"Anonymous. In the Foxtrot according to you the rise we get on step two is not maintained onto step three."
The quality that the book considers to be rise is maintained, what is not maintained is the actual altitude.
"Now you are quoting Len Scrivener Is is it 20 or is it 30 years ago that he wrote the technique book is wrong."
No, I'm not quoting with him. While I agree with him about what the dancing looks like overall, unlike him I don't consider the book to be wrong. It's accurate for what it describes, but what it described is only part of the picture, not the whole story. The book's idea of rise and fall is only one part of what creates the overall change in altitude.
"If we arrive on step two with height we must maintain that height onto the beginning of step three and then start to lower. Otherwise we would be lowering on step two which would mean that knee is bending too early."
The problem is that you are thinking only in terms of the action of the feet, ankles, and knees, and forgetting about the change in altitude due to the legs being apart, then together, then apart again. As the feet close at the end of step two, the altitude peaks. From there, it starts decreasing again, not because the knees are bending or the feet lowering (that happens later), but because the legs are separating. The book doesn't talk about this component of change of altitude, but it can be plainly seen in any video of good dancing - and the bigger the movement is, the more obvious it becomes.
"If it is possible try to keep an eye on their heads in relation to the wall to see the rise and fall of which there is plenty of."
You will see continuing gain of altitude until the end of step two, and loss of altitude beginning immediately thereafter.