I found in MEGA5, after bootstrap test, there was no scale bar in my bootstrap consensus tree. While in the original tree, there was a scale bar. Does anyone know why this happens in MEGA5?
I do not know of any program which gives consensus distances on a consensus tree. All consensus trees I have seen only show the topology of the consensus and not the consensus distances. I am not even sure how those distances would be computed for branches that were not supported by 100% of the trees. It is common, if you want to publish a tree with distances, to transpose the bootstrap values from the consensus tree onto another tree (maximum likelihood tree for example).
With RaXML, you can search for the best ML tree and bootstrap on the same run, and it will give you a file with the best tree (with scale bar) and the bootstrap values as branch labels.