EDITH H. JONES, Circuit Judge:

From an adverse summary judgment in this products liability case based upon a mother's ingestion of Bendectin during pregnancy, the plaintiffs appeal. We hold that the district court properly granted summary judgment by excluding an untimely-filed controverting affidavit. 723 F. Supp. 1188.

BACKGROUND

The Bernhardts' case was filed in January, 1985 and promptly removed to federal court in Mississippi. Plaintiffs asserted that Mrs. Bernhardt's use of Bendectin to control nausea during her pregnancy caused her daughter to be born without fingers on her left hand. In July, 1986, Merrell Dow Pharmaceuticals, Inc. ("Merrell Dow"), the corporate successor to Richardson-Merrell, Inc., moved for summary judgment based upon plaintiffs' inability to prove two essential elements of their case: (1) that Bendectin, in fact, caused the birth defects of Dana Bernhardt; and (2) that the alleged inadequacies of Bendectin's labelling proximately caused the injuries.

Merrell Dow contended that the plaintiffs could not prove causation because of the "timing of ingestion" of Bendectin by Mrs. Bernhardt. Answers to interrogatories and deposition testimony established that Mrs. Bernhardt first ingested Bendectin at day fifty-four or fifty-five post-conception. Merrell Dow submitted the affidavit of its medical expert, Dr. Jackson, who opined that this was too late in pregnancy to have caused the specific birth defects of Dana Bernhardt. He stated that, in the human fetus, the hand plate is formed by the thirty-eighth day following conception, and the finger rays have formed and the hand has begun to split into fingers by day forty-four. Consequently, a drug exposure after that point in pregnancy could not produce a child with congenital absence of the fingers. Dr. Jackson thus concluded that Mrs. Bernhardt's ingestion of Bendectin occurred too late in pregnancy to have caused Dana's hand deformities.

About a month later, the Bernhardts filed a memorandum in opposition to Merrell Dow's motion. They submitted no medical or other expert proof that Bendectin was taken by Mrs. Bernhardt at a time when it could have caused the absence of fingers on Dana's left hand. The Bernhardts did state that their counsel had been involved in trial in another case with the defendant and that "should the court desire," counter-affidavits of plaintiffs' experts would be submitted. They did not, however, move for an extension of time for further discovery or to supply affidavits. See Fed. Rule Civ.Proc. 56(f).

In May, 1988, the trial court denied Merrell Dow's motion for summary judgment, holding that Dr. Jackson's affidavit "assumed a genuine issue of material fact," i.e., when the mother's last menstrual cycle and the date of conception occurred. Merrell Dow moved for reconsideration of its motion, citing plaintiffs' interrogatory answers and deposition testimony which unequivocally established these dates. The Bernhardts' opposition to the request for reconsideration contained no admissible summary judgment evidence. Rather, it referred to testimony of a person identified as a Merrell Dow expert witness "in a prior case just a few months ago," which it says was at odds with the affidavit of Dr. Jackson. This averment may have reflected common knowledge to counsel who were involved in representing multiple plaintiffs or defendants in Bendectin litigation, but it certainly did not qualify as controverting summary judgment evidence.

Merrell Dow's motion for reconsideration evidently grabbed the attention of the district court, and his law clerk wrote to plaintiffs' counsel on November 1, 1988 with the following inquiry:

Recently the court reviewed the Defendant's motion to reconsider its denial of the defendant's motion for summary judgment. In considering this motion, the court wondered what expert testimony the plaintiff had to rebutt [sic] the defendant's contention that the fingers of a fetus were developed by the forty-fourth day after conception. Please point out in writing what expert testimony you have submitted to rebutt this conclusion.

(emphasis added). Six weeks later, in response to this letter, plaintiffs submitted an affidavit and curriculum vitae of Ph.D. Stewart Newman, which alleged that Mrs. Bernhardt ingested Bendectin at a time when it could have affected the development of Dana's fingers. Plaintiffs also submitted the previously-mentioned excerpt from the testimony of a Merrell Dow expert in another case, in which the expert agreed that limb development in an individual fetus may vary by as much as ten days from the average date following conception.

In April, 1989, the court granted Merrell Dow's motion for reconsideration and entered judgment for the defendant. The court refused to consider Dr. Newman's affidavit, finding that it was untimely filed. The court specifically held that, in the law clerk's letter of — November 1, there had been no request for additional evidence. The court had inquired only as to the then-existing record. The court noted plaintiffs' failure to comply with Rule 56(f) in order to supplement the summary judgment evidence. Reviewing the plaintiffs' references to their expert testimony in interrogatory answers furnished to Merrell Dow, the court found no genuine issues of material fact concerning (1) whether Bendectin is a teratogen in humans and (2) whether, even if the drug is a teratogen, it could have caused Dana's specific defect.

DISCUSSION

On appeal, the Bernhardts vigorously contest the interpretation of the November 1 letter, the exclusion of Dr. Newman's affidavit, the propriety of granting summary judgment even without the Newman affidavit, and whether the court based its ruling on a theory it developed sua sponte without giving plaintiffs a chance to argue or brief it. Our disposition requires a ruling only on the first three of these contentions.

Responding to plaintiffs' first contention, we hold that the district court did not abuse its discretion in excluding Dr. Newman's affidavit as untimely filed. The court's November 1, 1988 letter plainly requests the Bernhardts to point out what evidence previously submitted could refute Merrell Dow's contention that the Bendectin was taken "too late" to have caused Dana's specific birth defects. As the court later reiterated, this letter was not an invitation to submit further controverting summary judgment evidence. Thus, the Bernhardts took it upon themselves to attempt to supplement the summary judgment record nearly two and one-half years after Merrell Dow's motion was originally filed. The Bernhardts have never sought to excuse their failure to file at an earlier point in time an affidavit controverting that of Dr. Jackson. At no time did the Bernhardts seek a continuance to obtain evidence pursuant to Rule 56(f).

The express terms of the Federal Rules condemn the Bernhardts' late affidavit. Rule 56(c) permits the adverse party "prior to the day of hearing" to serve opposing affidavits. In this case, the parties had fully briefed the issues for summary judgment by August, 1986 and the court had issued its ruling by May, 1988. The Bernhardts' attempt to submit further evidence, certainly after May, 1988, without obtaining leave of court, contravened Rule 56(c). Rule 56(f) authorizes the opponent of a summary judgment motion to seek a continuance to obtain controverting affidavits or necessary discovery. This provision would become useless, and the summary judgment procedure uncontrolled, if a court could not enforce some limits on the timely submission of appropriate evidence. Similarly, when the Bernhardts finally tried to submit Dr. Newman's affidavit, they failed to move for an enlargement of time which, under these circumstances, could only be based upon a showing of excusable neglect. Fed.R.Civ.Proc. 6(b). This case is virtually analogous to our decision in Farina v. Mission Investment Trust, 615 F.2d 1068, 1075-76 (5th Cir. 1980), in which we determined that a trial court did not abuse its discretion in refusing to allow a controverting affidavit to be filed after the hearing on summary judgment. We conclude that the district court did not abuse its discretion in refusing to consider Dr. Newman's affidavit.

Accordingly, we analyze the defendant's motion for summary judgment based on the evidence exclusive of the untimely affidavit. Our task is to determine whether any genuine issue of material fact, requiring a trial, is raised in the evidence presented on the motion. Fed.R.Civ.Proc. 56(c). Further, "[w]hen the moving party has carried its burden under Rule 56(c), its opponent must do more than simply show that there is some metaphysical doubt as to the material facts. . . . In the language of the Rule, the non-moving party must come forward with 'specific facts showing that there is a genuine issue for trial.'" Matsushita Electric Industrial Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 586-87, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986) (citations omitted, emphasis in original). We may affirm a grant of summary judgment on grounds other than those stated by the trial court. Sapp v. Renfroe, 511 F.2d 172, 175 n. 2 (5th Cir. 1975).

So viewed, there is no genuine issue of material fact. Our decision is somewhat narrower than that of the district court, however, for we rely upon the undisputed affidavit of Dr. Jackson. We do not broach the broader questions whether Bendectin is a human teratogen, although if we did so, we would be bound by our court's prior holding in Brock v. Merrell Dow Pharmaceuticals, Inc., 874 F.2d 307 (5th Cir. 1989). Dr. Jackson's affidavit stated that the fingers of the human fetus are present by the forty-fourth day after conception. Consequently, because Mrs. Bernhardt did not take Bendectin until day fifty-four or fifty-five, her first ingestion of Bendectin occurred too late to have caused the absence of Dana's fingers.

The Bernhardts' timely response to Merrell Dow's motion relied upon three pieces of evidence. On examination, none of these actually contradict Dr. Jackson's affidavit. First, plaintiffs provided an answer of Merrell Dow to interrogatories posed in another case, which admitted that until fifty-six days after conception, the upper limbs of a fetus continue to develop. Utilizing a standard textbook by Moore, Developing Human, 4th Ed. W.B. Sanders Co. 1988, plaintiffs quoted a statement that until fifty-six days after conception, the upper limbs including the fingers are still developing. Finally, the Bernhardts cited previous testimony of an expert hired by Merrell Dow in another case, which stated that the speed at which a fetus develops varies and the critical period for development of the limbs may vary up to ten days from the average. The Bernhardts contend that because Dana's fingers may have been developing until the fifty-sixth day after conception, her mother's ingestion of Bendectin on the fifty-fourth day could have caused her to lack fingers.

This conclusion leaps too easily from the very general premises stated by the plaintiffs. To say, for instance, that the upper limbs are still developing until the fifty-sixth day after conception does not assist plaintiffs' case unless it is further shown that the fingers materialize during the last two days of that period. In other words, such evidence does not suggest with any specificity when the fingers would normally begin to appear, which would be the critical point for determining when a teratogen could have arrested their development. Thus, these statements may be generally true, but they do not contradict Dr. Jackson's more specific testimony about the development of the fetus's hand plate and finger rays. It is not enough for plaintiffs to contend that in a case requiring expert testimony, any reference to the work of an expert will suffice to create a genuine issue of material fact. The court has a duty to scrutinize the probative value of the evidence. See Viterbo v. Dow Chemical Co., 826 F.2d 420 (5th Cir. 1987). Here, the evidence adduced by the Bernhardts simply hits the wrong target. It fails to controvert Dr. Jackson's affidavit and therefore did not create a genuinely disputed fact issue on the timing of Mrs. Bernhardt's ingestion of Bendectin.

The Bernhardts further contend, in rather vague language, that the trial court erred by employing a federal rather than a Mississippi law standard of causation. The trial court, like our court, was ruling on the sufficiency of plaintiffs' evidence in the face of a summary judgment motion. As we have routinely held, this question raises federal evidentiary concerns, rendering resort to state law improper. See, e.g., Thompson v. Southern Pacific Transp. Co., 809 F.2d 1167, 1168 (5th Cir. 1987).

For the foregoing reasons, the judgment of the trial court is AFFIRMED.