A cosmological scenario is proposed where the dark matter (DM) and dark energy (DE) of the universe are two simultaneous manifestations of an inhomogenous dilaton. The equation of state of the field is scale-dependent and pressureless at galactic and larger scales and it has negative pressure as a DE at very large scales. The dilaton drives an inflationary phase followed by a kinetic energy-dominated one, as in the "quintessential inflation" model introduced by Peebles & Vilenkin, and soon after the end of inflation particle production seeds the first inhomogeneities that lead to galaxy formation. The dilaton is trapped near the minimum of the potential where it oscillates like a massive field, and the excess of kinetic energy is dissipated via the mechanism of "gravitational cooling" first introduced by Seidel & Suen. The inhomogeneities therefore behave like solitonic oscillations around the minimum of the potential, known as "oscillatons", that we propose account for most DM in galaxies. Those regions where the dilaton does not transform enough kinetic energy into reheating or carry an excess of it from regions that have cooled, evolve to the tail of the potential as DE, driving the acceleration of the universe.

The least-action principle (LAP) method is used on four galaxy redshift surveys to measure the density parameter Omega_m and the matter and galaxy-galaxy power spectra. The datasets are PSCz, ORS, Mark III and SFI. The LAP method is applied on the surveys simultaneously, resulting in an overconstrained dynamical system that describes the cosmic overdensities and velocity flows. The system is solved by relaxing the constraint that each survey imposes upon the cosmic fields. A least-squares optimization of the errors that arise in the process yields the cosmic fields and the value of Omega_m that is the best fit to the ensemble of datasets. The analysis has been carried out with a high-resolution Gaussian smoothing of 500 km/s and over a spherical selected volume of radius 9,000 km/s. We have assigned a weight to each survey, depending on their density of sampling, and this parameter determines their relative influence in limiting the domain of the overall solution. The influence of each survey on the final value of Omega_m, the cosmographical features of the cosmic fields and the power spectra largely depends on the distribution function of the errors in the relaxation of the constraints. We find that PSCz and Mark III are closer to the final solution than ORS and SFI. The likelihood analysis yields Omega_m= 0.37\pm 0.01 to 1sigma level. PSCz and SFI are the closest to this value, whereas ORS and Mark III predict a somewhat lower Omega_m. The model of bias employed is a scale-dependent one, and we retain up to 42 bias coefficients b_{rl} in the spherical harmonics formalism. The predicted power spectra are estimated in the range of wavenumbers 0.02-0.49h Mpc^{-1}, and we compare these results with measurements recently reported in the literature.

The measurement of cosmological parameters is investigated in a representation of the least-action method that uses a redshift-space dataset to simultaneously constrain the real-space fields $\delta$,$\b v$. This method is robust in recovering the entire evolution of the matter density contrast and peculiar velocities of galaxies in real space from current galaxy redshift surveys. The main strength of the method is that it permits us to break the degeneracy of the parameters $b$ and $\Omegam$ (customarily measured in the ratio $\beta\equiv \Omegam^{0.6}/b$ from redshift-space distortions), and these are evaluated in the current context separately. The procedure provides a simple numerical means to extract as much information as possible from a given sample, in the simplest linear bias model, before resorting to cosmic complementarity to resolve the degeneracy in the measurement of $\Omegam$. The same premise applies to more sophisticated choices of bias models. We construct a likelihood parameter $\lambda(b,\Omegam)$ to evaluate the relative likelihood of different values of $b$ and $\Omegam$. The method is applied to the \iras redshift survey with a low-resolution Gaussian smoothing length of 1200 \km within a spherical region $x_{\rm max} \sim 15,000$ \km and the reconstructed velocity field is then compared with POTENT-reconstructed velocities from the Mark III radial-velocity dataset within a radius $\sim 5000$ \km, which have been suitably prepared to account for Malmquist bias and other systematic errors. The analysis yields a likelihood for the parameters that is overall consistent with $\Omegam\approx 0.3$ and $b\approx 1.1$, thus lending support to a non-vanishing cosmological constant $\Omega_{\Lambda}\approx 0.7$ in a flat universe.

Brans-Dicke gravity is remarkable not only in that General Relativity and Mach's Principle find a common enlarged scenario where they are mutually consistent, but also in that it provides a very interesting quantum cosmological model within the inflationary paradigm. The interplay between the Brans-Dicke scalar $\Phi$ and the inflaton field $\sigma$ plays an important r\^{o}le during the course of inflation, and although the dynamics as such is governed by the potential, the onset and the end of inflation are determined by the values of both fields jointly. The relative position of the beginning-- and end-of-inflation curves (BoI and EoI respectively) is the most relevant factor in determining the resulting quantum cosmological scenario. The classification of potentials that is given in this paper is based on the criterion of whether the BoI and EoI boundaries enclose a finite or infinite area in the ($\sigma$,$\Phi$) plane where inflation takes place. It is shown that this qualitative classification distinguishes two classes of potentials that yield very different cosmologies and it is argued that only those theories in which BoI and EoI enclose a finite area in the ($\sigma$,$\Phi$) plane are compatible with our observable universe.

In this paper we investigate extended inflation with an exponential potential $V(\sigma)= V_0 e^{-\kappa\sigma}$, which provides a simple cosmological scenario where the distribution of the constants of Nature is mostly determined by $\kappa$. In particular, we show that this theory predicts a uniform distribution for the Planck mass at the end of inflation, for the entire ensemble of universes that undergo stochastic inflation. Eternal inflation takes place in this scenario for a broad family of initial conditions, all of which lead up to the same value of the Planck mass at the end of inflation. The predicted value of the Planck mass is consistent with the observed value within a comfortable range of values of the parameters involved.

A first- and second-order relation between cosmic density and peculiar-velocity fields is presented. The calculation is purely Lagrangian and it is derived using the second-order solutions of the Lagrange-Newton system obtained by Buchert & Ehlers. The procedure is applied to two particular solutions given generic initial conditions. In this approach, the continuity equation yields a relation between the over-density and peculiar-velocity fields that automatically satisfies Euler's equation because the orbits are derived from the Lagrange-Newton system. This scheme generalizes some results obtained by Nusser et al. (1991) in the context of the Zel'dovich approximation. As opposed to several other reconstruction schemes, in this approach it is not necessary to truncate the expansion of the Jacobian given by the continuity equation in order to calculate a first- or second-order expression for the density field. In these previous schemes, the density contrast given by (a) the continuity equation and (b) Euler's equation are mutually incompatible. This inconsistency arises as a consequence of an improper handling of Lagrangian and Eulerian coordinates in the analysis. Here, we take into account the fact that an exact calculation of the density is feasible in the Lagrangian picture and therefore an accurate and consistent description is obtained.

In the context of Brans--Dicke theories, eternal inflation is described in such a way that the evolution of the inflaton field is determined by the value of the Planck mass in different regions of the universe. The Planck mass is given by the values of the Brans--Dicke field, which is coupled to the scalar curvature in the Lagrangian. We first calculate the joint probability distributions of the inflaton and Brans--Dicke fields, in order to compute the 3--volume ratios of homogeneous regions with arbitrary values of the fields still undergoing inflation with respect to thermalized regions. From these volume ratios one is able to extract information on the values of the fields measured by a typical observer for a given potential and, in particular, the typical value of the Planck mass at the end of inflation. In this paper, we investigate volume ratios using a regularization procedure suggested by Vilenkin, and the results are applied to powerlaw and double--well potentials. The spectrum of density fluctuations is calculated for generic potentials, and we discuss the likelihood of various scenarios that could tell us whether our region of the universe is typical or untypical depending on very general bounds on the evolution of the Brans--Dicke field.

A scheme is developed which enables one to trace backwards in time the cosmic density and velocity fields, and to determine accurately the current-epoch velocity field from the current-epoch density field, or vice versa. The scheme implements the idea of Giavalisco \etal\ (1993) that the principle of least action should be used to formulate gravitational instability as a two-point boundary-value problem. We argue that the Eulerian formulation of the problem is to be preferred to the Lagrangian one, on grounds of computational simplicity, of ease of interfacing with observational data, and of internal consistency at early times. The scheme is successfully tested on an exact solution in one dimension, and on currently Gaussian fields in one and two dimensions. The application of the scheme to real observational data appears to be eminently feasible, though computationally costly.