by Charles Baum
group study the cycling of elements through ecosystems, and there’s
no better way to track an element than with isotopes. Carbon and ni-
trogen isotopes, both of which get progressively lighter as you move up
through the food chain, are her specialty. Fogel’s field areas tend to be
exotic: the crocodile-infested mangrove swamps of Belize, the parched
outback of Australia, the boiling springs of Yellowstone Park. Her grow-
ing scientific reputation and thoughtful mentoring style attract a steady
stream of postdocs and visitors.
I grabbed the stone and headed down the hallway mid-morning.
After the requisite niceties I got directly to the point: “I met Malcolm
Walter in Sydney.” I reached into my pocket and handed her the piece
of black chert. “Take a look at this.”
“Apex?” she asked. We all had been following the Schopf contro-
versy, and she knew that Apex Chert samples were pretty hard to
come by.
“Nope, this is apparently new. Same area but different. He’d like a
carbon-isotope value.”
Marilyn doesn’t generally betray excitement, but she immediately
knew the significance. This black fragment was one of the oldest rocks
on Earth. “Wow, that’s pretty neat!” She turned the object over in her
hands. “Yes, I think the machine is available this afternoon.” She paused.
“Why don’t you come by around two.”
I passed the time by breaking off a small piece of the chert, crush-
ing it in a mortar, immersing the powder in oil, and peering at the tiny
EARTH’S SMALLEST FOSSILS
57
glassy shards through a powerful microscope. How curious it was—
unlike any rock I’d seen before. Myriad tiny black specks, each a few
ten-thousandths of an inch across, clouded the otherwise clear, color-
less chert matrix. Unlike typical fossil microbes, which tend to occur in
clumps and filaments, these dots were uniformly dispersed. They cer-
tainly looked like carbon, but were they cells? Would they show a nega-
tive isotopic signature?
At two o’clock I showed up at Marilyn Fogel’s lab, as arranged.
Isotope experts rely on mass spectrometry, the experimental technique
of choice for measuring a sample’s isotopic ratio. Marilyn’s mass spec-
trometer for carbon-isotope analyses sits in one corner of a 20 × 40-
foot room crowded with scientific hardware. Little space is wasted, and
you have to exercise care not to bump into sensitive hardware when
squeezing between the various experimental stations. These days, mass
spectrometers tend to be highly automated and incredibly precise ma-
chines, though they still require meticulous maintenance and rigorous
standardization procedures to yield reliable results. With a machine
like Fogel’s, carbon-isotope analyses are relatively straightforward.
[Plate 4]
The mass spectrometer is a conceptually elegant analytical tool
grounded on two of the great physical laws of nature. Newton’s second
law of motion, F = ma (force equals mass times acceleration), enables
the separation of two atoms of different mass. As noted, carbon-12
and carbon-13 differ by about 8 percent in their mass, so if the two
isotopes are subjected to an identical force, then the carbon-12 atom
will accelerate about 8 percent faster than the carbon-13 atom. Mass
spectrometers accomplish this acceleration by applying a second fun-
damental law, related to electricity and magnetism: Magnetic fields
exert forces on electrically charged particles. The analytical technique
is to ionize the carbon atoms: Strip an electron from each—by zap-
ping them with a laser, for example—to yield carbon atoms with a
positive electric charge, then subject the ionized atoms to a powerful
magnetic field. In some mass spectrometers, a massive horseshoe-
shaped magnet bends the stream of carbon atoms (bending is a kind
of acceleration, as you discover when you ride a twisting roller coaster).
Carbon-12 atoms curve in a tighter arc than carbon-13, so the beam
of carbon atoms separates into two. Two detectors placed side-by-side
measure the relative amounts of the two isotopes. Alternatively,
sensitive electronic detectors measure the time of each ion’s flight:
58
GENESIS
carbon-12 atoms arrive at a target a fraction of a second before
carbon-13 atoms.
Fogel’s mass spectrometer works best with a small powdered
sample, so I had crushed and ground a chip of chert to specifications.
Rocks are generally a lot easier to prepare than plant and animal tis-
sues, which must be freeze-dried first. In duplicate, I carefully weighed
about a milligram of rock powder and tightly wrapped the powder
into a tiny ball of inert tin metal foil. Marilyn inserted a series of well-
documented carbon-isotope reference standards along with my
samples, each a crumpled metal sphere about the size of a BB, into
ports of the mass spectrometer’s automated sample holder. Then a
computer control system took over and we had to sit back and wait for
that one tantalizing number.
It takes only a few minutes per sample, but it seemed longer. The
standards always come first, of course; we have to make sure every-
thing is working properly. Finally, the machine spewed out a single
printed sheet of white paper, crammed with columns of numbers. One
number at the bottom was the key: –25.7 ± 0.5. The carbon was light—
just what we’ve learned to expect from ancient microbes! The analysis
also indicated that about a tenth of a percent of the chert’s mass was
carbon. The duplicate run soon followed: –25.9—satisfyingly consis-
tent results.
But even as we saw these enticing numbers, a nagging doubt re-
mained about the biological origin of the carbon. The oddly uniform
distribution of black specks in the chert looked nothing like fossil
forms. Indeed, the uniform spacing suggested a more chemical pro-
cess—a segregation of carbon from chert as oil drops separate from
water. Might there be nonbiological pathways to such a light carbon
signature? All known life-forms have a negative isotope signature, but
does a negative isotope value provide unambiguous proof of life? We
were convinced that the Australian chert had a fascinating story to tell.
I e-mailed Malcolm Walters right away with the exciting data and
a quick analysis of their possible significance. It would take a bit more
work, I thought, but these were certainly publishable results. His reply
came slowly, and with a disappointing surprise. Please stop working
on the samples, he asked. Evidently, Walter’s Australian colleague, pa-
leontologist Roger Summons, had been promised the chance to ana-
lyze the new find. Summons, a pioneer at extracting biomolecules from
EARTH’S SMALLEST FOSSILS
59
old rocks, had recently accepted a professorship at MIT and had al-
ready lined up a graduate student to do the work. It wouldn’t do for
our efforts to undermine that thesis project.
A deal’s a deal. I put aside the chert and we
nt back to the seemingly
endless list of other projects. But it sure was fun while it lasted.
EARTH’S OLDEST “FOSSILS”?
What does a negative carbon-isotope value tell us about an ancient
rock? This question came into focus following a surprising announce-
ment in the November 7, 1996, issue of Nature of the discovery of
Earth’s most ancient fossil carbon. The Earth’s oldest known rocks,
3.85-billion-year-old banded-iron outcrops from the remote island of
Akilia off the southwest coast of Greenland, reveal not the slightest
trace of anything that looks like a fossil. Nevertheless, these rocks may
contain a modest store of carbon. Even though the rocks have experi-
enced severe alteration through the ravages of temperature, pressure,
and time, some of that carbon is encased in the protective mineral
apatite. When Scripps Institution of Oceanography geochemist
Stephen J. Mojzsis (now at the University of Colorado) and his col-
leagues collected those rocks and performed the first carbon-isotope
analysis at UCLA in 1996, they were delighted to find light carbon, on
average a dramatic 3.7 percent lighter than reference limestone. No
known abiotic process produces that kind of value. That simple num-
ber, –37, was enough to convince many geologists that life had achieved
a firm foothold by that ancient date.
Such a result did much more than establish a world record for
ancient life. The work of Mojzsis and his colleagues seemed to narrow
the window for life’s origin, which presumably couldn’t have emerged
until after the last global sterilizing asteroid impact, roughly 4 billion
years ago. If signs of life persist in 3.85-billion-year-old rocks, then life
arose very quickly indeed.
But, as it turned out, the Akilia rocks posed problems. Earth’s old-
est rocks have been through a lot: heated and squeezed and contorted
beyond belief. Billions of years inevitably alter the fabric of a rock.
Mojzsis interpreted the Akilia formation, with its appearance of in-
tensely folded layers, as metamorphosed oceanic sediments—a per-
fectly reasonable residence for early cellular life. But when geologists
60
GENESIS
Christopher Fedo of George Washington University and Martin
Whitehouse of the Swedish Museum of Natural History performed a
more detailed geological analysis of the carbon-bearing outcrop, the
rocks proved to be an ancient molten igneous mass that gradually so-
lidified deep underground from temperatures approaching 1,000°C.
The carbon deposits must have formed under extreme metamorphic
conditions deep in the crust. Under no circumstance could those rocks
have contained life at the time of their formation.
Scientists quickly came up with a range of plausible explanations
for the light carbon. Heating experiments, which preferentially release
recent organic contaminants, revealed that some of the rocks’ carbon
is modern. It’s also possible that some natural non biological processes
also generate light carbon. Today much of the carbon cycle is regulated
by life, and all carbon compounds derived from living organisms are
isotopically light. But before the first microbial life, there could have
been equally vigorous geochemical processes that separated carbon-12
from carbon-13. If so, then isotopes alone can provide scant help in
recognizing life in Earth’s most ancient formations—or from rocks on
other worlds, for that matter.
At best, the isotopic evidence from Greenland is ambiguous. And
so, in their search for unambiguous proof of ancient life, paleontolo-
gists have had to turn to even more elusive fossils—fragments of life’s
oldest biomolecules.
5
Idiosyncrasies
The ability of the major atomic components of the cell to
combine into molecules of considerable complexity . . . is
enormous. However, the actual number of compounds that
are used in biology is relatively small, comprising only
hundreds of compounds.
Noam Lahav, Biogenesis, 1999
In a sense, you are what you eat. Nutrition labels on the side of every
packaged food underscore this biochemical fact: Fats, carbohydrates,
and proteins satisfy life’s energy requirements (i.e., calories) and pro-
vide life’s most basic molecular building blocks as well.
Carbon, the essential element of life, combines with other atoms
in every living cell to form the molecules of life. Even as ancient rocks
can entomb the original carbon isotopes from cells, so too, under the
right circumstances, they can preserve larger fragments of life’s
biomolecules. Such molecular remnants hold great promise for identi-
fying ancient life, because terrestrial life is so remarkably, uniquely id-
iosyncratic in its choice of chemical building blocks.
SYNTHETIC QUIRKINESS
Consider the example of life’s hydrocarbons, the molecular family that
includes waxes, soaps, oils, and all manner of fuels, from gasoline to
Sterno. All cells require a rich variety of these molecules, which incor-
porate long, chainlike segments of carbon and hydrogen atoms. Hy-
drocarbons, which we eat in the form of fats and oils, serve many
cellular functions, including the production of flexible cell membranes,
efficient energy storage, varied internal support structures, and more.
61
62
GENESIS
In life and in commerce, long hydrocarbon molecules are usually
made by linking smaller pieces end-to-end. When industrial chemists
want to synthesize hydrocarbons, or when these molecules arise by
natural nonbiological processes, the molecular chains are usually
lengthened one carbon group at a time. This process ordinarily yields a
suite of molecules with many different lengths, from just a few to many
dozens of carbon atoms long, but all formed by the same stepwise
mechanism.
Life builds hydrocarbons differently and in a strikingly idiosyn-
cratic way. In each cell an amazing tool kit featuring half-a-dozen dif-
ferent protein catalysts, collectively called the “fatty acid synthase,”
facilitates the assembly of hydrocarbon chains by adding units of three
carbon atoms to a growing chain and then stripping one away. The net
result is carbon addition by pairs. So life’s biochemistry is often char-
acterized by a preponderance of hydrocarbon chains with an even
number of carbon atoms: chains of 12, 14, or 16 carbon atoms occur
in preference to 11, 13, or 15. As a result, given a suite of molecules
from some unknown source and a mass spectrometer that can analyze
the size distribution of those molecules, it’s not too difficult to tell
whether the hydrocarbons came from living cells or from non-
biological processes.
Polycyclic compounds, an even more dramatic example of life’s
molecular idiosyncrasies, include a diverse group of carbon-based
molecules with several interlocking 5- and 6-member rings. A variety
of cyclic molecules are found everywhere in our environment. Even
before Earth was born, t
hey were produced abundantly by chemical
reactions in interstellar space and during star formation—processes
that littered the cosmos and seeded the primitive Earth with cyclic or-
ganic molecules. The PAHs (polycyclic aromatic hydrocarbons) found
in the Martian meteorite ALH84001 are examples of these ubiquitous
compounds. Cyclic molecules continue to be synthesized on Earth as
an inescapable by-product of all sorts of burning: They are found in
the soot of fireplaces and candles, the smoke of incinerators and forest
fires, and the exhaust of diesel engines. Travel to the remotest places on
Earth—the driest deserts of North Africa, deep ocean sediments, even
Antarctic ice—and you’ll find PAHs.
Every living cell manufactures a variety of polycyclic carbon com-
pounds but, as with hydrocarbon chains, the polycyclic compounds
IDIOSYNCRASIES
63
produced by life are much less varied than those produced by inor-
ganic processes. The 4-ring molecules called sterols, including choles-
terol, steroids, and a host of other vital biomolecules, underscore this
point. Literally hundreds of different 4-ring molecules are possible, yet
while the relatively random processes of combustion or interstellar syn-
thesis yield a complex mixture of cyclic compounds, life zeroes in al-
most exclusively on sterols and their by-products.
Again, cells employ a remarkably quirky synthesis pathway. The
first step in forming a sterol is to manufacture lots of isoprene, a 5-
carbon branching molecule (which the cell makes from three smaller
molecules). Six isoprene molecules line up end-to-end to form
CH3
A
—
=
CH2 = C – CH2 = CH2
Isoprene
B
Squalene
C
Cholesterol
Cells manufacture polycyclic molecules in an idiosyncratic three-step process. First, three small molecules link together to form isoprene (A). Then six isoprene molecules line up end-to-end to make squalene (B). Finally, squalene folds up into the 4-ring cholesterol molecule (C). In these and subsequent drawings of molecules, each short line segment represents a chemical bond between two carbon atoms.
64
GENESIS
squalene, with 30 carbon atoms—24 of them in a chain, with six single
carbon atoms branching off at regular intervals. This long molecule
then folds up into the 4-ring sterol backbone.