Academic journal article
By James, William H.
The Journal of Sex Research , Vol. 37, No. 4
The authors offer some propositions which I take to be logically flawed. These may be summarized as follows:
(A) the authors write in their abstract `(1) sex determination is random, or (2) sex determination is facultative or biased', (B) the authors claim (1) there is evidence against Weinberg's Rule--that among human dizygotic (DZ) twin pairs, there are almost exactly equal numbers of same-sexed (SS) and opposite-sexed (OS) pairs--and that (2) this evidence constitutes grounds that support `the biased or facultative position' (their p. 169).
I shall deal with these points in order.
In this context the word `facultative' may be taken to mean "tending to confer reproductive advantage as e.g. in producing more grandchildren." There are many examples of offspring sex ratios of subpopulations which differ significantly from those of their parent populations, such as the offspring sex ratios of men exposed to a number of deleterious chemicals (e.g., dioxin: Mocarelli et al., 1996; and boron: James, 1999); or men destined to suffer from various diseases such as prostatic cancer (James, 1990) or testicular cancer (Moller, 1998). However there are no grounds for supposing that these sex ratio biases are facultative: indeed it is invalid to infer (without further argument) a facultative function for an established sex ratio bias. The point has been satirized by Gould & Lewontin (1979). It is only from documented reproductive advantage that one may infer facultative sex ratios.
Most of the data in the authors' Table 1 was reviewed in the most recent study cited in that table (James, 1992). It is worth explaining why the evidence in that table (which at first sight looks overwhelming) remains indecisive.
In estimating the proportion of DZ twins which are SS and OS, it is usual to observe the following rules in regard to a randomly ascertained sample of twin pairs.
(1) All opposite-sexed twins pairs are DZ.
(2) For practical purposes, all monochorionic twin pairs are MZ.
When these two categories of twins are set aside, one is left with the subsample of same-sexed dichorionic pairs. These may be either MZ or DZ. These twins of unknown zygosity are then subjected to a series of tests by genetic markers (e.g., of blood groups). Each pair which is discordant on any marker is diagnosed DZ. However, after the testing, the remaining twin pairs (which are concordant on all markers) contain all the MZ pairs plus some undiagnosed SS DZ pairs which (by chance) happen to be concordant on all the markers so far employed. It is, in principal, possible to estimate this number of concordant SS DZ pairs by applying the same set of markers to the opposite-sexed pairs (which are known to be DZ). However, though the OS twins in many of the cited studies were tested on such markers, it is clear that this testing was, in general, not so rigorous (viz on not as many markers) as that on the SS pairs. This is for a good reason. The main purpose of the markers is to establish whether a pair of twins is MZ or DZ. But that is already known in the case of OS pairs! Hence the number of SS DZ pairs concordant for all markers tend to be over-estimated.
So, pace Coney and Mackey, no firm conclusion may be reached from the data in their table. With modern genetic techniques, the relative numbers of SS and OS DZ twin pairs will be established--but, as far as I know, accurate evaluation of the ratio has not yet been published.
In short, there is, as yet, no compelling evidence against Weinberg's Rule. And even if there were, that would not affect the status of the proposition that (some) sex ratio variation is facultative. The two propositions are logically independent as I noted (James, 1997a).
I should like to make some observations on whether (some) human sex ratio variation nevertheless is facultarive and, if so, whether it is controlled by the "woman as final arbiter. …