The A7 and A7r will almost certainly be VERY difficult to get the most out of from a lens perspective - the D3x and especially the D800(e) are notorious for being extremely picky about lenses. I have used both extensively, and I never found a zoom I really liked on the 800e (even the 24-70 f2.8 Nikkor wasn't quite up to the sensor). There are some primes that are good enough to show off those sensors - the only one I ever shot much with was the 105 Micro VR - but I have always used zooms (apart from macro lenses), simply due to the nature of my photography - I don't doubt that there are a decent range of superb primes that take full advantage of the sensor.
Assuming that the A7r is about as picky as the D800e, and the A7 is about as picky as a D3x, which doesn't seem a bad assumption, given the sensor similarities, how badly constrained will they be by available lenses? There seems to be little question that the 35mm Zeiss is a superb lens, and I wouldn't be at all surprised if the 55mm was equally good. I'm not sure about the Zeiss 24-70 zoom - some Sony Zeiss zooms are superb, but others, including the new 16-70 for NEX, aren't great. The Sony 28-70 kit lens will probably be cut to ribbons by those ultra-high resolution sensors, and that seems apparent in early reviews - I have seen few good words about that lens.
Adapters are another question entirely - in addition to losing AF and image stabilization, they are hard to get completely straight - one more joint between body and lens that can be off by some fraction of a mm... Adapters are also notorious for not working terribly well on ultra wide lenses. At least right now, I see the line up of truly good lenses for the A7 (r) as stretching from about a 24 (Leica, among others, on a top-quality adapter) through 35 and 55mm Sony Zeiss lenses to 75 and 90mm rangefinder lenses, again on a top quality adapter. There is no really good option for a macro lens, either - the only possibilities will be FF DSLR lenses on adapters, making them even bigger than they already are, and no AF or IS. Above 90, the fact that adapted lenses will generally have no AF (or extremely slow AF at best) or IS seems a huge drawback. If it is a superb lens, the 24-70 Zeiss will cover most of that range with AF and IS intact (although with no close focusing). There is also the option of really high end DSLR primes (Otus, etc.) on adapters, which will offer truly exceptional performance, at a huge cost in practicality.
Outside that limited range, or with less expensive or more versatile lenses within the limited range, will the A7 (r) offer better image quality than other mirrorless options, once body and lens are considered together? I would not be AT ALL surprised if an E-M1 with the 12-40 f2.8 Olympus or an X-E2 with the beautiful 18-55 Fuji can actually outresolve the Sony with the kit lens (I have NO question that the Sony sensor is better, but will the lens give back the sensor's gains)? I am sure that the A7r with the Zeiss zoom will outresolve the competition, but at twice the price, and by how much? Will the A7 with the Zeiss zoom actually outresolve the Olympus and the Fuji by enough to matter most of the time - or are the Sonys specialized cameras which only show their superiority with a limited number of primes, losing stabilization and even AF.
If the A7 and A7r do turn out to be severely lens-constrained, photographers will have to consider them VERY carefully, based upon their type of shooting. If you're a classic Leica rangefinder shooter, great - I'm 100% sure they'll turn in a better performance than anything except (possibly) an M240, at a much lower price (with the option of a couple of AF lenses). If I preferred primes at a range of moderate focal lengths, I'd have a Sony for sure. You also have the Otus and a couple of other options that offer medium-format image quality, at the cost of a ton of weight and bulk.
If you prefer zooms, you have ONE lens choice (the Zeiss), if it turns out to be good enough - there is a good possibility that it won't really outresolve a lesser camera with a better lens. With no real reviews of the Zeiss yet, I'd speculate based on the history of Sony Zeiss zooms and the few preliminary shooter's reports that it has (very roughly) a 33% chance of being no better than an E-M1 with the 12-40 or an X-E2 with the 18-55, a 33% chance of being only a little better (lens is letting the sensor down, but it is noticeably slightly better than the next-best mirrorless option), and a 33% chance of living up to at least the 24 mp sensor and possibly the 36, and being the best highly portable body/zoom combo by a wide margin. At first, there is no ultra wide zoom, and one telephoto zoom that may not live up to the sensor, only goes to 200mm, and is going to be big and heavy when it finally appears. If you want a macro lens, you're entirely out of luck so far... Sony's record on high-quality lenses for the NEX system isn't exactly stellar, either - more are starting to appear (and some, like the 10-18, seem very good), but the history has been a frustrating mishmash of multiple versions of the same consumer-grade zooms, with an occasional very nice prime thrown in.
I ended up expanding my m4/3 system after very careful consideration of adding a Sony instead (I'm a zoom and macro shooter, because I do nature, but on long hikes, so I want neither the weight nor the bother of a bagful of primes), primarily because I thought I could equal the performance of anything except unstabilized primes and the very expensive A7r/Zeiss combo in a system with a lot more versatility. I am NOT claiming that my choice will be right for everyone, simply raising the question of how hard the new Sonys will be to pair with lenses, and how this affects their final IQ in the print compared to other mirrorless options.