1) New measurements of tungsten isotopes in lunar rocks indicate that the Moon formed later than previously thought, between 62-150 million years after the formation of the solar system, challenging current models of early planetary formation. 2) This later formation of the Moon requires revising our understanding of the timing of events like the giant impact that formed the Earth-Moon system and the solidification of the lunar magma ocean. 3) The new timeline suggests Earth's core may have formed independently of the giant impact and that magma oceans on Earth and other terrestrial planets took longer to solidify than models predicted.

1. NATURE|Vol 450|20/27 December 2007 NEWS & VIEWS
than can be achieved with other methods. But necessitate minimizing false-positive results, Jonathan W. Uhr is in the University of Texas
Nagrath and colleagues’ one-step method does particularly when only a few putative tumour Southwestern Medical Center, 6000 Harry Hines
have its limitations. For example, it cannot cells are isolated. One way to solve this prob- Boulevard, Dallas, Texas 75390-8576, USA.
examine individual cells for genetic changes lem would be to count sufficient numbers of e-mail: jonathan.uhr@utsouthwestern.edu
associated with progress of the tumour. chromosomes or genes in candidate cells to
Nonetheless, Nagrath and colleagues’ results1 detect one of the hallmarks of a malignant 1. Nagrath, S. et al. Nature 450, 1235–1239 (2007).
2. Fidler, I. J. J. Natl Cancer Inst. 45, 773–782 (1970).
bring us closer to having a fully automated cell — a significant deviation from the two-
3. Méhes, G. et al. Am. J. Pathol. 159, 17–20 (2001).
instrument that can detect circulating tumour copies-per-cell norm. A simpler alternative 4. Meng, S. et al. Clin. Cancer Res. 10, 8152–8162 (2004).
cells with exquisite sensitivity. Such an instru- might be to test for increased expression of the 5. Riethdorf, S. et al. Clin. Cancer Res. 13, 920–928 (2007).
ment would allow routine monitoring of blood many known malignancy-associated proteins. 6. Meng, S. et al. Proc. Natl Acad. Sci. USA 101, 9393–9398
for tumour cells as part of a medical exami- In time, it might even be possible to detect and (2004).
7. Cristofanilli, M. et al. N. Engl. J. Med. 351, 781–791
nation, and could result in early detection therapeutically target or remove the cells’ organ (2004).
and treatment, with the chance of obtaining of origin at an early enough stage to prevent the 8. Went, P. T. H. et al. Hum. Pathol. 35, 122–128 (2004).
higher cure rates. But this approach would also cancer from metastasizing. ■ 9. Allard, W. J. et al. Clin. Cancer Res. 10, 6897–6904 (2004).
PLANETARY SCIENCE
A younger Moon Core formation
Silicate differentiation
Earth
Alan Brandon Moon
Old Giant impact
The most recent study of lunar rocks indicates that the Moon formed later New
than previously thought — a conclusion that requires our view of the early Old LMO start
New
history of the inner Solar System to be revised.
Earliest crust LMO final?
It was almost 40 years ago that we found out occurred in the first 60 million years or so of
what the Moon is made of: lunar samples col- Solar System history, before most 182Hf decayed 0 50 100 150 200 250
lected during the Apollo missions turned out away. It is useful for tracking metal–silicate dif- Time (million years)
to consist of rocky material similar in compo- ferentiation times, because W has an affinity
sition to that found on Earth. We also think we for iron-rich metal cores whereas Hf favours Figure 1 | The earliest history of the Moon and
know that the Moon formed from the accre- mantle silicate and oxide minerals. If core for- Earth. Time 0 marks the onset of Solar System
tion of molten and vaporized ejecta produced mation occurred when 182Hf was still actively accretion at 4.567 billion years ago, with the rest
by a collision between proto-Earth and a giant decaying, then a record is preserved in the of the scale denoting time after that event. The
previously inferred ages for the giant impact and
Mars-sized impactor1,2. But the answer to the amount of 182W present.
the onset of solidification of a lunar magma ocean
‘When?’ of Moon formation has remained elu- The amount of 182W measured in Earth (LMO)4–6 were based on older W-isotope data for
sive. On page 1206 of this issue, Touboul et al.3 samples4,5 indicates that W was removed into Moon rocks. Touboul and colleagues’ W-isotope
examine this question with tungsten-isotope Earth’s core about 30 million years after the data3 provide a new and later time for the giant
measurements of lunar rocks. Solar System began to accrete 4.567 billion impact and for the start of solidification of the
The mass of the giant impactor that led to years ago (Fig. 1). Before Touboul and col- LMO. The final solidification time of the LMO
the formation of the Moon was as much as leagues’ new study3, the lunar 182W data had is obtained from the 146Sm–142Nd chronology of
30% of Earth’s mass today, and this collision is yielded two conclusions. First, that formation Moon rocks8. The vertical band is the earliest
thought to represent the last significant growth of the Moon — and so the giant impact event — time for the giant impact that formed the Moon;
stage of our planet1. The post-collision disk of occurred around the same time as Earth’s core error bars represent 2σ uncertainty estimates for
each event. Data points for events on Earth are
ejecta and the new bulk Earth were a combi- formed at about 30 million years4,5. Second,
given for comparison.
nation of proto-Earth material and that of the that the lunar magma ocean largely solidified
impactor. The tremendous release of energy 10 million years after that6. This chronology
probably melted the Earth, producing a glo- provides a well-ordered Earth–Moon system of Touboul et al.3 show, the previous W isotope
bal-scale magma ocean. Iron-rich metal from that behaved exactly as we expected — rapid data for lunar materials were not completely
the proto-Earth and the impactor merged to accretion of Earth, core formation and corrected for 182W produced by the decay of
182
form the present Earth’s core, with Earth’s global-scale melting were coincident with Ta (tantalum). The isotope 182Ta forms when
mantle and crust forming from silicate and a giant impact, and were followed by rapid cosmic rays bombard the Moon. The excess
182
oxide (non-metal) material. The time intervals cooling and consequent solidification of the W from this process can result in spurious
between the various events — the giant impact, magma oceans of Earth and the Moon. ages using the 182Hf–182W clock.
the formation of Earth’s core, and the onset But some observations do not fit this simple When correctly accounting for 182W pro-
and solidification of magma oceans in Earth lunar history (Fig. 1). First, radiometric ages duced from 182Ta, Touboul et al.3 find that the
and the Moon — tell us when accretion was for the lunar crust, which was made from light amount of 182W in lunar samples indicates
completed and constrain the early differentia- minerals that solidified and floated to the top that the Moon could not have formed before
tion and cooling histories of the ‘terrestrial’ of the magma ocean, show that it began to form 62+ 90 million years after the initiation of Solar
−10
planets (Mercury, Venus, the Earth–Moon no earlier than 70 million to 150 million years System accretion, at least 16 million years after
system and Mars). after accretion of the Solar System began7. Sec- Earth’s core formed and no later than forma-
Enter the evidence from the 182Hf (hafnium) ond, the 146Sm (samarium) and 142Nd (neodym- tion of the first lunar crust (Fig. 1). These new
and 182W (tungsten) radiometric clock. The ium) radiometric clock shows that the lunar findings are in agreement with the slower rate
isotope 182Hf decays to 182W with a half-life of magma ocean largely began to solidify as late as of magma-ocean solidification obtained from
9 million years. This clock dates events that 215 million years8. Third, as the new lunar data ages for the lunar crust and 146Sm–142Nd data.
1169

2. NEWS & VIEWS NATURE|Vol 450|20/27 December 2007
E. A. CERNAN/NASA
50 YEARS AGO
“An experiment on ‘telepathy’
using television” by Donald
Michie & D. J. West — We
have made a small-scale trial
with the object of testing any
generalized extra-sensory
perception effect and looking
for individuals with strongly
manifested telepathic abilities
... The viewers were informed
that the cards would be drawn at
random from a pack consisting
of three types, depicting a canoe,
a wheelbarrow and a trumpet,
respectively. They were asked
to record their guesses on a
form printed in the TV Times
and post it to the programme
contractors … The pooled results
showed no significant deviation
from chance expectation. But
one entry, submitted by a Mr. B.
Downey, with 15 guesses out of
19 was considered sufficiently Moon rock — a picture
suggestive to justify further tests. from the Apollo 17 mission.
From Nature 21 December 1957. Harrison H. Schmitt is the
geologist–astronaut.
100 YEARS AGO
On the day of going to press, we
learn of the death of Lord Kelvin, The later time for the Moon’s formation we thought. Application of the 146Sm–142Nd
an announcement which will challenges the current view that the terrestrial clock to martian meteorites suggests that
be received with deep sorrow planets grew rapidly, and also challenges ideas the magma ocean in Mars took 60 million
throughout the civilised world about their early cooling histories4–6,9. It may to 100 million years to solidify11, likewise
… For the body of one who has mean that Earth and Mars took at least 50 mil- implying that Earth’s magma ocean probably
50 & 100 YEARS AGO
brought such honour to the British lion years, and possibly hundreds of millions took longer to solidify than some current mod-
nation, the only appropriate place of years, to reach their final mass (that is, 99% els predict. We need additional evidence to
of burial is Westminster Abbey. of their present size). By contrast, recent mod- further examine these issues on the earliest
els call for a more rapid accretion that took 30 history of the Earth–Moon system. The clues
ALSO: million years or less9. Interestingly, the first might lie in future samples returned from the
The increase in the efficiency of evidence from the 146Sm–142Nd clock of silicate Moon and Mars. ■
colleges and universities in this differentiation in Earth is about 40 million to Alan Brandon is at Astromaterials Research and
country is too pressing a need to 60 million years after accretion of the Solar Exploration Sciences, Johnson Space Center,
be dependent upon party politics. System began10. This probably records the NASA, 2101 Nasa Parkway, Houston, Texas
Unless our statesmen can be time of the onset of cooling of Earth’s magma 77058, USA.
made to realise the supreme ocean, which overlaps with the earliest new e-mail: alan.d.brandon@nasa.gov
importance of this matter and time for lunar formation and the giant impact
be persuaded to deal with it in a (Fig. 1). In turn, this implies that earlier core 1. Canup, R. M. Icarus 168, 433–456 (2004).
2. Pahlevan, K. & Stevenson, D. J. Earth Planet. Sci. Lett. 262,
patriotic manner, generously and formation in Earth did not necessarily coin- 438–449 (2007).
expeditiously, as if there were no cide with the development of a magma ocean, 3. Touboul, M., Kleine, T., Bourdon, B., Palme, H. & Wieler, R.
votes to retain or secure, we must and that, for a large terrestrial planet such as Nature 450, 1206–1209 (2007).
reconcile ourselves to the idea that Earth, rapid pulses of accretion, which could 4. Kleine, T., Münker, C., Mezger, K. & Palme, H. Nature 418,
952–955 (2002).
as manufacturing and distributing have been caused by a single giant impactor 5. Yin, Q. et al. Nature 418, 949–952 (2002).
people we shall in due course have that ultimately led to the formation of the 6. Kleine, T., Palme, H., Mezger, K. & Halliday, A. N. Science
to occupy a third or fourth place Moon, might have been necessary to initiate 310, 1671–1674 (2005).
among the nations of the world. In large-scale melting. 7. Norman, M. D., Borg, L. E., Nyquist, L. E. & Bogard, D. D.
Meteor. Planet. Sci. 38, 645–661 (2003).
Germany, the United States, and The long-lived magma ocean in the Moon 8. Rankenburg, K., Brandon, A. D. & Neal, C. R. Science 312,
now in Japan, rulers have learnt raises the question of how long Earth’s magma 1369–1372 (2006).
the lesson that efficient education ocean took to solidify. We do not know for cer- 9. Jacobsen, S. B. Science 300, 1513–1514 (2003).
10. Bennett, V. E., Brandon, A. D., Heiss, J. & Nutman, A.
and industrial success are related tain, but the implication of Touboul and col- Geochim. Cosmochim. Acta 71, A79 (2007).
to each other as cause and effect. leagues’ lunar W-isotope data3 is that accretion 11. Debaille, V., Brandon, A. D., Yin, Q. Z. & Jacobsen, B. Nature
From Nature 19 December 1907. and early cooling of Earth were not as rapid as 450, 525–528 (2007).
1170