The graph showing the effect of relative proportions of Phenidone and hydroquinone does not give a clue as to the need for sulfite. The earlier graph in the same chapter showing the effect of sulfite, ascorbic acid or hydroquinone on development by Metol shows that there was no increase in rate of development by Metol as hydroquinone was added. Even if the PQ mixture was synergistic without sulfite while the MQ mixture was not, the two sets of experiments show that we cannot conclude that sulfite is not needed for PQ synergism because the other ingredients of the PQ developer were not mentioned.
You must realize that the rate of development is only one aspect of developer, and distinguish this from a range of factors that must be considered in formulating practical developers. For the latter purpose, there are other requirements, such as photographic speed, compatibility with a wide range of emulsion types, tonality, granularity and keeping properties.
Now being specific to the rate of development, your reading as summarized above is correct. Hydroquinone can regenerate oxidized Phenidone more effectively than oxidized Metol. The presence of sulfite may not be very obvious in PQ rate of development, but it has a useful effect that may not be obvious depending on the quantities of hydroquinone. That is, sulfonated hydroquinone is effective in regenerating oxidized Phenidone but not oxidized Metol.
In practical b&w developers, it is customary to include sulfite or other forms of sulfur dioxide salts, because of the factors listed above. You can't deny the reasons behind this by just arguing the development rate.
Now, the MQ and the MA developers tested without sulfite were vastly different one from the other. We cannot tell from the graph whether the ascorbate was regenerating the Metol, or deactivating the oxidation products of the Metol. The author supposes that the sulfite deactivates while the ascorbate regenerates. Apparently, the hydroquinone does neither without sulfite, which means that the MQ mixture is not synergistic. I don't see the error in my logic. I see the lack of data other than what I get from my own experiments, and so would really like to do more reading. I'm sure the original reports had more detail.
Original report is one thing, but newer research relied more heavily on more direct and relevant aspect of electrochemistry. Electrochemistry was widely recognized to be a very useful tool in photography in 1960s, and in Kodak Rochester lab alone, a huge deal was made with this one branch of chemistry, using a range of techniques like cyclic voltammetry. People like Paul Gilman, J. E. Jones, et al. collaborated with their electrochemists to study a wide range of sensitizing dyes, which is also an electrochemical phenomenon. Carroll and West were also interested in this subject, but Brooker (Brooker and West are dye experts) retired around that time. Pontius, Thompson, Cole, Willis and others (they are electrochemists) worked on developers. Reading James 3rd edition is not a good idea for this particular subject, which saw significant progress after the publication of the book. James 4th edition is a better read (I don't recall why I sent you the 3rd... sorry!) If you appreciate the results of electrochemical research in developers and sensitizing dyes, you will realize how powerful the approach was. Electrochemical studies of photographically useful compounds were also very actively carried out in East Germany and Japan in 1960s.
Hydroquinone does regenerate oxidized Phenidone or Metol without sulfite, but sulfite makes this process a lot more efficient by removing oxidized hydroquinone by sulfonation reaction. Ascorbate does not require sulfite to remove its oxidized form from the system because it hydrolyzes by itself rapidly, but sulfite remains a useful addition in practical developers for the reasons I listed above.