bitul b'rov and kol d'parish m'ruba

This past Shabbos' daf, Zevhacim 73, finished the gemara's discussion of why an animal is not bateil in a herd, the simplest reason probably being that an animal is a davar chashuv, it is significant. The gemara follows up with a different line of questioning. If an animal is mixed into a herd and you get the herd moving, any animal that jumps out should be permitted based on the principle of kol d'parish m'ruba parish. For example, if there are 100 animals in the herd and one jumps out, odds are that it is one of the 99 heter animals and not the one prohibited animal. The gemara ends up concluding that this is not an acceptable solution because of a gezeirah derabbanan.

It seems clear from the gemara's question that there is a difference between bitul b'rov and holchin achar ha'rov (or kol d'parish -- same thing) or the whole discussion makes no sense -- if rov doesn't apply to a davar chashuv, how can kol d'parish m'ruba apply? The answer must be that these are two different animals (no pun intended). Bitul b'rov doesn't just mean that when a piece of treif meat gets mixed up in a pot with kosher meat we assume that any piece selected is kosher based on the odds. It means much more than that -- it means the treif meat has halachically ceased to exist. Not so the rule of kol d'parish, which is strictly a matter of odds. We know one member of the group is prohibited, but we assume that any one individual item selected must belong to the majority. The sugya in Zevachim assumed that even if we can't go so far as to utilize bitul b'rov, we still might have a way out using kol d'parish (see Shiurei R' Dovid Lifshitz on Chulin 95).

(As an aside, the obvious question is how both of these principles stem from the same pasuk of 'acharei rabim l'hatos' -- see R' Chaim in the stencils.)

This leaves me perplexed by a comment of the Shita M'Kubetzet. The gemara at first is troubled by the use of kol d'parish here, because if animals keep jumping out, at some point a majority of animals have been used up and you know you've taken the asur one. For example, imagine a marble jar filled with 9 blue marbles and 1 red one. If one marble falls out, odds are it will be a blue one -- kol d'parish m'ruba. If marbles keep falling out, e.g. 9 marbles fell out, the odds now tilt against you and are in favor of you're holding the red one. The gemara rejects this objection, but doesn't explain why. Whatever the reason, the Shita m'Kubetzet (#3) writes that this gemara proves that in a case of ta'aroves, e.g. a pot with a bunch of kosher pieces of meat that one trief piece fell into, you can eat the whole pot (other Rishonim disagree) irrespective of the fact that you will definitely be consuming that one treif piece. I don't understand -- the gemara is discussing kol d'parish. What does that have to do with hilchos ta'aroves? Bitul b'rov and kol d'parish m'ruba are two different animals!

Perhaps the Shita means that if the whole collection becomes permissable in a case of kol d'parish, it is certainly permissable in a case of bitul. However, it doesn't sound like that's what he means -- it sounds like the Shita is drawing a direct comparison between bitul and the gemara's case of kol d'parish. I didn't spend time looking around for an answer and nothing struck me offhand. Maybe those of you learning the daf have some ideas?