I have argued as long ago as 1998, and more recently in a paper published in this journal in 2003, that the beginnings of Sraffa's work for Production of Commodities—identified as the formulation in 1927 of some systems of equations, which later brought him to the equations we find in the book—owed much to his (re)reading of Marx, at the time when he started to prepare the lectures on ‘Advanced theory of value’, he was supposed to give in Cambridge in the Autumn of 1927. I have shown that probably the most important text for Sraffa at this stage was Marx's Histoire des doctrines économiques (the French edition of Theories of Surplus Value), which led him to discover the materialistic conception of cost and value of the Physiocrats, and their Tableau économique, which in turn brought Sraffa to study Marx's schemes of (simple) reproduction, which were the source of Sraffa's equations. In an article published in 2015 in this journal, A. Ginzburg disagrees, basing himself on a paper by Garegnani of 2003, where it is maintained that Sraffa's formulation of his equations was wholly independent of his study of Marx, and was in fact reached before ‘discovering’ Marx as a serious economic theorist. I point out that documents to be found in Sraffa's papers show that Sraffa's appreciation of Marx and his reproduction schemes are not later than the formulation of Sraffa's equations in the Autumn of 1927. Moreover, I show that Garegnani's reconstruction (endorsed by Ginzburg) of the origin of Sraffa's equations as deriving from an attempt to reduce the ‘real’ cost of a commodity to a quantity of a single commodity, necessary for its production, does not stand up close scrutiny. Some other arguments brought forward by Ginzburg to substantiate his claims against my thesis are also discussed, and found wanting.