OPINION

This private treble damage anti-trust action is here for the second time. See Estate of Le Baron, Sr., et al. v. Rohm and Haas Company and Monsanto Company, 441 F.2d 575 (9th Cir. 1971). The issue presented is a further evolvement of the evidentiary question addressed in the first appeal.

In accordance with the directions given in our prior opinion, the district court, upon remand of the proceedings, permitted plaintiff rice growers full discovery of the cost and profit data for Stam and Rogue, trade names of defendants Rohm and Haas, and Monsanto, respectively, for the herbicide propanil. The district court, in addition, permitted full discovery of the cost and profit data for a large number of other agricultural chemicals manufactured and sold by defendants.

In accordance with the procedures outlined in our prior opinion, the district court also held a hearing to determine the admissibility of the evidence offered by plaintiffs as a result of this discovery. The district court determined that the tendered evidence is inadmissible, and re-entered judgment in favor of defendants. Thereafter, the district court denied plaintiffs’ motion for reconsideration, whereupon plaintiffs took this appeal. We affirm.

In holding the proffered evidence inadmissible, the district court determined that the evidence is not probative of the existence of a conspiracy between defendants, and that any probative value such evidence might have is substantial ly outweighed by the complexities and confusion the introduction of the evidence would engender. On the motion for reconsideration, the district court also held that the evidence offered is not admissible as a basis for cross-examining defendants’ expert witness. On appeal plaintiffs take issue with each of these determinations.

In our view, the district court did not abuse its discretion in holding the evidence in question inadmissible. See Continental Baking Co. v. United States, 281 F.2d 187, 145-146 (6th Cir. 1960). The probative value of the evidence, slight at most, tends to show that the profit margins on Stam and Rogue are somewhat above the average gross and net profit margins of all agricultural chemicals manufactured and sold by defendants. But it falls short of demonstrating the “high profit margins” which, in our prior opinion, we said would be especially probative of the existence of a conspiracy. Moreover, there appear to be serious deficiencies in the tendered evidence. With reference to Monsanto, the discovered accounting records show only gross profits. As the district court pointed out, only the direct cost of manufacture is reflected in these figures, thereby excluding selling costs, administrative costs, research and development costs and other miscellaneous costs. In addition, there is reason to believe, as Rohm and Haas contends, that its net profit margin for all agricultural chemicals, to which the profit margin of Stam is compared, is understated. Both defendants attack the plaintiffs’ profit margin comparisons by asserting that evidence exists to show that, to be accurate, comparative analysis must be between products having similar market characteristics, e. g., products on the market a similar length of time.

As compared to this questionable probative value of the proffered evidence, the admission thereof at a jury trial would have resulted in extensive cross-examination and substantial countering evidence leading to obliteration of the basic issues of the case and confusion of the jury.

As the district court observed, each defendant would be entitled to make a detailed explanation of its accounting system, entailing expert testimony and cross-examination. Monsanto would be entitled to amplify the gross profits figures submitted for it by plaintiffs, by establishing the indirect costs involved in the manufacture and sale of Rogue. Both defendants would also be entitled to submit evidence bearing upon the understatement of profit margins caused by the pricing of the by-products and the inaccurate comparison analysis referred to above. While the cost and profit data submitted to the court may have had some probative value on the question of the existence of a conspiracy, such value was slight and was clearly outweighed by the difficulties the submission of this evidence would have brought to a jury trial. See Rule 403, Proposed Rules of Evidence for the United States Courts and Magistrates.

Plaintiffs’ main attack on this portion of the district court's holding is that the tendered evidence consists of contemporaneous business records of the defendants which reflect comparative profit margins which as a matter of law are relevant to the issue of competition, and thus conspiracy, and therefore must be admitted. Under this view, the district court has no discretion as to its admission, and all the court’s and the defendants’ objections go to the weight of the evidence and not its admissibility.

Plaintiffs rely primarily on Continental Baking Co. v. United States, 281 F.2d 137 (6th Cir. 1960). But the issue in Continental Baking, as in the prior appeal in this case, was whether evidence on profit margin and cost should necessarily be excluded as irrelevant, not the right of the trial judge, in the exercise of discretion, to exclude evidence where its probative value is slight and is outweighed by the difficulties its introduction will produce. In fact, the court in Continental Baking apears to affirm the discretion of the district court, on remand, to exclude the evidence, if unreliable. 281 F.2d at 145-146.

Moreover, this argument was settled adversely to plaintiffs on the prior appeal and that ruling is now the law of the case. In our prior opinion we stated: “At that time [on remand] the trial court in the exercise of its discretion may control the introduction of evidence so as to minimize problems of time and confusion. See United States v. Krulewitch, 145 F.2d 76, 80 (2d Cir. 1944).” 441 F.2d at 578 .

We also hold that the district court did not abuse its discretion in ruling that the proffered evidence is not admissible as a basis for cross-examination of defendants’ expert witness.

Plaintiffs argue that the tendered cost and profit evidence is necessary to cross-examine effectively defendants’ expert witness. That witness, plaintiffs contend, “was permitted to make an important judgment for the jury-— whether defendants’ profit margins tended to prove the absence of price competition.”

The expert expressed an opinion as to whether a high margin of profits indicated a conspiracy under these circumstances. But he did not base his opinion on the data which plaintiffs are attempting to introduce, nor in fact on any profit data of the defendant companies. In essence, the expert concluded that, no matter what the margin'of profit, in a duopoly situation the two prices would be identical and as high as possible. In addition, as defendants point out, it was the appellants’ attorney, himself, who initiated the questions concerning the cost information of the products and profit margins. The expert stated throughout that he did not rely on any cost or profit data, nor would his opinion change if he had received such data.

In controlling the scope of cross-examination the district court has wide discretion. Harries v. United States, 350 F.2d 231, 236 (9th Cir. 1965). The primary reasons which dictated that the court not allow the data introduced as evidence of conspiracy, undue confusion and consumption of time, support the decision not to allow the material as a basis for cross-examination.

Affirmed.

. In this opinion, we will assume familiarity with the holding and the litigation as described in our prior opinion. . During the period of the alleged conspiracy there is evidence that both defendants marketed a number of products, including products with substantial volume, with profit margins comparable to or exceeding those of Stam and Rogue. . Plaintiffs tendered net profits data concerning Stam, Rohm and Haas’ i>roduct, and approximately fifty other agricultural chemical products produced and sold by Rohm and Haas. But, as in the case of Rogue, these figures do not demonstate such high profits on Stam, as compared to the net profits on the company’s other products, as to be strongly probative on the issue of conspiracy. . As the district court pointed out, for internal accounting reasons, Rohm and Haas assigns a “cost” of ninety percent and a “profit” of ten percent to the by-products sold by its agricultural department whereas, the oomi>any asserts, there is actually no cost in in-oducing a by-product. Accordingy, an accurate reflection of by-product costs would substantially narrow the difference between the average net profit of agricultural chemicals manufactured and sold by Rohm and Haas, and the net profit on Stam. . According to defendants, additional factors of this kind include the product’s uniqueness, it value to the user, the seller’s risk of liability for crop damage, and the amount of customer service required.