It seems that there are 3 "versions" of the Moscow edition. I'll call them,
A, B and C.
1. Version A was published in the 50s as FLPH and Progress Publishers and
reprinted in the 60s. The features of version A are:
I: 807 pp, 1954, no Subject Index
(A) II: 546 pp, 1957
III: 923 pp, 1958 (59?)
2. In the 60s, they started to produce version B, probably beginning with
Vol III and, after this, books II and I. The editorial house was Progress.
The features of the books are:
I: 767 pp, 1970?, includes Subject Index.
(B) II: 551 pp, 1967?
III: 948 pp, 1966?
(I'm not completely sure about the dates of publication. They are simply
the earliest references *I* found to these books.)
3. L&W and International were still printing clones of version A in the
70s. It's possible that they shift to version B only in 1983 (L&W) and 1984
(Intl.) although, perhaps, vols II and III, version B, were introduced
before this date.
4. Version C is, of course, that of the Collected Works:
CW 35, I: 852 pp, 1996.
(C) CW 36, II: 546 pp, 1997.
CW 37, III: 982 pp, 1998.
The latter is actually no so important because there are not references to
these books.
5. In the literature there is a real mess regarding this edition because it
seems the books were sold mixing versions A and B and the editors didn't
say a word about this. To illustrate this problem, here I have a sample of
authors quoting the Moscow edition and the versions of each volume they
refer to:
Howard and King [1975], The Political Economy of Marx: A, B, B
Pilling [1980], Marx's Capital: A, A, A
Foley [1986], Understanding Capital: A, A, B
Carchedi [1991], Frontiers...: A, A, B
Weeks [1981], Capital and Exploitation: B, B, B
Catephores [1989], An Introduction...: A, B, B
Morishima [1973], Marx's Economics: A, ?, B
De Brunhoff [1976], Marx on Money: A, A, B
Roberts [1981], Value Categories... A, ?, B
(This means that, e.g. Howard & King [1975] quotes version A for Vol I and
version B for Vols II and III.)
Of course, the authors didn't know that they were working with different
versions of the books because the editors never stated that. It's usual to
refer to Moscow edition as "1967" probably because this is the copyright
date in most of the clones. I have found no author referring to the number
of pages of the volume they are quoting so it's difficult to know what is
the book they are citing.
This would deserve a better bibliographical research. For instance, maybe
the "versions" are really different "editions" involving changes beyond the
pagination. As Chris notes, in the case of Vol. I, version B included a
Subject Index that it is not in version A.
Alejandro Ramos