offthe strip. heemulsionwasscrapedoffofoneendofa strip, the opposite side of the other end was scraped, and then the strips were spliced with regular splicing cement. he seven 23 strips in a spectrogram were spliced together and the ends of the long strip were notched so they could be hooked together in a loop throng: the densitometer. The densitometer was equipped with a dial so that by setting it with the ”starting line" in the proper place, the line to be measln'ed could be quickly passed over the slit. A light inten- sity adjustment made it possible to take constant background readings for all lines, so the density readings for the lines could be converted to relative intensities directly by use of the chart shown on page 39. It will be noticed that the density readings shown’in the chart have an inverse relationship to those shown on the empirical characteristic curve. he reason for this is that the photocell used in the densitometer was a sensitive ligit meter, and consequently it was calibrated to register light intensity instead of line density. herefore, low readings were obtained frm lines of hid) optical density and vice-versa. he chart was made up to correct for this change automatically. Calculating relative £9322" Of the five elements, calcium, magnesium, barium, strontium and iron, thich were present in the samples, density readings were first recorded on the lines of all elements except iron. he lines which had been selected to represent these four elements were those of the following wave lengths (figure u): Calcium- M25 A; magnesium- 518k A; barium- hssh A; strmtium— 1607 A. Iran was discarded because small pieces of iron from the drilling bit were noticed in the orig- inal sample cuttings. Although those pieces or iron which were seen were removed, it was believed that other pieces may have 21+ remained to contaminate the sample as far as the quantity of iron was concerned. On sane spectrograms, the cyanogen back- ground was great enougi to interfere with the selected calcium line. Tests on the samples which had calciun lines that could be measured seemed to indicate that the use of the calcitmi lines in the ratios would not provide any data which would be usable for correlation. herefore, rather than try to reduce the cyan- ogen background on the poor samples, the calcium line was not wed and emphasis was concentrated upon what information could be obtained from the barium, magnesimn and strontium lines. Since it was not known what ratios might be most usable in correlation, various ratios were calculated for samples from wells in a selected portion of the strata in order to determine that would be the best ratio to use. Ratios of magnesium to strontium, bariun to strontium, and barimn to magnesiun were calculated. All. of these ratios appeared to be somewhat usable, but the ratio of barium to magnesium seemed to exhibit a. greater magnitude of change and appeared to be the best ratio. It was then noticed that strontium seemed to act somewhat as an internal standard in the samples, and for samples in which the barium- strontimn ratio was low, the magnesim-strontium ratio was big), thus explaining why the barimn-magnesitmi ratio had a greater amplitude of change. Also, it was noticed that the magnitmle of the bariun-strontium ratio was about twice as great as the mg- nitude of the magnesium-strontitm ratio. Tests showed that by doubling the intensity of magnesium in the magnesitmi-strontium ratio, the magnitude difference between the bariun-strontitmi ratio and the magnesitn-strontitm ratio was minimized. herefore, 25 it appeared that the best ratio to use was one involving the relative intensity of barim and twice the relative intensity of magnesium. his doubling of the magnesium intensity is strictly empirical and it is not believed to have adversely affected the final results. One more refinement of the ratio was used to help stabilize the results. his was the placement. of barium in both the miner- ator and denominator of the ratio to cause the results to behave more like percentages. herefore, the ratio of relative inten- sities which was used throughout the study was that of barium to twice the magnesium plus bariun. hat is, with R. I. referring to relative intensity, the ratio used was: ' Re Is 0: 38.1.11“ 5r. I. of MagnesiumT+ R. I. of Barium Plottg the ratios- When the samples were obtained, the depth at which they were collected was recorded. Knowing the elevation of the'wells, the positions of the samples in relation to sea level was determined. hen the ratio which was calculated for each sample was plotted on a horizontal scale against its sea level elevation on the vertical scale. Since each samle represented an interval of strata, the points were plotted at the middle values of the intervals . When all the points for a well were plotted, they were connected by straigit lines to males a "dog-tooth" curve to be used in correlation. Occasionally, there would be a gap in the sampling, so points on each side of such a gap were connected by a dashed line, indicating the lack of data. hese curves which resulted from plotting the ratios of barium to magnesium are shown in plate I. 26 For reasons which will be maplained later, it was deemed necessary to plot a "trend curve” for each well. he values for this curve were obtained by averaging the ratio values for every three adJacent points and plotting this value at the elevation of the middle point. hese cln'ves are shown in plate II. PRmEJURE Applying the various techniques outlined previously, the following procedure was followed in analyzing the samples: 1. 2. 3. 5. 7. Put about one gram of sample cuttings from each interval into their respective sample bottles and label. Place sample bottles in the bottle holders and onto the hot plate. Nearly fill all bottles with 6 n hydrochloric acid. Turn on hot plate to simmer the samples until they are well disaggregated. Remove samples from hot plate and allow insoluble particles to settle. Prepare negative electrodes by cutting carbons into two inch lengths and polishing ends with emery cloth. Put them in the electrode holders and place them in the drying oven. Load the electrodes by placing a drop of sample liquid on each with an eyedropper. Rinse eyedrOpper in two beakers of distilled water between each loading. When all electrodes are loaded, close oven carefully and allow the liquid to evaporate leaving the sample salts on on the electrodes. Do not remove electrodes from the oven until Just before they are ready to be arced. Prepare the positive electrodes by polishing the ends and 27 9. 10. 13. 1h. 15. 19. 20. sharpening them slightly. In darkroom, cut film and place it in plate holder. Put plate holder on spectrograph and adJust it to proper heigzt. Remove baffle. Install electrodes into clamps on spectrograph. Adjust negative electrode to proper heigat so that center of gap will be level with the center of the slit. Close power switch. Arcthe sample by lowering the positive electrode to contact the negative electrode and then raising it imediately to obtain the optimun electrode gap of about one half inch. When are is burning steadily, open shutter and expose for about four seconds. his time may be lengthened if the arc wanders . Close shutter. Open power switch. Remove electrodes from clamps and replace them in electrode holder. Move plate holder into position for next spectrun. Record the sample number and its position on the spectrogram. Repeat steps 10 to 15 for all samples (seven) on one spectrogram. - Close baffle and remove the plate holder from spectrograph. In darkroom, remove film from plate holder and develop as follows: Twc and one half minutes in Kodak D-19 developer, one half minute in running wash water, and approximately ten minutes in acid fix. Wash for convenient time and dry. After spectrogram is dry, put the identifying sample number on each spectrogram. Cut film into individual spectra and splice them together 28 as described previously. 21. Insert spliced film strip in densitometer and hook ends together. Line up "starting line" with edge of film track and set calibrated dial to starting point. 22. Turn on ligut. Rotate film to position for obtaining back- ground reading. AdJust light so that background reading is 100 on light meter. 23. Pass the lines to be read slowly over the slit and observe and record the lowest readings. 2h. Obtain the relative intensity for each line by use of the chart. 25. Calculate the correct ratio for each sample. 26. Determine the correct sample elevation and plot the ratio cm'ves. IRWIN OF RESUME 'Ihe dogtooth curves of the bariun-magnesiumratios for each of the wells are plotted in plate I. Since the Sunbury formation is the most distinctive horizon in the portion of the stratigraphic colunn analyzed, the tap of the Smbury was plotted and used as a reference line in correlation. 'Ihls horizon which was deter- mined on well logs, is shown as a black line on plate I. The Sunbury formation also appears to be distinctive as far as the barim-magnesim ratio is concerned because the ratio is low within the Smbury in all wells. 'nie red line correlating the lowest ratio values within the Sunbury does not exactly par- allel the line at the top of the Smbury, but this is not to be expected since variations may be caused by the Goldwater—Smbury 29 contact being either gradational or abrupt. However, a general relationship is apparent. Other correlations of the high and low points of the dogtooth ratio curves have been made as indi- cated by the red lines in plate I. Since the ratio correlations show a general relationship to the distinctive Sunbury formation, it is believed that the correlations in the Goldwater and lower formations may be as good, and thus a mans of subdividing these formations is established. Besides having high and low values which could be correlated, it was also noticed that each of the ratio curves had distinctive trends. It was difficult to determine exactly what the trends were on the dogtooth curves, so a trend curve for each well was constructed as shown in plate II. Each trend curve was made by averaging the ratios of three adjacent points and plotting this average at the elevation of the middle point. hen the trend curve of the average ratios was "painted" by making the line of the curve wide enougl to cover every plotted point. . For refer- ence purposes, ”base lines” were arbritrarily drawn for each curve, and these are the black lines in plate II. he trend curves show two siglificant relationships. First, all wells show a definite change in the trend of the average ratios at about the location of the Sunbury formation in the stratigraphic column. In the Goldwater formation, the trend curves all lie to the right of the base line except for minor variations in each well, while in the Sunbury and lower formations the trend has shifted so that the greater part of all curves lie to the left of the base line. he other relationship which appears to be present is that the position of the base line 30 varies from well to well. hat is, the ratio value of the base line which separates the upper and lower trends is different in each well. he relationships which have been pointed out in the dog- tooth and trend curves all are dependent upon the relative occur- rence of barium to magnesiun in the samples. If it is assured that the quantity of barium and magnesia: present in the sample is derived by precipitation of the elements into the sediments, it my be supposed that the ammmt of these elements in the sedi- ments is in turn dependent upon the concentration of each element in the sea water in which the sediments were deposited. Ranhama and Bahama (19.9, p. 295) list the amounts of elements supplied to sea water and the amount of each present in sea water. By making barium-magnesim ratios fran these data, it is seen that the ratio is much hiher in the source than in the sea water. hus, a higher ratio of bariun to magnesia: would be expected to be deposited near the shore, a fact that is supported by Ranksma and Sahama who say that barium is associated with near shore sediments. In this study, sediments most indicative of near shore depo- sition are the sands and sandy shales of the Berea formation. his formation, as picked on the electric logs, is thinner than the sample interval in six of the seven wells analyzed and there- fore any distinguishing characteristics may well be masked by the adjacent strata. However, well no. h in Ogemaw County penetrates llh' of the Berea sandstone between the elevations of -322' and 436'. his sandstone was probably deposited closer to shore than the shales above and below were, but both the dogtooth and 31 trend curves show a very low barium-magnesia: ratio for this interval instead of a high one as would be expected. herefore, the assumption that the concentration of barim and magnesium in the smples is the result of precipitation of the elements fran sea water is probably incorrect. Perhaps the cause of the anon- alous occurrence of a low bariun-magnesium ratio in the Berea sandstone lies in the fact that the samples which were analyzed in this study were rotary drill cuttings. Drilling mud used in ' rotary drilling mually contains barite to increase its specific gravity. hus, althomh the cuttings were washed and separated fran the drilling mud, there is still a great possibility that the samples may have been contaminated by the barite. herefore, it is possible that the bariun content which was determined in these samples did not originate by precipitation from the sea, but was introduced to the cuttings fran the drilling mud. he fact that the samples probably were contaminated by the barite does not mks it necessary to discard the results obtained, because it appears that contamination of the samples may have taken place as a selective process. If it is supposed that all of the samples were subject to contamination by barite in the drilling mud, and if when the samples were washed, it was much easier to remove the contaminating mud from the sands than frm the shales, the cuttings analyzed would be expected to have a higher barim-magnesium ratio in the shaly sections than in the sandy sections. his may explain why the Berea sandstone in well no. h has such a low ratio while the shales on both sides have a high ratio. Since the bariun-magnesiun ratios may be strongly influenced 32 by the drilling mud, the correlations which were made on the dogtooth curves cannot be considered to be time-line correlations. Instead, the correlations might be a type of rock unit correla- tion if the barium content in the samples varies according to the sediment type. Supposing this to be the case, interpretations as to the enviroment of deposition can be made from the results which were obtained. he low ratio trend values in the Sunbury and underlying formations possibly indicate that these sediments were deposited nearer to the shore than were the overlying Cold- water shales, and the high and low points on the dogtooth curves may indicate that the shore line fluctuated many times during the deposition of these sediments. Also, it could generally be as- suaed that a well near the center of the basin might have a base line of higher ratio than would a well near the shore line. Since well no. 3 in Clare County and well no. 1t in Ogemaw County have base line values of 270 and 275 respectively and the other wells have values of 300 or more, it my be concluded that the positions occupied by these two wells were nearer the shore than the other wells timing the interval in which these sediments were deposited. By comparing the relative localities of the wells on the location map on page 36, it is seen that this possible con- clusion is in accordance with the results of other studies which indicate that the nearest shore line during this time was to the southeast in the Findlay arch area. his agreement adds support to the assuaption that the correlations which have been made may be sane type of rock unit correlations. One other factor which may have influenced the results obtained in this study is the fact that cave-ins often occur in 33 the drill hole, especially when a shale section is being drilled. hus, the cuttings analyzed may have contained sane slmnp mater- ial fran higher in the well rather than Just the cuttings for the calculated depth. However, by not using amr cuttings which seemed to be extraneous to the sample as a whole, it is believed that much of this possible error was eliminated. In portions of the section where closely spaced sample control was present, the plotted points almost always fell on sane sort of curve, with no great deviations by any one point. Thus, it appears that the results were not greatly affected by sllmlping in the well, but this is not definitely known. CGCIUSICNS he results obtained in his stlxly could not prove or dis- prove he hypothesis that time-line correlations could be made by quantitative examination of elements in the solubles of sedi- ments because he samples were apparently contaminated. However, it is suggested hat further research along his line would have merit, especially if work were first carried out on closely spaced outcrop samples of known relationships, so hat the method could be tested against known results. In working wih outcrOp samples an important factor hat should be considered would be the effects of differential weathering on the samples. .Should he hypothesis prove to be true, he method could hen be extended to include greater outcrop areas, cores, and possibly well cuttings if non- contaminated elements were analyzed. he correlations which were made in this stuly appear to be valid rock unit correlations, but here are much better methods 3h available for making rock unit correlations, so this is of little significance. However, he fact that results which were capable of being correlated were obtained indicates that the spectrochem- ical ideas introduced here may have merit and could be worthy of me in furher research. he spectrographic technique which was developed was hor- ouglly tested and gave reproducible determinations even though may time—saving practices were employed. It is believed that if he spectrograph is to be used by he stratigrapher to make quantitative type determinations for correlation, a procedure such as used for his study must be employed or he time and costs of running many samples will be prohibitive. Further work along his line might be facilitated by use of a spark source for exciting he samples. he case of firing and timing a spark source would probably reduce possible errors due to selective volatilization of he constituent materials. Selective volatil- ization was not of great importance in this study because of the spectrographic nature of he elements analyzed, but in a study involving other elements, his could be an important factor to be considered. Finally, he possibility of extending the mehods developed in his paper to other types of sedimentary rocks should be emphasized. Facics studies of intervals bounded by time-lines are becoming increasingly important to he stratigrapher, so that if a spectrographic method such as developed here could be used for locating time-lines in any type sediment, it would certainly be a significant contribution to he field of strati- graphy and sedimentation. 35 ZNNH. «at... 2 am .H. i. eosbmm L2 9.3 838m 4mm :5sz 8.38m: 32: miSe m .. miSa maofloommzz Ewen miSa wgaoommwzzazz o .3850 3.30 2H HN _H. mm gwwomm Ifimhmm g I a mm 830% is a ”Es Izm jam mm: x05 dogmas duo? .m .N .H «mo mmmm m8 68 mm? Hmw: mmmma mmbma :fia’é .3850 anaemo do: we 83on“— NHQZH egg gums a penance m:m Mm m mNm ma m _ .s \F 3.3: _ m. 5;. 3.33% . h use _ _>»z:oo Emma. 34502 __ 2am a._.'.J unpack — ML a... '4de .m madman O [ Ill Il[. 35868 mange _ . 33925 Page .2 S n. _ m. H . it _ m . .n m." a aofi— 00—— :1 “J 5;! “1...;— gnaw: £39300 0.3.8 can 3888 Ho “.2: .8525 Haas .zONB PLATE 37 PLATE u... . L Suva... ”mm 38%..» ., .. WV. vL w“ A re; . a V .r... l.‘ ; u .. . a? ON. 1 5.. -3 .J .... .. u. \. . .. . i. .. L ; ,_ .. . .,. ,v.~ . .. . n ‘ VT _ u. R. X ‘! WL LQ/¥ a... :6. a . a w _ . .1 w. '2} (r. f _1 4 Hr; . ‘~ 1 I,’ ELEV it p , r- --< flay} .\.‘ o IIIII!.IIIV .-."-’. SK’» Goldwater fm. Sunbury fm, méLfivf -~aQUWW _ ‘ u ‘ W . W _ . b . W W V _ . _ _ _u s W _ . W . u 1 _ . . _ . ~ n W ._, W . W _ V W . c a — . l _ _ . a h v . a . W h a 1 ., W A W . n _. _ a V g. . v u _. . . I 1 ,v . _ __ W . . . V W m — _ q W. . _ h u . . l a W q . . l W . V _ ., W _ . .. W _ . m . fl — . ~ i . \ . W W W , W a . L _ V W H . u. a u l . .e W W n W V _. .. k a u W _ . .. .. . W n .a . . . {-4400 . W ., W 4 ~ 7 .. W v W _ . q .. W . _ _ W .. _ a ‘ ~ i _ _ . a W — . w ' W W . W W W V W W . W 4 _ .— . . . n W % W. W . V V W W o i i W . . \ . W W . W . ~ _ W W _ V W V l W . W .e W W . L o W . . i W W W W W . l a 4. ‘ - a W _, .V _ W e. _ 4. W , a W . 4 W . W _ W i i l i . ,. . v _ ._ . , l , W W W W . . . . , . _ . V W . _ . x . W . _ . l W h ~ . a .e u a . _ W . W _ ~ A . ., W W . W _ . . W, __ W W W , V W . v . , W. W. VV W W .._~ N. v . W W W . W W W W W W V w W W W . - li-‘ _. W _.V*.—.—~--—r»~—~a—o~-%Um. ,., : “—.u - .- r -. WWW—WWWWWW.-. _..-I I i ( . . A o C T‘xv‘flém v i "N h ‘--»~v..—i .__,.-, .__ 240a - -. W WM , ~-fi 1 J l omen. amen!" - mums mmsm cm:- 1. 585 21. 397 1+1. 326 61. 279 81. 225 2. 570 22. 392 1+2. 323 62. 277 82. 220 3. 555 23. 388 1+3. 321 63. 275 83. 217 1.. she 2h. 38. m7. 319 61». 272 81.. 213 5. 525 25. 382 hs. 316 65. 270 85. 208 6. 510 26. 379 1+6. 313 66. 268 86. 20h 7. 500 27. 375 In. 311 67. 266 87. 199 8. £790 28. 370 118. 309 68. 26h 88. 19h 9. 1.82 29. 365 1.9. 306 69. 262 89. 185 10. 1.75 30. 360 50. 30h 70. 260 90. 180 11. 1770 31. 357 51. 300 71. 256 91. 175 12. 1.60 32. 353 52. 297 72. 25h 92. 170 13. #51 33. 351 53- 295 73. 2&9 93- 165 11:. M2 31.. 3&9 51+. 292 7h. 21.7 91». 160 15. l+39 35. 3&3 55. 290 75. 21.5 95. 155 16. 1:33 36. 31:0 56. 289 . 76. 2171 96. 150 17. 1.25 37. 337 57. 287 77. 238 18. hi6 38. 33h 58. 285 78. 235 19. 1‘09 39. 332 59- 283 79- 232 20. 1.01 ho. 330 60. 280 80. 227 a- 'me values in he left hand columns are actually light intensities raher han optical densities as explained on page 2h. 39 DAM the data listed on he following pages was obtained by he procedures described previously and used to construct he curves of plates I and II. the data for he four wells in Ogemaw County are he only figures included here, but data for he hree wells in Clare County are given by Paige (1952). Only he values needed to plot he curves in plates I and II are listed wih his paper. Since he line densities were converted to relative intensities by use of he chart on page 39, only he relative intensity values will be included here. the first calm of he following data sheets gives he sumle code mnber and he second column lists he mean elevation of he sample interval. the hird and fourh calms give he relative intensity values of magnesia: and bariun respectively, while he fifth colunm lists he value of twice he relative intensity of magnesia plus he relative intensity of barium. the sixth column gives the ratio of he values in calunn four to he values in calm five, and is he value hat is plotted against he sample elevation in he dogtooh curves of plate I. 'Ihe sevenh calm is he average of the ratio values of hree adjacent samples and his average is listed opposite he middle sample of he hree. Column seven values were plotted against he sample elevation to locate he points by which he trend curves in plate II were plotted. ho wm. so. 1, comm coum (m1 # 16755). EIEVATIM: 1213' Relative Intensities Ba-ljg Ratio Sample Sea Level Average number Elevation 55 _B__a miBal Ba + (2215: Ba) Ratio 1 580 190 388 1368 280 2 570 277 260 811 . 319 301 3 560 332 290 951 301 307 1 550 160 388 1308 297 306 5 510 287 266 810 317 306 6 530 525 160 1510 305 308 7 S20 332 287 951 302 315 8 510 260 266 786 338 323 9 500 l#82 175 1139 330 339 10 190‘ 277 297 851 3119 338 11 1180 323 323 969 333 331 12 170 351 330 1032 320 321 13 160 289 268 816 317 327 11 150 238 219 725 311 310 15 110 270 199 739 270 301 16 130 182 109 1373 298 296 17 120 321 300 912 319 316 18 1110 297 295 889 332 321 19 100 525 175 1525 311 319 20 390 190 151 1131 315 320 21 380 277 275 829 333 321 22 370 510 500 1580 316 328 23 360 - - - - - 214 350 360 365 1085 336 311 25 310 330 256 916 280 311 26 330 525 1190 1510 318 317 27 320 319 3119 987 3514 335 28 310 330 330 990 333 333 29 300 321 290 932 311 329 30 290 2115 256 716 3M 331 31 280 319 326 961 338 333 32 270 510 500 1580 316 331 33 260 309 316 931+ 338 326 3h 250 300 287 887 3211 335 35 210 170 190 1130 312 326 36 230 525 175 1525 312 332 37 220 275 285 835 3111 326 38 210 297 285 879 321 328 39 200 279 262 820 320 328 10 190 510 525 1515 310 332 11 180 182 190 1151 337 310 12 170 283 295 861 313 310 113 160 280 287 8117 339 333 11 150 116 388 1220 318 332 15 110 219 256 751 310 338 16 130 260 287 807 356 339 17 120 280 261 821 320 337 18 110 295 297 887 335 330 11 m no. 1, 0mm comm (mar # 16755), Continued. Relative Intensities Ba-Mg Ratio Sample Sea Level Average Number Elevation 3g pg (fig +3212 Ba + (2142+ Ba) Ratio 19 100 287 290 861 336 335 50 90 311 311 933 . 333 339 51 80 215 262 752 319 351 52 70 285 3119 919 380 351 53 60 297 297 891 333 318 51 50 283 279 8145 330 326 55 10 160 125 1315 316 338 56 30 277 323 877 368 338 57 20 510 500 1520 329 317 58 10 208 217 633 313 335 59 00 277 277 831 333 335 60 -10 500 190 1190 328 315 61 -20 227 180 631 281 319 62 -30 292 306 890 311 320 63 -10 256 251 766 332 311 61 -50 313 332 958 3117 3311 65 -60 323 309 955 321 338 66 -70 332 315 1009 312 323 67 ~80 3115 300 . 990 303 330 68 -90 337 357 1031 3116 331 69 -100 353 370 1076 311 311 70 -110 311 311 933 333 339 71 -120 301 311 919 339 338 72 -130 297 309 903 312 317 73 -110 297 331 928 360 351 71 -150 219 268 766 350 351 75 -160 215 266 756 352 316 76 -170 266 270 802 336 338 77 -180 289 280 858 326 327 78 -195 266 219 781 318 327 79 -210 309 313 931 336 330 80 .220 261 266 791 335 338 81 ~230 292 306 890 311 311 82 -210 279 306 861 351 335 83 -250 321 285 927 307 336 81 -260 260 275 795 316 321 85 270 301 285 893 319 331 86 -280 321 313 955 328 317 87 -290 300 261 861 305 320 88 -300 182 170 1131 327 299 89 -310 311 225 817 265 293 90 -320 331 268 936 286 281 91 -330 287 238 812 293 292 92 -31+0 319 270 908 297 293 93 -350 112 360 1211 289 286 91 -360 101 300 1102 272 281 95 ~370 285 225 795 283 286 96 -380 331 292 960 301 295 WELL 110. 1, 0mm comm (PERMIT # 16755), Continued. Relative Intensities Ra-fig Ratio Sample Sea Level Average Number Elevation fig _R_a {gig +Ba2 Ba+§§g+ Ba) Ratio 97 -390 313 261 890 297 300 98 -100 300 256 856 299 293 99 -110 290 227 807 ' 282 297 100 4120 330 295 955 309 295 101 -130 323 268 911 291 305 102 -110 316 287 919 312 305 103 .150 182 500 1161 311 319 101 -160 306 268 880 305 311 105 4170 272 227 771 295 299 106 -180 313 261 890 297 309 107 -190 289 290 868 331 299 108 -500 525 379 1129 ‘ 266 302 109 -510 292 256 810 305 286 110 -520 337 272 916 288 292 1.1.1 -530 309 215 863 281 281 112 -510 319 219 887 281 272 113 -550 182 323 1287 251 268 111 -560 510 379 1399 271 262 115 -570 510 365 1385 261 273 116 -580 301 211 819 281 279 117 -590 295 2111 831 290 295 118 -600 309 277 895 310 265 119 -610 170 227 1167 195 211 120 .620 331 185 853 217 210 121 -630 332 185 819 218 225 122 -610 316 199 831 210 251 123 -650 3115 300 990 303 268 121 -660 319 215 913 260 299 125 -670 316 316 918 333 265 126 -680 326 165 817 202 289 127 690 323 319 965 331 279 128 -700 331 292 960 301 308 129 -710 351 283 985 288 283 130 -720 311 213 835 256 282 131 -730 297 256 850 302 279 132 -710 315 268 958 279 281 133 -750 311 232 851+ 272 276 131 -760 500 381 1381 277 286 135 -770 277 217 801 309 293 136 -780 323 268 9111 293 303 137 -790 280 217 807 306 286 138 -800 297 208 802 259 283 139 -810 313 217 873 283 277 110 -820 297 211 835 288 273 111 -830 275 180 730 217 275 112 -810 326 261 910 290 271 113 -850 301 211 819 281 '43 141111.110. 2, omw comm (mar 36! 16529). summon 1238' Relative Intensities Ba-flg Ratio Sample Sea Level Average Number Elevation 55 93 (Quad) Ba+(§4_giBa) Ratio 1 710 365 290 1020 281 2 700 116 311 1113 _ 299 231, 3 690 1151 331* 1236 270 311 1 680 211 287 769 373 312 6 660 M2 332 1216 273 293 7 650 116 379 1211 313 297 8 51“) 351 309 1011 305 3m 9 630 101 323 1125 287 293 10 620 120 337 1177 286 298 11 610 301 287 895 321 305 12 602 319 285 923 309 327 13 595 297 319 913 350 335 11* 585 109 112 1260 350 337 15 575 330 297 957 311 311 16 565 375 280 1030 272 287 17 555 351 283 985 287 286 18 515 332 295 959 308 300 19 535 337 295 969 .305 305 20 52 5 370 319 1059 301 301 21 515 321 280 922 301 328 22 505 1+16 353 1185 298 328 23 195 357 139 1153 381 328 21 185 392 315 1129 306 333 25 175 321 292 931 313 307 26 165 319 275 913 301 292 27 1155 279 199 757 263 290 28 115 319 309 1007 307 300 29 135 301 300 908 330 311 30 125 370 323 1063 301 301 31 1115 351 270 972 278 299 32 105 351 323 1025 315 301 33 395 292 262 816 310 302 31 385 375 292 1012 280 293 35 375 323 262 908 289 235 365 397 316 1110 285 289 37 355 357 297 1011 291 291 38 315 375 326 1076 303 299 39 335 357 301 1018 299 309 10 325 101 381 1186 321 311 11 315 300 280 880 318 331 12 305 353 382 1088 351 332 113 295 375 365 1115 327 336 11 285 392 381 1168 329 321 15 275 315 306 996 307 323 116 265 388 388 1161 333 32.1 17 255 365 151 1181 382 31.5 ‘18 215 360 315 1065 321 315 11 Sample Sea Level number IElevation '19 5o 51 52 53 51 55 56 57 58 59 60 61 62 63 61 65 66 67 68 69 70 71 72 73 71 75 76 77 78 79 80 81 82 83 81 85 86 87 88 89 90 91 92 93 91; 95 96 235 225 215 205 195 . 185 175 165 155 115 135 105 95 85 75 65 55 5 -5 -15 -25 -35 -115 -55 -55 -75 -85 -95 -105 -115 -125 ~135 -115 -155 -165 -175 -185 -195 -205 —215 -235 -215 -255 -265 -275 -285 -305 -315 Relative Intensities H8 316 330 160 112 101 392 290 370 397 109 170 139 332 357 277 365 353 292 376 3115 315 397 331 382 330 351 360 297 326 397 353 370 360 382 323 375 310 109 g; 313 365 379 101 330 39.-Lg Ratio WELL 110. 2, 0mm: comm (PERMIT # 16529), Continued. Average (2142 + 89) Ba -:— (2142 + 13a) Ratio 1&5 9115 1025 1299 1285 1136 1200 1019 1093 1227 123’4 1373 1257 1021 1071 331 356 292 312 323 316 130 323 352 337 315 302 3119 335 311 31.1 286 297 311 333 330 308 318 306 330 308 327 310 359 317 331 315 289 311 302 306 277 301 326 312 295 288 296 330 298 298 289 337 326 320 309 323 321 293 305 308 309 295 287 WELL N0. 2, WW 001m: (PERmZ‘I' # 16529), Continued. Relative Intensities Ba-yg Ratio Sample Sea Level Average Number E1avation 3 Ba 5313+ Ba) Ba -:- (21154-35) Ratio 97 -325 375 283 1033 271 261 98 -335 190 277 1257 220 291 99 -315 285 360 930 287 302 100 -355 353 300 1006 298 323 101 -375 381 306 1071 281 308 102 4:17 283 309 875 353 330 103 122 357 392 1106 351 317 1011 4127 283 283 8119 333 330 105 -137 300 260 860 301 307 106 -112 353 280 986 281 29 107 -117 360 290 1010 287 285 108 .152 - - - - - 109 -157 360 285 1005 281 297 110 -165 360 310 1060 320 292 111 -182 319 262 960 273 281 112 -187 306 213 825 258 277 113 4192 375 323 1073 300 277 1.11 -197 - - - - - 115 -502 382 285 1019 272 272 116 -515 133 280 1116 211 262 117 ~525 375 277 1027 270 255 1.18 -535 116 277 1109 250 263 119 -515 382 283 1017 270 263 120 -555 112 326 1210 270 270 121 -565 175 319 1299 269 267 122 -575 375 268 1018 263 258 123 -585 I133 277 11113 212 259 121 -595 139 330 1208 273 261 125 -605 109 300 1.118 268 278 126 -615 381 319 1087 281 285 127 -625 112 365 1219 292 271 128 -635 382 238 1002 237 261 129 -615 388 279 1055 261 257 130 —655 139 326 1201 270 273 131 -665 381 306 1071 285 270 132 -675 353 256 962 256 272 13 . -685 39 297 1081 271 267 1 -695 116 311 1113 272 273 135 -705 3811 287 1055 272 273 136 ~715 353 268 97': 275 266 137 -725 370 2119 989 252 259 138 -735 357 238 952 250 257 139 ~715 109 300 1118 269 258 110 -755 116 285 11.17 256 266 111 -765 - - - - - 112 -775 392 292 1076 272 219 1113 -785 3119 19“ 892 218 251 111 -795 392 280 1061 263 11211110. 3, 081m: 0011111 (1111111 # 1831). EIEVATIQU: 856' Sample Sea Level Hmber Elevation BuswmdePwNI-J PH :00 55 U) wWNNNNNNN Nb) HP ngHOWmdeFW$HOGmd 1» U1 55$5§£fi§5$833 100 88 73 55 37 25 15 5 -5 -15 -25 -35 -117 -60 -70 -80 -90 -lOO -110 -119 -131; -152 -165 -175 --185 -195 -205 ~215 -225 -235 -215 -255 Relative Intensities .3 217 227 217 211 217 2% 225 213 232 219 220 211 217 22 2% 227 235 235 321 283 232 232 235 235 232 238 321 235 227 215 219 211 215 238 211 235 309 227 217 22 323 227 232 232 330 2% 2% 93 217 175 211 217 219 238 211 199 201 219 208 165 175 213 39:15; Ratio Average (2,1143 4:38) Ba + (2143 4- Ba) Ratio 17 711 $9 675 6% W3 750 691 $5 668 717 618 707 6% 6% 701 662 6% 661 906 760 6w 705 7(2 5% 681 708 8% 290 so 715 7% 6? 680 333 279 358 311 335 318 319 318 306 333 321 318 250 316 321 311 307 293 280 255 310 312 330 326 318 328 326 319 311 315 323 218 289 300 313 277 211 278 3% 322 311 332 271 329 317 285 310 323 3116 Sample Sea level Ember Elevation 19 50 51 52 53 51 55 56 57 58 59 60 61 62 63 51. 65 66 67 68 69 70 71 72 73 71 75 76 77 4150 ~160 4470 -180 -1190 -500 -512 -521 -537 -560 -530 -590 -600 -610 Relative Intensities £8 251 235 232 235 235 316 235 311 232 217 219 235 227 319 227 232 235 217 235 P: 220 WHO. 1|», 0mm! GMT! Sample Sea level Runber Elevation (”de FWNI—l 613 633 626 618 612 605 595 595 Relative 21.6 fig. 213 313 357 2111 211 311 219 211 966 595 690 678 Intensities 11211110. 3, 0mm: comm (m # 1831), Continued 88.-Mg Ratio 317 323 319 326 319 319 312 259 263 311 255 (m1 # 1623). 2137mm: 38.-Pg Ratio 835 1197 1107 815 971 1069 1233 867 255 262 322 296 218 296 283 278 Average (2143-;- Ba) Ba + (2113 +138) Ratio 310 315 323 321 321 317 297 278 279 277 287 272 271 2% 800' Average £2135: Ba) Ba -:- (255.129.) Ratio Sample Sea Level Ember Elevation 9 10 11 12 13 11 15 16 17 18 19 20 21 22 23 21 25 26 27 28 29 30 31 32 33 31 35 36 37 38 39 10 11 12 13 M1 15 346 17 11.8 19 50 51 52 53 51 55 56 um :10. 1, 0mm: comm (m # 1623). Continued. 575 567 562 555 5117 5’11 535 530 525 Relative Intensities £6 351 112 510 151 1139 351 l139 g3 ($5180.! Ba+§§g+3a2 Ratio 89 1236 1011 1101 1265 1220 1009 995 738 859 918 952 1319 1039 1055 1019 1062 1092 BurygiRatio Average 207 221 188 223 273 211 272 329 112 331 287 335 277 276 263 276 288 280 290 281 275 269 212 272 298 271 281 291 293 275 252 281 306 281 293 301 303 286 263 281 276 282 306 285 273 288 281 278 276 283 -289 291 316 286 252 283 280 260 218 261 256 260 275 279 306 282 266 279 265 259 216 257 259 252 252 261 272 266 271 281 297 268 232 251 221 212 271 256 373 261 219 271 292 271 282 291 111111.110. 1, 00114111 0001111 (1mm # 1623), Continued. Relative Intena ities Ba-gg Ratio Sample Sea level Average umber Elevation 155 §_a @331 13a) Ba+ (fig +341) Ratio 57 222 387 210 1131 299 291 58 217 375 309 1059 a 292 310 59 207 3‘15 353 1013 338 303 60 191 357 375 989 278 301 61 182 160 370 1290 287 283 62 170 330 262 922 281 281 63 130 357 266 980 272 279 61 121 332 260 921 282 296 65 113 357 360 10711 3311 327 66 105 109 170 1288 365 331 67 87 332 277 911 291 310 68 62 109 306 1121 272 288 69 17 109 319 1167 299 295 70 10 392 360 1111 311 301 71 20 1109 331 1152 290 300 72 10 337 283 957 296 316 73 -5 381 1139 1207 363 336 711 ~22 316 310 972 350 339 75 -35 3811 337 1105 305 320 76 ~15 392 3115 1129 306 303 77 -61 375 319 1069 298 299 78 ~69 353 292 998 293 293 79 -77 365 295 1025 288 281 80 -85 388 277 1053 263 282 81 -97 365 3011 10311 2911 28h & ~117 351 295 997 296 291 83 ~11o 375 295 1015 282 281 81 -160 397 285 1079 261 256 85 -170 170 270 1210 223 251 86 -180 357 260 971 267 250 87 Duplicate - - - - - 88 -190 182 337 1301 259 272 89 -200 357 292 1006 290 277 90 .208 382 300 1061 282 271 91 -216 116 275 1107 ~ 219 273 m .223 101 326 1128 289 277 93 -231 388 319 1095 2% 291 9‘1 -239 326 279 931 300 300 95 -2117 1101 375 1177 308 306 96 -262 315 31.1 1001 311 303 97 -287 360 295 1015 290 31.1 98 -302 1151 1151 1353 333 286 99 -307 375 232 982 236 293 100 -312 382 3115 1109 311 267 101 -318 175 323 1273 251 281 102 -326 I139 353 1231 286 258 103 -335 139 270 1118 235 252 101 -315 175 290 1210 231 235 50 111311110. 1, 0mm: comm (PERMI'I‘ # 1623), Continued. Relative Intensities Belg Ratio ample Sea level Average Ember Elevation fig 1a (23554-1111) Ba-i-‘gg-i-Ba! tio 1.05 -355 182 297 1261 236 210 106 -365 116 279 1.1.1.1 . 251 237 107 -375 500 285 1285 223 2311 108 -385 190 289 1269 228 215 109 -395 510 2115 1265 193 211 110 -106 500 270 1270 212 205 1.1.1 -119 510 270 1290 209 221 1.12 -131 175 319 1269 251 218 113 4139 352 311 1095 281 280 1111 4153 352 311 1015 306 289 115 -160 321 217 889 278 297 116 4165 392 319 1133 308 275 117 -172 182 300 1261 238 259 118 480 525 313 1363 230 2M 119 4191 I133 309 1175 263 250 120 -503 510 375 1155 258 268 121 -511 139 319 1227 281 267 122 -528 510 357 1377 259 268 123 -518 139 31.1 1189 261 286 121 -575 151 160 1362 338 287 125 -585 116 297 1129 263 282 126 -596 388 211 1017 215 2 51 127 -607 170 321 1261 251 257 128 -615 112 330 1211 272 266 129 -630 525 392 1112 272 260 130 -655 I133 266 1132 235 217 131 -675 370 227 967 235 2‘15 132 -685 170 337 1277 261+ 257 133 -695 I170 353 1293 272 268 1311 -701 375 287 1037 267 265 135 -712 379 260 1018 255 262 136 -723 182 315 1309 261 256 137 -737 370 215 985 219 212 138 -752 500 268 1268 212 218 139 -767 1175 225 1175 192 200 110 -779 510 217 1267 195 202 111 -792 175 261 1211 218 213 112 -807 510 313 1393 225 217 113 -818 151 235 1137 207 228 111 -827 133 290 1156 251 227 115 -835 5110 311 1391 221* 236 116 -815 175 290 1210 231 211 117 -855 190 220 1200 183 211 118 -865 182 268 1232 217 211 119 -875 570 365 1505 212 215 150 -885 510 215 1325 185 51 313W Ahrens, L. n. (1950). Spectrochemical Analysis. Addison-Wesley Press, Inc. Iii-ode, w. R. (1913). Chemical @ectroscqu. (2nd Ed.) John Wiley and Sons, Inc' " Harrison, C. R., Lord, R. C. and loofbourow, J. R. (1918). Practical Spectroscm. Prentice-Hall, Inc. ICrmibein, w. 0. and 81033, L. L. (1951). m_ and _S___edi- mentation. W. B. Freeman and Compaq. Paige, n. s. (1952). A R9216.___ Method f__g_r_ ‘lh___e_ Correlation of Fine- Grained Sediments with the Aid 9_f the flectrogggh g Amlied_ to the Mississimian-Devonian Seguence_ in _C____lare Comm, Michigan. Unpublished Master' 8 Thesis, Michigan State University. Rankana, K. and Bahama, T. G. (1950). Geochemistzq. The Univer- sity of Chicago Press. 81088, 1.. L. and Cooke, 8. 13.13. (1916). goguoehemieai Sm 319651-95... of Limestones. Am. Assoc. of Petroleum Geologists 13u1.1., Vol .‘3'6"—"', No. 11, pp. 1888-1898. “1"”? USE. CEELY 151.711; " '- 0-13%” .1 Date Due * b' '1‘ 1‘;- ‘1 :fixle Demco-293 I!HNININNWNIIIIIU 3 1293