OK, I think that explains that - Wikipedia is making the first assumption identified below, rather than the other one that he prefers:

"If the only force connecting mirror matter with ordinary matter is gravity, then the consequences would be minimal. The mirror SB would simply pass through the Earth and nobody would know about it unless it was so heavy as to gravitationally affect the motion of the Earth.

However if there is photon - mirror photon kinetic mixing as suggested by the orthopositronium vacuum cavity experiment, then the mirror nuclei (with Z ′ mirror protons) will effectively have a small ordinary electric charge ǫZ ′ e. This means that the nuclei of the mirror atoms of the SB will undergo Rutherford scattering off the nuclei of the atmospheric nitrogen and oxygen atoms. In addition ionizing interactions can occur which can ionize both the mirror atoms of the space body and also the atmospheric atoms. The net effect is that the kinetic energy of the SB is transformed into light and heat (both ordinary an mirror varieties) and a component is also converted to the atmosphere in the form of a shockwave, as the forward momentum of the SB is transferred to the air which passes though or near the SB.

What happens to the mirror matter SB as it plummets towards the Earth’s surface depends on a number of factors such as its initial velocity, size, chemical composition and angle of trajectory. Of course all these uncertainties occur for an ordinary matter SB too. Interestingly it turns out that for the value of the kinetic mixing suggested by the Or- thopositronium experiment, ǫ ≈ 10−6, the air resistence of a mirror SB in the atmosphere is roughly the same as an ordinary SB assuming the same trajectory, velocity mass, size and shape (and that it remains intact). This occurs because the air molecules will lose their relative forward momentum (with respect to the SB) within the SB itself because of the Rutherford scattering of the ordinary and mirror nuclei as we will show in a moment. (Of course the atmospheric atoms still have random thermal motion). This will lead to a drag force of roughly the same size as that on an ordinary matter SB, implying an energy loss rate ....

The above calculation shows that the rate of energy loss of the SB in the atmosphere depends on its size and density. If we assume a density of ρSB ≃ 1 g/cm3 which is approximately valid for a mirror SB made of cometary material (such as mirror ices of water, methane and/or ammonia) then the body will lose most of its kinetic energy in the atmosphere provided that it is less than roughly 5 meters in diameter. Of course things are complicated because the the SB will undergo mass loss (ablation) and also potentially fragment into smaller pieces and of course potentially melt & vaporize. Thus even a very large body (e.g. R ∼ 100 meters as estimated for the Tunguska explosion) can lose its kinetic energy in the atmosphere if it fragments into small pieces.

...Returning to the most interesting case of large photon - mirror photon kinetic mixing, ǫ ≃ 10−6 which is indicated by the orthopositronium experiment, our earlier calculation suggests that most of the kinetic energy of a mirror matter SB is released in the atmosphere like an ordinary matter SB if it is not too big (∼ 5 meters) or fragments into small objects. It seems to be an interesting candidate to explain the 1908 Tunguska explosion (as well as smaller similar events as we will discuss in a moment).