Defendant, Olaf F. Aldren, appeals from a judgment for $8,000 entered against him upon a verdict by a jury in an action arising from his sale of some dairy cattle to the plaintiff, Carl V. Olson.
*295Briefly, the evidence in support of the verdict shows that on or about February 20, 1963, Aldren, a dairy farmer for more than 35 years, contacted Olson on a proposed sale of a part of his dairy herd. Olson had been engaged in dairy farming with his father since 1956, but had started in business for himself on a nearby farm in November, 1962, taking with him about 40 head of grade A dairy cows from their jointly owned herd. Negotiations resulted in Olson purchasing from Aldren 44 top quality grade A dairy cows at a price of $300 per head. During the negotiations Aldren advised Olson that he had not had any positive milk ring tests for five or six years and no reactors for about 14 or 15 years and guaranteed that the cattle purchased were free of Bang's disease. Aldren trucked the cattle to Olson's farm with the last delivery made on March 1, 1963.
The Aldren and Olson herds had been regularly tested for brucellosis or Bang's disease by personnel of the Animal Disease Eradication Division of the United States Department of Agriculture for several years before the sale date. Official records of the department showed that milk ring tests of the Olson herd made on July 1 and December 16, 1960, August 14 and September 25, 1961, January 8, May 28 and June 4, 1962, and February 8, 1963, each tested negative. Tests of the Aldren herd made on those same dates were positive. Aldren was advised of such results shortly after the tests were made.
Expert testimony established that a milk ring test is a screening test made on a composite sample of milk taken from the dairy herd. It is considered a highly reliable test and if positive, i. e., if brucella bacteria are found in the milk, there is a suspicion that some of the herd from which the milk was taken is infected with Bang's disease or as it is sometimes termed "contagious abortion." To definitely ascertain which cows, if any, are infected, blood tests should be taken of the entire herd. Cows found to have the disease, generally termed reactors, should be sold as beef cattle. To combat the disease, periodic retesting is recommended until all reactors have been disposed of and the herd has been culled of diseased animals.
*296On September 3, 1963, a separate composite milk sample was routinely picked up from the Olson and Aldren farms as well as others in that area. Olson's specimen for the first time in years revealed a positive milk ring test and Aldren's for the first time in about three years was negative. On September 14, 1963, Olson caused a blood test to be taken of his entire herd. It showed eight reactors, all cows purchased from Aldren, and one suspect. Blood tests were also made of Olson's father's herd and were negative.
Olson's claim for damages included losses on cattle he was required to sell in cleaning up his herd; loss of calves by abortion; veterinarian expense; and loss of profits. Such assignments of error as merit discussion follow.
Alleged error in submitting to jury question of partial release of liability
Upon discovering that his herd was infected with Bang's disease, Olson notified Aldren and they met at a sales ring in Rapid City on September 17, 1963. The eight reactors were sold on that date for a net price of $1114.66. The suspect was sold on October 14, 1963 for $122.87. The problem was discussed and it is undisputed that subsequently Aldren delivered to Olson three heifers. Aldren contended this was a final settlement of liability on losses sustained prior to those dates and Olson said it was only a partial payment to apply on his total damage when the herd was finally cleaned up. We have reviewed the testimony supporting the respective contentions and conclude the trial court properly submitted the matter to the jury for resolution. The record including instructions lacks clarity on the value of the three heifers, their ultimate disposition, and the credit to be allowed therefor on total damage, if the jury found it to be a partial payment.
*297Alleged error in receipt in evidence of sales of cows without competent proof they were infected with brucellosis
Eldon Odegard, an employee of the federal government, made blood tests of Olson's herd on March 31, 1964 and May 25, 1964. Odegard submitted written reports of such tests to Olson. The trial court refused to admit such reports under SDCL 1967 19-7-11, The Uniform Business Records as Evidence Act, and we believe properly so. They did not qualify as business records of Olson. Pruett v. Burr, 118 Cal.App.2d 188, 257 P.2d 690. Odegard was not called as a witness and the reports- were not otherwise identified by the custodian thereof or any other qualified witness. See Bentz v. Cimarron Ins. Co., 79 S.D. 510, 518, 114 N.W.2d 96, 99, for foundation necessary to admit business records. Based on such tests and reports Olson was permitted to testify that between April 3, 1964 and June 3, 1964, he sold 14 head of infected cows for a total price of $2018.16 and sales slips showing such sales were received in evidence.
Essential to Olson's claim for damages sustained by reason of such sales was competent proof that the cows sold were in fact infected. Olson was not qualified to establish such fact from his own knowledge, but had to rely on the report of Odegard. The report was obviously hearsay and since it did not qualify as a business record of Olson, his oral testimony to prove the same fact contained in the report should not have been allowed.
The trial court appears to have permitted such testimony on the assumption that the report could be used to refresh Olson's recollection. A memorandum made by another person may be used to refresh the recollection of a witness if, after inspecting it, he can testify from his own recollection, or remembers having seen it when his memory of the facts was still fresh, and recollects that he then knew statements therein to be correct. Brown v. Smith, 24 S.D. 231, 123 N.W. 689. Before a witness may refresh his recollection from a document it is necessary that both the use of the writing and the testimony given are relevant and the witness is competent to testify to the matters therein *298contained. See Annot., 82 A.L.R.2d 485, 500. The record does not show that Olson was competent to testify from his own knowledge that the cows were infected.
Without the foundational testimony of Odegard, or other competent evidence that the cows sold were infected, it is our opinion that the trial court erred in receiving into evidence the sales slips showing the sale of the 14 head claimed to have been infected as shown in said reports.
Damages for loss of profits
Olson made his first sale of the infected cattle on September 17, 1963. Thereafter, the herd was quarantined from May 25, 1964 to October 8, 1964, when the herd again tested clean. Olson contends that by reason of his forced sale of cows to clean his herd, his milk production decreased since it was not feasible to replace the cows until the whole herd was cleared. This could not be done until the quarantine was lifted. As a result he contends he sustained a loss of profits.
The trial court submitted the case to the jury on both express and implied warranty, as well as on fraud and deceit. Aldren does not challenge such submission on this appeal. In our opinion damages for loss of profits are sustainable under either theory if properly proved. See Annot., 87 A.L.R.2d 1317, 1379.
Following the South Dakota statute, SDC 54.0169(6) (7),1 the court instructed in part:
"the measure of damages for breach of warranty is the loss directly and naturally resulting, in the ordinary course of events, from the breach of warranty. In the case of breach of warranty of quality, such loss, in the absence of special circumstances showing proximate damages of a greater amount, is the difference between *299the value of the goods at the time of delivery to the buyer and the value they would have had if they had answered the warranty. However, where special circumstances are proven showing proximate damages of a greater amount, all damages which naturally may result from a breach of the warranty, accrue in favor of the party injured by such warranty if they are certain and determinate in nature or amount and are also directly attributable to the breach of the contract as their cause."
We believe the evidence supports giving this instruction.
The court also instructed under SDCL 1967 21-3-1 on the measure of damages in case liability was predicated on fraud and deceit and permitted assessment of exemplary damages on such theory. By special verdict the jury allowed no exemplary damages.
Although sometimes difficult to prove, the generally accepted rule is that, where it is shown that a loss of profits is the natural and probable consequence of the act or omission complained of, and their amount is shown with reasonable or sufficient certainty, there may be a recovery therefor. 25 C.J.S. Damages § 42. However, such damages must not be speculative, contingent, or uncertain and there must be reasonable proof of the amount thereof. Any reasonable method of estimating a prospective profit is acceptable. 25 C.J.S. Damages § 90. Absolute certainty is not required.
Olson attempted to show loss of profits by offering in evidence certain monthly production records he received from the dairy cooperative where he, and he and his father, had marketed milk from 1957 through 1964. These records showed the total pounds of milk marketed each month, the test, base price, surplus price, and Olson's proportionate share of association expense in connection with the marketing. It does not purport to show the number of cows milked during any period or the expense of production. Variations in production are not consistent. In some months after reactors were sold the amount of *300milk sold exceeded what had been produced and sold when the reactors were being milked. The trial court did not receive this evidence holding it too speculative. We agree.
Olson then attempted to testify that he remembered approximately what each cow was producing when sold although he kept no written records. He also said he could fix with reasonable certainty the cost and expense of feeding the cow and the cost of producing the milk so as to arrive at the net loss on each cow from the time of sale to when the quarantine was lifted. The court did not permit this evidence because of the absence of written production records on individual cows.
Although we feel the monthly production records of the marketing cooperative were inadmissible to prove loss of profits for reasons stated supra, we do believe Olson as an experienced dairyman may have been qualified to testify on estimated production and estimated expense of the dairy cows he was forced to sell, so that on a per day, week, or month basis, he could estimate with reasonable certainty the loss of profit from milk production on the cows sold. Essentially this is what the court permitted in Tuttle v. Bootes Hatcheries & Packing Co., D.C., 112 F.Supp. 705, in a case involving a loss of turkeys. We see no cogent reason why it cannot also be done in this case by adopting reasonable accounting procedures on estimated receipts and expenses. The fact that production records were not kept on each cow does not in itself disqualify Olson from fixing his loss of profit. Cases from which it appears damages for loss of profits were allowed in the absence of written records are Dillard v. Clutter, Tex.Civ.App., 145 S.W.2d 632; Hawkins v. Jack son, 97 Ga.App. 525, 103 S.E.2d 634; Lanesboro Produce & Hatchery Co. v. Forthun, 218 Minn. 377, 16 N.W.2d 326 and Frame v. Hohrman, 229 Minn. 468, 39 N.W.2d 881.
The trial court received as evidence Olson's federal income tax returns for the years 1962, 1963 and 1964. Attached to each such return was a Schedule F, Form 1040, Farm Income and Expense, showing gross income from dairy products and from some other sources, farm expenses, depreciation, and other information usually contained in income tax returns of *301farmers. The returns do show some decline in dairy product gross income in 1964 from 1963, but in our opinion the many variables reflected in income tax returns of farmers from year to year make such evidence highly speculative and insufficient to sustain an award for loss of profits. Differences in weather, feed supply, costs of operation, and prices, in general, are such as to give little probative value to an income tax return for the purpose of showing a loss of profit. In our opinion they should not have been received in evidence in this case.
In Bruha v. Bochek, 76 S.D. 131, 74 N.W.2d 313, an owner was allowed damages for loss of four acres of oats upon proof that the remainder of the field yielded about 30 bushels per acre and the market price was 59 cents per bushel. This court held that in the absence of evidence of the fair and reasonable cost of tilling, harvesting, and marketing, the court was without sufficient data from which it could determine the actual damage.
In our opinion in the case at bar, the jury did not have before it sufficient competent evidence from which it could determine damages for loss of profits.
Reversed.
BIEGELMEIER, P. J„ and RENTTO, J„ concur. ROBERTS, J., concurs in result. HANSON, I., dissents.. This section was repealed by Chapter 150, Session Laws of 1966, when South Dakota adoptéd the Uniform Commercial Code. See SDCL 1967, Section 57-8-36 et seq.