I saw this question in test. It says that insertion and deletion is easier in B+ tree compared to B tree. The given reason for this was the key to be deleted must be present in leaf node in B+ tree but the it could be anywhere in case of B tree.
But what i think is, for deletion in B+ tree, we have to go all the way down to leaf and then after deleting, the effect could propagate all the way up to root (merging and redistributing could be required at each level). Instead, if it would have been B tree, we could have found the required key without going all the way down to leaf and its effect could propagate up from their. Didn't we required less work in B tree as compared to B+ tree.
