The Evolution of Galaxies on Cosmological

The detection of planetary transits with high photometric precision has a rich potential of various informations, especially for those in close orbit. This contribution gives an overview and new perspectives on the subject, in particular:-planet radius measurement to constraint their internal structure-multicolor determination of the albedo of planets in close orbit-spectroscopic investigation of giant planet environments during transits-inclination of the orbit relative to the star rotation axis. Ground-based and space-based projects dedicated to these searches are reviewed. Potential artefacts are discussed: what they can be, how to discriminate them from real transits? 1. The planet transits method The aim of the transit method, rst investigated by Rosenblatt (1971) and Borucki & Summers (1984), is to search for extrasolar planets by the lumi-nosity drop of a star produced by a planet transit. The relative drop F=F , given by the surface ratio R 2 P =R 2 , runs from 10 ?4 for an Earth-like planet to 10 ?2 for a Jupiter. The low probability p = R =a of a transit (p = 0.5% at a = 1AU for a 1 R star) governs the search strategies discussed in section 2. 1.1. Expected scientiic results Assuming that a transit has been detected, it can be exploited for several science purposes. 1/ If the stellar spectrum allows for high precision radial velocity (RV) measurements , giving the product M P sin i, there is no ambiguity in the planet mass determination since the inclination i of the orbit is exactly =2. 2/ The most obvious output is the radius of the planet. Indeed, the imaging gives only the product A R 2 P where A is the planet albedo (in fact, the planet radius will also be given by the measurement of the thermal ux F th = 4R 2 P T 4 P of the planet, the temperature T P being given by the wavelength at maximum of the planckian spectrum, which will be measured by future mid-IR space missions). The value of the radius is infered from the value of F=F , assuming a value for R ; the latter can be estimated from the spectrum, the brightness and 1