© 2005 Nature Publishing Group

NATURE|Vol 438|3 November 2005 NEWS & VIEWS

35

50 YEARS AGO“Use and abuse of English inscience” — Another problemwhich is causing increasingconcern — to printers as well asto editors — is the frequent andindiscriminate use ofabbreviations in the form of asingle capital letter, or a group ofcapitals, to represent the name of a substance, or perhaps evenan adjective or adverb. Theprinter is concerned because a page of text sprinkled withcapital letters is not pleasing in appearance; and like othercraftsmen, he feels that hisefforts are being frustrated…The use of abbreviations,especially initial letters, is nowbecoming so fashionable amongscientists that one suspectsauthors sometimes go out oftheir way to use them. … thisfashion may, if not checked,defeat its own ends and producea veritable ‘Tower of Babel’.Indeed the time does not seemfar away when high-schoolpupils will have to learn a newtable of symbols apart fromthose atomic.From Nature 5 November 1955.

100 YEARS AGOA return has been published, we learn from the Pioneer Mail,regarding the measures adoptedfor the extermination of wildanimals and venomous snakesduring the year 1904. The totalmortality among human beingsreported to have been caused by wild animals was 2157, against2749 in 1903. The most notabledecrease occurred in Madras and the United Provinces,namely, from 438 and 404 in1903 to 237 and 193 in 1904respectively… The mortality fromsnake-bite rose from 21,827 to21,880. It is reported that in theSeoul district of the CentralProvinces anti-venin was usedwith success in two cases, andthe question of introducing more generally the treatment of snake-bite by potassiumpermanganate is under the consideration of the local Government. The totalnumber of snakes killed was 65,378.From Nature 2 November 1905. 50

& 10

0 Y

EARS

AG

O

of several millimetres. These oscillations havereceived scant attention to date. Some of themresult from changes in ocean heat contentassociated with internal perturbations of the ocean–atmosphere system, such as the El Niño–Southern Oscillation and PacificDecadal Oscillation4. But other processes,probably related to the ‘forcing’ effects of nat-ural climate variation, also play a role.

On page 74 of this issue, Church et al.5 use cli-mate simulations to reveal the effects of vol-canic eruptions on sea level between 1890 and

with either one gene type or the other. Thescattered losses produced the evolutionarymosaic at the level of species or even strain.

Croft et al.3 also provide evidence that a bac-terium, Halomonas, upregulates the biosyn-thesis of vitamin B12 when in the presence ofalgal exudates. Many algae are members of aloose taxonomic grouping known as the pro-tists. These are often unicellular and includesuch organisms as Amoeba and Paramecium.The phenomenon of endosymbiosis, in whichone organism (such as a bacterium) takes upresidence inside another, to mutual benefit,has been thoroughly studied in protists, especially with regard to the origins of thechloroplast and mitochondrion. But possibleectosymbioses — literally, more superficialrelationships — involving bacteria and algaehave received less attention.

An earlier investigation did indeed show thatThalassiosira and other marine diatoms, all ofwhich require vitamin B12, could be grownwithout the vitamin when bacterial cultureswere added to the diatom cultures10. Such stud-ies hinted at the existence of a symbiotic rela-tionship. But unlike Croft et al., the authors ofthis study did not identify the bacteria involvedor the specific genes (or enzymes) concerned,and they did not demonstrate upregulation of a bacterial gene in response to a chemical signal from the algae.

Croft and colleagues’ approach3 could prof-itably be adopted more broadly, because protistshave a much wider variety of basic biochemicalpathways than do either animals or plants. Wecan hope that the enzymatic pathways leadingto other amino acids, sugars, lipids and so forth— which have long been known to be diverse inprotists and to show similar evolutionarymosaic patterns11 — will likewise be examinedusing the genome data now available. It is likely

that additional symbiotic vitamin-B12-produc-ing bacteria will be identified, and that othervitamins are produced by symbiotic bacteria.

But non-symbiotic bacterial sources of vita-mins may be equally or more important. Forexample, concentrations of vitamin B12 in theoceans vary with season, and there is strongcircumstantial evidence that this vitamin isproduced on the ocean floor at depths wheredarkness makes it unlikely that an algal–bacterial symbiosis can exist12.

Clearly, the paper by Croft et al. doesn’tanswer all questions. But it greatly advancesour understanding of why the vitamin-B12requirements are so sporadic among the algae,and also points to an enticing variety ofresearch opportunities. ■

Robert A. Andersen is at the Provasoli–GuillardNational Center for Culture of MarinePhytoplankton, Bigelow Laboratory for OceanSciences, West Boothbay Harbor, Maine 04575, USA. e-mail: randersen@bigelow.org

1. Provasoli, L. & Carlucci, A. F. in Algal Physiology andBiochemistry (ed. Stewart, W. D. P.) 741–787 (Blackwell,Oxford, 1974).

2. Lewin, J. C. & Lewin, R. A. Can. J. Microbiol. 6, 127–134 (1960).3. Croft, M. T., Lawrence, A. D., Raux-Deery, E., Warren, M. J.

& Smith, A. G. Nature 438, 90–93 (2005).4. Famintzin, A. Bull. Acad. Sci. St Petersb. 17, 31–70 (1871).5. Pringsheim, E. G. Beitr. Biol. Pfl. 11, 305–334 (1912). 6. Provasoli, L., Hutner, S. H. & Schatz, A. Proc. Soc. Exp. Biol.

Med. 69, 279–282 (1948).7. Hutner, S. H. et al. Soc. Exp. Biol. Med. Proc. 70, 118–120

(1949).8. Swift, D. in The Physiological Ecology of Phytoplankton (ed.

Morris, I.) 329–368 (Univ. California Press, Berkeley,1980).

9. Guillard, R. R. L. & Ryther, J. H. Can. J. Microbiol. 8, 229–239(1962).

10. Haines, K. C. & Guillard, R. R. L. J. Phycol. 10, 245–252 (1974).11. Ragan, M. A. & Chapman, D. J. Biochemical Phylogeny of the

Protists (Academic, New York, 1978).12. Menzel, D. & Spaeth, J. P. Limnol. Oceanogr. 7, 151–154

(1962).

GLOBAL CHANGE

Sea level and volcanoes Anny Cazenave

Large volcanic eruptions cool the world ocean. In doing so, they temporarilyreduce the increase in ocean heat content and the rise in sea level attributedto warming caused by greenhouse-gas emissions.

Global warming is producing a rise in sealevel. Observations from tide gauges and satel-lite altimetry indicate that sea level has beenrising by 1.8 millimetres per year since 1950(ref. 1) and about 3 millimetres per year dur-ing the 1990s (ref. 2). The two causes are athermal expansion of sea water in response toocean warming, and the input of extra waterfrom the melting of glaciers and ice sheets onland3. But against the background of this over-all increase, global mean sea level displaysinterannual to decadal oscillations of the order

3.11 News & Views 031 CDS 28/10/05 4:50 PM Page 35

Nature Publishing Group© 2005

© 2005 Nature Publishing Group

Enzymes are proteins that catalyse chemicalprocesses by lowering the energy barriersbetween substrate and product, thus increas-ing the rate of reaction. They function by stabilizing transition states, the structuralintermediates formed in the rate-limiting stepsof chemical reactions. In some cases, the cataly-sis process involves the enzyme interconvert-

ing between alternative versions of its struc-ture, which can be followed using nuclearmagnetic resonance (NMR) spectroscopy1–3.On page 117 of this issue, Eisenmesser et al.4

demonstrate that the slow structural motionsbetween sub-states of an enzyme that takeplace during catalysis also occur in the freeenzyme in the absence of its substrate. This

NEWS & VIEWS NATURE|Vol 438|3 November 2005

36

2000. The simulations take account of anthro-pogenic forcing that stemmed from greenhousegases, aerosols and ozone, and the natural cli-mate forcing resulting from changes in volcanicactivity and the input of solar radiation. Theseforcings affect the oceans through warming (orcooling) the upper ocean, leading to an increase(decrease) of ocean heat content and hence anincrease (decrease) in sea level through oceanthermal expansion (contraction). Church et al.find that, in the first few months after large volcanic eruptions, there is a fall of several millimetres in global mean sea level. This is followed by a slow increase, lasting for a decadeor more, towards the pre-eruption state.

The link here is that large volcanic eruptionsinject particles and gases into the atmosphere,in particular sulphur gases that are convertedinto sulphate aerosols in the stratosphere, thelayer of Earth’s atmosphere immediately abovethe lowest layer, the troposphere. Their domi-nant effect is to increase the fraction of inci-dent radiation reflected by the planet, andhence reduce the amount of solar energyreaching Earth’s surface. The correspondingnet surface-air cooling and its consequenceson weather have been much studied6, unlikethe impact on ocean heat content and sea level.

Church et al.5 show that, because of thereduction of the net solar flux at the ocean sur-face, volcanic eruptions induce an immediatecooling of the surface layers, and so a decreasein heat content and sea level. The predictedabrupt fall in ocean heat content is well corre-lated with observations7. Although surface-airtemperature recovers within a few years, thecooling effects on the ocean persist for at leasta decade. This is because of the large heatcapacity of the oceans compared with that ofthe atmosphere and the slow redistribution ofheat by the ocean circulation5,8.

Observational analyses of historical oceantemperatures show a net warming of theoceans since 1950, contributing about 85% ofthe total increase in heat content of the wholeEarth system7 and consistent with the current

imbalance between the energy absorbed andemitted by the planet9. Model studies suggestthat most of this ocean warming results fromhuman activities and the associated increase inlevels of greenhouse gases10. However, duringthe past several decades, large volcanic erup-tions — Mt Agung, Indonesia (1963), El Chi-chon, Mexico (1982) and Mt Pinatubo,Philippines (1991) — have temporarilyreduced the anthropogenic ocean warming.Because the effects of subsurface ocean cool-ing can persist for one to several decades, theyhave masked at least part of the acceleratingincrease in sea level.

The simulations5 indicate that the eruptionof Mt Pinatubo in 1991 (Fig. 1) produced adrop of some 6 mm in sea level within about ayear, which was followed by a slow rise ofabout 0.5 mm yr�1 over the next decade ormore. Thus, about 0.5 mm yr�1 of the steepersea-level rise resulting from thermal expan-sion, estimated from ocean temperature dataover the past decade (about 1.5 mm yr�1, com-pared with the 0.4 mm yr�1 mean rate of the

Figure 1 | Mount Pinatubo erupts. The cooling effect of the sulphate aerosols produced by this andother eruptions helps to account for oscillations in the sea-level record5.

STRUCTURAL BIOLOGY

Proteins flex to functionYuanpeng J. Huang and Gaetano T. Montelione

Static pictures of protein structures are so prevalent that it is easy to forgetthey are dynamic molecular machines. Characterizing their intrinsicmotions may be necessary to understand how they work.

past 50 years)11, may reflect a recovery fromthe Mt Pinatubo eruption. This in turn mayexplain part of the higher rate of sea-level riseobserved by satellite altimetry since early 1993(3 mm yr�1, compared with 1.8 mm yr�1

recorded by historical tide gauges since 1950;ref. 1). Although it remains unclear whethersea-level rise over the past decade indicates anaccelerating trend, the study by Church et al.5

suggests that part of it (although not all) can beexplained by natural variability on interannualto decadal timescales.

Climate-model predictions indicate that sealevel will continue to rise in the coming decades,and even centuries, owing to thermal expansionof the ocean in response to anthropogenicwarming3. But Church et al.5 clearly demon-strate that large volcanic eruptions can (partiallyand temporarily) mask that effect. Their study isone step towards a better understanding of thesea-level record — which is essential if we are toimprove projections of sea-level rise, and pre-pare for the impact of that rise on vulnerablecoastal regions and island nations. ■

Anny Cazenave is at the Laboratoire d’Etudes enGéophysique et Océanographie Spatiales,LEGOS-CNES, 18 avenue Edouard Belin, 31401Toulouse, Cedex 9, France.e-mail: anny.cazenave@cnes.fr

1. Church, J. A., White, N. J., Coleman, R., Lambeck, K. &Mitrovica, J. X. J. Clim. 17, 2609–2625 (2004).

2. Leuliette, E. W., Nerem, R. S. & Mitchum, G. T. Mar. Geodesy27, 79–94 (2004).

3. Church, J. A. et al. in Climate Change 2001: The ScientificBasis (eds Houghton, J. T. et al.) 639–694 (Cambridge Univ.Press, 2001).

4. Lombard, A., Cazenave, A., Le Traon, P. Y. & Ishii, M. Glob.Planet. Change 47, 1–16 (2005).

5. Church, J. A., White, N. J. & Arblaster, J. M. Nature 438,74–77 (2005).

6. Robock, A. Rev. Geophys. 38, 191–219 (2000).7. Levitus, S., Antonov, J. I. & Boyer, T. P. Geophys. Res. Lett. 32,

L02604; doi:10.1029/2004GL021592 (2005).8. Delworth, T. L., Ramaswamy, V. & Stenchikov, G. L.

Geophys. Res. Lett. (in the press). 9. Hansen, J. et al. Science 308, 1431–1435 (2005).10. Barnett, T. P. et al. Science 309, 248–287 (2005).11. Antonov, J. I., Levitus, S. & Boyer, T. P. Geophys. Res. Lett. 32,

L12602; doi:10.1029/2005GL023112 (2005).

SIPA

PRE

SS/R

EX F

EAT

URE

S

3.11 News & Views 031 CDS 28/10/05 4:50 PM Page 36

Nature Publishing Group© 2005