In the past two decades, natural remanent magnetization carried by marine sediments and sedimentary rocks has been used extensively to monitor the history of the geomagnetic field and to constrain motion of crustal plates. But the origin of the magnetic minerals contributing to observed remanent magnetization is only now being resolved. The main reason behind that is that representative magnetic extracts are hard to obtain from marine sediments for direct observation. Previous attempts to separate the magnetic carriers in sediments and to examine their granulometry under the scanning electron microscope (SEM) have revealed the existence of large detrital, diagenetic, and meteoritic magnetite particles. These studies, however, have been limited in their ability to recognize the more magnetically stable and smaller single-domain fraction (<0.1 micron in size) of magnetite particles. Meanwhile, the existence of single-domain magnetite grains as a major remanence carrier in a variety of marine sedimentary deposits has been shown by various rock magnetic methods, but the origin of these grains was not revealed. The major part of this thesis is devoted to the development of better extraction techniques, employing the higher resolution transmission electron microscope (TEM) to identify the presence and to study the origin of these ultrafine-grained magnetite particles in marine sediments. Special emphasis is paid to search for the existence of magnetite precipitated by magnetotactic bacteria and to determine its role as the remanence carrier in certain sediments.
Several new occurrences of living magnetotactic bacteria have been discovered and investigated. Among them, the organic-rich mud from a shallow marine basin off the California coast (the Santa Barbara Basin) and the carbonate ooze from Sugarloaf Key, Florida are of particular interest. The former demonstrates that magnetotactic bacteria are able to live and flourish at depth in an open marine environment similar to that present over most of the world ocean floors; if the local marine sediments are able to preserve the bacterial magnetite particles, then they have an excellent chance for recording a stable remanent magnetization. The latter implies that there is a good chance for using magnetostratigraphy study on shallow water carbonates to unravel the history of their formation. In addition, the occurrences of magnetotactic bacteria and bacterial magnetite at a hypersaline lagoon (Laguna Figueroa) in Baja California, Mexico, a well-known and well-studied present-day analog of Precambrian stromatolites, suggest that stromatolites would be a good place to search for bacterial magnetofossils in Precambrian.
The magnetotactic organisms from all the newly studied occurrences have been isolated and examined. The magnetite crystals in them are similar in size and morphology to those previously found in magnetotactic bacteria from other environments. Three basic shapes of bacterial magnetite are cuboid, hexagonal prism, and tear-drop, which are all quite distinguishable from that (typical octahedra) of inorganically formed magnetite. In addition, all of the measured sizes of bacterial magnetite crystals fall well within the single-domain stability field of magnetite. It is this characteristic size and shape distribution of bacterial magnetite particles that enables the search for their occurrences in modern and fossil sedimentary records.
A set of calcite, aragonite, and recrystallized dolomite samples from Bahama Bank and a core sample from Laguna Figueroa that displays interlayers of flood derived sediments and laminated mats are studied to determine the possible diagenetic effects on bacterial magnetite. Euhedral bacterial magnetite crystals have been found in all three types of sediments from Bahama Bank. Apparently, the recrystallization process does not change or alter the identity of bacterial magnetite. In Laguna Figueroa core samples, the bacterial magnetite has only been observed in the surface layer (where the living magnetotactic bacteria were found) and flood derived sediments. No bacterial magnetite was detected from laminated mat samples, and rock magnetic study shows the disappearance of a significant portion of ultrafine-grained magnetite through depth in them. Iron reduction coupled with the oxidation of organic materials, which are rich in laminated mats and relatively scarce in flood derived sediments, is one possible explanation for these observations.
Numerous deep sea core samples have been examined to identify the presence of bacterial magnetite particles. To date, the oldest undoubtedly bacterial magnetite assemblage detected in deep sea core materials is from Miocene ODP Leg 101 sample 633A-023X-03. Some bacterial magnetite-like crystals have also been isolated from Oligocene DSTP Leg 73 samples, but they are not aligned in a chain or clumped together like bacterial magnetite particles extracted from modern environments are. Among varieties of deep sea sediments being studied, bacterially formed single-domain magnetite grains are found to be most abundant in calcareous sediments with high sedimentation rate, which might reflect the enhancement of preservation potential of ultrafine-grained magnetite during the period when massive carbonate deposition diluted the concentration of organic materials.
Some possible implications of surveying the fossil occurrences of bacterial magnetite were explored. One of them is using the presence of bacterial magnetite as an independent magnetic stability indicator. It seems clear that bacterial magnetite crystals should preserve their spatial orientations and magnetic remanence directions relative to the rock matrix, unless they are disrupted by major events of thermal, chemical, or physical alteration, which would result in producing a strong secondary component in the sample. Several sets of samples that have been shown by conventional paleomagnetic or rock magnetic techniques to contain either one single primary component or one main primary component plus a weak secondary component are analyzed to test this possibility. Bacterial magnetite has been found well preserved in some of them (e.g., Neogene carbonate samples from the Bahamas, Miocene Potamida Clay of Crete, Cambrian Sinskian Formation of Siberian Platform, etc.). On the other hand, no bacterial magnetite was detected from samples with well-documented overprinting records (e.g., materials from the Great Basin, Morocco, and Newfoundland).
Because bacterial magnetite formation requires iron-mediating enzymes and certain amounts of free oxygen, to trace back the earliest occurrence of bacterial magnetite in Precambrian would support constraints on some important biochemical evolutional sequences. Stromatolitic carbonate and chert samples with ages ranging from middle Archaean to late Proterozoic are studied. Euhedral bacterial magnetite chain has been found from Nama sedimentary rocks of South Africa (approximately 700-600 My) which represents the oldest bacterial magnetofossils reported to date. A chain composed of single-domain magnetite particles with fuzzy outlines has also been detected from the 2000 My Gunflint deposit. These findings support the currently accepted hypothesis about the timing of abrupt Precambrian atmospheric oxygen buildup. They also reflect the necessity for organisms to develop mechanisms for acquiring and storing extracellular iron after the Global Ocean "Rusting" event drastically reduced the availability of dissolved iron (normally in the ferrous state) in the hydrosphere.The geologic record shows this event probably occurred around Early Proterozoic as represented by worldwide-spread Banded Iron Formation deposition at that time.
