44:57 Hey Julien, so I think I covered it pretty well with Jonathan’s question, right, because that question went a bit to what were some of the extraordinary cases and I think that's related to your question, what would have let us to settle cases that we thought were significantly larger, and then, we might have expected based on our prior analysis. So, I'm not sure I have a whole lot to add there, just maybe to reiterate the points and then we set this in the past as well Julien. 45:32 We're nothing about consistent on these calls. And the reality is that each of these cases is an individual case and as you know, case by case by case by case across thousands of cases. And I give a lot of credit to our team. They've done a great job both the internal team and with outside support and trying to get our arms around this [uncertainty comp] [ph] from the very beginning and do mappings of the areas that were impacted and have a sense of what kind of household are in each neighborhood that was impacted? And obviously thinking about the tragic toll on too many families. And so, developing a number of estimates that led to the best estimate. 46:13 Initially, you might recall, we didn't provide you a best estimate. We were only able to provide you a low end of the [indiscernible] range, because we’re still at such a large part of the uncertainty come that we couldn't develop the best estimate as we got more experience under our belt we progressed, and we're able to shift through the best estimate, but to be candid about it, there’s been learnings and surprises along the way and not – if there are no surprises, there are surprises, right, with this kind of complex very large case. 46:43 And so, again going back to what I [indiscernible] in response to Jonathan’s question, we saw, and I think the biggest driver this quarter was seeing a small number of very large cases that were well beyond the scope of what we had and anticipated, making – taking a reserve for those cases that we've now settled taking an additional reserve amount anticipating that we might find more surprises in the rest of the tail that's remaining. And then of course, the second big factor that I mentioned in my prepared remarks was also adjusting for the number of [pointers] [ph] that we're seeing. 47:22 That's about all the color we can give you at this point, Julien, but it's been, I think a very deliberate process on the part of our team [indiscernible] very methodical and it's just to reality of the statistics and frankly the probability curve across thousands and thousands of cases.