23 January 1989
Hubbert: Well, I'd done previous work at Columbia University which I've discussed the last time, and I'd been using the Nadai theoretical work in connection with mechanics of faulting in my courses at Columbia. When I got to Houston, I soon discovered the senior geologist of Shell in that area was still talking about normal faults as being tension faults.
Hubbert: Yes. It was particularly there, but the principals were common enough. There's a long background history on that, faults, dating back to 1800 or more. They were described originally primarily as geometrical. We had normal the faults in this country and British coal mining country, the mines, and those were hanging wall blocks, the upper block hanging with respect to the lower one. And then there was the reverse fault, which was the other way around, where the upper fault block had moved upward with respect to the lower one. Geometrically, there were these two classes of faults. If you took the normal fault, then you took the two edges of the stratum, say, after the fault which had been together before the fault and you'd find that they had separated in a direction of extension away from the fault. Whereas in the thrust fault, these same two edges had overlapped, and there was a corresponded to a shortening of the section. Now, then it was a fairly easy intuitive transition to credit these two classes of faults to the forces of stress that are associated with it. So the reverse fault also became known as a thrust fault; the two sides were pushed together. Besides that, they were commonly associated with folds, which corresponded to horizontal compression. Then, by a simple step of intuition, when the section extended, that was naturally interpreted as tension. So normal faults came to be also referred to as tension faults, and reverse faults as thrust faults.
Hubbert: Not especially. They were considered to be more or less equivalent, if you accepted one of them. One of them was clearly geometrical, and the other one was used for the forces, stresses, presumed to exist. Now, when I was a student at Chicago in 1926, Chicago did not have anybody who taught petroleum geology. They arranged for a geologist with the then newly formed Amerada company to give some lectures and run a course on petroleum geology during the spring quarter. He also ran a surveying course in plane table surveying. He had a colleague in a sister organization called the Rikaid [unclear] Company by the name of Donald C. Barton. The first man was called Plummer. Donald C. Barton came in and gave special lectures on current and recent geophysical developments in the petroleum industry, torsion balance, seismograph, and so on.
Hubbert: Oh yes, he was one of the chief geophysicists with the Rikaid Companies. That company was a sister company of Amerada. They had a kind of a cousin relationship, according to the corporate structures. Also, it might be remarked that Freddy B. Plummer had formerly been a geologist with Shell.
Hubbert: Earlier. So one of the things of great contemporary excitement in the oil industry was the discovery of the large number of salt domes in the Gulf Coast, and that had been discovered by torsion balances and also by seismic work. Very large oil fields were being found around the peripheries of these salt domes. In fact the principal exploration for oil on the Gulf Coast was searching for salt domes at this time. I remember asking Plummer what other types of geologic structures there were on the Gulf Coast other than these salt domes.
Hubbert: Yes, 1926. I asked him specifically for example about thrust faults. His reply was approximately as follows: that thrust faults were due to compressing stresses, and the same stresses would cause the wrinkling or the folding of the sediments. Since there was no folding in the Gulf Coast sediments, you wouldn't expect any thrust faults. With regard to normal faults, normal faults were tension faults, but these Tertiary sediments in the Gulf Coast were largely unconsolidated sands and clays, with marginal tensile strength. Therefore you couldn't have any tension faults or normal faults in the Gulf Coast. Explorations were mainly around salt domes and were not expecting faulting in the Gulf Coast sediments.
Hubbert: Yes, that was orthodox geological thinking. Well, when I got to Shell, how many years later, in 1943...
Hubbert: Yes. Right. I was surprised to find the senior geologists, some very able and distinguished geologists, still talking about this. By now they had faults in the Gulf Coast. Normal faults were fairly abundant, but they were still calling them changing faults. In other words, they had no understanding, really, of the mechanics of faulting. So after I'd been there I guess about two years, it became more and more evident that there was need to educate our geologists with a better understanding of the mechanics of faulting. That time we were due to have the annual meeting of the scholars at the Geophysical Conference, about a three day, two or three day meeting. And I had decided that I was going to give a paper on that subject at that meeting, as a kind of an opening wedge on this question. In the meantime, within the year, I had a colleague by the name of Willy Hafner who was a Swiss geophysicst, PhD from the University of Zrich. He had a good scientific training in physics, mathematics.
Hubbert: He joined Shell, oh, back in the middle twenties, middle or late twenties. His assignment was as a torsion balance operator, and he was now at this time an interpeter of seismic results. He came in my office one day, and I had just ordered and received a book which I'd seen advertised in the bulletin of the American Society of Geologists and I ordered a copy.
Hubbert: The Anderson book, by British geologist, E. M. Anderson, and it was called as I recall, MECHANICS OF FAULTING AND DYKE FORMATIONS.
Hubbert: Not "dike." Well, it seemed fairly obvious that the man was talking about sound mechanical sense, although I didn't take the time to read it, I didn't have time at the moment.
Hubbert: No. I had this book. I'd flipped the pages and I saw that the man knew what he was talking about. Hafner came into my office one day about that time. I said, "Willy, here's one of the few books in print that I've ever seen in which the author is well instructed in geological things and in which the author perfectly understands what he's talking about mechanically." Well, Hafner borrowed the book promptly, took it home and digested it from cover to cover with great enthusiasm. So it turned out the time I was thinking about giving a paper at this conference, using this sandbox experiment type of analysis, Hafner arrived at the same conclusions and he gave a paper based on this Anderson book. So they were really complementary papers, from different methods of approach to the same thing, arriving at the same results. Well, we gave those respective papers. Hafner's was essentially a review of the Anderson book. My paper was a development from first principles of the mechanics of simple structures like normal and reverse faulting, and compressional fault and so on. Well, these two papers were bound. We had developed a very nice system of internal publication at that time, we used 8+ by 11 format nice and informal, and it was before the IBM typewriters we used the old oh, the kind that had the rotating wheel on it similar to a daisy wheel right now, in printers.
Hubbert: Well, the drum was by the vertical axis. Multiplex it was called. So we adopted those because they had interchangeable types for mathematical purposes, you could pull them off, do equations, and put them back on to do italics, etc. We had a pool of typists that did our typing for technical reports, and a standard format. We ran them off on offset printing, and we had a flexible cover on the outside. It was a very, very neat publication.
Hubbert: Practically the entire the company, within the company. They were given a rather widespread distribution within the company.
Hubbert: I can't say. Sometimes there might have been more. Sometimes there was much wider use for a paper, some papers, than others. There'd be a minimum of maybe a hundred or so, and the maximum might get up to several hundred.
Hubbert: Oh, this was entirely internal. This was internal company distribution. And if some of these were released for outside publication, well, then they might be redrafted to make them suitable for outside publication, in the format of the journal, and also eliminate irrelevant materials, and more or less streamline the subject for a general purpose, rather than oil company purposes.
Anyhow, we gave those two papers, and they were on top being bound together in one cover as a unit, and then they became in the nature of a textbook in our training course in geology. They were very widely used. I don't know how many hundred copies of those were ultimately released. And so finally after maybe ten years, no maybe five or six years, it occurred to me, look, here we're doing this thing. It has no smell of oil to it particularly, it's of general scientific interest, and the whole geological profession is practically in the same situation. They don't understand these things. They don't know anything about it, most of them.
Hubbert: No fighting about it. They were very interested. And so, realizing that this was of far broader scientific interest than just our internal use, I proposed that these two papers be released for publication, and be published together jointly. Well, the people who had to approve this thing agreed to release my paper but Hafner's paper they said was not original but largely a review of the Anderson book. Hafner, then on his weekends he went into a major period of study, and he took the G.B. Airy stress differential equation, it's similar to [unclear] equation only it's a higher order differential equation, from which you could derive stress components. Hafner took this as his starting point, and derived his own whole sequence of consequences of geological interest with regard to folding, faulting and various geological problems.
Hubbert: Yes. He was a hard-working man and very busy at his job, and he did this entirely while lesser people would be out playing golf. On his own. So then these two papers were sent together to the Geological Society of America, and after the usual refereeing, review and some slight modifications, they were published in the GSA [Bulletin] in 1951. My paper was on the mechanical bases of certain familiar geological structures. And if you can get in that book on structural geology, off the second shelf over there, I think it is.
Hubbert: It has a yellow back.
Hubbert: That, yes.
Hubbert: So here are the papers. Incidentally, I had this box built by the shop. This is loose sand; here's a partition that extends to the bottom of the box.
Hubbert: A crank, yes, to move this along. So moving it just a bit forward, there's that 60 degree depth. These markers are potted plaster of Paris, I put there, for visibility. They have no mechanical significance. But you have a 60 degree dip fault.
Hubbert: A little bit. But it's secondary. Here's the other side. There's the outcrop of these thrust faults, and here are the visible traces. See, there's one, two, three of them, in the foreground.
Hubbert: And you can see the traces of this one. Here's the same thing over here, same, from a different view, and again you'll see that these run from about the edge of the box at about a 30 degree dip. And the whole series of these experiments, here's a whole series of normal faults and their dips. Here's a whole series of the reverse faults, and their dips. Here is the angle of dip depth, and the interior of the sand, the was measuring away from the glass front. And here's the angle of dip of the reverse faults, not exactly 30 degrees, the range, 20 to 28 degrees or so. That's consistent with observations on the outside. Here's the development of the theory. So, if you start with this undisturbed state, you're dealing with essentially hydrostatic stress, where the horizontal stress and the vertical stress are both the same at a given dip. But as you move this block forward, the overburden is determined by the vertical stress. So what you're doing is loading up the horizontal stress. On this side, what you're doing is easing away on the horizontal stress, and so what happens is that when you do get failure, you get it here at about 30 degrees to the greatest principal stress. In this case the greatest principal stress is vertical. Here the greatest principal stress is horizontal.
Hubbert: But the slip surface in both cases is about 30 degrees to the greatest stress.
Hubbert: Not verbatim, but this is what I learned from Nadai. And I developed the theory here from first principles.
So then, we come to the next question here: this sand has no tensile strength. And yet you have perfectly good 60 degree faults. So that completely demolishes that notion of the normal faulting in the Gulf Coast being tension faults, or their not being any because they have no tensile strength. The zero tensile strength is approximately correct, but by this time we'd discovered that the faulting in the Gulf Coast was very, very common, and the faults were parallel to the strike, the sediments were [unclear] and the faults were mostly parallel to the strike. The fault plane was mostly dipping forward and sometimes reversed.
Hubbert: It was established knowledge by this time. There's an interesting sequence too that I got some years later. One of our senior geologists in the Gulf Coast Region was in charge of production geology. In other words, he and his group did detailed work on operating oil fields.
Hubbert: Yes, O. G. Wilhelm. Another Swiss.
Hubbert: Oh, about the midtwenties.
Hubbert: Yes. So, I'm commenting on this thing to him one day. Kind of reviewing his background history. And he told me something that I didn't previously know. He said that that notion of no normal faults in the Gulf Coast was the same argument made by Plummer which persisted until in the 1930's. He also said that in fact there was a geologist on the Gulf Coast who found that he had to postulate a normal fault in the subsurface work by showing his oil logs and so forth. The higher authority in The Hague sent in another man to correct this man's work and do it right, and delete those normal faults. He said that dogma persisted until the introduction of the electric logs about 1933, when the evidence, the correlation of the electric logs became unmistakable if there were normal faults. So this was a case of a geologic misinterpretation based on faulty understanding of mechanics. And yet it persisted for ten years or more, as a frozen dogma.
Hubbert: It was primarily being done by the oil industry. After all, they had all the interest in it. The outside people didn't have much to go on.
OK. Well, these two papers then were published. They were received very well, there was no argument ever that I know of about either one of them. They were totally accepted. So that was the first thing I did in structural geology with Shell. And then, another thing came up shortly after that. It was just internal.
Hubbert: No, no. I doubt it. It's the same name, obviously Germanic.
Hubbert: Well, OK. Something else occurred in parallel with that, just before these papers were published.
Hubbert: Yes. My opposite number, the director or associate director of exploration had joined the company in 1946. His name was J. P. Murphy. He was from the West Coast.
Hubbert: Yes, he was in the operating area. He was a petroleum engineer on the West Coast.
Hubbert: Oh, he had a degree in maybe petroleum engineering at Berkeley. It was probably a petroleum engineering degree, but he had ten or fifteen years of field experience on the West Coast. And so he came in, in charge of the production part of our research laboratory. We were in charge of exploration and production research.
Tape 18 side 2
Hubbert: In fact, he'd come in at a time when I was on a trip to Europe in 1946, and I gave this paper at the 6th International Congress on Applied Mechanics in Paris. And so he wound up by inviting me to give the same paper at his forthcoming meeting of production engineering conference, which I did.
Hubbert: Yes. We had the whole Geophysical Conference had been running for several years, and then parallel with that, the production men organized their conference.
Hubbert: No, a different time, but annually. He brought in not only the research staff but the operating people from various areas, about 50 or so people. Well, so at this particular time, a very interesting engineer from the West Coast, who happened to be one of Murphy's colleagues, was there. The two of them, Murphy and this fellow, I don't remember his name, were buddying around and trying to figure out what to do about a very serious problem in drilling, which was the problem of lost circulation. Now, I don't know whether you know what lost circulation is. When you drill with a rotary drill, you go through mud, way in mud down the drill pipe. And it goes out through the bit, and there's two things to cool the bit, otherwise it would get hot very soon.
Hubbert: It cools the bit and it picks up the cuttings. The main cutting brings the cuttings up to the surface, and there it's going through a sorting system on the surface, and the cuttings are separated out, and the mud reworked and it goes back down again. It's a circuit.
Hubbert: Right. Another part of this is a third thing the mud does. The mud pressure has to be greater than the presure of the fluids in the uncased hole. You see, you may have the whole case down 5000 feet, and you may be drilling lower than that. So, in the uncased part of the hole, that mud pressure must be greater than any anticipated pressures, fluid pressures on the exterior sense, so that you have a mudcape around the hole. If the fluids outside have a greater pressure than the mud, they'll break through that and flow into the hole, and if it's a gas, you'll get a blowout.
That was a very critical and important function. So what these men were talking about was something that I'd just heard of, but I didn't know anything about it. That is, that in the Gulf Coast wells, down around five or six thousand feet, until that depth, why, the pressures of the formation fluids are just about the hydrostatic pressure of water. But they go rather rapidly, within maybe a hundred feet or two, to quite high pressures, and if they're drilling with light mud, down to that depth and if they run into this transition zone with this light mud, and the sand they're drilling happens to contain gas, that can flow in.
Hubbert: Blow away the top and you can lose it all. So it's a very tricky thing. The obvious way to do it is to be drilling with mud which is heavier and has a higher pressure, but if the pressure gets too high, the mud doesn't come back. It goes down the hole, it doesn't return. In that case, you can lose pressure again by having the mud in the annulus also drop down, drop the pressure, and you get it all up.
All right. So they are scouting on a very narrow margin. If they don't have enough pressure, they can get a blow out, and if they have too much pressure, they get a lot of circulation and a blow out. Well, that was the question that these men were batting around, wondering what to do about it. And Murphy finally called me in on it, and said, maybe I could suggest something. So they outlined to me what the problem was. Well, I had heard about lost circulation, but I thought it was a cavernous condition in limestone or something of that sort that they drilled into, cavernous rathole or something. Not always. But I was told that the sediments that they were drilling into were exclusively sands and clays. And this mud was so tailored that it would not enter these types of rocks. You could leak a little bit of filtrate in, but the mud itself would just form a plaster on the surface. So, you can't get lost circulation by drilling into a hole that doesn't exist. And so the only other answer is that there must be a pressure induced fracture in the rocks. And the mud is going down this fracture. Which did not pre-exist because you reduce the pressure, and you regain circulation.
Well, so the question was asked me, they were getting this lost circulation at mud pressures that were considerably less than the total area weight of the overburden. You know the density of the rocks plus water, you know what the weight of vertical overburden pressure is. And they were opening these rocks up with a pressure that was less than the overburden pressure. And how is this possible? I said, "Well, gentlemen, now that I understand your problem, I think I can explain that to you." And I took my sandbox here, brought it in, and we did this little experiment, producing a normal fault.
Hubbert: Yes. We did this sandbox experiment with a normal fault. Now, you know what the stress is in the undisturbed state, so if you ease away here, you're easing away on the horizontal stress. The vertical stress remains unchanged. Which is the weight of the overburden. And so this normal fault is in response to the stress difference between the greatest and the least stress, and occurs at this angle about 30 degrees to the greatest stress. Now, since the Gulf Coast is full of normal faults, which geologically have been recurrent throughout a long period of geologic time, they didn't all happen at once, in other words. They happened in different times. This indicates, instead of stresses in the Gulf, that here it's all compressive. The least stress is normal to the strike of the faults, and the greatest stress horizontically is parallel to the strike of the faults. In fact, the two greatest stress is vertical. And the medium stress is parallel to the the strike of the fault. So, that's a stress state that would be inferred from the existence of these faults. Since these faults are recurrent, they must be hovering very close to that breaking point most of the time. Your chronic state of stress in this area is one in which the overburden is the greatest, and the least stress is perpendicular to the general region of the strike. And in the sandbox, there's approximately one-half of the overburden in which fracture occurs. So I showed him that this was a stress state, so therefore, I only need a mud pressure that is greater than this minimum thrust, which could be very much less than the total weight of the overburden. That's what they're dealing with when they're drilling into rocks in a state of stress. If the mud pressure is a little greater than the minimum stress, then you can open the rocks up by simply pressure applied in the walls of the hole and those fractures should be vertical, and should be parallel to the strike of the fault, of the original, and perpendicular to the least stress.
Well, that completely satisfied these gentlemen. Somewhere about this time, somebody published a paper in the AIME in which they displayed in this oil paper a lack of understanding of this kind of thing. I wrote a discussion of that paper, which was published in the AIME. I don't remember the original author and I don't remember the date. I wrote a synopsis of this viewpoint in that discussion, the only thing I published of that nature at that time. Then about a year later, the engineers of Standolind Oil, Oklahoma, which was the operating branch of Standard of Indiana, and then something else they kept changing their name but anyhow, their headquarters was in Tulsa. Two or three of their men published a paper in the trade magazine of AIME Transactions in which they described a new method of increasing the production by means of hydrolically produced fractures in the deep oil wells.
Hubbert: Yes, that paper was J. B. Clark, "A Hydraulic Process for Increasing the Productivity of Wells", 1949. My discussion of that earlier paper is not in there, I don't think. Well, anyhow, that paper produced a lot of excitement, and should have, because one of the problems of getting the oil out, especially out of old fields where the wells were running down, was to increase the rate of flow of the oil from the well. The engineering staff of Standolind, had been studying the so-called breakdown pressures in squeeze cementing as you're setting a pipe. You pump down cement, and you apply pressure and force it back up in the Annulus as a way of setting this pipe. And it was common to increase the pressure and force it up until finally they had what they called a breakdown. They literally fractured the rock. That's the maximum pressure they could put on it. Well, anyway, they finally carried on for, oh, two years or more, this study of pressures in squeeze cementing. Then they finally reversed the thing and said, well, look, why can't we use this for deliberately fracturing the rock for increasing the flow of toward the hole? So they did reverse it, and described in this paper. They deliberately fractured the rock, and they had a kind of a gell laden with sand, and so they applied pressure on this. They pumped a lot of this stuff down the hole, applied pressure, fractured the rock, and then you injected it into the fracture as far as you could. Later you ran in some kind of a solvent that took out the gell, and the sand propped the fracture open. You had a fairly coarse textured sand, and then you had a ready-made channel to the hole for facilitating the flow of residual oil, say, into the hole or if it's a low permeability sand, of increasing the flow, or if it's an old depleted well, of again rejuvenating it.
Hubbert: So this was actually one of the most important developments, in reservoir engineering, over a period of ten or fifteen years. It was very very important, and naturally attracted a great deal of excitement. Almost immediately this technique was adopted and the numbers of cases of treatment just bloomed out into the thousands in two or three years or so. It very rapidly built up.
Hubbert: Yes. I think it was Clark's paper, since this came out of the prior history of study of squeeze cementing. As I remember, he had one or two figures in which he had plotted the breakdown pressure versus depth, in a total of about 200 wells. As I recall there were two sets of figures for two different areas, but this was in the area of Oklahoma, Kansas, somewhere up there. Well, then there was a line showing the weight of the overburden. And only occasionally was one of these measured breakdown pressures above the weight of the overburden they averaged about .8 of the weight of the overburden. So, to determine whether these fractures were vertical or horizontal, they drilled a little well about 20 feet deep or so out on the laboratory grounds there in Tulsa. They fractured it, and they mined down to see what happened. When they got down about 15 feet or so, they found that the slurry or cement or whatever they'd used had gone out on the bedding plane, in a kind of a dish shaped injection. On the basis of that experiment and observation, they concluded that all these fractures they were dealing with were bedding plane fractures. Although the average weight of the breakdown, the average pressure at which breakdown occurred, was only 8/10 of the weight of the overburden.
Hubbert: They were generalizing from this little experiment they'd done on their grounds. So then they drew the conclusion that what it amounted to was, they were lifting the overburden with the pressure equal to about 8/10 of the pressure of the overburden. Well, about a year later, something took place at the annual meeting of the Society of Petroleum Engineers in Dallas. There was an afternoon session devoted I think entirely to this problem of hydraulic fracturing. In this case, a couple of authors from Atlantic that's before Atlantic and Richfield were put together; Atlantic had headquarters in Dallas, and one of these two authors was later chairman of the board of Atlantic Richfield right now, he went up the ladder and he became one of the top officials in Atlantic-Richfield at that time he was one of the operating engineers.
Hubbert: I don't think I do.
Hubbert: Anyhow, these two men from Atlantic gave a paper in which they argued from soil mechanics that if you have a static body, say, of sand Terzaghi was more or less the founder of soil mechanics, the analysis of this stress thing that the horizontal stress is less than the overburden in static equilibrium and is related to the Poisson ratio of the materials. Well, this was the standard development and is true for this particular condition. You're constrained so there's no horizontal extension, and under that constraint, you can figure this out, and the horizontal stress is less than the weight of the overburden because of the effect of the Poisson ratio. They used this argument, then, as the basis for arguing that the horizontal stress was less than the overburden, and therefore the fractures should be vertical.
Well, their conclusion was right but for the wrong reason. But the Standolind boys were completely up in arms over this thing.
Hubbert: Yes. They got a dogfight going that lasted nearly all afternoon.
Hubbert: Mostly Standolind against these two men from Atlantic-Richfield. And most of the rest of us just on the sidelines just kind of enjoying the show. (laughter). Well, there were two men at that time, J. B. Clark and somebody else of Standolind, and they finally were backed into a corner more or less or an untenable situation. Then they reverted to an article by their theoretical man by the name of Lubensky who had worked out the theory of this thing. This goes back to that first paper, that the reason they could lift the overburden with a pressure less than the overburden was because of the effective weight of the overburden was less than the total weight. They didn't bother to explain what was meant by the effective weight. And so in the discussion, Lubensky the theoretical man who was responsible for this notion of the effective weight of the overburden, was called in. I don't know whether he was called in, in person. He may or may not have been there, I don't recall. But I know his name was brought in, in defense. He had worked the thing out. Well, this was 1949 or so, as I remember the paper, maybe a little later.
Hubbert: Yes, his paper was in the middle fifties. Around 1954, '55, somewhere along in there. Somewhere in that area, in the middle fifties. Finally, in about 1954 or '55...oh incidentally Shell had set up a new division called Technical Services. It was a kind of a consulting firm to the rest of the company.
Hubbert: Yes. And it was both exploration and production.
Hubbert: Well, a dynamic situation. They decided they needed it, so they set it up.
Hubbert: I was transferred from research to this technical services. Also, that was the point of the new change of setting aside certain individuals and essentially relieving them of any ordinary duties, and giving them title of consultants. I was the first one of the first, if not the first such person, so designated. And I was given the title of chief consultant, I believe. But I had to tie it on to something, and what? And I gave that a considerable amount of thought, to what I was going to be consulted on. Well, what I wanted to do was to break up this old business of barriers. I didn't want to be told I can't do this or that isn't in your domain.
Hubbert: Well, not really, but nevertheless there was always the potential. You had to be careful not to get on the wrong man's turf. And especially the operating people.
Hubbert: Well, not that so much, but between the research staff and the operating people.
Hubbert: There was a certain jealousy on the part of the operating people towards the research lab. It was new and untried and suspect. So, they set up this technical services unit.
Tape 19 side 1
Hubbert: I was set up off by itself, and I was relieved of all administrative duties, and given this title of chief consultant.
Hubbert: Yes. There were no assigned duties, really.
Hubbert: I was pretty much a free lance. And as I say, I gave considerable thought as to what a chief consultant was about, so that nobody could say, "Get out of here, you're treading on my grass." So I finally decided on general geology.
Hubbert: I made it broad enough that almost anything I chose to do came under general geology. Well, one of the things was that I was by this time, I was accepted among the engineers. That battle was already over, or pretty well over. And nobody in other words, I was in good standing with the petroleum engineers as well as the exploration people. So anyhow, I was Transferred out of the research laboratory down to the headquarters office down at Houston. The research laboratory was about seven miles out on the edge of town. And so about this time, the question came up, in the production department, that here we've had this hydraulic fracturing going on now for about seven years or so, or six or seven years, with thousands of cases already, and yet there's still the overwhelming view is that these fractures are horizontal. So we needed to know, are they horizontal or are they vertical? And I was assigned the job of trying to find out. So that's when I was giving these lectures out of Stanford in the spring of that year, when this young Willis, one of the graduate students...
Hubbert: David Willis. As I told you before, I was so impressed with the boy, and since I'd been asked to stay until Saturday or Sunday, I forget which, by the dean, I told him, I said, "Look, I'm not out here recruiting, but I am so impressed with this young man Willis, and I could use a personal research assistant. I can't describe what he'll be doing, but it will be interesting, and a variety of things, and what I'm looking for is somebody who'll do what I'm doing and do it better."
Hubbert: No, I didn't know myself what I'd be doing.
Hubbert: I'd be doing all kinds of whatever comes by.
Hubbert: Oh yes. Yes.
Hubbert: Yes. No, there was no secrecy involved, we didn't know what we'd be doing.
Hubbert: But whatever it is, it'll be interesting.
Hubbert: OK. Well, Willis joined me. He got his PhD at the end of the year, and he joined me, and his recently acquired wife, in Houston, in about June. So we went into the research lab. We were downtown, though, and I had an office and he had an office, next door. I had a secretary. And my immediate assignment was to try find out what goes on in this hydraulic fracturing. Well, by this time, we had the records of several thousand fracturing jobs, with varying degrees of reliability in their data. We had to smoke out useful information. So, we went back as far as possible, right back where I discussed with Jim Murphy and this man back in 1946 or so, that these fractures ought to be determined by stress. In that connection, I should go back again to Anderson. In Anderson's book on mechanics of faulting and dike formation, he postulated that the dikes implanted themselves perpendicularly to the principal stress. He had considerable field evidence that that was so, in England and Scotland.
Hubbert: So, OK, we said, now let's look at the stresses we're dealing with here, and we'll make the assumption that these pressure-induced fractures will be like the dikes and implant themselves perpendicular to the principal stress in the materials. So then, I'd already had the basics of the stresses associated with normal and reverse faulting, and we were in mid-continent; most oil bearing regions are where normal faults are in stress-relaxed areas. The Gulf Coast, midcontinent, Pacific Coast on the other hand was a different situation. So we started in on this thing theoretically, and if you put a hole in this and there's a concentration of stress pressure around the hole, well known from engineering mechanics, and the theory of elasticity. So, we computed what the stress pattern would be, in a stressed material of the greatest, intermediate release stress, with oil in it. If we look at the stress in the horizontal plane, you have the minimum and intermediate stress, you know, intermediate is the minimum and we're not bothered with the stress of the overburden at the moment. So we put a hole in this. So then, around this hole, [draws] see, here's the unstressed state I mean state in the rocks. Now, we put a hole in here, instead of a continuum. So then what you have from in this plane, is that you're taking the pressure out of this hole. Suppose that the pressure at the moment is zero, say. That means then that the support that was formerly produced by air is being removed. That means that the stress from the outside, a distance from the hole, you have a stress of a certain magnitude, but as you approach the hole, the stress will get bigger.
Hubbert: And then it's asymptotic to the undisturbed state as you go away from the hole. But the theory of elasticity tells you how to handle this problem. And so that's where you're concerned with at the greatest stress. Now, if you take the least stress, you've got a similar thing except it's smaller.
Hubbert: Yes. But it's asymptopic to the least stress and to the greatest. So if we apply mud, the mud pressure...oh yes, if you simply apply a pressure P in there, it produces a tension all the way around this hole, similar to the walls of the pipe. So if the rock had tensile strength with no external the stress, it would be when this pressure gets up, when this tension is equal to the pressure equal to the tensile strength of the rock. So in this case, we're dealing with rocks of a negligible tensile strength, and so the question is, will we get away when the pressure reaches this amount? We've got this buildup here. So what we do is get a fracture that's perpendicular to the least stress. Here's the least stress in this direction. [sketches diagram].
Hubbert: You get a fracture which is perpendicular to the least stress, and it's induced in the pressure just greater than in that compression around the walls of the hole. But it consistently should be less than this hole here. I worked these things out on several occasions. Here's the picture, on the outside of the book.
Hubbert: This is Hafner's paper.
Hubbert: The structural geology book.
Hubbert: We worked on it together. Here is our analysis of the stress data and the effects that would occur, here. We had this sandbox experiment, [showing] what it's like, associated with normal thrust faults, and here are the calculations, and various conditions. Here you have two actual states, super posed, and various combinations. Here is the open hole with no external stress. Here the tensile stress is produced by pressure...
Hubbert: And the position. You work out these separate cases, and then...
Hubbert: ...get the results, you see.
Hubbert: Yes. Well, we borrowed it. We didn't work it from first principles. It was available already. We used Timoshenko and somebody's textbook on the theory of elasticity. So we didn't work it from first principles, we simply used the basic results that they'd established, duly citing our sources. Here, Timoshenko: Theory of Elasticity, McGraw Hill, 1934. And here's Kirsch: Die Theorie der Elasticitt. Well, we reviewed scads of oil data. And as this paper was being prepared, there was going to be a big Shell conference over in Amsterdam, worldwide scope.
Hubbert: Well, how often they had these high level conferences, I don't know, but this one was being held in the summer of 1955. This paper was being prepared ultimately as a report to be given at that top level Shell conference in Amsterdam. And we had to have it typed and sealed and signed and reproduced and everything in advance of that meeting, so we were working against a deadline, and the deadline was some time around oh, February or so, 1955. And we'd started to work on this about June 1st, 1954. But we got the job finished. It was shipped over in advance. And it was very well received by the highest level technical people in the Shell organization, and was given at the meeting, and accepted completely, with no significant criticism.
Well, then, that fall I gave the paper over there that fall, they decided to call in the field engineers in the US and Canada, and give them an orientation on this thing for use in their work out in the oil fields. So this conference was called, and there were, oh, I don't know, maybe a couple of dozen engineers from the US and Canada, all operating areas. Willis at the time was out in California on another job in connection with the Ventura field, and I was handling this conference all by myself. And what I discovered was that the theoretical argument was having no effect whatever on these men.
Hubbert: They were sure absolutely that they were making horizontal fractures. And no amount of theoretical argument would make any dent in them.
Hubbert: Well, I think it's reasonably clear. They'd been brainwashed, so to speak. In the first place, this was the original pioneer paper, it carried a lot of weight, and he'd drawn this conclusion. Every paper that was published after that used a figure with a pancake type of fracture. Service companies were running ads that they were available for doing these fracture jobs, and every one of them showed pancake bedding plane fractures. So when you look at those things, and it's universal for six or seven years, you get to where you're completely brainwashed. That's what happened to these people. So they didn't have any real evidence, but they'd been so thoroughly indoctrinated on this thing that they knew damned well these fractures were horizontal. And I wasn't getting anywhere with them at all. We had an all day meeting, and I think nobody was convinced of anything.
Hubbert: Yes. So the meeting broke up. Everybody went home. And Dave Willis came back to town, a few days after that, and I told him that I didn't get anywhere with these guys. I hadn't convinced them of anything. They just mentally said, sat. Well, Willis didn't say very much about it. Willis came back at the end of the week, and I gave him an account of this meeting, and the following Monday morning, he came into my office and wanted to show me something he'd been doing over the weekend. I had a stack of paper about so high on the desk, and said, I'd go through these things, and then you tell me about it. He came back about a half an hour later, getting more and more fidgety, he really wanted to show me what he'd been doing. And I found out that he had an experiment that he wanted to show me, and he had a certain time limit on it. He was worrying about what kind of an experiment could be devised that was could be produced in the laboratory, a demonstration, that would demonstrate this transform, that these fractures had occurred at right angles to greatest or the least, rather than stress. So he concluded that gelatine would be a suitable material. So he'd brought a globular-shaped fish bowl about so big that had two plane faces on it, about like this. And a supply of Knox gelatine at the grocery store, and he cooked this up and congealed it well, he poured it into the bottom of this, into his goldfish bowl. He needed some way to relax the stress, and so he made a plastic bag, a demonstration with plastic film all over the place. Here are the plain faces of his goldfish bowl, here, and coming off the top like that. Then he wished to be able to relax this stress, so he made a little bag of plastic, with water in it, and hooked it up to the outside. Then he poured the gelatin in here, and then he placed an Alkaseltzer bottle like so, with a lid on it, so he brought it up here, no, he brought it to here. This was congealed it was liquid, and he put it in the refrigerator to congeal it.
Hubbert: So it was up to here. For communication to the outside, for tubing, he had drinking straws. A lid with two holes in it. Anyhow, there's the bottle. And he ran warm water into it. And that melts the gelatine around here. Then he removes the bottle, then puts the lid back on the bottle, with two drinking straws communicating outside. So what he has here is an open hole, sucks in the gelatine, and then he follows that on up with another load of liquid gelatine. You congeal that. He'd been doing that all over the weekend, and that's what he had.
Hubbert: And he was getting real nervous because if this thing warmed up, it would liquefy. (laughter).
Hubbert: So in that situation, we dropped everything. I don't know why, but we had a few useful tools on hand already, an egg beater, a squeeze bulb syringe, and he brought in some plaster of Paris. We made up a plaster of Paris, worked up a plaster of Paris slurry, and so we pumped it down one of these tubes. Incidentally, this was all visible. It's all transparent so you could see what was going on.
Hubbert: So we pumped the plaster of Paris down these tubes. It began to fill up in the hole. And when it really got full, and began to go the other way, we started to get some real, just little tiny 45 degree things on these square corners. And then finally one of these took off and went right up the side of the hole. Seen from here it would be on this side, this way. But one of them took right up, perpendicular or parallel to these plane faces, perpendicular to the released stress. You pump the water out of this and relax the stress.
So it was perfectly obvious that we had a technique there that was sound. It would work. Not only that, but when you let it stand by, the stuff liquified. You could lift it out of the plaster of Paris setup, and so you could lift this fracture thing right out of the on the straws and you could dump the gelatine and simply have the thing for demonstration. So I said, OK, boy, you've got something here I think we can do something with. So I arranged for lab space down in the lab, and we went to work down at the research lab to do a more elaborate set of experiments. First we got a about a nine gallon aquarium. But it was rectangular and we still had this problem of relaxing the stress.
Hubbert: No, we could do that, but it was awkward. And besides it was a lot of gelatine. So we played with that for a week or two, and then all of a sudden it occurred to us: my God, how dumb can you get? All we needed was a flexible circular cylinder.
So we did this: We built a circular cylinder, and it was roughly 18 inches in diameter and about 18 inches long. So we could, somehow or other we could mold the open hole the same way that he had done, and then remove the seltzer bottle or glass tube whatever we used for it. Actually what we had was casings. In this case, we had two tubes in the gelatine and we had a movable inside one that telescoped, so then after we got it all set up, then we ran some warm water down the inside one and melted that out to remove it, and then we had two cased sections and an open hole in between. In that case, you take a pair of parallel boards and squeeze this. So you are handling the stress, so you start off here with a circular cylinder, but you squeeze this with a pair of boards, and your cylinder flexes like that. So your least stress then is this direction. Your greater stress is this direction. And so you've got it under control. Well, we did that. It worked beautifully. And some of the results are illustrated here. Here it is. There's your vertical fracture. We even stratified it. We put in layers of hard and soft gelatine, to see if the stratum would influence the demonstration. Here's the markings of the structure, you see, and the fracture was homogeneous.
Hubbert: Well, then we said, well, now, how about when the horizontal stress is under a greater stress? Then we could see that by wrapping a rubber around this, under tension, so you squeeze the thing on all sides. And when you do, that's what you get. Now, the reason for the mushroom shape is because this squeezes up and there's stress on the sides, and so these stresses are actually curvilinear. Now, in the stratified case, we did get these bedding plane fractures.
Hubbert: Yes. But corresponding to this.
Tape 19 side 2
Hubbert: We did quite a lot of these, a variety of these experiments. And then we also brought in these dike patterns. Here's a pattern of dikes out in Colorado.
Hubbert: No. These are called Spanish Peak Dikes. Spanish Peak is I guess right here (points). So you can see from the patterns, they're plainly symmetrical with respect to this origin, and here's a picture of some of these dikes, cutting vertically out across the sediments. So, we can produce horizontal structures, or we can produce vertical structures by producing the appropriate stress patterns. So the next part of this thing was that the people on in the Hague recommended that this paper be published. In fact we made a motion picture of this experiment.
Hubbert: Yes. Right. Wait a minute, there's one intermediate step there. Well, maybe it will come back. Anyhow, this paper was rewritten, I mean the original. It was a very big thing, so it was written down and rewritten down to a size suitable for presentation. And this demonstration...oh, what I was forgetting then was a spring meeting of production engineers in Houston. I don't think that was a general conference, though it may have been. Anyhow, we were going to give a presentation of this experimental work to that meeting, and so we had this thing all put together in the back room waiting, and we brought it in, with the light behind. The audience is on this side, and what they could see was the hole and everything. Then we started squeezing no, we had this and then we started pumping the slurry in, and all they see is something that looks kind of globular, because they're looking edgewise at this.
Hubbert: There's the hole coming, the casing coming down, and here's the open hole section. What they're seeing is something coming out like this. As far as I could tell, it's simply a kind of a globular mass. Then you remove the boards; there's a light behind this so they can see it all happening. We remove the boards and came around and they're looking at it as a knife edge. Well, that had a magical effect. It made Christians out of these people. And within a week they were sending in field confirmation.
Hubbert: One of them sent in an account of a fracture job out in West Texas. It was acidization job in limestone. What they had was an imprint of a rubber packer, soft rubber packer on the walls of the hole, before and after. Before there was no imprint, and after they had two vertical imprints going up like so. Somebody else called attention to a recent paper in West Virginia where somebody had fractured a gas well, and so they had pictures down the hole, kind of like looking down a shotgun barrel. No fracture. Then they fractured it. They did some more pictures down the same hole, and you could see the fractures going down the walls of the hole.
Hubbert: They weren't looking. (laughter). But this convinced them of the reality of the thing, and then they began to find confirmations. All right. So that was in the spring of this year. Then came the annual meeting of the AIME in Los Angeles; that's a big national meeting of the AIME, the petroleum engineerings part of it. We gave this paper at this meeting.
Hubbert: Along with the motion picture of the demonstration.
Hubbert: We couldn't do it with a big audience like that. They couldn't see it. But we had the motion picture of it. I think we've still got that somewhere.
Hubbert: I can't show it to you because I don't have a projector. But anyhow, up until this time, the petroleum industry was almost unanimous among engineers that these fractures were horizontal. That paper upset that thing, and the transition was fairly abrupt but not totally. But within a few years, practically all fractures reported in the literature were vertical, in these relaxed areas. California was different. You're dealing there where the greater stress is horizontal, and the immediate stress the greatest. You're dealing with a different situation. And we had a few cases in California where the results were ambiguous, and there was no cleancut case of whether you had a fracture or whether you didn't and so on. But leaving that out of it, as a separate problem, in the Gulf Coast and so on there was almost universal agreement that the fractures were vertical, except the shallow ducts, within the range of a few hundred feet or down to a thousand or so feet, fractures within that range were commonly horizontal. But greater than that, they were vertical.
Hubbert: Yes.
Hubbert: There were some of these Stanolind people, some of them had got a job somewhere else, with Conoco.
Hubbert: Yes. They were formerly Stanolind engineers, and got a job, but they got a patent. They did something kind of interesting. They did a laboratory experiment, I think backed by the Stanolind people, using cores. They pressured a sandstone core internally, and so they had two cases, cores cut perpendicular to the bedding and cores cut in various angles to the bedding. So in the case of the cores cut perpendicular to the bedding, when they pressured it up...wait a minute. No. The other thing was, using a penetrating fluid, just crude oil, will flow through the sand. In those cases, I believe they had bedding plane fractures. But if they used a non-penetrating fluid, say drilling mud, it wouldn't penetrate the sand. They got vertical fractures, natural fractures. So the two men,...
Hubbert: Yes. They had been Stanolind, I believe, and they got a job with Conoco, and they got a patent for a way of producing either vertical or horizontal fractures at will. If you used a penetrating fluid, you got a bedding plane fracture, and if you used a non-penetrating fluid, then you got a vertical fracture. So I wrote a discussion of that paper also, and refuted their whole approach because Willis and I in this paper maybe not here, certainly in the original company report did an elaborate analysis of that question of the stress distribution with a penetrating fluid and without one. I don't remember whether it's in this or not.
Hubbert: Yes. We had done an elaborate analysis of this stress too. We also had done an experimental analysis of whether you could have a pressure less than the weight of the overburden to support the overburden. We had two cases. One was where the fluid flowed through the rock, the other was where it didn't. We did this one. We had again a fairly large cylinder and we had a screen, an open chamber, and we had sand.
Hubbert: Yes. And so we had an inlet here of water, and we had two cases. One was where you could put an impermiable membrane underneath this screen, and build up pressure, P until you had an overburden which is sand plus water. When you did that, the pressure P was exactly equal to the weight of the overburden. But suppose you didn't do that: you removed the membrane and let the water circulate through the sand. Now, in this case, water pressure is distributed throughout, the pressure is dropping as you go up and is distributed throughout in other words, you're lifting elements all over the place. You're lifting the whole works from the bottom. Here you're lifting it internally, not volumetrically. But nevertheless, and we worked out the theory of this in the experiment, that the pressure at the bottom is in exactly the same as it was before, although it's acting upon the sand in a totally different way.
Hubbert: You have the same pressure at the bottom to lift it. What happens is very interesting. This goes into floatation, into what they call a fluidized state. When it reaches that critical value where the water is just enough to support the overburden. Now, if you place a lead sinker, see, like a fishing sinker on top of this sand and sand is very solid it just rests there. Then it reaches a fluidized state, and the sinker drops to the bottom.
Hubbert: Well, you've unlocked the sand grains and this thing, it's literally kind of a fluid state.
So we did that, and we confirmed, and we worked out the theory of this flow-through cylinder with and without penetrating and non-penetrating and so on. Well, we couldn't go into this, the company report was much bigger than this. But now, one thing that happened at that meeting, the Stanolind man named Lubensky came up to me and protested the conclusions that we had drawn there, and that they had evidence and experiments that showed that we were entirely wrong. I said, well, I'd be very interested in seeing what it was. He said, "I can't show them to you. In fact, I've told you too much already." But they were going to publish their paper. They never did.
Hubbert: No.
Hubbert: I don't know. He was still with Stanolind the last I know of. That was 1962. In '62 he was still in Stanolind. At that time, there was a big to-do over depositing nuclear wastes. And there was the National Research Council Committee and so on, on this waste disposal problem.
Hubbert: And also there was an interest in this matter. Scripps and some of the other institutes were trying to get in on the contract.
Hubbert: Yes. We were having a big conference in Houston, two or three days duration. It was also the time they were about to dig this so-called Mohole well.
Hubbert: Which was finally killed. Lubensky was there, from Stanolind, at that meeting and that's the last time I saw him. But he was representing Stanolind at that conference.
Hubbert: I don't know, but I presume from his name, I presume that he was probably European.
Hubbert: But where, I don't know. Anyway, within a few years, practically all these reported cases of fractures where the results were known, I say, and except in very shallow depths, the fractures were vertical.
Hubbert: The Schlumberger company is one of the largest service companies in the petroleum industry. They're French originally and their American branch is quite large. They have a research laboratory up in Connecticut, and they had an in-house scientific publication. It was in print but it was not available to the public. And I was up there. I was invited up to give a lecture to their staff, and I don't know how the fracture thing came in. Maybe that was independent. I talked about the energy problem. And I think they came back to me then a year later, wanting to discuss this fracture problem, because they were organizing a conference or something on hydraulic fracturing. So they published this synposis of my earlier paper in their in-house scientific journal. Subsequently they opened that journal and it's now available to the public by subscription. Something like the Bell Telephone Laboraties' journal and so on. And they propositioned me again about a year or so ago, that they were putting out a large volume on hydraulic fracturing, and they wanted me to write one of the papers. I said I couldn't do that, I didn't have the time and I wouldn't have anything new other than what I had written in that first paper. So I suggested they use that paper, because that was the one that turned the thing upside down when it happened. It was the introductory paper to the whole volume. Well, they agreed to do so, but if that book has ever been published, I haven't seen it. They didn't pay me much for the use of the paper. In summary: that paper of Willis and mine completely reversed the petroleum industry, of almost the universal opinion that they were producing bedding plane fractures to vertical fractures.
Hubbert: So the summer of 1955, at the Amsterdam meeting this also coincided with the meeting in Rome of the World Petroleum Congress. I forget which one that was, number 5 or something or other. They had met about every four years, so they met in Rome in 1955.
Hubbert: I guess it was. I believe it was the first one I attended. And I only attended that because I was invited by the program committee to give a paper on fracture reservoirs. Well, I wasn't interested in the subject, but I did it mainly because it gave me a chance to go to the meeting. But I did it with no enthusiasm at all. In fact, that's when Dave Willis first came on, and we wrote this paper on fracture resevoirs together. That was before we did the job on the hydraulic fracturing. And so we got that off to the World Petroleum Congress, and so, in going to these, the time schedule was essentially the following. The conference was in Rome probably about the end of May. It ran for altogether about eight or nine days, a week and a couple of extra days. As I was about to take off for that meeting, our exploration manager happened to come by my office, and raised the question of whether I would like a trip to the Swiss Alps. I said yes. I didn't know it, but Shell had a more or less standard arrangement they'd been following for years of having a local Alpine geologist take visiting people like myself, either individually or in groups, on a Lake tour of the Alps.
Hubbert: Yes. They'd been doing this, I didn't know it, for quite a few years. So the answer to that was: Yes. So I went, my wife and I went over, and we went by boat. We went over on the Andrea Dore beautiful boat and we landed at the port north of Rome. It's the rather important principal seaport of Italy, north of Rome.
Hubbert: No, that's south of Rome.
Hubbert: Never mind, it's not relevant to the point. We landed there, and we had a company person, an Italian woman, who looked after us and got us into a hotel, took us out to dinner. Then the next day we took a train through northern Italy, and on that, we met this Swiss geologist who was to take us on this trip. His name was Tom Locher, and he has a Ph.D. from the University of Zrich.
Hubbert: Yes. And he'd grown up in the Alps, and had taken students on these various field trips all through the Alps. So he was intimately familiar with the whole area, all over the place. But he had this prescribed itinerary, to do so and so today and Friday and one thing Thursday and so on. He had this rental car with a driver who met us in southeastern Italy, and we worked our way up over the central Alps and finally down into the region of Lucerne. There was a big lake there, Vierwaldsttter See, and we took a boat. Here is this enormous face of limestone standing maybe three or four thousand feet high and nearly vertical. So we looked at it with binoculars. Well, it's the Nappes where the folds are very impressive. But I told my guide, I said, "Look, I want to get somewhere where I can get my hands on the outcrop, I would like to look at it up close to. I'm interested in secondary structures and fractures and all that kind of thing." Well, he was following the prescribed route, and I more or less countermanded that and said, "Look, I'm want to take an extra day and I'm want to to take a trip inland here where I can get my hands on the outcrop."
Tape 20 side 1
Hubbert: So we did. And without a map, it's difficult to convey very much meaning except to say that we made a one day circuit around the interior of the Alps. And one place that he took me was up a dead end road about 15 kilometers up into the mountains, to a little village up there called Elm, a farming village and so on. But from that Village, here was a skyline that came in, around, jagged peak up there, another one over here, like this, and you could see very plainly by different color, one of these was dark, the other was light, I forget which. That was one of those big thrust faults, and this is what was left of the upper block. They were now exposed in these remnants of the peaks. But the surface had inclined gently downward, and then about 10 or 15 kilometers it was down to road level, where you could put your hands on it, and we did. So there was a little cabin there, the guides had. It was a very famous locality. And there was a ledge about so high, for getting up it had a little ladder on it. Here, we got up there, and here's this polished surface of rock, and then there was a zone about so thick of crushed up material, fault material. Then above that were practically non-deformed sedimentaries horizontally, limestone, sandstone type thing.
Hubbert: Beneath that, about 30 degree angle was truncated series of beds, like so. That was the fault from there to here, and the beds above, I'd say the evidence from...the fossils was that they were Permian. The stuff below was Tertiary. Very much younger. So you had these older rocks thrust over on a gently inclined surface, over a truncated younger rocks below. The surface was polished. There was no perceptible significant deformation in the rocks either above or below, although there may have been if you'd examined it more carefully. But in the outcrop, there was no significant deformation, except this dip Tertiary may have been folded just within a small part of it. Anyhow, it was a very impressive sight, this combination. In fact that whole trip was the most important part of the entire trip.
Now, let's go back to what's known as hydraulic fracture problems, because we were also dealing with the fact that we had abornomal pressures. We had this lost circulation problem, and the lost circulation occurred because the pressure suddenly jumped up to well beyond normal. You know, in a matter of two or three hundred feet. And in looking up data in the Gulf Coast on these abnormal pressures, there were pressures that were going up to within a few percent of the fault in the overburden. Well, this carried with it mechanical implications, because suppose that you had a sand that was essentially an unconsolidated sand, a stratum thing. Then you had all this overburden on top of it. Now, Van Hise in Wisconsin way back in 1918 or 1920 or thereabouts wrote a paper in which he described an experiment I think he may be really quoting somebody else's data but did it over again of having a rubber balloon over sand, where the atmospheric pressure of the sand, you could squeeze this and it was loose sand, but if you evacuated it, that sand became concrete. Pull it together, the grains locked. So all right, utilizing the same principle, he said, "Suppose we have loose sand down there at 10,000 feet deep. We've got all this right on top of it. But there's also water pressure in the sand. Suppose we build that water pressure up enough to where it's supporting this overburden with the water pressure, and we're dealing with loose sand. If we reduce this pressure, the burden comes down, like that balloon. It locks the sand and it's enormously stronger." So that came up in our consideration of the fracturing problem.
Because among other things, we were considering the fact that we were dealing with pressures, water pressures of those magnitudes, we had to allow for that, with regard to what effect it would have on the rest of the stress. Well, with regard to that, we had Terzaghi's book on theoretical soil mechanics, which is a fairly general text within the preceding decades at least. Terzaghi had resolved these stresses into two parts. He had the total stress, and he resolved it into water pressure and the other part was the solid pressure over and above, stress over and above that. He called the water pressure the neutral stress. The solid stress over and above that was what he called the effective stress. And this neutral stress would have no shear stress. Well, that was a useful resolution. But another problem we ran into, shortly before this, was when I was about to set up the research laboratory there on rock mechanics. Staff hadn't yet been hired and I was scouting around to people to get their ideas. I visited the big rock laboratory in Denver in the Bureau of Reclamation who built these big dams, were building these big dams in the Rocky Mountains and so on.
Hubbert: They had set up a laboratory there for testing rocks in connection with these large dams. They were taking cores about six inches in diameter, a foot long, or something of that sort, of both rock and concrete. And they set this up in a big tracks or testing machine, to bear down with natural stress. Then they could also jacket these specimens, and have an internal fluid pressure, as well as external pressure. They can apply external pressure outside the jacket, and axial stress. Then the inside of it, they had a fluid pressure inside.
Hubbert: So they did them both ways, jacketed and unjacketed. Well, when I visited there, I met the man in charge I don't remember his name at the moment. He outlined these experiments that they'd been doing, and startled me. He handed me a copy, a preprint copy of the paper that he was giving shortly at the International Congress on Large Dams over in Copenhagen, I think. He had a whole theoretical analysis of a whole bunch of these experiments based on the assumption that the internal pressure acting in this rock was only acting over the pore spaces, and therefore if you passed a sinuous surface through these pore spaces in the grain contacts, suppose that the pore spaces were say .7, the grain contacts were .3, then you had this fluid pressure acting only over .7 of what it would be if it were 100 percent. One of the things that he was after in this paper was to determine the so-called surface porousity, and he went back to Terzaghi again. Well, he did a bunch of experiments with jackets, variable combinations of stress, external and internal pressure and stress. His unknown was this magnitude of the surface porousity, and it came to be 1.0 plus or minus some small fraction.
Hubbert: Yes. So within the limits of the experimental data it's 1.0. And that was everything he did. Well, I rebelled against that. In fact, when I was talking to him, I objected to the whole idea. I brought this paper back, restudied it, and it was so tight, it was consistent throughout, and I was having trouble getting my teeth into it. Then Willis came along and we were working together on this thing, and we finally smoked it out. And that was, just forget about the idea of surface porousity, and look at the internal and external stresses as if to make that resolution. It's your fluid pressure goes through the whole damned work, water and fluid, and then what's left over and it has no shear stress what's left over then is residual stress is what you call the effective stress. We showed that these experiments that we were dealing with were consistent. The man's name is in here.
Hubbert: Yes, it played no role. That's my counting.
Hubbert: Douglas McHenry, The Uplift Pressure on the Shearing Strength of Concrete, Troisieme Congres de Grand... [unclear], Stockholm, 1948.
Hubbert: Yes. Well, we've got some discussion of that in the text, with our figures. And what we found out was that if you took away the water pressure, and looked at all that was left over, it behaved exactly as if you were dealing with the same rock without fluid pressure. Just look at that second part of it. It had nothing to do with surface porousity.
Well, in the meantime, by this time we had the laboratory setup and work going on under John Handlin in our laboratory. John was unhappy about what Willis and I were doing with our McHenry experiments, data. And I said, "Look, it's the only data we've got, we'll have to accept it." Well, he thought Dave Griggs had gotten some different results or somewhere or other. I said, "I can't help it, John, these are the only data I've got, and I have to use them." And we did. So later when Rubey and I were working on the overthrust fault problem, this was a fundamental question.
Handlin came into my office one day with a sheepish grin on his face and said he had something he thought I'd be interested in, and so he handed me a run here,
Hubbert: On breia sandstone, and that's under a great variety of fluid pressures, and external, and far more accurate data than McHenry had.
Hubbert: So that was the clincher.
Hubbert: Well, I said, "Well, look, I need to use this. You'll get credit for it of course, it's your work, but we need it for this paper." And so he gave it to us.
Well, OK. There was that problem, and Willis and I worked it out and did it properly, just based on McHenry's paper. Going back again, we're now moving toward the overthrust fault problem.
Hubbert: We made this trip through the Alps, and looked at this big fault, I guess going through the mountain top and then coming down here, seeing the displacement there. There's a Swiss publication on the whole system of faults in this area, and every l5 or 10 kilometers or more a horizontal displacement. This is about maybe 50 kilometers. So it was very, very impressive. Well, these overthrust faults had been an enigma in geology for the last hundred years or more. The first recognition of them was by a Swiss nobleman whose estate was in this region. He had recognized these big overthrusts. Somebody or other van de Linth. And published it in 1840, about. The British Geological Survey in the 1880's turned their men loose and a two of their senior geologists on the Scottish Highlands, Northwest Highlands, and there's a whole bunch of thrusts in there, the so-called Morean [unclear] Thrusts, and they studied those, and they had displacements there of 8 or 10 miles or so in those Morean thrusts. Then they began to develop the study of these things in the Rocky Mountains, the Lewis overthrusts, and the Appalachian Mountains, a whole series of big overthrusts, the Pine Mountains and so on in the Appalachians. These things as real geological phenomena were just popping up all over the place, and being studied in terms of the displaced blocks commonly over ten miles. Various papers were published in which an attempt was made to explain them mechanically, how you could push a block of rock that distance, of that length, and move it, in view of the friction of rock on rock. Well, it was a paradox, because it just turned out that if you applied enough stress to well, you couldn't, if you had a 10 or 15 mile block, you couldn't do it. You applied it from the rear, it would shear off into a local thrust fault. You couldn't move the whole block. Or if you slid it down, it took a 30 degree slope.
Hubbert: No, it was just kind of an impasse. There were published papers, but mostly, nothing came out of it.
Hubbert: So, looking at this impressive thing, and marveling, as one must, it's a very impressive phenomenon. Well, in my hotel room that night I was thinking about this, and the question that arose, I wondered how much overburden there was above this fault? How deep was it in the ground when it was moved? Now it's way on top of the mountain.
Hubbert: Yes, at the time it was formed, it was probably several kilometers deep. Maybe ten or more.
Hubbert: Well, I was just trying to it had to be, after all, you had the fragments of this thing left. And just from the general arrangement of geology, there must have been an enormous amount of overburden when it moved. So, if that was the case, these rocks had to be full of water, and you wonder what the water pressure must have been going back to Willis.
Hubbert: And it was obvious, the water pressure must have been just about the maximum possible. These rocks are in one of the greatest orogenies known in the history of geology. They're full of water, and the water must have been raised to a pressure that was the maximum the rocks would stand. It would be enough to just about float the overburden. And then, that would have blown up your idea, my God. That's the answer to this whole mechanical problem of how can you move these blocks. You're not dealing with the whole weight of the overburden, rock on rock. You're dealing with this residual stress which approaches zero as the water pressure approaches the weight of the overburden. So you can move it and the corresponding force required to move it, you're not changing the friction, you're changing the normal stress across the interface. And if you reduce that normal stress, that which exists in the shear stress, it's only 30 percent of it, or it's 36 minimal angle, in the shear stress then as related to that normal stress. As the normal stress approaches zero, so does the shear stress.
Well, then from this trip I went on to, I think I spent a little bit of time in England. I don't know.
Hubbert: I went to this meeting in Amsterdam. I guess I went directly there from this field trip. And during that meeting, a note was handed me from Professor Vening Meinez who was in the Dutch's own estimation one of the most distingushed scientists. It was a personal note to me inviting me and my wife to have dinner with him at his home on a certain evening. He was a professor at Utrecht and he lived in a family house that had been built by his ancestors, I don't know how many generations ago, big thatched roof house in a wooded area.
Hubbert: A very, very lovely place. We got there maybe 4 in the afternoon or so. So he and I were having a little informal visit and miscellaneous conversation before dinner in his study, and somehow or other I mentioned this overthrust fault problem. His face lit up and he said, "Have you published that?" And I said, "No, I just thought of it a couple of weeks ago, when I was seeing these things in Switzerland." "Well, please do," he said, "I want to use it." Well, after this trip, I went over there on a vacation, then I headed back to Houston. I gave a report of this to my superiors, said I'd like to write this up as a scientific paper. They agreed. But I was busy with forty other things and hadn't got around to working on it yet. About that time, I got word from my long time friend Bill Rubey in the Geological Survey that if I was coming to Washington he'd like very much to have a chance to discuss a problem with me.
Hubbert: No. So I came to Washington, the next time I was there, not too far away, and he and his wife and I had dinner at the Cosmos Club. He spent the entire dinner hour quizzing me about the reality of these abnormal pressures in the Gulf Coast and elsewhere, and I said, "Yes, they're real." And finally he said, "The reason I'm interested in it is, I have an idea that may have something to do with thrust faults." (laughs)
Hubbert: I said, "Well, now, just a minute. I'm about to write a paper on that subject, and I don't want to be accused of plagiarism." We were both embarrassed. Well, he'd been working for years, he had 25 or 30 years of work on overthrust faults they were mapping in southwestern Wyoming.
Hubbert: He'd been working on this thing almost his whole professional career, 25 or 30 years or so. Working his way toward the notion that the water pressure might have something to do with this thing. Well, we were both embarrassed. It was kind of an awkward situation. And then I kind of outlined what my own thinking on this thing had been and how I had gotten where I was, and he briefly mentioned, he'd been toying with this thing. He'd asked Mr. Meinger, head of the ground water people, some questions about it, also had talked to Howard Gulf about the realities of these pressures, and so we kind of left it in that position.
Hubbert: That must have been about '56. 1955, '56. The meeting in Rome was '55. And I was going to write this paper, and had permission to do so. I hadn't started it yet. So it must have been about the spring of '56.
Hubbert: Willis and my paper hadn't been published yet by the AIME, but we'd given this big paper at the company conference in Amsterdam.
Hubbert: So I got back and I told my boss this thing that had happened, that in a sense was kind of embarrassing, but in another sense it was an opportunity, because here was this man who was a senior geologist with the Geological Survey who had been working for years on this overthrust fault block problem, and had broad knowledge of the geology of such things, not only there but elsewhere. On the other hand, I thought I had a better command of the mechanics of the situation than he did. And it looked to me like a ready-made situation for a very fruitful collaboration.
Hubbert: Not at no. No. Well, my boss agreed so I told Rubey that I wanted to see him the next time I was in town, and when I did, I propositioned him on this thing. I said, "It looks to me like a ready-made situation for collaboration." Well, he was a bit negative about it, he didn't think he knew enough. And I said, "Well, look, you've got this enormous range of experience, of first hand geological work on this, that I haven't got. I've got an extensive knowledge of the mechanics of the thing. So I think that if we pooled our knowledge, we might be able to do something very useful." So we finally worked it out, OK, let's consider this. And we wrote a joint paper under a covering title but in two parts. The overall title, we decided then or later, would be "The Role of Fluid Pressure in the Mechanics of Overthrust Faulting." And then part 1 was the theoretical part.
Tape 20 side 2
Hubbert: "Part 1: Mechanics of fluid-filled porous solids and its application to overthrust faulting." And this would be joint authoriship, but I'd be the senior author and he'd be the junior author. Then Part 2, he'd be the senior author, write the geological part, and he would be the senior author and I would be the junior author. So that's called "Overthrust Belt in Geosynclinal Area of Western Wyoming in Light of Fluid Pressure Hypothesis."
Hubbert: Right. And we also agreed, something that bears upon what you're doing now. It's regrettable that in scientific literature rarely do the authors ever tell you how they got into this situation. It's as if you built the house and took the scaffold away, and you didn't know whether they even used a scaffold. So we agreed that we would write the introduction of our respective parts, of how we personally got from here to there or there to here. I did my part and Rubey did his part. If you haven't read those already, you'll find they're useful historical summaries.
Hubbert: OK. So we started to work on these things, me in Houston and him in Washington, but he was kind of moving along at a leisurely pace. We had by this time talked this around among our friends and what not, so much that it was becoming public knowledge. If we weren't careful, somebody would publish it before we did. So I kind of pressured him that we'd better get going on this. the Geological Society of America was going to meet that fall in Atlantic City, and so I proposed that we have an oral presentation besides at that meeting, just to kind of consolidate our claim. And we did. I think it was it must have been about 1956, must have been about '57, the next year, the Geological Society meeting. And so we gave the paper at the Atlantic City meeting. Then we went seriously to work on the writing, and each one of us wrote parts and shipped them back and forth to each other for criticism and review. I was working on part 1, he on part 2, but we were exchanging manuscripts, for major criticism and review. Somewhere along there in the middle of this thing, I came down with some kind of an ailment, and I was just almost like I am now, I was just inching along there for a while. I told him I felt like I was in a state of slow motion. But nonetheless I was doing this work at home. I can do that better at home than in the office, so what I do is write at home, and bring the manuscript down to my secretary and have her type it. She could be typing it while I was working on the next part, and back and forth there, and then if we got something reasonably in hand, ship it on to Bill Rubey for comment. So that's how the actual writing was done. And the paper was finally published in '59, as I remember.
Hubbert: Yes.
Hubbert: When I went to New York, I guess. I used to come down here for the Geophysical Union meeting that happened in April every year.
Hubbert: In Washington. And so I met some of these people at that time, Survey people. Also, work I'd been doing on oil transferred with the Geological Survey in 1934. And I was in charge of the Survey partly in Indiana, Kentucky and Alabama. And that was of interest to the geologists, because what I was doing was something they didn't know anything about, and they were rather intrigued by it, and so they signed me, had me signed on at the Geological Society in Washington to give a special paper on this at their annual meeting the following year, 1935, I believe it was. 1941, '35. I have my dates mixed up here. Anyway, in '35 again in April I gave a paper at a meeting of the Geological Society in Washington on this work. So I had direct connection with the Survey people, being a member of the Survey at the time. Oh yes, then the International Geological Congress met in Washington in 1933, and I attended that.
Hubbert: Right, and also Jim Gilluly. Jim Gilluly was one of the organizers of this Congress, one of the principal representatives of the Geological Survey.
Hubbert: And I was very impressed with this guy, and we got to be very good friends. So maybe about that time, Bill Rubey and Jim Gilluly and a whole bunch of people were personal friends of each other, and Tom Nolan, and a member by the name of H.G. Ferguson, who was my party chief. I mean, he was a supervisor over me when I signed on to the Survey in '34. So I was an insider with the Survey at that time, and I used to come down, any time I was working on something, why, they were my principal critics and people that I could discuss and argue it out with.
Hubbert: Well, one of them was this ground water paper. I told you, we had this beer bust thing, around 1936 or so. At the time I was reading this controversy, you know, over the Ship Canal. And I came to the conclusion that the experts of the Geological Survey were all off base on that, and the one who was more nearly correct was the Army engineer who was a geologist formerly with the Geological Survey. He was far more nearly correct in his analysis than the experts in the Geological Survey were. And that led in turn to what's wrong, what have the Survey people done that's wrong? And so it was out of this that I was in turn developing myself this theory of ground water motion. But at the same time, I was under the impression that the petroleum engineers had the thing under control. Muskat's book came out about 1937, and I thought, well, now, this is the last word of authority. A treatise. Its general appearance.
Hubbert: This distinguished McGraw-Hill series on pure and applied physics or something of the sort. It had a lot of credentials, and it had the appearance of being a very authoritative work. So I drew the conclusion that the ground water people didn't know what they were talking about, but it's hardly to be expected because they were mostly geologists without much knowledge of physics or engineers also without very much knowledge of physics. And they were all kind of brought up internally under the influence of Meinger who was the head of the works.
So I didn't expect too much. There was a lot of physics they were talking about that they didn't understand. But I thought the petroleum engineers had it under control. So I didn't get around to writing that paper until 1939.
Hubbert: Well, he's a very very nice person. He's a gentleman of first order, with very broad interests and tastes. Very well informed. He is generally rated by his colleagues as one of the top men in geological theory. There's another item that I might mention. I'm trying to remember exactly the year. In 1956 was when I gave this paper before the American Petroleum, on the nuclear energy in fossil fuels.
Hubbert: That paper was given in March, and a preprint copy subject to revision was distributed in the meeting, and I had a few copies over. Later on that summer, I was in some meeting or other in Denver. I believe the Geological Survey was having an in-house conference on a large class of phenomena like floods and a lot of earthquakes and what not, surface phenomena in geology, catatrophic type thing, and I was invited to attend that as an outsider. One of the papers given was by Bill Bradley, William H. Bradley, who was chief geologist of the Survey. He and I were staying at the same hotel, and I gave him a preprint copy of the paper I'd given at the AGI meeting. Well, he refused to fly planes and he travelled by train. He said he got a lot of work done in a Pullman.
Hubbert: In a compartment, quietly, on the trips. So I got a letter from him, a few weeks later saying that he had studied this paper on the way back from Denver to Washington, and was tremendously impressed by it, and it was the kind of thing that he would like to see the Geological Survey doing. Could he have 50 copies for distribution to his staff there, which is the geological division of the Survey. It was the kind of work he'd like to see done by his own people. Well, I wrote to him and said that this was only a preliminary, subject-to-revision, draft, and that I couldn't, I didn't feel like sending them to him until the final draft was OK'd. In due course, after this paper was published, I did send the 50 copies, and he did distribute it. About the same time, maybe that fall, or winter, the Geological Society of America met in New York, and I was there. An emissary from Bill Bradley propositioned me about coming with the Survey. And what I was offered was, that the Survey had two slots, and they were sure of appointments for super grade 18 appointments, and Bill Rubey was going to have one and they wanted me to have the other one.
Hubbert: The rank would be a senior scientist with no assigned duties.
Hubbert: Well, I would have been enormously impressed with it, except that the Survey pay at that time was less than half of what I was getting, and I was within seven years of retirement with a permanent pension.
Hubbert: And financially, I couldn't afford it. So, I'd have given my right arm to have taken the job, otherwise, but financially I couldn't do it.
Hubbert: Yes.
Hubbert: That was mostly on paper. I didn't have any contact with the people personally.
Hubbert: To back up a little further on that: while the paper was being written, I gave some lectures at MIT, and just one incidentally was on this subject. The other ones I think were on energy problems. And I just stated in the lecture, as far as the resolution of the stress and fluid pressure, well I threw the whole works, solid and fluids, and the residual one. I think following this, one of the people present had been one of the graduate students in soil mechanics, and he had reported back to his department as to what I had said. So I had an appointment the following day with the chairman of the department. I don't know whether it was the soil mechanics department but I think it was. And also present was a visiting professor from Norway, University of Oslo. Well, the Norwegian immediately lit into me, that he'd heard what I said and he didn't agree with it at all and so on. And I wasn't even quite sure what he was talking about. But it emerged that it was this business of the surface porousity thing all over again, and that this was one of the fundamental principles of soil mechanics. I said, well, I'd read a little about that, but I didn't know that anyone, I was surprised that anyone took it seriously any longer. It was just an idea that had a brief prevalence, I thought. "Oh no," he said, "it's a fundamental principle of soil mechanics." And I said, "Well, could you cite me some key literature on the subject? I'd like to read it." So the very first thing he cited me was the Terzaghi paper back in early 1930's I believe, and then the later Terzaghi paper and then one or two others in Britain.
Well, I went back. Part of these I'd read already. I was familiar with Terzaghi's paper and of course the other one was this McHenry. And then another paper too that I wasn't familiar with. And I studied these things, and I told Rubey, I said, "Look, we've got to face this thing squarely." And so we went to work on it. Also, there was Hoover Mackin.
Hubbert: Yes. Hoover Mackin called our attention to an episode in the Society of Civil Engineers, and he sent us his personal copy of the Transactions. It was something like 40 or 50 pages. One of the members had written a paper in which he'd had the temerity to quarrel with the veracity of this surface porousity thing in the uplift in dams. Some incredible number of discussions by members of the Civil Engineers just taking this guy apart.
Hubbert: Well, the man was from Norway, and also some papers were from the Geological Society of Britain. So, this thing just had to be faced up to, and so we took it apart, completed it, with this confirmation by the experiments of Handin and so on, as well as McHenry's and we sent in the paper. Well, after this paper was published, then you began to get the reaction, in print this man Laubscher, for one. He's a Swiss. One interesting thing. One of my colleagues in Shell was an engineer and he visited at the University of Illinois, and visited a professor of soil mechanics at the University of Illinois, and outlined to the man what Rubey and I had done. And I think he wrote a letter, of which I was given a copy, saying that he had looked up this Hubbert and Rubey thing, and he thought that these authors were getting themselves in very serious trouble. But since they laid down the gauntlet, that he'd better give it further thought before he did anything. We never heard from him again. Also, I was told, Rubey got this somewhere, that some of Terzaghi's friends went to him and said, "Have you seen this thing that Hubbert and Rubey did?" Well, yes, he had. "Aren't you going to do anything about it?" and he said, "No."
Hubbert: No. He apparently had ideas along this line in the past. But evidently, I don't know whether he said so, but he'd become less sure of himself since. And he wasn't going to tangle with it.
Hubbert: Well, I think that shows up in his writings. Indirectly.
Hubbert: Yes. Because if you take his later papers, and he's being far less explicit and positive about this thing than he was in the earlier ones. What I think was that he began to have personal doubts over the whole business.
OK. Well, Laubscher just came in like a bull in a china shop; all we could do was analyze what he'd said and write a reply. Another man, Moore, was a soil mechanics man at the University of Texas, and he had a colleague who was working part time for Shell in the summer time, and I discussed this with the colleague. He'd gone back to discuss it with his superior back at Texas. So this guy comes barging out also like a bull in a china shop, and we took him on. And Francis Birch came along from Harvard.
Hubbert: Yes. I don't remember now exactly. But it had to do with the shear strength and the leading edge of this fracture and so on. We had pointed out that before the fracture occurred, there was no slippage, and the fracture had to have a leading edge, but the stress around this leading edge was so concentrated that it amounted to very little force. You had intense stress but not much force. And in back of that, you were back of these great blocks, and so that this to as we called it, which was the basic strength, wasn't involved in after the fracture had formed, it wasn't involved ahead of the fracture, and you had a great stress concentration on the leading edge, and that leading edge was propagated essentially as a displacement.
Well, we went through that. Later, a Chinese boy who worked in our lab and finally wound up as a professor at the University at Zurich and he wrote a paper on this thing. I wrote a rather general reply to this, but it was along the same line, about this basic to. Well, actually, I did more work on those damned dicscussions than I did on the original paper.
Hubbert: But I don't think I came off second best.
Hubbert: I don't know, because I don't read SOIL MECHANICS LETTERS so they may have been there, I don't know. But I do know that this man from Britain who was apparently a complete advocate of this surface porosity thing, after we'd criticized him, he gave a paper, I think, at some soil mechanics meeting in Denver or somewhere, in which he kind of backed out of the whole thing, and pretended that he'd said so all along.
Hubbert: Actually it was the most productive period in my life, because of what I had. I had the resources of the Shell Oil company, the stenographic; I had a technical assistant during a good deal of that period, David Willis part time, a woman, Martha Churley, for about about seven years, and a secretary. So I was pretty much well, after I got out of the research I had administration, which took a lot of time.
Hubbert: Oh, probably two-thirds. So that any scientific work I did in that period was just practically bootlegged.
Tape 21 side 1
Hubbert: Well, once I got out of that, with this appointment as a consultant with no well-defined duties, I mean, like the fracture problem well, that was a natural, because we'd done the basic work on that. I was just simply assigned the job of looking into the thing and deciding or advising as to what goes on here.
Hubbert: Not very much. There weren't very many people.
Hubbert: Well, this group, most of these people weren't called consultants. There was a group set up. I gave it a title, I'm not sure I remember it now.
Hubbert: Special services.
Hubbert: It was called Special Services. We were available as a group to help out on this, that or the other problem that came along, principally in production.
Hubbert: But then I was with them four or five years, and then I requested to be transferred back to the research laboratory.
Hubbert: Because it was a better place to work. It had the facilities, a library and everything else, which I didn't have in this other place. I mean, I was in the headquarters office of the company, but I didn't have facilities.
Hubbert: So I wanted to get back, and so I did. I went back. I had my own setup. I had my personal office, roughly from the size of this area back here, and then there was a secretarial area in between and then there was my assistant's office down the way, and that was a self-contained little suite of offices. We did our own work within those confines.
Hubbert: And then I showed you these pictures. This is an aside, but anyhow, it was a new building, luxurious quarters.
Hubbert: They were one of the principal developers of the seismic instruments in the petroleum industry.
Hubbert: No.
Hubbert: No, I don't think that's quite what I said, or intended to say. I think that our research was much more fundamental than that of most other companies.
Hubbert: As I remember the figure, in our budget something like 30 percent of it was for fundamental issues. Most companies, it's closer to 5 percent. With most of them, they were concerned with directing the research for a particular utilitarian result.
Hubbert: Probably not. After all, an awful lot of revolution went on both in the company and the petroleum industry during that period. One thing about Shell that was distinctive, Shell officials were mostly people who had a very high level technical education, PhD's in geology, physics, chemistry, or major training in engineering, and the company officials when in the early stages were like a business aristoracy. They were born to the job type of thing. That very fact gave them an objectivity that you didn't have in the "office boy to president" regime. So by the time I came along, they'd been through a war and they'd been occupied by the Germans and had a hell of a rough time. In fact, in their technical forces on their staff, they had a geological group that they had working underground in the coal mines, mapping things like faults and fractures, just to keep them out of the Germans' slave labor camps that type of thing. They did a certain amount of geophysical work, gravity work, under German pressure in Holland, but they did the damnedest sabotages so they didn't find any oil. They didn't succeed because the Germans didn't find any oil, and this was just immediately after the war. Now, this is very interesting. I only visited the headquarters office about three times, once in 1946, and in 1955 and again in 1960. And I knew several of the board of managing directors, which is the highest level of management in the International Shell group. I've known some of these. By my time, they hired men out of universities, with top flight training, and then they had an internal training and education. Then they put them out to man the empire all over the place. In the jungles of Sumatra, and Borneo and etc. around the world. The war came on, and many of these people in the East Indies were interned by the Dutch and were in prison during the duration of the war. The people in Holland were thoroughly incidentally, they burned tons of laboratory reports, research reports.
Hubbert: To keep the Germans from getting them. They had to build a special furnace to burn the damn things. Paper doesn't burn very good, in compact form. And they had a kind of a management in absentia, in England, but a lot of their top people stayed back in Holland. Well, I met these people as they came over for official visits to our laboratory, and got to be well acquainted with them, like this man Van Dyke. I first met him the first time I was over there on a trip to Amsterdam, and when he'd come to Houston he'd almost invariably have dinner with Miriam and me. And when we were over there, why, we'd have dinner with him and his family. That type of thing. There was that kind of an interchange on a personal level. One man that I met in South America, in this hellish tropical climate down there. I was in the Shell offices, and was introduced to a man who was working at a drafting board, a kind of a sandy-haired red-faced man of about maybe 50. I sort of sized him up as an Irishman from the West Coast, and was surprised to discover he was a Dutchman. Well, a few years later he was one of the managing directors of the company. Here he was, working away on a technical job out in the oil fields when I met him. So they had a lot of those kinds of things. They hired a top level man, and they started him in pretty much at the bottom, and they kind of grew.
Hubbert: Another thing was that I had a reputation among these people of being what you might call a distinguished scientist, and this thing showed up very conspicuously when I was there in about 1960. There was a vice president from Western Canada whom I knew slightly, who was about 35 years old. He was over at headquarter's office for his first visit at the time I came through. I'd been to the International Geological Congress at Copenhagen and in Norway. And this man was temporarily a visitor from the outside, and he was very awed by this environment he was in. He looked and acted like an office boy. And they treated him like an office boy. They were entertaining him by buying him tickets to the girlie shows, while I was being entertained by the top people of the company.
Hubbert: I was simply received as a distinguished geologist. Another experience that I had, that same meeting, I guess, in Amsterdam. I met one of their scientific staff there, whom I hadn't met before, a man 35 or 40 years old. When I was introduced to him, he said, "Are you the phenomenal King Hubbert?" I said, "Well, I don't know about that. I'm King Hubbert."
Hubbert: Well, I can't answer that question because I didn't know their laboratories well enough. But as I say, on the basis of this intercompany comparison that I was involved in and knew about, we were well ahead in fundamental research. As far as the basic equipment was concerned, why, I'd have to say that the others were as well equipped as we were. I think maybe the educational level of our staff was higher than most of them. We had practically an equilateral exchange with the universities. Our staff could leave us and take top professorships in universities. It was that kind of an exchange. Most of these people couldn't do that.
Hubbert: They probably did. I don't know the details. But we had quite a few Shell fellowships, and sent students to universities. We had good relations with the universities. We invited outstanding professors who shared our interests down for visits, on an honorarium basis. Both to know what we were doing, and for us to get acquainted with them. Well, you see, when we set up this lab, it was just the end of the war, and there was a great manpower shortage. [unintelligible] and I were running the thing at the time, and we agreed on a basic and fundamental proposition, and that is that the laboratory was never going to be any better than the quality of the men that we hired. And if we made too many mistakes and got too many second class men in, you couldn't get a good man in the door. Good men tend to attract each other, and vice versa.
Hubbert: We knew that. So a lot of people that we interviewed, who had the proper academic credentials, we made a job offer to not more than one in four or so.
Hubbert: Here they were, they were from top universities, and we didn't offer them a job, in a period of manpower shortage.
Hubbert: Yes. Take this rock mechanics lab. That was written into the research program almost immediately, within the first year. I spent about three years before I hired anybody for the job.
Hubbert: Interviewing men all over the place. And saying no. Again, if I made a mistake, and got the wrong man, that program was essentially stymied. So finally after about three years I got John Handin who was working for his degree at UCLA and was in high pressure work. What I wanted was a team. I wanted lab work. I wanted theoretical work. And I wanted field work, all in kind of a teamwork, integrated together. So once I got John Handin, and we took over a Dutch boy who had very good technical training from Amsterdam in the lab. We took him on as more or less part of the theoretical group, Elmer O'Day[?], and then John got some of his colleagues in from UCLA, and Dave Griggs helped.
Hubbert: Yes. So that was the idea. Get a nucleus, then they'd bring in, they'd pick their colleagues, and they did. And so this soon became one of the foremost labs of its kind in the West, and still is. The group is still bound together. Well, after that some of them scattered. John Pucker for example made the chairman of the department of [unclear] at Syracuse University and later became vice president for academic affairs at Syracuse. Another one of our men left and went with the Livermore Laboratory in California. Now we've got a whole bunch of them at Texas A and M University.
Hubbert: In their school of geosciences or whatever they call it. We had that kind of give and take, where well, I personally did a lot of lecturing in universities all around the country, by invitation. I was a visiting lecturer at UCLA in oh, around 1960 or so. Later on, well of course, this is after I left I was a visiting professor at Berkeley, which is the highest level of the University of California. And I was a visiting professor, later full professor with tenure at Stanford.
Hubbert: OK.
Hubbert: Well, a lot of things were going on in the general geological field. We weren't deeply involved with it. Our research didn't involve earthquake seismology, which is one of the principal sources of information there. But we were all aware of what was going on, and were interested in it.
Hubbert: Well, some. Oh yes, a couple of our staff members, two very bright boys, one was named Ken Deffeyes and he's now professor at Princeton. Another, Peter Wilde, stayed a while. He's a professor at StonyBrook, New York University at StonyBrook.
Hubbert: Well, they came on somewhere in the latter part of the mid-fifties or so.
Hubbert: Well, they were leaving along about that time. These two men Wilde and Deffeyes: Deffeyes was very highly regarded by the president at Princeton, and he wound up with a professorship there. Wilde went to the University of Washington in oceanography. He finally wound up where he is now at StonyBrook.
Hubbert: Well, we had men of that quality.
Hubbert: Oh, there probably are. I don't know what would be appropriate.
Hubbert: Well, I opened my mail one morning, and there was a letter from the president of AAAS. I just looked up his name, it's Sinnott, I believe. He was later the dean of science at Yale. He was president of the AAAS. And what the letter said was it was received along about March, 1948 he said that the AAAS was having its centennial meeting in Washington in September, and the date's given, and that the program committee had decided to build the program around 15 symposia on subjects deemed to be of worldwide scientific important. For each symposium, they were inviting three speakers. And there would be other discussion, but there were three invited speakers for each symposium. One of these symposia was to be on energy, energy resources. The three invited speakers were Farrington Daniels, for solar energy, Eugene Wigner for nuclear energy, and you for fossil fuels.
Hubbert: I know something about it now, but at the time it was just out of the blue.
Hubbert: Yes. Yes. So my suspicion was that since I had no publications in this field, it must have come from somebody first hand, personally. I later learned that it was Jim Gilluly.
Hubbert: Who was on the program too. Jim told me later on. He said that they had some very obvious names among petroleum geologists who in his opinion were quite unqualified for saying anything of any significance, and he recommended me. And another, a Geological Survey man well, I blank out on his name for mineral resources, another field. Well, all these things were credited, my experience, very informal. So in reply to the man's letter, I wrote about a three or four page single space letter in which I practically wrote the paper in reply, what I proposed to do. The man incidentally was a biologist and ecologist, among other things. And so then I went ahead and did the investigation, the leg work had to be done, and wrote the paper. Again, it was put together in that very neat format. I have a sample copy of the type of thing.
Hubbbert: This is the AAAS meeting, and my colleagues, none of whom I'd met before. I didn't know Farrington Daniels or Wigner. But they were very very pleased with my paper, and so were a lot of other people. Well, immediately following that meeting, the Smithsonian, which I didn't know it but they published an annual volume of selected papers published during the preceding year. My paper was one of those in the Smithsonian REPORT for that year. And then SCIENCE published a slightly abbreviated version of it in January, 1949, the next year. The meeting was September, 1948. That was the first time I was ever swamped with mail for a published paper.
Hubbert: I must have had several hundred letters.
Hubbert: Just about 100 percent favorable. And overwhelmingly from biologists.
Hubbert: Well, the reason for that may be that at that time, SCIENCE was heavily, well, their papers were heavily biological.
Hubbert: So their principal readership was biological. That probably would account for that bias in the correspondence. But the reaction was very very good, and I showed you this letter, here, a couple of weeks ago, that I received from Gary Hardin, who's a senior emeritus biologist from University of California, Santa Barbara. He really liked that paper, he said, and was still impressed with it. But that paper was really a condensation of a thing that I'd been doing, prior work that I'd been doing for 15 or 20 years.
Hubbert: Yes. So outside of just the small details, as I say, I practically wrote the paper in my reply to Sinnott's letter. I wrote him a four page single space letter in which I said here and here and here is what I propose to do.
Hubbert: Yes, I had to have the legwork done, I had to dig up a lot of stuff. The other thing was that since it was an international meeting, I decided to use metric units, instead of using barrels and so on. I was using cubic meters, and that meant that there the data didn't have to be recomputed over, to be done in metric units.
Hubbert: It probably did with a lot of people, because if they weren't involved in it, they didn't know what I was talking about. I didn't get any negative reaction from the audience, maybe on account of I talked in familiar language.
Tape 21 side 2
Hubbert: ... but I got a response from two people, and this was fortunate. One of them was Wallace Pratt who was either still or just retired. I think he was vice president of Standard New Jersey. And the other was L. G. Weeks, who was the senior geologist of Standard of New Jersey, and had been devoting all his time to a world evaluation of oil resources.
Hubbert: I'd met him before. But Weeks I knew somewhat better. Well, at this meeting the following year, the United Nations was holding a three weeks' conference at Lake Success.
Hubbert: Not much. Wigner couldn't talk about he had to talk in generalities, everything as to nuclear energy was still top secret. Daniels, I didn't have a chance to talk with him very much. He was running a laboratory in Wisconsin on photosynthesis. But the PHYSICS TODAY, I believe, ran an article, kind of a synthesis of the three papers, under our triple authorship.
Hubbert: The following year.
Hubbert: He couldn't discuss nuclear energy. There was no opportunity when I saw him at that meeting, but even so he wouldn't have been able to talk.
Hubbert: Yes.
Hubbert: Somebody in the interior department knew who I was and invited me. I was invited by the Secretary of the Interior. Somebody on his staff had recommended me, that I be one of the representatives of the US government.
Hubbert: Of the meeting?
Hubbert: I wasn't on the program, so I was really there as a listener, and a discussant. But in the petroleum end of it, what they did was, they broke up, half a dozen or so parallel sessions on different subjects going on at the same time. Then about 4 o'clock in the afternoon, they all came together in the big auditorium for a plenary session, where there were one or more invited lecturers.
Hubbert: Yes. So one of the invited lecturers on petroleum was E.I. Leverson of Stanford. Leverson was an oil geologist, prominent oil geologist before he took this job at Stanford. In fact I'd visited him at Stanford a few years before, shortly after he was there. And I knew him fairly well. The day he was to give that lecture, I think he and I had lunch together. I was perfectly relaxed as to what he would do, because he was a certainly a well-informed oil geologist and when it came time for the meeting, I wasn't even down on the floor where the delegates were. I was up visiting a friend in the visitors' gallery. Leverson gave this invited address, and he reviewed some of the top evidence that you used in estimating oil and gas. It got along to where the estimate on the ground oil was about the best available was around 15 hundred billion barrels. And then he took off. He started cited how worried they were back in 1920. So he got off on that. In 1925 the estimate they estimated there'd be only so much left, by 1930 they'd already passed this, and so on. So then he got through that sequence, and he said, "Now, this figure of 1600 billion barrels, or 14, 15 hundred billion barrels, is only our current estimate, and that's probably only a small fraction of what we'll estimate at," he specified some future time, but not too far hence. Then he finally wound up with a grand slam, that the world could be assured of enough oil to meet all of its needs for the next 500 years.
Hubbert: Well, I nearly fell out of my seat. I was up here, relaxed, visiting with my friend, and good God almighty! And nobody said Boo. The next paper was a Frenchman talking on a mining subject in French. Well, I worked my way down to the delegates' floor, and determined that I was going to challenge him, if I could get the floor when the Frenchman got through. And I figured, whatever I said would probably be gibberish because I was nervous and jittery. But at least I'd get my toe in the door and I could revamp out my remarks for the record later on. So I called for the floor when the Frenchman got through, and I said this figure of 1500 billion barrels that Leverson had was about as good as present evidence would indicate, a very good figure, but that I was very concerned about his final statement, and I pointed out, I said, the overall history of petroleum production has to begin at zero, it has to go over a peak and finally come back, and the area beneath that curve is cumulative production. If you start plotting a curve for the last four or five hundred years, and start adding up, it's just an utterly preposterous amount of oil, there's no evidence whatever of that set figure. Well, L.J. Weeks got up and said that this figure that Leverson had used had been published in or cited in a paper by Chase Manhattan Bank. Well, Chase Manhattan Bank and Standard of New Jersey are pretty close associates. And he said they were actually his figures; he'd given them at a conference with the bank and the bank people had used them in one of their publications. He went on to say, as he'd said in print before, that his figures were he gave these margins, I'm not sure I'm quoting him correctly but probably the real figure might be as high as 50 percent higher than his estimate, or maybe 10 percent lower. But he thought that it was within that ballpark.
Hubbert: Yes. And with regard to onshore and offshore, he hadn't given an offshore estimate before, but he estimated the onshore I believe at about 600 billion barrels and offshore about another 400. At the present time, and that added up to about a thousand. But again with these margins, whatever. Well, Leverson was decidedly embarrassed and irritated over this thing. What came out of that was that it set up such a ruckus among the oil people there, the people who were interested in oil, including the Geological Survey and so on, that a day or two later they had an off session, that wasn't even on the program, on the matter of oil estimations.
Hubbert: It doesn't show up in the TRANSACTIONS but they did have such a meeting.
Hubbert: Well, it was a world wide audience. It included the director of the Geological Survey of Great Britain and the US Geological Survey and the Bureau of Mines people and some oil people and academics.
Hubbert: Yes. Of course I did. We had one man there, Ira Cram, who came out very strongly in support of Leverson, but later on changed his mind.
Hubbert: He was with Continental oil company, chief geologist, I believe. And then around 1960 or so, the AAPG got out a big two volume work of a re-estimation of the oil, at the request of the National Petroleum Council. Ira Cram was the editor of that.
Hubbert: The director of the Geological Survey of Britain had some things to say about Middle East oil but he couldn't really say anything because it was all secret. And this was before the right figures on Middle East were generally publicly known.
Hubbert: I guess about 50, 40 or 50.
Hubbert: I think it was a luncheon meeting, I'd say between one or two hours maybe.
Hubbert: Oh, it was just a kind of a straight up concern over differences of opinion over this whole question of oil. But at least it broke the ice on this thing. If I hadn't done it, why, I figure Leverson would have gone on the record without any discussion, as the, you might say, official opinion of the meeting. I broke it up.
Hubbert: Yes.
Hubbert: Maybe we'd better quit.