The writer has devoted a number of years to practical operations and to the study of geology in the oil fields. In consequence, he has been brought to investigate the theories advanced to account for the accumulation of oil and gas in commercial deposits. It is the result of these investigations and his personal conclusions that he wishes to sum up in this paper.

The writer is an advocate of the organic origin of petroleum found in pools. He has been led to believe that the present theories of oil and gas accumulations are incomplete and, in certain respects, incorrect, although they may embody certain elements of truth; that the forces that are called upon to explain the movement are only secondary forces in the process, and insufficient, by themselves, to cause this movement; and that the phenomenon of accumulation is of much larger order than heretofore admitted and bears an intimate relation with the general as well as with the local deformations of the crust and is a direct consequence and a mechanical effect of these deformations to which the term “diastrophic” has been applied. As a complement to this theory, the writer offers a new interpretation of the “rock pressure” and of the mechanism of the “sealing up” of this pressure in oil and gas pools.

The Anticlinal Theory

The theory most generally accepted to account for the accumulation of hydrocarbons in commercial deposits is the anticlinal or structural theory. This theory, the principles of which have been laid down by I. C. White, Orton, Winchell and others, and which has since been modified and perfected, is too well known to need a detailed statement. A clear interpretation of it may be found, in the paper of W. T. Griswold, in Bulletin No. 318 of the United States Geological Survey, and the reader is referred to it. In this theory, the force which is supposed to have caused the motion is the gravity of the hydrocarbons; and the principal factors which have intervened in the process are the structure and the general stratigraphic conditions of the rocks, their texture and porosity, and their water content. The oil and the gas are supposed to enter the porous rock that is to act as a reservoir, by some indefinite process. Any further movement toward accumulation is only considered as possible if a dip exists in this rock sufficient to overcome the friction, and practically impervious areas of the rock will be more or less perfect barriers against the movement. Then, if this rock is completely saturated with water, or if the hydrocarbons have entered it below the line of complete saturation, the oil and the gas will gradually move up the slope by the effect of buoyancy; the gas, with its lower specific gravity, occupying the higher places. Should the rock be dry, or if the hydrocarbons have entered the rock above the line of saturation, the oil will flow down, as long as gravity is sufficient to overcome the resistance to motion, and the gas will diffuse with the air or water vapor contained in the pores of the rock.

From this statement it will appear that the following assumptions are an absolute requisite for the anticlinal theory:

First, the structural deformations (dips, anticlines and synclines, domes, terraces, etc.) must be supposed to have existed previous to the introduction of the hydrocarbons in the porous strata; in other words, it must be assumed that structural deformation has preceded the movement toward accumulation. Indeed, in this theory, the preexistence of structural deformation is the very basis of the action of gravity and, at the same time, the theory implies that the forces which have caused the deformations have had no bearing on the movement of the hydrocarbons.

Second, gravity or buoyancy is to be considered the sole agency through which accumulation has been brought about and, as such, is supposed to be adequate to explain accumulation under any condition of dip.

To which the following considerations may be added: The movement toward accumulation would have to take place in the porous rocks when they are already solidified and partially cemented. There would be no connection between the causes which have led: To the introduction of the oil in the porous strata; to the movement of the fluids in these same strata; and to the “rock pressure” itself. Further, this theory does not provide for any satisfactory explanation as to the mechanism of the “sealing up” of the pressure in the pools.

The first assumption is begging the question. To the best knowledge of the writer, no attempt has been made by the authors of the theory even to discuss the point or to prove the accuracy of the inference. It seems as if, from the start it was admitted as self-evident that the gravity of the fluids was the only possible force entering into action. The writer will try to show later on why, in his opinion, this assumption falls short of truth and how its admittance may be explained by the fact that the problem has not been attacked from its true angle. On this first point, to say the least, the anticlinal theory is not established on proven ground.

The second assumption may be seriously contested. The force of motion due to buoyancy is a function of the sine of the dip. It would be maximum for a vertical stratum and null for a horizontal one. Many anticlines in the Eastern fields of the United States have a dip so low that a motion caused by buoyancy alone could hardly be understood. Dr. Ashburner has found a maximum dip in the Bradford region of 69 ft. per mile, or less than 1 ft. for 76 ft., while Carll has shown that the dip of the oil sands in the Venango belt, and the southern end of the Butler belt, rarely exceeds 34 ft. per mile (Sir Bowerton Redwood), or less than 1 ft. for 155 ft. In West Virginia, the maximum dip of many anticlines is less than 1 ft. for 50 ft., which corresponds to an angle of about 1° 8′ 45″.

For this last dip, the force due to buoyancy, which is supposed to move the oil up the slope, would be reduced to 0.0066 of the weight of the oil, as pointed out by Malcolm J. Munn, of the U. S. Geological Survey. A volume of 1 c.c. of oil would thus be submitted to a moving force of a few milligrams. The oil could be set in motion only if this force were greater than the resistance due to static friction, viscosity and surface tension. It is not possible, with the data now at hand, to submit the question to a complete mathematical treatment. As Van Hise points out, ‘‘the pore openings in sandstones are for the most part capillary”, and the flow of oil through the rock would have to obey the laws of flow through capillary tubes, which are not yet satisfactorily expressed. Further, the pore openings would be extremely irregular in shape and dimensions. But it is possible to reduce the problem to a simpler form. If, barring viscosity and surface tension, we would only consider the resistance due to static friction, we would obtain an inferior limit to the resistance. And if it were possible to show that the force of motion due to buoyancy is insufficient to overcome such a frictional resistance, in the case under consideration, it would be proof that the anticlinal theory can not be accepted, at least in its present form, and that other forces besides buoyancy are to be considered in order to explain the motion.

Let us imagine a globule of oil appressed against the roof of the sandstone stratum by the effect of buoyancy (Fig. 1). Let F be the force of buoyancy. This force may be decomposed into two components: F’ parallel to the dip, which would cause the globule to slide along the roof, and F” normal to the roof, which would appress the drop of oil against it.

The effect of this pressure is to create a resistance to the sliding, which is proportional to the pressure and directed down dip. This proportion, which remains constant when the nature of the surface remains the same, is the friction coefficient. Let it be c. The movement will become possible only if c < F’/F” or c < tang α, α being the angle of the dip. If the dip is 1 ft. to 50 ft., the tangent is about 0.02 and the condition reads c < 0.02.

There are no data available for the friction coefficient of oil and sandstone under the conditions stated. But one may get an idea of the order of magnitude of this coefficient by a simple experiment. Let us take a slab of some oil-bearing sandstone, some Berea grit, for instance. The lower surface of this slab must be planed, but not ground, so that the grains of sand remain entire. Let us immerse the slab in some salt water of proper density, in a horizontal position, and loosen a drop of oil from a syringe in the water, under the slab. The drop will rise to the bottom of it, and here it will remain if the slab is perfectly horizontal. Let us then incline the slab progressively and note the smallest angle of dip at which the drop will be set in motion. The tangent of this angle will be equal to the coefficient of static friction.

The writer has made this experiment repeatedly and has never succeeded in getting the drop in motion for any such angle as 2°, which would tend to prove that the coefficient is superior to the maximum limit required.

It is further to be noted that the spherical form is the general form of oil drops in an emulsion, where the particles of oil are, in great majority, very small; it is also the form which, every other thing being equal, offers the least resistance to motion. Hence, if the motion is shown to be impossible under this form, it will be equally impossible under any other form, and the whole theory falls through.

The change from a spherical form to an irregular one, by the flattening of the drops when compressed against the walls or when passing through irregular and narrow channels, the indentations, the penetration of the oil between the grains, the effect of viscosity, the capillary action—especially if gases be held in the emulsion—would be as many causes furthering an increased resistance to motion.

Now, the fact that the movement toward accumulation is not possible under the sole influence of gravity, with the low dips so characteristic of many oil regions, does not mean that gravity has had no influence in the process. It simply means that this force alone would be inadequate to explain the motion and that some other force or forces must have entered into play. But though gravity of hydrocarbons may not be a primary agent of their accumulation, it is a force which must be taken into account in any general theory of the process.

The Hydraulic Theory

In opposition to the anticlinal theory of oil and gas accumulation through gravity, Malcolm J. Munn has propounded another theory in which the moving force would be the hydraulic (not hydrostatic) pressure and capillarity of underground waters. This is termed the hydraulic theory.

“The fundamental idea of the hydraulic theory,” writes Mr. Munn, “is that moving water under either hydraulic or capillary pressure has been the direct agent of accumulation of oil and gas pools. To this idea may be added another of equal value—the pools of oil and gas are held in place by water under hydraulic and capillary pressure which effectively seals up all the pores of the surrounding rock and prevents the dissipation of pressure by diffusion.” This theory is very interesting and, as will be seen later on, the writer agrees with Mr. Munn on two points, viz., that water under hydraulic pressure has really been the primary agent of motion, and that the “sealing up” of the pools is a phenomenon of surface tension. But, as far as the writer can see, Mr. Munn has reduced the hydraulic pressure to that of underground waters circulating in the same way as they do now, and his statement relating to “hydraulic and capillary pressure” is rather undetermined.

“Capillary pressure” is a somewhat misleading term. If by this is meant the general action of forces due to surface tension, it is well to note that surface tension may create a resistance against the flow as well as a tendency toward it, as Mr. Munn himself seems to acknowledge when he comes to the “sealing up” of the pools. If it simply means the process of infiltration, the creeping of a liquid through capillary channels, this process has its limitations and would seem to be a process of dissemination rather than of concentration. On the other hand, hydraulic pressure of underground waters may be inadequate. Hydraulic pressure is a function of the square of velocity, and the velocity of underground waters is known to be extremely small. “ The motion of the ground water as a whole”’ writes C. S. Slichter, “is somewhat like the slow motion of very viscous sirup or the slowly creeping ice of a glacier.” It has been a shown, for instance, that the ground waters of the Arkansas River flow in gravels at a rate not greater than 3 to 5 m. a day. This would mean a velocity between 0.00361 and 0.0058 cm. per second, or less than from 13 to 21 cm. per hour. It is possible to show that, should the static coefficient of friction between oil and sandstone be greater than 0.02, as heretofore admitted, a velocity of 0.006 cm. sec. would not overcome the frictional resistance of oil in an inclosed pipe, on a horizontal plane, barring the additional resistances due to viscosity, surface tension, deformation, etc., which would interfere in this case. Though the analogy may not be entirely correct, it has a tendency to show that the influence of hydraulic pressure of underground waters is here doubtful and at least remote in importance. There will be always a great difference between the transportation, of a material in the state of solution, which may be followed by subsequent re-deposition, and the transportation of insoluble solid material, as oil in water. Further the theory is open to the same general line of objections as the anticlinal theory, and more so, as structural deformation, which is so intimately connected with the phenomenon of accumulation, becomes here entirely subordinate and incidental.

The Diastrophic Theory

The fundamental weakness of the theories which we have reviewed may perhaps be traced to the hesitation of the authors of these theories to choose deliberately between the different conceptions of the origin of oil itself. This origin once ascertained, it is evident that the surest way to determine how oil and gas may have accumulated in certain places, would be to try to follow the possible movements of the oil from the time of its first appearance in the strata, down to the pools where we find oil today. In order to be rational, a history ought to be complete; and we have no more right to limit the history of petroleum to the space of time during which the strata, in which we find it now, may have existed in the same present shape and condition, than we would have the right to reduce the story of a man to the last years of his life. It is further obvious that the history of petroleum will have to be entirely different, dependent upon its origin, either from emanations coming from the depths, or from organic decomposition in the strata themselves. In the writer’s opinion, many reasons favor the theory of organic origin for the petroleum found in commercial deposits, though this does not mean that petroleum of a different origin does not exist elsewhere. Hence, this is a point on which we have first to agree, or else agree to disagree. Should the organic origin of the petroleum that is found in pools be granted, the following interpretation is offered for the mechanics of its accumulation.

Oil developed by the process of decomposition of organic matter—whether vegetal, animal or both—would have first to exist in the sediments which contain the parent matter, in the state of dissemination. As Orton remarks, “disseminated petroleum is well nigh universal, but accumulations are rare.”

In order that such an accumulation of disseminated particles may take place, three elements are necessary: An adequate source of supply; a reservoir to hold the oil and accompanying gas under pressure, and a process of concentrating the disseminated material and conveying it to the reservoir. The source of supply will have to be ascertained in every individual case. It is supposed here to exist, for the sake of argument. The reservoir is easy to imagine, inasmuch as we are able to corroborate our theoretical views by actual knowledge. A porous stratum overlaid by an impervious one, may act as a reservoir. This is the general case. Or the reservoir may be constituted by a porous stratum highly inclined and cut off abruptly below ground by a fault, which thus seals the oil and gas and prevents their escape. This kind is found in some Californian fields and elsewhere. Or the reservoir may be constituted by joint cracks in a shale, as is the case in the Florence oil field, Colorado, etc.

The interpretation of the process of concentration and eventual migration, i.e., of accumulation, is more complicated. As a natural process, it must obey physical laws, and especially the laws which govern the motion of fluids and gases; but the forces which may cause the motion (gravity in its different forms, heat, surface tension), the agents through which these forces accomplish certain results (water, oil and gas in varying physical or chemical states), and the factors that intervene in the process (structure and texture of the rocks, porosity, stratigraphic conditions, water and gaseous contents, viscosity, capillarity, depth and time) may have or have had a widely varying range of influence, according to past and present conditions. These elements would have to be seriated, according to their probable rank of importance, and their relative degree of influence determined, before a tentative conclusion could be reached. It thus becomes evident that if some broad general principle may be laid down, a great number of variations are to be expected for each individual case.

A few general observations may help us from the start. According to the organic theory, which is accepted here, the ultimate source of the hydrocarbons is to be found in sedimentary organic deposits, and oil accumulations in commercial quantities (oil pools) are always associated with sedimentary strata. The immediate inference is that hydrocarbons must have been submitted, from the date of their origin, to the action of the forces that may have affected these same strata.

If we refer to a map of the known oil fields of the globe, we will notice, with Sir Bowerton Redwood, that “whilst petroleum exists very generally distributed throughout the world, the principal deposits occur along well defined lines, often associated with the principal mountain chains.” This remark is important and deserves to be emphasized. It means that petroleum deposits are mainly associated with the lines of lesser resistance of the globe, with the general direction of geosynclines and consequently with the areas of general orographic movement and deformation. The distribution of the Eastern oil fields of the United States and Canada along the Paleozoic trough, or geosyncline, of the Appalachian region, the location of the Western oil fields of the Pacific coast, from Alaska to California, followed by the Mexican oil fields, the deposits of the West Indies, of the northern coast of Venezuela, Columbia and a part of the Andes, along the general trend of a Mesozoic syncline; the oil fields of Galicia, Roumania, of the Caucasus, of Burma, of the Islands of Java, Sumatra, Borneo and New Zealand, following the path of another syncline, are extremely suggestive.

This first observation may be completed by a second. An oil field is always accompanied by a certain amount of local structural deformation, which sometimes is reduced to simple undulations, as is the case of the Appalachian region, or may reach a stage of high disturbance, with contorted strata, as is the case of the Galician oil fields. But local deformation has been observed everywhere. It becomes thus difficult to escape the impression that structural deformation, general as well as local, is more or less connected with the phenomenon of accumulation, and our inquiry is thus directed toward the possible action of the forces which have produced these deformations in the sediments from which hydrocarbons have proceeded and their possible action on the hydrocarbons themselves.

As stated before, oil would have first to exist in the stratum where organic decomposition takes place, in the shape of a finely disseminated matter. The sediments themselves would have to be deposited under water, in lagoons, marshes, deltas or at the bottom of the seas, in the form of, mud or ooze. Such a mud would be composed of extremely fine mineral particles, intermingled with water in large amount. The particles of oil would take the form of spherical globules under the influence of surface tension, and, as these globules would be larger than the mineral particles of mud, they would be mechanically held in the mixture. Experiments conducted by Murray Stuart, Assistant Superintendent of the Geological Survey of India and others, have clearly established this capacity of muds to hold oil in sediments by purely mechanical action.

The preceding condition may be termed the first phase of the process. It is characterized by a layer of mud or ooze, holding mechanically, between its very fine mineral particles, disseminated drops of oil and gas bubbles, in a “matrix of water,” as Murray Stuart expresses it.

When the conditions of sedimentation which allowed the deposition of the mud have changed, some other material may be deposited over this first layer; say some sand. Then, another change in sedimentary conditions will occur, and a new layer of material will cover the sand; suppose another layer of mud; and so on successive layers of various sediments will be piled up in the sedimentary syncline.

Now, the progressive increase of weight due to this accumulation of successive layers will cause a progressive compression to take place. The effect of compression on a mass of practically incompressible mineral particles imbedded in water is to draw these particles closer together, to reduce the space available between them, and to squeeze out more or less of the liquid contents. The limit of compression is reached when the particles come in contact. For this reason, the result would be somewhat different, according to the size of the component particles of the layers. Mud or ooze particles are exceedingly small. When, in the syncline, the lower layer of mud would be compressed, the resulting texture would be close, and the tendency would be for the fluids to be squeezed out to a great extent; whereas the grains of a layer of sand are much larger than the particles of clay, and such a layer may become incompressible when it still contains a great percentage of holes. It may thus act as a reservoir for liquids even under considerable pressure. The overlying sand stratum would then remain porous to a large extent, and the liquids, escaping from the underlying mud, would rise and become confined in it. This may be termed the second phase of the process, and is characterized by the progressive transfer of a large part of the oil and gas from the clayey layer in which they have originated to the overlying porous stratum in which they still remain distributed in a finely disseminated state. At this stage, the porous rock may be supposed to be saturated with an emulsion of oil and gas in water.

Finally, in the course of time, sedimentation is stopped in the syncline of deposition, and orogenic movement begins. Whatever may be the cause of this movement, it is a tendency toward a new adjustment which evinces itself by a lateral thrust. In the simpler form, one rim of the syncline remains immovable, and acts as a resistance, arid the other rim is brought nearer to the first one by a movement tangential to the crust, which squeezes the syncline as between the jaws of a vise. The strata are bent and lifted above the level of the waters.

It is possible to get an idea of the order of magnitude of the thrust and to compare it with the forces due to buoyancy and hydraulic pressure invoked in the preceding theories, and which we have found to be insufficient to explain the migration of the oil. For this, we have to consider that this new force exceeds the resistance to crushing offered by the most resistant sedimentary rocks. This crushing strength becomes thus an inferior limit to the force. Limestone, for instance, may be crushed under a pressure of between 400 to 1,200 kg. per square centimeter; so the thrust is to be measured at least in hundred kilograms to the square centimeter. We have already seen that the force of buoyancy, by which a cubic centimeter of oil is supposed to be lifted along a dip in the anti-clinal theory, is to be measured in fractions of a gram (in milligrams for a dip of 1 ft. for 50 ft.), and hydraulic pressure due to underground waters, with the velocity admitted, would be still less important. The magnitude of the force we are considering now—the tangential thrust—is then at least one million times greater per unit surface than the two other forces previously considered. Further, this force has left its imprint in all sedimentary strata, not only by the flexure of the strata or their relative displacement, but frequently by dynamo-metamorphic actions which it has brought about. These actions have taken place between the time the strata were lying at the bottom of the waters and the time they.have acquired their present situation. If then, any oil was ever existing at the

time of the deposition of the strata and has remained in some stratum since, it must equally have been subjected to the action of the thrust. The third phase of the process begins with this action of the thrust. But this needs a more complete analysis of facts.

When a square prism of a solid material, such as a piece of wood or an iron bar (Fig. 2), disposed horizontally on a support in order to avoid the action of gravity, and propped against, a resistance R at one end, is com-

pressed at the other end by a force P, in the direction of its axis, the compression is instantaneously transmitted to the resistance, from end to end. Then, if the pressure is sufficiently increased, the prism will bend (Fig. 3). This flexion, will extend the fibers on the convex side and contract them on the concave one, and there will be an intermediate layer which will be neither contracted nor extended by the flexion. This

is the neutral plane (Fig. 4). Superimposed to this first effect, there will be a general contraction along the axis, which will affect the whole prism.

If, instead of a prism of solid material, we operate on a prism made of some homogeneous plastic material, like moist clay, the result would not be exactly the same. The compression at one end would not be immediately transmitted at the other end, but only, progressively, in proportion as the inertia of the successive sections of the prism and the friction against the supporting stand would be overcome, the one after the other. And deformation would begin at the end where the thrust P is applied, even before any compression would become noticeable at the other end, if the prism were long enough.

Let us now imagine a prism composed of a series of parallel, superposed, horizontal and homogeneous layers, individually uniform in thickness, but variable in composition and resistance, such as clayey, sandy and calcareous layers would be. Suppose such a prism submitted to a uniformly distributed load, representing the load of superincumbent strata; the friction against the stand reduced by the interposition of a soft base, and proper precautions taken to avoid lateral deformation. And let us subject this prism to the same process of compression. Here again the compression would be transmitted progressively from the point of applied pressure to the point of applied resistance, and deformation would begin at the extremity nearer to the thrust. This deformation may vary, according to the, relative thickness and disposition of the “competent” and “incompetent” strata, as they have been termed, but would generally take the form of a local bend, some sort of anticline, which may be overthrown in the direction of the resistance. If now

the pressure should be continued, waves would rise in succession, from right to left, decreasing in height, the farther they reach toward the resistance, and the flexed prism would finally take the general appearance of Fig. 5. The mechanical result would be: (1) A general compression of the prism, which would decrease from P to R, where it might disappear, and (2) deformations by bending, with appressed folds toward P and lower anticlines toward R; the folds being generally overthrown in the direction of the resistance R.

Now, we have already seen that when a material is compressible, compression will tend to draw nearer to each other the solid particles of which the material is composed. Compression would thus tend to a reduction of volume and to a reduction of porosity, or capacity for liquids. The first mechanical effect of our experiment, then, tends to reduce the capacity for liquids of the material at a decreasing rate from P to R, from the point of applied pressure to the point of applied resistance.

To this would be superimposed the mechanical effect due to flexure. Anticlines would be extended on their convex side and synclines compressed on their concave side. If, then, the thickness of the prism is sufficient, and if we consider only its upper portion, above the neutral plane, the general effect would be as follows: A general decrease of compression and a general increase of capacity for liquids from P to R, superimposed on a wavy succession of local increases and decreases of capacity. This would create a succession of zones of increasing capacity from right to left (from P to R), individually elongated in a transverse direction to the prism, or in a parallel direction to the folds. If then the porous strata were saturated with a liquid, the tendency would be for the liquid to move from the more compressed parts to the less compressed or decompressed ones. This movement would take place along the lines of lesser resistance; viz., (a) from the clayey layers to the sand strata, which would complete the concentration of the oil emulsion, already begun by the effect of vertical compression, into the sandy layers, and (b) along the sand strata themselves. Pools would be constituted parallel to the flexures, which would become more important the nearer they approached the resistance, or increasing from right to left.

Further, compression and consequent migration of the emulsion through the sand would have a tendency to allow oil drops and gas bubbles to coalesce. Temperature may play a part in the phenomenon, as pressure would generate heat. This heat may be negligible in the zone of reduced compression, but may be important in the highly compressed one. Rocks may afford sometimes a certain amount of heat without noticeable changes; but organic liquids would be more or less modified. A slight increase of temperature would lower the viscosity of the oil and facilitate its migration. A sufficient rise may partially decompose it, and complex reactions may occur if this decomposition should take place under pressure, in the presence of water vapor. This side of the question will have to be investigated more completely, should the general theory be admitted. For there could be found possibly a partial origin of the different grades of oil, and the source of a large part of the gas which accompanies oil in the pools today. The complexity of the products would still be increased by the filtration of the oil through certain rocks, which an excessive compression may allow, or which capillarity may induce in other circumstances.

At any rate, gas and oil concentration in the zones of decompression or of reduced compression would result; and water, oil, and gas may finally settle in the reservoirs according to density.

On the other hand, we have seen that the anticlines would be extended along their crests and the synclines contracted along their bottoms, if we consider only the part of the prism above its neutral plane. The liquid would thus be squeezed out from the synclines and would have a natural tendency to collect, by the effect of compression, in the higher and more open parts of the anticlines. Later, if, for some reason, water were reduced in amount, the oil would follow the level of the receding waters and sink toward the bottom of the synclines, as long as porosity would permit.

It might occur, with a sufficient increase of the lateral thrust, that the material subjected to compression would break instead of bend, and that overthrusts and faults would result. These deformations would naturally take place in the more compressed region, on the side of the thrust. This would open a way of escape to the gas and to the liquids, and the affected parts of the oil-bearing strata might be deprived of their contents, by outflow or evaporation. The oil and gas would be lost and these parts of the layers would become dry.

As far as we have gone, we have only considered theoretical views and results of laboratory experiments. The material which we have been handling was simple, the individual layers homogeneous, of constant thickness and horizontally disposed; the forces were acting in the vertical plane of symmetry of the prism, etc. And from a starting point fraught with geometric simplicity we have reached simple and geometric results. The merit of this way of proceeding is that it has allowed us to understand more clearly than could be done otherwise, the trend of events when the elements of the problem are reduced to their simpler lines.

But in Nature, things are much more complicated. Let us consider, instead of a prism, a syncline of deposition and note the new set of conditions which may affect the result. Between the geometric dispositions of our previous scheme and the dispositions encountered in the field, we will remark at once the following differences:

The material will not be homogeneous, but varied to the extreme. Not only will the strata be different from one another in composition, texture, resistance, flexibility, etc., creating a lack of homogeneity in the vertical direction, but similar differences will be found in individual strata, so that the lack of homogeneity will become almost general.

The materials will not be geometrically disposed. The outlines of the compressed area, in horizontal projection, will be irregular, and its thickness at great variance from point to point. The layers will be incurved, presenting original bends of deposition. Their own thickness will be irregular, and the so-called “planes of stratification” will not be planes at all, but complicated surfaces.

If we consider one particular stratum, the pressure due to overlying formations will not be the same everywhere.

The thrust will be irregular, and to simple compression and flexion will be superimposed a torsional movement. Consecutive thrusts, varied in direction and importance may affect the same region. Further, an orogenic movement does not result simply in a series of more or less parallel folds, elongated in one direction; it is complicated by transverse or orthogonal undulations, which create successive elevated and depressed areas along the general strike of the folds; the anticlines have a tendency to be transformed into elongated domes, the synclines into elongated basins.

Finally, breaks of greater or lesser magnitude frequently occur in the folded masses, and folds and overthrusts take place.

For these reasons, the resulting structure of a syncline subjected to orogenic movement becomes sometimes very intricate. Further, the superposition to the simpler effect of lateral compression of a series of incidental effects, derived from irregular forces which enter into play, would put the problem out of reach of mathematical treatment.

But there remains a way open to us. We may consider such a deformed syncline in nature, and try to ascertain how much the facts collected in the field may agree with the general lines of our theoretical views.

A good example of a folded chain may be found in the mountain system of the Appalachian province, which extends from New York to Alabama with a general northeast-southwest trend, and in the immediate neighborhood of which some of the most productive oil fields of the United States are situated. The thrust is supposed here to have taken place from the southeast. The whole Paleozoic series of Strata to the floor of the crystalline Archean rocks—in some parts 40,000 ft. thick—have been involved in the system of flexure. The flexures are generally parallel to the main direction of the chain, and tend to be arranged en echelon in overlapping series. They are mostly unsymmetrical, i.e., they bear a “rear and front structure,” the steeper side generally facing northwest, away from the Atlantic Ocean. Bailey Willis, in his Mechanics of the Appalachian Structure recognizes four districts in the Appalachian belt. The eastern portion of the system is a district of “close folding,” extending nearly its entire length and terminated at the south by a district of “folding with cleavage.” The western part of the belt, on the north, through central Pennsylvania and the Virginias, is a district of “open folding.” South of it, and always on the western side of the belt, comes a district of “folding and faulting,” which passes through eastern Kentucky and Tennessee to Alabama.

The district of “open folding” terminates, on the west, in large undulations of low dip and merges finally into the monoclinal or gently undulating structure of northwestern Pennsylvania and eastern Ohio.

Beyond the southern region of “folding and faulting,” proceeding westward, comes the high plateau or tableland of Kentucky and Tennessee.

As a whole, and barring irregularities of detail, we find here the characteristics which have been outlined in our theoretical statement: Successive parallel folds, more closely appressed on the east (the side of the thrust) and decreasing in importance the farther we come west (toward the resistance), with a general tendency for an overthrow of the folds in the direction of the resistance.

All the oil and gas fields are located along the western side of this belt, on its outer margin; i.e., on the side that is farther from the point of application of the thrust. From Pennsylvania down to Alabama, their main direction closely follows the direction of the chain, and individual pools are frequently disposed in rudely parallel rows, with their major axis parallel to the folds. This parallelism between the positions of the oil and gas areas and the trend of the mountains has long ago been pointed out, especially for western Pennsylvania.

Another remark is that the oil belt is especially developed in front of the region of “open folding,” to the north, but rapidly decreases in importance as soon as it reaches the front of the region of “folding and faulting,” to the south. The effect of faulting would be to relieve the strains and, in so doing, to reduce the, amount of deformation farther west, as well as to open a way of escape to the fluids, if any. This double effect would reduce the tendency toward accumulation.

The Pennsylvanian oil belt offers a further interesting feature. When entering the State from the south, the mountain belt is at first bent eastward in its trend, and it takes again a northern course after crossing the Susquehanna. Fig. 6 represents this curve, which has the form of the letter S. Now, incurvations of this kind are frequently encountered in mountain chains, and they have been interpreted sometimes as the result of the resistance against orogenic movement due to an ancient mass of consolidated rocks (Massif d’ancienne consolidation) in the region M, acting as a pier which would deflect the direction of the thrust. The mechanical effect of such a deflection would be an increased pressure in the concave part of the curve, toward A, and a decompression in the region B, where the curve becomes convex. The fact that the principal oil areas of Pennsylvania and southwestern New York—which are among the most productive in existence—are located precisely in this zone of decompression and do not extend farther east in the compressed zone, is interesting.

The same relations between orogenic deformations and oil-producing areas may be observed in other fields, as, for instance, in the oil fields of central Europe. The chain of the Alps, followed by the Carpathian Mountains, the Transylvanian Alps and the Balkans, are interpreted today as a system of folded chains. The belt develops from southern France to Galicia, Bukowina, and northern Roumania in the form of an arc, with its convexity directed northward. Its eastern extremity presents the well-known sigmoidal inflexion formed by the Transylvanian Mountains and the Balkans. The thrust has been directed from the inside of the curve toward the outside. Everywhere, the oil fields are located on the exterior slopes of the belt; i.e., on the side that is farther from the point of application of the thrust, and the pools are distributed in rows parallel to the trend’ of the folds, frequently with their major axis in the same direction.

In these districts the deformations are no more the gentle anticlines and synclines or the terraces of the Appalachian region; the strata are highly disturbed, contorted and even faulted. The thrust in the Alpine belt has been much more powerful than in the Appalachian region, and in certain places overthrusts of considerable magnitude have taken place. In fact, the Alps proper and the Carpathians seem to have been submitted to folding and deformation at least at two different periods: First immediately after Oligocene times, and second during the middle Miocene; and the present state of affairs is the result of the cumulative effect of both thrusts. It seems reasonable to admit that the region has passed through a first stage similar in some respects to the one presented by the Appalachian belt, and during which the greater part of the accumulation may have been effected, before reaching the further stage of greater disturbance, in which a part only of the accumulated hydrocarbons seems to have been preserved, in consequence of special stratigraphic conditions. Similar relations may be observed in many other fields.

The preceding remarks have led the writer to believe that the facts observed in the field agree with the theoretical views previously expressed, and the hypothesis which he proposes for oil and gas accumulation may be summarized as follows:

First Stage.—The oil proceeding from organic remains, perhaps still in process of decomposition at the very origin of the movement, is at first distributed in the water-laden sediments of the geosyncline of deposition in the state of disseminated particles.

Second Stage.—The increasing compression, due to the continuous accumulation of superimposed strata, expels an increasing amount of the water of deposition with its contents of hydrocarbons, from the original layers, which, at the same time, are the most easily affected by the compression (argillaceous or limy sediments), to some other layers less affected by it (especially sands and gravels). This displacement may take an upward or a downward trend, the only condition being that the fluids must move along the line of lesser resistance, from a more compressed to a less compressed zone. This movement may take place between two entirely different strata, or between two layers of the same group of deposits, provided the layer acting as a temporary reservoir is less compressible or more, porous than the former one. (This latter condition obtains in the dolomitic layers of the Trenton rock.)

Third Stage.—As soon as orogenic movement begins, a more or less horizontal compression, due to the thrust, takes place and becomes added to the vertical pressure due to superincumbent weight. The result of the intervention of this new force is to create: first, a general increase of compression from the point of applied resistance to the point of applied lateral pressure; second, successive and parallel zones alternately compressed and decompressed, whose strike is normal to the direction of the thrust. The waters which saturate the strata are submitted to the effect of this unequal pressure and move from the highly compressed regions to the lesser compressed ones, carrying the hydrocarbons with them in their course, finally collecting in pools parallel to the folds. The movement would have to take place along the lines of lesser resistance, i.e., toward and along the more porous layers of the formation (sandy layers, etc.). Pressure would reduce the viscosity of the oil, favor the coalescence of the globules and perhaps induce some chemical changes of the hydrocarbons. The more probable places of accumulation would be the crest of anticlines, the summit of domes, the rims of terraces, or, in the main, the places where a change occurs in the dip or along the strike of the strata, in the form of convex edges or arches; for the reason that at these places the local reduction of compression, buoyancy, and resistance to motion due to a change in the direction of flow, would act together and accumulate their effect.

When the fluids reach the zones of lesser compression, if the physical condition of the reservoir in which they collect is such as to hold the hydrocarbons and prevent their escape, an equilibrium is established, and the final pressure in the pools must be equal to the original pressure less the losses of head encountered on the way.

Further, a progressive settlement would take place in the reservoir, according to gravity; water would congregate at the lower places, oil would have a tendency to collect at its surface, and gas—either brought with the water or dissolved in the oil and further released by decompression, or simply produced from the oil itself-—would reach the higher places.

Considered as a whole, the process would be a consequence of the mechanical principle of least action. In this process, the agent of transportation of the hydrocarbons would be water; the moving force would be hydraulic pressure created by vertical and lateral compression; and the extent of the movement would be variable in the extreme according to local conditions of the strata. But compression, especially compression due to the lateral thrust, whose action would be irreglar and continuous and would have to be extended over a long period of time, would not act as the permanent head of water which nowadays is the ultimate source of the velocity and of the hydraulic power of circulating underground waters.

Frictional resistance may reduce and even stop the motion of underground water once for all in a given direction; a thrust would act by a succession of jerks and repeat the effort again and again. There would be periods of activity, during which the compressive force would exceed the resistance of the rocks and deformations would ensue, followed by periods of rest, brought by the momentary relaxation of strains due to deformation itself and during which the compressive forces would accumulate; and the thrust would become the source of a periodically renewed energy. The consequent hydraulic pressure would follow a similar wavy movement, with periods of maxima, to which a maximum of velocity would correspond, and periods of minima or of rest, where frictional resistance might bring the movement to a stop. The action of the liquids would thus become similar to that of a water ram, with a maximum of efficiency periodically renewed.

Fourth Stage.—A new stage will be reached by the gradual reduction of the water contents of the strata, producing consequent changes in the level of complete saturation and in the local disposition of the pools, by gravity.

Fifth Stage.—Sometimes, a new period of folding may take place, in which the thrust may have or may not have the same direction as the previous one. New zones of compression and decompression may be created, and the liquids may be put again in motion. The results may become thus very intricate, especially if the strata are deformed to a large extent.

All gradations must be expected to be found in the oil fields between these two extreme sets of conditions. The Appalachian belt may represent the first set, which stops at the fourth stage of our description. The fields of central Europe would represent the further and more complex stage.

The research for oil becomes thus a problem of tectonic effects as well as of stratigraphy.

Rock Pressure

The origin of “rock pressure” has been traced to one of the following causes: Hydrostatic pressure, weight of superincumbent strata, gradual accumulation of the inclosed gas, capillary diffusion.

The hypothesis of hydrostatic pressure is untenable as a general one. Hydrostatic pressure would agree, to some extent, with the pressure found in Ohio and Indiana, but could not account for the heavy pressures encountered in western Pennsylvania, as shown by Prof. J. P. Lesley and J. F. Carll, nor for the pressure of most of the deep West Virginia wells. Further, this theory is contradicted by the well-known fact that flow and pressure are found to decrease in any given well with the age of the well. A certain amount of constancy in the flow and in the pressure ought to be expected under artesian action, which is not the case.

The weight of overlying strata, under present conditions of the rocks would be mechanically inadequate.

The progressive accumulation of the gas may be a cause of pressure, and, according to David T. Day’s experiments, capillary diffusion through Fuller’s earth bears a curious analogy with osmotic phenomena and the pressure due to this cause may be compared with osmotic pressure. But, if both theories may explain the origin of a certain pressure, they are not entirely satisfactory, in the writer’s mind, for the following reasons: The progressive accumulation of the gas would rapidly be checked by the increasing pressure itself, and it has not been shown that such a limitation would not occur before reaching the high pressures encountered in some wells. On the other hand, the action of capillary diffusion seems to require certain physical conditions which are not met everywhere, and therefore it can not be admitted as a general cause. Further, none of the preceding theories explain the following facts: At least in the eastern fields of the United States, the rock pressure, in the main, increases with the depth of the “sand,” and, at the same time, there seems to be a decrease of pressure with an increase of distance from the principal axis of the folding. The closed pressure in the Trenton limestone of Ohio and Indiana averages 200 to 300 lb. per square inch and only exceptionally reaches over 600 lb.; whereas the pressures in western Pennsylvania and in West Virginia, farther east, easily reach the double figure. In other words, it seems as if the pressure would increase, as a whole, in the same, direction as the compression to which the rocks have been submitted at the time of their folding, both vertically and horizontally. A partial origin of the rock pressure, at least, would thus have to be traced to orogenic deformation. The two other forces—due to progressive gas increase and capillary diffusion—may have, and possibly have played a more or less important part in the final result, but this effect would have been produced later, and, in this respect, is to be considered as subordinate.

How Capillary Pressure Seals an Oil Pool

One of the most interesting problems involved in the study of oil and gas accumulation is the process by which gas or oil may accumulate in the pay streaks under heavy pressure, without this pressure being dissipated through neighboring rocks or through the sandstones. Imperviousness of the superincumbent strata and of the oil-bearing bed itself has been frequently advocated. But unaltered rocks of the type encountered in oil fields are never impervious. From 10 to 40 per cent, of their bulk is made of pore space, and the pore space of the inclosing beds of shales averages no less than 6 per cent., which means that every square foot of so-called impervious rocks contains an average of 8½ in- of holes (M. J. Munn).

In his “hydraulic theory,” M. J. Munn has suggested that “the pools of oil and gas are held in place by water under hydraulic and capillary pressure which effectively seals up the pores of the surrounding rock, and prevents the dissipation of pressure by diffusion.” “Pressure in pools,” he writes, “is maintained by the expansive force exerted by gas.

Such gas could not diffuse because of the saturated conditions of the surrounding rocks. But Mr. Munn does not go further than a general statement, and no detailed explanation of the process by which such a “sealing up” is rendered possible has yet been given, as far as the writer knows. The following is proposed as a tentative explanation.

Hydraulic pressure is here discarded, this problem being a problem of statics rather than of dynamics, and in which the velocity is naught.

There is a peculiar and interesting phenomenon which appears to have escaped the notice of those who have tried to explain the “sealing up” of a pool. It has been pointed out by Jamin, a French physicist, and is illustrated by the experiment of “ Jamin’s tubes.” If a capillary tube is incompletely filled by water and the water distributed through the tube in such a way as to constitute a string of droplets, a pressure may be applied to one of the extremities of the tube which will not be transmitted to the other end. In other words, the string of droplets will act as a resistance. If there is a large number of drops in the tube, the difference of pressure at the ends arising in this way may amount to several atmospheres. The explanation follows: A drop of liquid, like water or oil, which does not wet the tube, will be limited on both ends by a meniscus (Fig. 7). The superficial tension which results in this form, is caused by the tension of glass and air, glass and liquid, and air and liquid. The two first sets of forces are parallel to the axis of the tube, and, being equal and directly opposed two by two at both ends of the drop, neutralize each other when conditions of equilibrium are formulated. The third set of forces, caused by the tension of air and liquid, is tangent to the meniscus all around the tube, at both ends, as shown in the figure, and makes with the axis of the tube an angle θ which is the “angle of contact.” The weight of the drop may be neglected here, as its action is insignificant with regard to that of surface tension. When the pressure is equal at both ends of the drop, the meniscuses are identical, and so are the angles of contact, θ and θ’. But if the pressure P at one end is increased, both meniscuses will alter their curvature. The meniscus in front of P will decrease its angle θ, which will tend toward zero, whereas the meniscus opposed to P’ will increase, its angle θ’, which will tend toward 90°. The result of this deformation is to increase the force T cos θ directed against P, and to decrease the force T cos θ’ directed with it. The difference will be a resistance against motion expressed by T(cos θ — cos θ’), which will draw closer to the limit T as the pressure is increased.

The same conditions would obtain in shales or clays capping a gas pool. There is nothing like a plane of separation between the gas of the pool and the water that fills the pores of the superincumbent rock; but there is a more or less irregular intermediate zone in which the gas and the water are commingled. The pores of the shale, by their juxtaposition, would constitute the Jamin tubes of the experiment, and these would be filled by the mixture, which would take the form of bubbles of gas intermingled with droplets of water.

Van Hise remarks that “the majority of the particles of most clays, shales and slates are much smaller than 0.0012 mm., and therefore the openings of the rocks are subcapillary.” He defines subcapillary openings as those which are 0.0002 mm. and less in diameter. Starting from these data, it is possible to calculate, with a sufficient degree of approximation, the thickness of the shales or of the clay which would seal up a given pressure. The writer has found that a few feet would be amply sufficient to seal up a pressure of 1,200 lb. per square inch.

It may be conceived that the same process would apply for the sealing of gas and oil pools laterally, along the dip and strike of the porous layers themselves, wherever the sealing is not already produced by changes in the nature of the “sand” from a pervious to an impervious one, or by the presence of water. Pore openings of sandstones are, for the most part, capillary; and should water not be present in the sand, oil may replace it for the Jamin tube effect. Calculation shows that the marginal zone thus constituted may reach a width of a few hundred feet. Pressure would decrease in this zone from the inside to the outside of the pool, progressively, a feature which is readily observed in the field.