With the reference to Marshall, the subject header seems an odd way to pose  
the question of putting price on the vertical axis.  
  
Rau did it, and published it in 1841.  
  
Mangoldt did it, and published it in 1863.  
  
Cournot (1838) and Dupuit (1844) put price on the horizontal axis.  So why  
did Rau reverse Cournot's axes?  And why did Mangoldt reverse Cournot's and  
Dupuit's axes, following Rau instead?  
  
You could direct the student to the very detailed paper by Thomas Humphrey,  
"Marshallian cross diagrams and their uses before Alfred Marshall:  The  
origins of supply and and demand geometry," Economic Review (Federal Reserve  
Bank of Richmond), March-April 1992, Volume 78, Issue 2, pp. 2-23.  
  
Note also:  If you think of yourself as a price-taking consumer, then of  
course price is the "independent variable."  And along the downward-sloping  
market demand curve one can fix price, then use the curve to determine  
market quantity demanded.  But one can also fix the market quantity and then  
use the curve to determine the highest price at which that quantity would be  
bought, treating market quantity as "independent."  The logic of the curve  
works both ways.  
  
Torsten Schmidt