letters to nature

2. Adkins, J. F., McIntyre, K. & Schrag, D. P. The salinity, temperature, and d18O of the glacial deep ocean. ..............................................................
Science 298, 1769–1773 (2002).
3. Mitchell, B. G., Brody, E. A., Holm-Hansen, O., McClain, C. & Bishop, J. Light limitation of
phytoplankton biomass and macronutrient utilization in the Southern Ocean. Limnol. Oceanogr. 36,
Hybrid fracture and the
1662–1677 (1991).
4. Martin, J. H. & Gordon, R. M. Northeast Pacific iron distributions in relation to phytoplankton
transition from extension
productivity. Deep-Sea Res. 35, 177–196 (1988).
5. Conkright, M. E. et al. World Ocean Atlas 2001: Objective Analyses, Data Statistics, and Figures, fracture to shear fracture
CD-ROM Documentation (National Oceanographic Data Center, Silver Spring, 2002).
6. Warren, B. Why is no deep water formed in the North Pacific? J. Mar. Res. 41, 327–347 (1983).
Jonathan M. Ramsey* & Frederick M. Chester
7. Haug, G. H., Sigman, D. M., Tiedemann, R., Pedersen, T. F. & Sarnthein, M. Onset of permanent
stratification in the subarctic Pacific Ocean. Nature 40, 779–782 (1999).
Center for Tectonophysics, Department of Geology and Geophysics, Texas A&M
8. Hillenbrand, C.-D. & Fütterer, D. K. Neogene to Quaternary deposition of opal on the continental rise
west of the Antarctic Peninsula, ODP Leg 178, Sites 1095, 1096 and 1101. Proc. ODP Sci. Res. 178, 1–33
University, College Station, Texas 77843-3115, USA
(2001). * Present address: Anadarko Petroleum Corporation, 1201 Lake Robbins Drive, The Woodlands,
9. Barker, P. F. & Kennett, J. P. Weddell Sea paleoceanography: Preliminary results of ODP Leg 113. Texas 77380, USA
Palaeogeogr. Palaeoclimatol. Palaeoecol. 67, 75–102 (1988). .............................................................................................................................................................................
10. Altabet, M. A. & Francois, R. Sedimentary nitrogen isotopic ratio as a recorder for surface ocean Fracture is a fundamental mechanism of material failure. Two
nitrate utilization. Glob. Biogeochem. Cycles 8, 103–116 (1994).
11. Andreev, A., Kusakabe, M., Honda, M., Murata, A. & Saito, C. Vertical fluxes of nutrients and carbon
basic types of brittle fractures are commonly observed in rock
through the halocline in the western subarctic gyre calculated by mass balance. Deep-sea Res. II 49, deformation experiments—extension (opening mode) fractures
5577–5593 (2002). and shear fractures1,2. For nearly half a century it has been
12. Narita, H. et al. Biogenic opal indicating less productive northwestern North Pacific during the glacial hypothesized that extension and shear fractures represent end-
ages. Geophys. Res. Lett. 29, doi:10.1029/2001GL014320 (2002).
members of a continuous spectrum of brittle fracture types3–6.
13. Sigman, D. M., Altabet, M. A., Francois, R., McCorkle, D. C. & Fischer, G. The d15N of nitrate in the
Southern Ocean: Consumption of nitrate in surface waters. Glob. Biogeochem. Cycles 13, 1149–1166 However, observations of transitional fractures that display both
(1999). opening and shear modes (hybrids) in naturally deformed rock
14. Sigmon, D. E., Nelson, D. M. & Brzezinski, M. A. The Si cycle in the Pacific sector of the Southern have often remained ambiguous, and a clear demonstration of
Ocean: seasonal diatom production in the surface layer and export to the deep sea. Deep-sea Res. II 49, hybrid fracture formation has not been provided by experi-
1747–1763 (2002).
15. Brzezinski, M. A. et al. A switch from Si(OH)4 to NO2
ments4. Here we present the results of triaxial extension experi-
3 depletion in the glacial Southern Ocean.
Geophys. Res. Lett. 29, doi:10.1029/2001GL014349 (2002). ments on Carrara marble that show a continuous transition from
16. Lefevre, N. & Watson, A. J. Modeling the geochemical cycle of iron in the oceans and its impact on extension fracture to shear fracture with an increase in compres-
atmospheric CO2 concentrations. Glob. Biogeochem. Cycles 13, 727–736 (1999). sive stress. Hybrid fractures form under mixed tensile and
17. Francois, R. F. et al. Water column stratification in the Southern Ocean contributed to the lowering of
glacial atmospheric CO2. Nature 389, 929–935 (1997).
compressive stress states at acute angles to the maximum prin-
18. Gildor, H. & Tziperman, E. Physical mechanisms behind biogeochemical glacial-interglacial CO2 cipal compressive stress. Fracture angles are greater than those
variations. Geophys. Res. Lett. 28, 2421–2424 (2001). observed for extension fractures and less than those observed for
19. Boyle, E. A. & Keigwin, L. D. Deep circulation of the North Atlantic for the last 200,000 years: shear fractures. Fracture surfaces also display a progressive
geochemical evidence. Science 218, 784–787 (1982).
change from an extension to shear fracture morphology.
20. Manabe, S. & Stouffer, R. J. Simulation of abrupt climate change induced by freshwater input to the
North Atlantic Ocean. Nature 378, 165–167 (1995).
In the laboratory, a shear fracture is produced under a compres-
21. Wang, Z. M., Mysak, L. A. & McManus, J. F. Response of the thermohaline circulation to cold climates. sive stress state, shows displacement parallel to the failure surface,
Paleoceanography 17, doi:10.1029/2000PA000587 (2002). and forms at a 208–408 angle to the maximum principal compressive
22. Sarmiento, J. L., Hughes, T. M. C., Stouffer, R. J. & Manabe, S. Response of the ocean carbon cycle to stress direction, j 1 (Fig. 1). An extension fracture forms under a
anthropogenic climate warming. Nature 393, 245–249 (1998).
23. Toggweiler, J. R. Russell, J. Shifted westerlies caused low CO2 during cold glacial periods. Eos 84
tensile stress condition, displays an opening-mode displacement
(Ocean Sci. Meet. Suppl.), OS21N-06 (2003). normal to the fracture surface, and propagates in an orientation
24. Sarmiento, J. L. & Toggweiler, J. R. A new model for the role of the oceans in determining atmospheric parallel to the maximum and perpendicular to the minimum
pCO2. Nature 308, 621–624 (1984). principal compressive stresses. Additional fracture types described
25. Froelich, P. N. et al. Biogenic opal and carbonate accumulation rates in the Subantarctic South
in laboratory samples, such as wedging and longitudinal splitting
Atlantic: The Late Neogene of Meteor Rise Site 704. Proc. ODP Sci. Res. 114, 515–532 (1991).
26. Orsi, A. H., Whitworth, T. III & Nowlin, W. D. Jr On the meridional extent and fronts of the Antarctic
during uniaxial compression, are often assumed to be forms of
Circumpolar Current. Deep-Sea Res. I 42, 641–673 (1995). extension and shear fractures1,2,7,8.
27. Rea, D. K., Basov, I. A., Janecek, T. R. & Palmer-Julson, A. Introduction to Leg 145: North Pacific Within the compressive stress regime, the empirical Coulomb–
Transect. Proc. ODP Init. Rep. 145, 5–7 (1993). Mohr failure criterion successfully predicts the brittle failure
28. Burckle, L. H., Gersonde, R. & Abrams, N. Late Pliocene-Pleistocene paleoclimate in the Jane Basin
region: ODP Site 697. Proc. ODP Sci. Res. 113, 803–809 (1987).
strength and orientation of shear fractures, at least to first
29. Brewster, N. A. Cenozoic biogenic silica sedimentation in the Antarctic Ocean. Geol. Soc. Am. Bull. 91, order9,10. As implied by Mohr’s hypothesis, the orientation of
337–347 (1980). the fracture surface is inclined to j 1 at a fracture angle, v, equal to
30. Diekmann, B., Fälker, M. & Kuhn, G. Environmental history of the south-eastern South Atlantic since p/4 2 f/2 or 2p/4 þ f/2, where tanf is the slope of the failure
the middle Miocene: Evidence from the sedimentological records of ODP Sites 1088 and 1092.
envelope in Mohr space 9. The failure envelope is commonly
Sedimentology 50, 511–529 (2003).
depicted as a continuous function that crosses the transition from
Supplementary Information accompanies the paper on www.nature.com/nature.
a compressive to a tensile stress state, and thus includes uniaxial
tensile failure11,12 (Fig. 1).
Acknowledgements This work was supported by the US NSF, Schweizer Nationalfonds, Deutsche A theoretical basis for continuity across stress states is provided
Forschungsgemeinschaft, and by British Petroleum and Ford Motor Company through the by the Griffith theory of fracture; the theory that forms the basis of
Carbon Mitigation Initiative at Princeton University. This research used samples provided by the modern linear elastic fracture mechanics1,2. Griffith theory captures
ODP. The ODP is sponsored by NSF and participating countries under the management of Joint
Oceanographic Institutions. We thank T. F. Pedersen and K. Gordon for isotope analyses, M. Soon
the essential physical processes of crack propagation, and provides
for analytical assistance, and C.-D. Hillenbrand for providing data from ODP Site 1096. J. D. Hays an energy-based failure criterion for uniaxial tensile failure. From
focused our attention on M. Winton’s work on the climate implications of the nonlinear this theory Griffith derived a criterion for brittle failure under
dependence of density on temperature. D. P. Schrag encouraged its pursuit as a general biaxial stress states, assuming a local tensile stress condition for
mechanism, and J. F. Adkins contributed the thoughts about the role of sea ice formation on the
Antarctic shelf.
crack propagation13. Further modifications treat failure under
compression where crack closure and friction are important, and
are generalized to treat failure under triaxial stress14–16. The
Competing interests statement The authors declare that they have no competing financial
interests. Griffith and modified-Griffith criteria predict a parabolic failure
envelope for mixed tensile and compressive stress conditions that
Correspondence and requests for materials should be addressed to D.S. ([email protected]). includes uniaxial tensile failure. The Griffith criteria have also been
NATURE | VOL 428 | 4 MARCH 2004 | www.nature.com/nature
©2004 Nature Publishing Group 63
letters to nature
combined with the Coulomb–Mohr criteria for compressive stress dog-bone samples are the generation of bending forces and jacket
states to form a composite failure envelope17,18. Such a continuous effects. Our experiments were conducted in a gear-driven, liquid
failure criterion is consistent with the hypothesis that a continuous confining-media, triaxial apparatus that is very square and stiff
transition from extension fracture to shear fracture exists and that (9.47 £ 108 N m21), and capable of deforming samples 45 mm in
hybrid fractures form under mixed tensile and compressive stress diameter24. Bending forces are minimized using this apparatus with
states (Fig. 1). Nevertheless, laboratory evidence documenting precisely machined samples. Undesirable jacket effects were
brittle failure in the form of a hybrid fracture produced under a obviated by using layered jackets composed of an inner layer of
mixed stress state is ambiguous8,19. latex or copper and an outer jacket of polyolefin separated by
The most comprehensive experimental study of brittle failure plasticene modelling clay22. The outer polyolefin is impermeable
under mixed stress states is that of ref. 19. Twenty triaxial extension to the confining media and the clay conveys confining stress to the
tests were conducted on five different rock types (granite, quartzite, necked portion of the sample. The clay is necessary because the
diabase and two dolomites) using a dog-bone sample geometry. polyolefin supports relatively large elastic stresses and induces
Dog-bone samples are cylinders of rock having a reduced central undesirable loading of the sample if it is in direct contact with the
neck within which an axial tensile stress can be generated via axial rock. An inner layer of latex restricts the clay from intruding into
extension under confining pressure, P c (refs 19, 20) (Fig. 2). rock pores before fracture, and allows very accurate measurement of
Although fifteen of the samples experienced an axial tensile stress failure strength because the latex is very weak. Detailed observation
at failure, the inconsistency in fracture orientation and failure of the fracture surfaces is precluded by use of latex because the latex
strength precluded conclusions regarding the existence of a con- ruptures after failure and the clay intrudes into the sample.
tinuous transition from extension to shear fractures4,19. The experi- Accordingly, a parallel suite of experiments were conducted using
mental data did, nonetheless, indicate that fracture under a mixed jackets with an inner layer of copper. The copper was sufficiently
stress state may occur at a tensile stress that is less tensile than the strong to ensure preservation of the fracture surfaces.
uniaxial tensile strength. Thirty-one extension experiments were completed at room
To investigate hybrid fracture further we followed the general temperature (22 8C) and an axial extension rate of 2 £ 1022
methodology of previous extension tests on dog-bone samples, but
made key modifications to the sample geometry and jacket design to
improve reproducibility. Samples were taken from the Lorano
Bianco Carrara marble of Italy. The marble is a relatively homo-
geneous and isotropic, nearly pure calcite rock having a grain size of
approximately 250 mm, extremely low porosity, and little crystal-
lattice preferred orientation21,22. Unlike the fillet-cut dog-bone
geometry used previously19,20, we used a large-radius notch-cut
geometry produced by grinding along a single continuous circular
arc (Fig. 2a). Photoelastic modelling and elasticity calculations
indicate that the large-radius notch cut produces the most uniform
stress condition, at least for uniaxial tensile loading23.
Two experimental difficulties that may arise during extension of

Figure 1 A representation using the Mohr diagram of the hypothesis that the brittle failure Figure 2 Photographs of the undeformed and deformed notch-cut dog-bone samples of
envelope and the transition from extension fracture to shear fracture is continuous. Below Carrara marble. a, Samples are cylindrical and the notch cut is made with a 91.5-mm
the Mohr diagram (plot of shear stress, t, versus normal stress, j) are schematics of the radius grinding wheel. The final sample is 90 mm long, with 43 mm diameter at the ends,
notch-cut dog-bone sample geometry displaying the three stress conditions and the and 28 mm diameter in the central portion of the notch. Arrows indicate the confining
associated types of fracture (extension, hybrid and shear). Three stress states satisfying pressure, Pc, applied along boundaries of the cylindrical sample that generate differential
the failure criterion are shown in the Mohr diagram by the point of tangency of the stress stress as the sample ends are unloaded. In the central portion of the notch cut, the
circles with the failure envelope. In the Mohr diagram, the fracture surface orientation is minimum principal compressive stress, j3, is in the direction of the cylinder axis and may
specified by the slope of the failure envelope at the point of tangency (see text). The be tensile (negative). The maximum and intermediate principal stresses are the same
predicted fracture angle, v (measured between the fracture surface and the direction of magnitude (equal to Pc ) and are oriented perpendicular to the cylinder axis.
the maximum principal compressive stress, j1), increases with increase in compressive b, Representative samples showing the progressive change in fracture orientation with
normal stress. Hybrid fractures are thought to occur under mixed tensile and compressive increase in Pc. Fractures form in the samples at the neck (central portion of the notch); the
stress conditions and form at small but non-zero v. other (matching) halves of the samples are not included in the photograph.
64 ©2004 Nature Publishing Group NATURE | VOL 428 | 4 MARCH 2004 | www.nature.com/nature
 letters to nature
mm s21. The results of 18 experiments with copper jackets and 13 display the highest total area of cleavage planes relative to area of
experiments with latex jackets were combined to define failure comminuted material. Conversely, fracture surfaces formed at the
strength, fracture orientation, and fracture surface morphology as a highest pressures display the lowest total area of reflective cleavage
function of confining pressure (Table 1). Fracture surfaces are planes relative to area of comminuted material. The thickness of the
approximately planar at the sample scale, but are complex at the discoloured zone adjacent to the fracture surface also increases with
grain scale. Fracture orientation is specified by the maximum angle increasing confining pressure.
between the fracture surface and the plane normal to the cylinder Our observations of the progressive increase in fracture angle
axis (which contains the j 1 direction, Fig. 2a). Overall, there is an with confining pressure, from extension fracture at low pressure to
increase in fracture angle with confining pressure (Figs 2b and 3a). shear fracture at high pressure, provides the first convincing
The axial stress in the central region of the neck, which is the laboratory evidence for the existence of hybrid fractures that
minimum principal compressive stress, j 3, is determined from constitute a continuous transition from extension to shear fractures.
measurements of axial force and confining pressure19,20. A tensile The results reported herein support previous interpretations that
axial stress (negative j 3) was achieved at failure in all tests run at some natural fractures may be hybrids or conjugate shear fractures
confining pressures less than 130 MPa (Fig. 3b). with narrow dihedral angles. The surface morphology of the hybrid
On the basis of fracture orientation, fracture surface morphology, fracture is compatible with the step-tread geometry of the fractures
and stress state at failure, three major classes of fractures are
identified within the total population. Fractures formed at the
lowest confining pressures (7.5 and 60 MPa) are oriented approxi-
mately perpendicular to j 3 (parallel to j 1; Fig. 3a), form when j 3
equals 27.8 MPa (Fig. 3b; close to the uniaxial tensile strength of
6.9 MPa determined25 for Carrara marble), display a surface mor-
phology consistent with an opening-mode displacement, and may
be classified as extension fractures1,2. The fracture surfaces are
somewhat undulating at the sample scale, and display discrete,
highly reflective intragranular cleavages. In most cases the cleavage
surfaces in neighbouring grains are not coplanar, so the surfaces are
rough at the grain scale.
Fractures formed at confining pressures of 130 to 170 MPa are
inclined 138 to 228 from j 1 (Fig. 3a), form under a compressive
stress state (Fig. 3b), display surface features consistent with shear
displacement, and may be classified as shear fractures1,2. The
surfaces are covered with a powder of comminuted grains and
contain short grooves. The grooves are parallel to the maximum
inclination (that is, in the direction of the maximum shear traction)
and record the direction of slip. Cleaved crystals, as seen on
extension fracture surfaces, are not present. In cross-sectional
view the marble is discoloured adjacent to the fracture surface,
which probably reflects a microfractured zone associated with the
through-going macroscopic fracture.
Fractures formed at confining pressures between 70 to 120 MPa
are inclined 38 to 138 from j 1 (Fig. 3a), form under a mixed tensile
and compressive stress state (Fig. 3b), and display transitional
surface characteristics expected for hybrid fractures3,4. Specifically,
the surfaces exhibit patches of discrete, reflective, cleaved crystals
between areas of comminuted material with slip lineations. The
morphology of the fracture surfaces show a strong correlation to
confining pressure. Fracture surfaces formed at the lowest pressure

Figure 3 Fracture angle and fracture strength as a function of confining pressure.

Extension, hybrid and shear fracture are distinguished on the basis of fracture orientation
Table 1 Combined results of Carrara marble triaxial extension experiments
and surface morphology (see text). a, Fracture angle increases monotonically with
Pc ¼ j1 j1 2 j3 j3 Fracture angle Fracture
(MPa) (MPa) (MPa)* (degrees)† morphology
confining pressure for hybrid and shear fractures; fracture angle is constant for extension
............................................................................................................................................................................. fractures. The fracture angle is the maximum angle between the fracture surface and the
7.5 15.4 27.9 2.4 Extension plane normal to the cylinder axis; the plane normal to the cylinder axis contains j1 and j2
15 22.5 27.5 1.8, 2.1 Extension
30 37.8 27.8 1.1 Extension directions. Orientation of the fracture surface was determined by the best-fitting plane to
40 – – 1.4, 2.1 Extension 24 evenly distributed measurements of surface height (along the cylinder axis);
60 67.9 27.8 1.9, 2.1 Extension measurements are made using a drag profilometer with accuracy of 50 mm. b, Fracture
70 80.6 210.6 3.7 Hybrid
80 89.7, 89.8 29.7, 29.8 5.5 Hybrid strength increases approximately linearly with confining pressure for hybrid and shear
90 96.9 26.9 6.1 Hybrid fracture, and is approximately constant for extension fracture. Fracture strength is
100 104.0 24.0 7.3 Hybrid represented by the magnitude of the minimum principal stress in the neck of the sample at
110 – – 11.1 Hybrid
120 123.0 23.0 12.4 Hybrid the ultimate load-bearing capacity. There is an abrupt (possibly discontinuous) change in
130 130.0 0 13.4 Shear fracture strength at the transition from extension fracture to hybrid fracture at a Pc
140 137.5 2.5 15.6 Shear between 60 and 70 MPa. Fracture strength predicted by the Griffith criteria for biaxial
150 145.7 4.3 19.9 Shear
170 – – 21.6 Shear (j1 . j3, j2 ¼ 0) and triaxial-extension (Pc ¼ j1 ¼ j2 . j3) states-of-stress are
.............................................................................................................................................................................
* Determined from latex jacket experiments.
shown by solid lines. The Griffith criterion produces a parabolic failure envelope in Mohr
† Determined from copper jacket experiments. space.
NATURE | VOL 428 | 4 MARCH 2004 | www.nature.com/nature
©2004 Nature Publishing Group 65
letters to nature
classified as hybrids in the Mount Desert Island granite4,26. This ..............................................................
geometry is consistent with fracture formation through linkage of
an array of en echelon, interacting, extension cracks27. Coral mucus functions as
Although a continuous transition from extension to shear frac-
ture was hypothesized on the basis of Griffith theory and the
an energy carrier and particle
extrapolation of the Mohr envelope into the tensile field, our data
do not fully support such an approach. Contrary to Coulomb–
trap in the reef ecosystem
Mohr analysis, the fracture angles observed for hybrid and shear Christian Wild1, Markus Huettel2, Anke Klueter3, Stephan G. Kremb4,
fractures are systematically less than predicted by the slope of the Mohammed Y. M. Rasheed5 & Bo B. Jørgensen1
failure envelope. In addition, the failure strengths of the marble for
mixed tensile and compressive stresses are not consistent with a 1
Max Planck Institute for Marine Microbiology, Celsiusstrasse 1,
continuous, parabolic failure envelope such as that predicted by the D-28359 Bremen, Germany
2
Griffith criterion (Fig. 3b). Additional study is needed to determine Department of Oceanography, Florida State University, Tallahassee,
the specific processes that govern formation of hybrid fractures, but Florida 32306-4320, USA
3
given that hybrid fractures are transitional to extension and shear Australian Institute of Marine Science, PMB No. 3, Townsville MC,
fractures, it is likely that a complete understanding will require an Queensland 4810, Australia
4
Alfred Wegener Institute for Polar and Marine Research, Department Benthic
appropriate fracture mechanics theory that considers the mechan-
Ecosystems, Am Handelshafen 12, D-27570 Bremerhaven, Germany
ical interaction of stepped cracks4,27. A 5
Marine Science Station, University of Jordan and Yarmouk University, PO 195,
Received 1 December 2003; accepted 13 January 2004; doi:10.1038/nature02333. Aqaba, Jordan
.............................................................................................................................................................................
1. Jaeger, J. C. & Cook, N. G. W. Fundamentals of Rock Mechanics (Chapman and Hall, London, 1979).
2. Paterson, M. S. Experimental Rock Deformation—The Brittle Field (Springer, Berlin, 1978).
Zooxanthellae, endosymbiotic algae of reef-building corals, sub-
3. Hancock, P. L. in Geology in the Real World—the Kingsley Dunham Volume (eds Nichol, I. & Nesbitt, stantially contribute to the high gross primary production of
R. W.) 155–164 (Institution of Mining and Metallurgy, London, 1986). coral reefs1, but corals exude up to half of the carbon assimilated
4. Engelder, T. Transitional-tensile fracture propagation: a status report. J. Struct. Geol. 21, 1049–1055
by their zooxanthellae as mucus2,3. Here we show that released
(1999).
5. Muehlberger, W. R. Conjugate joint sets of small dihedral angle. J. Geol. 69, 211–219 (1961). coral mucus efficiently traps organic matter from the water
6. Sibson, R. H. Brittle-failure controls on maximum sustainable overpressure in different tectonic column and rapidly carries energy and nutrients to the reef
regimes. Am. Assoc. Petrol. Geol. Bull. 87, 901–908 (2003). lagoon sediment, which acts as a biocatalytic mineralizing filter.
7. Griggs, D. T. & Handin, J. W. Observations on fracture and a hypothesis of earthquakes. Geol. Soc. Am.
Mem. 79, 347–364 (1960).
In the Great Barrier Reef, the dominant genus of hard corals,
8. Hoskins, E. R. The failure of thick-walled hollow cylinders of isotropic rock. Int. J. Rock Mech. Min. Acropora, exudes up to 4.8 litres of mucus per square metre of
Sci. 6, 99–125 (1969). reef area per day. Between 56% and 80% of this mucus dissolves
9. Handin, J. On the Coulomb-Mohr failure criterion. J. Geophys. Res. 74, 5343–5348 (1969). in the reef water, which is filtered through the lagoon sands. Here,
10. Mogi, K. Effect of the intermediate principal stress on rock failure. J. Geophys. Res. 72, 5117–5131
(1967).
coral mucus is degraded at a turnover rate of at least 7% per hour.
11. Leon, A. Über die Rolle des Trennbruches im Rahmen der Mohrschen Anstrengungshypothese. Detached undissolved mucus traps suspended particles, increas-
Bauingenieur 15, 318–321 (1934). ing its initial organic carbon and nitrogen content by three
12. Nadai, A. Theory of Flow and Fracture of Solids (McGraw Hill, New York, 1934).
orders of magnitude within 2 h. Tidal currents concentrate
13. Griffith, A. A. Theory of rupture. Proc. Int. Congr. Appl. Mech. 1, 55–63 (1924).
14. McClintock, F. A. & Walsh, J. B. in Proc. 4th US Nat. Congr. Appl. Mech. Vol. II, 1015–1021 (Am. Soc.
these mucus aggregates into the lagoon, where they rapidly settle.
Mech. Eng., New York, 1962). Coral mucus provides light energy harvested by the zooxanthel-
15. Murrell, S. A. F. The effect of triaxial stress systems on the strength of rocks at atmospheric lae and trapped particles to the heterotrophic reef community,
temperatures. Geophys. J. R. Astron. Soc. 10, 231–281 (1965).
thereby establishing a recycling loop that supports benthic life,
16. Murrell, S. A. F & Digby, P. J. The theory of brittle fracture initiation under triaxial stress conditions. I
and II. Int. Geophys. J. R. Astron. Soc. 19, 309–334 and 499–512 (1970). while reducing loss of energy and nutrients from the reef
17. Brace, W. F. An extension of the Griffith theory of fracture to rocks. J. Geophys. Res. 65, 3477–3480 ecosystem.
(1960). Zooxanthellae provide a substantial part of the energy require-
18. Secor, D. T. Role of fluid pressure in jointing. Am. J. Sci. 263, 633–646 (1965).
19. Brace, W. F. in State of Stress in the Earth’s Crust (ed. Judd, W. R.) 111–180 (American Elsevier, New
ment of their hosts by transferring photosynthetically fixed carbon
York, 1964). to the coral4. High arabinose contents in the carbohydrate fraction
20. Schock, R. N. & Louis, H. Strain behavior of a granite and a greywacke sandstone in tension. J. Geophys. of coral mucus indicate that much of the fixed carbon is released as
Res. 87, 7817–7823 (1982). mucus, because arabinose is not usually a constituent of animal
21. Pieri, M., Burlini, L., Kunze, K., Stretton, I. & Olgaard, D. Rheological and microstructural evolution
of Carrara marble with high shear strain: results from high temperature torsion experiments. J. Struct.
cells5. Many hard and soft corals release mucus continuously, with a
Geol. 23, 1393–1413 (2001). species-specific composition5. Known functions of this release are
22. Ramsey, J. M. Experimental Study of the Transition from Brittle Shear Fractures to Joints (Texas A&M protection against fouling6, desiccation during air exposure at
Univ. Press, College Station, TX, 2003).
extreme low tide7 and sedimentation8.
23. Neuber, H. Theory of Notch Stresses (McGraw-Hill, New York, 1937).
24. Handin, J., Friedman, M., Logan, J. M., Pattison, L. J. & Swolfs, H. S. in Flow and Fracture of Rocks The adhesive mucus has the ability to trap particulate matter
(eds Heard, H. C., Borg, I. Y., Carter, N. L. & Raleigh, C. B.) 1–28 (AGU Geophys. Monogr. 16, from the water and the light-collecting coral surfaces. Mucus and
American Geophysical Union, 1972). trapped particles are transported over the coral surface by ciliary
25. Jaeger, J. C. & Hoskins, E. R. Stresses and failure in rings of rock loaded in diametral tension or
compression. Br. J. Appl. Phys. 17, 685–692 (1966).
currents and are released to the water9 at a magnitude such that
26. Engelder, T. The analysis of pinnate joints in the Mount Desert Island granite: implications for post- mucus can dominate suspended matter around reefs10,11. Our
intrusion kinematics in the coastal volcanic belt, Maine. Geology 17, 564–567 (1989). measured carbon to nitrogen ratios of 5–14 point towards the
27. Reches, Z. & Lockner, D. Nucleation and growth of faults in brittle rocks. J. Geophys. Res. 99, potential value of coral mucus as a food source; however, its role in
18159–18174 (1994).
the cycling of matter in coral reef environments is largely unknown.
Acknowledgements We thank M. Finn, A. K. Kronenberg and J. S. Chester for discussions and We investigated release rates, chemical composition, contents
suggestions, J. N. Magouirk and E. C. Powell for assistance in the laboratory, and J. S. Chester for and fate of mucus produced by Acropora, the coral genus with the
reviews of the manuscript. This work was partially supported by US NSF. highest area of coverage (15.8%) on the reef rim (crest plus slope) of
Heron Island (238 27 0 S, 1518 55 0 E), a ring-shaped platform reef in
Competing interests statement The authors declare that they have no competing financial the Great Barrier Reef, Australia. Submerged Acropora released 1.7 l
interests.
of mucus per m2 of reef area per day, excluding the rapidly
Correspondence and requests for materials should be addressed to F.M.C. dissolving mucus fraction. In situ measurements carried out after
([email protected]). air exposure, a regular event caused by extreme low tides12 and
66 ©2004 Nature Publishing Group NATURE | VOL 428 | 4 MARCH 2004 | www.nature.com/nature