Cabinets along a wall are secured to the wall. With Cabinets in an island you need to consider the situation to determine how much support you need. Since we are DIYers we consulted with the granite supplier/fabricator to determine the support needed. Then DH did a little more just to be sure. Here is our island. Note that we have a 1.5 " plywood base with a heating element. So our island top is heavy than just granite. Anyway the same principles apply. See the electric outlets on the back of the cabinets? That enclosure holds the electric and an L bracket bolting the cabinet to the floor. I think the L bracket could have been put inside the cabinet but his was how DH chose to do it. The SS leg has an L bracket at the rear (toward window) that is just barely visible. DH was preparing for worst case scenario.

Many people on this forum have noted that marbles etches at the blink of an eye. We don't have experience with that yet, because we're only just beginning to move back into our new kitchen and we're still using our temp kitchen for food prep. My marble island is two levels: The prep (lower) level is taken up by a prep sink and giant butcherblock cutting board, so the remaining marble surface is likely only to be affected by water. The eating (counter height) level is where we plan to be "careful", using placemats and runners under plates and platters--that is, until the first etching starts (boo-hoo), and then we'll see how careful we'll be after that!

Update: After the advice on this thread, We went with the single tier. The island back was built out a little more to give us more room behind the sink. Everything turned out great. The Granite is Lennon which has whites, browns, greys, and some copper/bronze. Also, there have been a few posts lately about granite costs/overages and trying to reduce budget by going with limited slabs. If you want a nice uniform look, I would suggest picking out multiple slabs so you can get a nice match/book match. See below how my granite vendor accomplished this. Other experts on here can go into other details. I love the way everything turned out. Can't notice the seam in the middle of our island unless you broke out a magnifying glass. There is a seam right in the middle but you cant see it.

I review the plans of everyone working on my house. [...] My general contractor and the granite fabricator/installer will warrant their work, and like everything else going into my house, I want to double check, and I want them to justify their plans if they differ from what I have learned from Professor Google. We don't have their plans though, if that sketch was given to you as their plan for support, then you are in trouble and need to hire another professional. To evaluate their plan we need to know what system they plan to use to limit flex. Make a wall beyond the cabinets, or close in the ends. for additional support. I am not a huge fan of walls that dissect long pieces of granite unless the people building the walls really understand what they are doing. I think the post is a much better idea that accomplishes substantially the same thing as a wall. ----------------------------------- When we talk about supporting granite we are talking about one thing. Limiting the tensile pressures. Granite, just like concrete, has great compressive strength and rotten tensile strength, when granite flexes you get compressive force on the concave side and tensile force on the convex side. Moving from theory to reality in granite we really have to do three things, (1) limit flex, (2) limit movement and (3) avoid isolation. Many inexperienced people will address movement and flex while ignoring isolation and when that happens you get the result that Joseph posted above. If we look at the brackets that are often used for granite islands we can see that their real purpose isn't to stop movement but to avoid creating a fulcrum so that the entire sheet of granite moves together. http://www.countertopbracket.com/countertop-island-support-brackets-hidden-p/ib.htm I guarantee you that something heavy placed on that overhang will cause the countertop to tip slightly in that direction, but the brackets ensure that the movement is carried past the edge of the cabinets so that the entire piece moves rather than creating a fulcrum at the back of the cabinets that isolates the overhang from the cabinets. Too much movement must be avoided because then you get the shock of the cabinets moving which can be bad so walls in long islands may be necessary but when they are you must avoid isolating them. In other words, in Joseph's example above had the brackets been extended and tied into the cabinets in such a way that the entire piece was tied together or tied to the cabinets sufficiently that everything moved together, the break is unlikely to have occurred. Technically the brackets in Joseph's post above did exactly what they were designed to do. They eliminated breakage at the fulcrum...They just created a new fulcrum because their designer didn't understand the problem. From a physics stand point, a post is nothing but a very short wall. You can get substantially the same results from a post that you can from a wall, by adding cost. Essentially the white arrows would be a wall while the red arrows will be a post. So long as the post doesn't flex more than a wall (which it will compress less than a wall), mechanically you can achieve the same thing. However, you create more isolation points, all of the red arrows need to either have very little movement at all, or be tied together in such a way that they move together (note the arrows actually form a circle). Achieving no movement is much harder than tying them together in such a way that the limited movement extends through the area. A 2" tubular frame under the entire granite surface would absolutely be preferred, as that is the best way to carry movement to the entire counter. My original advocacy for plate was in opposition of L brackets rather than a tubular frame, to avoid isolation. I think you can achieve much the same result with a steel plate (a 36" steel plate is roughly equivalent to about 20 brackets) but I am not opposed to a tubular frame, especially if you are going as deep as 2". Anyway, this is just my 2 cents. In the end, to evaluate the solution that your contractor is using we need to know that.