Some of the surviving fragments of the Brazil magnesium, that purportedly had their origin in the explosion of a UFO,  have been subjected to surface, internal and isotopic analyses. 

The surface composition of four of the specimens has been determined to better than 1 ppm (part per million).  There are some similarities, but also significant differences, so it is clear that the specimens were subjected to different influences.  Some of the impurities (such as sodium and calcium) may be due to sea water or sand, but many of the impurities are incompatible with such contamination. Some of the impurities (titanium, chromium, iron, cobalt, selenium, strontium, yttrium, niobium, palladium and barium) may point towards an origin in a technological device or devices. 

Two specimens of Brazil magnesium, together with four comparison specimens, have been subjected to internal analysis by a laser ablation ICP-MS instrument (inductively coupled plasma mass spectrograph). This analysis shows that the Brazil specimens contain calcium at a few thousand ppm, and (as found by the Colorado Project) both strontium and barium at a few hundred ppm.  One specimen also contains titanium at a few hundred ppm.  This analysis indicates that the existing Brazil samples are not as pure as magnesium specimens readily available in the 1950s. 

Some of the specimens have been subjected to isotopic analysis. The only departures from normal isotopic ratios are small differences that may be attributed to fractionation as a result of heat treatment. 

The origin of these fragments remains a mystery. There is no evidence that the specimens are of extraterrestrial origin. 

The material that is here referred to as the “Brazil magnesium” has an important place in UFO research. It was the only material specimen investigated by the Colorado Project, which is the only unclassified UFO research project funded by a United States government agency (Condon & Gillmor 1969). (Footnote 1) 

The Brazil magnesium was first mentioned in a short article published in the Rio de Janeiro newspaper El Globo on September 14, 1957, reproduced in its original Portuguese in Appendix 1. This article, headed “UM FRAGMENTO DE DISCO-VOADOR!” [A Fragment from a Flying Disk!] was written by the El Globo society columnist, Mr. Ibrahim Sued, who reported that he had just received an interesting letter which he reproduced in his column (Sued 1957).  Professor Pierre Kaufmann of the Instituto Presbiteriono Mackenzie, Sao Paulo, Brazil, has kindly reviewed this column, and supplied the following translation: 
 
 

The above letter was read by Dr. Olavo Fontes, a resident of Rio de Janiero. Dr. Fontes (who passed away in 1968 at the early age of 44) was an M.D. and chief of the gastroenterology section of the National School of Medicine in Rio de Janiero. He was also an investigator of UFO reports.  Fontes telephoned Sued and arranged to visit Sued in his apartment that same day. Fontes asked if he could take possession of the samples, to which Sued readily agreed. Fontes has described the material shown him by Sued in his apartment as follows: 
 
 

The above excerpt is taken from a report that Fontes sent to APRO (the Aerial Phenomenon Research Organization, Tucson, Arizona) on November 30, 1957. This report was subsequently reproduced, with minor revisions and excisions,  as an article (Fontes 1962) in a book by Coral Lorenzen (1962). Fontes was affiliated with APRO, of which Coral Lorenzen and her husband Jim Lorenzen were the founders and directors. It appears, from his article and from APRO files, that Fontes took possession only of the material that had been sent to Sued’s apartment. There is no indication that he took possession either of the letter Sued had received at his home, nor the duplicate letter with additional specimens that had been send to Sued’s office. This has proved to be an unfortunate oversight. From information provided by Fontes, there is no way to be sure that the specimens did originate near Ubatuba (which is why they are here referred to as the “Brazil magnesium”), no way to be sure that they originated in an aerial event, and no way to be sure that the event (if it occurred) happened in 1957. These questions have been pursued by Professor Kaufmann. The results of his investigations will be presented in a separate article. In the present article, we discuss the results of composition analyses of the specimens. 

Fontes received from Sued three specimens, that he refers to as “Samples 1, 2 and 3.” Photographs of Samples 2 and 3 are reproduced as Fig. 1 of Fontes (1962). Their lengths were about 1/4 inch and 3/4 inch, respectively. Sample 1 was never photographed. 

Fontes first took the specimens to the Mineral Production Laboratory in Rio de Janeiro, a Division of the National Department of Mineral Production in the Agriculture Ministry of Brazil, where they were delivered to Dr. Feigl, the Chief Chemist. Feigl’s assistant, Dr. David Goldscheim, made a preliminary examination of a chip of Sample 1, and determined that it was a metal. Sample 1 was then divided into several pieces. Two were left with the laboratory, and Fontes retained the rest (together with Samples 2 and 3). Goldscheim sent one piece of Sample 1 to the Spectrographic Section of the Mineral Production Laboratory, where it was investigated by Dr. Luisa Maria A. Barbosa, a chemical technologist, using a Hilger mass spectrograph, model DMA 1-412 (a high-quality instrument). In her report dated September 24, 1957 (Fig. 2B of Fontes 1962), Barbosa states 
 
 

Fontes asserts that the sensitivity of the spectrograph was 1 ppm (one part per million), but it is not clear whether or not Fontes is quoting the analyst in making this statement.  It is important to note that the Hilger mass spectrograph evaporates the material in an arc discharge and is therefore destructive. Fontes received a copy of the spectrogram, which is reproduced as Fig 2C of his article. 

One of the pieces of Sample 1 was sent to the Laboratory of Crystallography of the Geology and Mineralogy Division of the National Department of Mineral Production.  An X-ray diffraction analysis was carried out by Dr. Elysiario Tavora Filho (then a Professor of Mineralogy at the National Chemical School). The results of these tests are reported in Fontes' article. This test was presumably non-destructive. 

At Fontes' request, a second spectrographic analysis of Sample 1 was carried out by Mr Elson Texeira at the same laboratory, using the same Hilger spectrograph. The spectrogram from Texeira’s analysis is reproduced as Fig. 3B of Fontes (1962). Texeira’s report, reproduced as Fig. 3A of Fontes (1962), reads in part: 
 
 

This appears to be the origin of the assertion, frequently repeated, that Sample 1 was "absolutely pure." This statement has been repeated many times in the UFO literature, with the implication that the material was extraterrestrial in origin. Fontes (1962, p. 115) states: 
 
 

However, the details of Texeira’s report do not support his assertion that the specimen was “absolutely pure,” since his report also contains the following statement: 
 
 

It appears, from this statement, that Texeira was claiming only that the specimen he analyzed was comparable in purity to a “chemically pure” specimen he had in his laboratory. Such a specimen is not 100% pure. As a gauge of the purity of magnesium specimens then available, we may refer to Fontes (1962) who wrote (p. 115): 
 
 

Here, as elsewhere in studying the Brazil magnesium, we are left with a puzzle: even if Texeira had used the ASTM standard for comparison, analysis by the Hilger spectrometer should have shown the presence of calcium, at least, since that spectrograph was certainly capable of detecting elements at the 0.1% (1,000 ppm) level. 

Texeira’a report raises another puzzle, when he states: 
 
 

One would like to know why the Hilger spectrograph did not show the usual contaminants from the carbon rod. Clearly, there is the possibility that the Hilger spectrometer malfunctioned, but it would be surprising if a similar (presumably rare) malfunction had occurred for both the Barbosa and Texeira analyses. 

On November 4, 1957, Fontes gave one of the remaining pieces of Sample 1 to Major Roberto Caminha of the Brazilian Army, who had the specimen analyzed at the IMT (Military Institute of Technology). A few months later, Fontes gave another piece of Sample 1 to Commander J.G. Brandao of the Brazilian Navy; this specimen is believed to have been analyzed at the Navy arsenal in Rio de Janiero. Neither piece was returned, and Fontes received no information about any analyses that were made at either laboratory. 

In her book (Lorenzen 1962), Coral Lorenzen adds an interesting postscript to Fontes’ report, dealing with his later attempts to track down the person who sent the material to Sued, and with evidence of military interest in the Brazil magnesium. Her postscript reads in part: 
 
 

In late 1957, Fontes conveyed the remaining piece of Sample 1, and also Samples 2 and 3, to the Lorenzens at APRO. If the piece from Sample 1 had been carefully identified and preserved, and subsequently tested non-destructively, this could have put an end to the speculation that Sample 1 differed from Samples 2 and 3, and that Sample 1 was “100% pure,” in the sense that it contained absolutely no impurities. As it is, no specimen that was in APRO’s possession was ever shown to be 100% pure. 

Coral Lorenzen (1962) recounts their experience in trying to get part of the material analyzed by a U.S. Air Force laboratory: 
 
 

The Lorenzens were able to make an informal arrangement with an APRO member, who was a theoretical physicist at a national laboratory, for an analysis to be made at that laboratory.  The physicist secured the assistance of a physical chemist, a chemist specializing in spectrographic analysis, and a metallographer.  A spectrographic analysis of Sample 2 was carried out on September 18, 1958, by means of an Applied Research Laboratories two-meter spectrograph with a dispersion of 5 Angstrom per millimeter.   The spectrographic analyst confirmed that magnesium was the major ingredient, but he detected aluminum, silicon and iron in the 100 to 1,000 ppm range (see Table 1).  Lorenzen's colleague wrote that "as far as this writer has been able to ascertain, no commercial alloy of magnesium exists with a composition at all like that of the sample.  The metal of the sample is of no conceivable use for mechanical purposes, or for the conduction of electricity." 

Mrs. Coral Lorenzen also arranged for part of Sample 2 to be spectroscopically analyzed by the Dow Chemical Corporation.  The results of their analysis were transmitted in a letter dated December 15, 1961, to Mrs. Lorenzen  from Dr. R.S. Busk,  Research Director of the Dow Metal Products Department of the Dow Chemical Company, Midland, Michigan. Busk’s analysis (Table 1, Dow Sample 2) indicates that calcium is present at 100 ppm, and strontium and barium each at 30 ppm.  It should be noted that these results have been misrepresented in the UFO literature.  W.W. Walker and R.W. Johnson prepared a report for APRO in 1969 - 1970, but their report was not published until 1992 (Walker & Johnson 1992), when it appeared in the Journal of UFO Studies with an introduction by Swords (1992). Unfortunately, Walker and Johnson misread Dr. Busk's table, so that entries in their version of the Dow analysis are all too high by a factor of 100. 

In 1967, the Lorenzens contacted Dr. Edward U. Condon, who was serving as director of the Colorado Project, to examine UFO evidence under contract with the U.S. Air Force.  Analysis of the Brazil magnesium specimen was assigned to Dr. Roy Craig, who has given a narrative account of his experiences with the Colorado Project (Craig 1995). The results of his investigations into the Brazil magnesium are summarized in the Condon Report (Condon & Gillmor 1969, pp. 94 – 97). 

According to Craig, he contacted and visited Dr. Busk early in 1968. Busk informed Craig that Dow had, for about 25 years, produced a number of batches of very pure magnesium by the process of repeated sublimation, and provided him with a specimen of triply sublimed magnesium.  Craig was advised that the most sensitive tests for impurities would be neutron activation analysis. He therefore arranged to take a specimen of the Brazil magnesium and (for comparison) a specimen of the Dow triply sublimed magnesium, to the Alcohol, Tobacco, and Firearms Laboratory, Washington, D.C.  This visit took place on February 5, 1968, and the specimen was analyzed by Mr. Maynard J. Pro, whose report on the analysis was mailed to Craig on February 29, 1968. 

The results of Pro’s analysis of the Brazil specimen and of the Dow specimen are included in the Condon Report (Condon & Gillmor 1969, pp. 94 – 97). They are summarized in Table 1. Clearly, this specimen of the Brazil magnesium was not "100% pure."  Indeed, it was not as pure as the triply sublimed Dow specimen. 

In his report, Condon comments that 

Like many of Condon’s statements in his report (see, for instance, Sturrock 1987), this comment is not quite accurate: the specimen analyzed at the Alcohol, Tobacco and Firearms Laboratory was a piece of triply sublimed magnesium, not a piece of “regular commercial metal.” 

On the other hand, the results of the ATF analysis (see SU-E in Table 1) did seem somewhat unusual to the magnesium experts at the Dow Chemical laboratory.  The Brazil magnesium contains significant amounts of both barium and strontium.  As Craig reports 
 
 

Craig learned that the Dow Metallurgical Laboratory had over the years produced a number of experimental batches of magnesium containing strontium. However, there is no indication that any of the batches contained exactly the same impurities as those found in the Brazil magnesium. 

Dr. Donald Beaman and Dr. Laurence Solaski of the Dow Chemical Company, during a meeting with the author on January 17, 1983, also expressed surprise at the presence of strontium in the specimens. They pointed out that the strontium would have to have been added, since it is not a “natural” impurity in magnesium production. Dr. S. Lawrence Couling, of the Battelle Columbus Laboratories, expressed a similar opinion during a telephone conversation with the author on May 11, 1984. He remarked that the presence of strontium in magnesium metal is “very, very unusual.” He did not know of any place in magnesium technology where strontium is used. 

I became interested in the Condon Report in 1976 and prepared an evaluation of that report that was subsequently published in this journal (Sturrock 1987).  I thereby became interested in the only material that had been investigated by the Colorado Project, namely the Brazil magnesium specimens. With the kind cooperation of the Lorenzens, I was able to arrange for some analyses in California. The most useful of these was also the first. I was able to arrange for an isotopic analysis to be carried out at the Division of Geological and Planetary Sciences of the California Institute of Technology, Pasadena.  In October 1976, Mr. Typhoon Lee and Dr. P.A. Papanastassiou performed a mass-spectrographic analysis aimed specifically at determining the isotopic composition of the magnesium (Lee & Papanastassiou 1976). They were able to determine that, with an accuracy of 0.04% (400 ppm), there is no significant difference between the isotopic composition of the specimen I had provided and that of normal terrestrial magnesium that had been subjected to normal fractionation processes such as sublimation. 

Through the kind cooperation of M. Jean-Jacques Velasco of the French space agency CNES (Centre National d’Etudes Spatiales) in Toulouse, I was able to arrange for an independent determination of the isotopic ratio. This was carried out in 1986 by Professor J.C. Lorin and Dr. A. Havette of the Laboratoire de Mineralogie-Cristallographie of the Pierre and Marie Curie University in Paris. Their report (Lorin & Havette 1986) confirmed that the isotopic composition of the specimen they analyzed differs from the normal terrestrial composition by less than 0.2% (2000 ppm). Lorin and Havette also determined that the specimen contained calcium at 8,500 ppm and strontium at 700 ppm. 

The Lorenzens kindly transferred ownership of the remaining specimens to me in 1987.  It should be noted that, by that time, the association of the remaining specimens with the original three specimens (A, B and C), had been completely lost.  The specimens had not been carefully protected and tracked. In my discussions with the Lorenzens, I learned that two specimens were out on loan.  One was in the possession of Mr. Robert Achzehnov of Costa Mesa, California; I subsequently retrieved this specimen from Mr. Achzehnov in 1986.  The other has a more interesting history. 

Mr. Harold Lebelson, a journalist, had expressed an interest in the Brazil magnesium in 1978.  As a result, a specimen (the same specimen that had been analyzed by the Colorado project) was given into his care by the Lorenzens.  He took this specimen to Professor Robert E. Ogilvie of the Metallurgy Department at MIT. The results of Ogilvie’s analysis were reported by Lebelson in an article in OMNI magazine (Lebelson 1979), which reads in part: 
 
 

The first part of this report is not accurate. I visited Ogilvie and discussed his analysis with him in June 1982, when he informed me that he had in fact detected impurities in the Brazil specimen, including calcium at a few thousand ppm, and strontium, iron and zinc at lower concentrations.. 

In 1986, with the agreement of the Lorenzens, I contacted Ogilvie to see if I could take possession of the specimen he had been analyzing. However, on telephoning him, I was dismayed to learn that he no longer had the specimen.  According to Ogilvie, Lebelson had telephoned him in 1984 and advised him that someone would visit Ogilvie on Lebelson's behalf to retrieve the specimen.  Soon thereafter, a gentleman turned up at Ogilvie's laboratory and took possession of the magnesium.  Ogilvie did not recall the person's name, did not check his credentials, and did not ask for a receipt.  All that Ogilvie could remember about the visitor was that he said he was from the IBM plant in Fishkill, New York.  I telephoned Lebelson to ask what had happened to the magnesium as I wished to retrieve it, but Lebelson responded that he had never telephoned Ogilvie to authorize anyone to pick up the specimen. So, another piece of the Brazil magnesium was lost. 

 Ogilvie more than once recounted to me a conversation with Beaman who, according to Ogilvie, had told him that, in the late 1950s, an Air Force lieutenant brought a magnesium specimen from Ubatuba to his laboratory for analysis. Beaman had found calcium, strontium, and other elements in the specimen. The officer took away all records, putting them in a briefcase that was chained to his wrist. However, when I later asked Beaman about this episode, he stated that it had never happened. 

 During our conversations, Ogilvie made a very interesting suggestion. He pointed out that the Brazil magnesium seemed to originate in an event that occurred in September 1957, only one month before the Russians launched Sputnik One (October 4, 1957). Ogilvie hypothesized that the Russians had attempted to launch a Sputnik (that he referred to as “Sputnik Zero”) in September, but it did not make the required orbit, and crashed near Ubatuba. This seemed a plausible hypothesis, since it would explain the apparent interest of military and intelligence agencies in this material. However, I addressed an inquiry to Dr. Vladimir Rubtsov of the Research Institute on Anomalous Phenomena in Kharkov, who replied as follows:   

In 1999, I finally acquired a sample of magnesium produced in Russia in the 1950s. The results of my comparison of the Russian magnesium and the Brazil magnesium are given in Section 3. 

In 1977, there was an interesting exchange of correspondence between the President's Science Advisor, Dr. William Press, and the Administrator of NASA, Dr. Robert Frosch.  This exchange was unusual in that it was released to the news media.  In a letter dated September 14, 1977, Press advised Frosch that the White House was receiving inquires about UFO reports and asked if NASA would be willing to investigate the subject. (Footnote 2) As reported by Professor Richard C. Henry (1988) of Johns Hopkins University, who had been working temporarily at NASA Headquarters for the period in question, this request was duly considered at NASA Headquarters. On December 21, 1977, Frosch replied to Press declining to initiate a new study but stating, in part, 

At that time, I had little understanding of the operation of government agencies. I took the offer at face value, and wrote to NASA to ask where such material should be submitted.  These inquiries led me to meet with Mr. David Williamson, Jr., of “Code AX, Special Projects.” After discussing two or three items that might be analyzed at NASA centers, Williamson agreed to arrange for the analysis of a specimen of the Brazil magnesium.  This specimen was forwarded to Dr. Richard Williams at the Johnson Space Flight Center, Houston, Texas.  I had several telephone conversations with Dr. Williams.  I attempted to find out just what kind of machine he would be using and its sensitivity.  He was not too precise in his answers.  I learned that it was "an old ARC machine," and that the sensitivity was not very good, not as good as other machines then in existence.  When pressed, Williams guessed that his instrument should be able to detect impurities with abundances of 100 ppm or more. 

In due course, I received a letter dated December 15, 1981 from Williams, giving the results of his analysis. His letter reads, in part,   

When I reported these results to Ogilvie at a later date, he remarked laconically “I’m afraid their system isn’t very good.” 

I made several other attempts to get accurate composition analysis of the Brazil specimens, without great success, until I was able to commission analyses at commercial laboratories.  In subsequent sections, I report the results of analyses carried out at the laboratories of Elemental Research Inc. in Vancouver, Canada. 

Several other analyses of the Brazil magnesium have been carried out over the years, due to the generous cooperation of the Lorenzens. One notable analysis is the metallurgical investigation carried out in 1970 by Walter W. Walker, then Associate Professor of Metallurgical Engineering at the University of Arizona, Tucson, and Dr. Robert W. Johnson, then Development Metallurgist in the Advanced Materials Division of the Materials Research Corporation in Orangeburg, New York. Their report was finally published in 1992 (Walker & Johnson 1992) in the Journal of UFO Studies. The same issue of that journal also contains a valuable historical introduction to the Brazil magnesium by Michael D. Swords (1992), Professor in the General Studies Department at Western Michigan University in Kalamazoo, and a commentary by Walker (1992) reviewing the available information about the material. 

Fontes (1962), in describing the specimens that he acquired from Sued, commented on their appearance as follows:  "three small pieces of a dull gray solid substance...Their surfaces were not smooth and polished, but quite irregular and apparently strongly oxidized....the surfaces of all samples were covered with a whitish material....The fine, dry powder filled the fissures and cracks on the surface of the first sample." It is not easy to reconcile the gray appearance of the specimens with the writer's account of their origin. If a magnesium object were suddenly fragmented, the fragments would be shiny, not dull gray. Magnesium becomes dull only slowly.  In a reasonably dry climate, it remains shiny for months and even years.  In a damp climate, it will corrode more rapidly. 

However, if a complex device were to explode, the metal parts may well be contaminated by material from parts that were not constructed of magnesium.  Furthermore, the person who sent the letter and specimens to Sued had claimed that the event occurred near the water's edge (in or near Ubatuba).  If this were the case, it is possible that magnesium fragments were heated to a high temperature and then dropped into sea water.  They might then acquire a coating of minerals from the sea water and from the sand, and this could account for the whitish-gray surface material. 

As a result of this speculation, it appeared that it would be worth while to determine the composition of material on the surface of the specimens.  If it were found that the surface abundances were a good match to the element abundances of sea water and of the sand on one of the beaches in the Ubatuba area, that would tend to corroborate the writer's story, and it might even help identify the actual location, since different beaches in that area have different types of sand. 

In April 1978, the surface composition of one of the specimens (SU-D) was determined by analyst Mr. Chris Zercher of the Center for Materials Research at Stanford University using a KEVEX electron microprobe.  (Footnote 3) Zercher found that the principal constituent was magnesium, but detected calcium, chlorine, iron, silicon and titanium, each at about 2,000 ppm.  The material was also analyzed by a Laue diffraction analysis, and found to be mainly Mg(OH)2. 

More recently (January 1997), Professor Michael Kelley, also of the Center for Materials Research at Stanford University, carried out a similar analysis using an XPS (X-ray Photoelectron Spectroscopy) instrument, but this proved to be much less sensitive than we required, and he was able to identify only a few elements.  According to Kelley, the gray surface material contains the following elements: oxygen 22.9% (by number), magnesium 10.6%, carbon 66.1% and chlorine 0.4 %, indicating the presence of Mg O and Mg Cl2.  (Note that some elements - including oxygen, carbon and chlorine do not show up in the positive-beam analyses that were used in the SIMS-type (Secondary Mass Ionization Spectrometer) instruments to be discussed later.) 

On consulting Dr. Robert Odom, Manager of Contract Research at Charles Evans and Associates, Redwood City, California, in the spring of 1997, I was advised that the most sensitive surface analysis would be provided by a device developed in that laboratory that is referred to as ToF-SIMS (Time-of-Flight Secondary-Ion Mass Spectrometer).  The top monolayer of the surface of the specimen is zapped with a beam that is pulsed, the pulse duration being of order one nanosecond.  Ions produced in this way are accelerated by an electric field, travel some distance, and are analyzed by a mass spectrometer, taking into account the different flight times of ions of different mass.  Information from the mass spectrometer, which yields the mass-to-charge ratio, is registered as a function of time. This procedure, which has a sensitivity of better than 1 ppm, is regarded by the analysts as a qualitative technique for determining the composition of the top 1 – 4 monolayers. By comparison the analysts regard ICP (inductively-coupled-plasma) instrumentation as capable of providing quantitative information about the bulk composition of specimens. 

 Of course, surface analysis can be carried out only on specimens that have not yet been mounted and polished, and several of the Brazil magnesium specimens have at various times been mounted and polished for analysis. This led me to select specimens SU-C and SU-D for ToF-SIMS analysis.  The results are shown in Table 2.  The instrument gives the “intensity” of measurements of the various elements relative to magnesium. 

There are significant differences between the surfaces of SU-C and SU-D.  For instance, the abundances of lithium, sodium and titanium in SU-C are about ten times the levels of the same elements in SU-D.  Furthermore, SU-C shows a significant trace of palladium, that was not detected in SU-D.  It is clear that, judging from the surface composition, the specimens are not homogeneous. 

With the kind cooperation of Professor Kaufmann, I was able to obtain samples of sands from two beaches in the neighborhood of Ubatuba: Praia Anchiete and Praia Enseada. These were analyzed by Dr. Hugh Gotts of the Materials Analysis Group at Philips Semiconductors in  Sunnyvale, California, on March 28, 1997.  The samples were dissolved in a mixture of aqua regia and hydrofluoric acid and analyzed for bulk elemental composition by means of a Thermo-Jarrell-Ash AtomScan 25 ICP [inductively coupled plasma] Optical Emission Spectrometer (ICP-OES). The instrument was calibrated using standards containing 0, 1, 3 and 5 ppm of each element.  The calibration was then verified by running quality-control standard solutions containing 1 ppm of each element. In addition, to assure that there was no instrumental drift during the analysis, several standard zeroes and 1 ppm standards were interspersed with the extracts. 

Table 2 shows the bulk abundances of elements in seawater and in sand from these beaches, for the same elements that are found at a level of 10 ppm or more in the surface impurities of SU-C and SU-D.  Although the sodium and calcium in SU-C and SU-D may have originated in seawater and sand, it is clear that most of the impurities in the surface material cannot be explained in that way.   It is also notable that the sands in the Ubatuba area have a high iron content, but the iron content of SU-C is comparatively low, and the iron content of SU-D is undetectable. 

Later in 1997, I became aware of the capabilities of the Elemental Research laboratories in Vancouver, British Columbia, Canada, including a laser-ablation inductively-coupled-plasma mass spectrometer that can give abundances with high sensitivity and consumes only amounts of order 1micrometer3 (10-12 cm-3). I originally planned to use their services for internal analysis, but I learned that their instrument could be used also for analysis of the top few micrometers of a surface. I therefore arranged for them to carry out these surface analyses of two specimens that had been sent to them for internal analysis: SU-Ib and SU-Ja.  The results of their analysis are also given in Table 2.  We see that these two specimens differ from each other, and also from SU-C and SU-D. 

On comparing SU-Ib and SU-Ja, we see that, except for silicon, calcium, chromium, zinc and selenium, SU-Ja has lower impurity levels than SU-Ib; of these elements, silicon and calcium could have been derived from sand. 

Comparing SU-Ib and SU-Ja with SU-C and SU-D, we see that boron, phosphorus, and zinc are present at a higher level in SU-Ib and SU-Ja than in SU-C or SU-D; on the other hand, lithium, strontium, and niobium are present at lower levels in SU-Ib than in SU-C and SU-D. This suggests that different specimens have experienced different environmental conditions, either in their original setting, or in subsequent handling and storage. Although some of the surface impurities (such as sodium and calcium) may have been derived from seawater or sand, other elements (such as titanium, chromium, iron, cobalt, selenium, strontium, yttrium, niobium, palladium and barium) are more likely to have had a technological origin. 

With the kind cooperation of M. Jean-Jacques Velasco of the French space agency CNES, I arranged in 1986 for a specimen of the Brazil magnesium, then in my possession, to be analyzed by Professor J.C. Lorin and Mme. A. Havette of the University of Paris.  Lorin and Havette used a Cameca SIMS (secondary ion mass spectrometer) instrument.  They detected calcium at the 8,500 ppm level, substantially higher than levels previously quoted.  They also detected strontium at 700 ppm, similar to the level found by the Colorado Project (see Table 1). 

In 1997, I arranged for new analyses of some of the specimens then in my possession.  I began with SIMS analysis at a local laboratory, but the analyst found that the results were erratic.  He ascribed this to the fact that the samples had not been properly mounted.  He advised me that, to get meaningful results from SIMS analysis, it would be necessary not only to mount and polish the specimens and cover them with conductors such as carbon or gold, but also to have some independent information as to the likely impurities.  It therefore became clear that SIMS analysis was not the optimum procedure. 

I learned that the Elemental Research laboratories in Vancouver, Canada, is one of the few laboratories in the world that have a laser ablation ICP-MS instrument  I was advised that this instrument can provide measurements at the 1 ppm level or better, using only a micron-cube of a specimen, over the complete range of atomic numbers. 

In September 1997, I arranged for Elemental Research to analyze two samples, SU-Ia and SU-H.  They also analyzed, at the same time, the following comparison samples: Dow CP-d (part of the Dow sample provided to, and analyzed by, the Colorado Project); ALFA-a and ALFA-2a (two samples of very pure magnesium purchased from the ALFA Corporation, Ward Hill, Massachusetts); and ISO-A (a sample of isotopically certified magnesium purchased from National Institute of Standards and Technology [NIST]).  The Dow specimen was generously made available to me by Dr. Roy Craig, who had been responsible to the analysis of the Brazil magnesium on behalf of the Colorado Project.  The results of these analyses are shown in Table 3, where I display all elements for which there was a non-zero measurement for any one of those six specimens. 

There is good agreement between the measurements of the composition of SU-Ia and SU-H, except that silicon shows up in SU-H and not in SU-Ia, and titanium appears to be more abundant in SU-Ia than in SU-H.  Calcium was detected at the 3,000 to 5,000 ppm level, which is similar to the estimate made by the analysts at the University of Paris.  Strontium was found to be present at 600 to 900 ppm, and barium at about 300 ppm; this result is approximately consistent with the results of the Colorado Project analysis.  Ignoring estimates below 1ppm, we see that barium appears (at 15 ppm) in only one of the comparison specimens, and strontium in none of the comparison specimens. 

As in our analysis of the surface composition, we note that the major impurities are calcium, strontium and barium, in that order.  These all belong in the same column (column 2) of the Periodic Table as does magnesium.  This suggests that the Brazil magnesium was produced  from material that contained not only magnesium but also calcium, strontium and barium. 

As mentioned in Section 1, Professor Robert Ogilvie of the Massachusetts Institute of Technology suggested to me that the Brazil magnesium may have come from the crash of an attempted launch of a Sputnik by the USSR before the first successful launch of Sputnik One in October 1957.  Ogilvie referred to this hypothetical spacecraft as "Sputnik Zero."  I have made inquiries into this possibility, and I have been advised by contacts in Russia that no such event occurred.  However, I decided to compare the composition of the Brazil specimens with magnesium produced in the Soviet Union.  Through the kind cooperation of Dr. Pavel Detkov of the Solikamsk Magnesium Works in Solikamsk, in the Perm region of Russia, I obtained information concerning the primary impurities of high-quality magnesium produced in the Soviet Union in the 1950s. Another correspondent in Russia kindly obtained for my analysis a small sample of magnesium alloy used in 1950-era MIG aircraft. Table 4 shows impurities in SU-H, ALFA-2a, ISO-A, a Solikamsk  specimen, and the MIG specimen, as determined by analysis at Elemental Research. We see that SU-H has higher levels of boron, calcium, strontium, and barium than the other four specimens. 

It appears that the Solikamsk magnesium has very low impurity levels, comparable with those of the ALFA specimen. It is notable that there is no trace of either strontium or barium in the Solikamsk magnesium.  The MIG material is basically a magnesium-aluminum alloy, and is therefore quite different from the Brazil magnesium. The MIG material contains only traces of strontium and barium. Clearly, the composition of the Solikamsk magnesium and of the MIG magnesium-aluminum alloy provides no support for Ogilvie’s interesting hypothesis that the Brazil magnesium came from the crash of a Russian spacecraft that preceded Sputnik One. 

As mentioned in Section 1, in December 1975 I arranged to have a specimen of the Brazil magnesium analyzed at the Meteoritic Laboratory of Professor Gerald Wasserberg of the California Institute of Technology.  I received a report from Dr. Typhoon Lee and Dr. D.A. Papanastassiou in October 1976.  Lee and Papanastassiou did not provide me with their raw data, nor with separate estimates of the abundances of the three isotopes (24Mg, 25Mg, and 26Mg), but they did present an analysis of their data indicating that the measurements were consistent with fractionation of normal magnesium.  When magnesium is heated, it tends to lose the lighter isotopes preferentially: the change in the abundance ratio 26Mg/24Mg should be twice the change in 25Mg/24Mg.  Lee and Papanastassiou informed me that their measurements were a close fit to this expectation. 

More recently, I have attempted to obtain precise measurements of the isotopic composition of the Brazil magnesium, but this has not been easy.  The first laboratory that I approached led me to believe that they could make accurate measurements by means of secondary ion mass spectrometry (SIMS) analysis, but they did not deliver on this promise. 

I next turned to Charles Evans and Associates in Redwood City.  They first attempted to measure the isotopic ratio using an unmounted specimen, but the results were very erratic, and they decided that it would be essential to mount the specimen, polish it, and then gold-coat it. Charles Evans and Associates also advised me that it is essential to have comparison specimens.  Accordingly, I supplied their analyst, Dr. Jack Cheng, with one specimen of the Brazil magnesium (SU-A); two specimens of triply sublimed Dow magnesium that had been provided to me in the 1970s (Dow A and Dow E); a specimen that had been used at Johnson Space Center for comparison, derived originally from the Baker Company (Baker A); and part of the triply-sublimed magnesium that the Dow Chemical Company had provided to the Colorado Project in the 1960s (Dow CP).  The Baker specimen was kindly provided by Dr. Williams of Johnson Space Center, and the Dow CP specimen was kindly provided by Dr. Craig of Durango, Colorado (formerly at the University of Colorado). 

Charles Evans and Associates took great care in the mounting, polishing and coating of the specimens and they made a number of runs with SIMS instrumentation.  The results are shown in Figure 1.  The measured isotopic ratio for Dow CP is very close to that expected of normal magnesium, for which 25Mg/24Mg = 0.127 and 26Mg/24Mg = 0.139.  The solid line in Figure 1 is the track to be expected if fractionation occurs due to heating.  We see that all specimens lie on that track.  It is curious that the isotopic ratios for the Dow CP specimen are quite close to the values for normal magnesium, since this specimen and also the specimens Dow A and Dow E were all produced by triple sublimation.  The mechanism for the production of the Baker specimen is unknown, but it may well have been sublimation since that is the normal procedure used to purify magnesium. 

We see that the Brazil specimen SU-A is the furthest from normal composition.  However, it is on the same track as the other specimens.  One may therefore infer that a specimen with the same isotopic composition as SU-A could be produced from normal magnesium by multiple sublimation.  Hence this analysis does not point towards a non-terrestrial origin for the specimen SU-A. 

However, these results are somewhat surprising. Sublimation (or any other form of fractionation) moves the isotopic composition away from the normal ratios (25Mg/24Mg = 0.127 and 26Mg/24Mg = 0.139 = 0.139), along the track shown in Figure 1.  However, it also tends to purify the magnesium, although this tendency is contingent upon the sublimation thermodynamics of the ensemble of elements. One therefore tends to expect that the specimen with the most deviant isotopic composition will also be the purest specimen.  However, we see that SU-A is the furthest from normal composition, but it is less pure than the DOW-CP specimen, for example.  The former has high abundance of Ca, Sr and Ba, whereas the latter is almost free of these three impurities. 

In addition to the abundance analyses reported in Sections 2 and 3, Elemental Research also analyzed certain specimens with greatly increased mass-to-charge resolution in the neighborhood of the values appropriate for the magnesium isotopes. The results are shown in the four panels of Figure 2. Panels (a) and (b) show the scans obtained for two specimens of the Brazil magnesium, SU-Ia and SU-H. Panel (c) shows the scan obtained for specimen ALFA-a, one of the standards used in this work. Panel (d) shows the scan obtained for specimen Iso-A, part of the magnesium isotopic standard obtained from the National Institute of Standards and Technology. We see that the isotopic compositions of these four specimens are indistinguishable: the isotopic composition of the Brazil magnesium is clearly compatible with terrestrial origin. 

          • It was the only material specimen investigated by the Colorado project. The Colorado Project staff conducted a good analysis, as far as it went. However, it is regrettable that the Project did not send an investigator to Brazil. The investigator may or may not have been able to obtain more information from inquiries in the Ubatuba area. However, he could have consulted with Brazilian authorities. Such inquiries might have turned up more information about the event (if real), and the investigator would have been able to address some serious questions to the analysts at the National Department of Mineral Production. It is also likely that, even without sending an investigator to Brazil, the Project might have been able to send an inquiry through official channels, and so obtain the results of tests carried out by the Brazilian Army and Navy. 

          • From the very beginning, there have been claims that some of the specimens were “ultra-pure,” more pure than magnesium produced on Earth at that time, with the implication that the magnesium was of extra-terrestrial origin. These claims are very suspect. Analysis of the material that was in APRO possession was by no means ultra-pure – it contains calcium at the 1,000 ppm level. Hence the claim of ultra-purity has rested on the possibility that Sample 1 (the subject of the original analysis in Rio de Janeiro) was significantly different from Samples 2 and 3, which were sent to APRO. However, in his letter transmitting these samples, Fontes mentions that he is also enclosing a small remaining piece of Sample 1. Unfortunately, the Lorenzens did not keep a careful log of the specimens, so it is not possible at this time to identify that piece, or even to be sure that it is still part of the remaining specimens. 

Furthermore, looking back on the reports prepared by the analysts at the Mineral Production Laboratory, it is clear that they did not justify Fontes’ claim that Sample 1 had been found to be ultra-pure. The analysts stated that they found no impurities in the sample, but they added that neither did they find impurities in a comparison specimen, and neither did they detect the usual impurities attributable to the carbon rods. Perhaps their equipment was not working properly. Perhaps the analysts were having a bad day. Or perhaps the Laboratory was deliberately withholding information. Government agencies, both in Brazil and in the United States, seem to have been either incompetent or uncommunicative in respect to the Brazil magnesium. The Brazilian Army and Navy both withheld from Fontes the results of their tests. The US Air Force either destroyed a specimen or withheld information from the Lorenzens. An analyst at a NASA laboratory claimed that he was unable to detect any impurities either in the Brazil specimen or in a comparison specimen. 

Investigators with the Colorado Project did draw attention to the presence of strontium and barium in the Brazil magnesium. According to experts at the Dow Chemical Company, no one knew of any production procedure that would lead to these impurities, and no one knew of any producer who deliberately added these elements to magnesium. It may or may not be significant that magnesium, strontium and barium all belong to the same chemical group (group 2). 

Analysis of the surface material yields a rich mixture of elements, and this fact deserves further investigation. It would be helpful to get expert advice on whether these elements could arise from contamination in a natural setting (air, water, sand, etc.), or whether they indicate that the specimens were at one time associated with materials usually found in a technological setting. 

As far as one can tell from analyses carried out to date, there is no case for believing that the Brazil magnesium specimens had an extraterrestrial origin. On the other hand, it has not proved possible to identify where the material was produced. 

However, composition analysis covers only one area of questioning concerning the Brazil magnesium. Inquiry into a UFO case normally involves investigation of the witness or witnesses. Unfortunately, the identity of the person who sent the material to Sued remains a mystery. At my instigation, Sued published a request for further information in his column dated August 21, 1985, but there was no response. 

In an attempt to gain some information relative to the origin of the Brazil magnesium, Professor Pierre Kaufmann of Sao Paulo has carried out some inquiries in the Ubatuba area, with interesting but by no means conclusive results. These investigations will be reported in a separate article. 

FOOTNOTES   

RECEBEMOS: “Prezado Sr. Ibrahim Sued. Leitor assiduo de sua coluna e seu admirador, quero proporcionar-lhe um verdadeiro furo jornalistico a respeito dos discos voardores, se e que acredita nos mesmos. Eu tambem nao acreditava no que ouvia falar e somente lia, ate ue, alguns dias atras, perto de Ubatuba,  em pescaria com varios amigos, vi um disco-voador! Aproximou-se da praia em incrivel velocidade, parecendo prestes a abater-se sobre as aguas, quando, a um impulso fantastico, elevou-se rapidamente. Atonitas, seguiamos com os olhos esse espetaculo, quando vimos o disco explodir em chamas, saindo em milhares de pedacos que pareciam fogos de artificio -- apesar de ser doze horas, ou seja, meio-dia --, com um brilho fortissimo. Esses pedacos cairam quase todos no mar, mas muitos pequenos pedacos cairam perto da praia, tendo nos recolhido um bom numero desse material, tao leve que parecia papel. Aqui junto uma pequena amostra desse material, que nao sei a quem devo confiar para analisar. Nunca li que houvesse sido recolhido um disco-voador  ou que se tivessem recolhido pedacos de um disco, a nao ser que as autoridades militares or tenham feito e usado de sigilo. Estou certo de que este assunto bastante interessara ao brilhante cronista, e mando-lhe esta em duplicata, para o jornal e para a sua residencia …” Do admirador, (assinatura ilegivel), junto, recebi detritos de um metal estranho. 

Processing, such as sublimation, that involves heating the material, will lead to changes in the ratio of the abundances of the various isotopes. The lighter isotopes “boil off” more readily, so that the mixture changes in favor of the heavier isotopes. Magnesium has three stable isotopes, with atomic weights 24, 25 and 26. The evaporation rate is inversely proportional to the atomic weight, and this leads to a relationship between the three rates of change of the abundances. 

If we write the fractional abundances of the three isotopes as F24, F25 and F26, and if the abundances change by  ?F24, ? F25 and ? F26, these changes will be related by 

REFERENCES 

Clark, J. 1998, The UFO Encyclopedia (2nd ed.; Detroit, MI: Omnigraphics). 
Condon, E.U., & Gillmor, D.S. (eds.) 1969, Scientific Study of Unidentified Flying Objects (New York, NY: Bantam Press). 
Couling, L. 1984, telephone conversation with author. 
Craig, R. 1995, UFOs: An Insider’s View of the Official Quest for Evidence (Denton, TX: University of North Texas Press). 
Fontes, O.T. 1962, Physical Evidence: A Report on the Investigation of Magnesium Samples From a UFO Explosion Over the Sea in the Ubatuba Region of Brazil, in The Great Flying Saucer Hoax: The UFO Facts and their Interpretation, by C. Lorenzen (New York, NY: William-Frederick Press), 89 – 132. 
Henry, R.C. 1988, J. Sci. Exp. 2, 93. 
Jacobs, D.M. 1975, The UFO Controversy in America (Bloomington, IN: Indiana University Press). 
Kaufmann, P. 2000, private communication. 
Krenzke, F.J., Hays, J.W., & Spell, D.L. 1958, Journal of Metals (January issue), 28 - 30. 
Lebelson, H. 1979, OMNI, November issue, 130, 132. 
Lee, T., & Papanastassiou, D.A. 1976, letter to author dated October 19, 1976. 
Lorenzen, C.  (1962) The Great Flying Saucer Hoax: The UFO Facts and their Interpretation (New York, NY: William-Frederick Press). 
Lorin, J.C., & Havette, A. 1986, Isotopic and Elemental Characterization of a Magnesium Sample of Unknown Origin Collected in Brazil in 1957 (Paris, France: Laboratoire de Mineralogie-Cristallographie, Universite P. et M. Curie). 
Sturrock, P.A. 1987, J. Sci. Exp. 1, 75. 
Sued, I. 1957, El Globo, September 14, 1957 (Rio de Janiero, Brazil). 
Swords, M.D. 1992, J. UFO Studies, New Series, 4, 1 – 5. 
Walker, W.W., & Johnson, R.W. 1992, J. UFO Studies, New Series 4, 6 – 25. 
Walker, W.W. 1992, J. UFO Studies, New Series 4, 26 – 37. 

FIGURE CAPTIONS 
Figure 1. Plot of the determination of the isotopic ratios 25Mg/24Mg and 26Mg/24Mg, as measured by Charles Evans and Associates by SIMS (secondary ion mass spectrometry) analysis, for the following specimens: (a) DOW CP, the Dow sample of triply sublimed magnesium used by the Colorado Project; (b) DOW A, a sample of triply sublimed magnesium supplied to me by Dow Chemical Company; SU-A one of the Brazil magnesium specimens; Baker A. a magnesium standard used by Johnson Space Flight Center; and DOW E another sample of triply sublimed magnesium supplied to me by Dow Chemical Company; Specimen (a) is located closest to the idea ratios 25Mg/24Mg = 0.127  and 26Mg/24Mg = 0.139 = 0.139. 

Figure 2. Isotopic composition as determined by Elemental Research Inc., using high-resolution mass scans of  ICPM (inductively coupled plasma mass spectrometer): (a) Brazil magnesium specimen SU-Ia; (b) Brazil magnesium specimen SU-H; (c) magnesium standard