The Error It Cannot See

Fourth article in this silo. Trust accounting is the most heavily regulated bookkeeping in Canada and the least written about outside the rulebooks themselves.

The Verdict, Stated First

Five claims, in descending order of confidence.

One. Ontario requires a monthly trust reconciliation no later than 25 days after the end of the period covered by the bank statement, and requires it even in months with no transactions.

Two. The test is a three-way comparison, not a bank reconciliation, and the third leg is what makes it a trust control rather than a cash control.

Three. On our own analysis the test is blind to same-amount misposting between clients, which is our reasoning from its structure rather than a finding we obtained.

Four. On our own arithmetic the 25-day window is far tighter than it sounds, leaving a firm with a one-day-a-week bookkeeper roughly two working days.

Five. And the ten-year retention period means dormant client ledgers are carried and reconciled for a decade, which lengthens every monthly close.

The third is the one worth an article, ours. A control that passes while the breach it exists to prevent is occurring is worse than understood.

Worse than understood because of how it fails, ours. A control that visibly breaks prompts investigation, and one that silently passes produces confidence.

A Warning About Dates

Standing feature of this silo. Ours.

Four observations.

All rules described here were verified on 29 August 2026 and are those of the Law Society of Ontario.

By-Law 9 is amended periodically, and the version we consulted records amendments across multiple years[3].

Trust accounting rules are provincial and differ materially between law societies, which we address in its own section and do not attempt to survey.

And the structural analysis outlasts the rules, ours. The blind spot in a three-way reconciliation is a property of the arithmetic, and it holds in any jurisdiction that uses the same test.

So the two halves of this article age differently, ours. Treat the timings and thresholds as current to a date and the structural analysis as durable, which is the same instruction every article in this silo carries.

Our Grades For These Claims

Applying the scheme this publication uses throughout.

Grade A for the existence and section numbering of the requirements, from the Law Society of Ontario's own materials[1][2][3].

Grade B for the operational detail, including the 25-day timing and the three-way structure, which reach us through practice guidance and commercial sources citing the Law Society[4][5][6].

Grade A for our own analysis of the blind spot, which requires no source because it follows from the structure of the test.

Grade A for our own arithmetic, with every firm figure invented and marked.

Grade D for provinces other than Ontario, which we have not researched beyond noting that they differ.

A Note On Method

Everything here is verified to 29 August 2026.

We obtained the Law Society of Ontario's frequently asked questions on reconciling a trust account, which cites the operative By-Law 9 provisions[1], and its summary of By-Law 9 record keeping requirements[2].

We obtained a Law Society copy of By-Law 9 itself, though only in fragments[3].

We did not obtain the full current text of By-Law 9, and describe its requirements as the Law Society's guidance and commercial practice sources characterise them. That is a real limitation on an article about compliance.

All arithmetic and the reconciliation logic table are ours. The firm, its matter volume and its residual balances are invented.

This article discusses professional regulation and bookkeeping practice. It is not legal or accounting advice, and a licensee should work from the current By-Law and their law society's guidance.

The Three Figures

What the reconciliation actually compares, which is more than a bank reconciliation.

The test compares the adjusted bank statement balance, the trust account's own journal or general ledger balance, and the combined total of all individual client ledger balances. All three must match[5].

By-Law 9 requires a detailed reconciliation made monthly of each trust bank account[3], with the operative provisions at section 18 and the comparison requirements at subsections 18(8) and 22(2)[1].

Four observations, ours.

The first two figures are an ordinary bank reconciliation. Any business does that, and it proves the firm's record of the account matches the bank's.

The third figure is what makes it a trust control. It proves that the money in the account is fully allocated to identified clients, with nothing unassigned and nothing over-assigned.

The third leg is also the one most often prepared badly, in our experience, because it requires every client ledger to be listed and totalled rather than a balance to be read off a system.

And a firm that runs only the first two has a bank reconciliation, ours. It does not have a trust reconciliation, whatever its software calls the report.

That is worth checking rather than assuming, ours. Ask to see the client ledger listing that supports the total, and if nobody can produce one, the third leg is not being performed.

The Blind Spot

Our own analysis of the test's structure. This requires no source because it follows from the arithmetic.

Consider which pair of figures each error breaks.

A deposit recorded but not yet at the bank breaks A against B. A bank fee not recorded breaks A against B. A cheque written but not presented breaks A against B. Trust money never deposited breaks A against B.

An amount keyed wrong on one client ledger breaks B against C. A client ledger omitted from the listing breaks B against C.

And then: a receipt posted to the wrong client for the correct amount. The journal total is unchanged. The client ledger total is unchanged, because one ledger rose by exactly what another should have. The bank is unchanged. All three figures agree and the reconciliation passes.

Four observations.

The same holds for a disbursement charged to the wrong client's ledger, for the same reason and with the same result.

This is not a weakness in how the test is performed. It is a property of what the test measures, which is aggregate allocation rather than correct allocation.

A perfectly executed reconciliation, prepared on time by a diligent bookkeeper and reviewed by a licensee, will report no exceptions while the error sits in the ledgers.

And we state this as our reasoning rather than as a finding, ours. No source we obtained says it, and a reader should satisfy themselves by working an example rather than taking our word.

One clarification on scope, ours. We are not saying the reconciliation is a poor control; it catches every error in the first list and those are the common ones.

We are saying it is incomplete in a specific and knowable way, and that a firm relying on it alone has an uncovered exposure it can name.

What It Looks Like On Paper

Our own worked example, with invented figures.

A firm receives $12,000 in trust for Client A and the receipt is posted to Client B's ledger.

Client A's ledger should show $12,000 and shows nil. Client B's ledger should show nil and shows $12,000. The total of client ledgers is $12,000, which is correct. The trust journal shows $12,000, which is correct. The bank holds $12,000, which is correct.

Four observations.

The reconciliation is clean. Three figures, all $12,000, no exceptions to investigate.

Now the firm pays a disbursement on Client B's matter out of trust. It is paying with Client A's money, and the ledger it draws on says the funds are there.

The position only surfaces when Client A asks for their money or when somebody reads Client A's ledger for an unrelated reason.

And by then the firm has a shortage to fund from its own resources, ours, plus a disclosure question it did not have the day before.

Note how long that can run, ours. Nothing forces the discovery, so the interval between the misposting and its surfacing is set by when somebody happens to look at one of the two ledgers.

Why That Is Not A Bookkeeping Problem

The reframing that makes this worth a licensee's attention. Ours.

Four observations.

A posting error between two general ledger accounts is a bookkeeping problem, corrected by a journal entry and nobody is harmed.

A posting error between two client trust ledgers is a use of one client's funds for another client's purpose, which is the thing trust accounting rules exist to prevent.

The intention is irrelevant to the outcome and largely irrelevant to the obligation. The rule is about whose money paid for what, and a clerical origin does not change the answer.

Which is why we would treat this as a control question rather than an accuracy question, ours. The correct response is a detective control aimed at it specifically, not more care during posting.

What Actually Finds It

Four controls that address the blind spot, ours, and none of them is the monthly reconciliation.

A negative client ledger balance report. The most valuable single report in a trust system, because a misposting frequently leaves one ledger overdrawn even when the total agrees.

A per-matter review at file closing. The lawyer responsible for the file is the only person who knows what should have gone through it, and closing is the moment they will look.

A client ledger listing reviewed by matter rather than totalled. The reconciliation needs the total; catching this needs the detail, and those are different documents.

And an ageing of dormant balances, since a ledger that should be nil and is not is either a residual or a misposting, and both need clearing.

Four observations.

Only the first is likely to be automated in a small firm's software, and it is often available and unrun.

The second is free and is the strongest of the four, because it uses knowledge no bookkeeper has.

The third costs the time to read a listing, which for a firm with a few hundred open matters is not large.

And none of them is required by the By-Law as we read it, ours. They are our suggestions for a gap we identified ourselves, and a licensee should form their own view.

Twenty-Five Days

The timing requirement, and it is more specific than most firms treat it.

Ontario requires monthly reconciliation and allows the process to be completed no later than 25 days after the end of the period covered by the financial institution's monthly statement[4].

Four observations, ours.

Note what the clock runs from. It is the end of the statement period, not the month end and not the date the statement arrives.

For most accounts those coincide, and for some they do not. A statement period ending mid-month moves the deadline accordingly.

Practice guidance suggests reviewing the comparison and its supporting documentation by the 25th of each month[6], which is the same instruction expressed as a habit.

And the deadline is for completion, not for starting, ours. A reconciliation begun on day 24 and finished on day 28 was late, however diligently it was done.

One more distinction inside the requirement, ours. The comparison is a document to be produced, not a state to be achieved, so a firm whose figures happen to agree but which prepared no comparison has not met the obligation.

What That Is In Working Days

Our own arithmetic, and it is why small firms miss the deadline.

Twenty-five calendar days sounds generous. Subtract the time before the statement is available and convert to business days.

If the statement is available on day 3, there are 22 calendar days and about 15 business days. On day 5: 20 calendar, about 14 business. On day 7: 18 calendar, about 12 business. On day 10: 15 calendar, about 10 business.

Now apply the bookkeeper's schedule. Against roughly 14 business days in the window, a bookkeeper working four days a week has about 11, working three has about 8, working two has about 5, and working one day a week has about 2.

Four observations.

A firm with a one-day-a-week bookkeeper has roughly two working days inside the window to reconcile, identify breaks, chase the bank and correct the records.

That is not enough for a month with a genuine break, ours, and it is the structural reason small firms file late rather than a diligence failure.

The business-day conversion is ours and is approximate. It ignores statutory holidays, which in several months would reduce the figures further.

And the practical response is scheduling rather than effort. A firm in this position should book the bookkeeper's day in the first half of the window, so a break discovered on the first visit can be resolved on the second.

There is a second scheduling point worth making, ours. The window is the same twenty-five days for every firm regardless of size, so the smallest practices operate the tightest version of the requirement.

Required Even If Nothing Happened

A point that catches dormant practices.

In Ontario the reconciliation is required every month even if no transactions occurred[5].

Four observations, ours.

So a trust account holding a single dormant balance still generates twelve reconciliations a year.

Not every jurisdiction takes that position. The Northwest Territories does not require monthly reconciliations when no activity has occurred, and otherwise requires reconciliation at least annually at each fiscal year end[4].

That contrast is useful for a firm practising in more than one jurisdiction, because the Ontario habit is the safe one and the reverse is not.

And it argues for closing accounts you do not need, ours. An empty trust account is not free; it is twelve compliance events a year.

With one caution before closing anything, ours. A dormant balance belongs to somebody, and clearing the account is a separate question from clearing the funds, which has its own process we have not researched.

Stale-Dated Cheques

A recurring reconciling item with a specific treatment.

Practice guidance directs that cheques outstanding more than six months be reversed, with the client liability reinstated in the client's trust ledger and the cheque reissued if appropriate. It notes that clearing rules permit a cheque more than six months old to be cashed, so a firm should check its institution's policy and consider a stop payment before reissuing[6].

Four observations, ours.

The reinstatement step is the one that gets missed. Reversing the cheque without restoring the client ledger creates exactly the misallocation this article is about.

The stop payment warning is a real exposure rather than a formality. Reissuing without stopping the original leaves both instruments live, and the account can be drawn twice.

Six months is also a useful ageing trigger for a different purpose. A trust cheque nobody has cashed in six months frequently means the payee has moved or the matter resolved differently.

And this is where reconciling items become residual balances, ours, which is the next section.

The Ledgers That Never Close

Our own arithmetic on an invented firm. Every figure is ours.

A practice closing 400 matters a year. If 1 percent leave a residual averaging $45, that is $180 a year and $1,800 across 40 dormant ledgers over a decade. At 2 percent and $85: $680 a year, $6,800 across 80 ledgers. At 3 percent and $120: $1,440 a year, $14,400 across 120 ledgers.

Four observations.

The money is not the problem. Fourteen thousand dollars across a decade is immaterial to most practices.

The ledger count is the problem. A hundred and twenty dormant client ledgers must be listed and totalled in every monthly comparison, which lengthens the close permanently.

And they degrade the control described earlier. A listing padded with dormant balances is harder to read for anomalies, so the residuals hide the misposting.

Which is the practical argument for clearing them, ours. Not to recover the money, but to keep the report short enough that somebody will read it.

Ten Years Plus Current

The retention obligation, which drives the accumulation above.

A commercial summary of By-Law 9 records retention periods of 10 years plus current for records of trust and general account transactions, receipts and disbursements of trust money, money received and disbursed for each client's trust account, and signed electronic trust transfer requisitions and confirmations; and 6 years plus current for records of transfers of trust money between client trust ledger accounts and for records of cash received[8].

A commercial bookkeeping firm rather than the Law Society, flagged, and we did not verify these against the current By-Law.

Four observations, ours.

The distinction between the two periods is worth noting. The core trust records run longer than the cash and inter-ledger transfer records.

The phrase "plus current" is doing real work, since it means the obligation is longer than ten years from the transaction in every case.

Ten years is substantially longer than the CRA's ordinary six-year record retention period, so a firm applying a general business retention policy to its trust records will destroy them early.

And that mismatch is the practical point, ours. A law firm needs two retention schedules, not one, and the shorter one must not be applied to trust.

The risk concentrates at a predictable moment, ours. An office move or a system migration is when records get culled, and it is the point at which a general policy is most likely to be applied to everything at once.

No ATM, No Debit

A restriction that surprises firms moving from ordinary business banking.

In Ontario, ATM withdrawals and transfers from a trust account are flatly prohibited. Funds may leave only by cheque or through electronic transfer systems meeting specific security requirements[5].

Four observations, ours.

This removes the fastest and least documented ways to move money, which is the point rather than an inconvenience.

It has a banking consequence at account opening. A firm should ask the institution to suppress card access on the trust account rather than rely on nobody using it.

And it interacts with modern payment habits in a way worth flagging. Interac transfers, bill payments and pre-authorised debits are not obviously within the permitted methods, and a firm should confirm rather than assume.

We could not establish from the sources we obtained exactly which electronic systems qualify, ours, and a licensee should get that answer from the Law Society rather than from this article.

Dual Authorization In A Two-Person Firm

Where the control model meets the reality of a small practice.

For internet banking transfers the process requires a signed requisition form, with the licensee signing a Form 9A initiating the transfer, and dual authorization: one person inputs the transaction details using a unique password and a second person authorizes it using a different password[5].

Four observations, ours.

Dual authorization requires two people. A sole practitioner with a part-time assistant has a scheduling problem embedded in a control requirement.

The obvious workaround defeats the control entirely. Two passwords held by one person is not dual authorization, whatever the system records.

Which means the practical choice for a very small firm is often to use cheques instead, since a cheque requires a signature the licensee provides personally.

And that is an unglamorous conclusion we would nonetheless stand behind, ours. A control you can actually operate beats a faster one you have to circumvent.

The Cash Ceiling

The limit and its recording obligations.

By-Law 9 restricts cash. A licensee may accept more than $7,500 Canadian in cash only in connection with specified exceptions, and foreign currency must not convert to more than $7,500 Canadian unless an exception applies. Where cash is accepted, the by-law mandates a book of duplicate receipts for every cash transaction, with the receipt recorded in either the trust receipts journal or the general receipts journal depending on where the cash was deposited[7].

A commercial software vendor's blog rather than the Law Society, flagged.

Four observations, ours.

The ceiling is per transaction in connection with a matter rather than a daily limit, on our reading, and a firm should confirm the precise scope.

The duplicate receipt book is a physical control in a largely digital process, which is deliberate, since a bound sequential book is hard to alter after the fact.

The retention period for cash records is the shorter one at six years plus current[8], not the ten-year period applying to trust transactions.

And the exceptions are the part to look up rather than remember, ours. We have not listed them because getting that list wrong would be worse than sending a reader to the source.

That is a general policy in this silo rather than a gap here, ours. Where a rule turns on a list of exceptions we either obtain the list or name the omission, because a partial list reads as a complete one.

Outside Ontario

A material limitation, given its own section. Ours.

Four observations.

Trust accounting is regulated by each provincial law society, and the rules differ in timing, format and reporting.

Two contrasts we did obtain: the Northwest Territories does not require monthly reconciliation when there has been no activity, and the Law Society of Alberta operates a Trust Safety Program with annual compliance activities going beyond standard reconciliation[4].

We did not research any other jurisdiction, and a firm outside Ontario should treat the timing and procedural detail here as inapplicable.

What does transfer is the structural analysis, ours. Any jurisdiction using a three-way comparison has the same blind spot, because it is a property of the arithmetic rather than of the rule.

And the same is true of the working-day squeeze in principle, ours, though the specific numbers depend on each jurisdiction's deadline and a firm elsewhere should redo that calculation with its own.

What Happens When It Does Not Balance

The obligation on discovering a break, which is stricter than ordinary bookkeeping practice.

Any discrepancy, whether a shortage or an overage, must be investigated and resolved immediately[5].

Four observations, ours.

Overages are included, which surprises people. Money in the trust account that belongs to nobody identified is as much a problem as money missing.

The word is immediately, not by the deadline. A break found on day 10 is not something to carry to day 25 and explain in the comparison.

A shortage is generally funded from the firm's own resources, on ordinary trust principles, which converts a bookkeeping error into a cash call on the practice.

And we would note what we did not establish, ours. We did not determine the reporting obligations that attach to a shortage, which vary by size and cause and are exactly the point at which a licensee should call the Law Society rather than an accountant.

Mixed Trust Accounts

The default structure, and why the client ledger leg exists at all. Ours.

The Law Society's record keeping resource addresses mixed trust accounts as a category alongside general accounts[2], and practice guidance directs that a comparison include mixed accounts along with pass books, GICs, term deposits, electronic registration, estate and power of attorney accounts[6].

Four observations.

A mixed trust account holds funds for many clients in one bank account, which is ordinary and is why the client ledger total is the control that matters.

Without that third leg the structure would be indefensible. One bank balance covering fifty clients tells you nothing about whether any individual client's money is there.

The listing must reach beyond the main account. Term deposits and GICs held in trust are client funds and belong in the comparison, and they are the items most often left out because they sit on a different statement.

And that omission produces a break of a particular shape, ours. Client ledgers exceeding the bank balance because a GIC was not counted looks alarming and is a listing error rather than a shortage.

Which is worth knowing before the panic, ours. A shortage and an incomplete listing present identically on the comparison, and the first thing to check is whether every account holding client money is on the schedule.

If You Run A Small Firm

Practical, and not legal or accounting advice. Ours.

Four points.

Run a negative client ledger balance report every month. It is the single control most likely to catch what the reconciliation cannot, and most systems already produce it.

Have the responsible lawyer review the client ledger at file closing. They know what should have gone through the matter and no bookkeeper does.

Book the bookkeeper's time in the first half of the 25-day window, so a break found on one visit can be fixed on the next.

And clear residual balances actively, to keep the client ledger listing short enough that somebody reads it rather than totals it.

A fifth, if the firm has the people for it, ours. Rotate who prepares the comparison and who reviews it, since the same person doing both across many months is the condition under which a consistent error becomes invisible.

If You Support One

For our own profession. Ours.

Four points.

Confirm the third leg exists. A report reconciling only bank to ledger is a bank reconciliation, and a firm may believe it has more than it has.

Ask for the client ledger listing, not the total. The total satisfies the comparison and the detail is where anomalies live.

Check the retention policy separately from the client's general one, because ten years plus current is longer than most business schedules.

And do not advise on By-Law 9 compliance from secondary sources, which includes this article. The Law Society publishes the by-law and guidance, and that is where an opinion should come from.

We put ourselves on that list deliberately, ours. Most of the operational detail here is a commercial source's characterisation of the by-law, and this article is a starting point for a question rather than an answer to one.

What To Do

Complete the reconciliation within 25 days of the end of the statement period, not 25 days from month end and not from the date the statement arrives.

Reconcile every month, including months with no transactions, which Ontario requires and some jurisdictions do not.

Verify all three figures agree: adjusted bank balance, trust ledger balance, and the sum of every client ledger.

Add a control aimed at same-client misposting, because on our own analysis the three-way test cannot detect it.

Reverse cheques outstanding over six months and reinstate the client liability, and consider a stop payment before reissuing.

Keep trust records ten years plus current, which is longer than an ordinary business retention schedule.

Suppress card access on the trust account, since ATM withdrawals and transfers are prohibited.

And confirm your own province's rules, because the timing and procedure described here are Ontario's.

The Limits Of This Analysis

Several caveats matter. This article discusses professional regulation and bookkeeping practice and is not legal, accounting or compliance advice; a licensee should work from the current By-Law and their own law society's guidance. All rules were verified on 29 August 2026 and By-Law 9 is amended periodically. We did not obtain the full current text of By-Law 9. We obtained the Law Society of Ontario's frequently asked questions and its record keeping summary, which cite the operative provisions, and a Law Society copy of the by-law itself in fragments only; every substantive requirement described here is therefore the Law Society's guidance or a commercial source's characterisation rather than our own reading of the instrument, which is a real limitation on an article about compliance. The 25-day timing, the three-way structure, the prohibition on ATM withdrawals, the Form 9A and dual authorization requirements, and the cash and retention rules all reach us through commercial sources citing the Law Society, flagged individually in the text, and we did not verify any of them against the by-law. We did not establish which electronic transfer systems satisfy the security requirements, and say so rather than infer. We did not list the exceptions permitting cash above $7,500, deliberately, because an incomplete list would be worse than none. We researched no province other than Ontario beyond two contrasts noted in the text. The reconciliation logic table and the blind spot analysis are entirely our own reasoning from the structure of the test; no source we obtained states it, and a reader should verify it by working an example rather than accepting our account. All arithmetic is ours: the 400 matters, the residual rates and amounts, and the business-day conversions are invented, and the business-day figures ignore statutory holidays. And the four detective controls we suggest are our own recommendations for a gap we identified ourselves, not requirements of the by-law as we understand it.

Related Reading

Niche Industry Finance The Cliff At Fifteen Thousand Another industry where the rule is precise and the published explanations are not. Niche Industry Operational Finance Per Loaded Mile, Not Per Mile Published cost figures for Canadian owner-operators differ by a factor of five. Most are US data in US dollars. And you are paid per loaded mile while... Behavioural Finance When The Measure Eats The Goal A control that passes while the thing it protects against is happening.

Frequently Asked Questions

What does a trust reconciliation actually compare?

Three figures: the adjusted bank statement balance, the trust account's own journal or general ledger balance, and the combined total of all individual client trust ledger balances. All three must match. A report comparing only the first two is a bank reconciliation, not a trust reconciliation.

What error can the reconciliation not detect?

On our own analysis of the test's structure, a receipt or disbursement posted to the wrong client for the correct amount. One client ledger rises by exactly what another falls, so the client ledger total, the journal and the bank all remain correct and the reconciliation passes. No source we obtained states this; it follows from the arithmetic.

Why does that matter more than an ordinary posting error?

Because the firm is then holding one client's money against another client's ledger, and may disburse on the second matter using the first client's funds. That is the outcome trust accounting rules exist to prevent, and a clerical origin does not change it.

When is the reconciliation due in Ontario?

No later than 25 days after the end of the period covered by the financial institution's monthly statement. The clock runs from the statement period end, not from month end and not from the date the statement arrives, and completion is what must happen inside the window.

Is it required in a month with no transactions?

In Ontario, yes. Some jurisdictions differ: the Northwest Territories does not require monthly reconciliation when no activity has occurred and otherwise requires it at least annually.

How long must trust records be kept?

On a commercial summary we did not verify against the by-law, ten years plus current for trust and general account transactions and for receipts and disbursements of trust money, and six years plus current for inter-client ledger transfers and cash records. That is longer than an ordinary business retention schedule, so a general policy applied to trust records will destroy them early.

Can a sole practitioner satisfy the dual authorization requirement?

Not alone. Internet banking transfers require one person to input the transaction with a unique password and a second person to authorize with a different password. Two passwords held by one person is not dual authorization. For very small firms the practical answer is often to use cheques, which the licensee signs personally.

References

1. Law Society of Ontario, Frequently asked questions about reconciling a trust account, citing the operative provisions of By-Law 9 including section 14, section 18 and subsections 18(1) and (2), 18(8), 18(10), 22(2) and 23(2), covering the requirement itself, the minimum requirements for the comparison, the trust bank statement, and the retention of original or electronic image copies of records showing deposits. Note: the Law Society of Ontario's own guidance page and our source for the section numbering of the requirements. We obtained the page's citation structure rather than its full explanatory text. lso.ca
2. Law Society of Ontario, Summary of By-Law 9 Record Keeping Requirements, a practice resource for lawyers covering types of accounting systems, bank accounts in a law practice, general retainers, cash receipts, general accounts and trust accounts including mixed trust accounts. Note: the Law Society's own summary. We obtained its table of contents rather than its substantive text, and rely on it only to establish that the resource exists and what it covers. lso.ca
3. Law Society of Ontario, By-Law 9, Financial Transactions and Records, made 1 May 2007 and amended subsequently, obtained as a Law Society hosted PDF. Fragments obtained include the requirement for a detailed reconciliation made monthly of each trust bank account; provisions on special trust accounts related to a client's real estate transaction and the transfer of excess money inadvertently paid in; a time limit on holding money in a special trust account; and headings on the preservation of financial records required under sections 18, 19, 19.1 and 20. Note: a Law Society copy of the by-law itself. WE OBTAINED FRAGMENTS ONLY, not the full current text, which is the principal limitation on this article. lso.ca
4. Legal practice software vendor's guide to trust accounting rules by province, stating that the Law Society of Ontario requires monthly reconciliation but allows the process to be completed no later than 25 days after the end of the period covered by the financial institution's monthly statement; that the Northwest Territories does not require monthly reconciliations when no activity in the account has occurred and otherwise mandates reconciliation at least annually at the end of each fiscal year; and that the Law Society of Alberta operates a Trust Safety Program outlining compliance activities that go beyond standard reconciliation requirements. Note: a commercial software vendor, NOT the Law Society, flagged. Our source for the 25-day timing and the two provincial contrasts. Not verified against the by-law. clio.com
5. Legal information publication on trust account rules in Canada, citing the Law Society of Ontario, stating that the reconciliation compares the adjusted bank statement balance, the trust account's own journal or general ledger balance, and the combined total of all individual client ledger balances, and that all three must match; that in Ontario the reconciliation must be completed within 25 days of the end of each monthly bank statement period and is required every month even if no transactions occurred; that any discrepancy, shortage or overage, must be investigated and resolved immediately; that ATM withdrawals and transfers from a trust account are flatly prohibited in Ontario and funds may leave only by cheque or through electronic transfer systems meeting specific security requirements; and that internet banking transfers require a signed Form 9A requisition and dual authorization, with one person inputting the transaction using a unique password and a second person authorizing using a different password. Note: a commercial legal information site citing the Law Society, NOT the Law Society itself, flagged. Our source for the three-way structure and the transfer restrictions. Not verified against the by-law. legalclarity.org
6. Trust accounting software vendor's help documentation on monthly trust reconciliations for Ontario lawyers and paralegals, advising review of the trust comparison and all supporting documentation by the 25th of each month; that all client trust funds be included covering mixed accounts, pass books, GICs, term deposits, electronic registration, estate and power of attorney accounts; that reconciling items such as bank errors and posting errors be cleared each month and explained and supported by documentation; and that stale dated cheques outstanding more than six months be reversed, the client liability reinstated in the client's trust ledger and the cheque reissued if appropriate, noting that clearing rules permit a cheque more than six months old to be cashed so a firm should check its institution's policy on stop payments before reissuing. Note: a commercial software vendor's help documentation, NOT the Law Society, flagged. Our source for the stale-dated cheque treatment. trustreq.ca
7. Trust accounting software vendor's blog on cash handling rules for Ontario lawyers and paralegals, stating that a licensee may accept more than $7,500 Canadian in cash only in connection with specified exceptions; that foreign currency must not convert to more than $7,500 Canadian unless an exception applies; that By-Law 9 mandates specific record-keeping practices where cash is accepted; that a book of duplicate receipts must be maintained for every cash transaction; and that the cash receipt must be recorded in either the trust receipts journal or the general receipts journal depending on where the cash was deposited. Note: a commercial software vendor's blog, NOT the Law Society, flagged. Our source for the cash ceiling and its recording obligations. We deliberately did not reproduce the list of exceptions. trustreq.blog
8. Bookkeeping firm's summary of By-Law 9 recordkeeping obligations and retention periods, giving 10 years plus current for a record of all trust and general account transactions, for a record of all receipts and disbursements of trust money for clients including date, method, amount, document identifier and purpose, for a record of all money received and disbursed for each client's trust account including any unexpended balance, and for a record of signed electronic trust transfer requisitions and confirmations; and 6 years plus current for a record of all transfers of trust money between client trust ledger accounts including the purpose of each transfer and for a record of all cash received including date, person, amount, client, file number and signatures. Note: a commercial bookkeeping firm, NOT the Law Society, flagged. Our only source for the retention periods, which we did NOT verify against the current by-law. bookkeepingmatters.ca

This article discusses professional regulation and bookkeeping practice and is not legal, accounting or compliance advice. The full current text of By-Law 9 was not obtained. Most operational detail here comes from commercial sources citing the Law Society of Ontario rather than from the by-law, flagged individually. The reconciliation logic and blind spot analysis are the authors' own reasoning from the structure of the test and are stated as such. All arithmetic is the authors' own and every firm figure is invented.