Discrete source counts and the rotation of the local Universe in hierarchical cosmology

Attention is given to four reasons for believing that the upper limit on the rotation of the Universe ω set by isotropy of the 3K background may not be appropriate to the local system because of its hierarchical structure. In particular, recent work of Rubinet al. (1973) on the anisotropy of Hubble's parameter (H) as determined by certain galaxies is examined. The anisotropy inH is a 1st order harmonic effect, inconsistent with an origin in an acceleration of the expansion of the Universe (U _α;4≠0), but explicable as being due to a large peculiar velocity of the Local Group. This compromises limits set on ω by isotropy of the 3K field, as does the realization that only weak limits can be set if the last-scattering surface (z _*) is notz _*→∞ but is at smallz _* (as expected in a hierarchy). In a rotating Universe, the 3-spaces of constant density cannot be orthogonal on the world lines of matter: a number test of Gödell based on this is generalized and applied (after consideration of Galactic obscuration) to the local Universe, by taking data on clusters of galaxies from the Abell and Zwicky catalogues. Data from the former give only a marginally significant result for the component ω₁ of ω in one direction, but a bootstrap argument is applied which takes significance over from Abell's data (considered as a class of galaxies) to Zwicky's data (taken as a class of clusters), giving a statistically significant result on the hypothesis that clusters are the fundamental units of the Universe: it seems likely that ω₁r?(const)r^-n with 0?n?1 over the interval 500–1000 Mpc (H=60 km s^?1 Mpc^?1) with a total rotation of ω<2ω₁, and ω₁ = 1.2 (+0.25) x 10^-18 s^-1 evaluated on data out to 10³ Mpc. Strictly, the quoted value of the rotation only applies to a region of space that in some sense has an isotropic limit: if the actual hierarchy has a large density-dependence away from a local origin (i.e., large thinning factor), then the numerical value of the rotation is smaller than the quoted value but still finite and significant.