Nuclear Museum Logo
Nuclear Museum Logo

National Museum of Nuclear Science & History

General Leslie Groves’s Interview – Part 4

In this interview, General Groves talks about his responsibilities as the director of the Manhattan Project as well as the responsibilities of his subordinates, including Colonel Kenneth D. Nichols and General Thomas Farrell. Groves also discusses the relationship that he had with Vannevar Bush and James B. Conant and their role in the Project as administrators and science advisors.

Date of Interview:
June 22, 1965
Location of the Interview:


[We would like to thank Robert S. Norris, author of the definitive biography of General Leslie R. Groves, Racing for the Bomb: General Leslie R. Groves, the Manhattan Project’s Indispensable Man, for taking the time to read over these transcripts for misspellings and other errors.]

Stephane Groueff: Now, I wanted to ask you one thing that is not very clear for me: How closely did President Roosevelt follow the progress of the project?  How often would he inquire and through what channels?

General Groves: I do not know whether Roosevelt ever inquired. But Bush saw him from time to time and kept him posted. Bush also talked to Harry Hopkins at times. As far as I know, that was the only communication with the president.

Groueff: Harry Hopkins knew everything? 

Groves: Apparently he did. I do not know how much he knew until fairly recently. He knew what any outsider would know. He did not know any of the details. All he knew was they are getting along and they are building plants. He might have been told there was an electromagnetic plant and a gas diffusion plant at Oak Ridge, but I doubt if he could remember the names. 

Groueff: I see. But you did not feel any kind of urgent or pressure coming from the White House?

Groves: No, nor did I find any coming from the War Department either. They all knew it was a big job. They all knew that it was to take time to do it. I think that the attitude of the War Department was influenced a great deal by the original comments by Somervell when I was asked to take charge or told to take charge. That was reinforced by the natural disposition of Marshall, who did not interfere with subordinates. Then it was strengthened a great deal by my conduct at our one meeting in which I said, “I have got to get out of here if you are through talking, I want to get to Oak Ridge.”

Well, the idea of anybody moving that fast was just not understood by them. I know that from what Somervell told me afterwards. He said, “That was the smartest thing anybody ever did. You just made me look like a million dollars, because I told him what you would do if you took charge of it.” The minute you get through the original conversation and before waiting for a proper protocol, or anything like that. You just said well, if I do not leave now, I will miss my train, and I want to be in Oak Ridge tomorrow morning.

Groueff: He was the president, too?

Groves: He was president to me.

Groueff: That was Styer?

Groves: Styer was also.

Groueff: Well, among your superiors—President Roosevelt, the vice president, Stimson, and General Marshall—who was the person who was the most [00:03:00] active or who was in direct challenge of being in touch with you and following?

Groves: Nobody.

Groueff: Nobody?

Groves: No. They depended entirely, I think, on the Military Policy Committee to tell them if it was not going well.

Groueff: That was the three officers?

Groves: No, that was Bush—Conant as his alternate—Styer, and Purnell. That is what they depended on. Also, they had been very definitely told by Somervell, I think beforehand, that if I took it over, they would expect to see me run it up. They might have some difficulties. It is like what Styer told Bush, he said, “He may be hard to get along with at times, but he will get it done. That’s what we want, isn’t it?”

Groueff: There was no need to be impatient, or to push?

Groves: Well, there was a need to. I have often said that if I had been in General Marshall’s place, I would have required reports from me.

Groueff: But he did not?

Groves: But he did not.

Groueff: How often would you report to him?

Groves: I do not know the exact dates, but the first time was when I took the report to the president from the Military Policy Committee. That would be early in December of ’42. The next report was at August of ’43. I did not take that over.

Groueff: For several, eight months, for eight or nine months—

Groves: No reports.

Groueff: No reports? And he did not ask through some someone?

Groves: No. Now, I may have taken an occasional memorandum over to him where I wanted him to do something like a letter of introduction to someone going overseas. But I do not think that occurred at that time. I think that all came later.

Groueff: Didn’t the Policy Committee report to him [Roosevelt] more often?

Groves: No. All he knew was that I had been assigned to it. He assumed that I was doing the best I could. That was all there was to it. 

Groueff: But that sounds surprising to me because after all, the atomic bomb would be a very major element in the whole strategy.

Groves: Yes. That is why I say that I felt that it was a mistake on his part. But it is very convenient from my standpoint. I saw no reason to go over and tell him that we were not doing well, which is all that I could have told him at that time. That we could not get the barrier developed. He did not want to know anything about my problems with the barrier. That was my problem to settle. It worked out very well. But as I said, if I had not produced, it would have been a terrible affair. I do not think that anyone in his position should have allowed me to go freewheeling as it were. Now, I did see an assistant to the secretary, Harvey Bundy, the father of McGeorge, from time to time on various matters, and he undoubtedly told Stimson.

Groueff: You did not see Stimson?

Groves: No, but I would not have told him anything about what really counted on this thing. I was always after something. I needed help to get somebody or something.

Groueff: That is not the report on the state of this?

Groves: No, there is no report. The only reason the August report was rendered was because I heard suddenly and without any warning about the Quebec Conference. I decided that if Roosevelt was going to be here and Churchill was going to be there, why we might get stung badly, if I did not get a report out. Particularly, because they were going to discuss the Quebec Agreement or what became Quebec Agreement.

For that reason, I remember calling Nichols in a hurry and telling him to get up to Washington quickly and that I had a report to write. I started it right away. He got up there, I imagine by air. By that time air was a little bit better than it had been. We used it coming this way. We did not use it going down because going down all you lost was the night train—you did not mind. After it was finished, I had made arrangements to get it up to General Marshall through the Secretary of the General Staff.

As soon as it was finished, I think it was seen by Bush, Conant, Purnell, and Styer by just taking it around by hand. Then Nichols took it up and gave it personally to General Marshall. Marshall took it and handed it to his secretary of the general staff and said, “Put it in the safe. I am afraid it is too late.” It was, and that was the second one. Then there were, I think, two more written reports.

Groueff: That is all?

Groves: That is all.

Groueff: In other words, from that I assume that 1942, the atomic bomb was not an element of the present strategy of the war?

Groves: No. The original idea of General Marshall was, I think, that it was to be something that we would fight the war without. If it came, then we would be prepared to use it. That was the attitude, whether he had it or not. It is the attitude I took immediately and from right straight through.

The first time that it was ever suggested that it should have a bearing on the strategy of the war was the assurance in my opinion that if it worked, it would end the war and end it in August (if it worked to the full power). If it did not work to the full power, then we did not know when it would end it.

But Harvey Bundy talked to me one day—I guess it was the time of Okinawa when the casualty list started to come in. Maybe it was before Okinawa or before Iwo Jima, but he asked me if we should hold up these offensives in order to use and wait for the bomb.

I was very strongly against that. The expression that I used was that the opponent is reeling on the ropes. If you are a boxer, you try to finish him. You do not give him any respite. I said if you pull back and do not do anything for three or four months, he can get strong again. We do not want him to get strong again. This bomb may not work.

Groueff: Yes. You could not count on the bomb.

Groves: You should not count on it as a sure thing.

Groueff: That was the official policy?

Groves: That policy was accepted by Marshall and by Stimson and by Admiral King and Arnold.

Groueff: Of the two men, Marshall and Stimson, who was closer to the project?

Groves: Stimson was.

Groueff: Stimson.

Groves: Yes, but not originally. I would say because I did not have any dealings with Stimson until the fall of ’44. That is, I do not think I even saw him. I would have to search the diary to see. But I do not know just when it was. But in any case, I had started seeing him in the fall of ’44. Then by April of ’45, I was seeing him all of the time. I was, in his eyes, the most important person in the War Department because he knew that Marshall would run the war alright on that end but this was the thing that he was looking forward to.

Groueff: But for two years, you would not see either Stimson, or Marshall, or the president?

Groves: No, I only saw the president once, that is Roosevelt.

Groueff: All of your contacts were with the Policy Committee?

Groves: Yes, but those were not contacts in the sense of supervision. In the Policy Committee minutes, which I kept and wrote up later, they started off with maybe that much on decisions reached at the meeting. Then gradually a new item came in called “General Groves reported the following decisions.” Then gradually what I had decided and reported on came up there and sort of squeezed the decision out of it so that the last few had almost no decisions reached at the meeting.

Groueff: I think I should correct then the popular belief that America was preparing in secret the new weapon with which to finish the bomb [misspoke: war]. The question was who were the people who knew it and who prepared it. The answer was the president, and Stimson, and Marshall. You can imagine those people just impatiently waiting and practically calling every day, “Is it ready or not to drop it?” And when it is ready, they will drop it and the war will be finished. Now, I see that and I learned that for about two years, they did not inquire or you did not report it.

Groves: No, remember that Bush was talking to the president from time to time and undoubtedly every time he saw the president, he told him about this. But he did not go to see the president for this particularly excepting in connection with our negotiations on the Quebec Agreement. 

Now, the Quebec Agreement was discussed between Bundy and me a great deal, I am sure. The Quebec Agreement was really the product of Bush. Now, originally Bush saw a great deal of Stimson and of Marshall, but later on and towards the end of the war—this you cannot put in—they [Stimson and Marshall] got tired of spending so much time with him [Bush].

Groueff: He talked too much?

Groves: He talked a great deal. He always had ideas of how they should run the war. They were always very polite, but they tried to dodge him. That is quite obvious.

Groueff: Conant would not go directly?

Groves: No, Conant always went through Bush unless Bush was out of town. Conant, of course, became one of my scientific advisors along with Tolman. That was done at Conant’s suggestion; I did not ask him, he asked me. He said he thought it would be a good idea.

Now, why he did it, I do not know. But I think he is completely above board in saying that he thought that it would ease my difficulties with Los Alamos, particularly, because they would feel that he as a scientist—although he was a retired scientist—got along well with those people. They did not look on him with contempt as Szilard did.

Groueff: And a Harvard president had a certain prestige.

Groves: Yes. They all were interested in the possibility of being invited to be professors at Harvard after the war. We did not go into that, but I knew it just as well as he did. He thought also that Tolman would be a good means of getting in under Oppenheimer’s skin because Tolman had been very close to Oppenheimer for years. He had been an older mentor whose brainchild came up fine. Tolman is the graduate dean at Cal-Tech and Oppie taught down there half of the time and half of the time at Berkeley.

Groueff: He respected him.

Groves: He respected him and you might say looked on him as an elder statesman. He did not think that Tolman had as good a brain as he did. By that time, he was confident of his own ability. Tolman could talk to him, if we needed somebody to.

Groueff: He had a good judgment?

Groves: He had very good judgment. He was not anywhere near as good as Conant.

Groueff: Conant was your real advisor when you needed it?

Groves: Well, one thing that these people were used for was not so much as advisors but, for example, if we had trouble with a chemical process anywhere, I would ask Dr. Conant to look into it. Conant might go and spend three or four days at Oak Ridge in one little section of the electromagnetic plant. Well, I could not afford to do that. Being a chemist, he could – and all of their language, he was familiar with. It was very straightforward for him. Everything of that kind, you see, he could do.

The same way, Tolman could sit in on a meeting at Los Alamos for two or three days at a time. That just meant that these people thought they were getting right in with me, you see.

Groueff: Yeah, as if you were present. But did they work on the basis of full-time every day? They would come every day to you?

Groves: No. Conant had his office up with Bush. He did not have an office with me. He was running Harvard, too. He was not always in Washington.

Groueff: It was only when needed?

Groves: Yes, when needed and when he had free time.

Groueff: There was no such thing as somebody next to you?

Groves: No. Tolman had his office over in the National Academy of Sciences. He would come in at frequent intervals. Sometimes, he would spend the whole day there. I did not have to go to see them as a matter of protocol; they always came in there because they knew how busy I was.

Groueff: And during the period of the actual making of the bomb, your major assistant then was Nichols? And Conant was a major sort of advisor?

Groves: You might say this: first, it was Marshall who was the major assistant, you might say, as the district engineer. When he was relieved and Nichols took his place, then Nichols was. Then at the end along about January of ’45, when Stimson made me get an assistant—which he should have done long before and I should have done without his saying anything about it—then we brought in General Farrell. It was arranged that Farrell would take my place in case anything happened to me. But only on an understanding between Farrell and myself that he would then defer to Nichols on everything until he knew better.

Groueff: I see, because Nichols, if I understand correctly, was under the procurement and construction side but not on things like Los Alamos or military policy.

Groves: Well, he had the general supervision administratively of Los Alamos, but I ran it, which made it difficult; there was friction at times. At Oak Ridge, he ran that completely, that operation. We both ran Hanford with the officer out there, Matthias.

Groueff: But Matthias told me that they had very free hands there.

Groves: Matthias? 

Groueff: Yeah. 

Groves: Yes, he did, but then everybody had a free hand.

Groueff: Right, and he said that theoretically he should be under Nichols, but in practice there was never a problem with that; Nichols never bothered him. He was in direct touch with you.

Groves: That is right. But actually you might say Los Alamos was completely under me excepting for the paperwork. Hanford was half and half with my having the greater influence there because Matthias came to me and also DuPont came to me – generally, the bigger people in DuPont.

Oak Ridge was entirely under Nichols. The University of Chicago Laboratory, Nichols had almost entirely expected I would go through there. The reason for that was that I felt that Compton liked Nichols better than he did me, I am sure of that. So I just told Nichols, “Compton has much more respect for you than he does for me. You run Chicago.” That was alright. 

Then the laboratories in New York, they were all under Nichols. I kept pretty close touch there, but not completely. Nichols really did more with that than I did. Berkeley, I would guess that I probably did more than Nichols did with Berkeley. But the whole thing was—as I had told Nichols and Marshall ahead of him—I said, “I am going to interfere all of the time. I know it. Do not get mad about it or if I shortcut you because I am going to do it. But if there is ever conflict between us, the subordinate has got to tell us both right away or at least tell you so that you can straighten it out.”

Then in the security, Nichols had his own security people. They were under Colonel Parsons at the end. But I had a security section in my office after we took over from the War Department. That was headed by Lansdale. We had friction every once in a while because Lansdale would shortcut Parsons.

Groueff: There were two organizations, Nichols and yours? 

Groves: Yes. I only had this little bit of an office, you see. The only reason for having the security that way was because we had all of these contacts with the FBI. You had to have somebody who was smart and able and had to be practically in here in Washington. Then Nichols had a liaison office here in Washington for handling procurement. That was bigger than all of the rest of my office put together.

That officer reported to Nichols and he came to see me whenever he needed help. He did not have to go to Nichols. He would say, “I can’t do this.” He would come to me first, and said, “Can you give me a hand?” Then when the overseas intelligence came in, I kept that entirely under Lansdale. When the planning for action came in, I kept that entirely and put Farrell on it.

Groueff: The targets?

Groves: The targets and everything to do with that.

Groueff: Nichols did not have anything to do with that?

Groves: No. The reason that I did not want Nichols to succeed me was not because I did not have confidence in him—although I think it would have worked better with Farrell in there—but Nichols would have been thought of as being young. He was. He was younger than I was, you see, by about ten years. Also, he was not used to Washington and making his way around Washington.

Groueff: Congress and public relations?

Groves: Well, not Congress so much, but the military in Washington. He could not have gone in and faced down Styer and told him, “No, we are not going to do that.” If Clay had started to pinch off his supplies, which he would have done immediately, he could not have fought with Clay. Clay would have beaten him.

Groueff: Clay?

Groves: Lucius Clay. He was down in Somervell’s office in charge of procurement of supplies. Alright, what else have we got now?

Groueff: Now, how close did you personally follow the development of the barrier and all of these series of developments?

Groves: I did not follow it in the sense of studying it or making minor decisions but I followed it in the sense of making all of the major decisions of it excepting towards the end, when I told Nichols and we were desperate. I said, “You just go ahead, you are authorized to initiate anything that does not cost over five million without even asking me if you think it has got the slightest chance of success.” Now, I did go out to Houdaille-Hershey and Nichols was not there. I am sure because I came up from Oak Ridge.

Groueff: You went to Chrysler also about the diffusers?

Groves: Yes, I went there. But I went to Decatur and I had a meeting there. As a result of that meeting and inspection, we ripped out all of our process material and started a new system in there with a later type of barrier, which is the later method of manufacture essentially.

Groueff: But you did not go into the details of the Norris-Adler barrier and mixed barrier?

Groves: No, not particularly. I was familiar with them. But I would say I was familiar with them rather than had detailed knowledge of them.

Groueff: You were not briefed every day about it, or I mean, every week?

Groves: No. My whole policy was I went where things were going rough—where I could do some good. I was not there to just supervise and see that everything was lovely.

Groueff: Now, as I told you of the K-25 plant is supposed to be very classified. It is one of the admirable things that I have to describe as material proof of the whole project. It is the one building which is the most impressive. I feel a little bit uncomfortable not knowing what it looks like.

Groves: Of course, you can see the outside of it.

Groueff: The outside, yes. It is very impressive, right; I saw pictures.

Groves: It is very impressive.

Groueff: How did it look from inside from the eyes of a complete layman?

Groves: Well, I think you can get a description of that down there.

Groueff: Do you think they will give it to me?

Groves: I think so, if you talk to Larson.

Groueff: I have no idea if it is like a big factory with enormous rooms or is it separated things?

Groves: Well, you have almost got to talk to Clarke Center or Larson because I have forgotten even how many floors were in it because there was this very expansive floor space. On one floor would be the equipment itself, which was sealed up in containers so you could cut them off. Then, the instrumentation went right straight up above that. Originally, I think we had a tremendous number of employees there just on instrumentation. They had to read each one individually.

Later that was all concentrated into the one room, you see. But I think that is the important thing to put in, that there was not time to make the design. I think that was ahead of the time that that started to be done anyway. I do not know when was the first time you had a central control room for a chemical operation. But that would be an interesting thing to put down.

Groueff: Yes, I think it was probably the first really automated enormous plant in the world, no?

Groves: Well, I think you might be able to get something on employment there. At least you can get how many men we had during the war to get it in operation. And then how in the course of the operation of the Manhattan District and in the implementation of the work that they had already started before they turned over had been reduced. It was reduced to a mere handful.

Groueff: From what I understand, it is a real miracle of engineering.

Groves: Yeah, you just go in there and you do not see anything. 

Groueff: It does not look like a plant with a lot of workmen and mechanics. Everything is automatic. You do not see anybody. It is silent. 

Groves: You see one occasional person around. And you will not see the instruments because the instruments are not used anymore.

Groueff: It runs automatically?

Groves: It is all run from a central room in one corner of the building that is about as big as the living room and hallway in this apartment, something like that.

Groueff: Not bigger?

Groves: No.

Groueff: When you think that since 1944 or ’45, it works. From what I understand, there was never a major break for twenty years. It is a fantastic thing. Half of the public does not know that the factory that you put together twenty years ago, it is still running and everything is perfect.

Groves: Yeah. Now, there have been changes and improvements made, mostly with a view to further automation, I would say. Offhand, I know this: they told me down there, I asked them how they got along with all of their labor unions and they said, “Well, every time they raised the rates on us, why we cut the payroll.”

He says they have not found that out yet. Apparently, if they raise the rates by ten percent, their actual costs for labor after that is usually five percent less. They take advantage of that to reduce the people that they found are excess.

Groueff: There is one delicate point that I have to see how to present. When I talked to Dunning long ago and in the few things that he has written, he refers to the whole process and to the K-25 as the “Dunning-Booth plant,” the “Dunning-Booth method,” or the “Dunning-Booth cascade.” Now, when I talked to all of the Kellex people, they say the contrary and they are a little bit indignant. Keith said that there is not one single component, pumps or barrier, or pipes, or valves, which came out of Columbia. He says that it is true that they work on the same problems—I will not discuss which is better—but it happens that the ones that were put in K-25 without an exception were all developed by Kellex or other people. But none of them, for instance they had a pump at Columbia, but they used the Kellex pump, Swearingen. They did the same thing with valves, et cetera. That is a rather delicate—

Groves: No, it is not at all. You can get around that very easily by just saying that the whole policy was one where you would compartmentalize this thing. First came the theoretical science, which could be written down on paper and sketched. Then came the scientific development that showed that you could make a pump that would go part way. Then came the engineering development that developed a pump that would really be competent and take all of the things that went on before. Then finally, that pump would be taken and it would be manufactured and there would be certain changes made there to make it easier to manufacture. It is not the Dunning-Booth pump. That is just idiotic. Kellex is right. Actually Dunning did not claim the barrier, did he?

Groueff: No, but he claims that the whole principle and that they invented the cascade. Now, Keith says that his organization, especially Manson Benedict, developed the cascade as it is in K-25 and not the people like Dunning and Booth. In other words I think the tendency of Dunning is to present the thing as his.

Kellex and all of this, they did a fantastic job. But it is like let’s say you invent something and you find a manufacturer to produce it. But it is you that is the inventor and the credit should go to you. Because in other words if it was not done by Kellex, it would have been done by somebody else. But the thing is to invent that.

Groves: Well, of course, the thing is that most of the work that Dunning did, I do not think it did as much good to be perfectly frank about it. I think all of the work was done by Kellex and by the lab after [Laughlin] Currie came into the picture. People like Hugh Taylor at Princeton and people like Mack—I think he came from Ohio State—and others were the ones who really did something. But Dunning was just a big blow off and it still is. He has made a terrific reputation out of being and talking about himself. 

But if there is one thing about Dunning—I think you could give him the credit this way—Dunning was a very strong proponent from the start while other men at Columbia were doubtful. I would say other Columbia professors were doubtful—that takes away the ones who came in from outside so you do not have as many enemies. Dunning never lost faith.

Groueff: Keith says the main contribution of Dunning is that he was so enthusiastic and optimistic. His energy was contaminating the others; he would come and say, “I got it! I got it! It is marvelous. I have the barrier!” Then they would all come around and of course, when they touched it with their fingers, it would fall to pieces. But Dunning was still enthusiastic and they needed that. But I see the slant there.

Groves: Yes, I think for once Dobie Keith has put his finger right on it. Dunning’s great contribution was his enthusiasm and optimism. I think anything else he did, it did not amount to much. So I think if you trace out how these things were developed, then you do not have to call it a Dunning-Booth affair. I would have something that said how this all was developed. In other words it started off with a theory based on the known discoveries of the past such as Chadwick’s discovery of the neutron and the like, and Einstein’s theory of relativity. These were the bases on which the whole project was founded. 

Then there was the theoretical thinking that went on that people wrote down on paper using just plain pads and pencils. Then came preliminary laboratory work, which was done by scientists. Then when it reached a point where it seemed feasible, then engineering development moved in and made feasible equipment. Equipment that was feasible to build and to operate.

Groueff: I want to start my book there and I want to explain that which was not clear to myself before I started these interviews: that from the laboratory experiments, even the successful ones, their translation in doing this project is something practically completely different. It is not like in a current thing you invent something and you send it to be made. Making it becomes a new challenge and new inventors feel that actually. 

Groves: Yes, it was an inventive field like all industrial engineering development. It is inventive but still it is not anything that requires that great spark. It is something that you can tell a man, for example, supposing you took that typewriter and you said, “I want to put two extra keys on it.” Well, that would be a terrible job. I can tell you that we have tried to do it. It is just almost impossible. That is not a case of original invention. That is a case of an engineer who sits there and says, “How can I solve this problem?” He goes on taking everything. But on the other hand, the man that first invented the typewriter, he had the original spark.

Groueff: I think the wrong idea is to try to give credit and to see things in black and white. I think that all of them had part of the truth and they were right.

Groves: Yes well just take for example the British, [who] kept saying that you couldn’t handle these cascades. You would have a break and it would take seventy days to get this plant back into shape again. Well, the seventy days was all wrong. We know that today. I said, “Why, all you have to do is jump that particular section.”

They just could not understand how that could be done. They said if you jump it in time, it would not do any good then because it would upset things. I said, “But do you think that gas that it is just a shade different from the other can ever tell the difference?” I said, “You could not tell it—”

I said it will all equalize immediately. The reason we have these elaborate electrical installations at Oak Ridge for the gas diffusion plant was this idea that we could not shut down for as much as a second or it would take us seventy days to get back into operations.

Groueff: It wasn’t true?

Groves: I do not think it was true at all. But I was not going to take any chances. If there was a feeling on the part of competent people and they were confident that could happen, it was foolish for me to permit it to happen. Whether I believed in it or not, I just said we are going to make this plant so that it is not subject to a break down like that.

Groueff: Yeah. I think it can be done in a balanced sort of way.

Groves: I think so. I think your big problem now is just to get started writing.

Groueff: Yeah.

Copyright 1965 Stephane Groueff. From the Stephane Groueff Collection, Howard Gotlieb Archival Research Center at Boston University. Exclusive rights granted to the Atomic Heritage Foundation.