These are appeals pressed by a stock brokerage firm and seven individual defendants found guilty on various fraud and conspiracy charges of a fifty count indictment after a protracted jury trial of some fifteen weeks.- Some appellants urge insufficiency of the evidence as grounds for reversal while others claim the case against them was rather weak and inconclusive — thereby magnifying the prejudicial effect of claimed misconduct and excessive zeal on the part of the prosecution and of alleged errors on the part of the Court. There is little choice, therefore, but to delineate in some detail the scheme or schemes charged to the appellants as developed in a trial transcript of approximately 9000 pages.
The only convicted defendant not appealing, Texas-Adams Oil Co., Inc. (Texas-Adams), was the central “glue” in the fraudulent schemes charged by the government. Count 1 was a so-called “wire fraud” count, 18 U.S.C. § 1343; in it defendants Crosby, McCarthy1 and Pet-tit were accused of using interstate telephone wires in furtherance of a plan which resulted in the defendants “purchasing” control of Texas-Adams using that company’s own funds as consideration, thereby defrauding the other stockholders. Counts 2-32 charged Crosby,2 Meredith, Mittelman, Pettit and Texas-Adams with violations of 18 U.S.C. § 1341, the Federal “mail fraud” section. While each count was based on a separate alleged mailing, the “scheme or artifice” charged was the same in all cases. The government generally sought to prove that the defendants had made or knowingly sanctioned fraudulent misrepresentations to the investing public and to brokers and businessmen dealing with Texas-Adams; that defendants had engineered various corporate mergers and purchases, with little or no business substance, whose sole or principal purpose was the flooding of the market with worthless or overpriced Texas-Adams share certificates; and that defendants had fraudulently authorized the issuance of corporate stock without consideration in return for “kickbacks” from the proceeds of the sales of those shares. Further, the Count One “scheme” was incorporated by reference into these mail fraud counts.
The next seventeen counts charge all the aforementioned defendants, save Crosby, plus Goldberg, Gordon, Reicher and Philip Gordon & Co., Inc. with wilfully and knowingly using the mails to sell common stock of Texas-Adams — • which stock was not covered by a regis*932tration certificate filed with the Securities and Exchange Commission, 15 U.S. C.A. §§ 77e(a) (1) and (2), 77x. Many of the mailings concerned a multi-colored brochure, allegedly containing many fraudulent misrepresentations, entitled “Texas-Adams — an industrial empire is under way.” 3 The final count was the ever present conspiracy indictment. In it were accused McCarthy and all the appellants in addition to Joseph Castagna and Guy Gully, acquitted by the jury. Numerous co-conspirators, not defendants, were also named — many of whom testified at the trial. The thrust of the alleged conspiracy was the use of the mails to foster the sale of unregistered securities; all of the appellants were convicted on this count.4
I. The Texas-Adams Story
A. Pre-Natal Activity
Relevant proof adduced by the government at trial actually went back in time several months before the principal “management” defendants obtained control of Texas-Adams. During the summer of 1955 McCarthy and Mittelman were engaged in negotiations with one Claget Sanders to purchase from the Ajax Oil Company concededly valuable oil producing properties in what were known as the Bonanza and Poplar fields in Montana and Wyoming. The price agreed upon was $1,250,000.
It was at about this time, in June, 1955, that the American Montana Oil & Gas Corporation, later used as a dummy buyer in the fraudulent takeover of Texas-Adams, was organized by Mittelman, a New York attorney; however, no organizational meeting was held until October, at which time Crosby, Pettit and McCarthy were elected its officers.
An early linking of appellant Crosby to McCarthy and Mittelman was shown by Basil Zavoico. He was a petroleum consultant approached by Crosby in May of 1955 to obtain engineering reports on the Bonanza and Poplar properties. A month later Crosby again retained him to prepare a statement reflecting the financial structure of a Dumont Airplane & Marine Instrument Co. after an anticipated purchase by that corporation of the Ajax properties. (Zavoico was told that American Montana held an option to purchase from Ajax — and would in turn sell to Dumont.) The terms of the projected Dumont-American Montana “sale” would have netted a $2 million profit for the latter company.5
Apparently lacking funds to swing the Ajax purchase, McCarthy and Mittelman then concluded an agreement, in Denver, with Homestead Oil & Uranium Co. of Colorado. They told Holland, president of Homestead, that they already had control of Dumont — and agreed, in return for a loan of $50,000 to be used as partial *933payment for the Ajax properties, to merge Homestead into Dumont.6 The instrumentality set up for this merger was still another penniless, dummy corporation, Sundown Oil Co. of Colorado, which had as officers and directors McCarthy, Mittelman and Paul R. Jones, a co-conspirator.
Although the record is silent on this point, plans for the acquisition of Dumont must have collapsed sometime that summer. Rather, an agreement was made between one Satiris Fassoulis and McCarthy whereby “groups” headed by those two men were to obtain control of Continental Car-na-var Corp. — at which time Continental would purchase all the stock of Sundown (which was to have acquired the Ajax properties). Holland and Homestead were in turn reassured by an agreement wherein Continental agreed to replace Dumont and merge with the uranium company.
The financial problems of McCarthy and Mittelman during this period are painfully obvious. The Ajax commitment called for 1% million dollars — the sole amount visibly raised was the $50,-000 from Homestead in return for the promise of merger. The “McCarthy group” was to have produced $277,000 to purchase control of Continental in conjunction with Fassoulis; in order partially to meet their commitments in this direction, Mittelman was forced to go to Denver to obtain from Holland release of the same $50,000 which had been banked in Billings, Montana pending closing of the Ajax deal. This money was then forwarded to the New York attorneys of the Limestreet Corporation, from whom McCarthy and Fassoulis were purchasing control of Continental. In the end, both the Ajax and Continental deals fell through for lack of funds— but both of these episodes shed light on the eventual Texas-Adams machinations.
B. An “Industrial Empire” Is Begun
While Crosby’s connections with the foregoing transactions can only tangentially be inferred, he here enters the scene, center stage. In August of 1955 he, with co-conspirator Lawrence Rosen, placed an ad in the Wall Street Journal seeking the controlling interest in a publicly held corporation. Among those responding to the ad were the controlling shareholders of Texas-Adams Oil Co., Inc., defendants Gordon, Goldberg and Reicher. They had incorporated Texas-Adams two years before and had financed it with a public issue exempt from registration under Regulation “A.” The company’s headquarters were in the offices of the brokerage firm, Philip Gordon & Co., Inc. Texas-Adams had been rela*934tively dormant during its short history— only in March of 1955 did it acquire an interest in some Texas oil producing properties which earned about $10,000 in gross income by the time of the sale of control to McCarthy and his associates in late September of that year.
Preliminary negotiations were held among Crosby, McCarthy and Goldberg and agreement was reached on the sale of 283,000 shares of Texas-Adams for $105,000 — $45,000 at closing, $20,000 in sixty days and $40,000 in ninety days; the lion’s share of the stock certificates was to be escrowed pending payment on the time notes.
Closing of the transaction was twice postponed because of the buyers’ financing difficulties, the second postponement being accompanied by a $1,000 check, on September 25, 1955, to evidence the continuing good faith of the McCarthy group. Defendant Pettit had by that time joined the purchasing group as that check was traced from his account in a Salt Lake City bank.
The closing was finally consummated on September 27, 1955. The buyers’ arrangements for “financing” were most interesting. Crosby and McCarthy approached Joseph Spiotta, a New Jersey money lender, seeking $40,300 to meet the closing installment of $45,000. He agreed to furnish that sum on the strength of a representation that Texas-Adams would use the City National Bank & Trust Co. of Hackensack, New Jersey (where Spiotta was a director) as a repository for funds, and on the strength of a guarantee that Spiotta’s account in that bank would be credited with $42,000 before he ever parted with his money.
The Texas-Adams bank account, some $60,000, had been kept with the Underwriters Trust Company in New York. When the buyers represented that they were going to move that account to the New Jersey bank, the sellers agreed to switch the funds, prior to closing, to the Chemical Com Exchange Bank, scene of the closing and one of two New York banks serving as correspondent to the City National in Hackensack.
On September 26th, the buyers opened three new accounts in Hackensack in the names of Sundown, American Montana and Texas-Adams; a token deposit of $100 was placed in each account. Proper corporate resolutions and signature cards for all the companies were filed with the bank and the corporate seal of Texas-Adams was exhibited and used for that purpose. At the same time a set of checks and deposit slips was prepared reflecting a series of transfers to take place the next day. $51,700 was to be transferred from the Texas-Adams account at Chemical Corn to its City National account. Two Texas-Adams checks totaling $51,250 would then transfer funds to Sundown and American Montana ; almost all of the amount deposited in Sundown’s favor would be immediately credited to American Montana, the putative purchaser of Texas-Adams. Finally, an American Montana check for $42,000 was made out to Spiotta’s account to “reimburse” that gentleman.
Next morning, the day of the closing (but before Texas-Adams changed hands), Crosby attempted to make a wire transfer to New Jersey. A Chemical Corn official refused to make that transfer because the resolutions and signature cards gave authority only to the “old” management. Crosby quickly cured that defect by executing new resolutions and signature cards — once again using the Texas-Adams seal. The money was wired to the City National and Spiotta, after personally checking by telephone to make sure his account was $42,000 richer, “loaned” American Montana the funds necessary for the purchase.
Only $42,000 of Texas-Adams funds was actually used in the purchase of control of the corporation; $9,250 remained in the American Montana account in Hackensack. This money too was quickly siphoned off, however. Pettit, whose $4,700 with Spiotta’s $40,300 had made up the purchase price, was handsomely rewarded in the form of two American Montana checks, a total of $6,700, drawn by Crosby and credited to Pettit’s Salt Lake City account. The remaining *935$2,550 Crosby paid to himself through a dummy and co-conspirator, George Jannis, about whom we shall hear more later. Thus the “buyers” successfully contrived not only to purchase Texas-Adams with its own money, but also enriched themselves, from the till, for their labors.
C. Dynastic Operations
It is indeed difficult to telescope the fantastic activity of Texas-Adams and its controlling proprietors during the ten months or so after acquisition by McCarthy, Crosby and company in September of 1955. Texas-Adams’ outstanding stock ballooned, during that period, from 750,000 to almost ten million shares; not one share, however, was at any time the subject of a registration statement filed with the S.E.C.
Many of the deals sought by Texas-Adams never reached fruition, usually for lack of necessary funds. Two of these were the so-called “Longview, Texas” acquisition and the Taylorcraft merger. An option to purchase the Long-view properties (oil) for $275,000 was entered into in the name of one of the ever available corporate dummies, Sundown Oil Company of California. The search for this sum of money led McCarthy and Pettit to Alabama in November, 1955. There Paul Jones and Sterling Graham, a Pittsburgh broker and also a co-conspirator, were engaged- in underwriting public issues of stock on two “intrastate” corporations, Warrior Products Company and Gypsy Mining Company, under the watchful eye of the Alabama Securities Commission. These corporations apparently were organized to buy some of Jones’ seemingly boundless supply of “mining properties.” Graham told McCarthy that the Alabama commission would not allow a Texas-Adams merger with either of the local corporations and suggested a merger between Texas-Adams and Taylorcraft, a Pennsylvania manufacturer of light aircraft into which Warrior was to have been merged.
Taylorcraft, which was experiencing ready cash difficulties of its own, was not the answer to Texas-Adams’ need for cash to close the Longview oil deal.7 At this point, Jones came up with another, more suitable deal. He suggested that Texas-Adams buy from the Osage Mining Co. (another corporate repository for some of the Jones “properties”) a lease on a feldspar mine in Spruce Pine, North Carolina for 375,000 freely transferable Texas-Adams shares. Jones would “accept” only the proceeds from the sale of 75.000 of those shares while Texas-Adams would receive, for its use in Texas, the proceeds from public subscription of the remaining block.
McCarthy not only accepted this offer, but, in implementing it, went a long step farther. The actual purchase of the mining lease was made in the name of American Montana — which company agreed to resell the feldspar mine to Texas-Adams for 650,000 shares plus a cash payment of $75,000. Much of the rest of the Osage story is too obscure and complicated to be drawn out in detail here. It is enough to say, however, that the sale, kickback and resale procedure was used as a vehicle to sell to the public hundreds of thousands of unregistered Texas-Adams shares; the same certificate for 375.000 shares was caused to be broken down and issued by two different stock transfer companies; the feldspar mine in question was inactive, flooded with water, and worthless; and, finally, Texas-Adams never bothered even to take title on the lease.
While the Osage transaction is perhaps the most artful example of the McCarthy *936stewardship of Texas-Adams, passing mention may be made of other corporate depredations. More than 70,000 shares of stock were issued, without consideration, to George Jannis, a chauffeur, valet and handyman of Crosby’s, for imaginary loans and services rendered. These shares were cleared for trading by defendant Mittelman’s opinion letters and the proceeds, after token payments to Jannis, went to enrich appellant Crosby at least.
Mergers were accomplished with the Empire Drilling Company and the United States Sulphur Company. In the former transaction a purchase of Empire stock was misrepresented as a purchase of all the assets of that company.8 Further, machinery was set up whereby creditors and “interest holders” in wells being developed by Empire could trade their interests in Empire for Texas-Adams stock. All told, 3,100,000 shares of Texas-Adams were issued for these purposes. Aside from the fact that the U. S. Sulphur properties appear to have been worthless, the government proved no real infirmity in that transaction. On top of the quarter of a million shares actually used to buy out the United States Sulphur Company, however, an additional 150,000 shares were issued under fraudulent circumstances, purportedly in connection with the Sulphur Company deal, to L. W. Ltd., a Canadian dummy corporation which Mittelman “served” as vice-president.
The final notable transaction was the so-called Comigu deal, in which Texas-Adams bought, for two million shares, twenty prospecting permits on supposed tantalum mining properties on the bank of the Sinn Marie River in French Guiana. The seller was acquitted defendant Guy Gully and, like Osage, the transaction apparently involved a substantial kickback of shares which were then sold to the public through Castagna, McCarthy’s young brother-in-law.9 Once again, as in the case of Osage, Texas-Adams never bothered to take title to the permits.
Although the aforementioned dealings were highly effective in keeping a steady flow of Texas-Adams shares entering the market, it was a key to the success of these schemes to maintain the price of those shares at a fairly high level. This was accomplished by a high pressure public relations campaign coupled with the use of patently fraudulent misrepresentations in the company’s public reports. Within a week of the September 27, 1955 takeover of Texas-Adams, Crosby and McCarthy contacted co-conspirators W. Ware Lynch and Russell Birdwell Associates, Inc. to start rolling a publicity campaign.
Lynch obtained newspaper coverage on the change in Texas-Adams control and management; the McCarthy name was fully exploited.10 Later, there was extensive reporting of the Empire merger which stressed tremendous growth in the assets of Texas-Adams to a false figure of more than nine million dollars. In November of 1955, a “Report to the Stockholders” was sent to brokers dealing in Texas-Adams. Among major misrepresentations there made were statements that prior to the new management Texas-Adams had a yearly income of $35,000; that acquisition of the “Bonanza Field in Montana” had been accomplished, with resultant additions to income of $33,-000-$34,000 per month; that the new management had acquired, for $300,000 and 850,000 shares of stock, some east Texas oil acreage with $24,000 net monthly income and $4 million proven reserves in each of five well locations; that negotiations had been completed to take over *937Osage, replete with “300,000 in cash and a producing feldspar mine * * * [and] valuable samples of uranium and mica”; and that Homestead, when merged, possessed cash totaling $140,000.
Many of the foregoing — and other— misstatements of fact were made more specifically in a series of Texas-Adams pro forma balance sheets issued with the aid of Herman Lubin, an accountant and co-conspirator, in February and March of 1956. What the original report to the stockholders lacked from a purely artistic standpoint was amply supplied by the production and dissemination of an attractive, pictorialized brochure which substantially mimed the bald misstatements of fact and gross overstatements of assets contained in the earlier literature.
Such methods were generally successful, in conjunction with active promotion by various brokers in New York, Pittsburgh and Denver, in keeping Texas-Adams common priced at an artificially high level. The government adduced proof, however, that in at least one instance more direct market manipulation was necessary. Mass unloading of stock by the former creditors and interest holders of Empire was threatening, in February of 1956, to deflate seriously the market in Texas-Adams. In response to this danger, those shareholders were urged to sell back to the corporation at a guaranteed rate of one dollar per share. To support this corporate extravagance, brokers in turn were induced to buy the shares from Texas-Adams and pass them on to the public at that price. The brokers were encouraged to enter into these dealings by an offer of two bonus shares of Texas-Adams for every share bought at an artificial level; 200,000 shares emanating from the Comigu-produced “gravy” were placed in escrow to provide this bonus stock.
II. Sufficiency of the Evidence
A. The “Management” Defendants
While Crosby, Pettit, Meredith and Mittelman do not dispute the fact that the evidence against them properly warranted submission of the case to the jury, they all argue that the record is equivocal as to their guilt — and that the factual issues involved are such close ones that even minor errors at trial would be deserving of a reversal by this court and a remand for a new trial. We disagree.
The proof against Crosby on the Count One scheme is overwhelming. His counsel would have us believe there was a genuine jury issue as to whether Crosby was in “good faith” in this transaction— i. e., whether the defendant considered the raiding of the Texas-Adams treasury a “loan” to American Montana. Ignoring the evidence that Crosby immediately caused to be drained the $9,000 not used in the purchase of control, and even if the record failed to indicate that American Montana not only was penniless but also had “no visible means of support,” only an incredibly naive jury could be expected to credit such a story. His culpability on the mail fraud and conspiracy counts is similarly evident. The feeble contention that the Jannis assaults on the investing public were in legitimate pursuit of a deserved “finder’s fee” is hardly worth mentioning.
Pettit and Meredith both challenge the degree of actual knowing and fraudulent participation proved by the government. Meredith, the only defendant who took the stand in his own behalf, attempted to define his role in the scheme as that of a mere errand boy. Both were experienced businessmen before they came east for their alliance with McCarthy; yet both would have us believe — and claim the jury might easily have found — they were credulous schoolboys completely taken in by the impressive, free-wheeling Texas oil “operator,” McCarthy. Apart from the substantial incredibility inherent in these claims, there was more than ample independent evidence supporting scienter on the part of both defendants.
Pettit received back more than his original investment, from the Texas-Adams treasury, within two days of the takeover of control; a later payment, from proceeds of sale of Osage stock, was made to a Kathrine McDavitt, a lady friend of *938Pettit. A number of witnesses testified either that Pettit misrepresented facts11 or was present and participating at conferences where McCarthy did the same. Meredith’s activity on behalf of Texas-Adams was astounding; he was constantly back and forth all over the country picking up and delivering stock certificates, money and instructions. He signed innumerable checks and documents, and testimony indicated he was active in Denver in the trading of Texas-Adams stock for Empire interest certificates. While the heavy weight of the evidence would seem to indicate he was a “follower” rather than a “leader,” this is a far cry from raising a doubt as to lack of criminal knowledge.
Mittelman, both at trial and here on appeal, has attempted the Herculean task of arguing away every single bit of damaging evidence produced by the government. We have carefully considered his extended discussions of the trial evidence but cannot agree there was any real doubt as to his guilt. Any conclusion of Mittelman’s innocence must rest upon the hypothesis that Holland and Carroll, the moving parties in Homestead Oil & Uranium, plus Kimmes, a Denver broker and co-conspirator, perjured themselves; that Harrison Miller, a New York broker, was mistaken as to dates when he testified that Mittelman vouched for Jannis in mid-November of 1955; that Purcell, an experienced securities attorney, was mistaken in his testimony concerning Mittelman and the Empire transaction; and that Mittelman was unbelievably homswoggled by McCarthy due to an incredible naivete on the part of an experienced attorney.
It is true that some apparently daifming evidence has been convincingly explained away. The overall involvement, however, is clearly shown. The Jannis, Empire and Comigu opinion letters, asserting the free transferability of those shares, are not adequately explained. Long after he concededly knew the truth about Jannis, Mittelman sold 5,000 shares of unregistered stock in Jannis’ name for the account of his sister. Defendant’s letter of April 30,1956 to the stock transfer company stating that 150,000 shares of Texas-Adams being issued to L. W. Ltd., a dummy corporation of which Mittelman was vice-president, was for part payment of the purchase price of United States Sulphur Company is particularly damning. Despite appellant’s meticulous analysis of the evidence, it is hard to see how the jury could have found Mittelman anything but guilty.12
B. The “Broker” Defendants
Gordon, Goldberg and Reieher, the stockholders in Philip Gordon & Co., Inc., all contend the proof against them was insufficient to warrant submission to the jury. Guilty verdicts on the three substantive counts and the conspiracy count from which they appeal all involve use of the mails to effect wilful and knowing sale of unregistered securities. The proof is clear that no Texas-Adams stock was ever registered—and that defendants knew that the particular certificates involved in Counts 46-8 were not registered. There were, however, opinion letters claiming the shares to be exempt from registration. Two basic elements thus remain; Were the shares exempt? If not, does the record adequately prove the necessary knowledge on the part of *939these defendants to support the element of wilfulness? We hold that, while the shares should have been registered, there was insufficient proof from which the jury legitimately could have inferred wilfulness and criminal intent.
All the stock involved in the three substantive counts here involved stems from the two million shares issued to Gully in connection with the Comigu transaction. Philip Gordon & Co., Inc. received them for sale from Castagna. The Mittelman legal opinion letters concerning these shares asserted the stock was exempt from registration and free for trading; the opinion was based on S.E.C. Rule 133 which provides that for purposes of the criminal statute dealing with sale of unregistered securities no “sale” shall be deemed to have been made “so far as the stockholders of a corporation are concerned where * * * a plan or agreement for a statutory merger or consolidation or reclassification of securities, or a proposal for the transfer of assets of such' corporation to another person in consideration of the issuance of securities” is approved by a binding majority of the stockholders. This exemption, which “has permitted the offering of literally billions of securities without registration” 13 probably fits the circumstances of the Comigu transaction; this court has specifically held, however, that the Rule 133 exemption applies only to the original exchange in which stock is transferred to the merged corporation’s stockholders - — and not to later resales if they otherwise are within the registration requirement. Securities and Exchange Commission v. Culpepper, 2 Cir., 1959, 270 F.2d 241, 247-248; see also Securities and Exchange Commission v. Mono-Kearsage Consol. Mining Co., D.C.D.Utah 1958, 167 F.Supp. 248.
Similarly, none of the statutory exemptions are here applicable.14 Philip Gordon & Co., Inc., and the individual defendants are clearly “dealers” within the purview of the Act;15 their sales would have been exempt transactions, therefore, only if made more than forty days “after the first date upon which the security was bona fide offered to the public.” The purchases of these shares by Gordon & Co. were from Castagna, nominee for controlling interests of Texas-Adams, the issuer. The Act defines Castagna himself as an “issuer” under those circumstances — which fact, in turn, made *940the broker defendants technically “underwriters.” 16 Thus, by recourse to the definitional sections of the Act, we can circuitously, but inevitably, reach the otherwise immediately apparent truth- — • that the sales by Philip Gordon & Co., Inc. were the first public offerings of those shares, thereby rendering inapplicable the “forty day” exemption. Finally, since the defendants bought and sold the stock for their own accounts, the exemption accorded “brokers’ transactions” is unavailable.
Since all of the government’s evidence tending to tie the brokers into the overall conspiracy is relevant, in some degree, on the substantive counts for sale of unregistered securities,17 the sufficiency of the evidence on both types of offenses may be treated together. Reieher was a 50% stockholder of the brokerage house, Gordon and Goldberg evenly dividing the remaining one half interest. Philip Gordon & Co., Inc. was not a big house and all three appellants were active in its management; Reicher and Goldberg supervised the sales force.
Chronologically, the government points first to the brokers’ role in the sale of Texas-Adams to the McCarthy group. Although the proof failed to show directly that these defendants aided or knew of the arrangements by which they were paid out of the treasury of Texas-Adams, much is made of the fact that the buyers apparently used, two days runlling, the corporate seal in furtherance of their fraud. A Gordon & Co. employee testified that the seal was ordinarily kept in the office safe but he could not vouch for its presence at any one particular time. The attorney who handled the sale for the brokers could not remember whether he had given or loaned the seal to the buyers.
The second installment of $20,000 on the purchase price of Texas-Adams was raised through the sale of a large bloc of shares, for investment, to one Jack Baker. Part of this bloc, a certificate for 50,000 shares, was sold to Baker by Philip Gordon & Co. for the account of American Montana and the actual payment of the $20,000 was made directly from Baker to Goldberg.18
Other evidence relied upon by the government includes the fact that Goldberg sanctioned the cashing of one check payable to Jannis although such was not the usual practice of the house; that Goldberg, in a hotel meeting with Kimmes and the chief officers of Texas-Adams, “suggested” a broker bonus arrangement to help stabilize the price of Texas-Adams shares near the one dollar mark; that Goldberg and Gordon questioned a salesman for a rival brokerage firm concerning the latter’s source of “cheap shares” and that the salesman remarked that he smelled something “rotten”; and that the brokers, in pushing the sales of Texas-Adams stock, disseminated fraudulent literature put out by the alleged co-con*941spiratorial accountant and public relations firm. Finally, the government presses upon us the more than 500,000 shares sold to the public by Philip Gordon & Co., Inc. under various opinion letters—shares emanating from almost all of the various “transactions” of Texas-Adams—and insists that the extent to which the defendants did rely, or were entitled to rely, on the opinion of counsel was for the jury to determine.
The course of conduct of the three brokers in selling so many shares proves little19—it merely presents, in multiple instances, the same question posed by the three substantive counts here in issue: Were the brokers justified in selling unregistered stock on the representation of counsel that such shares or transactions were exempt from the registration requirement?
The evidence that the buyers used a Texas-Adams seal in implementation of the takeover fraud is far too weak a base from which to infer guilty knowledge on the part of these defendants. The gaps are too broad and substantial. To hold the defendants on this evidence, it must be inferred that it was “the” seal, rather than a facsimile, that was used by Crosby; that the “defendants” (or one of them, with the blessing of the others) gave the seal over to the buyers; and that it was done with the intent to aid and abet the fraud. The government’s evidence concerning the second installment payment through the sale of 50,000 shares of Texas-Adams held by American Montana is similarly weak and insubstantial. Appellee argues that the brokers “knew” Texas-Adams legitimately could not have declared such a large stock dividend to controlling American Montana only a few weeks after acquisition— and that American Montana would have paid the brokers directly if it had funds. The fallacies in this reasoning are clear. There are numerous transactions which might have been the base for legitimate issuance of this stock to the parent corporation ; to state only the most obvious, American Montana might have transferred property, which they were loath to liquidate, to Texas-Adams as a means of obtaining stock, in turn to obtain funds, to pay the brokers.
Since the broker appellants were a principal outlet for over the counter sales of Texas-Adams stock, there is nothing strange about “hotel meetings” with the top officers of the corporation. The two meetings testified to prove little. The testimony concerning the first of those occasions shows only a concern on the part of the defendants that the market in Texas-Adams, in which they had considerable personal stake,20 was being seriously undercut by “cheap stock” from unknown (to them) sources. The effect of Goldberg’s suggestion of a bonus share arrangement at the latter meeting is compromised by the fact that Gordon & Co. bought or sold none of the Empire stock in question and received none of the bonus shares.21 Moreover, there was evidence that in this area of speculative mining, oil and uranium stocks in the over the counter market, the selling price of a security almost never has any relation to its intrinsic value and that substantial commission and bonus arrangements are the rule rather than the exception.22
*942Concerning the mailing of the fraudulent literature furnished by McCarthy et al., the record shows that it was standard procedure in the over the counter market for mailings to emanate from the brokers rather than the corporation. Jealous guarding of customer lists by the brokerage houses caused this practice. The claim that the defendants, as ex-controllers of Texas-Adams, should have known of the falsity of one assertion in the “Report to the Stockholders” is more serious. We believe, however, the single sentence in question was sufficiently ambiguous so that defendants were not required, at the peril of criminal liability, to challenge its accuracy.
As to the substantive counts alone, evidence is similarly lacking to refute appellants’ apparent reliance on the representations of counsel. It is key here that the government nowhere showed that the brokers knew the details of the alleged transfer from Gully to Castagna— or that they knew the latter’s relation to and probable domination by McCarthy.23 Thus there was no evidence that the brokers knew or should have known they were “underwriters” in terms of the Securities Act. Mittelman’s original opinion letter to Philip Gordon & Co. concerning the Gully stock relied solely on Rule 133’s special no-registration treatment for statutory mergers, etc.; his later letter to the brokers’ counsel, however, stressed a combination of an original Rule 133 exemption and subsequent freedom to trade which allegedly arose because they were no longer “control stocks.” While such a theory is probably invalid in light of this court’s decision in the Culpepper case, supra, we cannot say that on its face the legal opinions tendered were so patently erroneous as to permit the jury to speculate on the good faith of the defendants.
It is always a difficult task in a complicated omnibus conspiracy case like this one to decide at what point the government has produced enough evidence against individual defendants to put their fates in the hands of the jury. Because of the tremendous potential for collateral prejudice inherent in these cases, however, the courts must be extremely scrupulous in sorting out the evidence against single defendants and weighing the same —divorced from the all too often damning influence of the background conspiracy. The case against these three defendants presents a peculiarly difficult problem in this regard because the prime issue is one of guilty knowledge and criminal intent. It is undisputed that the brokers’ acts in fact furthered the conspiracy; the claim is persuasively made, however, that they were merely “doing business as usual” and, to their best knowledge, according to acceptable standards of that business. The statutory and administrative regulatory scheme over this area is far from a model of clarity; such a situation is aggravated when, as here, criminal liability is based on the failure to comply with the law. We think it is all too facile an answer for the government to rely on the experience and supposed expertise of these brokers; the fact is that they purported to rely on opinion letters of an attorney, whose expertise is presumed to be even greater. The scanty extrinsic evidence produced by the government against these defendants implies guilt only if we first assume guilty knowledge and purpose; when used to prove that basic element of the *943crime it is neither substantial nor convincing. The trial court erred in not directing a verdict of acquittal on all counts for defendants Gordon, Goldberg, Reicher, and Philip Gordon & Co., Inc.
III. The Conduct of the Trial
The four remaining appellants have advanced to this court a large number and wide variety of alleged errors which, they claim, warrant reversal of their convictions. Although almost all of the appellants complain of what they characterize as a prolix, rambling and disconnected presentation of the government’s case, only Pettit and Mittelman strongly urge upon us the contention that they were unduly harmed by a combination of poor and confusing presentation by the government and prejudicial conduct on the part of the trial court.
As noted, the trial was a long one of three and one half months. Our own sketchy and incomplete statement of the facts gives some indication of the breadth and complexity of the schemes submitted to the jury for digestion. While the proof, as organized and offered by the government, may not have met some hypothetical concept of perfection, the presentation of the case was, on the whole, fairly orderly and logical. Difficulties presented by the very scope of the alleged frauds and by reason of the fact that the testimony of some witnesses was relevant to more than one aspect of the case were inevitably multiplied by problems, especially troublesome in a long trial with many cross-examining defendants, concerning the ready availability and procurement of witnesses.24 Upon our reading of the record, we find that any direct or veiled charge of actual misconduct leveled against the government’s attorneys is wholly without substance.
Complaints about the trial court’s general handling of the case stem primarily from the same source of difficulty. Appellants tell us that “everything” was let in “subject to connection”; that in his charge Judge MacMahon made a few factual errors and generally favored the government in his summation of the relevant facts; and, finally, that the charge which took some eight hours to deliver was too long and confusing. Some specific claims of error will be discussed, infra; it here suffices to say merely that there is, practically, little a trial court can do in these cases except take large portions of evidence subject to connection. Moreover, there was presented in the instant case overwhelming independent evidence connecting each of the remaining appellants to the alleged frauds and to the conspiracy. The factual charge of the court was, in general, amply supported by the record; any small errors made are explained by the long and complicated nature of the case—and neutralized by the court’s cautionary instruction to the jury.
The attack on the length of the charge is frivolous. Just as this type of trial, almost of necessity, must fall short of perfection in clearly and concisely presenting all the issues to the jury, the court’s charge is inevitably similarly infirm. It may be true that the concentration span of a “real world” jury is, like that of any other twelve humans, a limited one. So long, however, as our appellate practice places so much weight on the hypothetical effect of a judge’s charge, see dissenting opinion of Judge Frank in United States v. Farina, 2 Cir., 1950, 184 F.2d 18, 21, extended instructions to the jury in many cases are a necessity.25 That the instant case, with a transcript approaching ten thousand pages and nine individual defendants accused on various charges of a fifty count indictment, falls within the *944aforementioned category is only too evident.
Examination of the record shows that the charge of prejudicial conduct by the lower court is without foundation. Defendant Mittelman particularly complains of alleged unfair limitation of his cross-examination and summation. Administering a trial involving nine separate defendants and their attendant counsel is at best a trying task. Further, it here appears that the attorney for Mr. Mittelman had had consistent difficulty in marshalling his relevant facts and in conducting non-repetitious questioning of the witnesses. Under the circumstances, the trial judge was acting completely within his discretion and in furtherance of his duty to expedite the trial when he cut off Mittelman’s attorney after he had gone more than a half hour over his self-requested five hour summation period. See United States v. Campisi, 2 Cir., 1961, 292 F.2d 811.
IV. Attacks on the Indictment
Appellants’ general complaint that the indictment was unforgivably prolix and a “pleading monstrosity” is closely akin to their objections to the trial presentation by the prosecution and the judge’s lengthy charge. The government can hardly be faulted for prolixity when it appears that the very complexity and deviousness of the alleged swindles are the basic causes of confusion. Just as the ingenuity of the white collar criminal and quasi-criminal is mirrored in the complicated regulation of the securities industry,26 the indictment in this case is a rather faithful picture of the activity charged to the defendants. Objection to use of the “incorporation by reference” device is merely a quibble; surely there is enough governmental waste without requiring complete reproduction, from count to count, of all allegedly relevant pleadings previously enumerated.
More specific assaults on the indictment are made. Mittelman argues that Counts One and Two are repugnant because the takeover scheme is incorporated by reference into the later count while Count One does not name him as a defendant; his alternative argument is that if there is no repugnancy then Count Two is of necessity duplicitous — and that his conviction on that count must fail on any account. Closely related to appellant Mittelman’s attack on the first two counts is the argument made by all defendants that there was a fatal variance between the pleading and proof on the conspiracy count. Relying on Kotteakos v. United States, 1946, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1557, they assert that the government proof tended to establish not one scheme, but a multiplicity of disjointed, unrelated criminal ventures — at most, a fatally amorphous “conspiracy to do evil.” In support of this argument, they magnify the fact that some of the alleged parts of the “scheme” did not accrue directly to personal interests of all the defendants.
Appellants’ reliance on Kotteakos is misplaced. The Supreme Court there decided only that a variance in the proof was prejudicial. That multiple conspiracies rather than one had been proven was conceded by the government. In that case one defendant had engaged in a number of quite isolated transactions, each with a different co-defendant, to obtain loans from government agencies on false representations. The Court of Appeals, sub nom. United States v. Lekacos, 2 Cir., 1945, 151 F.2d 170, 173, aptly analogized the case to a fence who disposes of goods stolen by a number of independent persons; the fact that they know him to be a fence does not mean all conspire together.
The proof in the instant case reveals a far more integrated scheme. The government argues correctly that the evidence establishes, basically, a central plan to effect the issuance and sale of worth*945less or grossly overpriced unregistered corporate securities to an uninformed and deluded investing public. The implementation of corporate “deals” and mergers as a cover for the fraud was an almost constant factor; the use of both kickbacks and resales by compliant dummy corporations—plus the issuance of stock for falsely declared consideration— reflects a relatively consistent, if varied, pattern of operation. Recurrent personal misrepresentations dovetailed neatly with the fraudulent publicity campaign to create a false public image of Texas-Adams. Mittelman’s opinion letters consistently supplied still another necessary element in the success of the fraud. The non-broker defendants indisputably worked together in the Texas-Adams venture; indeed, the only possible questions raised at trial concerned the extent to which certain of these defendants either were duped by or blindly followed the others.
A mail fraud scheme consisting of a large number of similar transactions, conducted by a constant nucleus with differing subordinates, was held basically a single conspiracy in United States v. Cohen, 2 Cir., 1945, 145 F.2d 82, 87-88. We have also declared with reference to determining whether the proof is of one or two conspiracies that when a conspirator “embarks upon a criminal venture of indefinite outline, he takes his chances as to its content and membership, so be it that they fall within the common purposes as he understands them.” United States v. Andolschek, 2 Cir., 1944, 142 F.2d 503, 507. It is certainly no problem that not all the details were formulated at the outset; “A single scheme to defraud may involve a multiplicity of ways and means of action and procedure. It may be such that the complete execution of it would involve the commission of more than one criminal offense. Mere details may be changed and the scheme remain the same.” Mansfield v. United States, 5 Cir., 1946, 155 F.2d 952, 953. Whether a scheme is one conspiracy or several is primarily a jury question, since it is a question of fact as to the nature of the agreement. There was ample evidence here to justify a finding that the non-broker defendants were a part of an overall scheme to acquire and exploit a public corporation—for the ultimate purpose of selling to the public unregistered common stock. See Developments— Criminal Conspiracy, 72 Harv.L.Rev. 920, 929-30, 933 (1959); Note Federal Treatment of Multiple Conspiracies, 57 Colum. L.Rev. 387, 393 (1957).
Mittelman’s claim of duplicity in Count Two founders on the very nature of the mail fraud offense—with its attendant “scheme or artifice,” 18 U.S.C. § 1341. The government, in presenting its case, must of necessity offer evidence as to the acts of the defendants which make up the relevant scheme. Aside from the Count One takeover transaction, the government’s proof in this case showed (together with acts which were, aside from the scheme, innocent) any number of separate “crimes” under state or federal law. It is difficult to see how the common law duplicity doctrine can retain any meaning or vigor within the context of this particular statutory offense.27 As long as the defendants clearly are “charged” with only one offense (i. e., the mailing in furtherance of the scheme), the count is not duplicitous. See Owens v. United States, 5 Cir., 1955, 221 F.2d 351; Weiss v. United States, 5 Cir., 1941, 122 F.2d 675, 680; United States v. McAlpine, 7 Cir., 1942, 129 F.2d 737.
The fact that Mittelman was not named as a defendant in Count One—■ while the takeover fraud was pleaded as part of the Count Two scheme—does not make these two counts repugnant. The takeover scheme charged in the first count was a localized, short lived offense completed in September of 1955. The “grand *946scheme” charged in Counts 2-32 was of much longer duration and encompassed a much broader scope of activity—including the acquisition of control of a publicly held corporation. The fact that the government did not charge Mittelman with direct criminal responsibility, as a principal or otherwise, for the takeover of Texas-Adams does not prevent it from accusing him of knowing participation in the larger scheme, which took in many acts for which he may not have been directly responsible.
V. Rulings on Evidence
Considering the length and difficult nature of the trial, appellants press surprisingly few claims of error concerning disputed rulings on evidence. Mittelman complains that evidence of the takeover fraud was erroneously admitted against himself and Meredith who were not named in the first count; he further claims error in the admission of the testimony of Holland, Carroll, Sanders and Zavoieo concerning the period of the Homestead-Ajax negotiations before the McCarthy group gained control of Texas-Adams. Pettit asserts the court erred in allowing in, against himself, a letter addressed to Jannis by Crosby—and also claims that evidence concerning money paid by Texas-Adams to the Worth Investment Co., of which Pettit was an officer, was inadmissible.
Mittelman’s argument concerning the evidence of the fraud in the acquisition of control of Texas-Adams is disposed of by our prior discussion of the interrelationship of Counts One and Two. The pre-Texas-Adams transactions, although not pleaded in the indictment, were relevant in several respects. The evidence showed an early connection not only with McCarthy but also with Crosby; it indicated the known falsity of one of Mittelman’s opinion letters written to cover stock issued to Jannis; it was relevant to the question of Mittelman’s knowledge of the falsity of the representation, in the Texas-Adams “Report to the Stockholders,” that Homestead had a $140,000 treasury at the time of the merger. In some aspects the testimony, especially a charge by Holland that Mittelman had apparently failed to return $19,000 of Homestead funds, was quite prejudicial and might have been excluded; as it was relevant to important aspects of the case, however, the court acted within its discretion in admitting it.
Pettit’s objection to the CrosbyJannis letter, which cited Crosby, McCarthy, Mittelman and Pettit as principals for whom Jannis had acted as nominee, is without merit. When admitted, although he allowed it only against Crosby, Judge MacMahon rather ambiguously told the jury that later on, if the conspiracy was sufficiently established, it could be considered against other defendants. Defendant did not object to this. Any possible prejudice in connection with the original instruction was cured in the court’s charge, wherein the jury was specifically instructed that “it is evidence against Crosby only and may not be considered by you against any other defendant.”
The Worth Investment Co. was a corporation organized in Colorado by McCarthy, Pettit and Thomas Hudson, the Denver attorney and co-conspirator who most of the present defendants contend was, with McCarthy, the moving force in Texas-Adams. Worth was organized to finance conditional sales contracts for a vending machine company in which Hudson was somehow interested. The damaging aspect of this evidence was that $50,000 of Texas-Adams funds were sent west to finance the Worth Investment operations. The Worth Investment Co. episode was not pleaded in the indictment and Pettit claims it was wholly irrelevant and prejudicial as proof of another crime.
The evidence was properly admitted. Proof of personal use of Texas-Adams funds was highly relevant to the issue of criminal intent and knowing complicity in the fraud and conspiracy. It is somewhat ironic that while Pettit spends a considerable portion of his brief arguing the alleged fact that there is no evidence that he personally gained at the *947expense of Texas-Adams shareholders, he at the same time asserts that evidence showing use of $50,000 of Texas-Adams funds to finance outside personal business interests is irrelevant.
YI. The Charge to the Jury
The challenged portions of the charge include that concerning the Worth Investment Co. and the instructions on the proper weight to be given Mittelman’s character evidence.
After reciting to the jury its recollection of the facts surrounding the Worth Investment episode, the court instructed the jury thus:
“If you find, beyond a reasonable doubt, that one of the defendants, while an officer or director of Texas-Adams, caused or planned to cause, money, stock or other property belonging to the corporation to be paid to him, to his fellow officers or directors, or to a private corporation belonging to him or them with intent to benefit himself, and with knowledge that the corporation was not receiving, or would not receive fair return for what it would pay out, you may conclude that he violated his trust and committed a fraud.”
As has been noted, the indictment did not specifically charge any of the defendants with criminal liability on account of the Worth transaction; the evidence was properly admitted as relevant to criminal intent. Upon this formulation, we think the trial court’s charge was somewhat ambiguous in failing to delineate clearly the thrust and purpose of the Worth Investment evidence. The jurors should have been instructed that if they found the circumstances of the $50,000 transfer to Worth to have been fraudulent, they could then weigh that fact when considering the good or bad faith of the defendants involved concerning the actual crimes specifically charged in the indictment.
The lower court’s failure to clarify completely the role of the Worth evidence cannot, however, be deemed cause for reversal. The infirmity pointed to is encompassed in one sentence of an eight hour charge. It is important to note that the jury was not told it could convict on the basis of the Worth evidence — merely that “(y)ou may conclude that he violated his trust and committed a fraud,” a proposition which is technically correct. Elsewhere in its charge, the court spelled out in detail the necessary elements for conviction on the mail fraud counts — presence of the scheme to defraud, the need to establish the complicity of each defendant with independent evidence, and the use of the mails. The court was similarly thorough on the Securities Act counts and the conspiracy. The error here was one merely of form and cannot be deemed prejudicial.
Appellant Mittelman’s attack on the charge as to character evidence, noted in the margin,28 is without merit. Whatever the validity of the traditional distinction, the court correctly instructed the jury on the role of character evidence as going to prove community attitudes *948rather than personal belief. Finally, the appellant attempts to isolate a single sentence of the charge as incorrectly stating the law: “But if from all the evidence you are satisfied beyond a reasonable doubt that Mittelman is guilty, a showing that he previously enjoyed a reputation of good character does not justify or excuse the offense, and you should not acquit him merely because you believe he has been a person of good repute.” When read alone, this sentence would appear to be a questionable statement of the role of character evidence in the jury’s deliberation and might call for reversal. Searway v. United States, 8 Cir., 1910, 184 F. 716; United States v. Klass, 3 Cir., 1948, 166 F.2d 373. In the paragraph immediately preceding this statement, however, the jury was clearly instructed that it should consider evidence of good character, together with all other evidence, in determining whether guilt had been proven beyond a reasonable doubt. The court further specifically told the jury that evidence of good character, alone, may create a reasonable doubt. In this context, it is clear that the court’s reference to “all the evidence” in the challenged sentence included the evidence of good reputation—and that appellant’s cited cases, supra, are inapplicable.
VII. McCarthy’s Guilty Plea
All of the appellants assert the trial court erred prejudically in failing to grant motions for a mistrial when co-defendant McCarthy pleaded guilty toward the end of the first week of trial. The change of plea was accepted out of the presence of the jury; because counsel for most of the defendants felt a cautionary instruction would only serve to call the jury’s attention to—and emphasize—the guilty plea, the court did not inform the jury of the circumstances concerning McCarthy’s subsequent absence from the case and courtroom. The court did, however, charge them at length concerning the proper means of individually determining the guilt or innocence of each defendant.
Three weeks after the jury returned with its verdict attorney Garlock, counsel for Mittelman, moved for a new trial, which motion was joined in by other defendants, on the strength of affidavits executed by Mr. Garlock and his wife. The affidavits asserted that the Gar-locks had encountered by chance, four days after the conclusion of the trial, one of the jurors in the case. The juror allegedly told the attorney that the jury had heard radio accounts and had read newspaper reports of the guilty plea; that said newspapers were brought into the jury room where they were discussed by members of the panel; and that the juror ventured the opinion that McCarthy’s plea “in effect said that the other defendants were also guilty.” The court denied the motions for a new trial and also declined to hold a hearing to question the jurors concerning the effect on their decision of the change in plea.
It was not error to refuse to grant a mistrial when McCarthy pleaded guilty. Certainly the guilty plea cannot be used as evidence against the remaining defendants, Babb v. United States, 5 Cir., 1955, 218 F.2d 538; United States v. Hall, 2 Cir., 1950, 178 F.2d 853; United States v. Toner, 3 Cir., 1949, 173 F.2d 140; but see Grunberg v. United States, 1 Cir., 1906, 145 F. 81, 86. But the federal courts have uniformly held it not error, if proper cautionary instructions are given, for the jury to be informed during trial that one or more defendants have pleaded guilty, or even for the jury to be present when the pleas are entered. Wood v. United States, 8 Cir., 1960, 279 F.2d 359; Davenport v. United States, 9 Cir., 1958, 260 F.2d 591; Richards v. United States, 10 Cir., 1951, 193 F.2d 554; Schliefer v. United States, 3 Cir., 1923, 288 F. 368. Appellants attempt to distinguish these cases on the strength of McCarthy’s counsel’s strong opening to the jury and because McCarthy “held himself out” as innocent at the start of trial. Such a distinction is meaningless; all defendants who start trial “hold themselves out” as innocent and the fervor or effectiveness of an opening statement cannot call for a different result. As in United States v. Falcone, 2 *949Cir., 1940, 109 F.2d 579, 582, “The existence of some kind of conspiracy was proved beyond peradventure, and the only important issue was whether these appellants were connected with the enterprise.” The appellants’ presentation of their cases, generally heaping all possible blame on McCarthy, strongly supports the thesis that there was no resultant prejudice from the change of plea.
The question raised by the denial of defendants’ motion for a new trial based on the jury’s knowledge and use of news accounts of the McCarthy plea is more difficult. Despite Lord Mansfield’s original flat prohibition of impeachment by juriors of their verdict, Vaise v. Delaval, 1 T.R. 11, 99 E.R. 944 (K.B. 1785), the government properly concedes that there is no bar to admitting jury testimony to prove the fact they read the newspaper, Mattox v. United States, 1892, 146 U.S. 140, 13 S.Ct. 50, 36 L.Ed. 917. But it seriously contends we cannot consider testimony by jurors as to the actual effect the papers had on their decision, if any, and that this distinction is in accord with the opinion in Mattox. McDonald v. Pless, 1914, 238 U.S. 264, 35 S.Ct. 783, 59 L.Ed. 1300, expressed a still more hostile attitude toward the admission of jury testimony to impeach verdicts, barring not only testimony as to matters solely within the knowledge of each individual juror but also matters known to the jury in common but not to outsiders, such as the fact that their verdict was reached by averaging the damage awards each thought proper. The Court had earlier stated, in Hyde v. United States, 1912, 225 U.S. 347, 384, 32 S.Ct. 793, 56 L.Ed. 1114, that jurors will not be heard on matters of impeachment provable solely by their own testimony and not subject to corroboration. McDonald and Hyde were cited with approval by the Supreme Court as late as 1953, Stein v. People of State of New York, 346 U.S. 156, 178, 73 S.Ct. 1077, 97 L.Ed. 1522, for the statement that jurors will not be heard to testify their verdict was a compromise.
In Jorgensen v. York Ice Machinery Corp., 2 Cir., 1947, 160 F.2d 432, this court indicated doubt as to the rule against impeachment and placed its decision sustaining a verdict which affidavits asserted to have been reached by majority vote in order to permit the foreman to go home in a family crisis, rather on the ground that the irregularity was not such as to call for upsetting the verdict. But we have since refused to hear jurors’ testimony that they relied on legal knowledge imparted to them in another case, Rotondo v. Isthmian S. S. Co., 2 Cir., 1957, 243 F.2d 581. We there stated the rule as permitting testimony that evidence reached the jury outside court, but as excluding testimony as to “the reasons that in fact induced them to find their verdict.” 243 F.2d 581, at page 583.
The juror’s testimony that he interpreted McCarthy’s plea as implicating the remaining defendants and that the verdict was influenced in fact would thus not be admissible. This is what Mattox stated; It is what we held in Rotondo. Marshall v. United States, 1959, 360 U.S. 310, 79 S.Ct. 1171, 3 L.Ed.2d 1250, is no authority to the contrary. There the trial judge before verdict polled the jury to obtain assurance they would not be influenced by the offending news article; the Supreme Court did not order a second inquiry after verdict but decided the matter was of a nature too inflammatory not to have affected the jury. In Krogmann v. United States, 6 Cir., 1955, 225 F.2d 220, a new trial was ordered, in part because the judge had reread to the entire jury an offending headline without quizzing them all as to whether it would affect them, but the events occurred before verdict rather than after. Even Wigmore, one of the McDonald rule’s sharpest critics, who would permit jury testimony to attack compromise and quotient verdicts as well as to note extraneous influences, and would also permit asking the jury about probable effects before verdict, 8 Wig-more on Evidence § 2350, draws the line *950on post-verdict motive testimony by jurors ; he would invoke the parol evidence rule to exclude such inquiry, 8 Wigmore on Evidence § 2349.
There are many cogent reasons militating against post-verdict inquiry into jurors’ motives for decision. The jurors themselves ought not be subjected to harassment; the courts ought not be burdened with large numbers of applications mostly without real merit; the chances and temptations for tampering ought not be increased; verdicts ought not be made so uncertain. Whenever information erroneously reaches the ears of the jury in open court, the judge’s instruction to disregard it may be disobeyed; yet it is not suggested that in such a case the jury ought be polled as to whether any of them relied on the forbidden knowledge. Even though presumably the jurors themselves know best, the question is determined, as is the question whether an error was prejudicial, on the basis of the nature of the matter and its probable effect on a hypothetical average jury. The same reasons apply when the improper knowledge reached the jury outside the courtroom. It was not error to refuse to examine the jurors as to their mental processes.
Stripped of the offered proof that the jury erroneously considered McCarthy’s guilt as indicative of the guilt of the others, the question presented is virtually the same as on the motion for a mistrial. We assume that Mattox v. United States, supra, makes admissible the statement that the jurors read in the newspapers the accounts of the McCarthy plea. The unanimous Circuit Court authority that such a plea may be accepted in open court is implicit recognition, however, that the probable effect on the jury of such knowledge is not sufficiently harmful to require a new trial; we agree. The situation here presented is a far cry from the infinitely more damaging information obtained by the jury in Marshall (defendant’s own prior felony conviction) or in Mattox (two previous murders by the defendant; newspaper stated that the evidence was so conclusive that even the defendant’s family had given up hope). Proper cautionary instructions were here given; the existence of the conspiracy itself was overwhelmingly proven; defendants at trial strove mightily to put all blame on McCarthy’s shoulders; and, finally, the fact that two of the defendants were acquitted weighs against the conclusion that the jury was prejudicially affected.
VIII. The Jencks Act Claim
All of the appellants have joined in defendant Crosby’s contention that the cross-examination of numerous government witnesses may have been unduly restricted because of the trial court’s refusal to order produced for defense counsel certain “witness sheets” 29 and notes of interviews conducted by the two Assistant United States Attorneys who prosecuted the defendants. When demand was first made for these papers, the Court, in accordance with the procedure prescribed by the Supreme Court in Campbell v. United States, 1961, 365 U.S. 85, 81 S.Ct. 421, 5 L.Ed.2d 428, had both prosecutors take the stand; an extensive voir dire was held looking into the circumstances under which the notes were taken and the subsequent “witness sheets” prepared — to determine whether the statements in question were “substantially verbatim.” 18 U.S.C. § 3500; Palermo v. United States, 1959, 360 U.S. 343, 79 S.Ct. 1217, 1225, 3 L.Ed.2d 1287. The Court decided that question in the negative and denied production to defense counsel; the notes and sheets were *951marked as Court Exhibits and were made part of the record on appeal.
The Supreme Court in Palmero held that in the light of both the language of the Jencks Act and its legislative history Congress had looked to the production of “only those statements which could properly be called the witness’ own words,” and which “reflect fully and fairly” what was said; a principal fear evinced was that of the distortion inherent in selective summaries and statements taken out of context. Finally, the Court stated that, within the Congressional framework, the “final decision as to production must rest * * * within the good sense and experience of the district judge * * * subject to the appropriately limited review of appellate courts.” 360 U.S. 343, at page 353, 79 S. Ct. 1217, at page 1225.
With the Palermo ease as a guide, inspection of the notes and “witness sheets,” and a perusal of the voir dire examination of the government attorneys, make it clear that the trial court did not err in denying production. The prosecutors testified that there had been omissions in the taking of notes — that often no notes were taken for long periods of the interviews. They added there was no attempt to use the witnesses’ own words; rather that these notes and jottings were prepared for the purpose of providing a general framework of reference from which this long and complicated case could be presented. They repeatedly answered in the negative the query whether the notes were verbatim. They testified that they made constant additions, interpolations and interpretative asides along with fragmentary recordations of the witnesses’ stories.
The notes and sheets largely bear out the testimony of the United States Attorneys. The witness sheets prepared from the handwritten notes are not reports of interviews, much less substantially verbatim ones. They have all the appearance of being plans for the conduct of examination of witnesses at trial. Many are divided into topics by headings preceding sentence fragments to indicate the expected testimony of the witness; others are made into more complete sentences, separately stated to make the individual points desired to be elicited. In no sense can they be said to be “the witness’ own words” or “continuous, narrative statements.” Moreover, most of them contain material clearly not elicited from witnesses during the interview. Finally, the fact that the sheets were not prepared contemporaneously with the interviews, in conjunction with the testimony that they and the notes were far from identical, might alone warrant refusal to order production.
The case for production of the notes themselves is little stronger. Notes of interviews with many witnesses had been destroyed prior to trial; the prosecutors testified that this had been done in good faith, with no thought of subverting the Jencks Act, to prevent cluttering of files after the typewritten witness sheets had been prepared. See United States v. Thomas, 2 Cir., 1960, 282 F.2d 191. They also stated that the remaining notes were substantially identical in form to those that had been destroyed. The notes, of course, cannot definitely corroborate the testimony of the prosecutors that substantial gaps existed in the note taking. In other respects, however, the notes which remain do substantiate the testimony given on voir dire. Generally, they appear to be composed of fairly cryptic references to the highlights of the respective interviews — to be used as reminders at trial. Even the notes contain some matter extrinsic to the interviews, cross references to the introduction of documentary evidence, identification of individual defendants, and occasional references to inter-relationships with expected testimony of other witnesses. Finally, legislative history strongly indicates that one evil Congress sought to remedy in passing the Jencks Act was the possible production of notes made by government attorneys in preparing their case. 103 Cong.Rec. 15931 (1957) (Remarks of Senator Hruska); S.Rep. No. 981, U.S.Code Cong. & Adm. News, 85th Cong., 1st Sess., 1861, 1867.
*952IX. Other Contentions
Some of appellants’ other contentions may be dealt with summarily. Mittelman. claims his Count Two conviction cannot stand because it is completely inconsistent with the jury’s acquittal of him on Count Twelve. It is by now well settled, however, that “consistency in the verdict is not necessary,” Dunn v. United States, 1931, 284 U.S. 390, 393, 52 S.Ct. 189, 190, 76 L.Ed. 356; Steckler v. United States, 2 Cir., 1925, 7 F.2d 59, 60; defendant’s attempt to distinguish these landmark decisions, and a host of other Federal cases, is completely without support in logic or precedent. Mittelman also claims that the “public offering” exemption incorporated into the criminal provision of the Securities Act is so amorphous as to make the criminal statute unconstitutionally vague — and that the Court below erred in its charge to the jury concerning the applicability of that exemption. While the term “public offering” in the statute alone might possibly be open to an attack on grounds of vagueness, the judicial gloss placed on this legislation by the Supreme Court in Securities and Exchange Commission v. Ralston Purina Co., 1953, 346 U.S. 119, 73 S.Ct. 981, 97 L.Ed. 1494, two years before the acts here charged, cured any defect which might have existed. As for the correctness of the charge, the trial court adopted most of the very language of the Supreme Court in Ralston Purina. See also Gilligan, Will & Co. v. S. E. C., 2 Cir., 1959, 267 F.2d 461, certiorari denied 361 U.S. 896, 80 S.Ct. 200, 4 L.Ed.2d 152. Any of the numerous arguments raised by the four appellants whose convictions we are affirming not specifically answered in this opinion, are either totally lacking in merit or inapplicable to them.
The judgments of conviction of defendants Crosby, Meredith, Mittelman and Pettit are affirmed. The convictions of Goldberg, Gordon, Reicher and Philip Gordon & Co. are reversed and as to them the indictment is dismissed.
. William J. McCarthy, president of Texas-Adams during most of the period pertinent to the indictment, changed his plea to guilty during the first week of the trial. Appellants moved for mistrial on the basis of that change of plea and assign the Court’s denial as error; this contention will be discussed infra.
. Crosby was named, in this group, only in Counts 2-7, 12 and 14.
. In Counts 39-49 the mailings involved were of share certificates of Texas-Adams. Philip Gordon & Co., Inc., and the three individual broker defendants, Gordon, Goldberg and Reicher were not charged in Counts 39 and 40.
. The trial court dismissed thirteen of the counts before the case was given to the jury, four of the mail fraud counts (3, 8, 28 and 29) and nine of the counts dealing with the use of the mails to sell unregistered securities (33, 35-7, 41, 42, 44, 45, 49). Crosby was found guilty on Counts 1-7 and 50, but acquitted on 12 and 14; Meredith was convicted on all counts submitted to the jury except Count 1 (he was not named) ; the jury convicted Pettit on the surviving counts between 1 and 22, 39, 40, 43 and 50 — and acquitted him on 23-32, 34, 38, 46-8; Mittelman was found guilty only on the first mail fraud count and on the conspiracy. The brokers were convicted on all counts on which their names were submitted to the jury, 34, 38, 46-8 and 50. The government, on appeal, has confessed error on Counts 34 and 38; the proof there showed that the brochures allegedly mailed to facilitate sales of unregistered securities were not mailed until after the sales proved in those counts.
. Later, after the takeover of Texas-Adams, Crosby again contacted Zavoico and had him prepare a report on the projected sale of the Ajax properties — this time to Texas-Adams. The Sundown Oil Company was to pass on the Ajax fields (price to Sundown one and a quarter million) to Texas-Adams for $1 million in cash, $1 million in ten j'ear notes, and 1,250,000 shares of Texas-Adams stock valued at that time at one dollar a share.
. Homestead was apparently in trouble of its own. A public issue of $300,000 of its capital stock had been authorized without S.E.C. registration pursuant to a Regulation “A” exemption. The underwriters, however, had failed to obtain subscriptions for the entire issue; the promoters were faced with the possible necessity of undergoing the process of registration with its attendant public disclosures. Such a course may well have been a particularly uncomfortable one as Homestead had purchased certain “uranium properties” from one Paul R. Jones, a co-conspirator who appears to have specialized in acquiring worthless mining claims and reselling them to captive or cooperative corporations at a lucrative profit.
Probably to ease tbe difficult situation Homestead’s promoters found themselves in, Mittelman and McCarthy undertook to complete sale of the subscription issue. 1.800.000 of the original public issue of 5.990.000 shares at $.05 remained unsold. Mittelman “bought” those shares with $90,000 borrowed from Howard P. Carroll, underwriting broker for Homestead, and an associate. He then borrowed the $90,000 back from Homestead, immediately repaid the personal loans, and headed back east with the stock, which he was to attempt to sell. Apparently, a half million shares were left with Jones in Birmingham, Alabama and Mittelman turned over the remaining certificates to Satiris G. Fassoulis in New York. Fassoulis had agreed to help dispose of the Homestead stock in order to help the McCarthy group straighten its financial skirts in connection with other proposed mergers and purchases being negotiated in New York, see infra. Homestead was finally merged into Texas-Adams after tbe “management” defendants gained control of that corporation.
. This, however, did not stop Texas-Adams from going ahead with negotiations for a Taylorcraft merger — perhaps in the hope of eventually grasping the Warrior funds which the aircraft company was to obtain. A Texas-Adams-Taylorcraft merger was approved by the respective Boards of Directors and McCarthy and Crosby were elected to the five man Taylorcraft Board. The plan was never submitted to the stockholders of Taylorcraft for necessary ratification, however, and the final merger was never consummated. This fact, likewise, did not deter Texas-Adams from making untrue representations concerning the value of Taylorcraft in its market literature.
. The significance of this misrepresentation is discussed below.
. As noted before, both Castagna and Gully were acquitted by the jury. Gully’s defense, apparently accepted by the j'ury, was that his endorsement had been forged on the allegedly kicked back certificates. Whether the shares wore kicked back by agreement or stolen and forged, the role played on the Texas-Adams side of the transaction was patently illegal.
. McCarthy is the brother of Glenn McCarthy—of Texas oil and Shamrock Hotel fame.
. Especially convincing, in this regard, was the testimony of the Longview, Texas, oilman, John O. Robbins, Jr., J. A. 816a, one of the few important witnesses who could not be impeached in any way by the hint of fraudulent involvement on his own part.
. On this point, appellant Mittelman urges his acquittal on some 30 odd counts while he was convicted on only Counts Two and Eifty. While his inconsistency contention will be discussed infra, it will here suffice to point out that there were factors, other than directly relevant evidentiary ones, which may have led the jury to convict on only two counts. The collateral consequences to Mittelman, an attorney, may have weighed with the jury—plus the fact that at trial he produced an impressive array of character witnesses.
. Loss, Securities Regulations (1951), page 336.
. “The provisions of section 77e of this title shall not apply to any of the following transactions:
“(1) Transactions by any person other than an issuer, underwriter, or dealer; transactions by an issuer not involving any public offering; or transactions by a dealer (including an underwriter no longer acting as an underwriter in respect of the security involved in such transaction), except transactions taking place prior to the expiration of forty days after the first date upon which the security was bona fide offered to the public by the issuer or by or through an underwriter and transactions in a security as to which a registration statement has been filed taking place prior to the expiration of forty days after the effective date of such registration statement or prior to the expiration of forty days after the first date upon which the security was bona fide offered to the public by the issuer or by or through an underwriter after such effective date, whichever is later (excluding in the computation of such forty days any time during which a stop order issued under section 77h of this title is in effect as to the security), and except transactions as to securities constituting the whole or a part of an unsold allotment to or subscription by such dealer as a participant in the distribution of such securities by the issuer or by or through an underwriter.
“(2) Brokers’ transactions, executed upon customers’ orders on any exchange or in the open or counter market, but not the solicitation of such orders.” Section 4 of the Securities Act of 1933, as amended, 15 TJ.S.C.A. § 77d.
. “The term ‘dealer’ means any person who engages either for all or part of his time, directly or indirectly as agent, broker, or principal, in the business of offering, buying, selling or otherwise dealing or trading in securities issued by another person.” Securities Act of 1933, § 2(12), 15 U.S.C.A. § 77b (12).
. “The term ‘underwriter’ means any person who has purchased from an issuer with a view to, or offers or sells for an issuer in connection with, the distribution of any security, or participates or has a direct or indirect participation in the direct or indirect underwriting of any such undertaking ; but such term shall not include a person whose interest is limited to a commission from an underwriter or dealer not in excess of the usual and customary distributors’ or sellers’ commission. As used in this paragraph the term ‘issuer’ shall include, in addition to an issuer, any person directly or indirectly controlling or controlled by the issuer, or any person under direct or indirect common control with the issuer.” Securities Act of 1933, § 2(11), 15 U.S.G.A. § 77b(11).
. The sales in question, as noted, were of Gully-Comigu stock, a transaction which took place rather late in the short lifetime of Texas-Adams’ greatest activity. Therefore, a showing that the brokers were privy to or aware of a consistent course of fraudulent conduct on the part of the officers or attorneys of Texas-Adams would be relevant to the issue of their good faith in trading the unregistered securities.
. The balance of the agreed upon purchase price for the investment shares was paid by Baker, in two smaller checks, to Meredith and Lawrence Rosen.
. Aside from the extensive dealing in Texas-Adams stock by defendant Philip Gordon & Co., Inc., four other brokerage firms, the majority of which were named as co-conspirators, sold either more shares than Gordon & Co. or almost as many. In addition to those houses “pushing” or “dealing in” the stock, Texas-Adams was generally traded by many other brokerage firms during the period of the McCarthy leadership.
. In most instances, apparently, these defendants were buying and selling on their own accounts—with attendant risk to themselves—rather than acting merely as brokers.
. The government suggestion here is that this is explained by the fact that the price stabilization needed was a purely local problem in Denver. This overlooks the fact that by far the largest number of bonus shares went to a Pittsburgh brokerage firm.
. Good examples of this practice may be found in the record of the instant case. Both Texas-Adams and Homestead Oil & *942Uranium Co. were originally financed by a public issue, under a Regulation “A” exemption, to the tune of $300,000. Contempoi’aneous with the organization of the corporation and the public issue of the bloc of shares, the organizers and often the underwriters are issued investment shares greatly swelling the outstanding capital stock. When the large, underwriting commissions on these speculative issues are then paid, the public’s stock bas by that time been effectively “watered” often by more than 50%.
. In contrast to other brokers named as co-conspirators, there was a complete failure to connect these bx’oker defendants with any of the fraudulent transactions by McCarthy and company underlying the eventual public offerings of stock.
. This problem was most acutely presented here because witnesses had to be subpoenaed from Denver, Pittsburgh, North Carolina, Texas and other farflung parts of the country.
. It should be noted that Judge MacMahon handled the problem well under the circumstances. The jury was given a short break at regular intervals of approximately one hour; these recesses came at logical stopping points in the presentation.
. Compare the field of federal taxation where the combination of a highly sophisticated tax bar and an obstinate, if sliglitly less quick, legislature have resulted in our present tax laws.
. This, of course, does not mean that the government is free to offer evidence of other unrelated “crimes” under the guise of the “scheme or artifice” of the mail fraud offense,
. “There has been testimony as to. the previous good character of the defendant Mittelman. You should consider such evidence of good character, together with all the other facts and all the other evidence in determining his guilt or innocence. Evidence of good character may, in itself, create a reasonable doubt where, without such evidence, no reasonable doubt would exist.
“But if from all the evidence you are satisfied beyond a reasonable doubt that Mittelman is guilty, a showing that he previously enjoyed a reputation of good character does not justify or excuse the offense, and you should not acquit him merely because you believe that he has been a person of good repute.
“It may be that those with whom he had come in contact previously have been misled, or that he did not reveal to them his real character or acts.
“The testimony of a character witness is not to be regarded by you as expressing the witness’ personal opinion of Mittelman’s character, nor is it to be taken by you as the witness’ opinion as to his guilt or innocence.
“The guilt or innocence of Mittelman is for you, and you alone, to determine.”
. According to testimony given by the two Assistant United States Attorneys who tried the case for the government, handwritten notes were first taken of interviews with various witnesses. Afterward, typed “witness sheets” were prepared; although the notes were used as a base, often much of the notes was excised and other information and cross references to exhibits were added — to result in a “trial aid” in the direct examination of government witnesses.