r a 1:: 2:2:glggfi; PLACE IN RETURN BOX to remove thin checkout from your record. TO AVOID FINES return on or betore dete due. I DATE DUE DATE DUE DATE DUE MSU Is An Affirmative AotloNEwel Opportunity Inetttuion W1 This is to certify that the thesis entitled ‘ 7' ”flP/r09rA/4/C’ /rt VIIV‘IQA/row 57/ J"'¢— [ta—r0 {an/Iwfl/ 57¢”“7/"9 W/Jk’“? 744‘. /WNI» Eff/r- r//fi/Ir-¢. )4 [7‘1“ r, "4 ' resente & p d by ' [4704 ’Z‘o’é’if"”lé.d"wy L. has been accepted towards fulfillment of the requirements for ‘ MUS degree inwofi ’7 . W9. : 7 . M ajoyfirofessor Date (fie. ,1 /0 ’1" /€’¢J>, - ' A l 3 u - "' ’ ‘ D ' , , ‘ . a ‘ ‘4 " I ' ,w'f». . '3 \ , A ‘ ." ' ‘~ . I “ ‘1 ‘ ‘ . 7 »" I f V ‘ (‘ ) .. . - ,— ‘1 o A J I 9 ' I ‘ \t I .a - ,' - , I , / - 7 I 3 I 1 ‘ ~ _ ,. r / ’ . y I , 1 -‘ ‘ < 3 ~ . ‘. ‘ ~ ‘ ‘ J -‘ s I ‘ f... > _ ', ’ t/ - , A » - fl .7 ’ ‘— -_ v ‘ I r K ‘ t V / ’ 1 . .’ ‘ , . ¢, ’ '1, . ‘. 1 i .. 1' . - ’ . ‘ \ "\ ( I‘ \ I l 'J 1‘3 ‘I l I I n ‘ \ I J . ,. , ‘ / I I I / \ we , , . . x- . _ ~ 4 .I ' ‘ . 7 ‘ V \. 4 \ l v— - \{ K" ' \ A - n ‘ -( /’\ L ‘ "\ \ 4’ f / \ , w r V/ ~ ,. l _ \ , , 1“ ‘ , i .— l at ‘n. \ z __ l I‘- "x“ ‘ ' , ‘ ‘\‘ I i I I :_ , I .t.. " k - - . ' - . ‘_, '. 'R - ; ~ .. x . \ a ‘. ! , ' - . , I s ‘ I N I “ ' ., ‘ | ' I . 3 . ‘7 ( ‘ . ‘4' , t _ , JI ' f! I 6 Is ‘7 x '\ \ ‘ I \ ‘ _ ‘ L . ‘ " . :~ "’.‘ ’ .3 ' ‘ A -.I ‘ r" ‘t t I ) > . v4 , . \ .‘ ‘ - 1/ . ’ l u . . ‘ l ‘ .' /‘ l . \ __ ‘ t. _. . ; - t- 7 ‘r . ~ . - \ —~ , t‘ , I ~."- , x , ’ \ ,A e. - I“ - \ \ ,. . \ l ' - ‘ x ‘ '/ ~ r t ‘ \ ~ 4 A I t K t \ n r . r J V/. ’ . \ -— l‘! I I. ‘ I'| ' ‘r \. \' \ '_ V. x" , ~ , '_ $ 1 I . _’ 1‘ I V t I ‘ ...Jf_ “— i . ‘ _ z \ ' .4-\‘-"' — 'x- '~ ~ Id. ' \kv-r H‘ - Y \' r " Y > Ill. - -" ‘ r‘ V_ ’ V‘ - . A A w - - ’l I — .. ‘ s t ' I :r w J ’4‘- ‘\ a i . ‘. A’ ". a I I .‘ I A I. A .' I t " L . “’Iv v N ~ln' ‘ .t /! w \ A PETROGRAPHICAL INVESTIGATION OF THE LONGITUDINAL DEPOSITION WITHIN THE MASON ESKER RELATIVE TO ITS ORIGIN By Ralph LeRoy Erickson A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geology and Geography 19h8 IHESIS \\w\\v\ ACKNOWLEDGEMENTS The writer wishes to express his sincere appreciation to Dr. B. T. Sandefur under whose direction the problem was undertaken. Dr. Sandefur gave freely of his time and knowledge both in the field and laboratory. Dr. S. G. Bergquist's assistance in the preparation of the thesis and arrangement of aid in the laboratory is deeply appreciated. ' The writer also wishes to express his thanks to Dr. J. w. Trow for checking the manuscript, and to Dr. F. w. Poster for his aid and suggestions in preparing the maps and graphs used in this thesis. 216976 TABLE OF INTRODUCTION. . . . . . . LOCATION. . . . . . . . . PURPOSE OF STUDY. . . . . METHOD OF INVESTIGATION ; General. . . . . . . Field Sampling . . . Laboratory Sampling. Leaching . . . . . . Sieving. . . . . . . Separation e e e e 0 Mounting for Microscopic CONTENTS Stlldy 0 Identification of Heavy Minerals Roundnees and Sphericity Measurements. CONCLUSIONS . . . . . . . Page m-qoxoxmp-rrp-w P‘ la n) c: NH 0%) INTRODUCTION One of the most interesting glacial deposits is the esker, a long narrow ice-contact ridge commonly sinuous, and composed chiefly of stratified drift. Eskers range from a few feet to more than a hundred feet in height, from.twenty to several hundred feet in width, and tram a fraction of a mile to nearly 150 miles in length. The sides are generally steep; the crests are smooth. Theorigin of eskers has been a controversial subject since they were first recognized as glacial deposits. Will- iam M. Davis!!- !Davis, William.M., The Subglacial Origin of Certain Eskers: Boston Seciety_of natural History, Proceedings, vol 25, pp h77-h99. 1895. believed that eskers were sub-glacial channel deposits. His explanation presumed a stagnant and decayed.marginal zone of the ice sheet. Water resulting from the bgsgljmelting of the ice, together with water from.surface melting, which becomes subglacial through the medium.of crevasses, is gathered into streams. Considerable detrital material is picked up by - these streams. Should these subglacial streams become over- loaded, a certain amount of detrital material will be depos- ited. If the stream is diverted, the deposit will be left in the abandoned tunnel and gradually will develop the steep lateral slopes and other fermal features of eskers during the 2. slow melting of the retaining ice walls. In 1895, warren Uphemi *Uphem,‘warren, Evidences of the Derivation of the Kames, Eskers, and Moraines of the North American Icesheet Chiefly from.its Englacial Drift: Geological Societyfiof America, Bulletin, vol 5, pp 71-86, 18911,. supported the englacial theory of drift transportation and deposition. He believed that the drift was not carried for- ward beneath the ice, being pushed or dragged along in con- tact with the land, but that the drift was carried within the ice in the lower part of the ice sheet. Such a hypoth- esis states that as the glacier thins the deposited sediments are gradually lowered to the ground. I. O. Crosbye .L *Crosby, I.O., The Origin of Bakers: Boston Society of Natural History, Proceedings, vol 50, pp375-hll, 1902. opposed Davis's subglacial channel theory, and proposed the superglacial stream.theory, whereby the sand and gravel were deposited in streams flowing on the surface of the glacier and later lowered to the ground in the same manner as the englacial theory above. A.D. Trowbridge* *Trowbridge, A.D., The Formation of Eskers: Science, new series, vol hO. p 1&5, l9lh, abstract. 3 suggested that eskers were formed by slow recession of the edges of glaciers during the deposition of kames and result- ed in the drawing out of the kames into serpentine-like ridges. This would mean headward building of the esker in successive segments, each.marked by a delta where the esker stream entered a glacial lake. Richard F. Flint* *Flint, Richard F., American Journal of Science, 5th series, vol 15, pp Lao-hits, 1928. ' advanced still another possibility: that several glacial deposits which have been described as eskers are actually crevasse fillings and are less closely associated with eskers than with marginal lake and outwash.deposits. It is probable that eskers are formed in several dis- tinct ways and places: in subglacial channels, in super- glacial channels, in englacial channels, headward building of successive kames, and crevasse fillings, all with or without the active agency of water. All these theories are not doubt essential to a complete explanation of eskers. LOCATION The Mason esker is one of the longest observed in Michigan and is approximately twenty miles in length. It extends from the gravel pit at the corner of Main and Shep- ard streets in Lansing, southeast through Mason, the county seat of Ingham County, and has its southern terminus in 1+ the Charlotte morainic system southeast of Mason (Fig 1). PURPOSE OF THE STUDY The purpose of this paper is to illustrate the value of a detailed petrographic study of the esker sand as an I aid in determining the origin of a particular esker. This study is based upon seven field samples taken from.the Mason esker at two- or three-mile intervals. METHOD OF INVESTIGATION General. I A.detailed petrographic study of any sand involves fre- quency counts of heavy mdnerals and roundness and spher- icity measurements of quartz grains. Meat sand is com» posed chiefly of quartz; therefore the quartz grains may not serve as a key mineral in the identification or recogni- tion of a particular sand. However, most samples contain several manor constituents which have a high specific grav- ity. These minerals may be amphiboles, pyroxenes, garnets, zircons, titantite, etc. If, in some manner, these heavy minerals can be separated from the quartz grains, identi- ‘fied and counted, they may serve as a factor in the corre- lation and recognition of esker sand. Roundness and sphericity are two fundamental prop- erties of sedimentary particles, and are the most recent to be studied quantitatively and statistically. Early observ- ers noted the modification of shape that took place by transportation. F. J. PettiJohna- *Pettijohn, F. J., Manual of Sedimentary Petroggaphy, D. Appleton Century Company, p 278, New'YOrk, 1938. states that the factors that control sphericity and round- ness are; (1) the original shape of the fragment, (2) the structure of the fragment, as cleavage or bedding, (3) the durability of the material, (h) the nature of the geologic agent, (5) the nature of the action to which the fragment is subject and the violence of the action, and (6) the time or distance through which the action is extended. If a roundness and sphericity value can be obtained for quartz grains of a particular deposit, this value would serve as a correlation possibility with other deposits. ‘This study, therefore, is concerned with the comparison of the heavy mineral suites and the roundness and sphericity values of quartz at the seven stations studied in the Mason 3 81591. 0 Field Sampling. The seven sample stations along the Mason esker were chosen not at regularly spaced intervals but wherever commercial gravel companys had carried on operations. These gravel pits presented a good vertical face through the center of the esker and permitted an accurate sand sample to be obtained from top to bottom. Extreme caution had to be exercised in order to insure 6 an equal concentration of sand from each horizon. A channel- ing method was used whereby vertical channels were drawn downward and the sediment dislodged was c aught in a scoop. The sand from each channeling operation was collected in a bucket. After collecting a complete sample:from top to bottom.of the esker, a Jenes sample splitter was used. For the purpose of this study an 800 gram.samp1e was retained from.each station. Laboratory Sampling. ' Since these seven field samples consisted of unconsol- idated sand and gravel, no disaggregation problems were en- countered and the analysis was'begun as soon as a test sample of approximately hOO grams had been split from the field sample. The sand was poured into an ordinary quart Jar, filled with water and the mud removed by decantation. In view of the fact that this is a petrographic study of esker sands, the presence of mud would have been objection- able e Leaching. After the washed sand had been thoroughly dried, 200 grams of each sample was retained. A lOfl’HCl solution was added to each of these samples to remove all carbonates which also are an objectionable inclusion in a petrographic study. This process was repeated.until no more reaction was visible. In order to give some idea of the concentra- tion of carbonates in each of the seven samples, the resi- 7 due was carefully weighed and the results plotted in graph form.(Fig. 2). These percentages are of little importance however, insofar as the homogeneity or heterogeneity of the mineral suites are concerned, since the carbonates are largely secondary in origin. The graph simply shows that carbonates were present at all stations and in roughly the same concentration. Sieving. After leaching and drying, the residues were placed in a Ro-Tap sieving machine for a 15 minute period. Six sieves were used in the Ro-Tap: h8, 65, 100, 150, 200 meshes per inch. Each sieve size was weighed to the near- est milligram. A graph comparing weight percentages of the sieve sizes at all stations in shown in Fig. 3. The sieving was done primarily to separate the sample into convenient size units for further study, and not as an instrument of mechanical analysis. The theory of siev- ing involves a number of complexities which limit the accuracy of the operation. E. A. Mitschlerlichfi Mitschlerlich, E. A., Bodenkunde fur Land-und Forstwirte, Berlin, 1905, p 57. ' ‘ pointed out a number of years ago that sieving sorts grains not only according to size but also to shape. This might be illustrated by considering lath-shaped and spherical grains. The largest sphere that can pass through a given 8 mesh is a sphere with a diameter equal to that of the mesh, whereas a latheshaped grain with such a diameter and.52y length could pass through the sieve. This naturally would allow for a great differential in volumes of the grains. Therefore sievflng is not truly a size analysis. Despite the validity of the above criticiem, sieving is still a widely accepted practice and does have a purpose. The criticism is included prflmarily as a warning to those who read the graph not to interpret it with a view towards homogeneity or heterogeneity of the test stations. The Mason esker is crossbedded and stratified with coarse gravel lenses pinching in and out and highly suggestive of stream.deposits. A sample taken from one gravel pit which contains mostly fine sand would not be expected to have the same size frequency as the sample taken from the pit with many coarse gravel lenses. The sieving separated the samples into convenient size units for the most vital and important study of this problem; the heavy mineral frequency counts and the round- ness and sphericity measurements of quartz grains. Separation. The separation of the heavy minerals from the light was accomplished by means of bromoform, a gravity separa- tion, in which the heavy minerals sink and collect while the light minerals such as quartz and feldspar will float. One gram of each of the first three sieve sizes (20-h8, 9 h8-65, 65-100) and 500 milligrams of the last three sieve sizes (100-150, 150-200, 200 plus) for each of the stations in the esker were weighed. It was discovered by trial and error that this amount could be easily separated and still provide enough heavy minerals to study. In order to elim- inate as much error as possible, each sieve size was quart- ered by pouring the sample into a conical pile on a sheet of paper and separating the sand into four quarters by cutting it along two diameters. Alternate quarters were retained and combined and the process repeated until approx- imately one gram or 500 milligrams remained. Bromoform (tribrom-methane) was used as the separating agent. Bromoform has a specific gravity of 2.89 at 10 degrees Centigrade. This is sufficient to separate quartz and feldspar from.the heavier minerals. Commercial bromo- form has a low specific gravity because of dissolved alco- hol. The specific gravity was increased by adding water and decanting the alcohol-water phase. After decantation, the bromoform-water mixture was poured into a separating funnel, drawn off, and passed through several thicknesses of filter-paper which absorbed the dispersed water. The separation apparatus consisted of a filtering funnel with a short length of rubber tubing and a pinch- cock fitted to the stem. The funnel should be as broad as possible so that the floating minerals may spread over the heavy liquid. The grains may be agitated to overcome any tendency to "ball up". A second funnel with a filter paper 10 was placed below the first. A beaker under the second funnel caught the filtrate. The funnels should be con- structed to remove easily the heavy mdneral concentrate and they should not be so constricted that clogging occurs. A watch glass was placed over the funnel to reduce evapor- ation. After separation, the heavy minerals were allowed to flow into the lower funnel and were caught on filter paper. The heavy minerals were washed with alcohol to remove the bromoform. After the heavy minerals had been removed a new filter paper was inserted in the lower funnel. This paper caught the light minerals as the bromoform was allow- ed to flow through the funnel. The heavy and light minerals of the six sieve sizes for each station were separated simultaneously with a bat- tery of six separation units. The "lights" and "heavies" were each weighed and the "lights" put into vials and labeled. A small horseshoe magnet removed all the magnetite from the "heavies'. The magnetite was removed in order to facilitate the heavy min- eral frequency count. Mounting for Microscopic Study. A representative fraction of the minerals to be used for mounting must be chosen with care. Quartering was again necessary. The writer used the method suggested by ll PettiJohn*, *Pettijohn, F.J., Op. Cit., p 557. in which four rectangular sheets of paper, th.inches, were placed together in such a manner that each overlaps one half of the other, and altogether form a square. The , minerals, heavy or light, were poured into the center of the square. The pieces of paper were pulled apart, alter- nate quarters rejected, alternate quarters combined and the process repeated until a small sample was obtained. The samples obtained by quartering were mounted in a permanent mounting medium. The selection of this mount- ing medium was important since the refractive index of the medium.plays a.major role in the identification of heavy minerals, and aids in the measurement of roundness and sphericity of quartz grains. Most light minerals have low indices; most heavy min- erals have high indices. Therefore, a synthetic resin with a refractive index of 1.66 was used. This resin divides the range of indices of the heavy minerals, and thus facilitates identification. In mounting the light minerals (quartz, index 0-1.5hh, e41.553) this medium of index 1.66 markedly increases the relief of the grains. The details are accentuated for study and.measurement with the camera lucida. Since the heavy minerals of the first two sieve 12 sizes (20-h8, h8-65) were rock particles instead of min- erals, they were discarded. PettiJohn* *Pettijohn, F.J., Op. cit., p 519. points out that "it is known from observation that the heavy minerals of a sand are largely concentrated in its finer grades. This is due to the fact that for certain dominant size of quartz and feldspar there is a smaller size of magnetite and like heavy minerals, which are deposited together because they have what Schone called the same settling rates. Consequently, by screening out the coarse fractions, the heavy minerals will be materially concentrated.” Heavy minerals of the four sizes 65-100, 100-150, 150- 200, and 200 plus, were mounted on slides for each of the seven stations along the esker. A portion of the remain- ing unmounted heavy minerals were mounted by using the conventional Canada balsam (index 1.5h) as the mounting medium. The appearance of certain minerals was compared in the two different media. Only sieve size 20-h8 was discarded in mounting the light minerals. Identification of Heavy Minerals. The identification of heavy minerals was based upon the use of the polarizing or petrographic microscope. Min- erals may be distinguished in a number of ways, but for ‘ the non-Opaque mineral grains the optical method is the most acceptable. The polarizing microscope is a compound 15 microscope and differs from.the ordinary compound instru- ment since it is equipped with two prisms which produce plane polarized light. The manner of passage of the polarized light through the minerals serves as the ident- ification of that mineral. A mechanical stage was used to expedite the mineral count. This stage holds the mount to be studied and is so constructed that the mount can be moved in either of two straight paths at right angles in a horizontal plane. This eliminates the possibility of overlapping micro- scopic fields. The mounts of three different sieve sizes (65-100, 100-150, 150-200) for each station were counted and ident- ified. The ZOO-plus size was ignored since the grains ' were very small and difficult to identify. This meant counting the heavy minerals on 21 slides with an average of h00 grains per slide. Approximately 8&00 heavy min- eral grains were counted and identified in this study. Eighteen different mineral species were encountered; fourteen were constant through all seven stations and all sieve sizes. It is interesting to note that the most fre- quently occurring mineral is actually a composite aggregate or compound mineral. The grains are greenish white, semi- opaque with a "dirty" appearance. H. B. Milner* Milner, H.B., Sedimentary‘Petrography: D. Van Nostrand 00., New'York, 1929, p 89. 1h points out that "the petrography of sedimentary rocks is frequently complicated by an inherent difficulty of identifying certain rock- fragments, composite grains (compound minerals), iron- or carbon stained flakes, green or white semi-opaque grains of "dirty" appearance, and a number of other possibilities well appreciated by those who spend much time in searching mineral concentrates under the microscope." Careful investigation of the h8-65 sieve size re- vealed that larger fragments of the same composite grain were composed of quartz, iron ore, and chloritic matter. Following is the list of the minerals found in the Mason esker and the characteristics by which identifica- tion was made: Hornblende: Complex silicate of Fe, Mg, Ca, A1, and Na. Crystal system: Color : Index Birefringence Optic Figure Elongation Pleochroism : O. .0 .0 Monoclinic. Var Arfvedsonite - blue green. Var Common Hornblende - green to brown. 8», 1.658’10698; b, 1e670" 1.719; c, 1.679-1.722. e 026'e 027e Biaxial negative. Positivee Marked. Distinctive Features: Grains elongate; pris- matic; inclined extinction; marked pleochro- ism; common. Clinopyroxene: (Augite, Diallage, Diopside) Crystal systemc: Color 3 Index : Birefringence : Optic Figure : Extinction : Monoclinic. Brownish gray to gray-green. a, 1.696-1.700; b 1.702- 718- c 1.71h91.752. O -eoa3e Biaxial positive. hhth degrees. 15 Distinctive Features: Grains usually elongate worn cleavage fragments; poorly rounded; high index; high birefringence; large extinction angle. Garnet: R"R"' (SiOh) where R" is Mg,Fe",Ca,Mn. . R'H is A1,Fe"',Cr. Crystal system: Isometric.-s Color : Pink and colorless. Index : 1e70-1e90e Distinctive Features: High relief; iso- tropism; conchoidal fracture. Chloritic Matter: Essentially silicates of A1,Fe, . Mg, and Hydroxyl. Crystal system: Monoclinic. Color : Dirty green. Birefringence : .005-. 009. Distinctive Features: “Ultra-blue" abnormal interference color; compound polarization; pale green color; low birefringence. Zircon: ZrSiO . rystal s§stem: Tetragonal. Color : Colorless. Index : e-l.285-1.99l; o-1.926- 1. . Birefringence : .O . Optic Figure : Uniaxial positive. Elongation : Positive. Extinction : Parallel. Distinctive Features: Euhedra common; pyram- idal terminations; basal grains rare; rod shaped inclusions; high index. Monazite: (Ce, La, Nd, ,Pr)§O g.P2 20 0 *Crystal system: lini Color : ‘Yellow. Index : a, l. 786- 1.800; 1.788- 1.801;O c, 1.857- 1. 8A9. Birefringence : .0h9-.O O. Optic Figure Biaxial positive. Pleochroism : Faint, X-light yellow, Y- dark yellow, Z-greenish yellow. Extinction : 2- 10 degrees. Cleavage : Perfect basal. Distinctive Features: Grains rounded, equi- dimensional often lying on 001; euhedra rare; exhibit same color between crossed nicols as in ordinary light owing to high birefringence; high relief; light yellow color. 16 E idote: Ca2 (A1,Fe) 31 O OH). 'IL"TF' 5 M 3 1 ( rystal2 system: 0 oc inic Color : Bottle een. Index : a, 1.72 -1. 729; b,1.7hZ- 1.765; c, 1750 -l. 780. Birefringence : .028-.051. Optic Figure : Biaxial negative. Pleochroism : Distinct. X-colorless, Y- bottle green, Z-colorless. Extinction : 2-5 degrees. Cleavage : Perfect basal. Distinctive Features: Grains equidimensional, subrounded; distinct pleochroism; bottle green color; high index. rystal system: Monoclinic. Color : Pale yellow, light brown. Index : a, 1.900; b, 1.007; . c,.2.05h. Birefringence : . . Optic Figure : Biaxial positive. Elongation : Negative Pleochroism. : Weak. Extinction : 51 degrees. Distinctive Features: Conchoidal fracture; diamond shaped euhedral grains; exhibit same color under crossed nicols as in ordinary light owing to high birefringence; many grains fail to show complete extinction in white light due to high dispersion. The grain turns bluish as the extinction position is reached. Hypersthene: (Mg,Fe)SiO ‘Crystal system: Oréhorhombic. Color : Pale pink and green. Index : a, 1. 665-1. 715; b, 1.669- 1.728, c, 1.67h-1. 731. Birefringence : .OO9-.016. Optic Figure : Biaxial negative. Elongation : Positive. Pleochroism : Marked. X-pink,‘Y-yellow, Z-green. Extinction : Parallel. Distinctive Features: Worn elongate cleav- age fragments. Highly colored, thin, brown, plate-like inclusions (schiller structure); low birefringence; parallel extinction; striking pleochroism. Orthopygoxene: (Enstatite,Bronzite) (Mg,Fe)8105. 17 Description same as hypersthene except enstatite is nearly colorless and is bi- axial positive. Enstatite, bronzite,and hypersthene are members of an isomorphous series in the orthorhombic pyroxene group. All Others: Rarely grains of biotite, tremolite, zoisite, leucoxene, tourmaline, rutilc, and staurolite are encountered. Since collectively they make up less than 1% of each heavy mineral suite, their description is omitted. The results of this heavy mineral frequency count are presented in tabular form on pages 25, 26, and 27. It is immediately apparent that the mineral suites and frequencies are similar for all seven stations. This establishes a definite homogeneity for the seven min- eral suites.' Roundness and Sphericity Measurement of Quartz Grains. The quartz grains were studied in an effort to deter- mine if a relationship could be established between the quartz grains at the seven different stations of the es- ker. In other words, how does the roundness and spher- icity compare at the various stations? Hakon Wadella *Wadell, Hakon, Volume, Shape and Roundness of Quartz Particles: Journal of Geology, vol A5, pp 250-280, 135. was one of the first to differentiate between sphericity and roundness and to show that these are independent var- iables. Wadell pointed out that roundness is concerned 18 with the sharpness of the corners of a grain whereas sphericity is a ratio between the length and breadth of a grain. Wadell devised a method for measuring roundness and sphericity but it is very time-consuming. Geology stu- dents at Michigan State College combined the ideas of Wadell and N. Allen Rileyt. *Riley, N. Allen, Projection Sphericity: Journal of Sedimentary Petrology, vol 11, pp 9h-97, 19h1. Riley devised a method by which sphericity can be meas- ured rapidly with accuracy. Mr. George T. Schmitt, grad- uate student at Michigan State College, devoted a great deal of time and.effort to this problem and is respons- ible for the development of the method used in this paper. A Mr. Schmitt drew a concentric circle protractor similar to the one used by Wadell, on white paper rather than celluloid” By means of a camera lucida, the images of the grains were projected to the concentric circle pro- tractor. The diameters of the inscribed and circumscrib- ed circles (sphericity) of each quartz grain, and the arc of the corners of each grain (roundness) could be obtained simultaneously. The writer used the method described above with Wadell's formula for roundness: (Egg; : P (r/H) is the sum of the roundness values of the corners. 19 N is the number of corners in the given plane. P is the total degree of roundness. and Riley's formula for sphericity: ¢ : 1 \\5c 1 is the diameter of the inscribed circle. Dc is the diameter of the circumscribed circle. Two sieve sizes (h8-65, 65-100) were measured. Some authors believe that finer particles would be subjected to an excessive amount of fracture and therefore would not give satisfactory roundness and sphericity values. At least fifty quartz grains per slide, with an average of ten corners per grain, were measured. Fourteen slides with a total of 7000 measurements was sufficient to give a fair roundness and sphericity value. The comparison of the values obtained for the two sieve sizes at the seven stations is shown by Figs 5 and 6. It is immediately apparent that there is a close correlation between the roundness and sphericity values for all samples. It is interesting to note that the roundness increases from north to south. It is an accepted fgct that the subglacial streams in this area flowed southward. It has also been generally assumed that sand grains become progressively more round as they are transported. Theoretically, the direction from which a sediment came could‘be determined if a progressive in- crease in roundness were detectable. The roundness fig- 20 ures obtained for the Mason esker support the theory that the subglacial stream.must have flown southward. CONCLUSIONS Earlier in this paper the theories that have been advanced by well-known geologists for the origin of eskers were summarized. It was also pointed out that probably all of these theories were necessary for a complete ex- planation of the origin of eskers. This paper is con- cerned with the Mason esker and the application of a de- tailed petrographic study of the esker sands as an aid in the determination of the origin of that particular esker. We are now ready to compare the results of this petrdgraphic study with each theory and attempt to deter- mine if theIPesults favor any one theory for the origin of the Mason esker. Inspection of the graphs (Figs. 7,8,9,10,1l, and 12) shows immediately that the heavy mineral suites are ident- ical and that the mineral frequencies are very similar for all seven stations in the esker. This strongly suggests that the Mason esker is a continuous, homogeneous deposit and that deposition was simultaneous along its length. These results lend support to Davis's* *Davis, W.M., 0p cit. subglacial channel theory whereby the sandsczarried by the 21 subglacial stream were deposited in a continuous narrow channel. It would be expected that the composition of the deposit at the beginning of the channel should be very nearly the same as at the end. These results do not exclude the possibility of en- glacial or superglacial origin. However, Trowbridge's* *Trowbridge, A.D., Op. cit. theory is difficult to explain in the light of the re- sults obtained in this study. Such a theory proposes head- ward building of the esker by successive kames. This dumping action at the margins of the retreating ice sheet would result in a more heterogeneous mineral suite. If a time factor were involved this would be especially true. It is not known how long it took the glacier to retreat the length of the Mason esker or if it were an uninter- rupted retreat. It seems possible that shortly after the ice front had advanced into a new lithologic province, climatic conditions might change causing the glacier to retreat. The detrital material carried within the ice would be deposited. The mineral suite and frequency found in the first kame deposited during the retreat would be governed to some extent by the lithologic prov- ince in which the kame was deposited. This would obvious- ly cause a variation in mineral suites in a series of successive kames as Trowbridge postulates. 22 Perhaps an even stronger argument against the "drawn out kame" theory is presented if we assume that the foremost part of an advancing glacier will pick up the bulk of its detrital material from the earth's sur- face. As the ice advances, scouring and gouging deeper into the earth's crust, new material is incorporated into the ice. Widely varying strata might be encountered as the ice cut downwards If the strata were the same for great depths, mineral suites would still differ since the uppermost rocks may consist of alteration products. For example, hornblende may alter to chlorite, biotite, mag- netite; olivine may alter to antigorite, magnetite or some pyroxene; pyroxene may alter to the secondary horn- blende, uralite. If the ice were eroding an igneous structure, differ- ent minerals might be encountered as the ice approached the center of the structure. This may be explained by the order of crystallization around the periphery of an igneous mass, the more basic minerals crystallizing near the edge. With a retreat of the glacier, the ice front would have a heavy concentration of the minerals first encountered by the gouging, scouring, advancing ice. The first kame de- posited would contain these minerals and with headward building of successive kames, the mineral suites and fre- quencies should vary. However, under the same conditions (new lithologic provinces, varying strata, or igneous structures), the 23 deposits of a subglacial stream would still be homogeneous since this is an aggrading stream. Such a stream does not scour and erode the surface over which it flows. The sediment which is deposited throughout the length of the stream channel has been well sorted. Any new source mat- erial would be distributed by the stream along the length of the channel since it is a simultaneous deposition. It is possible that the mineral suites might vary in a sub- glacial stream but the most important fact in this study is that the mineral suites do 223 vary. Therefore, it is difficult to accept any explanation for the origin of the Mason esker other than Davis's subglacial stream chan- nel theory. . The results of the sphericity and roundness measure- ments of quartz grains give further support to Davis's subglacial theory. Figs. 5 and 6 show that the roundness and sphericity values are similar at all stations. Round- ness increases slightly to the south, the direction of flow of the subglacial stream. This slight change in roundness suggests progressive abrasion of the grains as they are rolled southward. The sphericity is also nearly identical for all sta- tions, but decreases southuard. Stream abrasion may in: crease roundness but decrease the sphericity of grains. Quartz has no cleavage. The rolling and abrasion causes 2).; quartz grains to become more elliptical thus decreasing the sphericity. The retreating kame hypothesis again comes under criticism since the roundness and sphericity of the indiv- idual particles in each kame probably would not be ident- ical. In conclusion, this petrographic study clearly indi- cates that the Mason esker is a subglacial stream channel deposit. The results prove conclusively the advantage of a statistical approach as an aid in determining the origin of glacial land forms. 25 FREQUHCY PERCENTAGE OF HEAVY MINERALS 65-100 Sieve Range Statlon!F 1 1A 2 _*5D A. 5 6 Compound Aggregate 36.h 39.6 57.3 38.0 35.8 55.9 57.5 Hornblende (arf.) 18.5 17.2 17.1 20.0. 19.5 18.5 20.6 Chloritic Matter 11. 5 11.9 12.9 15. 9 10.2 10.11 8. 5 Garnet (Pink) 9.8 K'6.5 8.7 6.7 9.h 6.1 6.7 Garnet (Colorless) 7.8 5.0 7.0 2.6 8.0 9.5 8.5 Hornblende (Common) 6.0 h.6 u.5 6.u. 6.6 8.A h.9 Clinopyroxene 5.0 7.8 6.6 7.5 5.2 'h.h 6.7 Eplootc 5.2 h.6 '2.1 0.9 5.7 1.7 1.28 Hypersthene 1.6 1.5 2.8 0.9 0.9 1.0 5.0 Titantte 0.h. Tr o.h 0.9 0.9 1.0 0.6 All Others 0.0 0.8 0.11 08:41.0 1.8 1.7 TABLE 1 26 FREQUENCY PERCENTAGE OF HEAVY MINERALS 100-150 Sieve Range Station 1 1A 2 5 A 5, 6 compound Aggregate 50.h. 27.2 50.1 2h.7 29.8 25.9 26.2 Hornblende (Arf) 15.0 19.0 19.2 15.7 15.0 18.9 17.5 Hornblende (Common) 12.0 9.2 16.0 16.8 16.0 15.9 8.8 Clinopyroxene 10.0 9.7 6.6 8.6 7.0 7.h 8.2 Garnet (Colorless) 8.7 8.1 7.6 7.8 10.0 10.9 8.6 Garnet (Pink) 5.5 h.7 h.7 6.8 6.5 6.9 6.5 Chloritic Matter 8.1 9.7. 7.5 5.8 8.2 5.7 7.2 Epidote 5.9 h.5 2.5 h~h 0.7 5.6 5.8 0;thopyroxene 2.6 0.8 Tr 2.1 0.5 Tr 1.7 Hypersthene 1.6 2.0 2.5 71.8 1.9 1.8 1.h Titanite 0.6 0.7 Tr 1.6 Tr 1.8 1.h zircon . Tr 2.2 1.1 1.5 1.5 2.2 1.7 lbnazite 1.6 . 1.2 1.8 1.5 2.6 1.0 2.1 All Others 0.0 1.0 0.8 1.0 0.7 0.0 2.1 TABLE 2. 2? PREQDENOT'PEEOENTAGE 0F HEAVY MINERALS 150-200 Sieve Range Station 1 11 2 5 A. 5 6 Compound Aggregate 22.8 26.h. 27.6 25.5 2h.9 27.0 2h.5 Hornblende (Arf) 20.5 20.0 22.8 20.8 19.8 21.9 19.5 Hornblende (Common) 11.2 10.5 10.7 9.5 10.6 10.5 9.7 Clinopyroxene 7.0 7.5 7.7 7.6 7.7 7.9 8.1 Garnet (Colorless) 11.7 10.7 11.2 11.8 10.8 8.5 10.7 Garnet (Pink) 1.6 11.6 1.9 11.7 5.2 2.0 5.7 Chloritic Matter , 5.9 5.1 5.8 5.h, 5.0 7.2 h.0 zircon ~ 5.9 h.0 1.9 5.5 1.5 h.6 5.h Menazite 5.1 1.8 1.1 2.h. 1.1 5.9 5.0 Epidote 5.5 5.7 2.7 5.0 5.2 5.5 2.7 Titanite 5.1 1.2 1.h 1.1 2.0 5.5 2.0 Hypersthene 2.5 h.0 2.7 2.2 1.8 Tr 5.h Orthopyroxene 1.6 1.2 Tr 1.6 Tr Tr 1.0 All Others 0.0 1.5 2.5 1.5 1.6 0.0 0.7 TABLE 5 FIG. OOZD>m_mOz Om. 41m mmmocmzo< Emifimcjoz Om 41m Im><< §_me>_um ..a<. v|. I.) ‘0 (I ——---Isllu-u - EEO? )1 —i_v-- U I i ' IIPII) .191 _—-F_hbhbbtb 67" l l _—h-!Pblwlul J u D I" ‘ - "Nf‘."\ V y/////////////////i///M////,//é - 7//////////////////,//4 wwwwwwwwww FIG3 WEIGHT PERCENT ANALYSIS OF SAND 7 \\ SEVE RANGE SEVE RANGE mEVE RANGE SEVE RANGE SIEVE RANGE m\\ \\\\\\\\\\\\\ SEVE RANGE 7/////////////////////////% 3 mm //////////////////// A A A 7 A 7 A A, % A A wwwww m: % A A A A A FREQUENCY PERCENTAGE OF HEAVY MINERALS 65")00 SIEVE RANGE STA | STA. IA O OMOSITE AGGREGATE ' '1 ’ ' NORNGLENDEIARE) / CHLORITIG MATTER GARNET (PINK) WTE AMBATE HORNBLENOE (ARE) CHLORITIG MATTER GARNET (PINK) GARNETKBOLMLESS) GARNET (WMESS) MRNKENDE WON) HOR'BLEI’OE (mm CLINOPYROXENE CLINOPYROXENE EPIOOTE EPIDOTE NYPERSTHENE HYPERSTHENE TITANITE TITANITE ALL OTHERS ALL OTHERS . . ?‘ .....—.-. -. ‘.-..~..w-’—-e .. - . o . .- 9'. e ’7Yf.‘ 0| '3 i n?»- .1... STA. 2 STA. 3 O COMPOSITE AGGREGATE ’ MTE AGGREGATE ) HORNBLENDE (ARE) :4 , ’ A HORNBLENDE (ARE) :- CHLORITIC MATTER CHLORITIG MATTER . ;; GARNET (PINK) GARNET (PINK) - :j: GARNET (COLORLESS) GARNETIGOLORLESS) 'I: HORNBLENDE (COMMON) : ' : , NORNsLENOEIOOMMON) , : CLINOPYROXENE = CLINOPYROXENE - '; EPIDOTE EPIDOTE L. T HYPERSTHENE HYPERSTHENE TITANITE TITANITE ALL OTHERS ALL OTHERS - 1 STA. 4 STA 5 O °///////.////7/////////,°'////7// o .t/// /,1 ///. //7 WITE AGGREGATE HORNBLENDE (ARE) CHLORITIC MATTER GARNET (PINK) GARNET (COLORLESS) 7////// 7;} MOSHE AGGREGATE NORNGLENDE (ARE) GRLORITIc MATTER GARNET (PINK) GARET (COLORLESS) HORNBLENDE (COWON) “ORNBLENDE WON) CLINOPYROXENE ’ CLINOPYROXE NE EPIDOTE EPIDOTE HYPERST HENE HYPERSTHENE . TITANITE TITANITE ‘ “L‘- OTHERS ALL OTHERS v D . no! I - '9-07. III-«v.9 .wvo'Oi ' ‘I . I iIm-y, ._ STA. 6 o 5 u OOMPOSITE AGGREGATE 7/27//////<27////;{//;¢’/////////: HORNBLENOE (ARE) WAVZAZ/AV/fl/fl/A' CHLORITIC MATTER ”6'? GARNET (PINK) GARNET (COLORLESS) .. . HORNGLENDHOOMMON) CLINOPYROXENE / ' EPIDOTE HYPERSTHENE TITANITE ALL OTHERS . «1,. /- ,, //,1 1’1 FIG. 8 FREQUENCY PERCENTAGE OF HEAVY MINERALS IOO'ISO SIEVE RANGE STA l HORNBLENDE (ARFI V/A/fA////7/// HORNBLE NOE (COMMON) '/// ’L-AA'A/Aé CLINOPYROXENE ////////A': GARNET (COLORLESS) ////////A GARNET (PINK) / CHLORITIC MATTER EPIOOTE ORTHOPYROXENE HYPERSTHENE TITANITE 2IRCON MONAZITE ALL OTHERS STA 2 O I COMPOSITE AGGREGATE ///A/ /’/ ///A c)’/////////A/////////A HORNaLE NOE (ARE) ’A/////////////////A HORNRLEmEmOMMON) /////////A////// CLINOPYROXENE GARNE T (COLORLESS) //////// GARNET (PINK) CHLORITIC MATTER ' EPIOOTE ORTHOPYROXENE HYPERSTHENE TITANITE ZIRCON MONAZITE ALL OTHERS STA 4 O COMPOSITE AGGREGATE ///// /////// /////////A/////////A HORNBLENDE (ARF) HORNBLENDE (COMMON)/ CLINOPYROXENE GARNET(PMK) CHLOanO MATTER EMOOTE ORTHOPYROXENE HYPERSTHENE TITANITE vRCON MONAZITE ALL OTHERS STA. 6 CWPOSITE AGGREGATE HORNBLENOE (ARE) HORNBLENOE COMMO CLINOPYROXENE GARNET (PINK) CHLORITIC MATTER ; EPIOOTE ORTHOPYROXENE HYPERST HENE TITANITE ZIRCON . MONAZITE ALL OTHERS STA IA COMPOSITE AGGREGATE ’//////////// //////A//////A HORNBLE NOE (ARF) ’///////////////////, HORNaLENOE (COMMON ///A///// CLINOPYROXENE / G‘ARNET (COLORLES GARNET (PINK) ’ EPIOOTE ORTHOPYROXENE HYPERSTKNE TITANITE ZIRCON MONAZITE ALL OTHERS STA. 3 COMPOSRE:AGGREGATE.AAdéééQAAAAé/xszA”’ HORNBLENDE(ARF) AAAAV/QAAW HORNBLENOEICOMMo y /AA4¢7CA6077A7 CUNOPYROXENE AA GARNET(COLORLESS)4¢AA¢7 GARNET (PINK) /////// CHLORHIC MATTER 077/ EMOOTE ORTHOPYROXENE HYPERSTHENE TITANITE vRCON MONAZITE ALL OTHERS STA. 5 O O COMPOSITE AGGREGATE '/////////A/////////ACIA/:44 HORNBLENDE (ARE) ’/A/////AA/// //////A HORNaLEN OE (COMM /////////A//A CLINOPYROXENE GARNET (COLORLEx/ GARNET (PINK) EPIOOTE ORTHOPYROXENE HYPERSTHENE TITANITE ZIRCON MONAZITE ALL OTHERS ’///////;//o//////;////////A FIG. 9 FREQUENCY PERCENTAGE OF HEAVY MINERALS ISO'ZOO SIEVE RANGE STA I (I 5 7///////// /////////// ”A //////// //////A/////// // /// // ’//// CWPOSITE AGGREGATE HORNBLENDE (ARF) HORNBLENDE (CWMCIN) CLINOPYROXENE GARNE T COLORLESSI GARNET (PINK) CHLORITIC MATTER ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE ORTHOPYROXENE ALL OTHERS STA 2 o C COMPOSITE AGGREGATE 7////////A/////////A’//////A HOR N BL E NOE (AR F) 7//////////////////A’/A HORNBLE NOE (COMMON) 7//////AA CLINOPYROXENE ’ GARNETICOLORLESS) ’ GARNET (PINK) ’ CHLORITIC MATTER ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE ORTHOPYROXENE ALL OTHERS STA 4 COMPOSIT E AGGREGATE C’////////////////////////A HORNBLE NOE (ARE) /’ , HORNeL ENOE (COMMON/ ////// CLINOPYROXENE ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE OR THOPY ROXENE ALL OTHERS STA. 6 COMPOSITE AGGREGATE ’////////;//;///////////A HORNBLE NDE (ARF) HORNBLENDE (COMM 0 CLINOPYROXENE GARNE T (COLORLESS) : GARNET (PINK) CHLORITIC MATTER ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE ORTHOPYROXENE ALL OTHERS STA, IA COMPOSITE AGGREGATE .1 "3 1 HORNaLENOE (ARE) ~ ' ,, HORIBLENOE (COMMON) } g. , GLINOPYROXENE ' '- ’ GARNET(COL0RLESS) / GARNET (PINK) A. CHLORITIC MATTER ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE ORTHOPYROXENE ALL OTHERS STA. 3 COMPOSITE AGGREGATE HORNBLE NOE (ARE) HORNRLENOE (COMMON) 17” CLINOPYROXENE ' GARNET (COLORLESS) ‘ ’ GARNET (PINK) CHLORITIC MATTER ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE ORTHOPYROXENE ALL OTHERS STA 5 O COMPOSITE AGGREGATE ///////,/// HORraLENOE (ARE) A’ HORNaLENOE (COMM ' ,1 CLINOPYROXENE ' GARNET (COLORLESS) GARNET (PINK) / CHLORITIC MATTER ZIRCON MONAZITE EPIOOTE TITANITE HYPERSTHENE ORTHOPYROXENE ALL OTHERS I 3 OWING SIMILARIT)! OE THESEVEN CUMUL I I ATIVE -CURVES I SH GR IAPH 5 7 #Ar — —~9- ~77, ENTING THE I | I 1 I I r--—O __.,._ —EAT —-—— 0————..— - I | I ___f-— HEAVY MINERAL- __._. _ ,,‘_ ,,_ ___ REPRES . EREOUENCY I I I I I .wITHIN- THLMASON- ESK ER — —+———— . I I I SIEVE RANGE I65-IOQ I I IIIIIIIITII) mmero 44¢ uzmxom>aozpmo _ u+_z<»rr uzuxpazma>1 ,wk-oo_am. mauxomyaoz_do II III -I- IIIII- I + -IIII 202208 mozmamzaox. AmmuJ;OSOo.Fuz¢4m 21!:IILII::.I., -.I1- §i_A. +uz¢<.u h amph4: ortmoqxu (III) A III. . )IT I.IIIIII r2353 u A 1m 0 E :04 -I-I III-fi-I- .Lmaozkmo . H .M S w . . . . T A. whooim Y I F V M E «3.2: 0:20;: 6 I mm H M 3.2:: Amzx.aoz_.3. l _..H S .zozsoovmozmizmo: _ mug N If: 331353: T- G N 33233 utmoazoo r N .E . .I .. - I. I - . I I S . W E mu... W Max:312 . S .E I I - 1 III. R A. H _ III .W- I- I I . R G GRA‘PIII SHOWING SIMILARITY 0E THE-SEVEN CUMULATIVE CURVES I w-.. L‘_. .— JREPRESENTI’NG‘THE' HEAVY MINERAL FREQUENCY ’ WITHIN. THE. MASON ESKER _ ISO’BOO mEVEIRANGE . mzwxpo 44¢ uzmxom>uotho .wzuthaua>x mp_z<»rr 309$ wp_~