Reports of the National Center for Science Education

The Mysterious "Spheres" of Ottosdal, South Africa

Cut face of Ottosdal concretion showing typical radial fibrous internal structure of these objects. Like some of the Ottosdal concretions, this specimen consists of two intergrown concretions. This specimen is 6.2 cm wide.


The "Ottosdal objects" are spherical and subspherical objects that were found in 3.0 to 3.1 billion-year-old (Precambrian) pyrophyllite deposits in South Africa (Jackson 1992). The objects have been the subject of much attention and speculation by various fringe groups, including Christian and Hindu creationists and advocates of "ancient astronauts". These fringe groups argue that the objects are either actual or possible "Out-of-Place Artifacts" (OOPARTs), which are supposedly direct evidence of a civilization that existed either billions of years ago or before the Biblical Flood. Advocates of "ancient astronauts" further speculate that the Ottosdal objects were manufactured by intelligent extraterrestrials.

The oldest known article that advocates an artificial origin for the Ottosdal objects is Barritt (1979). This article appears in the October 2, 1979, issue of the National Enquirer as a short version of Barritt (1982), which repeats and adds much additional material to the descriptions and discussion presented in Barritt (1979).

Barritt (1982) was published in the June 11, 1982, issue of Scope Magazine. In 1982, this magazine was well known for its sensational stories and photographs. In addition to comments by an anonymous Wonderstone "mine official", Barritt (1982) includes comments from Brenda Sullivan, a South African representative of the Epigraphic Society of Arlington, Massachusetts, and Roelf Marx, Curator of the Klerksdorp Museum. According to this article, Sullivan speculated that the objects were artifacts and clear evidence of "a higher civilisation, a pre-flood civilisation about which we know virtually nothing." Barritt (1982) noted that Marx and JR McIver, a professor in the Department of Geology of the University of the Witwatersrand, in Johannesburg, South Africa, lacked a satisfactory explanation for the origin of the objects. Barritt (1982) also quoted Marx as allegedly stating that a specimen of the Ottosdal objects slowly rotated on its axis while locked in a "vibration-free" Klerksdorp Museum display case.

Later, Jochmans (1995), a young-earth creationist, included the Ottosdal objects in his list of "top ten outof- place artifacts" and described the objects as being composed of "manufactured metal" and a "nickel-steel alloy which does not occur naturally." He clearly claims that these objects are artificial in origin. In his short discussion of the objects, Jochmans (1995) repeats the claim, possibly taken from Barritt (1979, 1982), that Marx had observed one of the objects slowly rotating on its axis while locked in a "vibration-free" display case. Inspired by Jimison (1982) — whose 1982 article appeared shortly after Barritt's and may have been derived from it — Hindu creationists Cremo and Thompson (1993, 1999) published a short description of the Ottosdal objects after corresponding with Marx. They argued that the Ottosdal objects are a possible example of artifacts having been found in geologic strata as old as 2.8 billion years. They discounted the identification of these objects as limonite concretions made by AA Bisschoff, a geologist at the University of Potchefstroom, because the objects were supposedly harder than steel, had grooves that appeared unnatural, and did not have the form and other characteristics of concretions.

On February 25, 1996, the National Broadcasting Company, a US television network, broadcast "The Mysterious Origins of Man" (for a description, see BC Video, 1996). The program contained a short segment on the Ottosdal objects. It described these objects as "metallic spheres" with fine grooves encircling them. The program claimed that anonymous "lab technicians", later revealed by Cremo as working for the Emerald City Metallurgical Engineering Company, could not find any explanation for the grooves. BC Video (1996) confused the Klerksdorp Museum with the Ottosdal pyrophyllite mines by stating that the objects were found in mines at Klerksdorp. The "Stratographic Column" [sic] web page (BC Video 2003) stated: "Perhaps the oldest artifacts ever discovered are these metallic spheres found in Klerksdorp, So. Africa."

In a web site, which briefly appeared on the Internet (Anonymous 2001), a three-grooved Ottosdal object was promoted as an alien artifact called the "Cosmos". In addition to rehashing material from a number of other sources, this web page offered the opinion of Elizabeth Klarer, a South African psychic and UFO enthusiast. She proposed that this Ottosdal object had been placed in the pyrophyllite by an "advanced race" and has an "optic disc", which "contains secrets of the universe". She predicted that a "chosen person" would open the optic disc and use its "secrets" to save the earth. Most importantly, the "Cosmos"web site (Anonymous 2001), contained several close-up photographs of a three-grooved Ottosdal Object from various angles.

For a brief period of time, a Klerksdorp Museum web page (Klerksdorp Museum 2002), contained the text from a letter from John Hund of Pietersburg, South Africa. This letter provided an account, which remains unsubstantiated, of the alleged results of an examination of an Ottosdal object by the California Space Institute, a multi-campus research unit of the University of California. The letter stated that scientists at the California Space Institute tested an Ottosdal object and concluded that its balance "... is so fine, it exceeded the limit of their measuring technology ..." and "... to within one-hundred thousandths of an inch from absolute perfection ..." This implication of these alleged findings is that no known natural process can explain the formation of the Ottosdal object. The letter also stated, by way of further qualifications, that the California Space Institute was the organization that made gyroscopes for the National Aeronautics and Space Administration (NASA).

Numerous other web pages and message boards have discussed the Ottosdal object after Klerksdorp Museum (2002). Typically, they consist of rehashed, quoted, or paraphrased material from Cremo and Thompson (1993, 1999), Jochmans (1995), Govradhan Hill Publishing (1996), Heinrich (1996), Klerksdorp Museum (2002), or some combination of these sources. However, little of what is on these pages represents any new or better information.


Discussion of the physical nature and origin of the "spherical" objects by conventional scientists is limited to Nel and others (1937) and popular articles by Cairncross (1988), Pope and Cairncross (1988), and Heinrich (1996, 1997). Nel and others (1937), who first described the geology and physical characteristics of the pyrophyllite deposits, simply report the occurrence of pyrite concretions within them. In response to Barritt (1982), another article, and an episode of a 1980s South African Sunday television program called "50-50", Cairncross (1988) and Pope and Cairncross (1988) argued that the Ottosdal objects are natural concretions. Cairncross (1988) noted that the grooves on these objects are often exhibited by concretions and reflect the layering of the sediments in which they grew. In an internet report on these objects, Heinrich (1996) speculated that the objects were possibly of metamorphic origin. Firsthand observations of specimens of the Ottosdal objects by Heinrich (1997) noted that these objects are neither the "perfectly round"nor "singular"objects as claimed by creationists and other fringe groups. To demonstrate the true nature of these objects, it is necessary to examine both the objects and the literature that has grown around them systematically.


To investigate the physical nature and origin of the Ottosdal objects, the pertinent literature was reviewed. This review included studying popular articles, books, and web pages, and various scientific papers on the geology of the Precambrian strata containing them, relevant mineralogy, concretion formation, and various other topics. Additionally, attempts were made to verify the various opinions and observations, which had been posted to various web pages, for example at the Klerksdorp Museum (2002).

I was also able to examine the actual specimens of the Ottosdal objects to determine their physical properties. Susan J Webb of the University of the Witwatersrand and Allan Frazier of Online Minerals acquired five Ottosdal objects for me to examine. After being photographed, three of these specimens were sliced on a trim saw. A sample from one specimen was analyzed using petrographic techniques. Samples from two specimens, Ottosdal-2 and Ottosdal-4, were analyzed using X-ray diffraction techniques. In addition, a sample of pyrophyllite taken from the same mine as the objects was analyzed with petrographic and X-ray diffraction techniques.

Two images. Examples of concretions from the Ottosdal pyrophyllite quarries showing variations in size and morphology. A shows sub-spherical Ottosdal concretion with poorly developed longitudinal groove. This is the same concretion whose internal structure is shown in the first image. B shows disk-shaped Ottosdal concretion composed of hematite.


Barritt (1982) shows a photograph exhibiting the empty spaces left by Ottosdal objects in the face of a cut in the pyrophyllite quarry. The photograph shows that the objects are not randomly scattered through the pyrophyllite, but occur as a very narrow layer, perhaps in volcanic deposits that were later metamorphosed to pyrophyllite.


A number of sources describe the Ottosdal objects as being spherical. Barritt (1982) initially describes them as having three longitudinal grooves and being "... so perfectly made that they look though they were cast from a mould". Barritt (1982) quotes both Marx and Sullivan as referring to these objects as "spheres". Pope and Cairncross (1988) describe the objects as being "almost perfect spheres", while Cairncross (1988) simply described them as being "round." Cremo and Thompson (1993, 1999) and Govradhan Hill Publishing (1996) further claim that the Ottosdal objects are "metallic spheres" and are "isolated and perfectly round". They state that at least one of these objects exhibits three grooves. They show a photograph in which it appears spherical. BC Video (2003) and John Hunt, as quoted in Klerksdorp Museum (2002), simply described the objects as "metallic spheres".

Two images. C shows Specimen composed of individual Ottosdal concretions intergrown like soap bubbles. D shows Ottodsal concretion with single longitudinal groove. This specimen is composed of wollastonite.

In contrast, various sources also describe the Ottosdal objects as having shapes that are neither true spheres nor "perfectly round". For example, a photograph on the last page of Barritt (1982) shows a three-grooved Ottosdal object that is clearly an ellipsoid. Barritt (1982) also gives the dimensions of a specimen in the Klerksdorp Museum as being "exactly" 3.3 cm (1.3 inches) high and 4.0 cm (1.6 inches) long. Barritt (1982) further contradicts himself and other fringe publications by quoting an anonymous mine official as stating that all of these objects are "oval" in shape. Jochmans (1995) also contradicts himself by describing them as "... metallic spheroids look [sic] like flattened globes ..." Finally, Roelf Marx (personal correspondence in 1996, including an "information sheet" on Ottosdal objects) notes that the Ottosdal objects, which he has observed, are not all spheres, but "some" of them are "oblong in form". From these descriptions, it is apparent that the authors have either greatly exaggerated the spherical nature of these objects or have been very careless in their descriptions of their shapes.

As shown in photographs that were once posted to the Cosmos web page, Anonymous (2001), the Ottosdal object exhibiting three grooves is not perfectly spherical as various authors claim. Judging from the photographs, this three-grooved object appears to consist of two Ottosdal objects that have closely intergrown together. Additional photographs of another grooved Ottosdal object in the Klerksdorp Museum, which were sent to me by van Heerden (personal correspondence, including an article, an "information sheet," and pictures of Ottosdal objects, in 2007), also clearly show that the object is not perfectly spherical.

Hund, as cited in Klerksdorp Museum (2002), claimed that an Ottosdal Object examined by the California Space Institute was balanced "... so fine, it exceeded the limit of their measuring technology ..." and "... to within one-hundred thousandths of an inch from absolute perfection ..." In personal correspondence in 2002, Arnold, who works at the California Space Institute, indicated that he remembered examining an Ottosdal Object, that Hund had loaned them. However, Arnold denied that anyone told Hund that the object had the extraordinary properties described in the letter as quoted by Klerksdorp Museum (2002). He suggested that there was "some error in transmission" and that Hund had completely misunderstood what had been told him. In addition, Arnold noted that the claim made by Hund that the California Space Institute makes gyroscopes for NASA is completely false. Judging from my correspondence and from personal examination of actual Ottosdal objects, the claim that the California Space Institute found them to be perfectly balanced and shaped spheres lacks any substance and credibility.

A careful examination of the Ottosdal objects demonstrates the imaginary nature of the "perfectly spherical" descriptions given by various authors. As first noted by Heinrich (1997), the Ottosdal objects, which were collected from the Wonderstone mines by Webb and Frazier, exhibit a wide range of shapes including spheres, flattened spheres, discs, and clusters of two to four spheres grown together like soap bubbles. Although three specimens are roughly spherical, they definitely are not "perfectly round"as various fringe group authors claim. All of these Ottosdal objects, including the "Cosmos" illustrations by Anonymous (2001), are well within in the range of shapes exhibited by natural concretions.

The size of the Ottosdal objects varies over a relatively small range. Cairncross (1988) notes that these objects vary in size from a few millimeters to several centimeters. Barritt (1979, 1982) reports that they are as large as 10 cm (4 inches) in diameter. Marx (personal correspondence in 1996) reports that these objects vary in size from 3 to 5 cm (1.2 to 2 inches) in diameter. The five specimens that were studied for this paper varied from 3.6 to 8.5 cm (1.4 to 3.3 inches) in length and 1.3 to 5.2 cm (0.5 to 2.0 inches) in height. The ratio of height to maximum length of the five objects studied varied from 0.30 to 0.83.


A variety of descriptions of the composition of the Ottosdal objects have been published. For example, Jochmans (1995) claims that the Ottosdal objects are composed of a "... nickel-steel alloy, which does not occur naturally ..." The source of this claim is unknown, although it might be an imaginative elaboration of the descriptions by Barritt (1982), where they are described as "metal spheres". According to Barritt (1979, 1982), an anonymous mine employee reported that there were two types of Ottosdal objects. The employee described the first type as being solid all of the way through and composed of a bluish-white "metal" having a reddish tinge and embedded flecks of white "fibres". The second type was hollow with a thin skin and was more common. Barritt (1979, 1982) adds that this "skin" is about 0.5 centimeter (0.2 inch) thick with a sponge-like whitish center. Descriptions of these objects given by Cremo (1993, 1999) and Govradhan Hill Publishing (1996) appear to be a summary of the descriptions given by Barritt (1982). Marx (personal correspondence in 1996) reports that the Ottosdal objects have a hard concentric shell that exhibit "perfectly concentric grooves" that surround either a spongy substance or material resembling charcoal. Cairncross (1988) describes two types of Ottosdal objects. One type exhibits a brassy metallic color and the other exhibits a dark earthy brown color. Based only upon visual inspections, Cairncross (1988) speculated that the former might be composed of pyrite (an natural iron sulfide mineral) and the latter of siderite (natural iron carbonate). According to Marx (personal correspondence in 1996) and Cremo and Thompson (1993, 1999), Bisschoff concluded that the specimens, which he examined, consist of limonite. The color of the five specimens of Ottosdal objects that were studied by the author were dark reddish-brown, red, and dusky red as defined by the color chart of the Munsell Color Company (1975).

The internal structure of three Ottosdal objects, specimens Ottosdal-1, Ottosdal-2, and Ottosdal-4, was determined by cutting them open with a trim saw. All three of these objects exhibit a spectacular radial structure, which breaks into concentric shells. They are clearly natural concretions. Internally, the concretions were found to be both porous and friable. One of two noticeably "grooved spheres"which was cut on the trim saw exhibited faint ghosts of flat laminations cross-cutting its radial structure. A prominent internal lamination was specifically associated with the external groove. The cut surface also failed to support the claim that grooves had been artificially cut into the specimen.

The analysis of two Ottosdal objects, specimens Ottosdal-2 and Ottosdal-4, by X-ray diffraction techniques revealed that they consist of two different minerals. As confirmed by petrographic and two X-ray diffraction analyses, specimen Ottosdal-2 consisted of hematite, a common naturally occurring iron oxide. Xray diffraction analyses by MA Holmes of the Geosciences Department at the University of Nebraska (personal correspondence in 2007, including X-ray diffraction data and diagrams) demonstrated that specimen Ottosdal-4 consists of wollastonite (CaSiO3), a common metamorphic mineral, along with minor amounts of hematite and goethite, a hydrated iron oxide. Holmes also confirmed that Ottosdal-2 consisted of hematite.


Marx (personal correspondence in 1996), Cremo and Thompson (1993, 1999), and Govradhan Hill Publishing (1996) also claim that some of the Ottosdal objects are harder than steel. Marx further implies that this hardness is typical of all, not just one or some, of the Ottosdal objects. An examination of the five Ottosdal objects collected for this study found none of them to be harder than 4.0–5.0 on the Mohs scale (a rating of 7–8 is typical of hardened steel). Marx, who openly admits to having "no geological training", and Cremo and Thompson (1993, 1999), and Govradhan Hill Publishing (1996), whose source for the hardness claim was apparently Marx, are clearly mistaken about these objects' being harder than steel.


In correspondence sent to Bruce Cairncross (1988) and me, Marx stated that a reporter had falsely quoted what he had said about the rotation of the objects. According to him, it was true that the Ottosdal objects had rotated in their museum cases. However, he unequivocally stated that the claim by Barritt (1979, 1982) that the Klerksdorp Museum display cases were free of outside vibrations is completely false. According to his correspondence, Marx clearly told the reporter that vibrations from underground blasting in local gold mines regularly vibrated the museum's display cases and caused the Ottosdal objects to rotate. Judging from Marx's firsthand accounts, it is clear that the claim that these objects rotated under their own power is completely false.


The descriptions of the physical characteristics and properties of the Ottosdal objects found in the literature of fringe groups badly distort reality. They also show a profound lack of expertise by fringe authors in making basic observations concerning the physical characteristics of the objects that they are discussing.

The first-hand evidence indicates that the Ottosdal objects are composed largely of hematite, wollastonite, pyrite, or some combination of these minerals. Trained geologists, Nel and others (1937) and Cairncross (1988), concluded that the Ottosdal objects are composed of pyrite within the pyrophyllite deposits. The presence of Ottosdal objects composed of hematite and wollastonite is proven by X-ray diffraction and petrographic analyses. Given the difficulty of identifying fine-grained minerals from visual inspection alone, it is understandable that Cairncross (1988) confused either hematite or wollastonite with siderite. In addition, hematite and geothite are often called "limonite" when they occur as a massive earthy mass lacking any observable crystals. Thus, the identification of some of these objects as consisting of limonite by AA Bisschoff is a general specimen description for these minerals when detailed mineralogical analyses are lacking.

The internal structure of the hematite Ottosdal objects indicates that they are natural concretions that are pseudomorphs after original pyrite concretions. It is well known that limonite, goethite, and hematite will form such pseudomorphs in these situations. This transformation occurs when oxidizing chemical reactions transform pyrite into limonite, goethite, or hematite while keeping the external shape of the pyrite. The porous and friable nature of the hematite concretions is likely the result of a decrease in the volume of the concretions as they were transformed from pyrite to hematite.

The Ottosdal object composed of wollastonite is also readily explained as a natural concretion. The Wollastonite often forms as the result of the interaction of silica-rich fluids with calcium carbonate during the metamorphism of volcanic deposits to pyrophyllite, which also silicified adjacent beds of lava (Nel and others 1937). The relict structure of the object is also typical of natural deposits.

In contrast to the various observations provided by the fringe-group literature, the sizes and shapes of the Ottosdal objects fall within the range of shapes observed for natural concretions. The intergrown nature, which some of the objects exhibit, is quite typical of natural concretions. The observed and reported sizes of these objects fall well within the size range of concretions, which can vary from a few millimeters to over 6 meters (up to 18 feet) (Dietrich 1999; Raiswell and Fisher 2000).

The longitudinal grooves exhibited by some of the Ottosdal objects, as noted by Cairncross (1988), were caused by sediment laminations. The grooves in the concretions represent individual laminae within the host sediments. These laminae were slightly finergrained than overlying and underlying sediments. As the concretion grew within the sediments, it grew at a slightly slower rate within these laminae than in adjacent layers, which resulted in the formation of the grooves. How this process can produce longitudinal grooves and ridges on spherical and subspherical concretions is well illustrated by innumerable iron oxide concretions found within the Navajo Sandstone of southern Utah called "Moqui marbles" (Chan and others 2004). The longitudinal ridges and grooves exhibited by these concretions are more pronounced and irregular than those in Ottosdal objects because the sediment in which they grew is coarser than the sediments in which the Ottosdal objects formed.

It is also clear from this investigation that the fringe-groups literature contain blatantly incorrect information about the physical character of these objects. For example, the various claims that the Ottosdal objects are perfectly round are refuted by both direct observation of the actual specimens and published photographs of them. In addition, the supporters of these objects are non-natural in origin are completely wrong in their claims that the objects rotate in "vibration-free" cabinets, are "perfectly balanced," "are hard as steel", and are composed of a "... nickel-steel alloy, which does not occur naturally ..." Jochmans (1995) even incorrectly noted that the objects were found in a silver mine. It is quite clear that the those who argue for an artificial origin for these objects have based their interpretation on misconceptions and misinformation about the physical characteristics of these objects. As a result, they completely failed to make a credible case that these objects are anything other than interesting, but completely natural, geological concretions.

Finally, the case of the Ottosdal objects is not unique. It appears that lay people often mistake concretions of various shapes for intelligently designed and manufactured artifacts. For example, the Moeraki Boulders of New Zealand, which are natural "cannonball" concretions, have been mistaken for the sail weights of Chinese junks. Natural concretions found by explorers on Seymour Island, Antarctica, were misidentified as artifacts. Concretions from the bottom of the Bay of Cambay (Khambat) have also been mistaken for ancient artifacts (Heinrich 2002). In a similar case, Kuban (2006) argues that an alleged shoe print mentioned by Cremo and Thompson (1993, 1999) and other fringe archaeologists and creationists, as having been found in Triassic strata within Nevada, is "... most likely a broken ironstone concretion ..."


An examination of the Ottosdal objects indicates that they and their grooves lack any indication of being artificial. They are just another example of how concretions have been mistaken for intelligently designed and manufactured objects. The misidentification of natural objects as the by-products of "intelligent design" is an important lesson that needs to be learned by many fringe group members.


I thank Allan Fraser, Susan J Webb,and Desmond Sacco for their successful efforts at obtaining specimens of Ottosdal objects for my study. I also thank H van Heerden for pictures of Ottosdal concretions currently on display in the Klerksdorp Museum and Roelf Marx and Frans Waanders for giving copies of hard-toget handouts and articles concerning these concretions. Finally, I thank Kevin R Henke for taking the time and trouble to review this article for me.

By Paul V Heinrich
This version might differ slightly from the print publication.