CHAPTER 6

How Safe Are Nuclear Reactors?

Routine Operations

          Discussion of the safety of nuclear reactors has two components: (1) The normal day-to-day operations in which the reactors are permitted to release radioactivity into the environment, and (2) the accident situation wherein the reactor may release large quantities of radioactivity into the environment.

          In normal day-to-day operations, nuclear power plants are permitted by law to release radioactivity in the form of radioactive atoms to the environment in gaseous and liquid discharges. There are essentially two regulations concerned with these releases. The regulation which represents the primary standard is the dosage that could be delivered to an individual or to the population-at-large. We have discussed this primary standard earlier in this book and have indicated that the standard is much too high. The maximum permissible concentrations of the various radionuclides in air (MPCa) and water (MPCw) that are permitted to be released outside of the restricted area of a nuclear reactor are called secondary standards.

          The primary standard should be derivable from the secondary standards. But, as we indicated earlier, the secondary standards, the maximum permissible concentrations that are listed in Title 10 of the Code of Federal Regulations, do not permit this. The MPCs that are tabulated in Title 10 of the Code of Federal Regulations apply only to the situation where individuals are breathing the contaminated air or drinking the contaminated water. They do not take into account the fact that the contaminated air and the contaminated water will result in the contamination of the foods consumed by man. This is an extremely important factor in terms of the dosage that would be received by man from reactor releases.

          The proponents of the nuclear power industry state that the exposure from the nuclear power plants would be considerably lower than those of the guidelines. They indicate that individuals in the near vicinity of nuclear reactors would be exposed to no more than 5-10 mr/yr value and that as individuals lived further and further away from the reactor, their exposure would drop off very rapidly from this 5-10 mr/yr value. Moreover, they indicate that the design objectives and the operation of existing power plants are such that the actual releases of radioactivity from the power plants are no more than 1 percent of the releases allowed by the AEC's MPC values.

          The State of Minnesota has proposed emission standards that are 50 fold lower than those of the AEC. At a symposium titled "Nuclear Power and the Environment", held at the University of Minnesota in October, 1969, members of the audience repeatedly asked the question, "If reactors are only going to release the small amount of radioactivity that you indicate, then why are you so reluctant to make guidelines more restrictive and adopt the Minnesota regulations?" Congressman Craig Hosmer, a member of the Joint Congressional Committee on Atomic Energy, stated that if the standards were lowered, he doubted if the reactors could operate safely. Commissioner Theos Thompson made essentially the same statement in testimony before the Joint Committee on Atomic Energy.

          In recent hearings before the JCAE, the following exchange took place between AEC Commissioners James Ramey and Thompson and Congressmen Chet Holifield and Hosmer. We will reprint their exact words, with only a bit of interpretation.

      Chairman HOLIFIELD . . . One other point I wanted to bring out was that the newer plants and the plants that are now being put on the line commercially and which do not have experiments involved in their continuous operation show consistently a concentration limit of less than 1 percent.
      Is that not right, or am I wrong in reading the chart?

      Dr THOMPSON. That is correct. You are reading the chart correctly.
      I would like to make the statement, though, that there may be times, when, even in spite of careful inspection—which is always done—and the checkout to assure that the surface of these fuel elements is free from uranium, the effluent levels will rise above this one percent but still be well within the current part 20 limits.
      It is therefore important that we have what I will call an operating cushion.
      You talked with Dr. Totter the other morning and you asked him whether there was a reasonable cushion of safety between the effects of radiation and the present standards which are set up.
      What I am talking about is another cushion between the part 20 standards and the normal operating level. That cushion is important for the reliability of these plants as a part of an electrical utility system. Assume a utility builds a nuclear reactor, and then, say, they go up from one percent to three percent of the part 20 limit. Then, if we have set a lower limit at one percent, this reactor would have to be shut down. But it would not really be shut down because of a safety reason but simply because somebody had arbitrarily established a very low limit.

(Dr. Thompson means that, if the limits of radiation at the reactor were set at 1 percent of the part 20 limit, but the operators found they could not keep the limits this low, then the reactor would have to shut down.)

      This would materially reduce the reliability of this plant as a power source.
      I think the AEC has an obligation, as a responsible group, to be sure that the reliability of these plants is not reduced by making these standards too low just arbitrarily.

(By reliability he means here its ability to continue to provide power. He is not talking about reliability in terms of safety.)

      Chairman HOLIFIELD. Of course, if safety standards are too restrictive, the various attempts to comply with a too rigid standard would increase the probability of trouble from a technological standpoint; is that not true?
      Dr. THOMPSON. If the radioactive effluent standards are too rigid, in my own mind at least, there are some very grave worries that I have concerning whether this may not reduce the ultimate safety of the reactor plant itself. If one begins to push too hard on holding down effluents, one may as a result affect reactor safety adversely. For instance, say we hold up all the tritium in the plant. This tritium makes very high levels of tritium in the air inside the containment. Then the tendency will be not to inspect the plant so often.
      Another example. In the boiling water reactor, there are those who would cut down the effluent that is released through the stack too strenuously and too early before technical feasibility for doing this has been demonstrated. It may well be that, as one moves to a very long holdup of gases in the boiling water reactor effluent system—and a lot of the gases which come out from this plant are really hydrogen and oxygen which are disassociated in the core of the reactor—there is a possibility that unless one is very careful you will induce an explosive hazard where no hazard was there before.
      Therefore there is a very close interaction between effluent discharge levels and safety of the reactor.
      I am somewhat concerned that we will move from a more safe reactor to a less safe reactor if we push the effluents down more than we should on a reasonable basis. I believe we are on a reasonable basis right now.

(In other words, the more you restrict the levels of radioactivity loosed on the world outside the plant, the more you risk a possibly catastrophic explosion at the nuclear reactor.)

      Representative HOSMER. You brought up this matter of the cushion. You used as an example if you go from one to three percent.
      Dr. THOMPSON. I picked the number one to three.
      Representative HOSMER. Let us call that a size three cushion. But you go to 100 percent under the same limitations. Would you then be using a size 100 cushion?
      Dr. THOMPSON. That is right.
      Representative HOSMER. So, let us get into the reasonableness of the size of the cushion. We know that the limitations are established on the basis that you can go up to 100 and still do no damage to individuals and the public but some people seem to think that there is not enough known so that that might not be an absolute guarantee.
      So, why don't we think in terms of reducing the legal size of the cushion to what would be reasonable?
      If you say you want to go up to three percent, maybe size three cushion or maybe size 10 cushion, to give you some extra latitude, some elasticity, you know, to assure the public again and again and again that their health and safety is being cared for?

(We think Rep. Hosmer's statement here indicates, perhaps better than anything else could, the total confusion that exists in regard to the possible hazards we face. An important question is why do the Joint Committee and the AEC assure us that we are in no danger, even though they themselves confess to a great deal of confusion and uncertainty?)

      Mr. RAMEY. Mr. Chairman, may I comment on that?
      Chairman HOLIFIELD. Yes; proceed.
      Mr. RAMEY. I think we do have the standard for guidance here and it is the standard that is under the FRC of holding the levels as low as practicable.
      We have looked at this rather carefully. We are still looking at it as Dr. Thompson has indicated, but there are these factors that we have to take into account in balancing this, these trade-offs between reactor safety and the safety from the effluents. It might be possible to give some guidance as to what is practicable, how this could be handled. But it is not likely to be something that sets some limit in terms of radioactivity. It is more likely to be guidance in terms of design and in terms of operating procedure on how the utilities now are holding these levels down in these ranges. Because every once in a while you may have to go up above any particular limit and be near your 100 percent factor.
      Representative HOSMER. Mr. Ramey, with the older reactors that Dr. Thompson has just discussed, the Humboldt Bay reactor, for instance, the technology has now proceeded to where the practical limits observed in the normal course of operation are by a factor of 100 below the legal limits prescribed in the licensing process.
      Since the technology has developed and since the practical limits are being observed, all I am trying to seek is some accommodation between the present legal limit and the practical limit at which the elasticity, the cushion, would be adequate but at the same time the legal limits reduced.
      Mr. RAMEY. As I say, Mr. Hosmer, I think if we look on part 20, that number is initially set, as has been brought out, as a very conservative number in the first place with a great number of factors of conservatism in it.
      Representative HOSMER. Part 20 has already been described by Dr. Thompson as a "dynamic and living thing."
      Mr. RAMEY. That is right. I think the way we are looking at it is in terms of within it, and in accordance with the FRC guidelines, of how one might provide guidance on what is practicable below the part 20 numbers.
      Now, operating experience has shown that these are at a fraction of the part 20 numbers on these types of reactors.
      There are transient situations which may exceed this experience. For example, in Minnesota, in this Minnesota permit, what they have taken as an average and made that the limit. Anybody knows when you set a limit based on an average that sometimes you are going to go over that average and at other times you are going to go under it.
      So, if you set it at the average and as an absolute limit, you are going to be violating it.
      Representative HOSMER. I am not talking about an average. I am talking about an average plus a reasonable cushion and asking if size 99 to 100 is a reasonable cushion or wouldn't size 25, for example, be a reasonable cushion.
      Dr. THOMPSON. Mr. Hosmer, we don't have at the moment any way to set a reasonable cushion. There is not that sort of experience. So we should not move and make that cushion smaller until such experience exists.[1]

          Basically what the AEC Commissioners are saying is that they don't want to change the standards until they know how much radioactivity will be released. If the reactors are going to release only 1/100th of the present allowable release rates, then why should the AEC be so reluctant to lower the standards by at least a factor of 10? The only conclusion that a reasonable person can come to is that the AEC does not firmly believe that the reactors will be able to operate at these lower release rates.

          This 1 percent release rate is a design objective. Dr. Thompson recognized this in his testimony and also recognizes that an operating plant may exceed these objectives by a wide margin. A little later on in his testimony, Dr. Thompson stated:

      Frankly, at this stage in the development of atomic energy I think it would be premature to set this, say, three percent cushion or 10 percent cushion, in an arbitrary manner. I think we ought to take a look at the large plants that are coming on line and see how they are going to do. I think they will be at the same levels as present plants but we also need a fair amount of cushion . . .[2]

          Mr. Wilfrid Johnson, another AEC Commissioner, supported Dr. Thompson's position a little later in the testimony:

      Mr. JOHNSON. I wanted to add, with regard to the same point that Dr. Thompson brought up that we do need the flexibility in the levels, in part, because they have to apply broadly over various kinds of plants, such as chemical processing plants, as well as reactors. They are also related to occupational exposures.
      There is no way to completely divorce the matter of effluents of a plant from the occupational exposure that the employees get. They are related matters.
      On top of that, we must consider new plants that come along. They will have different kinds of releases and the limits have to apply to them, too.
      If we were too rigid, we would have nothing but boiling water and pressurized water reactors from now on. If we get to liquid metal cooled fast breeders, the effluent problem will be different. Hopefully, they will be better, but we know they will be different. We need flexibility for these reasons.[3]

          (Here Commissioner Johnson admits that nobody knows what the effluent problem will be in the fast breeder reactors which the AEC assures us are the only final solution to our power problems. They have announced plans for such a reactor at Meshoppen, Pennsylvania. Presumably they must wait until this reactor is in operation before they will know how much radiation will be released in its operation! )

          To a considerable extent, the amount of radioactivity released to the environment by an operating nuclear power reactor depends upon the integrity of the fuel rods in the reactor. The large reactors that are planned, and are being constructed, in this country today have thousands of these fuel rods inserted into the core of the reactor. These fuel rods can develop small pin holes. The radioactivity generated within the fuel rods then leaks through these pin holes and into the water which is moderating the reactor. (See diagram of reactor core in Chapter 1, page 38.)

          In a boiling water reactor the gaseous products will be released through the stack. The reactor is not able, completely, to contain this water which bathes and moderates the fuel elements and collects the radioactivity which leaks from the rods. Therefore, radioactively-contaminated water accumulates within the reactor housing. This radioactive waste water is then released into the cooling water and returned to the river or to the ocean.

          Consequently, the degree below the maximum permissible concentrations that a given reactor will be able to operate depends upon the integrity of its fuel rods, as well as the integrity of all the valves, nozzles and pipes in the plumbing and cycling system of the reactor. The reactors presently under construction are planned to operate for some 20 years. Plans are to change the fuel rods only once every two or three years. These reactors are considerably larger than the reactors upon which we have any experience to date.

          The combination of these things indicates that we do not really know how these reactors will operate as they begin to age and as their fuel rods begin to age. It may well be that the natural aging process of the reactor, variations in quality control, and operator errors will cause it to creep up to the maximum permissible concentrations that are presently allowed by the AEC. They might even exceed those levels!

          Since nuclear power reactors are being proposed at a rate which indicates they will be supplying a very substantial fraction of our future electrical power needs, we will be presented with a fait accompli in the future. Even if these reactors do not operate at their design specifications, it will be difficult to shut them down because we will need the power. If we shut them down, sizeable sections of the country will experience periods of brown-out. We might, therefore, be forced to live with whatever radioactive emissions the reactors require. Once we have made a very sizeable commitment to nuclear-generated power, we must face the fact that we will be stuck with the commitment.

          The discussion above indicates that the present generation of reactors is no more than an experiment. The public is told that the guidelines are safe. But they are not safe! The public is told that the radioactive emissions will be only 1 percent of the guidelines. This is a design objective. An objective that the AEC Commissioners are not in the least certain will be met. The AEC is adhering to its guidelines in order not to inhibit the development of the nuclear power industry by engineering, operational, or quality control failures. The public is simply required to take the risk inherent in this "Cushion."

Accidents In Present-Day Reactors

          In addition to the uncertainties in the day-to-day release, uncertainties exist about chances for a major accident. Dr. Walter H. Jordan, Assistant Director of the Oak Ridge National Laboratory and a member of one of the AEC's reactor safety boards, stated in a recent article in Physics Today:

      The important question still remains: Have we succeeded in reducing the risk to a tolerable level, that is, something less than one chance in 10,000, that a reactor will have a serious accident in a year?
      Have we succeeded in reducing the hazard to such a low level? There is no way to prove it. We have accumulated so far some 100 reactor years of accident-free operation of commercial nuclear electric power stations in the U.S. This is a long way from 10,000 so it does not tell us much.
      The only way we will know what the odds really are is by continuing to accumulate experience in operating reactors. There is some risk but it is certainly worth it.[4]

          How safe are nuclear reactors? Let us quote from consulting engineer, Adolph Ackerman:

      As an independent consulting engineer I have been active for many years in alerting the engineering profession to its overriding responsibilities in design and construction of safe atomic power plants. The simple fact is that none of the atomic power plants currently in operation or under construction have been designed with the traditional concepts of engineering responsibility and ethical commitment for maximum public safety.[5]

          Mr. Ackerman spelled out his reasons for this statement quite clearly in a recent article. Professor Robert L. Whitelaw, of the Virginia Polytechnic Institute and formerly Project Engineer for the design and construction of the power plant for the nuclear ship, N. S. Savannah, commented on this paper by Ackerman in the IEEE Transactions on Aerospace and Electronic Systems (vol. AES-5, no. 3, May 1969):

      I wish to endorse fully the principal argument advanced by A. J. Ackerman in his paper and, perhaps, strengthen the impact of his paper with this brief discussion.
      His principal argument has been confirmed by my own experience of the past fifteen years on nuclear projects and problems of various kinds This experience included preparing proposals and nuclear hazards evaluations in a variety of nuclear power plants, both commercial and military.
      It has been my observation that, despite the enormous amount of meticulous detail which the ACRS regularly requires on every projected power plant to satisfy itself that there is no "credible accident" that can threaten the public (or even the operators)—and despite the volumes of paper and hours of presentations consumed on this topic, and no doubt well-intentioned—there is still by common consent an unwritten agreement to treat as "incredible" the most fearful of all nuclear accidents that can occur in any plant with a highly pressurized primary system Such an accident is, of course, the explosive rupture of the primary vessel itself, which is ruled out of the list of credible accidents for the simple reason that there is no adequate answer short of putting the plant underground or inside a mountain, as Ackerman has pointed out.

          Dr. Edward Teller, often called the father of the hydrogen bomb and one of the most outstanding supporters of the AEC, has stated:

      A single major mishap in a nuclear reactor could cause extreme damage, not because of the explosive force, but because of the radioactive contamination. . . . So far, we have been extremely lucky . . . But with the spread of industrialization, with the greater number of simians monkeying around with things they do not completely understand, sooner or later a fool will prove greater than the proof even in a foolproof system.[6]

          On September 10, 1970, in Livermore, California, Dr. Teller told the Livermore Chapter of the Society of Professional Engineers that reactors were safe, but they should be put underground.

          How safe are nuclear reactors? Let us quote from a letter of the AEC's Advisory Committee on Reactor Safeguards concerning a reactor planned for Midland, Michigan.

      . . . The number of permanent residents within five miles of the plant site was estimated to be 41,000 in 1968, mainly in the city of Midland and its environs.
      The applicant has established criteria for, and has begun the formulation of a comprehensive emergency evacuation plan . . .

          In considering the safety of nuclear reactors, it is important to recognize that each nuclear reactor in this country is an experiment. Each reactor is different from all other reactors and whether or not it will operate and/or operate safely depends upon the outcome of the experiment.

          One of the reasons for this is that the AEC has not funded safety research at an appropriate level. This was recently pointed out by Mr. Joseph M. Hendrie, Chairman of the Advisory Committee on Reactor Safeguards, in a letter to Dr. Glenn T. Seaborg, Chairman of the Atomic Energy Commission, dated November 12, 1969:

      DEAR DR. SEABORG: The Advisory Committee on Reactor Safeguards (ACRS) wishes to reemphasize some previous recommendations concerning the need for safety research in several important areas in which the effort has not been sufficient. The Committee has been recently informed that overall reactor safety funding for FY 1970 and 1971 will be considerably below the AEC estimates of need for the water reactor safety research program, as well as for safety research on seismic effects, on sodium-cooled fast reactors, on high-temperature graphite-moderated, gas-cooled reactors, and on environmental effects. As a consequence, many safety research activities have not been initiated, have been slowed, or have been terminated. The Committee reiterates its belief in the urgent need for additional research and development in these areas, and refers in the paragraph below to earlier statements of the Committee on these subjects.

Water Reactors

      In its letter to Mr. Hollingsworth of March 20, 1969, the ACRS stated its belief that ". . . more effort should be devoted to gaining an understanding of modes and mechanisms of fuel failure, possible propagation of fuel failure, and generation of locally high pressures if hot fuel and coolant are mixed, and that effort should commence on gaining an understanding of the various mechanisms of potential importance in describing the course of events following partial or large scale core melting, either at power or in the unlikely event of a loss-of-coolant accident." The Committee has strongly recommended safety research of this kind several times during the last three years; the Regulatory Staff has also strongly supported such work. However, only small or modest efforts have been initiated thus far.
      In its comments of March 20, 1969, the Committee also recommended that ". . . considerable attention be given now to the potential safety questions related to large water reactors likely to be proposed for construction during the next decade. Large cores, higher power densities, and new materials of fabrication are some of the departures from present practice likely to introduce new safety research needs or major changes in emphasis in existing needs.
      The Committee further recommended that consideration be given to ". . research aimed specifically at improving the potential for siting of large water reactors in more populated areas than currently being utilized; for example, studies should be undertaken to develop reactor design concepts providing additional inherent safety or, possibly, new safety features to deal with very low probability accidents involving primary system rupture followed by a functional failure of the emergency core cooling system."
      It appears that, because of funding limitations and for other reasons, the recommendations of the ACRS will not be implemented at this time.

Liquid-Metal-Cooled Fast Breeder Reactors (LMFBR)

      The ACRS, in its report on safety research of November 19, 1963, stated that "Recent renewed emphasis on the long range role of large fast breeder reactors points up the need for a well developed, long term, comprehensive research program on the safety of such reactors. A strong research program started now should develop information very useful to the first generation of very large fast reactors." The Regulatory Staff and the ACRS have recently undertaken a preliminary review of a proposed site to be used for construction of a 500 MWe LMFBR. Construction permit reviews, of one or more LMFBRs, are anticipated in the next few years
      While an extensive LMFBR safety program plan has been formulated, and a growing program in LMFBR safety has been started, many safety-related design decisions will have to be made by applicants and the regulatory groups without the benefit of needed safety research, in part, because of a lag in the implementation of studies of high priority matters.
      . . . In summary, the Committee again emphasizes the importance of safety research to the protection of the health and safety of the public and urges that adequate funding be provided to permit timely pursuit of work in all high priority areas.[7]

          In a letter of November 12, 1969, to Mr. Robert E. Hollingsworth, General Manager of the U. S. Atomic Energy Commission, Mr. Hendrie stated:

      . . . The water-cooled reactor safety research program in PBF (power burst facility) should concurrently investigate, with high priority, the mechanisms and phenomena associated with the initiation, growth, and propagation of fuel pin failure, including the circumstances under which melting of fuel could progress beyond one fuel element. Such a situation could develop in a large power reactor because of a local reduction in heat removal rate (as by-flow blockage), a locally abnormal power density (as by incorrect enrichment of fuel), or a more widespread perturbation in power or flow. These experiments are required in order to ascertain the probability of a local incident progressing into a serious accident and, if possible, the course and consequence of such a sequence of events.[8]

These complaints, by the AEC's Advisory Committee on Reactor Safeguards, suggest that the present reactors and those under construction are far more experimental than we might have imagined.

          It is significant to note, particularly in relationship to the ACRS concern over loss of coolant, which it considers as an unlikely event, and Dr. Teller's statement about "simians monkeying around," that Mr. E. P. Epler discusses an emergency cooling system failure in the Oak Ridge Research Reactor in the July-August, 1970, issue of Nuclear Safety. In this case three human errors and four design errors contributed to the incident. In his conclusions, Mr. Epler states:

      The errors and failures cited are not individually unusual, although it would ordinarily be expected that they would be corrected during early operation and system shakedown Engineered protection systems are not operated routinely and, as a consequence, error and failure modes can lie dormant and unsuspected, only to appear when emergency operation is required . . .
       The incident was not the result of a single failure but resulted, amazingly, from seven failures or errors in each of three identical channels, a total of 21 failures If any one of these had not occurred, the reactor would not have been operated without emergency cooling. It is also noteworthy that this incident happened in a plant with an outstanding safety and availability record . . .[9]

          Aside from the chance of a serious accident, these delays in safety research or its counterpart, proceeding too rapidly with the development of the nuclear energy program, may have forced us into the position where we shall have to accept far more risk for our electrical power than was necessary. Moreover, we may end up with a less reliable source of power. The reactors may have to be shut down frequently because of unforeseen engineering problems.

          For example, in the May 14, 1970, issue of Nucleonics Week there is a fairly long discussion on the problems that developed with furnace-sensitized stainless steel in critical areas of the reactors. This article indicates that trouble was encountered at the reactors at Oyster Creek, Tarapur, Nine-Mile Point, and LaCrosse. These problems developed in furnace-sensitized stainless steel safe ends and other miscellaneous supports in the reactors.

          A somewhat similar problem developed in the Indian Point reactor (May 20, 1970) where small pieces of material were found circulating in the cooling water. Since these reactors were constructed to meet critical power needs, it appears quite possible that brown-outs will occur when nuclear reactors fail. The possibility looms larger as we proceed to larger plants, each plant being a significant part of the energy supply.

Accidents In Fast-Breeder Reactors

          The comments above concerning the water moderated reactors apply even more pertinently to the fast-breeders. Dr. Edward Teller expressed quite well the concern of many scientists and engineers, relative to fast breeders, when he wrote in Nuclear News:

      For the fast breeder to work in its steady-state breeding condition you probably need something like half a ton of plutonium. In order that it should work economically in a sufficiently big power-producing unit, it probably needs quite a bit more than one ton of plutonium. I do not like the hazard involved. I suggested that nuclear reactors are a blessing because they are clean. They are clean as long as they function as planned, but if they malfunction in a massive manner, which can happen in principle, they can release enough fission products to kill a tremendous number of people.
      . . . But, if you put together two tons of plutonium in a breeder, one tenth of one percent of this material could become critical.
      I have listened to hundreds of analyses of what course a nuclear accident can take. Although I believe it is possible to analyze the immediate consequences of an accident, I do not believe it is possible to analyze and foresee the secondary consequences. In an accident involving a plutonium reactor, a couple of tons of plutonium can melt. I don't think anybody can foresee where one or two or five percent of this plutonium will find itself and how it will get mixed with some other material. A small fraction of the original charge can become a great hazard.[10]

          In his book, The Careless Atom, Sheldon Novick describes a number of accidents that have occurred with nuclear reactors. One of these occurred at the Fermi Reactor, 30 miles from Detroit, Michigan. This is our first and only large scale fast breeder. In this accident some of the fuel rods had melted. The situation described above by Dr. Teller had occurred. Mr. Novick quotes Walter J. McCarthy, Jr., Assistant General Manager of the Power Reactor Development Company that owned the reactor, as stating that the possibility of a secondary and very serious accident was "a terrifying thought."

          The terrifying thought involved the possibility of the melted fuel reassembling into a critical mass and resulting in an explosion that could lead to the consequences foretold by Dr. Teller. It was a month before careful attempts were begun to remove the damaged fuel elements. When nothing happened, everyone breathed a sigh of relief.

          Dr. Teller says, "So far, we have been extremely lucky." But is Dr. Jordan's statement that the risk ". . . is certainly worth it" really true?

          How safe and reliable are nuclear power reactors? Apparently, no one really knows. The United States is engaged in a gigantic experiment. The stakes which each individual must gamble in this experiment may be extremely high, possibly even his life.

§
1. In Environmental Effects of Producing Electrical Power. Hearings before the Joint Committee on Atomic Energy, 91st Congress, Ist Session held Oct. 28-31 Nov. 4-7 1969. Washington, D.C., U.S. Government Printing Office, 1969, Part i, pp. 203-205.
   
   
   §
2. lbid., p. 206.
   
   
   §
3. Ibid., p. 209.
   
   
   §
4. Walter H. Jordan, "Nuclear Energy: Benefits versus Risks," Physics Today (May), 32-38, 1970.
   
   
   §
5. Personal communication.
   
   
   §
6. As seen in Eugene Register-Guard (Oregon), October 7, 1969.
   
   
   §
7. In AEC Authorizing Legislation. Fiscal Year 1971. Hearings before the Joint Committee on Atomic Energy, 91st Congress, 2nd Session, held March 11, 1970. Washington, D.C., U.S. Government Printing Office, 1970, Part 3, pp. 1619-1620.
   
   
   §
8. lbid., p. 1622.
   
   
   §
9. E. P. Epler. "The ORR Emergency Cooling Failure." Nuclear Safety 11 (4), 323-327, 1970.
   
   
   §
10. Edward Teller, "Fast Reactors: Maybe." Nuclear News (August 21, 1967).