Conservative MOND from first principles

The primary argument levelled against Milgrom's MOND is that it has no theoretical support, even though considerable effort has been expended in attempting to provide it. Against that criticism, MOND irrefutably enjoys an expanding portfolio of success and so is almost certainly tapping into something fundamental. But what? Over roughly the same period that MOND has been a topic of controversy, Baryshev, Sylos Labini and others have been claiming, with equal controversy in earlier years, that, on medium scales at least, material in the universe is distributed in a quasi-fractal $D\approx 2$ fashion. There is a link: if the idea of a quasi-fractal $D\approx 2$ universe on medium scales is taken seriously, then there is an associated characteristic mass surface density, $\Sigma_F$ say, and an associated characteristic acceleration scale $a_F = 4\pi G \,\Sigma_F$. The whole success of MOND is predicated upon the idea of a critical acceleration scale, $a_0$. It is an obvious step to make the association $a_0 \sim a_F$ and then to consider the MOND critical acceleration boundary simply as a marker for a characteristic mass surface density boundary separating 'galaxy' from an environment characterized by $\Sigma_F$. This provides a route to the synthesis of conservative MOND from first principles. The radial acceleration relation (RAR) for conservative MOND when applied to the SPARC sample is the unity line. There is no mass discrepancy.