The LHCb experiment has been conceived to study CP violation and other rare phenomena in B meson decays with very high precision. This should provide a profound understanding of quark flavour physics in the framework of the Standard Model, and may reveal a sign of the physics beyond. In order to achieve these goals, the LHCb detector must have a high track reconstruction efficiency, π– K separation capability from a few to ∼100 GeV/c, very good proper-time resolution of ∼ 40 fs and high trigger efficiencies, not only for final states including leptons but also for those with hadrons alone. The detector described in the Technical Proposal (TP) [1], approved in September 1998, was designed to fulfil those requirements. This document describes a reoptimization of the detector, that has been made to reduce the material budget and to improve the trigger performance. At the time of the TP the material budget up to the second Ring Imaging Cherenkov detector (RICH2) was 40% of X0 (10% of λI), where X0 (λI) is the radiation (nuclear interaction) length. This increased to 60% (20%) by the time the Outer Tracker Technical Design Report (TDR) [2] was submitted in September 2001, due to various technological constraints. Additional material deteriorates the detection capability of electrons and photons, increases the multiple scattering of charged particles, and increases occupancies of the tracking stations. With a larger fraction of nuclear interaction length, more kaons and pions interact before traversing the complete tracking system. The number of reconstructed B mesons therefore decreases, even if the efficiency of the tracking algorithm is maintained high for those tracks that do traverse the full spectrometer. This leads to a noticeable loss in the number of reconstructed B mesons from many-body final states. For example, one of the most promising CP violation measurements, from Bs → DsK decays, requires five charged tracks (including one for tagging) to be reconstructed. For these reasons, an effort has been made to reduce the material budget back to the level at the time of the TP. The trigger is one of the biggest challenges of the LHCb experiment1. It is designed to distinguish minimum-bias events from events containing B mesons through the presence of particles with a large transverse momentum (pT) and the existence of secondary vertices. Events are first triggered by requiring at least one lepton or hadron with a pT exceeding 1 to 3 GeV/c (Level-0) reducing the event rate to 1MHz. It was realised that the robustness and efficiency of the second trigger level (Level-1) could be significantly improved by not only using information from the Vertex Locator (VELO), as done in the TP, but also adding pT information to tracks with a large impact parameter. This can be achieved by associating the high-pT calorimeter clusters and muons obtained at Level-0 to the tracks found in the VELO [3]. A complementary approach that is more efficient for hadrons is to get a rough pT estimate from the tracking. This requires the introduction of a small amount of magnetic field in the region of RICH1. The design of RICH 1 then has to be modified in order to protect its photon detectors from the field.