Of course, there are a number of areas in which cosmological theory and observations are not in obvious agreement. The space available does not permit an exhaustive review, so I will concentrate on the following topics which seem to me to be the most important: big bang nucleosynthesis, galaxy centers, dark matter substructure, and angular momentum. Fortunately, some areas which had seemed problematic now seem less so. For example, the low values of the quadrupole and octopole CMB anisotropies reported by WMAP are revised upward in an improved analysis, and are now in good agreement with the CDM predictions . And the possibility that the first year WMAP data imply that the visible universe might be topologically complex and smaller than the horizon  has now been stringently constrained by the absence of pairs of circles in the WMAP data . Another topic that has received more attention in the press than is warranted by the science is the discovery of massive galaxies at high redshifts . Although these observations challenge some overly simplified theories of galaxy formation , there are plenty of sufficiently massive halos at the relevant redshifts (with 8 = 0.9) to host the galaxies in question, as Figure 1 shows.
Figure 1 Curves show the cumulative number density of dark matter halos more massive than 1011 to 1015 M (from top to bottom, as labeled), as a function of redshift, calculated using an improved Press-Schechter formula . Points show the estimated comoving number densities of several observed populations, as follows. Hexagons: galaxies with stellar masses greater than 1011 M (lower point) and 2.5 × 1010 M (higher point), obtained by integrating the stellar mass function from SDSS+2MASS; open square: Extremely Red Objects; open triangle: K20 galaxies; crosses: sub-mm galaxies; six pointed stars: Lyman break galaxies, filled square: quasars. For all populations, CDM predicts that there are enough dark matter halos massive enough to plausibly host the observed objects. (From Ref. , which gives the references for the data points plotted.)