The great "conclusive presumption" to which the cases refer is currently found in Family Code section 7540. It was formerly set forth in section 621 of the Evidence Code, and before that in subdivision 5 of section 1962 of the Code of Civil Procedure. The language of the statute is, in its entirety: "Except as provided in Section 7541 [which provides for the use of blood tests], the child of a wife cohabiting with her husband, who is not impotent or sterile, is conclusively presumed to be a child of the marriage." As the Supreme Court remarked back in Estate of McNamara (1919) 181 Cal. 82, 91 [183 P. 552, 7 A.L.R. 313] (McNamara), a literal reading of the text would suggest that the presumption would only apply if the wife was cohabiting with the husband at the time of birth. However, originating with considered dicta in the McNamara case, the words "wife cohabiting with her husband" have been judicially construed to mean cohabiting at the time of conception, not cohabiting at the time of birth. It is too late in the day for us to consider interpreting the statute according to its otherwise plain meaning; the trial court here was therefore correct in identifying the time of conception as the benchmark for the cohabitation issue. Without mentioning any authority or elaborating on the point, the McNamara court observed that "putting upon the section the meaning it undoubtedly should have, namely that issue of a wife cohabiting with her husband at the time of conception must be indisputably presumed legitimate . . . ." ( McNamara, supra, 181 Cal. at p. 91, italics added.) The court did not say why the section "undoubtedly should" be interpreted to refer only to conception, but it is easy to guess: That interpretation is the only one which would be entirely consonant with conception being the result of licit sexual relations, even though it would mean that the conclusive presumption would not necessarily apply in the case of the proverbial shotgun wedding. Whether the 1919 high court's gloss on the text qua text (i.e., its refusal to read it literally) could ultimately withstand critical scrutiny is, at this juncture, an academic matter. As the Court of Appeal observed in City and County of San Francisco v. Strahlendorf (1992) 7 Cal. App. 4th 1911, 1914 [9 Cal. Rptr. 2d 817], the statute has been interpreted that way since McNamara and has been "universally accepted by legal commentators." (See Strahlendorf, supra, 7 Cal. App. 4th at p. 1914.) While nominally a rule of evidence, there is no doubt that the conclusive presumption is a substantive rule of law, an observation first made by the California Supreme Court in Kusior v. Silver (1960) 54 Cal. 2d 603, 619 [7 Cal. Rptr. 129, 354 P.2d 657] ("A conclusive presumption is in actuality a substantive rule of law . . ."), later reiterated by the Court of Appeal in In re Marriage of B. (1981) 124 Cal. App. 3d 524, 528 [177 Cal. Rptr. 429], and Vincent B. v. Joan R. (1981) 126 Cal. App. 3d 619, 623 [179 Cal. Rptr. 9], and then finally picked up in Justice Scalia's plurality opinion for the United States Supreme Court in Michael H. v. Gerald D. (1989) 491 U.S. 110, 119-120 [109 S. Ct. 2333, 2340, 105 L. Ed. 2d 91] (Michael H.). As the Vincent B. court said, the conclusive presumption is essentially a social policy statement made by the Legislature "that the integrity of the family unit should not be impugned." ( Vincent B., supra, 126 Cal. App. 3d at p. 623.) Sometimes, the conclusive presumption has not been applied as a matter of its inherent statutory scope. For example, in McNamara there was simply too long a period from the time when the husband and wife even might have been cohabiting to the child's birth, and so the court simply refused to "apply" the statutory presumption, because the "laws of nature" (i.e., biology) precluded it. (See McNamara, supra, 181 Cal. at p. 96 ["The courts must reason in accordance with the usual operation of the law of nature . . ."]; see also Whitney v. Whitney (1959) 169 Cal. App. 2d 209, 217 [337 P.2d 219] [conclusive presumption did not apply where period was 297 days from conception to birth].) Similarly, in Anderson v. Anderson (1931) 214 Cal. 414 [5 P.2d 881], the period of cohabitation between husband and wife to the birth of the child was too short, and so the court again refused to apply the presumption. (See id. at p. 417 ["it is presumed . . . that things have happened in the ordinary course of nature"].) Similarly, the presumption is not applied (as a matter of its scope) where there has been no substantive cohabitation. (See Steven W. v. Matthew S. (1995) 33 Cal. App. 4th 1108, 1115 [39 Cal. Rptr. 2d 535] [separated wife's stolen weekend tryst with husband not enough to constitute cohabitation for purposes of presumption]; Comino v. Kelley (1994) 25 Cal. App. 4th 678, 681 [30 Cal. Rptr. 2d 728] [marriage in name only, no sex]; City and County of San Francisco v. Strahlendorf, supra, 7 Cal. App. 4th at pp. 1914-1915 [because husband and wife not living together at time of conception, it was error to apply conclusive presumption]; See also Kusior v. Silver, supra, 54 Cal. 2d at p. 616 [cohabiting means " 'living together as husband and wife' "].) Nor is it applied if there is no valid marriage. (See Alicia R. v. Timothy M. (1994) 29 Cal. App. 4th 1232, 1238 [34 Cal. Rptr. 2d 868] ["While the state has a legitimate interest in promoting marriage and not impugning a family unit, that interest cannot be served here where there is no marital union or family unit to disrupt"].)