why does fresh naphtha mean you dont get as much dmt n-oxide? Dmt n-oxide is not soluble in naphtha at all anyways so you shouldnt have it with fresh or reused naphtha.
In any case, the whole question whether you should use fresh naphtha or not, its all about advantages and disadvantages which you have to measure. Yes it is possible that reusing naphtha may mean the product is yellower when freeze precipitating, but you can always recrystalize it later with small amount of solvent (therefore being much less of a waste than using fresh naphtha for every pull). Also, you could wash your solvent before re-using it (doing a couple of basic and then acidic washes) which should get rid of most (all?) impurities, though any dmt that hadnt precipitated before will also be washed away when doing the acidic wash.
I dont use naphtha anymore for extractions but when I did, I did not notice any significant difference in yield amounts (though I never did a side-by-side test to see exactly the difference, if you want to do that sometime we would all be grateful to hear the results). So considering the only difference is a bit of yellowing, which many people dont mind, and even if you did mind it could be recrystalized, then I would think that reusing naphtha is better, but each one has to decide for themselves. You could also do a mix, so using fresh naphtha for the first two pulls, for example, and re-using naphtha for later pulls, or whatever.
By the way, as for yield drop from later pulls, of course it also has to do with DMT depletion, using fresh naphtha should also mean that later pulls will generally have less dmt (with a few exceptions, sometimes a 'freak' may happen that first pull has nearly no yield and second pull has a lot, etc, but there are many variables for that. But if you would graph it, then yes in general later pulls will always have smaller yield because of less dmt)
now the question you can ask yourself is: Has all (or most, or enough) of the dmt been extracted after all these pulls have been made? If you have 1+%, I would say that you should be happy with the results. Never throw anything away before you have your final yield, because of its small, it probably means there's more to be extracted. But in any case it is a fact that naphtha/hexane/heptane are not very effective solvents at pulling dmt. Limonene and aromatic hydrocarbons like xylene (or solvents like DCM) are way way more effective for dissolving dmt. Even if you have done many naphtha pulls and even if its fresh naphtha each time, and if in the later pulls nearly nothing comes out anymore, and if you then do a pull with one of those effective solvents, a significant amount of DMT will probably come out.
So another option of extracting is extracting with one of those solvents, and after evaporating them (or better, specially in the case of limo, salting them out, evaporating the acidic solution and basing it), which should yield you a gooey impurer product, you can recrystalize with small amounts of naphtha. This way you will get a bigger yield with less pulls of a solvent that can be reused and very little naphtha. Or just look at
BLAB tek and extract all the way with limonene, which you can get pure white crystals anyways without having to resort to petrochemicals, and yields are great.