© 1999 Macmillan Magazines Ltd

correspondence

556 NATURE | VOL 398 | 15 APRIL 1999 | www.nature.com

already used for crop production. Thereforeexpansion of agriculture to produce therequired organic nitrogen would necessitateuse of marginal land or furtherencroachment on natural ecosystems,resulting in soil degradation and loss ofbiological diversity.

Additionally, agronomic technology hasadvanced considerably beyond the‘conventional’ practices used by Drinkwateret al. In industrialized countries, themould-board plough has been largelyreplaced by no-till and conservation tillagepractices. Single applications of fertilizerare being replaced by multiple applicationswith amounts and timing matched to croprequirements. These practices result inimproved soil conservation and quality;greater efficiency in fertilizer use; higherand more stable yields; and they alsominimize the land area required for cropproduction.

Progress towards a ‘greener’ agriculturewill come from continued improvements inmodern high-yield crop-productionmethods combined with sophisticated useof both inorganic and organic nutrientsources, water, crop germplasm, pestmanagement and beneficial organisms.Thomas R. Sinclair*, Kenneth G. Cassman†*USDA-ARS, Agronomy Department, University ofFlorida, Gainesville, Florida 32611-0965, USA†Department of Agronomy, University of Nebraska,Lincoln, Nebraska 68583-0915, USA

1. Drinkwater, L. E. et al. Nature 396, 262–265 (1998).

2. Tilman, D. Nature 396, 211–212 (1998).

Green revolutionstill too green

Sir — L. E. Drinkwater et al.1 reportpromising isolated studies onenvironmental quality and productivity incrop production. But these studies shouldnot be used to discredit high-yield cropproduction technologies globally.

We do not dispute David Tilman’s call,in his News and Views article2

accompanying the paper by Drinkwater etal., for a greener agriculture based onecological principles. But we do not thinkthat as things stand we have the ability tocope with the probable doubling in globalfood demand in the twenty-first centurywith only organic nutrient inputs.

The organic nitrogen productionschemes investigated by Drinkwater et al.used essentially twice the land area of their‘conventional’ system to produce equivalentamounts of grain. But the best arable land is

Fertile grounds fora lively debate

Sir — I am not surprised that HermannBondi1 was puzzled by my correspondencewith Roger Short2,3, which concerned therelative importance of fertility andagriculture in the population explosion.

I myself was puzzled by Short’s response,because I was not arguing that fertility isinfluenced by nutrition; in fact I had beennaively unaware that this is the subject of alively controversy.

I simply meant that an increase infertility (from whatever cause) could not byitself have resulted in population growthbecause the additional babies would havestarved. Without the agricultural advancesof this century, the world population wouldnot now be approaching six billion, nomatter what the fertility rates were.Stephen G. WarrenDepartment of Atmospheric Sciences, University ofWashington, Seattle, Washington 98195-1640, USA

1. Bondi, H. Nature 397, 644 (1999).

2. Warren, S. G. Nature 397, 101 (1999).

3. Short, R. Nature 397, 101 (1999).

with limited long-term prospects. Thissituation is only alleviated by the benefactionof senior scientists and charitablefoundations, and occasional directives inselected areas which allow young scientists todevelop independent research.

Further obstacles exist in therecruitment process: new positions areoften focused on narrow research areas andonly advertised locally (in Danish). Recentwell-intentioned legislative changes havenot fully addressed these problems.

Such an inflexible system (which oftenobliges scientists to spend their entire careerin the same institute) is ill-equipped toadapt to the rapid development of newareas in basic research. The only surprise isthat Danish science has remained socompetitive for so long. How long this willcontinue to be the case is unclear whenthere is little to attract young scientists.Without a competitive basic researchcomponent, the ability to foster novelapplied research (so beloved of the presentgovernment) will be severely eroded.

Similar criticisms have been levelled atSweden. In its case at least some of thesecriticisms are now being taken to heart, ascan be seen by the establishment of openlycompetitive, well funded, junior facultypositions at the new Centre for MolecularMedicine in Umea. We hope that similarinitiatives will be taken in Denmark. Michael Ibba, Thomas BentinCentre for Biomolecular Recognition,Institute of Medical Biochemistry and Genetics,Panum Institute, Copenhagen University,2200 Copenhagen N, Denmark

1. May, R. M. Science 281, 49–51 (1998).

2. Horton, B. Nature 395, 412–413 (1998).

Denmark lacks coherentpolicy on basic research

Sir — Recent months have seen a lively,and at times bitter, debate in the Danishmedia on the future direction of scientificresearch. Much has been made by thegovernment of the need to shift towardsmore applied research to maximize theshort-term benefits of public investment.However, we would suggest that morecritical problems exist that must beaddressed immediately to ensure the long-term health of Danish science. Chiefamong these are a poorly funded andmisdirected policy on basic researchfunding, and conditions of employmentthat restrict the research opportunities ofyoung scientists.

Danish science is moderately wellfunded1. We have modern facilities, anexcellent level of technical support and abuoyant biotechnology sector2. What issorely lacking is a coherent policy on thefunding and nurturing of basic research.Entry-level appointments (assistantprofessor) have a heavy teaching load and nosupport for scientific staff. Young scientistscannot improve their situation by writinggrant applications, since the fundingavailable to the research councils allows little,if any, support for salary components. Suchrestrictions are making assistantprofessorships increasingly unattractive,

equator4 contain seasonal freeze-thawwedge structures in permafrost regolith thatindicate marked seasonal changes oftemperature3,4.

The ‘snowball Earth’ hypothesis for low-latitude glaciation5 is difficult to simulate6,7

and conflicts with geological evidence. Thelate Neoproterozoic stratigraphic record8

shows no sign of the drastic lowering of sealevel that would accompany globalglaciation. Moreover, seasonal freeze-thawstructures could not form near the Equatoron a snowball Earth because the very lowglobal temperatures would inhibit seasonalvariation9.George E. WilliamsDepartment of Geology and Geophysics,University of Adelaide, SA 5005, Australia

1. Hoffman, P. F. & Maloof, A. C. Nature 397, 384 (1999).

2. Williams, D. M. et al. Nature 396, 453–455 (1998).

3. Williams, G. E. Earth Sci. Rev. 34, 1–45 (1993).

4. Schmidt, P. W. & Williams, G. E. Earth Planet. Sci. Lett. 134,

107–124 (1995).

5. Hoffman, P. F. et al. Science 281, 1342–1346 (1998).

6. Crowley, T. J. & Baum, S. K. J. Geophys. Res. 98, 16723–16732

(1993).

7. Walsh, K. J. & Sellers, W. D. Global Planet. Change 8, 219–230

(1993).

8. Preiss, W. V. Geol. Surv. S. Aust. Bull. 53, 438 (1987).

9. Sellers, W. D. Palaeogeogr., Palaeoclimatol., Palaeoecol. 82,

217–224 (1990).