This appeal is from a grant of summary judgment to defendants in an action in which plaintiffs asserted that defendants had violated the securities laws and civil RICO. We recognize there are legal actions that serve to punish or deter defendants or to determine property rights. But the signature of the causes of action before us is harm to plaintiffs: actual damages or consideration paid in the securities laws actions, and injury to plaintiffs’ business or property in the RICO cause of action. Here plaintiffs, who are investors, may have feared they would suffer harm, but they actually suffered no out-of-pocket loss since their investments were fully repaid, plus interest. Thus, while we have no difficulty in finding that plaintiffs’ causes of action might well subject defendants' to liability, plaintiffs can prove no damages.
Plaintiffs-appellants are Commercial Union Assurance Co. pic, Commercial Union Pen*610sions Management, Ltd., Overbrook Nominees, Ltd., Strand Nominees, Ltd., Gemini Overseas Corporation, and Mercia Zurich, AG. (collectively appellants). They appeal from an order entered on May 5, 1993 and final judgment entered on May 6, 1993, by the United States District Court for the Southern District of New York (Pollack, J.), granting defendants Michael Milken and Lowell Milken’s motion for summary judgment on appellants’ claims under § 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b) (1988), and Rule 10b-5 promulgated thereunder, 17 C.F.R. § 240.10b-5 (1993); § 12(2) of the Securities Act of 1933 as amended, 15 U.S.C. § 77Z(2) (1988); and the Racketeer Influenced and Corrupt Organizations Act (RICO), 18 U.S.C. §§ 1961, 1962 and 1964 (1988 & Supp. II 1990). Athough we affirm the district court’s grant of summary judgment tó defendants, our rationale for doing so is different.
BACKGROUND
We review the background briefly. In March 1986 Ivan F. Boesky, the now notorious takeover-stock speculator, reorganized the form of his risk-arbitrage business from a corporation to a limited partnership. The new partnership, known as Ivan F. Boesky & Co., L.P. (the name has subsequently been changed to CX Partners, L.P.), was financed by a large debt offering and the private placement of $250 million in limited partnership interests. Because of the highly volatile nature of the risk-arbitrage business, the prospectus for limited partners cautioned prospective purchasers that the investment entailed a high degree of risk. Potential investors were required to submit investor suitability questionnaires prior to purchasing their interests. The questionnaire inquired as to whether the investor understood the nature and risks associated with the investment, explained that there was no guarantee of any financial return and that a limited partnership was not a liquid asset and should not be relied upon for current needs or personal contingencies.
Working closely with Boesky on the reorganization was Drexel Burnham Lambert Incorporated (Drexel), and appellee Michael Milken. The now-defunct Drexel was at the time the preeminent underwriter of high-yield — so-called “junk” — bonds. Milken was the Senior Vice President and Manager of Drexel’s High Yield and Convertible Bond Department that handled the Boesky reorganization. Appellee Lowell Milken, Michael Milken’s brother, also a Drexel employee in the same department, assisted his brother generally and on the Boesky deal in particular. For the underwriting services associated with placing the debt portion of the reorganization, Drexel charged Boesky $26.6 million. This fee was paid partly in cash and partly by Drexel taking an equity interest in the partnership. Drexel was also a major purchaser of partnership interests.
Meanwhile, on March 21, 1986 appellants purchased a total of approximately $10.5 million of the partnership interests directly from Boesky’s employees, without Drexel’s assistance. Specifically, Commercial Union Assurance Co. pic, Commercial Union Pensions Management, Ltd., Overbrook Nominees, Ltd., Strand Nominees, Ltd. (collectively Commercial Union) purchased $7,503,175, Gemini $1,001,927, and Mercia Zurich $2,006,572. Appellants’ purchases represented 4.2 percent of the total capital raised by the partnership.
Several months later, in November 1986, the government disclosed that Boesky had entered into a plea and cooperation agreement related to his insider trading activities. Professor David R. Herwitz of the Harvard Law School was appointed liquidation trustee of the partnership. As one might expect, in March 1987 appellants along with nearly two score originally named-plaintiffs, none of whom are appealing, filed their initial complaint against Boesky, Drexel and others, seeking to recover the capital they had invested. About two years later, in June 1989, after further public revelations of the insider trading scandal, the present complaint was amended by the addition of defendants-ap-pellees Michael and Lowell Milken.
Appellants’ claims were grounded on alleged violations of § 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and Rule 10b-5 promulgated thereunder, § 17 C.F.R. § 240.10b — 5; § 12(2) of the Se*611curities Act of 1933 as amended, 15 U.S.C. § 77i(2); and on the Racketeer Influenced and Corrupt Organizations Act, 18 U.S.C. §§ 1961, 1962 and 1964. The crux of appellants’ theory seeking recovery against the defendants is that had they known about the criminal activity of Boesky and the Milk-ens — which we need not recount — they would not have purchased their partnership interests.
Six months after the Milkens were joined in the complaint, appellants began receiving cash payments' resulting from their ownership of partnership interests. On January 7, 1990 Professor Herwitz distributed nearly the full amount of their invested capital. More precisely, Commercial Union received $7,482,189, about $21,000 less than its investment of $7,503,175. Gemini and Mercia Zurich received the exact amounts of their investments. A subsequent distribution on December 2, 1991 returned an additional $1,092,116, a yield of 10.2 percent on appellants’ invested capital. A third-party settlement of $463,934 made on October 7, 1991 yielded 4.4 percent to appellants on their capital investment. A de minimis distribution was made on April 9,1992. Thus, appellants have received $12,047,076, or 114.6 percent of their initial capital investment and they still own their partnership interests. Our calculations reflect only payments made to those appellants presently before us, therefore they differ slightly from those of the district court.
Throughout this period Michael Milken was attempting to resolve the criminal charges the government had brought against him. On April 24, 1990 he pled guilty to six felony charges, including conspiracy to violate the federal securities laws, securities fraud, mail fraud, and assisting in the filing of a false tax return. In conjunction with this plea, Milken agreed to pay $400 million to a restitution fund created in settlement of the Securities and Exchange Commission’s (SEC) civil enforcement action, and a $200 million criminal fine.
In an attempt to resolve the pending civil litigation, in March 1992 Milken and his brother entered into a complex settlement arrangement (Global Settlement) with plaintiffs, substantially funded by $500 million contributed by Milken. Appellants, as was their privilege, opted out of this settlement and elected to pursue their then pending litigation against defendants. Accordingly, the district court instructed them to file an amended complaint that would identify the parties and claims severed from the original complaint, and to file a statement in support of their RICO claim against defendants.
At the time when this appeal was heard defendants had not answered the amended complaint. Milken has filed a general denial and admitted the facts both as stated in his criminal allocution and in a letter from his attorney to the sentencing judge. Milken also raised four affirmative defenses. Lowell Milken has also filed a general denial, asserted his Fifth Amendment privilege and raised 12 affirmative defenses. It is on this amended complaint (a 314-page opus) that defendants moved for summary judgment. On April 30,1993 while this motion was pending, and following instructions from the district court, appellants filed a new amended complaint. Neither the Milken defendants nor Boesky have been deposed in the instant action due to a discovery stay issued by the district court.
In a reported decision, Commercial Union Assurance Co. v. Ivan F. Boesky & Co., 824 F.Supp. 348 (S.D.N.Y.1993), Judge Pollack granted summary judgment to defendants and dismissed appellants’ complaint. He also invited defendants to seek Rule 11 sanctions against appellants, but stayed his decision on the sanctions pending the outcome of this appeal. From the grant of summary judgment this appeal ensued.
DISCUSSION
We review a district court’s grant of summary judgment de novo, assessing the record in a light most favorable to the nonmoving party, see Bryant v. Maffucci, 923 F.2d 979, 982 (2d Cir.), cert. denied, - U.S. -, 112 S.Ct. 152, 116 L.Ed.2d 117 (1991), and applying the same standard as that applied by the district court, see Burtnieks v. City of New York, 716 F.2d 982, 985 (2d Cir.1983). Under Fed.R.Civ.P. 56(c) summary judgment *612should be granted if there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law.
I Appellants’ RICO Claim
The RICO claim was dismissed by the district court on alternative grounds: one on a finding that appellants had not suffered compensable damages under RICO, and the other based on a legal ruling that the claim was insufficient on its face because the existence of an ongoing enterprise separate and apart from the purported pattern of activity asserted had not been pleaded. See Commercial Union, 824 F.Supp. at 350. Dismissal of the RICO claim must be affirmed because appellants’ claim lacks that most fundamental of legal elements necessary to support a viable cause of action—any demonstrable damages.
As an initial matter, we reject appellants’ contention that the payments by Professor Herwitz were settlements and not a return of appellants’ capital investment. This conclusion is warranted given that the monies paid to appellants derived directly from the proceeds of the sale of assets held by the partnership, as appellants’ counsel conceded. The complex “settlement” appellants constantly refer to was simply a reordering of partnership interests to facilitate the Boesky partnership liquidation.
The need to classify the distributions made to appellants is apparent, given the statutory basis for their claim. Civil RICO provides a treble-damages remedy, but that remedy only inures to persons who have been “injured in [their] business or property by reason of’ a RICO violation. 18 U.S.C. § 1964(c). See Bankers Trust Co. v. Rhoades, 859 F.2d 1096, 1100 (2d Cir.1988) (“plaintiff [must] suffer injury in fact”), cert. denied, 490 U.S. 1007, 109 S.Ct. 1642, 104 L.Ed.2d 158 (1989); see also Sedima, S.P.R.L. v. Imrex Co., 473 U.S. 479, 496, 105 S.Ct. 3275, 3285, 87 L.Ed.2d 346 (1985) (a RICO plaintiff “only has standing if, and can only recover to the extent that, he has been injured”). Appellants’ principal contention is that they are entitled to a trebling of their damage award before any offset through settlements, restitution, recoupment or otherwise. This proposition ignores controlling precedent and mischaracterizes the monies distributed to them.
We recently ruled that after a RICO claim has been successfully collected it is “abated pro tanto, prior to any application of trebling.” Stochastic Decisions, Inc. v. DiDomenico, 995 F.2d 1158, 1166 (2d Cir.), cert. denied, — U.S. -, 114 S.Ct. 385, 126 L.Ed.2d 334 (1993). In Stochastic, plaintiff held several outstanding state-court judgments against the defendants, which defendants had artfully avoided paying through the use of fraudulent transactions. Plaintiff brought a civil RICO action and argued that it was entitled to a trebling of the amount of the outstanding judgments based on the amount outstanding at the time it brought its claim. We refused to award treble damages because collection of the claims had not been frustrated when litigation began. See id.
Appellants’ RICO claim in this case is strikingly similar. They assert they are entitled to a trebling of the full amount of their invested capital ($10.5 million), none of which had been returned when they initiated suit. But, as related, they have now recouped not only their initial investment, but also have received 10.2 percent return on their capital, exclusive of additional funds paid to them as the result of several third-party settlements. Appellants’ recovery stands in contrast to the risks—of which they were advised in the prospectus and investor suitability questionnaire—entailed in the purchase of the partnership interests. If a portion or all of their investment in the partnership was unrecoverable, a treble damage award might be appropriate, assuming the other RICO requirements were satisfied. But damages as compensation under-RICO § 1964(c) for injury to property must, under the familiar rule of law, place appellants in the same position they would have been in but for the illegal conduct. Here appellants have already been placed by defendants in that position. Hence, in the instant case, without provable damages, no viable RICO cause of action may be maintained.
Appellants’ reliance on Liquid Air Corp. v. Rogers, 834 F.2d 1297, 1310 (7th Cir.1987), *613cert. denied, 492 U.S. 917, 109 S.Ct. 3241, 106 L.Ed.2d 588 (1989), and other similar non-controlling cases is misplaced. Liquid Air was considered and distinguished in Stochastic, see 995 F.2d at 1166, and is equally unavailing here. The defendants in Liquid Air returned the property at issue after a RICO judgment had been entered against them. See 834 F.2d at 1310. Appellants in the case at hand received their capital investment back in full less than ten months from the time they commenced suit. We recognize that some cases cited by appellants suggest that trebling prior to any offset is appropriate given the punitive and remedial role of treble damages. Nevertheless, we are disinclined to permit the trebling of damages by upholding a non-viable lawsuit.
Since we affirm the grant of summary judgment to defendants on appellants’ RICO claim, we need not discuss the district court’s alternative holding dismissing the RICO cause of action because it failed to allege an ongoing criminal enterprise, separate and apart from the purported pattern of activity asserted.
II Appellants’ 10b-5 Cause of Action
We turn next to consider appellants’ 10(b)-5 claim. Damage is an element of a 10b-5 cause of action that seeks a monetary award. See Royal Am. Managers, Inc. v. IRC Holding Corp., 885 F.2d 1011, 1015 (2d Cir.1989). As appellants recognize, they have not suffered any direct pecuniary loss under 10b-5 for the reasons discussed with respect to the RICO claim. Accordingly, in an attempt to satisfy this element and rehabilitate their 10b-5 claim, they have set forth three alternative theories: out-of-pocket, benefit-of-the-bargain, and disgorgement damages.
A. Out-of-Pocket Damages
Given appellants’ recovery of their capital investment plus interest, their out-of-pocket theory damage calculation relies on the inclusion of a presumed award of nine percent prejudgment interest for the four years that have passed since they instituted this action against appellees. The 14.6 percent return received by appellants struck the trial court as a suitable prejudgment interest substitute. Ruling that appellants were not entitled to any further prejudgment interest in their Rule 10b-5 cause of action, the district court dismissed this claim. See Commercial Union, 824 F.Supp. at 350.
The decision whether to award prejudgment interest and its amount are matters confided to the district court’s broad discretion, and will not be overturned on appeal absent an abuse of that discretion. See Rolf v. Blyth, Eastman Dillon & Co., 637 F.2d 77, 86 (2d Cir.1980). Here, the trial court’s decision is examined in light of the fact that its determination effectively and entirely foreclosed appellants’ Rule 10b-5 cause of action. Defendants insist the prejudgment interest issue is irrelevant because there is no judgment on which to base an award of such interest. While initially attractive, this argument is simplistic. Appellants aver the lack of return on their capital investment is the damage they have incurred. Because lack of return is the kind of rationale usually relied on for awarding prejudgment interest, we must analyze the district court’s disposition using the guides for determining whether or not to make an award of prejudgment interest.
A decision to award prejudgment interest “should be a function of (i) the need to fully compensate the wronged party for actual damages suffered, (ii) considerations of fairness and the relative equities of the award, (iii) the remedial purpose of the statute involved, and/or (iv) such other general principles as are deemed relevant by the court.” Wickham Contracting Co. v. Local Union No. 3, IBEW, 955 F.2d 831, 833-34 (2d Cir.), cert. denied, - U.S. -, 113 S.Ct. 394, 121 L.Ed.2d 302 (1992). The district court summed up its reasons for denying any further interest as follows:
The present overblown complaint and the circumstances involved in this litigation commencing with the first and continuing through the fifth amended complaint are such that fundamental considerations of fairness and litigative propriety and normal restraint warrant the Court in the exercise of an informed discretion to deny *614the grant of an overlay of interest on the original investment on top of the 14.6 percent addition which the [appellants] have received on their capital investment. What has been repaid to the [appellants] more than adequately satisfies any equitable demand for interest on the capital recovery. ...
Commercial Union, 824 F.Supp. at 350. This reasoning may not be said to constitute an abuse of discretion.
In deciding whether an award of prejudgment interest is warranted, it must be remembered that this is an equitable remedy and courts must be careful that an award does not overcompensate a plaintiff. See Wickham Contracting Co., 955 F.2d at 834. Clearly the investment in the Boesky partnership was extremely risky, one where the investors might well have lost their capital, even absent any wrongful conduct by defendants. Appellants were fully aware of the risk and chose to invest in spite of it.
Appellants accuse the district court of “working] backwards” to arrive at an interest rate of 3.8 percent per annum when the proper rate should be 9 percent — the statutory rate under New York law. They declare our decision in Norte & Co. v. Huffines, 416 F.2d 1189 (2d Cir.1969) (per curiam), cert. denied, 397 U.S. 989, 90 S.Ct. 1121, 25 L.Ed.2d 396 (1970), requires the trial court to make specific findings before departing from the statutory rate. This reading of Norte is too sweeping. To the contrary, we there instructed the district court to make findings in that particular case of the personal wrongdoings of defendants. See id. at 1191-92. Setting the rate of interest is another component of a district court’s equitable discretion when it awards prejudgment interest.
Appellants rely on Myron v. Chicóme, a Seventh Circuit decision suggesting that “an award of prejudgment interest is particularly appropriate in cases involving investment fraud.” 678 F.2d 727, 733 (7th Cir.1982). This reliance ignores the fact that appellants received a return on their investment that may fairly be characterized as an interest award for purposes of Rule 10b-5. Further, in an attempt to imply a bias on the part of the district judge, appellants assert the denial of additional interest was related to their decision to opt-out of the Global Settlement. No evidence in the record supports this assertion and it is not apparent from a reading of the district court’s opinion, as appellants suggest. Any such motivation would of course be improper and beyond those factors a trial judge may consider when determining whether or not to award prejudgment interest. See Wickham Contracting Co., 955 F.2d at 833-34. No such basis appears from the present record. Thus, the out-of-pocket theory does not help appellants establish an entitlement to damages.
B. Benefit-of-the-Bargain Theory
The benefit-of-the-bargain theory is equally unavailing. Benefit-of-the-bargain damages in a Rule 10b-5 action are not available unless they can be calculated with reasonable certainty. See Barrows v. Forest Labs., Inc., 742 F.2d 54, 59-60 (2d Cir.1984). Appellants rely primarily on Osofsky v. Zipf, 645 F.2d 107 (2d Cir.1981), for the proposition that benefit-of-the-bargain damages are readily available in a Rule 10b-5 action. Yet, in Barrows we read Osofsky as establishing the “reasonable certainty” requirement. See Barrows, 742 F.2d at 60 (“The holding of Osofsky ... turn[s] ... on the distinction between damages that are speculative and those which are certain.”); see also Levine v. Seilon, Inc., 439 F.2d 328, 334 (2d Cir.1971) (governing rule under 10b-5 is “that a defrauded buyer of securities is entitled to recover only the excess of what he paid over the value of what he got, not, as some other courts had held, the difference between the value of what he got and what it was represented he would be getting”).
In this case, appellant buyers of partnership interests cannot demonstrate to a “reasonable certainty” what the benefit of their bargain was or what it should have been. They maintain that damages could be calculated by “extrapolation of the returns the Partnership would have earned from March 21, 1986 through January 7, 1990, based on evidence of the returns the Partnership actually earned during its first three months of operations, supported by evidence *615of returns paid by other arbitrage firms during that same period.” On its face, there is nothing “reasonably certain” about any computation from this extrapolation. The concept of recovery of damages for a benefit-of-the-bargain is founded on the agreement made, not on a proposed hypothetical agreement built on outside evidence and on speculation regarding what the parties might have done or received had the circumstances surrounding the agreement been different. See Barrows, 742 F.2d at 60. Accordingly, appellants’ “extrapolation” cannot be the basis for a Rule 10b-5 damage award founded on a benefit-of-the-bargain theory.
C. Disgorgement
Appellants’ disgorgement theory, which was not addressed by the district court, is similarly without merit. To recover under such a theory appellants would have to prove that defendants’ alleged illegal profits from appellants’ $10.5 million investment exceeded the amount appellees have thus far disgorged. Since Michael Milken has agreed to pay a total of $1.1 billion, the disgorgement theory borders on the frivolous. See, e.g., Litton Indus., Inc. v. Lehman Bros. Kuhn Loeb Inc., 734 F.Supp. 1071, 1076 (S.D.N.Y.1990) (“[0]nce ill-gotten gains have been disgorged to the SEC, there remains no unjust enrichment and, therefore, no basis for further disgorgement in a private action.”).
Ill Appellants’ § 12(2) Claim
For the same reasons as those discussed with respect to appellants’ RICO and Rule 10b-5 claims, appellants have not suffered compensable damages under § 12(2) of the Securities Act of 1933. While the district judge held defendants were not statutory sellers for purposes § 12(2), see Commercial Union, 824 F.Supp. at 351, we can affirm on the alternative damages ground since the parties argued the damages issue in the district court and briefed it before us, see Viacom Int’l Inc. v. Icahn, 946 F.2d 998, 1000 (2d Cir.1991) (affirming on grounds other than those relied upon by the district court), cert. denied, — U.S. -, 112 S.Ct. 1244, 117 L.Ed.2d 477 (1992).
Section 12(2) states that a party “may sue ... to recover the consideration paid for such security with interest thereon, less the amount of any income received thereon, upon the tender of such security, or for damages if he no longer owns the security.” 15 U.S.C. § 77i(2). Because appellants admittedly still own the partnership interests, their remedy, if any, is rescission. See Randall v. Loftsgaarden, 478 U.S. 647, 655, 106 S.Ct. 3143, 3148, 92 L.Ed.2d 525 (1986). Under the rescissory measure of damages appellants would be entitled to a return of the consideration paid for the partnership interests plus prejudgment interest, less any income received on the interests. See id. at 656, 106 S.Ct. at 3149.
Aonellants’ § 12(2) claim fails for substantially the same reasons as their KiOU and Rule 10b-5 claims. They paid $10.5 million for their interests and the district court determined they were only entitled to the 14.6 percent return on their investment. We realize that Congress specifically provided in § 12(2) for an award of interest. But we do not read § 12(2) as insisting upon any set rate of interest. Rather—as in the Rule 10b-5 context—under § 12(2) an award of prejudgment interest rests in the sound discretion of the trial court after careful consideration of the factors set forth in Wickham Contracting Co., 955 F.2d at 833-34.
In sum, appellants’ total damages are $12,047,076 (their original investment plus 14.6 percent interest), which must then be offset by appellants’ recovery of that amount from the distributions made by Professor Herwitz and the third-party settlements received. See Randall, 478 U.S. at 660, 106 S.Ct. at 3151 (noting that an “implicit offset for a return of consideration ... confined to the clear case in which such money or property is returned to the investor” would be proper under § 12(2)). The net result of these computations is that appellants have not suffered compensable damages under § 12(2). Allowing them to maintain a cause of action under such circumstances would constitute a waste of judicial resources and a thwarting of Congress’ aim in enacting § 12(2). See H.R.Rep. No. 85, *61673d Cong., 1st Sess. 9 (1933) (under § 12 the buyer can “sue for recovery of his purchase price, or for damages not exceeding such price”) (quoted in Randall, 478 U.S. at 656, 106 S.Ct. at 3149).
Nonetheless, we must say that had appellants suffered compensable damages, their cause of action, contrary to the district court’s view, would have been sufficient to withstand a motion for summary judgment. Privity between the buyer and seller is no longer required in a § 12(2) action. See Pinter v. Dahl, 486 U.S. 622, 646, 108 S.Ct. 2063, 2078, 100 L.Ed.2d 658 (1988) (broker acting as agent for seller is liable as statutory seller within meaning of § 12(2)); Wilson v. Saintine Exploration & Drilling Corp., 872 F.2d 1124, 1126-27 (2d Cir.1989). Since appellants did not purchase the interests directly from Drexel or the Milkens, defendants will be liable under § 12(2) only if they “solicited” the sale. As the Supreme Court teaches, the term seller includes the person “who successfully solicits the purchase, motivated at least in part by a desire to serve .his own financial interests or those of the securities owner.” Pinter, 486 U.S. at 647, 108 S.Ct. at 2078; see also Wilson, 872 F.2d at 1125 (“statutory sellers [are] only those who actually solicit the sale of securities for financial gain”). At the other end of the liability spectrum, the Supreme Court in Pinter stated that “§ 12’s failure to impose express liability for mere participation in unlawful sales transactions suggests that Congress did not intend that the section impose liability on participants collateral to the offer or sale.” 486 U.S. at 650, 108 S.Ct. at 2079.
In Wilson, grappling with the line drawn by Pinter, we ruled that a law firm that performed strictly ministerial acts was “collateral” to a transaction, but “brokers who might act on the seller’s behalf for a profit” could be liable. Wilson, 872 F.2d at 1126-27. In any event, regardless of where the line is drawn exactly, there exists a material issue of fact as to defendants’ role in appellants’ purchases of the partnership interests sufficient to defeat a summary judgment motion. This conclusion finds further support in the fact that appellants never had the opportunity to depose either Boesky or the Milkens. But because appellants have no compensable damages, the grant of summary judgment on the § 12(2) claim must be affirmed on that ground rather than on the district court’s incorrect reasoning that no question of fact existed as to whether defendants were statutory sellers for § 12(2) purposes.
CONCLUSION
Consequently, for the reasons stated, the judgment of the district court is affirmed.