At a glance
A Geological Survey of India geoscientist co-authored a study of 128 iron-rich olivine grains inside cosmic spherules, tracing the particles to ordinary and carbonaceous chondrite parent bodies.
A cosmic spherule is what remains of a speck of space dust that hit the atmosphere fast enough to melt. Most are a few tenths of a millimetre wide, smaller than the full stop at the end of this sentence. Inside 53 of them, researchers at India's National Institute of Oceanography found 128 crystals of iron-rich olivine that never fully liquefied, surviving a plunge that turns silicate rock into glass.
Those crystals have now been matched, grain by grain, against the chemistry of meteorites. The results, published in Meteoritics & Planetary Science, do not point to a single source. Roughly a fifth to a quarter of the grains resemble ordinary chondrites, the most common meteorites to land on Earth. Somewhere between a sixth and a third look carbonaceous. The largest share, anywhere from 41% to 62% depending on whether the host spherule came from Antarctic ice or Indian Ocean sediment, sits in the overlap where the two families cannot be told apart. The 53 spherules themselves were recovered from Antarctic ice and from deep-sea sediments of the Indian Ocean.
Why that matters now: Earth is being swept by roughly 5,200 tonnes of micrometeorites a year, and almost none of that material arrives with a label. If a chemical test can sort the dust by parent body, then a handful of grains sieved from ocean mud becomes a probe of the asteroid belt and the comet population that no spacecraft has to visit.
What survives a fall through a furnace
Entry heating is brutal and brief. A particle arriving at more than ten kilometres per second meets air thick enough to stop it in a second or two, and its surface reaches temperatures that melt silicate rock. Whatever melts loses its original chemistry to evaporation. A finished spherule is therefore a chemical average of everything that went into it, cooked and stirred.
Relict grains are the exception. They are pieces of the original mineral that stayed solid while the material around them turned to liquid, like finding unburnt fragments of vegetable in a pot of soup that boiled down. Because they never melted, they retain the trace-element signature of the body they came from. That is what makes them worth chasing through thousands of particles under a microscope.
The mineral in question, olivine, is a magnesium-iron silicate with two end members: forsterite, which is magnesium-rich, and fayalite, which is iron-rich. Geologists describe the mix as the Fa number, the mol per cent of the fayalite component. The NIO team restricted its analysis to grains with Fa above 10, meaning substantially iron-bearing. Magnesium-rich olivines, the common kind, had already been studied extensively. The iron-rich minority had been looked at far less.
Both mineral types trace back to chondrules, the millimetre-sized droplets that melted in the solar nebula before the planets formed. Type I chondrules carry magnesium-rich olivine. Type II chondrules carry the iron-rich variety. Find a type II crystal inside a cosmic spherule and you are holding a fragment of something that formed in the first few million years of the solar system.
Fifty-three particles under an electron beam
D. Fernandes, N. G. Rudraswami, M. Pandey and V. P. Singh worked on two collections that sample very different histories. Antarctic micrometeorites come from blue ice near India's Maitri research station, where the ice has trapped falling dust and kept it largely free of terrestrial contamination. Deep-sea sediments from the Indian Ocean hold spherules that settled onto the seafloor over tens of thousands of years, exposed to seawater and to the slow rain of terrestrial sediment.
Composition was measured by electron probe microanalysis, which fires a focused electron beam a couple of micrometres across at a polished grain and reads the X-rays that come back. That gives the abundances of major and minor oxides, and from them the element ratios the team needed: iron to manganese, iron to magnesium, plus iron and magnesium against silicon, normalised to the composition of CI chondrites, the most chemically primitive meteorites known.
The paper, Chemical compositions of Fe-rich relict olivines from cosmic spherules, reports the full dataset of 128 grains across the 53 particles, spanning several spherule types rather than focusing on a single texture.
Photo by EqualStock on Unsplash
What the numbers said
The minor oxides came out in narrow bands: manganese oxide between about 0.1 and 0.8 weight per cent, chromium oxide from roughly zero to 0.7, calcium oxide from zero to 0.6, and aluminium oxide up to about 0.2. These are small numbers, and that is the point. Trace elements behave differently in different chondrite groups, so their ratios act as a fingerprint even after major elements have been disturbed.
Plotted as molar iron-to-manganese against iron-to-magnesium, the grains fell along clear trends that straddle ordinary and carbonaceous chondrite fields. When the team assigned probable precursors, the split was 21% to 23% ordinary chondrite, 17% to 36% carbonaceous chondrite, and 41% to 62% overlapping both. The ranges differ between the Antarctic and deep-sea samples, which the authors report as the observed spread rather than a single pooled figure.
A second line of evidence came from pairing olivine with pyroxene in the same particles. Plotting Fa against Fs, the equivalent iron content in pyroxene, the assemblages lined up closely with equilibrated ordinary chondrites. The iron and magnesium ratios against CI, 0.5 to 1.0 and 1.1 to 1.7 respectively, overlap the ranges reported for interplanetary dust particles, for samples returned from Comet 81P/Wild 2, and for the asteroids Itokawa and Ryugu.
One comparison runs through the whole paper. Iron-rich relict olivines turn up in only a small percentage of cosmic spherules. Magnesium-rich relict olivines are far more common and cluster tightly around ordinary and carbonaceous chondrite compositions. The scarcity of the iron-rich type, the authors conclude, points to a supplementary contribution of ordinary chondritic material to the dust that feeds Earth's micrometeorite flux, layered on top of a carbonaceous-dominated background.
The overlap is the honest part of the result
Two-thirds of the grains in the most crowded bin cannot be assigned. That is not a failure of the analysis so much as a limit of the method. Iron is one of the elements that boils off most readily during entry, and the same heating that preserves a relict core can strip iron from the melt around it, shifting bulk and mineral compositions away from the parent value. The authors note that iron-rich grains are also more susceptible to thermal alteration than magnesium-rich ones, which is part of why they scatter.
The approach is elemental, not isotopic. Oxygen isotopes have long been the tiebreaker in cosmochemistry, because oxygen diffuses slowly and survives heating better than cations do. Without that measurement, a grain sitting inside the overlap stays ambiguous.
Collection bias complicates the picture further. A 2024 review of micrometeorite collections in Philosophical Transactions of the Royal Society A, co-authored by Rudraswami, catalogues the problems: magnetic separation misses some spherule types, seawater dissolves others before they are sampled, and terrestrial weathering alters primary minerals in sediments that have sat on the seafloor for tens of thousands of years. The Antarctic and deep-sea populations are not the same population, and the differing precursor percentages in this study reflect that.
Why a non-specialist should care about dust
Numbers first. Measurements from Antarctic snow at Dome C put the mass of micrometeorites reaching the ground at about 5,200 tonnes a year, with a total input before atmospheric entry estimated near 15,000 tonnes. Around nine-tenths of the incoming dust is destroyed on the way down. That flux dwarfs the mass delivered by conventional meteorites, which comes to less than ten tonnes a year.
This is not an abstraction for astronomers. Pandey's earlier work with the same Goa group examined what happens to that dust after it lands, estimating how much iron from dissolving and altering cosmic spherules enters seawater and becomes available to phytoplankton. In remote ocean regions far from continental dust, that extraterrestrial iron is part of the nutrient budget that drives primary productivity.
The parent-body question also connects to sample-return missions. Japan's Hayabusa2 brought back about 5.4 grams of regolith from the asteroid Ryugu, and analysis showed the material is close to CI chondrites. The Stardust mission returned grains from Comet 81P/Wild 2. Both appear in this paper as compositional comparisons. What the NIO team has shown is that a grain sieved from Indian Ocean mud, free of any mission cost, can land in the same compositional space as material returned from a known asteroid or comet.
For a geologist at the Geological Survey of India, that is the practical payoff. Cosmic spherules are the cheapest sample of the inner solar system available, provided you can read them.
Photo by Gyan Shahane on Unsplash
What comes next
The obvious follow-up is oxygen isotope work on the same grains, which would resolve much of the 41% to 62% overlap. New collections are also changing the sample base. The 2024 review points to fossil micrometeorites preserved in ancient sedimentary rock, which would extend the record of dust flux back through geological time, and to air-sampling campaigns that catch particles before they are weathered by ice or seawater.
More returned material is on the way, from Bennu and beyond, and each mission adds a calibration point for interpreting the terrestrial record. The dust that falls tonight has been falling for the whole of Earth's history. The work of sorting it into its sources is being done, one 128-grain sample at a time.
- cosmic-dust
- cosmic-spherules
- relict-olivine
- micrometeorites
- mayank-pandey

