JERRY E. SMITH, Circuit Judge:

Fidelity Deposit Company of Maryland ("F D") appeals a judgment entered pursuant to a jury verdict in favor of the Federal Deposit Insurance Corporation ("FDIC") 827 F. Supp. 385. We find no reversible error and affirm.

I. A.

F D was the fidelity bond insurer of the now defunct Capital Bank and Trust Co. ("Capital" or "Bank") in Baton Rouge. Capital went bankrupt in October 1987 as a result of the fraudulent acts of its chief lending officer, Allie Pogue.

In late 1986, Capital suspected that Pogue was making loans in exchange for bribes. Richard Easterly, its president, became suspicious of Pogue in August 1986. Easterly called upon Susan Rouprich, Capital's vice president and in-house attorney, and the Bank's internal audit department under Paula Laird, to investigate. The resulting report indicated that there were a number of undisclosed business relationships between Pogue and some of the loan customers.

Soon after the report, Easterly, purportedly contemplating the prompt notice-of-loss requirement in the Bank's fidelity bond, prepared and filed a notice-of-loss letter with F D. In February 1987, Capital filed a proof of loss that detailed some of the loans that Pogue had approved to persons from whom he had received financial benefits.

B.

There are four main groups of loans at issue in this case that involved improprieties by Pogue.

Jimmy Scott loans

Jimmy Scott and his affiliates obtained millions of dollars in loans from Capital, including the disputed Carrol Herring loan. All of these loans were approved by Pogue and certainly resulted in a major loss to Capital. Jimmy Scott bribed Pogue into approving the loans by buying him a duplex and giving him gifts.

The Herring deal was a $375,000 loan that Pogue arranged from Capital to Carrol Herring. The loan proceeds were utilized in a series of transactions, through attorney J. Glenn Dupree, to buy property from Pogue and pay off Pogue's mortgage on the property. Pogue and Dupree were indicted and pled guilty to giving and taking a kickback on the Herring loan.

LAREEL loans

Pogue and one of his customers, Wayne Bunch, became partners in several business ventures, including Aspen Partnership, which owned a four-plex in Baton Rouge. The partnership owed a $136,101.61 mortgage on the property, with a balloon payment for the balance due at the end of September 1986. Because Bunch was in financial trouble, Pogue developed a scheme to rehabilitate their finances.

The Aspen Partnership sold the four-plex to Thomas Keene, who was the operative of Louisiana Real Estate Equity, Limited ("LAREEL"). Keene then syndicated the project to some investors. Keene paid off the Aspen Partnership loan with a personal check, using funds provided by LAREEL. Pogue greatly benefited from this sale.

Following this transaction, the LAREEL loans occurred. LAREEL agreed to "syndicate" other Bunch projects by offering the units to "investors" who put no money down, but signed notes to Capital for amounts far in excess of Bunch's liability on the units. LAREEL and Keene pocketed the excess cash in the loans. Pogue obtained the individual investor financing from Capital.

Pogue recommended to the executive committee of Capital that the series of loans be made to the LAREEL investors on the Bunch projects, but never disclosed that he had recently benefited from actions by LAREEL with respect to the four-plex. LAREEL and Keene greatly profited from the LAREEL loans.

Pogue knew that several of the "investors" were not credit-worthy. Even though several of the LAREEL loan applications were rejected by one loan officer, Pogue ignored the recommendation and ordered that the loans be made. Keene refused to testify at the trial, asserting his Fifth Amendment privilege when questioned about the loans.

Quadrant/Thompson loans

In July 1984, Pogue and Bunch sold a four-plex to the Quadrant Partnership, whose partners were Theodore Jones, an attorney, and Robert Killingsworth and Jimmy C. Thompson, two real estate promoters. All were loan customers of Capital. Quadrant assumed a $135,291 mortgage owed by Pogue and Bunch on the four-plex and was supposed to pay $20,000 in cash to Pogue, but only $6,666 in fact was paid. A note for the balance secured by a second mortgage on the four-plex was pledged for the $13,334 balance.

On the day the sale closed, Pogue approved a $12,000 loan by Capital to Quadrant for the stated purpose of a down payment on the purchase of a four-plex for syndication purposes. A letter from Quadrant to the attorney for the bank that held the first mortgage on the property stated that the acquisition of the property was done as an accommodation to Pogue. The "accommodation" eventually resulted in over $1,000,000 in loans to support syndications by Quadrant and its related entities, resulting in significant losses. Pogue never disclosed his personal dealings with Quadrant to Capital.

Robert Harger loans

In late September 1982, Pogue and Bunch sold a group of fourplexes for $725,000 to Oakbourne Apartments Partnership, Ltd., a company owned by one of Pogue's Capital loan customers, Robert Harger and Associates. Oakbourne assumed the $545,000 mortgage on the property, though Pogue and Bunch remained personally liable on the loan. According to the sales agreement, $180,804 was to be paid in cash, and the rest was to be paid by the assumption of the mortgage. In reality, only $15,804 was paid in cash, and Oakbourne issued a $165,000 note to Pogue and Bunch, secured by a second mortgage on the property.

Capital loaned the Harger company money, approved by Pogue, while these dealings were ongoing. Harger eventually ceased payments on the $545,000 debt, and the lender began to pressure Pogue and Bunch. Harger, however, continued to pay Pogue on the $165,000 note. The loans, which Pogue approved from Capital to Harger, allowed Harger more easily to pay Pogue the money it owed him and to shelter him from liability on the first mortgage.

Pogue resigned from Capital effective at the end of August 1986. He was hired as the president of Acadia State Bank ("Acadia") in Baton Rouge. Acadia also made loans to Harger after Pogue arrived. Proceeds were used to catch up on the $545,000 loan on which Pogue and Bunch were personally liable.

C.

Capital's chairman, Embree Easterly, and its president, Richard Easterly, testified that if they had known of Pogue's personal relationships with loan clients, they would not have allowed him to handle loan decisions for those persons or entities. Following Capital's filing of the proof of loss, F D refused to pay the Bank's claim under the fidelity bond. As a result, Capital filed suit against F D in April 1987. The bond expired on October 1, 1987; the Bank failed at the end of October 1987. The FDIC took receivership of the Bank and stepped into its shoes for purposes of the lawsuit.

The trial lasted for three weeks in November 1992. The jury was given 27 separate verdict forms corresponding to 27 separate loan transactions. Verdicts were returned in favor of the FDIC on 17 of the loan transactions, totaling $5.313 million. Because the limit on the bond is $4 million, the district court entered final judgment for $4 million plus interest.

II.

The fidelity bond in this case covers "[I]oss resulting directly from the dishonest or fraudulent acts of an employee committed alone or in collusion with others . . . with the manifest intent to cause the Insured to sustain such loss." The bond applies "to loss discovered by the Insured during the bond period."

After the evidence had been presented at trial, the jury was presented with an interrogatory for each questionable loan. Each interrogatory contained four separate questions with respect to the particular loan:

1. Did Allie Ray Pogue commit a dishonest or fraudulent act in connection with this loan?

2. Did a loss result directly from Allie Ray Pogue committing a dishonest or fraudulent act with the manifest intent to cause Capital Bank a loss and to obtain a financial benefit for himself or others?

3. Did discovery, as defined in the Bond, occur on this loan prior to October 1, 1987?

4. What is the amount of the loss on this loan resulting directly from Allie Ray Pogue's dishonest or fraudulent acts?

The FDIC had to prove four distinct elements for each loan. First, each loss must have been discovered within the bond period. The relevant section of the bond reads:

This bond applies to loss discovered by the Insured during the bond period. Discovery occurs when the Insured becomes aware of facts which would cause a reasonable person to assume that a loss covered by the bond has been or will be incurred, even though the exact amount or details of loss may not then be known. . . .

In interpreting these clauses, courts have held that "discovery of loss does not occur until the insured discovers facts showing that dishonest acts occurred and appreciates the significance of those facts; suspicion of loss is not enough." FDIC v. Aetna Casualty Sur. Co., 903 F.2d 1073, 1079 (6th Cir. 1990) (citing, inter alia, United States Fidelity Guar. Co. v. Empire State Bank, 448 F.2d 360, 364-66 (8th Cir. 1971)). See also California Union Ins. Co. v. American Diversified Sav. Bank, 948 F.2d 556, 564 (9th Cir. 1991).

The FDIC had to prove dishonest or fraudulent acts on Pogue's part and causation between the loss and the fraudulent or dishonest acts. See First Nat'l Bank v. Lustig, 961 F.2d 1162 (5th Cir. 1992). The FDIC also had to show the amount of the loss resulting from the fraud.

F D challenges a number of the jury's findings for certain loans. In order to obtain a reversal, F D must show that no reasonable juror could have found in favor of the FDIC even when viewing all of the evidence in the light and with all reasonable inferences most favorable to the FDIC. Boeing Co. v. Shipman, 411 F.2d 365, 374 (5th Cir. 1969) (en banc).

A.

F D asserts that the FDIC did not prove that the Bank discovered the losses from the LAREEL and Herring loans within the bond period. F D claims that the proof of loss, which Capital filed, did not pinpoint these particular loans as losses. The FDIC counters that these losses were part of one huge "loss" that resulted from all of Pogue's dishonest actions. The FDIC argues that, even if the Herring and LAREEL transactions were later-discovered, they were part of the same "loss" that was discovered during the bond period.

Section Four of the bond plainly limits the coverage to "loss discovered by the Insured during the bond period." Section Three states that the total liability is limited to loss resulting from "all acts or omissions by any person (whether Employee or not) or all acts or omissions in which such person is implicated." Thus, "loss" is a broad term that covers all losses from the acts of an employee, but the "loss" must be discovered within the bond period.

As a result, we conclude that Capital's proof of loss need not have pinpointed every single loan loss. We, however, decline to adopt the FDIC's broad suggestion. We are unwilling to create a situation in which, as long as the FDIC can find some acts or omissions within the bond period committed by the same actor, it has unlimited time to investigate and add later losses caused by unrelated actions. We instead will look for loans that arose out of the same pattern of conduct or scheme that was originally discovered. See, e.g., Howard, Weil, Labouisse, Friedrichs v. Ins. Co. of N. Am., 557 F.2d 1055, 1059-60 (5th Cir. 1977) (dishonesty found in "totality of actions," loss sustained was "a single loss albeit the product of more than one act" though part of "a single ongoing episode"). We also note that under the broad language from Section Four of the contract, to uphold the finding on this issue, we must decide only that a jury could have concluded that a reasonable person, within the bond period, given the information available, would have "assum[ed] that a loss covered by the bond had been incurred or [would] be incurred, even though the exact amount or details of loss" were not known then.

There was plainly enough evidence from which the jury properly could have concluded that the LAREEL and Herring transactions were discovered within the Bond period. The proof of loss implicated parties who were involved in the transactions that constituted the LAREEL and Herring loans. Scott, who received money from the Herring loan, was a prominent part of the proof of loss; Keene, who was involved in the LAREEL transaction, was also mentioned in conjunction with Bunch.

Strictly speaking, the proof of loss did not mention every single action by Pogue that had something to do with the LAREEL and Herring transactions. Nevertheless, because the FDIC did present evidence that at least some of the acts and omissions related to the LAREEL and Herring transactions were discovered during the period, we will not disturb the jury verdict.

B.

F D argues that the FDIC presented no evidence that Pogue was involved in approving the Morning Glory/LAREEL loans. F D claims that no record evidence shows that Keene and Pogue knew each other at the time of the Morning Glory loan. Moreover, F D asserts that no one at the Bank testified that the Bank would not have approved the Sable Chase and Bayou Fountain loans had it known about the Aspen Partnership sale to Keene. Finally, F D contends that the FDIC never proved that Pogue intended to cause losses on the LAREEL loans.

The evidence indicates that Pogue and Bunch, who was a Capital loan client, were business partners in more than one venture in early 1986 when the LAREEL loans began. Evidence indicates that Pogue and Bunch sold a piece of property to the LAREEL operative, Keene, in February 1986. Keene paid off Pogue's personal debt on a loan to the Aspen Partnership, which was Pogue and Bunch's venture.

The LAREEL loans for the Sable Chase and Bayou Fountain projects were transactions in which LAREEL agreed to find "investors" for Bunch projects. The investors, who put no money down, received loans from Capital. Pogue allowed Capital to make the loans with the knowledge that the "investors" were largely uncreditworthy, that the income from the projects would not support the large loan payments, and that the Bank probably would suffer a loss as a result. Moreover, Keene and LAREEL made substantial commission income from the loan activity.

Pogue used his position to pressure the loan officers working under him into approving LAREEL loans that they had previously rejected. Pogue concealed his relationships with lending clients, such as Bunch, from Capital's president and chairman, and they would not have allowed him to approve loans if they had known. The FDIC showed that Pogue was a sophisticated loan officer. The jury might have inferred that he would not have approved the LAREEL loans to the risky investors had he not had a personal stake in the outcome.

F D is correct, however, that the FDIC failed to show any connection between Pogue and Keene with respect to the Morning Glory transaction. The sale of the four-plex that occurred between the Aspen Partnership and Keene was initiated in January 1986 and closed in February. The Capital loans for the Morning Glory condominiums occurred in January 1986, before the four-plex transaction. By contrast, the loans for the Sable Chase and Bayou Fountain projects occurred after the four-plex sale.

Moreover, the FDIC produced a memorandum from Pogue to the executive committee of the Bank recommending approval of the Sable Chase and Bayou Fountain loans. The memorandum occurred in March 1986, after the four-plex deal and the Morning Glory loans. The memorandum to the executive committee recommending the approval of the Morning Glory loans was from Mark Byouk and not from Pogue. Testimony plainly indicated that proceeds from the Sable Chase deal went directly to Keene, but evidence does not indicate that Keene received proceeds from the Morning Glory deal.

Finally, Karl Daggett testified that he was ordered by Pogue to approve investors for the LAREEL projects after he had rejected many of the applications. Daggett's testimony, however, also indicated that he was instructed to approve the loans by Pogue sometime after the disbursement of Morning Glory funds. Accordingly, we modify the judgment with respect to the Morning Glory loans but affirm with respect to the other two LAREEL projects.

C.

With respect to the Scott, Harger, and Quadrant/Thompson loans, F D claims that the causation standard that we approved in First Nat'l Bank v. Lustig, 961 F.2d 1162 (5th Cir. 1992), was not met. Lustig requires that the FDIC show that a loan would not have been made "but for" the fraudulent conduct of the employee. Id. at 1167-68. Moreover, F D claims that the FDIC never showed a dishonest act with respect to these other loans; rather, they assert that only a pattern of dishonesty was shown.

As with the LAREEL loans, the FDIC presented evidence of dishonest acts with respect to all of the loan groups. Scott admitted to bribing Pogue to influence him to make loans. This was a charge to which Pogue had pled guilty. The FDIC introduced the letter in which Quadrant indicated that it purchased a four-plex as an accommodation to Pogue. Harger testified as to the purchase of four-plexes from Pogue and Bunch that Pogue did not disclose.

As indicated above, the FDIC presented testimony from Bank officers that the loans would not have been made if they had been aware of Pogue's personal relationships with the various clients. Moreover, as before, Pogue's high position and sophistication could have given rise to inferences that he would not have made several of the loans if he had been honest. The jury might also have given weight to the testimony of Pogue's subordinates that he cajoled them into favoring certain borrowers. There was plainly enough evidence presented to meet the causation requirement.

A general pattern of dishonesty, rather than a dishonest act for each loan, is sufficient in this circuit. See Fidelity Deposit Co. v. USAFORM Hail Pool, Inc., 523 F.2d 744, 757 (5th Cir. 1975), cert. denied, 425 U.S. 950, 96 S.Ct. 1725, 48 L.Ed.2d 194 (1976). "There does not have to be a finding of fraud or dishonesty with respect to every disbursement." Id. Thus, as to each group of loans, the jury was entitled to find that Pogue was motivated by separated instances of bribery to make multiple loans.

III.

F D argues that the February 1987 proof of loss did not adequately satisfy the requirements of section 5(b) of the bond. The main substance of F D's claim is the same as the earlier argument that we rejected, i.e., that the proof of loss did not contain specific losses that were later added by the FDIC. Furthermore, the purely factual issues of the adequacy and timing of the proof of loss were raised only after the trial. F D claims that the issue was raised in its FED.R.CIV.P. 50 motion and that it is an essential element of the bond claim.

There was not a specific ruling on the adequacy and timing of the proof of loss. F D did not ask for an instruction to the jury on this issue. In fact, it does not appear that F D even argued this theory at trial. As a result, we reject this assertion.

IV.

F D claims that the jury's answers were inconsistent and that, when inconsistent answers are given to special interrogatories, the court must grant a new trial. Specifically, F D questions the finding of liability on the first Quadrant/Thompson loan and not the other one and inconsistent answers on the Scott and Robert Harger loans.

Jury verdicts are supposed to be reconciled, if possible, to validate a verdict when answers appear to conflict. White v. Grinfas, 809 F.2d 1157, 1161 (5th Cir. 1987). "The test of consistency is whether the answers may fairly be said to represent a logical and probable decision on the relevant issues as submitted." Central Progressive Bank v. Fireman's Fund Ins. Co., 658 F.2d 377, 382 (5th Cir. Unit A Oct. 1981) (citation and internal quotation marks omitted). There is nothing necessarily inconsistent about the verdicts in this case, as the FDIC had to prove each one of four elements with respect to each loan. The proof certainly overlapped, as has been explained, but not on every loan.

The twenty-seven special interrogatories essentially were separate jury verdicts. The apparent "inconsistency" in this case was not the type of inconsistency that is present where a jury returns conflicting answers on necessarily related jury questions. See, e.g., Royal Netherlands S.S. Co. v. Strachan Shipping Co., 362 F.2d 691 (5th Cir. 1966), cert. denied, 385 U.S. 1004, 87 S.Ct. 708, 17 L.Ed.2d 543 (1967). As a result, a new trial would be improper.

V.

F D argues that the court improperly instructed the jury that it could draw an inference from an invocation of the Fifth Amendment privilege against self-incrimination by a non-party. F D claims that the FDIC never established corroborating evidence linking Pogue to the loan claims before allowing the jury to draw inferences.

There were several witnesses who purportedly had had relationships with Pogue and who invoked the Fifth Amendment at trial. Much of F D's argument with respect to the lack of corroborating evidence relates to the argument that the FDIC failed to prove a dishonest act as to each of the loans claimed. We have rejected this argument, supra. Moreover, the court instructed the jury not to find liability based solely upon an adverse inference from a witness's invocation of the Fifth Amendment.

F D's main argument is that the court improperly allowed the invocation of the Fifth Amendment, by a non-party or non-agent of a party, to be the basis of an inference against a party. In general, the decision as to whether to admit a person's invocation of the Fifth Amendment into evidence is committed to the discretion of the district court. Farace v. Indep. Fire Ins. Co., 699 F.2d 204, 210 (5th Cir. 1983). The admissibility of a non-party's exercise of the Fifth Amendment against a party, however, is a legal question that we must review de novo. Nevertheless, if such evidence is not inadmissible as a matter of law, the district court's specific determination of relevance and its evaluation of a potential FED.R.EVID. 403 problem are reviewed for abuse of discretion.

F D argues that inferences from the invocation of the Fifth Amendment are not allowed when a non-party asserts the privilege. We find no support for such a proposition. The Fifth Amendment "does not forbid adverse inferences against parties to civil actions when they refuse to testify in response to probative evidence offered against them." Baxter v. Palmigiano, 425 U.S. 308, 318, 96 S.Ct. 1551, 1558, 47 L.Ed.2d 810 (1976). We acknowledge that no party has refused to testify in this civil action, but "[a] non-party's silence in a civil proceeding implicates Fifth Amendment concerns to an even lesser degree." RAD Servs., Inc. v. Aetna Casualty Sur. Co., 808 F.2d 271, 275 (3d Cir. 1986) (citing Rosebud Sioux Tribe v. A P Steel, Inc., 733 F.2d 509, 521 (8th Cir.), cert. denied, 469 U.S. 1072, 105 S.Ct. 565, 83 L.Ed.2d 506 (1984)).

Because there is no constitutional bar to the admission of this evidence, it is admissible if it is relevant and not otherwise prohibited by the rules. FED.R.EVID. 402. Certainly, evidence of this nature is generally relevant. In this case, a jury could determine that a witness who colluded with Pogue took the Fifth Amendment to avoid disclosing that collusion. District courts must evaluate each witness separately when making a relevance determination; F D fails, however, to identify specific instances in this case where a witness's invocation of the Fifth Amendment was irrelevant.

Under FED.R.EVID. 403, evidence will be excluded if its probative value is substantially outweighed by the danger of unfair prejudice. F D has not identified how specific invocations of the Fifth Amendment prejudiced it in this case. Rather, it argues that the admission of this type of evidence is, in effect, prejudicial as a matter of law.

F D argues that it is improper to allow a non-party's invocation of the Fifth Amendment to be used against a party when that non-party is neither an agent nor an employee, officer, director or voting member of the party. The concern is that a non-party who stands in no special relationship to the party at the time of trial may purposefully invoke the privilege solely to discredit the party. The classic example would be a disgruntled former employee who invokes the privilege to hurt his former employer.

Other circuits have held that the fact that the witness no longer serves the party in an "official capacity" does not present a bar to requiring the witness to assert the privilege in front of the jury. See Cerro Gordo Charity v. Fireman's Fund Am. Life Ins. Co., 819 F.2d 1471, 1481 (8th Cir. 1987); RAD Servs., Inc., 808 F.2d at 274-79; Brink's Inc. v. City of New York, 717 F.2d 700, 707-10 (2d Cir. 1983). In each of those cases, the witness was a former employee of a company that was a party to the litigation.

While the court in Cerro Gordo Charity, 819 F.2d at 1481, found that the ex-employee still retained some loyalty to the party, thereby negating any danger of invocation of the privilege solely for the purpose of harming the employer, the RAD Servs. court found that the "absence of an opportunity to cross-examine the invoker and the lack of proof regarding his continued loyalty to the employer . . . cannot per se exclude from the jury the witness's refusal to testify." RAD Servs., 808 F.2d at 276.

Similarly, we refuse to adopt a rule that would categorically bar a party from calling, as a witness, a non-party who had no special relationship to the party, for the purpose of having that witness exercise his Fifth Amendment right. As the Third Circuit indicated:

First, a witness truly bent on incriminating [a party] would likely offer damaging testimony directly, instead of hoping for an adverse inference from a Fifth Amendment invocation. Second, the trial judge could test the propriety of an invocation to ensure against irrelevant claims of privilege. Third, counsel may argue to the jury concerning the weight which it should afford the invocation and any inferences therefrom.

Id. Thus, district courts will have to evaluate these situations on a case-by-case basis.

In this case, any danger that the jury might have found that Pogue had committed dishonest acts merely from his association with witnesses who invoked the Fifth Amendment, thereby unduly prejudicing F D, was avoided by the instruction that the jury was not to find liability absent evidence corroborating the relationships between the invoking witnesses and Pogue.

There is no question that the evidence is relevant. F D fails to make a competent argument as to why it was unfairly prejudiced by the admission of the evidence. In fact, it is the invoking party who is generally thought to be the one unfairly prejudiced in these situations. See HEIDT, supra, at 1124. In this case, accordingly, there was no abuse of discretion.

F D also argues that the district court erred by not cautioning the jury that the Fifth Amendment may be invoked by an innocent party. F D correctly asserts that model jury instructions do contain such a provision. When challenging a jury instruction, a party must demonstrate that the charge as a whole creates "substantial and ineradicable doubt whether the jury has been properly guided in its deliberations." FDIC v. Mijalis, 15 F.3d 1314, 1318 (5th Cir. 1994) (citation and internal quotation marks omitted). Moreover, we will not reverse if we believe, "based upon the entire record, that the challenged instruction could not have affected the outcome of the case." Id.

F D has not met the Mijalis standard, given the jury instruction as a whole and the quantum of evidence produced by the FDIC. The charge given in this case is substantially similar to the instruction that the Third Circuit approved in RAD Servs. and, even if erroneous, was harmless error.

VI.

F D contends that the district court erred by admitting evidence that the first loan that Pogue made as president of Acadia was to a Harger company and that the funds were used to catch up on payments of the $545,000 loan on which Pogue and Bunch were personally liable. The court admitted the evidence as res gestae of the acts at issue in the case and, in the alternative, as FED.R.EVID. 404(b) "other acts" evidence showing evidence of intent, plan, knowledge, and absence of mistake.

The rule 404(b) ruling was not an abuse of discretion, and the evidence was properly admitted. Strictly speaking, the doctrine of "res gestae" (as traditionally understood) and rule 404(b) cannot be alternative justifications. The old doctrine of "res gestae" has been supplanted by FED.R.EVID. 803. Before the Federal Rules of Evidence were promulgated, res gestae was understood to encompass four distinct hearsay exceptions: "(1) declarations of present bodily condition; (2) declarations of present mental state and emotion; (3) excited utterances; [and] (4) declarations of the present sense impression." Wabisky v. D.C. Transit Sys., 309 F.2d 317, 318 (D.C. Cir. 1962). FED.R.EVID. 803 now explicitly accounts for these exceptions.

In a normal situation, prior bad acts would have to pass rule 404(b) muster and, if a hearsay problem was raised, would have to meet any hearsay objections. In this case, the court did not use the term "res gestae" to mean an exception to the hearsay rule. F D acknowledges, and there is no question, that the evidence at issue was not of the type that would raise a hearsay problem. Instead, the court used the term "res gestae" to represent its feeling that the evidence at issue actually dealt with acts that were directly at issue in the case. In other words, the district court held that Pogue's Acadia activity was a continuation of his Capital activity.

We need not evaluate the propriety of this holding, as we find that the court did not abuse its discretion in admitting the evidence under rule 404(b). Moreover, F D fails to show that it had a substantial right affected by the admission of the evidence. See United States v. Jimenez Lopez, 873 F.2d 769, 771 (5th Cir. 1989).

VII.

F D argues that this is a fraud case and, therefore, that the actions must be proved by clear and convincing evidence. The FDIC counters that, while Pogue's dishonesty may have risen to the level of fraud, the action between the FDIC and F D is for breach of contract. The Eighth Circuit has characterized cases like this one as breach of contract actions and not fraud. See First Am. State Bank v. Continental Ins. Co., 897 F.2d 319, 323 (8th Cir. 1990). We agree with the FDIC on this issue. The suit is on the fidelity bond contract, which the FDIC alleges that F D has breached. No proof of fraud is technically required, especially against F D.

VIII.

F D argues that the court's award of pre- and post-judgment interest was erroneous. The parties agreed that Louisiana law would govern this issue.

F D contends that the court erroneously awarded interest from April 29, 1987, the date that Capital filed suit, when most of the losses had not yet been demanded. F D claims that the proof of loss, on file at that time, did not contain the LAREEL and Herring losses, as earlier noted.

The district court held that under Louisiana law, pre-judgment interest begins to accrue from the date of judicial demand, regardless of whether the damages are unliquidated, disputed, or not ascertainable until judgment. See Cotton Bros. Baking Co. v. Industrial Risk Insurers, 941 F.2d 380, 391-92 (5th Cir. 1991), cert. denied, ___ U.S. ___, 112 S.Ct. 2276, 119 L.Ed.2d 202 (1992).

F D urges that there was no default on the contract with respect to the LAREEL and Herring losses at the time of the original judicial demand. This argument is related to the earlier claim that these losses had not been specifically discovered and mentioned in the original proof of loss. Because there was evidence that this wrongdoing had been timely discovered by the Bank, the court's award was correct.

IX.

In summary, our reasoning on the Morning Glory loan reduces the original losses awarded by $1,203,674.24 from $5,313,004.17 to $4,109,329.93. Because this amount is still above the $4 million limit on the fidelity bond, the judgment of the district court is AFFIRMED.