I pity Astronomers. They can see the objects of their affection—stars, galaxies, quasars—only remotely: as images on computer screens or as light waves projected from unsympathetic spectrographs. Yet many of us who study planets and asteroids can caress pieces of our beloved celestial bodies and induce them to reveal their innermost secrets. When I was an undergraduate astronomy major, I spent many a cold night looking through telescopes at star clusters and nebulae, and I can testify that holding a fragment of an asteroid is more emotionally rewarding; it offers a tangible connection with what might otherwise seem distant and abstract.

The asteroidal fragments that fascinate me most are the chondrites. These meteorites, which constitute more than 80 percent of those observed to fall from space, derive their name from the chondrules virtually all of them contain—tiny beads of melted material, often smaller than a rice grain, that formed before asteroids took shape early in the solar system's history. When we examine thin slices from chondrites under a microscope, they become beautiful to behold, not unlike some of the paintings by Wassily Kandinsky and other abstract artists.

Chondrites are the oldest rocks that scientists have ever touched. Radioisotope dating shows that they hark back more than 4.5 billion years to the time before the planets formed, when the solar system was still the turbulent, rotating disk of gas and dust astronomers dub the solar nebula. Their age and composition reveal that they consist of the primordial materials from which the planets, moons, asteroids and comets ultimately assembled. Most researchers believe that chondrules formed when silicate-rich clumps of dust melted to form individual liquid droplets during highly energetic events. The droplets quickly solidified and accreted—along with dust, metals and other materials—to form chondrites, which later grew to become asteroids. High-speed collisions among asteroids caused them to fragment and chip; eventually some of the debris fell to Earth as meteorites. The tangibility that appeals to me is thus not just a matter of aesthetics—these meteorites are fossils from the birth of the solar system, our firmest links to the conditions under which our own planet Earth took shape.

Yet as anthropologists know, finding fossils is just the first step to reconstructing history. The finds need to be put into a context. Inferring the birthplace of the different chondrites and their environmental cradles, however, has been difficult because we have had surprisingly scant data about the detailed structures of these varied rocks. A few years ago I made a systematic examination of the full range of the physical properties of chondrites, filling in many of the critical gaps that existed. With these data, I constructed a rough map of the structure of the ancient nebula from which the chondrites emerged.

Remarkably, the dust distribution in the map, crude as it is, resembles that of some T Tauri stellar systems. T Tauri objects vary erratically in luminosity and are enshrouded in extensive atmospheres, so they are thought to be young—or “pre–main sequence”—stars. Many of them are surrounded by dusty disks. The concordance of the solar-nebula dust map with the structure of certain T Tauri systems supports the idea that the latter are the progenitors of solar systems like our own. Chondrites thus probe not only our own deep past but offer insight into other young stellar systems in the Milky Way. Likewise, as scientists learn about the physics of these systems, they will better understand the processes that led to the formation of our own asteroids and planets.

Chondrite Characteristics

To explore the primordial solar system by analyzing chondrites, planetary scientists must first have an accurate accounting of the rocks' properties. Researchers have sorted chondrites into about a dozen basic groups distinguished by such features as their bulk chemical composition; their mix of isotopes (elements with the same number of protons but different numbers of neutrons); the number, size and types of their chondrules; and the amount of compacted dusty matrix in which the chondrules and other materials are embedded. Because each chondrite group has a distinct and narrow range of physical, chemical and isotopic characteristics, different groups that have fallen to Earth must have hailed from separate asteroids. Investigators have concocted many imaginative models to explain how the different chondrite groups initially formed, which involve such processes as gas turbulence, magnetic fields and differences in the velocities at which particles settled to the nebular midplane. The bottom line, however, has often been a vague conclusion that the various kinds of chondrites formed under “different conditions.”

Hoping to gain a better handle on what those conditions were, in 2009 I began digging through the literature with the intention of constructing a table that listed the essential properties of the major chondrite groups. Once I had the table in hand, I intended to search for correlations among the properties that might shed light on the history of each group. But the table I constructed was more than half empty; few researchers, it seemed, had been interested in gathering these kinds of data.

The only option was to do it myself. To that end, I parked myself at a microscope and examined 91 thin sections of 53 different meteorites from different chondrite groups. At a thickness of 30 microns, many minerals become transparent and their optical properties can be studied. The sections reveal a wide variety of chondrules that vary in size, shape, texture and color. Analyzing thousands of chondrules can certainly be tedious, but by persisting in this exercise in “microscopic astronomy,” I managed to fill out the table in just a couple of months. My findings did not fully resolve the “different conditions” conundrum, but the results did extend and refine ideas about where the different chondrite groups originated in the solar nebula and what their local environments were like.

Consider, first, a rare class known as enstatite chondrites that comprises only about 2 percent of all the chondrites observed to fall to Earth. These rocks are named for what typically is their most abundant mineral—enstatite (MgSiO3)—and they come in two forms, labeled EH and EL for the high or low amounts of total iron they contain. Scientists have discovered that the abundances of specific isotopes of nitrogen, oxygen, titanium, chromium and nickel in these chondrites resemble those of Earth and Mars, and they have thus surmised that the enstatite chondrites probably formed within the orbit of Mars, appreciably closer to the sun than the inferred locations of other chondrite groups.

A second set, the so-called ordinary chondrites, comprises three separate but closely related groups—labeled H, L and LL—that vary in the amounts and forms of iron they contain. “Ordinary” refers to their prevalence; together they constitute 74 percent of observed meteorite falls. The great abundance of all three groups indicates that they came from a region of the solar system that is gravitationally favorable for delivering meteorites to Earth.

John Wasson of the University of California, Los Angeles, has proposed that the ordinary chondrites hail from a region just sunward of the center of the asteroid belt, situated between the orbits of Mars and Jupiter. Asteroids lying at a distance of 2.5 times Earth's own distance from the sun (2.5 astronomical units) would complete three orbits around the sun in 12 years; Jupiter, lying at a distance of 5.2 astronomical units, completes just one orbit in this same time interval. Such a resonance means that Jupiter's enormous gravity tugs regularly on these asteroids and eventually steers many of them toward the inner solar system. In Sweden, scientists have found dozens of ordinary chondrites in 470-million-year-old rocks—a sign that ordinary chondrites have indeed been pelting Earth for more than 10 percent of our planet's 4.5-billion-year history.

A third group—the rare Rumuruti, or R, chondrites (named for the site in Kenya where the only observed fall occurred)—resembles ordinary chondrites in many of its chemical properties. They have much more matrix material, however, and significantly higher abundances of the oxygen isotope 17O relative to the lighter isotope 16O. High temperatures in the nebula tend to equalize isotopic abundances, and the farther an object gets from the sun, the more likely it is that variations in oxygen isotopes will be preserved. This imbalance in isotopes suggests that the R chondrites formed farther from the sun than ordinary chondrites.

High temperatures also break down organic compounds, which tend to be found more abundantly in the diverse class of meteorites known as carbonaceous chondrites than in other chondrite groups. Thus, carbonaceous chondrites almost certainly orbited at even greater distances from the sun than R chondrites did. The carbonaceous chondrites themselves comprise six major groups that can each be assigned more specific nebular positions on the basis of their chemical, isotopic and structural properties.

Dust to Dust

Apart from their chemical composition, the internal structures of chondrites also reveal much about the amount of dust in the immediate environs where they formed. Dust has been crucial in all stages of the solar system's evolution. As the original cloud of material that produced the sun and planets collapsed, dust grains became more effective at trapping infrared radiation; the resulting increase in temperature at the center of the cloud eventually led to the formation of a protostar. Later, dust (and, at greater distances from the center, ice) settled toward the nebular midplane and coagulated into larger clumps, eventually forming porous bodies, known as planetesimals, ranging in size from a few meters to tens of kilometers. Some of these planetesimals melted. The planets ultimately formed from a diverse suite of such melted and unmelted planetesimals; comets and asteroids most likely accreted from unmelted planetesimals having more uniform compositions.

One clue to the abundance of dust at the location where a particular chondrite group formed is the presence of dust-laden shells surrounding a silicate core in chondrules. Chondrules in certain carbonaceous chondrites, for example, commonly include a core, or “primary” chondrule, encased in a secondary spherical shell of melted, or igneous, material similar in composition to the primary chondrule. Frequently the secondary shell is itself surrounded by a tertiary shell termed an igneous rim, which is composed of finer mineral grains than are present in the central core [see micrograph at left].

Many meteorite researchers have suggested that the secondary shells were created when some original chondrule, having solidified after the initial melting event, acquired a porous dusty shell and then experienced a second, intermediate-energy event that melted the shell but not the interior chondrule. Subsequently, events of lower energy or shorter duration, or both, produced the igneous rim. Simply put, chondrite groups that contain numerous chondrules displaying the “nested shell” structure appear to have formed in dusty environments.

Multiple remelting episodes interspersed with periods during which the chondrules became enmeshed in dust would naturally produce larger chondrules with a thick secondary shell and thick igneous rim. Thus, the presence of such features is indicative of a substantial amount of dust in the environment where the chondrules formed. Chondrules encased in dust would also cool off more slowly than others because heat could not radiate away quickly. The relatively slow cooling would have, in turn, facilitated the evaporation of volatile elements, such as sodium and sulfur. Although most of the volatiles would have condensed on nearby dust (ultimately to be incorporated into the chondrites), some fraction of them would have been lost. The sodium and sulfur contents of the chondrite groups containing these large, dust-laden chondrules should therefore be lower than those of chondrite groups whose chondrules formed in dust-poor environments. I found that this is indeed the case.

By combining this and other information with the presumed locations of the parent asteroids, I developed a crude map of dust abundances across the early solar system [see box on these two pages]. The enstatite groups, which presumably formed sunward of Mars's orbit, sat in what must have been a dust-poor region; they have, for instance, few chondrules with shells or rims, and those chondrules that do sport rims have thin ones. The ordinary and R chondrites, which are next farthest from the sun, show more signs of dustiness—for example, the proportion of chondrules with igneous rims is higher and the rims are thicker than in the enstatite groups.

The dust concentration appears to have peaked in the region occupied by the carbonaceous chondrite groups with the largest chondrules and the greatest numbers of chondrules encased in secondary shells and igneous rims (those known as the CR, CV and CK groups). It then tapered off toward the locations of two carbonaceous chondrite groups (CM and CO) even farther from the sun. (Chondrules in these groups are much smaller, and far fewer of them show secondary shells and igneous rims.) The total amount of dust diminished still farther in the vicinity of the most distant carbonaceous chondrite group (CI), which contains no chondrules at all. (These are true chondrites nonetheless because the main criterion for class admission is having a chemical composition similar to that of the sun's nonvolatile elements.)

The dust distribution in this nebular map led me to conclude that our early solar system was probably similar to many of the T Tauri stars observed today—young stars like the early sun that have not yet settled down to the main business of stable hydrogen burning. The dust pattern resembles published astronomical observations of several protoplanetary disks around T Tauri stars. Because the masses of these particular disks (about 2 percent of the sun's mass) are similar to that inferred for the solar nebula, it seems that the disks provide good models for the nebula during the period of chondrule formation and chondrite assembly.

Heated Disagreement

Just what processes created chondrules are not yet understood. The first thing that any model of chondrule formation must explain is the evidence for repeated melting. The process must also have been widespread or else it would not have led to chondrules occurring in almost every chondrite group. Unfortunately, no convincing heating mechanism has been found that accounts for all chondrule properties. The multiple melting of so many chondrules rules out any of the proposed one-shot phenomena, such as supernova shock waves or gamma-ray bursts from deep space. The heat source must have been capable, on the one hand, of melting some entire chondrules (including ones several millimeters in size) but, on the other hand, of melting just the thin dust mantles around other chondrules while leaving their interiors intact. Some researchers have suggested a repeating, pulsed heat source, such as lightning bolts, but no consensus has been reached on the feasibility of generating lightning in the solar nebula.

The chondrule-formation model currently popular with astrophysicists involves shock-wave heating in the nebula. Shock waves could have been produced, for instance, by material falling into the nebula from outside. Propagation of the shock waves through dusty nebular regions could have produced enough heat to cause chondrule melting. Models relying on shock waves have their own flaws, however. First, shock waves have yet to be observed in protoplanetary disks; their existence is unproved. Second, shocks would heat huge numbers of chondrules at once but seem incapable of melting only the outer surface of individual chondrules (to form secondary shells and igneous rims) while leaving the chondrule interiors relatively cold. A third apparent flaw is that shock waves, which are localized phenomena, seem unlikely to have produced chondrules in widely separated nebular regions. The principal mechanism for the formation of chondrules thus remains a mystery.

Fifty years ago in these pages, meteorite researcher John A. Wood observed, “Only recently have we begun to study chondrules as entities. They contain a wealth of information … about the processes that have acted on them. We may be able to learn about the nature and evolution of the solar nebula, the formation of the planets, some stages of the evolution of the sun and the time scales for all these processes.” Half a century later scientists still have much to learn, but the picture provided by these primordial messengers of the solar system is at last coming into focus.