© 1999 Macmillan Magazines Ltd

of the Universe since the Big Bang. Yet becausegravitating matter and a positive cosmologi-cal constant pull and push the expansion ofthe Universe in opposite ways, the data fromSNe Ia alone still allow for many combina-tions of Vm and VL.

Another approach derives a constraint ona different mix of the cosmological parame-ters by quantifying the degree to which localmass in different parts of the Universe tugs onnearby galaxies. Although most of the appar-ent motion of galaxies results from the expan-sion of the Universe, some excess or ‘peculiar’velocity arises from the gravitational pull ofnearby mass concentrations10. Zehavi andDekel1 obtained limits on Vm of the Universeby using observations of these peculiar veloc-ities. These measurements favour Vm ! 1,which is a notable result from researcherswho once provided the strongest evidence insupport of a Universe flattened by mass (thatis, Vm 4 1)11.

The area of parameter space highlightedby combining these peculiar velocity resultswith the supernova results is seductivelyclose to the theoretical preference for a geo-

metrically flat Universe. Yet, the danger ofblindly combining (and trusting) the resultsof different measurements is to lose sight ofthe systematic uncertainties involved inindividual techniques. For the peculiarvelocity methods, if one or more assump-tions are not valid (for example, that the ini-tial mass fluctuations are normally distrib-uted) then the implications may not be cor-rect. Similar challenges face the conclusionsdrawn from supernovae data, such as thepossibility of supernova evolution12,13 orexotic intergalactic dust, which could scatterlight from supernovae making them appearfurther away than they are14.

By pushing the observations of super-novae to greater distances (and thereforefurther back in time) it should be possibleto conclusively confirm or refute the initialindications that we are living in an accelerat-ing Universe. We expect a younger, smallerUniverse to be denser and to experiencegreater deceleration from gravity, so theapparent brightness of supernovae shoulddecrease more slowly with distance than islikely to occur under the influence of system-atic effects.

Finally, measurements of tiny ripples inthe radiation left over from the hot Big Bang— the cosmic microwave background — arebeginning to point towards a Universe whichis geometrically flat. These first hints willsoon be surpassed by data from the MAP andPlanck satellites due early next year. By com-bining enough independent constraints itshould be possible to arrive at a set of cosmo-logical parameters that is not vulnerableto systematic uncertainties in any singlemeasurement. Although some cosmologistshave voiced concern that the recent studiesimply that we live at a special time — that is,shortly after the transition from a decelerat-ing to an accelerating Universe — can therebe any time more special than when we firstbegin to learn the answers to our Universalquestions?

Adam G. Riess is at the Space Telescope Institute,Baltimore, Maryland 21218, USA.e-mail: ariess@stsci.edu1. Zehavi, I. & Dekel, A. Nature 401, 252–254 (1999).

2. Turner, M. S. Publ. Astron. Soc. Pacif. 111, 264–273

(1999).

3. Eisenstein, D., Hu, W. & Tegmark, M. Astrophys. J. 518, 2–23

(1999).

4. Lineweaver, C. H. Astrophys. J. 505, 69–73 (1998).

5. Garnavich, P. M. et al. Astrophys. J. 509, 74–79 (1998).

6. Bahcall, N. P. et al. Science 284, 1481–1488 (1999).

7. Riess, A. G. et al. Astron. J. 116, 1009–1038 (1998).

8. Perlmutter, S. et al. Astrophys. J. 517, 565–586 (1999).

9. Schmidt, B. P. et al. Astrophys. J. 507, 46–63 (1998).

10.Strauss, M. A. & Willick, J. A. Phys. Rep. 261, 271–431

(1995).

11.Nusser, A. & Dekel, A. Astrophys. J. 405, 437–448 (1993).

12.Drell, P. S., Laredo, T. J. & Wasserman, I.

http://xxx.lanl.gov/abs/astro-ph/9905027

13.Riess, A. G., Filippenko, A. V., Li, W. & Schmidt, B. P. Astron. J.

(in the press). http://xxx.lanl.gov/abs/astro-ph/9907038

14.Aguirre, A. & Haiman, Z. http://xxx.lanl.gov/abs/astro-

ph/9907039

15.Roos, M. & Harun-or-Rashid, S. M.

http://xxx.lanl.gov/abs/astro-ph/9901234

news and views

NATURE | VOL 401 | 16 SEPTEMBER 1999 | www.nature.com 221

Figure 1 Window on the Universe. To discoverthe values of the parameters Vm (mass density)and VL (cosmological constant) that define theevolution of the Universe, cosmologists overlaydata from different astrophysical techniques tocreate a window in the allowed parameter space.Zehavi and Dekel1 have combined data fromstudies of supernova and peculiar velocities tostrengthen the case for a positive cosmologicalconstant. The results of a similar study15 usingnine independent astrophysical measures areshown here. The cross marks their best guess,where Vm 4 0.3 and VL 4 0.7, and the contoursrepresent relative confidence limits of 68% and98%. The solid line corresponds to a flatUniverse (Vm & VL 4 1), separating an openUniverse from a closed one. The dotted line (VL

4 0) separates a Universe that will expandforever from one that will eventually collapse inthe Big Crunch, whereas the dashed lineseparates a Universe with an expansion rate thatis decelerating from one that is accelerating.

ClosedOpen

Accelerating

Decelerating

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

–0.10.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Expand forever

Eventually collapse

m

Λ

ΩΩ

ΩΩ

100 YEARS AGOOne result of the rapid growth of seismologyis the suggestion of Dr. Mario Baratta thatprovision should be made by insuranceagainst the damage to buildings caused byearthquakes in certain countries. He showsthat, since the beginning of the seventeenthcentury, less than forty earthquakes havebeen responsible for deaths of more than150,000 persons in Italy alone. Moreover, totake but one example, the great loss of lifeduring the Ischian earthquake of 1883 wasdue to the fact that the buildings hadalready been damaged by the earthquakesof 1828 and 1881. Dr. Baratta points outsome of the conditions that must determinethe amount of the premium that should bedemanded by insurance societies. The mostimportant is the degree of seismicity of thedistrict; but this would be modified byothers, such as the nature of the surface-rocks, the character of the buildings, &c.One advantage of compulsory insuranceagainst earthquakes in a country like Italywould be that partially damaged buildingswould be at once rebuilt or repaired, andthis would tend to diminish the loss of life inthe future. From Nature 14 September 1899.

50 YEARS AGOSocial Biology and WelfareBy Sybil Neville-Rolfe…In 1905, at the age of twenty, the widow of anaval officer announced to her relatives herintention “to study prostitution and venerealdisease and try to get rid of them”. Sheknew of no organisation or senior friendunder whose tutelage she could begin. Withcourage and energy, Mrs. Neville-Rolfeplunged into the obscurity and obscurantismsurrounding the problems of sex irregularity;learning the hard way, but backed by theirrepressible gifts of her personality, she hasgiven the last forty-five years to by nomeans unsuccessful efforts to induce thepublic and officials, both at home andabroad, to face squarely and openly some ofthe biological and social factors involved inhealthy and unhealthy sex and familyrelationships in the community. …Unfortunately, the book itself is not a usefulcontribution to the campaign for a rational,humane and continuously constructiveapproach to the problems of social hygiene,the complexities of which grow the more weknow about the field. … Mrs Neville-Rolfe’sgifts are more those of a crusader than of awriter. From Nature 17 September 1949.