A numeric sampling of the above checks has been evaluated to determine compliance with the American Bankers Association Specifications (147 R 3, THE COMMON MACHINE LANGUAGE FOR MECHANIZED CHECK HANDLING AND THE SUPPLEMENT TO 147 R 3).

The number of recurring casualties on the non-compliance items are listed below. The remarks column indicates any areas of trouble, what ever, and is reported both in the case of unacceptable and marginal work.

This report refers only to the batch of checks described above and does not constitute an endorsement by the Bank of America of this print ing concern and its ability to print magnetically encoded checks.

A numeric sampling of the above checks has been evaluated to determine compliance with the American Bankers Association Specifications, (147 R 3, THE COMMON MACHINE LANGUAGE FOR MECHANIZED CHECK HANDLING AND THE SUPPLEMENT TO 147 R 3).

The number of recurring casualties on the non-compliance items are listed below. The remarks column indicates any areas of trouble, whatso ever, and is reported both in the case of unacceptable and marginal work.

This report refers only to the batch of checks described above and does not constitute an endorsement by the Bank of America of this print ing concern and its ability to print magnetically encoded checks.

As you may be aware, Bank of America has been advising us of an ongoing
problem concerning quality control problems with our BOA checks printed
by NorthEast Business Forms.

On April 25, Tom Sherman, Hugh van Hengel and myself met with two Bank
of America loan officers, at which point a conference call with Karen
DeKlerk, of BOA's Customer Check Support group, was made. According to
Ms. De Klerk, basically because of various quality problems with the MICR
line of our checks, we have a check reject rate far in excess of industry
standards. Because of this high reject rate, our account has required
special handling by BOA, equating to approximately $950.00 a month in
added labor costs. Tom Sherman then briefly discussed the problem with
NorthEast Business Forms and asked them to expect a call from Ms. DeKlerk
to try working with them in resolving the problem.

Subsequently, last week this situation was again discussed at a meeting
with BOA, Tom Lynch and myself. Apparently, this problem has yet to
be resolved, and BOA can not understand why our level of rejects can
not be reduced to the industry standard. I am at a loss to explain it
as well, and at this point it is damaging our overall relationship.

I would appreciate it if at this point, you could look into the problem
and shepherd it to a satisfactory resolution. We face both permanent
damage to our relationship, as well as the $950.00 per month special
handling charge.

If I can be of any assistance in this process feel free to contact me.
Though let me stress that we should move as rapidly as possible in
solving this problem, even if it means changing our current printer.

MEMORANDUM

June 20, 1984

Re: "Misencoded" Checks

Prior to late 1981 Hutton maintained its principal

branch reimbursement account at Chemical Bank. Checks for this account were preprinted "Pay to the order of E. F. Hutton & Company Inc." and sent to all the branches for use in avoiding negative bank ledger balances. A Hutton branch would deposit a Chemical check in its branch bank to reimburse itself, the check would clear through the banking system and be presented to Chemical, and the Company would fund payment of each day's clearings by wire transfer. The Company ordered several thousand of these checks from the printer at a time, divided them into small batches, and sent them out to branches as needed.

The Company's chief check printer is Northeast Forms, Inc. Northeast prints stock checks for the Company in batches of 100,000 and holds them until the Company orders smaller batches (usually between 1,000 and 10,000) for its branches. To order checks the Company's purchasing department sends the printer a "Z-Release" form, which authorizes him to release a certain number of checks from stock (Exhibit 1). This release specifies the form number of the checks being ordered, the address to be printed on the checks, and the check numbers to be printed in Arabic numerals on the upper right of the check. If

special instructions accompany the order, the statement "See attached specification sheet" will appear on the Z-release.

The specification sheets include a "Specifications for Printing Checks" form, which sets forth the bank name and address, the transit and routing code, and the microencoding instructions (Exhibit 2). This is attached to a "Specifications for Printing Forms" sheet that designates the form number of the checks, the quantity, the check numbers, and the priority of the order (Exhibit 3).

The coding that was missing from some checks should have appeared at the bottom of each check on the "MICR" (magnetic ink character recognition) or microencoded line. The MICR line of all of Hutton's checks contained the account number and a routing code for clearing. What was missing from some batches of checks was the sequential check number (i.e., check numbers 1, 2, 3, etc.) on the MICR line, although the sequential check numbers were printed in Arabic numerals in the upper right-hand corner of all checks.*

It is perhaps misleading to describe the checks at issue as "misencoded", since they were properly (albeit incompletely) encoded. The sole function of placing check numbers

See Exhibit 4 for an example of a check with a microencoded sequence number. Note that "06933" appears at the extreme left of the MICR line.