I agree with the Court that the jury’s million-dollar verdict must be set aside under the Statute of Frauds. But there is no reason to set aside the jury’s quantum meruit verdict. As the jury answered that question and no one challenges its factual or legal sufficiency, an appellate court cannot set it aside on the possibility that another jury might award a higher or lower figure. As the Court does not render judgment for the plaintiff in that amount, I dissent to that extent.

*56American law has traditionally recognized three damage measures for breach of contract: expectancy, reliance, and restitution.1 Expectancy damages award the benefit of a plaintiffs bargain;2 reliance damages compensate for the plaintiffs out-of-pocket expenditures;3 restitution damages restore what the plaintiff has conferred on the defendant.4

But fraud cases uniformly list only two damage measures: expectancy and reb-anee.5' Restitution is generally not listed, perhaps because it as an equitable rather than a legal remedy,6 or because it is available even without a showing of fraud. For example, a party may recover in quantum meruit when valuable services are furnished and knowingly accepted by a defendant in circumstances giving reasonable notice that the plaintiff expected to be paid.7 Thus, defendants who promise to pay for such services with no intent of doing so would be liable for fraud, but they would be liable even if they said nothing. In such circumstances, there is no reason a jury could not award restitution damages for fraud.8

In this case, the court of appeals affirmed a judgment for $1 million in expectancy damages because the defendant, Quigley, did not object to the charge. But the charge on fraud damages instructed jurors to consider only one element: “The reasonable value of Robert Bennett’s com-pensable work at the time and place it was *57performed.” This is a restitution measure, not expectancy, and in fact was identical to the instruction in the charge for quantum meruit damages. Because restitution can be recovered in fraud cases, the problem was not with the charge but with the evidence Bennett tried to squeeze into it. There was no evidence that generating geologists are paid $1 million in cash for their services; the evidence showed they are paid royalty interests, which are sometimes worth $1 million and sometimes worth nothing. As the Statute of Frauds prevents enforcement of oral contracts for royalty interests, it likewise prevents an action for damages measured by that amount.9 So I agree with the Court that the courts below erred in entering judgment as they did.

The charge on quantum meruit damages instructed jurors to consider precisely the same element of damage, but they gave a very different answer: $2,500. Neither party objected that these answers were conflicting or requested further deliberations.10 The reason is quite clear from the record — Bennett told jurors to write millions in both blanks (for a royalty interest), and Quigley told them to write $2,500 in both blanks (for a daily fee for services), so they did some of both. While there was evidence that would have allowed jurors to write anywhere from $500 to $20,000 in the quantum meruit blank, they wrote $2,500. We cannot reverse this finding because a second jury might answer it differently. For both fraud and quantum meruit, Bennett asked only for the value of his services, and the jury found that value was $2,500.

That being the case, there is no reason to remand for further review or a new trial; we can resolve both the apparent conflict and the appeal here. The standard for doing so is quite clear: we must preserve both answers if there is any reasonable basis for doing so:

It will never be presumed that jurors intend to return conflicting answers, but the presumption is always to the contrary. Courts properly refuse to strike down answers on the ground of conflict, if there is any reasonable basis upon which they may be reconciled.11

There was no evidence regarding expenditures Bennett made from his own pocket (reliance damages), and due to the Statute of Frauds there was legally insufficient evidence regarding the bargain he hoped to strike (expectancy damages). Accordingly, the only damage evidence jurors could credit was that regarding a daily fee, and their finding in the quantum meruit award cannot be disregarded.

Quigley argues that Bennett’s quantum meruit claim is barred because he did not file suit within four years after finishing his work on this mineral prospect.12 But the parties never agreed Quigley would be paid anything until the leases were sold; it is undisputed that Quigley never asked for payment until after they were. Although Bennett’s alleged oral contract was unenforceable, it establishes that limitations did not begin to run until payment was due later. We so held more than sixty years ago in a case involving an unenforceable promise to devise property upon the prom-isor’s death:

*58But when a contract to devise property-in consideration of services is established, the obligation to pay value for the services, which the law substitutes for the unenforceable promise, is performable also at the death of the promisor; and if there has been no prior repudiation of the contract by him, the cause of action accrues and limitation begins to run at the time of the promisor’s death.13

Accordingly, I would reverse the court of appeals’ judgment based on the jury’s verdict regarding fraud damages and render judgment on their verdict regarding quantum meruit. Because the Court does only the former, I concur in part and dissent in part.

In this case we consider whether evidence of the value of a royalty interest in minerals can be considered in determining compensation for geologic services when the services were not rendered pursuant to a written agreement. We determine that it cannot.

In early 1997, Michael Quigley decided to sell his interest in oil and gas leases known as the Samano leases. Robert Bennett, a geologist, agreed to help Quigley by analyzing the leases and assisting with a sales presentation to Louis Dreyfus Natural Gas. Bennett did not expect to be paid; he agreed to do the presentation as a favor to a sick colleague who had been working with Quigley. While Bennett was preparing for the presentation, Quigley asked him to do additional work as to the leases, including preparing more color graphs and maps. Bennett testified that when he told Quigley he did not have time to do the additional work because he had his own projects to work on, Quigley told him, “Don’t worry Bennett, I’ll take care of you.” Bennett spent three or four days working on the maps and graphs and then participated in the presentation to Dreyfus in April or May of 1997. Dreyfus decided not to purchase the leases, but Quigley kept the maps Bennett prepared. Quigley continued to market his leases and in April 1998, Coastal Oil & Gas bought them. *53Quigley secured an overriding royalty interest as part of the transaction.

After the leases sold to Coastal, Bennett mentioned his compensation to Quigley but the two did not discuss the matter in depth. Around August 2001, after Coastal drilled two producing wells on the Samano leases, Quigley and Bennett met for the purpose of determining Bennett’s compensation. The issue was not resolved.

In February 2002, Bennett sued Quig-ley, asserting causes of action for quantum meruit, conversion and fraud, and seeking attorneys’ fees. The case was tried to a jury. Bennett presented evidence that he was a generating geologist, and generating geologists are usually compensated by receiving overriding royalty interests in the prospects they generate. He testified that he personally only accepted mineral prospect work for compensation of a royalty interest in the prospect. Proof was offered at trial that the value of a 1% royalty interest in the past and estimated future production of the Samano leases was approximately $4 million. Evidence was presented that geologists who worked for cash compensation rather than overriding royalty interests earned between $500 a day and $20,000 per job.

The jury found in Bennett’s favor on all three theories of liability. Damages were found in the amount of $2,500 on the quantum meruit claim,1 $1 million on the fraud claim,2 and $1 million on the conversion claim.3 In connection with the quantum meruit claim the jury found that (1) by August 2001 Bennett should have discovered that Quigley was not going to pay, and (2) Quigley fraudulently concealed from Bennett that Quigley was not going to pay, thereby causing Bennett to delay filing suit. The jury also found reasonable attorney’s fees for Bennett in the amount of $185,000 for trial and appeal. Bennett elected to recover on his fraud claim because that gave him the greatest recovery. The trial court entered judgment based on the jury findings in response to the fraud questions and answers.

Quigley appealed. He asserted that Bennett’s recovery on his fraud claim violated the statute of frauds, the parties had no “binding agreement” to support a fraudulent inducement claim, and the quantum meruit and conversion claims were barred by limitations. He also asserted legal and factual sufficiency challenges. The court of appeals, based in part on Quigley’s having told Bennett “well do us a favor, work what you can ... and I’ll pay you for your time doing it,” deter*54mined that Quigley had an agreement with Bennett for Bennett to provide geological services, affirmed the judgment on Bennett’s fraud claim, and did not address the other jury findings. The court of appeals held that the fraud damages question submitted an improper measure of damages but Quigley did not object to the submission; Quigley did not preserve error as to the proper measure of damages; and there was evidence to support the jury finding based on the charge submitted. 2005 WL 763257.

In this Court Quigley asserts, in part, that the court of appeals erred in affirming the award based on fraud.4 Referencing Haase v. Glazner, 62 S.W.3d 795, 798-99 (Tex. 2001), he argues that there was no agreement for Bennett to receive a royalty interest as compensation. Bennett does not disagree. In his brief Bennett says that “[bjecause Bennett did not claim an express agreement giving him an overriding royalty interest, Bennett sought, under quantum meruit and fraud, restitution damages based on the value of the services he performed.” But Quigley also urges that allowing Bennett to recover damages based on the value of an overriding royalty interest, under these facts, violates the statute of frauds. Bennett disagrees with the latter position; we do not.

An overriding royalty interest in an oil and gas lease is considered an interest in real estate that falls within the statute of frauds. Consol. Gas & Equip. Co. v. Thompson, 405 S.W.2d 333, 336 (Tex. 1966). Absent a writing, an agreement to transfer such an interest is unenforceable. See id. Allowing recovery of the value of a royalty interest when the interest itself could not be recovered because the statute of frauds bars recovery would circumvent protections of the statute. Haase, 62 S.W.3d at 798-99 (noting that when the bargain violates the statute of frauds and is unenforceable, allowing recovery for benefit-of-the-bargain damages would deprive the statute of any effect). Thus evidence of the value of a royalty interest, which was what Bennett contended for as compensation, cannot be given any weight or effect and legally cannot be considered as evidence supporting the jury’s finding. See id. at 799; see also Merrell Dow Pharm., Inc. v. Havner, 953 S.W.2d 706, 711 (Tex. 1997) (noting, in part, that a no evidence assertion will be sustained if the court is barred by rules of law from giving weight to the only evidence offered to prove the matter at issue).

Absent evidence of the value of a royalty interest, the only evidence of damages was testimony as to cash-based compensation for a geologist. The testimony regarding cash-based compensation is some evidence of the value of Bennett’s work. It is, however, legally insufficient to support the entire $1 million fraud damages finding and the court of appeals’ judgment must be reversed. See Formosa Plastics Corp. USA v. Presidio Eng’rs and Contractors, Inc., 960 S.W.2d 41, 51 (Tex. 1998).

As noted above, there is some evidence that Bennett suffered damages from Quig-ley’s actions which the jury found constituted fraud. Thus, we cannot render judgment for Quigley on the fraud claim.

Citing Scott v. Walker, 141 Tex. 181, 170 S.W.2d 718 (Tex.Com.App. 1943), the concurring and dissenting justices would render judgment for Bennett on the quantum meruit findings. They note that when suit was filed in February 2002 it was timely as *55a matter of law. The reasoning is that Bennett’s quantum meruit cause of action did not accrue until the leases were sold in April 1998 when the unenforceable promise to transfer a royalty interest was breached and the obligation to pay value which the law substitutes for the unenforceable promise was also breached. We disagree that Scott supports rendition for Bennett. In Scott, the Walkers lived with and rendered housekeeping and personal services to A.E. Bower over a period of years until Bower died. Id. at 720. The Walkers sought compensation for fourteen years of services rendered to Bower. They also contended that Bower agreed to compensate them, in part, by devising his house and property to them when he died. The Court noted that such an oral agreement to transfer real property would be unenforceable because it violated the statute of frauds, but that if such an agreement existed the law would substitute an obligation to pay value for the services rendered in exchange for the promise to devise property. The Court further stated that it was important that the Walkers establish an agreement by Bower to devise property because the Walkers’ cause of action for breach of the substituted obligation to pay for their services would accrue at the time the oral agreement to transfer property was breached: at Bower’s death. Otherwise, limitations would begin to run on their claims for compensation at the time the services were rendered. Id. The Court held that there was no jury finding that Bower made the claimed agreement to devise his house to the Walkers, and that the Walkers’ claims to be compensated were barred by limitations except for the final two years. Id. at 722.

In this case, neither Bennett nor Quig-ley takes the position that there was an agreement for Bennett to receive a royalty interest when the leases sold. Bennett specifically claims only compensation measured by the value of a royalty interest. The jury was not asked to determine if Bennett and Quigley agreed that Bennett would receive payment upon sale of the leases or on any other particular date. Nor was there a jury finding as to such a date. We disagree that under the record before us Scott’s holding mandates a conclusion that Bennett’s quantum meruit cause of action accrued when the leases sold.

We believe the proper course is to remand the case to the court of appeals for consideration of the issues and contentions of the parties which that court did not previously address. Accordingly, we grant the petition for review and without hearing oral argument reverse the court of appeals’ judgment. See TEX. R. APP. P. 59.1. The case is remanded to the court of appeals for further consideration in light of this opinion.

Justice BRISTER filed an opinion concurring in part and dissenting in part, in which Justice HECHT and Justice WILLETT joined.

Justice GREEN did not participate in the decision.