In this suit the Administrator seeks to recover damages for a violation of the *708Price Control Act. 50 U.S.C.A.Appendix, § 901 et seq. The question for decision is whether the trial court erred in directing a verdict for the defendants.

The Administrator claimed that the latter sold a quantity of beans, comprising 3,099 sacks of 100 pounds each, at a price 10 cents per cwt in excess of the ceiling fixed by regulation. The fact of the sale was stipulated, and it was agreed, also, that the commodity sold had been stored as Lot No. 417 in the warehouse of the Southern Pacific Milling Company at Greenfield, California, to be cleaned for the owners. The dispute concerned the grade of the beans, whether, as contended by the Administrator, they were U. S. No. 1 small white beans, or whether, as insisted by appellees, they were U. S. choice handpicked. If of the latter grade the selling price was not above ceiling.

To prove the grade the Administrator offered in evidence an inspection certificate issued by an official of the United States Department of Agriculture. This certificate identified the beans as Lot No. 417 stored in the warehouse named above, and under the head of “remarks” stated that 1,000 bags only were exposed for sampling. The grade and class were-certified to be U. S. No. 1 small white beans. The certificate, made out apparently in the customary form, states that it is issued in compliance with the regulations of the Secretary of Agriculture governing the inspection of beans and peas. The court sustained an objection to its admission.

A statute of the United States (58 Stat. 738, 5 U.S.C.A. § 575, 1945 Cumulative Pocket Part) provides: “Market-inspection certificates issued by authorized agents of the Department of Agriculture shall be received in all courts of the United States as'prima facie evidence of the truth of the statements therein contained.” Tris statute was not called to the attention of the trial judge.

Confronted with the adverse ruling the Administrator introduced testimony to the effect that a licensed sampler of the Department, Richardson, was directed by his superior to draw a sample of the beans. The assistant manager of the Milling Company testified that Richardson called at the warehouse for the purpose of sampling the lot named, was shown where the beans were stored, and presently returned to the witness’ office “with sample bag of beans, the customary amount they bring in.” The sampler’s report, introduced as an exhibit, identified the beans as drawn from the Lot No. 417. The inspector who issued the certificate testified that the sample bag received from Richardson contained between 4 and 5 pounds, and from this he determined the grade.

There was other testimony, not so favorable to the Administrator. Manager Henry of the Milling Company testified that prior to Richardson’s visit he, Henry took samples from Lot No. 417 for the prospective purchaser; that he understood samples were supplied by the purchaser to the office of the Department of Agriculture at Los Angeles, where a grading certificate was issued of which the witness later saw a copy; and in response to a query interpolated by appellees’ counsel the witness testified, over objection, that the certificate classified the beans as “choice handpicked.” • The certificate was not produced.- Henry said that after Richardson’s visit he examined the Lot, and, aside from the bags he himself had sampled, saw not more than 40 which bore evidence of having been punctured with a sampling tool. His testimony was not, however, positive in respect of the extent of Richardson’s sampling and was prefaced by the remark that he did not know whether Richardson “took the samples out of all these beans.”

At the close of the Administrator’s case the court, having first cleared the atmosphere by admitting the proffered certificate for what it might be worth, granted appel-lees’ motion for a directed verdict. The judge fortified his ruling by observing that there was no direct evidence as to the taking of the sample by Richardson, and he thought it had not 'been shown that the beans sampled were from those sold. On the ground stated the ruling was clearly wrong. But appellees insist that the regulations of the Secretary of Agriculture required a sampling of at least 10% of the bags; and they say the testimony showed conclusively the taking of an insufficient *709sample. In sum, the argument is that the prima facie case made out by the certificate amounts to no more than a disappearing presumption which here was entirely dissipated.

For purposes of the decision we assume the validity of the doubtful premise that the statute imparts to the certificate merely the force of a disappearing presumption. But the balance of the argument we are unable to accept. Preliminarily, a study of the regulations of the Secretary fails to bear out the claim that a sampling of a definite percentage of the bags was requisite to a valid sample. The regulations themselves1 seem not to cover the point. Those bearing upon it are §§ 58.13 and 58.46.2 Under the latter, samples are to be drawn “in accordance with methods approved by the chief of the service.” In this connection appellees have supplied us with a set of instructions issued by the Agricultural Marketing Service of the Department under date of August 15, 1933. Presumably these state the approved methods of sampling. Paragraph 3 provides: “At least 10 percent of the bags in a carload or other lot of beans and peas should be probed. In sampling small lots, from 15 to 25 percent of the bags should be probed. In any case a sufficient number of the bags should be probed to satisfy the sampler that a representative sample has been obtained. * * * ” This instruction is not couched in mandatory terms. We judge that a measure of discretion was intended to be lodged with the sampler, who is to satisfy himself in all cases that he has taken a representative sample. The verbiage of the instruction takes on added significance when contrasted with the peremptory language of other instructions in the group.3

More to the point, however, and less open to legitimate difference of opinion is the inconclusive nature of the proof relied on as an impeachment of Richardson’s sampling. The presumption of regularity attendant upon the performance of official duty cannot fairly be said, as a matter oí law, to have been dissipated by the oral evidence on the point. Nor did the unproduced grading certificate mentioned by the witness Henry detract from the prima facie showing. His testimony as to the contents of the certificate ought to have been excluded as incompetent. We think, therefore, that the question whether there had been a representative sampling was for the jury.

A minor point remains to be noticed. It is claimed that appellees, who were growers of the beans, were not processors within the meaning of the applicable regulation, MPR No. 270, the second section of which provides: “ ‘Processor’ means a person, other than a dealer, who owns the kind and variety (or class) of listed commodity being priced and processes it or causes it to be processed.” Appellees clearly fall within this definition. They owned the beans and delivered them to the Milling Company, which, pursuant to their instructions, cleaned them. The term “process” is defined in the regulation as meaning, among other things, “to clean or otherwise prepare the listed commodity for shipment.” Appellees’ point is that while they caused the beans to be processed in this manner, the processing was not for shipment but *710merely preliminary to sale, and that the buyer did the shipping. This seems too narrow an interpretation. The regulation does not concern itself with the identity of the shipper; and in this instance ultimate shipment was contemplated when the beans' were placed in the warehouse to be processed and sold.

Since appellees were processors as-well as growers they are not within the exemption of “Sales by growers to processors or dealers” of Section 1, subdivision (1) of MPR270.

Reversed,