Mozambican grass seed consumption during the Middle Stone Age.
ABSTRACT The role of starchy plants in early hominin diets and when the culinary processing of starches began have been difficult to track archaeologically. Seed collecting is conventionally perceived to have been an irrelevant activity among the Pleistocene foragers of southern Africa, on the grounds of both technological difficulty in the processing of grains and the belief that roots, fruits, and nuts, not cereals, were the basis for subsistence for the past 100,000 years and further back in time. A large assemblage of starch granules has been retrieved from the surfaces of Middle Stone Age stone tools from Mozambique, showing that early Homo sapiens relied on grass seeds starting at least 105,000 years ago, including those of sorghum grasses.
- SourceAvailable from: Brooke Starkoff[Show abstract] [Hide abstract]
ABSTRACT: International Journal of Exercise Science 7(2) : 128-139, 2014. The Paleolithic (Paleo) diet is one modeled after the perceived food consumption of early human ancestors of the Paleolithic Era, consisting of mainly meat, fish, fruit, vegetables, eggs, and nuts. The purpose of this study was to examine the effects of a Paleo diet on blood lipids, including high-density lipoprotein (HDL), low-density lipoprotein (LDL), non-HDL cholesterol, triglycerides (TG), total cholesterol (TC), and the ratio between TC and HDL (TC/HDL) in a healthy population. Healthy subjects of both genders (24 males, 20 females) were asked to eat an ad libitum Paleo diet for 10 weeks. Prior to the intervention, body weight, body fat percentage (BF%), maximal oxygen consumption (VO2max), TC, TG, HDL, and LDL were measured. These measurements were repeated following 10 weeks of a Paleo diet. As a whole, there was a significant increase in non-HDL (107.1±6.0 mg/dL to 120.2±6.5 mg/dL; P<0.01), LDL (93.1±5.4 mg/dL to 105.6±6.1 mg/dL; P<0.01), TC/HDL (3.0±0.2 to 3.3±0.2; P<0.05), and TC (168.8±5.4 mg/dL to 178.9±6.6 mg/dL; P<0.05) in healthy subjects following a Paleo diet. When stratified into groups based on initial blood lipid levels, deleterious changes were found in those with optimal HDL (82.1±3.2 mg/dL to 68.6±4.8 mg/dL; P<0.05), non-HDL (86.6±3.9 mg/dL to 101.4±4.8 mg/dL; P<0.01), TC (157.2±0.7 to 168.2±0.9 mg/dL; P<0.05), TC/HDL (2.5±0.1 to 2.7±0.1; P<0.05), and LDL (69.1±3.1 mg/dL to 83.5±4.1 mg/dL; P<0.01), whereas those within sub-optimal stratifications showed no significant changes. Subjects also decreased body weight (80.7±2.6 kg to 77.5±2.4 kg; P<0.001) and BF% (24.3±1.2% to 20.7±1.2%; P < 0.05). Our results demonstrate that an ad libitum unrestricted Paleo diet intervention is associated with deleterious changes to blood lipids in healthy subjects, despite concurrent improvements in body composition and cardiorespiratory fitness. Future research should focus on determining recommendations that embrace the positive aspects of the Paleo diet, while minimizing any deleterious impact on blood lipids in a healthy population.International Journal of Exercise Science. 01/2014; 7(2):128-139.
- Journal of Human Evolution 01/2014; · 3.87 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Eastern Africa is an important area to study early populations of Homo sapiens because subsets of those populations likely dispersed to Eurasia and subsequently throughout the globe during the Upper Pleistocene. The Middle Stone Age (MSA) archaeology of this region, particularly aspects of stone-tool technology and typology, is highly variable with only rare cases of geographic and temporal patterning. Although there are differences in timing and perhaps frequency of occurrence, those elements that make up the MSA lithic tool kit are also found at contemporaneous sites elsewhere in Africa and Eurasia, making it difficult to identify a unique archaeological signal for hominin dispersals out of eastern Africa. Rather, regional variation appears to be the outcome of possibly long-term interactions between particular physical and social environments experienced by hominin populations.Current Anthropology 12/2013; 54(S8):S234-S254. · 2.93 Impact Factor
, 1680 (2009);
Mozambican Grass Seed Consumption During the Middle Stone Age
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by
, you can order high-quality copies for your
If you wish to distribute this article to others
The following resources related to this article are available online at
here. following the guidelines
can be obtained by
Permission to republish or repurpose articles or portions of articles
Updated information and services,
): January 5, 2011 www.sciencemag.org (this infomation is current as of
version of this article at:
including high-resolution figures, can be found in the online
can be found at:
Supporting Online Material
, 1 of which can be accessed free:
cites 13 articles
2 articles hosted by HighWire Press; see:
This article has been
This article appears in the following
registered trademark of AAAS.
is aScience2009 by the American Association for the Advancement of Science; all rights reserved. The title
Copyright American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on January 5, 2011
S4). The latter (N = 27) comprise fallow deer,
elephant, and bone fragmentsassigned to general
categories (e.g., artiodactyls, canids, and uniden-
recovered, mainly of Microtus (table S4). The
mammals’ spatial distribution reveals no distinct
patterns (fig. S4).
Analyses of Level 2 indicate that hominins
carried out different activities in two distinct lo-
northwestern area, resulting in a dense concentra-
tion of microartifacts (Fig. 1 and fig. S1). Other
noteworthy aspects of this activity area include
fish exploitation (Fig. 3) and the use of chopping
tools (fig. S3).
Greater variation was seen in the activities
carried out near the hearth. Although flint knap-
ping around the hearth was less intensive, basalt
and limestone knapping was spatially restricted
to the hearth (Fig. 1 and fig. S1). The hearth area
also served as a focal point for biface modifica-
tion and for activities involving the use of chop-
ping tools, side scrapers, end scrapers, and awls
(fig. S3). The percussors and the pitted stones
suggest that nut processing may have involved
the use of fire, as recorded for modern hunter-
gatherer societies (1, 29). In addition, the differ-
ential preservation of fish and crabs, along with
their spatial distribution, suggests that they were
consumed near the hearth.
The spatial organization of hominin activities
in Level 2 thus resulted in discrete patterning of
various categories of finds. The evidence from
Pleistocene hominins carried out different activ-
ities at discrete locations. The designation of dif-
ferent areas for different activities indicates a
formalized conceptualization of living space,
often considered to reflect sophisticated cogni-
tion and thought to be unique to Homo sapiens
(3). Modern use of space requires social organi-
zation and communication between group mem-
bers, and is thought to involve kinship, gender,
age, status, and skill (2).
References and Notes
1. L. R. Binford, In Pursuit of the Past: Decoding the
Archaeological Record (Thames and Hudson, London,
2. L. Wadley, Camb. Archaeol. J. 11, 201 (2001).
3. S. McBrearty, A. S. Brooks, J. Hum. Evol. 39, 453
4. M. Vaquero, I. Pasto, J. Archaeol. Sci. 28, 1209 (2001).
5. N. Alperson-Afil, E. Hovers, Euras. Prehist. 3, 3 (2005).
6. J. D. Speth, E. Tchernov, in The Middle Paleolithic
Archaeology of Kebara Cave, O. Bar-Yosef, L. Meignen,
Eds. (Peabody Museum of Archaeology and Ethnology,
Harvard University, Cambridge, MA, 2007), pp. 165–260.
7. N. Goren-Inbar et al., Science 289, 944 (2000).
8. C. S. Feibel, in Sediments in Archaeological Contexts,
J. K. Stein, W. R. Farrand, Eds. (Univ. of Utah Press, Salt
Lake City, 2001), pp. 127–148.
9. C. S. Feibel, in Human Paleoecology in the Levantine
Corridor, N. Goren-Inbar, J. D. Speth, Eds. (Oxbow,
Oxford, 2004), pp. 21–36.
10. N. Goren-Inbar, A. Lister, E. Werker, M. Chech, Paléorient
20, 99 (1994).
11. N. Goren-Inbar, E. Werker, C. S. Feibel, The Acheulian
Site of Gesher Benot Ya'aqov. The Wood Assemblage
(Oxbow, Oxford, 2002), vol. I.
12. N. Goren-Inbar, G. Sharon, Y. Melamed, M. Kislev,
Proc. Natl. Acad. Sci. U.S.A. 99, 2455 (2002).
13. R. Rabinovich, S. Gaudzinski-Windheuser, N. Goren-Inbar,
J. Hum. Evol. 54, 134 (2008).
14. N. Goren-Inbar et al., Science 304, 725 (2004).
15. N. Alperson-Afil, Quat. Sci. Rev. 27, 1733 (2008).
16. S. Ashkenazi, K. Klass, H. K. Mienis, B. Spiro, R. Abel,
Lethaia 10.1111/j.1502-3931.2009.00178.x (2009).
17. See supporting material on Science Online.
18. B. A. Purdy, in Lithic Technology: Making and Using
Stone Tools, E. Swanson, Ed. (Mouton, Paris, 1975),
19. J. Sergant, P. Crombe, Y. Perdaen, J. Archaeol. Sci. 33,
20. B. Madsen, N. Goren-Inbar, Euras. Prehist. 2, 3 (2004).
21. G. Sharon,N. Goren-Inbar, J.Isr.Prehist. Soc. 28,55(1999).
22. S. Ashkenazi, U. Motro, N. Goren-Inbar, R. Biton,
R. Rabinovich, J. Archaeol. Sci. 32, 675 (2005).
23. Medium- and large-sized crabs often indicate selection
for consumption; see (30).
24. M. Goren, R. Ortal, Biol. Conserv. 89, 1 (1999).
25. C. Dimentman, H. J. Bromley, D. F. Por, Lake Hula:
Reconstruction of the Fauna and Hydrobiology of a Lost
Lake (Israel Academy of Sciences and Humanities,
26. I. Zohar et al., Palaeogeogr. Palaeoclimatol. Palaeoecol.
258, 292 (2008).
27. R. L. Elder, G. R. Smith, Palaeogeogr. Palaeoclimatol.
Palaeoecol. 62, 577 (1988).
28. The remains of freshwater turtles (Mauremys caspica)
(NISP = 3, MNI = 2) (table S4) consist of costals
(carapace bony plates). Two fragments belong to adult
turtles, whereas the third belongs to a juvenile. The small
size of the sample precludes further conclusions and its
distribution reveals no distinct patterns (fig. S4).
29. J. E. Yellen, Archaeological Approaches to the Present:
Models for Reconstructing the Past (Academic Press,
New York, 1977).
30. S. Ashkenazi, A. D. Huertas, B. Spiro, J. Isr. Prehist. Soc.
37, 135 (2007).
31. Supported by Israel Science Foundation grant 300/06 to
the Center of Excellence Project “The Effect of Climate
Change on the Environment and Hominins of the Upper
Jordan Valley between ca. 800 ka and 700 ka as a Basis
for Prediction of Future Scenarios.” The study of burned
flint microartifacts was supported by grants from the
German-Israeli Foundation (GIF I-896-208.4/2005) and
the Israel Science Foundation (886/02). The study of fish
remains was supported by the Irene Levi Sala CARE
Archaeological Foundation and carried out at the
Department of Zoology and the I. Meier Segals Garden
for Zoological Research at Tel Aviv University and at the
Natural History Museum (NHM) in London and Brussels.
We thank M. Goren for his help and support in the
study and collection of freshwater fish; W. Van Neer
(NHM Brussels), M. Richter, P. Campbell, and O. Crimmen
(NHM UK) for providing access to labs and collections
as well as helpful comments; A. Ben-Nun, head of
the GIS center at the Hebrew University of Jerusalem,
and I. Sharon and I. Gilead for greatly assisting
in establishing the methodology for this study;
M. Haber and S. Gorodetsky, who edited the
manuscript with their usual professionalism; and
two anonymous reviewers for their constructive and
Supporting Online Material
Materials and Methods
Figs. S1 to S4
Tables S1 to S7
17 August 2009; accepted 13 October 2009
Mozambican Grass Seed Consumption
During the Middle Stone Age
The role of starchy plants in early hominin diets and when the culinary processing of starches
began have been difficult to track archaeologically. Seed collecting is conventionally perceived
to have been an irrelevant activity among the Pleistocene foragers of southern Africa, on the
grounds of both technological difficulty in the processing of grains and the belief that roots,
fruits, and nuts, not cereals, were the basis for subsistence for the past 100,000 years and
further back in time. A large assemblage of starch granules has been retrieved from the surfaces
of Middle Stone Age stone tools from Mozambique, showing that early Homo sapiens relied on
grass seeds starting at least 105,000 years ago, including those of sorghum grasses.
Proterozoic carbonate rocks (1) located at 1300 m
he Mozambican cave site of Ngalue
(12°51.517′S, 35°11.902′E) is part of the
Niassa Rift (Fig. 1). The cave formed in
above sea level. There is a 20-m-long corridor
leading into dark chambers, which have the
most habitable space, with a useable floor area
covering >50 m2and a ceiling height of ~8 m.
The portion of the sequence and the artifacts
studied here were deposited throughout the so-
called “Middle Beds” (2), a Middle Stone Age
clast-supported and time-averaged unit with light
yellowish brown sediments that are rich in an-
gular cave spall, lithics, animal bones, and teeth.
The deposits span a time range from 105,000 to
42,000 years ago (2). Excavation in 2007 re-
trieved 555 quartz artifacts.
For this study, I chose 70 stone tools (~12%
of the Middle Stone Age assemblage) from all
main technotypological types to take into ac-
count the broadest range of potential plant uses:
scrapers (35%), core tools/grinders (25%), points
(15%), flakes (7%), and miscellaneous tools (18%)
(Table 1). I selected tools from across the entire
industrial scatter across a 13-m transect running
along the largest cave chamber. These include
Department of Archaeology, University of Calgary, Alberta,
T2N 1N4, Canada.
18 DECEMBER 2009VOL 326
on January 5, 2011
cobble-sized core implements that have the right
size and weight to be used as grinders of vege-
table material: Cores and core scrapers make up
more than one-third of the entire assemblage.
Special pieces include a rhyolite grinder/core axe,
a ground cobble, and a faceted quartz mortar. The
last two implements were flaked on one side to
create a dish. All three appear to be covered
with red or orange pigments, and in one case
there is a patina over the ochre. The lithics have
a mean mass of ~85 g (range, <1 to 1000 g) and
maximum length of ~50 mm (Table 1). About
20% lack any starch residue (n = 12 stone
tools), but 80% have some. In all, 2369 granules
were identified on these, and each tool on
average has 41 granules (range, 1 to 654). The
average number of grains on lithics is 270
times larger than that in the site’s free-standing
sediments. Moreover, the mean number of
granule types on stone tools is about 125 times
the number of classified morphotypes retrieved
from modern topsoils outside the cave. Three-
fourths of the starch grains come from scrapers
(50.5% of the total) and core/grinding tools
(25.5% of the total). About 64% of the total
assemblage is well preserved and displays
features comparable to those seen in fresh
modern specimens. Large depressed circles
(Fig. 2F) are noticeable in the center of 639
specimens (27% of the total starch assemblage).
Enlarged fissures (Fig. 2G) total 190 (8%). There
are 37 instances (~1.5% of the total) in which
clumps fused, with gelatinized grains that show
flat reliefs, expanded size, and a loss of
birefringence. Because the starch-containing ar-
tifacts come from dark chambers of the cave,
these biogenic polymers cannot derive from in
situ, naturally growing plants. Starch preserva-
tion over such a long period of time might be
due to the formation of a molecular film over the
stone surface through adsorption, which, once
complete, retards or abolishes the utilization of
the adsorbate by microbes (3).
Of the 2369 grains retrieved, 89% (n = 2112)
are Sorghum spp. (Table 1). Sorghum shows an
extremely variable complex of cultivated, wild,
and weedy taxa that defy formal taxonomy (4, 5).
All domesticated sorghums derive from Sorghum
bicolor subsp. arundinaceum (5), and they group
in three complexes, one of which is restricted to
Ethiopia. Modern sorghum grows naturally in
the Zambezian Miombos of the study region,
including Sorghum bicolor subsp. bicolor, race
Kafir; and Sorghum bicolor subsp. arundina-
ceum, the common wild sorghum. In general,
Sorghum bicolor makes starches with several
shapes, roughly polygonal and rounded, between
5 and 25 mm across (the average for modern
sorghum granules is 15 mm) (6). The sorghum
grains from the prehistoric assemblage have a
mean maximum length of 16.6 T 3.6 mm (n =
264; minimum, 6.5 mm; maximum, 25.3 mm),
which is very close to the modern value. The
ancient starch grains display morphologies like
those seen in modern Sorghum bicolor subsp.
bicolor (7–9) and in the reference materials from
Sorghum bicolor subsp. arundinaceum. On the
basis of granule shapes (7–9), another subset of
427 prehistoric, putative sorghum granules is
made up of 43% corneous endosperm (or-
thogonal subspheres) (Fig. 2, A and B), 39%
Table 1. Starch and stone tools.
Starch on lithicsLithics (n) Granules (n)Mean (range)
Scrapers (n = 25)
Cores, core tools (n = 18)
Micro discoidal core
Points (n = 10)
Point, thick base
Flakes, blades (n = 5)
Levallois flake, retouched
Other tools (n = 13)
Small pebble tool
Fig. 1. Study area and site location.
VOL 32618 DECEMBER 2009
on January 5, 2011
floury endosperm (rounded grains) (Fig. 2C),
and 18% from undetermined provenances with-
in the grain (tabular shapes) (Fig. 2D). Six grains
have vase shapes (Fig. 2E) like those of sor-
ghum and closely related taxa (9).
The starch on Mozambican lithics shows
that woody plants were also exploited during the
Middle Stone Age. Lenticular granules with
ellipsoid-to-orbicular bodies, lamellae, tenuous
centric slits, depressions, pocking, and an
incision along the equatorial plane (Fig. 2H)
(n = 83) represent 3.5% of all recovered starches.
The mean size is 23.5 T 6 mm (range, 8 to 36.6
mm). This morphology has been discovered in
the seeds, legumes, nuts, and mesocarp from
several species in the Fabaceae, Malvaceae, and
Apocynaceae (3). Another group of starches
includes a pear-shaped body with one tapered
end, ellipsoid lamellae, an eccentric cross,
sometimes a cuneiform slit, and creases (n =
68; Fig. 2L). It represents 3% of the total
assemblage. This sample shows a wide range
in size: 17 to 67 mm (mean, 37.6 T 10.8 mm).
The closest match is in the starches from the
pith of the Arecaceae; specifically, the trunk of
the African wine palm (Hyphaene petersiana).
An additional starch morphotype is pointed (n =
11, Fig. 2J). It has a hyperellipsoidal shape (it
resembles a rod in three dimensions) with a
highly eccentric cross and bent arms. The
appearance of this type is restricted to three
tools: a grinder, a core axe, and a Levallois
flake. Its mean maximum length is 40 mm
(range, 21 to 68 mm). It matches starches from
the Musaceae, namely, those from the mesocarp
and corm tissues of the African false banana
(Ensete ventricosum). Lastly, I found 12
clusters of compound or fused small granules
(~2 mm) on a grinder/core axe. The closest
equivalent is found in two members of the
Hypoxidaceae: Hypoxis hemerocallidea (the
African potato) and Hypoxis iridifolia.
Fifteen stone tools yielded large numbers of
altered starch granules. The most frequent
damage pattern is documented on stone tool
no. 61, which presents 449 instances of flattened
granules with an enlarged centric hole and a
bulge around the edges (Fig. 2F). Marked
transversal fissures appear in at least 70
additional granules from the same stone tool
(Fig. 2G). The alterations are associated with
other modifications such as loss of birefrin-
gence, breakage, surface roughening, and radi-
al fissuring. Several possibilities may account
for the observed modifications, including aging
and bacterial attack, culinary-induced modifica-
tions to the native starch grain (for example,
from boiling), and/or hydrolysis during fermen-
tation and/or grinding (10, 11). Establishing the
mode of use of starch resources through granule
alteration patterns, however, requires a different
type of research and additional proof beyond the
altered granules alone.
The Mozambican data show that Middle
Stone Age groups routinely brought starchy plants
Fig. 2. Selected archaeological granules from Sorghum spp. include the following: (A) from stone tool
no. 23, (B) from no. 18, (C) from no. 20, (D) from no. 2, (E) from no. 2, (F) from no. 56, and (G) from
no. 2; (H) from no. 2, probably a seed type from woody taxa; (I) from no. 66, probably from the trunk
of the African wine palm Hyphaene spp.; (J) from no. 18; probable source, the African false banana
(Ensete ventricosum); and (K) from no. 19; likely source, the African “potato” (Hypoxis spp.). Selected
modern reference material is from the grain endosperm of Sorghum bicolor subsp. arundinaceum
(Poaceae) (L to P). Tabular shapes are shown in (N) (abaxial and side views, respectively). Vase shapes
appear in (O) to (P). Lenticular starches from (Q) are from the legume of Delonix spp. (Fabaceae). Starch
granules from the trunk of Hyphaene petersiana (Arecaeae) are exemplified in (R). The starch from the
corm of Ensete ventricosum (Musaceae) is shown in (S), and the starch from the corm of Hypoxis
hemerocallidea (Hypoxidaceae) is in (T).
Fig. 3. Examples of high-starch–bearing grinding
and pounding implements from Ngalue cave.
18 DECEMBER 2009 VOL 326
on January 5, 2011
to their cave sites and that starch granules got
attached to and preserved on stone tools. I can-
not prove that starch from all stone tools rep-
resents direct tool function (12, 13). Core tools
and scrapers were exposed to starches more
often than other tool types. Three-quarters of the
starch assemblage comes from chipped stone,
not ground or polished tools. African Middle
Stone Age lithic repertoires do not yield large
quantities of dedicated grinders that would dem-
onstrate the processing of seeds. These early
grinders are simply modified cobbles and core
tools (Fig. 3) (14), with a suspected use that
conforms to the technological action known as
“diffuse resting percussion” and “pounding”
(15), which allow the grinding of plant
materials. It is not clear why the tools should
be mostly coated with grass starches and not so
much with other types of starch. It is possible
that high-starch–bearing grass refuse built up
considerably in the cave’s main chamber at
times of human occupation, thus coating both
tools that were used in the processing of grass
seeds and others that were not. These data imply
that early Homo sapiens from southern Africa
consumed not just underground plant staples
(16) but above-ground resources too.
References and Notes
1. S. Lächelt, The Geology and Mineral Resources of
Mozambique (Direcção Nacional de Geologia, Maputo,
2. J. Mercader, Y. Asmerom, T. Bennett, M. Raja, A. Skinner,
J. Hum. Evol. 57, 63 (2009).
3. J. Mercader, T. Bennett, M. Raja, Quat. Res. 70, 283 (2008).
4. J. M. J. de Wet, J. R. Harlan, E. G. Price, in Origins
of African Plant Domestication, J. R. Harlan,
J. M. J. De Wet, A. B. Stemler, Eds. (Mouton, Paris,
1976), pp. 453–478.
5. J. M. J. de Wet, Am. J. Bot. 65, 477 (1978).
6. National Research Council, Lost Crops of Africa.
Volume 1: Grains (National Academy Press, Washington
7. G. B. Cagampang, A. W. Kirleis, Starch 37, 253 (1985).
8. K. G. Duodu et al., J. Cereal Sci. 35, 161 (2002).
9. A. Caransa, W. G. M. Bakker, Starch 39, 381 (1987).
10. L. H. Harbers, J. Anim. Sci. 41, 1496 (1975).
11. M. Sujka, J. Jamroz, Int. Agrophys. 21, 107 (2007).
12. L. Wadley, M. Lombard, J. Archaeol. Sci. 34, 1001 (2007).
13. M. Lombard, J. Hum. Evol. 53, 406 (2007).
14. P. Van Peer et al., J. Hum. Evol. 45, 187 (2003).
15. S. A. de Beaune, Curr. Anthropol. 45, 139 (2004).
16. H. Deacon, S. Af. Archaeol. Bull. 48, 86 (1993).
17. This work was supported by the Canada Research Chairs
program, Canada Foundation for Innovation, Social
Sciences and Humanities Research Council of Canada (file
no. 10-2007-0697; CID:148244), Faculty of Social
Sciences/Department of Archeology at the University of
Calgary, and the National Geographic Society.
Supporting Online Material
Materials and Methods
Tables S1 and S2
18 September 2009; accepted 20 October 2009
Universality in Three- and Four-Body
Bound States of Ultracold Atoms
Scott E. Pollack,* Daniel Dries, Randall G. Hulet
Under certain circumstances, three or more interacting particles may form bound states. Although the
general few-body problem is not analytically solvable, the so-called Efimov trimers appear for a
system of three particles with resonant two-body interactions. The binding energies of these trimers are
predicted to be universally connected to each other, independent of the microscopic details of the
interaction. By exploiting a Feshbach resonance to widely tune the interactions between trapped
ultracold lithium atoms, we find evidence for two universally connected Efimov trimers and their
associated four-body bound states. A total of 11 precisely determined three- and four-body features are
found in the inelastic-loss spectrum. Their relative locations on either side of the resonance agree
well with universal theory, whereas a systematic deviation from universality is found when comparing
features across the resonance.
by Efimov (1) in 1970. Efimov showed that three
particles can bind in the presence of resonant
two-body interactions, even in circumstances where
any two of the particles are unable to bind.
When the two-body scattering length a is much
larger than the range of the interaction potential
r0, the three-body physics becomes independent
of the details of the short-range interaction.
Surprisingly, if one three-body bound state exists,
then another can be found by increasing a by a
universal scaling factor, and so on, resulting in an
infinite number of trimer states (2). Universality
is expected to persist with the addition of a fourth
particle (3–7), with two four-body states associated
with each trimer (5, 7). Intimately tied to the three-
body state, these tetramers do not require any
additional parameters to describe their properties.
na is the universally connected series of
three-body bounds states first predicted
Ultracold atoms are ideal systems for exploring
these weakly bound few-body states because of
their inherent sensitivity to low-energy phenome-
na, as well as the ability afforded by Feshbach
resonances to continuously tune the interatomic
interactions. Pioneering experiments with trapped,
ultracold atoms have obtained signatures of
individual Efimov states (8–12)—as well as two
successive Efimov states (13, 14)—via their
effect on inelastic collisions that lead to trap
loss. Evidence of tetramer states associated with
the trimers has also been found (13, 15). Al-
though the locations of successive features are
consistent with the predicted universal scaling,
systematic deviations as large as 60% were
observed and attributed to nonuniversal short-
range physics (13). In the work presented here, we
use a Feshbach resonance in7Li for which a/r0
can be tuned over a range spanning three
decades (16). This enables the observation of
multiple features that are compared to universal
We confine7Li in the |F = 1, mF= 1〉 (where F
is the total spin quantum number and mFis its
projection) hyperfine state in an elongated, cylin-
drically symmetric, hybrid magnetic—plus—
optical dipole trap, as described previously (16).
A set of Helmholtz coils provides an axially
oriented magnetic bias field that we used to tune
the two-body scattering length a via a Feshbach
resonance located near 737 G (17). For a > 0,
efficient evaporative cooling is achieved by
setting the bias field to 717 G, where a ~
200a0(a0is the Bohr radius), and reducing the
optical-trap intensity. Depending on the final trap
depth, we create either an ultracold thermal gas
just above the condensation temperature TCor a
Bose-Einstein condensate (BEC) with >90%
condensate fraction. For investigations with a <
and proceed with optical-trap evaporation, which
is stopped at a temperature Tslightly above TC.
In both cases the field is then adiabatically
ramped to a final value and held for a variable
hold time. The fraction of atoms remaining at
each time is measured via in situ polarization
phase-contrast imaging (18) for clouds where
case of lower densities.
Analyzing the time evolution of the number
of atoms in the trap determines the three-body
loss coefficient L3(8, 13, 19), as well as the
four-body loss coefficient L4(15). Recombination
into a dimer is a three-body process because a
third atom is needed to conserve both momentum
and energy. For a > 0, the dimer can be weakly
bound with binding energy e = ħ2/(ma2) (where
m is the atomic mass and ħ is Planck’s constant
h divided by 2p), whereas for a < 0 there are
only deeply bound molecular dimers. The
recombination energy released in the collision
is sufficient to eject all three atoms from the trap
fora<0,aswellasfora>0whene ≳ U (where
U is the trap depth). In the case of the BEC data,
this latter condition holds fora ≲ 5000a0. None-
any recombination event because, even for a
larger than 5000a0, we observe rapid three-body
Department of Physics and Astronomy and Rice Quantum
Institute, Rice University, Houston, TX 77005, USA.
*To whom correspondence should be addressed. E-mail:
VOL 32618 DECEMBER 2009
on January 5, 2011