On Appellee’s Petition for Rehearing
PER CURIAM:From the Government’s petition for rehearing in this case, it appears that the following clarifying observations are in order.
I
The Government has correctly inferred from our opinion that a showing that “auditing services have been peddled to the general public on the basis of wholly non-religious pseudo-scientific representations” would support a verdict of false labeling. 133 U.S.App.D.C. at-, 409 F.2d at 1161. We gave as further explication of this view a *1164passage from Weiss, Privilege, Posture and Protection: “Religion” in the Law, 73 Yale L.J. 593, 604, 605 (1964), and citations to two instructive New York cases, People v. Cole, 219 N.Y. 98, 113 N.E. 790, L.R.A.1917C, 816 (1916), and People v. Vogelgesang, 221 N.Y. 290, 116 N.E. 977 (1917). Our basic point is that, in order to raise a religious defense to a charge of false statement (here misbranding), the person charged with the alleged misrepresentation must have explicitly held himself out as making religious, as opposed to medical, scientific or otherwise secus, lar, claims.
The Government now argues that there was sufficient evidence in the record to permit the jury to find false labeling on the basis of “wholly non-religious pseudo-scientific representations.” In the thousands of pages of Scientology literature introduced at trial, it finds passages which appear to be based on secular rather than religious claims, such as the claim that Scientology is “a precise and exact science, designed for an age of exact sciences,” and that “[n]o other subject on earth except physics and chemistry has had such grueling testing (proofs, exact findings).”1
This argument misconceives the ground upon which we reversed. We did not find insufficient competent evidence to support a verdict, noy did we find that all literature submitted to the jury as “false labeling” was religious doctrine. Rather we found that some of that literature was at least prima J facie religious doctrine, and that the jury, as it was instructed,2 could have found against the E-meter by finding false statements in “labeling” which was at the same time religious doctrine. See main opinion, 133 U.S.App.D.C. at -, 409 F.2d at 1161. And, of course, where a jury’s general verdict may have rested upon grounds improper for First Amendment reasons, a reviewing court will not pause to speculate whether the jury’s verdict was actually, reached on other, and permissible, grounds. Stromberg v. California, 283 U.S. 359, 367-368, 51 S.Ct. 532, 75 L.Ed. 1117 (1931); Gregory v. City of Chicago, 394 U.S. 111, 89 S.Ct. 946, 22 L.Ed.2d 134 (1969).
II
In its petition, the Government finds unclarity in our failure to state whether or not a second trial may follow in this case. Of course, it is not within our power authoritatively to declare the res judicata effect of our own decision. However, since the basis of our reversal was that the case was improperly framed, rather than that the Scientologists’ devices and literature were absolutely pro*1165tected, or that the evidence was insufficient, it would appear that a new trial would be in order.
If a new trial does follow, a further observation may be helpful. We have held that as a matter of statutory construction compelled by the constitutional doctrine of United States v. Ballard, 322 U.S. 78, 64 S.Ct. 882, 88 L.Ed. 1148 (1944), religious claims cannot be found “false labeling” within the meaning of the Food, Drug and Cosmetic Act. Thus it is incumbent on the trial judge to rule in the first instance whether each item of alleged false labeling makes religious claims and hence cannot be submitted to the jury for the factual determinations of whether it is a label for the device in question and whether it is false.3 See Jacobellis v. Ohio, 378 U.S. 184, 187-188, 84 S.Ct. 1676, 12 L.Ed.2d 793 (1964).
Finally, it should be noted that the Government up to this time, including its motion for rehearing, has not challenged the bona fides of appellants’ claim of religion. In the event of any new trial, as indicated in the panel opinion, it would be open to the Government to make this challenge. If the challenge is made successfully, the First Amendment question would, of course, disappear from this case. See United States v. Kuch, D. D.C., 288 F.Supp. 439 (1968).
Circuit Judge McGOWAN does not join in this clarification of the majority opinion, and continues to adhere to his dissenting opinion.. In our main opinion, we have already noted that some claims made on behalf of the E-meter and the auditing process, and introduced into evidence at trial, had no discernible relation to whatever religious content Scientology might have. See 133 U.S.App.D.C. at —-, 409 F.2d at 1158, n. 39 and n. 40.
. A tremendous volume of Scientology literature was submitted to the jury, and the jury was charged that if the literature “accompanied” the E-meter in the sense described in the main opinion, 133 U.S. App.D.C. at-, 409 F.2d at 1157 it was labeling. Thus there was neither a winnowing out by the judge of the religious material, nor an instruction to the jury that it could not find such material to be false labeling.
It is true that no such sifting process and no such instruction was requested by appellants. Appellants framed their First Amendment point in more broadside fashion as described in the main opinion, 133 U.S.App.D.C. at -, 409 F.2d at 1154. However, even though the narrower Ballard point was never raised in the form of objections to evidence or suggested jury instructions, we regard it as subsumed in the broader free exercise objections actually made. In any case, a denial of First Amendment rights is one of those exceptional instances where an appellate court will notice error in the charge even where no objection is made at trial. 5 J. Moore, Federal Practice If 51.04 (2d ed. 1968) ; Shokuwan Shimabukuro v. Higeyoshi Nagayama, 78 U.S.App.D.C. 271, 273, 140 F.2d 13, 15, cert. denied, 322 U.S. 755, 64 S.Ct. 1270, 88 L.Ed. 1584 (1944).
. With respect to this item-by-item determination, the inquiry should be whether that item puts forward its allegedly false claims respecting the E-meter or auditing on a “wholly non-religious” basis. Of course, this does not preclude a finding that an item (book, pamphlet, advertising flier) makes out a self-sufficient non-religious claim for Scientology services, to which a religious appeal has been merely tacked on.