Parity gravitation tests arise from orthodox physical theory (e.g., teleparallel gravitation), are rigorously derived, reduce to practice, and can falsify broad swaths of contemporary physics without contradicting any prior observation. You see how this outrages theorists and their untestable unorthodox extensions of theory. Somebody should look.
Paul, 1178 words? $25K grows quartz in both hands: chemical purity (no lascas' aluminum), dryness (Grade A), and low dislocation density (etch pits/cm^2). I've talked with academic groups. Senior faculty grant funding wants zero-risk, zero-innovation, PERT-charted productivity. A third class of gravitation experiment that accepts a parity test is being constructed by better people. Years to go.
The Eot-Wash group "vision" is zero-risk, zero-innovation, http://www.npl.washington.edu/eotwash/publications/pdf/lowfrontier2.pdf
It disavows a parity Eotvos experiment opposing spacetime geometry with test mass geometry. 420+ years of perfect failure does not justify a different approach. My paper will be published in a *mathematics* journal, early 2010. Referees passed it. They will not be embarrassed if it works to spec.
Gal - it is not about molecules or compartmentalization. The Coupe du Roi cancels chirality by pairing homochiral entities. The entire mass distribution is considered. A countable number of anonymous unit mass points with overall three finite moments of inertia comprise the test mass (metaphoric " shoe"). Create the inverted set and quantitatively compare. Inversion occurs in many ways. An exact method is required. Turning a shoe inside-out reverses it, as does an odd number of mirror reflections along axes or mixed axes, relative to the origin or to an external point of view.
In 3-space, find the three principle moments of inertia. Their axes intersect at the center of mass as (0,0,0). Build Cartesian (x,y,z) and locate each point (+/- x,+/- y,+/- z). The parity-inverted set is (-/+ x,-/+ y,-/+ z). Fast, easy, exact. A candidate test mass snugs QCM-calculated CHI = 1, perfect normalized parity divergence, by a few angstroms radius from any contained coordinate - and sustaines. A qualifying test mass also QCM calculates COR =1 (the identity element, from graph theory) and DSI = 0 (direct symmetry index - a set's similarity to itself). DSI is... complicated. http://petitjeanmichel.free.fr/itoweb.paper.JMP.1999.petitjean.pdf
Summary: Nobody knows if maximally divergent left and right shoes falsify vacuum isotropy in the massed sector. Said shoes are rigorously defined, reduce to practice, and insert into existing apparatus. Go ahead folks, dump in 1000 "1" votes. You cannot answer the question, I can, and all physics depends on its outcome. The limit to physics is the politics of fear having replaced endeavor, process (management) not product (experiment).