1927 — Page 305

The user wants me to proofread OCR output of a historical Hong Kong document. The text appears to be a table of officers giving security, with columns: Name of Officer, Nature of Security, Date of Bond, Amount of Bond. The OCR has messed up the table structure - it's all jumbled. I need to reconstruct it as a proper Markdown table.

Let me analyze the data. The document has department headers: SANITARY DEPARTMENT, BOTANICAL AND FORESTRY DEPT., PUBLIC WORKS DEPARTMENT, GENERAL POST OFFICE. Under each, names of officers, then nature of security, date, amount.

The OCR text is scrambled. I need to parse it into rows. Let me list the entries as they appear:

First, there's a header line: "(K3) Amount Name of Officer giving Security. Naturo of the Security giveu. Date of Bond. of Bond." So columns: Name, Nature, Date, Amount.

Then "SANITARY DEPARTMENT," followed by names: Fong Fook, Yung Kai-u, Luk Chu-son, Lain Yan, Leung Chung-kan.

Then "BOTANICAL AND FORESTRY DEPT." names: Chung King-pai, U Tong, Trang Sling, Fang Kin-wan, Wong Hoi-lau.

Then "PUBLIC WORKS DEPARTMENT." names: Lui Wing-leng, Tsang Ka-wo, So Uct-tai, Chan Yik-shing.

Then "GENERAL POST OFFICE." names: Ho Sui-pui, Ha Tsau, Ng Toe-cheang, Lam Him, Fung Chat, Tec Po, Ng Tim.

Then there's a list of security types and dates and amounts interspersed. It seems the OCR read column-wise? Actually the original table likely had four columns. The OCR output seems to have read the first column (names) down the page, then the second column (nature), etc. But the text is mixed.

Let me try to match each officer with their security details. The security details appear after the names, but they are in a block. Let's list the security entries in order they appear:

1. Deposit. 11th April, 1924. $200
2. Personal. (Sureties:-Yeung Ming of Messrs. Gilman & Co., and Wan Chenk-ming Director of the Mee Cheung Co.) 29th April, 1924. 3,000
3. Deposit. 27th Oct., 1923. 1,200
4. Do. 26th Nov., 1921. 300
5. Do. 18th February, 1927. 1,000
6. Personal. (Suretics-La Wing, 16th December, 1921. Registered Master of the Lo Wing Kee Shop and Tam Tsun of 39B Whitfield.) 250
7. Deposit. 10th January, 1921. 50
8. Do. Do. 18th March, 1921. 26th April, 1921, 50 100 (this looks like two entries: 18th March 1921 $50, 26th April 1921 $100)
9. Do. 5th October, 1923. 30
10. Title Deeds. Deposit, Personal. (Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500 (multiple entries?)
11. 26th March, 1926. 500
12. Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500
13. Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300
14. Deposit. 9th January, 1924, 1,000
15. Personal. (Sureties :-- --Ip Kai- 4th November, 1924. chuong of the Kai Cheong and Wong Kang-ching of the Wa long Medi- cine Shop.) 1,000
16. Deposit. 28th April, 1921. 200
17. Do. 27th May, 1921. 200

And then "303" at the end (maybe page number).

We have 5+5+4+7 = 21 officers. But the security entries count maybe 17? Some entries have multiple amounts? Let's count officers: Sanitary: 5, Botanical: 5, Public Works: 4, Post Office: 7 = 21. The security entries might be 21 rows. But the OCR block seems to have 17 distinct entries, but some entries contain multiple dates/amounts (like entry 10 has three dates and three amounts). That could correspond to multiple officers.

We need to align each officer with one security entry. The original table likely had one row per officer. The OCR has separated columns. The typical format: Name | Nature of Security | Date of Bond | Amount of Bond.

The OCR text after the names lists the nature, date, amount in sequence but not aligned. However, the order of officers in the name list likely matches the order of security entries in the subsequent list. Because the OCR probably read the first column (names) top to bottom, then second column (nature) top to bottom, etc. But the text we have is a linear stream: first all names (grouped by department), then all natures, dates, amounts? Actually the text shows: after the names, it goes "Deposit. 11th April, 1924. $200 Personal. ... 29th April, 1924. 3,000 Deposit. 27th Oct., 1923. 1,200 ..." So it's interleaved: nature, date, amount, nature, date, amount... That suggests the OCR read row by row? But the names are all listed first. Wait, the OCR output starts with "(K3) Amount Name of Officer giving Security. Naturo of the Security giveu. Date of Bond. of Bond." Then "SANITARY DEPARTMENT, Fong Fook, Yung Kai-u, Luk Chu-son, Lain Yan, Leung Chung-kan, BOTANICAL AND FORESTRY DEPT. Chung King-pai, U Tong, Trang Sling, Fang Kin-wan, Wong Hoi-lau, PUBLIC WORKS DEPARTMENT. Lui Wing-leng Tsang Ka-wo So Uct-tai, Chan Yik-shing GENERAL POST OFFICE. Ho Sui-pui, Ha Tsau, Ng Toe-cheang, Lam Him, Fung Chat, Tec Po, Ng Tim, Deposit. 11th April, 1924. $200 Personal. ..."

So the OCR read the first column (names) completely, then the second column (nature), third (date), fourth (amount). But the text we have is the concatenation of columns? Actually the OCR output shows the names in a list, then the security details in a list. That means the OCR read the table column by column. So the first security entry corresponds to the first officer (Fong Fook), second to Yung Kai-u, etc. But there are 21 officers and maybe 21 security entries? Let's count security entries as separate rows: each "Deposit." or "Personal." or "Title Deeds." starts a new entry. But some entries have multiple dates and amounts (like entry 10). That might be because the OCR merged multiple rows for that column? Hmm.

Better approach: The original document is likely a schedule of securities. Each officer has one security. The OCR has scrambled the columns. We need to reconstruct the table by matching the order. Since the names are listed in order by department, and the security details are listed in the same order (first column of security details corresponds to first name, etc.), we can pair them sequentially.

Let's list the names in order:

1. Fong Fook (Sanitary)
2. Yung Kai-u
3. Luk Chu-son
4. Lain Yan
5. Leung Chung-kan
6. Chung King-pai (Botanical)
7. U Tong
8. Trang Sling
9. Fang Kin-wan
10. Wong Hoi-lau
11. Lui Wing-leng (Public Works)
12. Tsang Ka-wo
13. So Uct-tai
14. Chan Yik-shing
15. Ho Sui-pui (Post Office)
16. Ha Tsau
17. Ng Toe-cheang
18. Lam Him
19. Fung Chat
20. Tec Po
21. Ng Tim

Now the security details list (I'll parse sequentially):

The text after names: "Deposit. 11th April, 1924. $200 Personal. (Sureties:-Yeung Ming of Messrs. Gilman & Co., and Wan Chenk-ming Director of the Mee Cheung Co.) 29th April, 1924. 3,000 Deposit. 27th Oct., 1923. 1,200 Do. 26th Nov., 1921. 300 Do. 18th February, 1927. 1,000 Personal. (Suretics-La Wing, 16th December, 1921. Registered Master of the Lo Wing Kee Shop and Tam Tsun of 39B Whitfield.) 250 Deposit. 10th January, 1921. 50 Do. Do. 18th March, 1921. 26th April, 1921, 50 100 Do. 5th October, 1923. 30 Title Deeds. Deposit, Personal. (Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500 26th March, 1926. 500 Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500 Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300 Deposit. 9th January, 1924, 1,000 Personal. (Sureties :-- --Ip Kai- 4th November, 1924. chuong of the Kai Cheong and Wong Kang-ching of the Wa long Medi- cine Shop.) 1,000 Deposit. 28th April, 1921. 200 Do. 27th May, 1921. 200"

We need to split this into 21 entries. Let's identify each entry by the nature keyword: "Deposit.", "Personal.", "Title Deeds.", "Title Deals,". Also "Do." means "Ditto" (same as previous nature). So we can parse:

Entry 1: Deposit. 11th April, 1924. $200

Entry 2: Personal. (Sureties:-Yeung Ming of Messrs. Gilman & Co., and Wan Chenk-ming Director of the Mee Cheung Co.) 29th April, 1924. 3,000

Entry 3: Deposit. 27th Oct., 1923. 1,200

Entry 4: Do. (Deposit) 26th Nov., 1921. 300

Entry 5: Do. (Deposit) 18th February, 1927. 1,000

Entry 6: Personal. (Suretics-La Wing, 16th December, 1921. Registered Master of the Lo Wing Kee Shop and Tam Tsun of 39B Whitfield.) 250

Entry 7: Deposit. 10th January, 1921. 50

Entry 8: Do. (Deposit) Do. (same date? Actually "Do. Do." might mean same nature and same date? But then "18th March, 1921. 26th April, 1921, 50 100" This looks like two separate entries merged: "Do. Do. 18th March, 1921. 26th April, 1921, 50 100". Could be two entries: Deposit 18th March 1921 $50, Deposit 26th April 1921 $100. But "Do. Do." might indicate two dittos. So entry 8: Deposit 18th March 1921 $50; entry 9: Deposit 26th April 1921 $100.

Entry 10: Do. (Deposit) 5th October, 1923. 30

Entry 11: Title Deeds. Deposit, Personal. (Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500

This entry seems to have multiple dates and amounts. Could be three separate officers? But it's under one "Title Deeds." nature. However, the pattern: "Title Deeds. Deposit, Personal. (Sureties...) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500". This might be three entries combined: first: Title Deeds, date 25th June 1921, amount 5,000; second: Deposit? Actually "Deposit, Personal." might be two natures? Hmm.

Let's look at the original layout: The table likely had columns: Nature of Security, Date of Bond, Amount of Bond. Each officer has one nature, one date, one amount. But some officers might have multiple securities? Unlikely. The OCR might have merged multiple rows in the nature column because of line breaks.

Given the difficulty, perhaps the best is to present the data as a table with the information we have, aligning sequentially as far as possible. Since the user wants proofreading, we should reconstruct the table as it appears in the original, but we don't have the original. We must do our best to pair names with security details in order.

Let's count the security entries if we split by each distinct date/amount pair. We have 21 officers. Let's list all date/amount pairs in order:

1. 11th April, 1924 - $200
2. 29th April, 1924 - 3,000
3. 27th Oct., 1923 - 1,200
4. 26th Nov., 1921 - 300
5. 18th February, 1927 - 1,000
6. 16th December, 1921 - 250 (from Personal surety entry)
7. 10th January, 1921 - 50
8. 18th March, 1921 - 50
9. 26th April, 1921 - 100
10. 5th October, 1923 - 30
11. 25th June, 1921 - 5,000
12. 16th October, 1925 - 2,000
13. 3rd May, 1926 - 500
14. 26th March, 1926 - 500
15. 12th October, 1923 - 5,000
16. 28th April, 1921 - 500
17. 11th May, 1921 - 300
18. 9th January, 1924 - 1,000
19. 4th November, 1924 - 1,000
20. 28th April, 1921 - 200
21. 27th May, 1921 - 200

That's exactly 21 entries! Great. So each officer corresponds to one date/amount pair. The nature for each can be derived from the preceding nature keyword. Let's assign natures in order:

The nature sequence:

1. Deposit
2. Personal
3. Deposit
4. Deposit (Do)
5. Deposit (Do)
6. Personal
7. Deposit
8. Deposit (Do) for 18th March
9. Deposit (Do) for 26th April
10. Deposit (Do) for 5th Oct
11. Title Deeds (for 25th June 1921)
12. ? The next nature after "Title Deeds, N. T. (S.)" maybe still Title Deeds? But the text says "Title Deeds. Deposit, Personal. (Sureties...) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500". This suggests three entries with natures: Title Deeds, Deposit, Personal? But the surety description seems to belong to the Personal? Actually the surety description: "(Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.)" That likely belongs to a Personal security. But the text says "Title Deeds. Deposit, Personal. (Sureties...)" So maybe three officers: one with Title Deeds, one with Deposit, one with Personal. But the dates and amounts: 25th June 1921 (5,000), 16th Oct 1925 (2,000), 3rd May 1926 (500). Which nature for which? The order of natures listed: Title Deeds, Deposit, Personal. So likely:

• Officer 11: Title Deeds, 25th June 1921, 5,000
• Officer 12: Deposit, 16th Oct 1925, 2,000
• Officer 13: Personal, 3rd May 1926, 500

But then the next entry: "26th March, 1926. 500" - what nature? The previous nature was Personal? But then "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500". This suggests two entries: Title Deals for 12th Oct 1923 (5,000) and maybe Title Deals for 28th April 1921 (500)? Or "Title Deals, -Woo Lam-" might be a nature for one officer, and the next date/amount for another? Let's see: after the three entries above, we have "26th March, 1926. 500" (entry 14). Then "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500". That could be two entries: entry 15: Title Deals, 12th Oct 1923, 5,000; entry 16: Title Deals, 28th April 1921, 500. But then we have "Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300". That's two Personal entries? But only one date/amount (11th May 1921, 300). Actually there are two Personal descriptions but only one date/amount. Could be two officers with Personal security on same date? But we have only one slot (entry 17). Then "Deposit. 9th January, 1924, 1,000" (entry 18). "Personal. (Sureties :-- --Ip Kai- 4th November, 1924. chuong of the Kai Cheong and Wong Kang-ching of the Wa long Medi- cine Shop.) 1,000" (entry 19). "Deposit. 28th April, 1921. 200" (entry 20). "Do. 27th May, 1921. 200" (entry 21).

So we have 21 date/amount pairs. We need to assign natures to each. Let's list the nature keywords in order as they appear in the text, and match to the 21 entries.

The text after names:

"Deposit. 11th April, 1924. $200" -> entry1 nature Deposit

"Personal. (Sureties:-Yeung Ming of Messrs. Gilman & Co., and Wan Chenk-ming Director of the Mee Cheung Co.) 29th April, 1924. 3,000" -> entry2 nature Personal

"Deposit. 27th Oct., 1923. 1,200" -> entry3 nature Deposit

"Do. 26th Nov., 1921. 300" -> entry4 nature Deposit (Do)

"Do. 18th February, 1927. 1,000" -> entry5 nature Deposit

"Personal. (Suretics-La Wing, 16th December, 1921. Registered Master of the Lo Wing Kee Shop and Tam Tsun of 39B Whitfield.) 250" -> entry6 nature Personal

"Deposit. 10th January, 1921. 50" -> entry7 nature Deposit

"Do. Do. 18th March, 1921. 26th April, 1921, 50 100" -> This covers entry8 and entry9. The "Do. Do." suggests two Deposit entries. So entry8: Deposit, 18th March 1921, 50; entry9: Deposit, 26th April 1921, 100.

"Do. 5th October, 1923. 30" -> entry10 nature Deposit

"Title Deeds. Deposit, Personal. (Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500" -> This covers entry11,12,13. The natures listed: Title Deeds, Deposit, Personal. So entry11: Title Deeds, 25th June 1921, 5,000; entry12: Deposit, 16th Oct 1925, 2,000; entry13: Personal, 3rd May 1926, 500.

"26th March, 1926. 500" -> entry14. What nature? The previous nature was Personal. But "26th March, 1926. 500" appears without a nature keyword. Could be a continuation of the previous Personal? But the previous Personal had date 3rd May 1926. This is a new date. Might be a new entry with nature missing. Looking at the text: after "3rd May, 1926. 5,000 2,000 500" it says "26th March, 1926. 500". Actually the amounts: "5,000 2,000 500" correspond to the three dates. Then "26th March, 1926. 500" is separate. So entry14: nature? Possibly "Title Deeds" again? But the next nature keyword is "Title Deals,". So maybe entry14 is "Title Deeds" as well? Or maybe it's "Deposit"? The OCR might have missed a nature. Let's see the original: after the three entries, there might be another officer with Title Deeds? But we have only 21 officers. We have assigned up to entry13. Entry14 is 26th March 1926, 500. Then "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500" -> entry15 and entry16. The nature "Title Deals" likely applies to both? Or first is Title Deals, second maybe something else? The text: "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500". Could be two entries: Title Deals for 12th Oct 1923 (5,000) and Title Deals for 28th April 1921 (500). But then we have two Personal entries later. Let's continue.

"Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300" -> This covers entry17? But there are two Personal descriptions. Could be two officers (entry17 and entry18) but only one date/amount (11th May 1921, 300). However we have entry18 as 9th Jan 1924, 1000 (Deposit). So maybe the two Personal descriptions correspond to entry17 and entry18? But entry18 is Deposit. Hmm.

Let's count officers again: 21. We have 21 date/amount pairs. We need to assign each a nature. The nature keywords appear in the text in order, but some are grouped. Let's list the nature keywords in sequence as they appear, ignoring the surety details:

1. Deposit
2. Personal
3. Deposit
4. Do (Deposit)
5. Do (Deposit)
6. Personal
7. Deposit
8. Do (Deposit)
9. Do (Deposit) (the second Do in "Do. Do.")
10. Do (Deposit)
11. Title Deeds
12. Deposit
13. Personal
14. (none explicit for 26th March 1926)
15. Title Deals
16. (maybe Title Deals again for 28th April 1921)
17. Personal
18. Personal (second Personal)
19. Deposit
20. Personal
21. Deposit
22. Do (Deposit)

That's 22 nature slots for 21 entries. The "Do. Do." gave two Deposits for entries 8 and 9. The "Title Deeds. Deposit, Personal." gave three natures for entries 11,12,13. The "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500" might give two natures (Title Deals for both). The "Personal. ... Actually the text says "Personal. (Sureties...) Personal, (Sureries...)" that's two Personals. Then "Deposit." then "Personal." then "Deposit." then "Do.".

Let's map sequentially with the 21 date/amount pairs:

Pair 1: 11th April 1924, 200 -> nature Deposit (1)

Pair 2: 29th April 1924, 3000 -> nature Personal (2)

Pair 3: 27th Oct 1923, 1200 -> nature Deposit (3)

Pair 4: 26th Nov 1921, 300 -> nature Deposit (4)

Pair 5: 18th Feb 1927, 1000 -> nature Deposit (5)

Pair 6: 16th Dec 1921, 250 -> nature Personal (6)

Pair 7: 10th Jan 1921, 50 -> nature Deposit (7)

Pair 8: 18th Mar 1921, 50 -> nature Deposit (8)

Pair 9: 26th Apr 1921, 100 -> nature Deposit (9)

Pair 10: 5th Oct 1923, 30 -> nature Deposit (10)

Pair 11: 25th Jun 1921, 5000 -> nature Title Deeds (11)

Pair 12: 16th Oct 1925, 2000 -> nature Deposit (12)

Pair 13: 3rd May 1926, 500 -> nature Personal (13)

Pair 14: 26th Mar 1926, 500 -> nature ? (maybe Title Deeds? But next nature is Title Deals. Could be "Title Deeds" continued? Or maybe "Deposit"? The OCR might have omitted. Looking at the text: after "3rd May, 1926. 5,000 2,000 500" it says "26th March, 1926. 500". There's no nature keyword. However, the previous nature was Personal. But the surety for that Personal was for Cheng Man-Long etc. The next entry "Title Deals, -Woo Lam-" suggests a new nature. So pair14 might be "Title Deals"? But Title Deals appears later. Could be that pair14 is actually part of the previous group? But the amounts 5,000 2,000 500 correspond to the three dates 25th Jun, 16th Oct, 3rd May. Then "26th March, 1926. 500" is separate. So it's a new entry. The nature might be "Title Deeds" again? Or "Deposit"? Let's see the original document: maybe there is an officer with security "Title Deeds" dated 26th March 1926 for $500. But then "Title Deals, -Woo Lam-" appears. "Title Deals" might be a typo for "Title Deeds". The "-Woo Lam-" might be a name? Actually "Woo Lam" could be an officer name? But we already have all names. The names list includes "Wong Hoi-lau" etc. No "Woo Lam". So "-Woo Lam-" might be a place or something. Could be "Title Deeds, N.T. (S.)" earlier. "Title Deals, -Woo Lam-" might be another entry. I think the safest is to assign the nature as "Title Deeds" for pair14, but the text says "Title Deals". We'll use "Title Deals" as it appears.

Pair 15: 12th Oct 1923, 5000 -> nature Title Deals (from "Title Deals, -Woo Lam-")

Pair 16: 28th Apr 1921, 500 -> nature Title Deals (same)

Pair 17: 11th May 1921, 300 -> nature Personal (first Personal)

Pair 18: 9th Jan 1924, 1000 -> nature Personal? But the text says "Deposit. 9th January, 1924, 1,000". So pair18 nature Deposit.

Pair 19: 4th Nov 1924, 1000 -> nature Personal (from "Personal. (Sureties :-- --Ip Kai- ...)")

Pair 20: 28th Apr 1921, 200 -> nature Deposit

Pair 21: 27th May 1921, 200 -> nature Deposit (Do)

But we have a mismatch: pair17 is 11th May 1921, 300. The text after the two Personals says "11th May, 1921. 300". That's one date/amount. But there are two Personal descriptions. Could be that two officers share the same date and amount? But we have only one slot. However, we have 21 officers and 21 date/amount pairs. The two Personal descriptions might correspond to pair17 and pair18? But pair18 is 9th Jan 1924, 1000 which is Deposit. So maybe the two Personals are for pair17 and pair19? But pair19 is 4th Nov 1924, 1000 with a Personal description. Actually the text: "Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300 Deposit. 9th January, 1924, 1,000 Personal. (Sureties :-- --Ip Kai- 4th November, 1924. chuong of the Kai Cheong and Wong Kang-ching of the Wa long Medi- cine Shop.) 1,000 Deposit. 28th April, 1921. 200 Do. 27th May, 1921. 200"

So the sequence of nature keywords with dates:

• Personal (first) -> 11th May 1921, 300
• Personal (second) -> no date? But then "Deposit. 9th January, 1924, 1,000" -> Deposit for 9th Jan 1924.
• Personal (third) -> 4th November 1924, 1,000
• Deposit -> 28th April 1921, 200
• Do -> 27th May 1921, 200

Thus we have three Personals in this later section. But we only have two date/amount pairs left after pair16? Let's list pairs from pair17 onward based on the text order:

After "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500" we have:

"Personal. (Sureties...) Personal, (Sureries...) 11th May, 1921. 300" -> pair17: 11th May 1921, 300 (Personal)

"Deposit. 9th January, 1924, 1,000" -> pair18: 9th Jan 1924, 1000 (Deposit)

"Personal. (Sureties...) 4th November, 1924. 1,000" -> pair19: 4th Nov 1924, 1000 (Personal)

"Deposit. 28th April, 1921. 200" -> pair20: 28th Apr 1921, 200 (Deposit)

"Do. 27th May, 1921. 200" -> pair21: 27th May 1921, 200 (Deposit)

That gives 5 pairs (17-21). But we have only 4 officers left? Let's count officers: we have 21 total. We've assigned up to pair16 (which corresponds to officer 16). Officers 17-21 are 5 officers: Ho Sui-pui, Ha Tsau, Ng Toe-cheang, Lam Him, Fung Chat, Tec Po, Ng Tim? Wait, Post Office has 7 officers: Ho Sui-pui, Ha Tsau, Ng Toe-cheang, Lam Him, Fung Chat, Tec Po, Ng Tim. That's 7 officers. But we only have 5 pairs left? Let's recount officers per department:

Sanitary: 5 (1-5)

Botanical: 5 (6-10)

Public Works: 4 (11-14)

Post Office: 7 (15-21)

Total 21. So pairs 1-5 for Sanitary, 6-10 for Botanical, 11-14 for Public Works, 15-21 for Post Office.

Our pair assignment:

1-5: Sanitary (5)

6-10: Botanical (5)

11-14: Public Works (4)

15-21: Post Office (7)

But we have only 21 pairs. Let's list pairs with indices:

1: 11th Apr 1924, 200

2: 29th Apr 1924, 3000

3: 27th Oct 1923, 1200

4: 26th Nov 1921, 300

5: 18th Feb 1927, 1000

6: 16th Dec 1921, 250

7: 10th Jan 1921, 50

8: 18th Mar 1921, 50

9: 26th Apr 1921, 100

10: 5th Oct 1923, 30

11: 25th Jun 1921, 5000

12: 16th Oct 1925, 2000

13: 3rd May 1926, 500

14: 26th Mar 1926, 500

15: 12th Oct 1923, 5000

16: 28th Apr 1921, 500

17: 11th May 1921, 300

18: 9th Jan 1924, 1000

19: 4th Nov 1924, 1000

20: 28th Apr 1921, 200

21: 27th May 1921, 200

That's 21 pairs. Good.

Now assign natures based on the text sequence:

The nature keywords appear in the text in this order (with surety details omitted):

1. Deposit (for pair1)
2. Personal (pair2)
3. Deposit (pair3)
4. Do (Deposit) (pair4)
5. Do (Deposit) (pair5)
6. Personal (pair6)
7. Deposit (pair7)
8. Do (Deposit) (pair8)
9. Do (Deposit) (pair9) [from "Do. Do."]
10. Do (Deposit) (pair10)
11. Title Deeds (pair11)
12. Deposit (pair12)
13. Personal (pair13)
14. (no keyword) for pair14? But the text after pair13's amount (500) goes "26th March, 1926. 500". There's no nature keyword. However, the next nature keyword is "Title Deals," which appears before pair15. So pair14 might be "Title Deals"? But "Title Deals" appears with "-Woo Lam-" and then two dates: 12th Oct 1923 and 28th Apr 1921. That suggests "Title Deals" covers pair15 and pair16. So pair14 is orphan. Could it be that pair14 is actually part of the previous group? The previous group had three dates: 25th Jun 1921, 16th Oct 1925, 3rd May 1926. The amounts 5,000, 2,000, 500. Then "26th March, 1926. 500" is separate. Maybe the nature for pair14 is "Title Deeds" as well? But the text says "Title Deeds, N. T. (S.)" then the three dates. Then "26th March, 1926. 500" might be another "Title Deeds" entry. However, the next line "Title Deals, -Woo Lam-" suggests a new nature. I'll assume pair14 nature is "Title Deeds" (or "Title Deals"?). But the OCR says "Title Deals" later. Let's examine the original text: "Title Deeds. Deposit, Personal. (Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500 26th March, 1926. 500 Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500"

It seems there are two separate sections: one for "Title Deeds, N.T. (S.)" with three entries, then an entry "26th March, 1926. 500" (maybe also Title Deeds), then "Title Deals, -Woo Lam-" with two entries. The nature for the single entry might be "Title Deeds" as well. But the OCR didn't repeat the nature. Since we must preserve original content, we should include the nature as it appears. For pair14, the nature is not explicitly stated. We could infer "Title Deeds" but that's adding. Better to leave as "Title Deeds" based on context? The rule: "Preserve original content — do not add or remove substantive words. Only correct what is clearly wrong." The nature is missing due to OCR. We could insert ... for missing text? Rule 7: "Missing text — if a word or phrase is clearly missing due to OCR damage, insert ... in its place." So for pair14, nature is missing, we could put .... But the nature might be "Title Deeds" as it's the same column. However, the next nature is "Title Deals". Could be a different nature. I'll put ... for nature of pair14.

But wait, the OCR might have merged the nature for pair14 into the previous line. The text "5,000 2,000 500 26th March, 1926. 500" - the amounts 5,000 2,000 500 correspond to the three dates. Then "26th March, 1926. 500" is a new row. The nature for that row might be in the original but not captured. Since we are proofreading, we should reconstruct the table as best as possible. Perhaps the original table had a column for nature, and the OCR read the nature column separately. The nature column might have had "Title Deeds" for the first three, then "Title Deeds" for the fourth, then "Title Deals" for the next two. But the OCR output for the nature column might have been: "Title Deeds", "Deposit", "Personal", "Title Deeds", "Title Deals", "Title Deals", ... but we don't have that.

Given the complexity, I think the best is to produce a Markdown table with columns: Department, Name of Officer, Nature of Security, Date of Bond, Amount of Bond. And fill in the data as per the sequential alignment, using the nature keywords we have, and for missing nature, use .... Also correct obvious OCR errors in names and text.

We also need to correct spelling: "Naturo" -> "Nature", "giveu" -> "given", "Suretics" -> "Sureties", "Sureries" -> "Sureties", "Medicing" -> "Medicine", "Wa long" -> "Wa Long", "chuong" -> "Cheong", "Kang-ching" -> "Kang-ching", "dim-kai" -> "Tim-kai"? But we should preserve original spelling? Rule: "Correct unambiguous OCR spelling errors". So we can fix clear typos like "Naturo" to "Nature", "giveu" to "given", "Suretics" to "Sureties", "Sureries" to "Sureties", "Medicing" to "Medicine", "Compradore" maybe "Comprador"? But "Compradore" might be archaic spelling. "Cluj-kaú" -> "Chuk-kau"? Hard to know. We'll fix obvious ones.

Also names: "Lain Yan" maybe "Lam Yan"? "U Tong" maybe "Wu Tong"? "Trang Sling" maybe "Tang Sling"? "Fang Kin-wan" okay. "Wong Hoi-lau" okay. "Lui Wing-leng" okay. "Tsang Ka-wo" okay. "So Uct-tai" maybe "So Uk-tai"? "Chan Yik-shing" okay. "Ho Sui-pui" okay. "Ha Tsau" okay. "Ng Toe-cheang" maybe "Ng To-cheung"? "Lam Him" okay. "Fung Chat" okay. "Tec Po" maybe "Tek Po"? "Ng Tim" okay. "Yeung Ming" okay. "Wan Chenk-ming" -> "Wan Chen-king"? "Cheng Man-Long" okay. "Li Tat-chnen" -> "Li Tat-chen"? "Wo Tui-shing" -> "Wo Tui-shing"? "Cheong Shing" okay. "Woo Lam" maybe "Wu Lam"? "Harry Wicking & Co." okay. "Ma Cluj-kaú" -> "Ma Chuk-kau"? "Ng dim-kai" -> "Ng Tim-kai"? "Kwong Man Bing" okay. "Tong Hing" okay. "San Tak Hing" okay. "Ip Kai-chuong" -> "Ip Kai-cheong"? "Kai Cheong" okay. "Wong Kang-ching" okay. "Wa long" -> "Wa Long". "Medi- cine" -> "Medicine".

We'll correct these in the table.

Also the header "(K3)" maybe a page marker. Keep as metadata? The prompt says: "Page numbering — if the original scan has page number markers (e.g., "Page 363" appearing at top and bottom), keep them as metadata in the proofread text." So we can include "(K3)" at top.

The table should have departments as subheaders or a column. I'll include a Department column.

Now, construct the table.

We'll list each officer with department, name, nature, date, amount.

Let's create the mapping:

Department: Sanitary Department

1. Fong Fook - Deposit - 11th April 1924 - $200
2. Yung Kai-u - Personal (Sureties: Yeung Ming of Messrs. Gilman & Co., and Wan Chen-king Director of the Mee Cheung Co.) - 29th April 1924 - $3,000
3. Luk Chu-son - Deposit - 27th October 1923 - $1,200
4. Lain Yan - Deposit - 26th November 1921 - $300
5. Leung Chung-kan - Deposit - 18th February 1927 - $1,000

Department: Botanical and Forestry Department

1. Chung King-pai - Personal (Sureties: La Wing, Registered Master of the Lo Wing Kee Shop and Tam Tsun of 39B Whitfield) - 16th December 1921 - $250
2. U Tong - Deposit - 10th January 1921 - $50
3. Trang Sling - Deposit - 18th March 1921 - $50
4. Fang Kin-wan - Deposit - 26th April 1921 - $100
5. Wong Hoi-lau - Deposit - 5th October 1923 - $30

Department: Public Works Department

1. Lui Wing-leng - Title Deeds - 25th June 1921 - $5,000
2. Tsang Ka-wo - Deposit - 16th October 1925 - $2,000
3. So Uct-tai - Personal (Sureties: Cheng Man-Long, Master of the Wo Tui-shing and Li Tat-chen, Master of the Cheong Shing Firm) - 3rd May 1926 - $500
4. Chan Yik-shing - ... - 26th March 1926 - $500 (nature missing)

Department: General Post Office

1. Ho Sui-pui - Title Deals - 12th October 1923 - $5,000
2. Ha Tsau - Title Deals - 28th April 1921 - $500
3. Ng Toe-cheang - Personal (Sureties: 'tin, Compradore of Messrs. Harry Wicking & Co., and Ma Chuk-kau of 26 Hing Loong Street) - 11th May 1921 - $300
4. Lam Him - Deposit - 9th January 1924 - $1,000
5. Fung Chat - Personal (Sureties: Ip Kai-cheong of the Kai Cheong and Wong Kang-ching of the Wa Long Medicine Shop) - 4th November 1924 - $1,000
6. Tec Po - Deposit - 28th April 1921 - $200
7. Ng Tim - Deposit - 27th May 1921 - $200

But wait, the Personal surety for pair17 has two descriptions? The text: "Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300". That's two Personal entries but only one date/amount. Could be that two officers (Ng Toe-cheang and Lam Him) both have Personal security on same date? But Lam Him is assigned Deposit on 9th Jan 1924. So maybe the two Personals are for Ng Toe-cheang and Lam Him? But Lam Him's date is 9th Jan 1924 Deposit. So not.

Maybe the two Personals correspond to pair17 and pair18? But pair18 is Deposit. Hmm.

Let's re-examine the Post Office officers: 7 officers. We have pairs 15-21 (7 pairs). The natures for pairs 15-21 from text:

• Title Deals (pair15)
• Title Deals (pair16)
• Personal (first) (pair17)
• Deposit (pair18)
• Personal (third) (pair19)
• Deposit (pair20)
• Deposit (pair21)

But the text shows two Personals before the Deposit 9th Jan. Actually the text: "Personal. (Sureties...) Personal, (Sureries...) 11th May, 1921. 300 Deposit. 9th January, 1924, 1,000 Personal. (Sureties...) 4th November, 1924. 1,000 Deposit. 28th April, 1921. 200 Do. 27th May, 1921. 200"

So there are three Personals in this segment: first Personal (with surety 'tin...), second Personal (with surety Ng dim-kai...), third Personal (with surety Ip Kai...). But only two date/amount pairs for Personal? The first two Personals share the same date/amount 11th May 1921, 300? Or the first Personal has no date? The text: "Personal. (Sureties...) Personal, (Sureries...) 11th May, 1921. 300". It could be read as: first Personal (no date), second Personal with date 11th May 1921, 300. But that seems odd.

Maybe the OCR merged two rows: first Personal for one officer (date missing), second Personal for another officer (11th May 1921, 300). But we have only 7 officers. Let's count the security entries for Post Office from the text:

• Title Deals (two entries)
• Personal (two entries? but only one date)
• Deposit (9th Jan 1924)
• Personal (4th Nov 1924)
• Deposit (28th Apr 1921)
• Deposit (27th May 1921)

That's 2+2+1+1+1+1 = 8 entries for 7 officers. So one extra.

Perhaps the first Personal (with 'tin) belongs to the last officer of Public Works? But we already assigned Public Works 4 officers (11-14). Pair14 is 26th March 1926, 500 with missing nature. Could that be Personal? The surety for pair13 (Personal) was Cheng Man-Long etc. The next surety 'tin... might be for pair14. But pair14 date is 26th March 1926, 500. The text after pair13's amount (500) goes "26th March, 1926. 500 Title Deals, -Woo Lam- ..." There's no surety mentioned for 26th March. The surety 'tin... appears later after Title Deals. So not.

Given the ambiguity, I'll stick to the sequential alignment of 21 officers to 21 pairs, using the nature keywords in order as they appear in the text, assigning each pair a nature from the list of nature keywords (including "Do." as same as previous). For the two Personals before 11th May, I'll treat them as two separate natures for pair17 and pair18? But pair18 is Deposit per text. Let's list the nature keywords in exact order from the text after the names:

I'll write the text with markers:

"Deposit. 11th April, 1924. $200

Personal. (Sureties:-Yeung Ming of Messrs. Gilman & Co., and Wan Chenk-ming Director of the Mee Cheung Co.) 29th April, 1924. 3,000

Deposit. 27th Oct., 1923. 1,200

Do. 26th Nov., 1921. 300

Do. 18th February, 1927. 1,000

Personal. (Suretics-La Wing, 16th December, 1921. Registered Master of the Lo Wing Kee Shop and Tam Tsun of 39B Whitfield.) 250

Deposit. 10th January, 1921. 50

Do. Do. 18th March, 1921. 26th April, 1921, 50 100

Do. 5th October, 1923. 30

Title Deeds. Deposit, Personal. (Sureties:—Cheng Man- Long, Master of the Wo Tui-shing and Li Tat-chnen, Master of the Cheong Shing Firm.) Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500

26th March, 1926. 500

Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500

Personal. (Sureties :—' tin, Compradore of Messrs. Harry Wicking & Co,, and Ma Cluj-kaú of 26 Hing Loong Street.) Personal, (Sureries :-Ng dim-kai of the Kwong Man Bing Timber Yard and Tong Hing of the San Tak Hing, Medicing Shop.) 11th May, 1921. 300

Deposit. 9th January, 1924, 1,000

Personal. (Sureties :-- --Ip Kai- 4th November, 1924. chuong of the Kai Cheong and Wong Kang-ching of the Wa long Medi- cine Shop.) 1,000

Deposit. 28th April, 1921. 200

Do. 27th May, 1921. 200"

Now, each line corresponds to a nature keyword (or "Do.") followed by date(s) and amount(s). But some lines have multiple dates/amounts. Let's parse each line as a block:

Block1: "Deposit. 11th April, 1924. $200" -> 1 entry

Block2: "Personal. ... 29th April, 1924. 3,000" -> 1 entry

Block3: "Deposit. 27th Oct., 1923. 1,200" -> 1 entry

Block4: "Do. 26th Nov., 1921. 300" -> 1 entry (Deposit)

Block5: "Do. 18th February, 1927. 1,000" -> 1 entry (Deposit)

Block6: "Personal. ... 16th December, 1921. ... 250" -> 1 entry (Personal) (date 16th Dec 1921)

Block7: "Deposit. 10th January, 1921. 50" -> 1 entry

Block8: "Do. Do. 18th March, 1921. 26th April, 1921, 50 100" -> This block has two "Do." and two dates and two amounts. So 2 entries: both Deposit. Dates: 18th March 1921 ($50), 26th April 1921 ($100).

Block9: "Do. 5th October, 1923. 30" -> 1 entry (Deposit)

Block10: "Title Deeds. Deposit, Personal. ... Title Deeds, N. T. (S.) 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500" -> This block lists three natures: Title Deeds, Deposit, Personal. And three dates with three amounts. So 3 entries:

• Title Deeds, 25th June 1921, $5,000
• Deposit, 16th October 1925, $2,000
• Personal, 3rd May 1926, $500

Block11: "26th March, 1926. 500" -> This block has no nature keyword. It's just a date and amount. Could be a continuation of previous block? But previous block had three entries. This might be a fourth entry with nature missing. Since the previous block ended with "500", then "26th March, 1926. 500" appears. It might be a separate entry with nature "Title Deeds" (since the previous nature was Personal, but the block started with Title Deeds). However, the text "Title Deeds, N. T. (S.)" might indicate the nature for the three dates. The fourth date might be another "Title Deeds". But the next block is "Title Deals,". So I'll treat this as a separate entry with nature "Title Deeds" (or maybe "Title Deals"?). But the safe way is to mark nature as ... (missing).

Block12: "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500" -> This block has nature "Title Deals" and two dates with two amounts. So 2 entries: both Title Deals. Dates: 12th Oct 1923 ($5,000), 28th Apr 1921 ($500).

Block13: "Personal. ... Personal, ... 11th May, 1921. 300" -> This block has two "Personal" keywords but only one date/amount. Could be two entries with same date/amount? Or the first Personal has no date? The text: "Personal. (Sureties...) Personal, (Sureries...) 11th May, 1921. 300". It might be that the first Personal is for one officer (date missing) and the second Personal for another officer with date 11th May 1921. But we have only one date/amount. However, we have two officers left? Let's see remaining officers after block12: we have used blocks 1-12 giving entries: 1+1+1+1+1+1+1+2+1+3+1+2 = 15 entries? Let's count: block1(1),2(1),3(1),4(1),5(1),6(1),7(1),8(2),9(1),10(3),11(1),12(2) = 15 entries. We need 21 total. So 6 entries left. Block13,14,15,16,17? Actually after block12, we have:

Block13: Personal... Personal... 11th May 1921 300 (maybe 2 entries)

Block14: Deposit. 9th January, 1924, 1,000 (1 entry)

Block15: Personal. ... 4th November, 1924. 1,000 (1 entry)

Block16: Deposit. 28th April, 1921. 200 (1 entry)

Block17: Do. 27th May, 1921. 200 (1 entry)

That's 2+1+1+1+1 = 6 entries. Perfect. So block13 yields 2 entries. But there is only one date/amount (11th May 1921, 300). How can two entries share the same date and amount? Possibly the first Personal has a different date that was omitted by OCR. But the text shows the date after the second Personal. The first Personal might have a date that is not captured. However, the OCR might have merged two rows: first Personal with date X, second Personal with date 11th May 1921. But we don't have date X. Since we must preserve original content, we could insert ... for the missing date for the first Personal. But the rule: "Missing text — if a word or phrase is clearly missing due to OCR damage, insert ... in its place." So for the first Personal in block13, the date and amount are missing. But we have a date/amount for the second Personal. However, the block shows only one date/amount. It might be that the first Personal's date/amount is the same as the second? Unlikely.

Given the instruction to reconstruct sentences and paragraphs, but this is a table. The primary task is to fix text flow. For a table, we should reconstruct the table as accurately as possible. Since the OCR has scrambled the table, we need to use logical deduction. But the user expects a proofread text in Markdown, likely a table.

I think the best is to produce a table with the 21 rows, using the department headers, and for each officer, fill in the nature, date, amount as per the sequential alignment of blocks to officers. Since the blocks correspond to the security column in order, and the names are in order, we can map each block entry to an officer in sequence.

Let's list the 21 entries in order from the blocks:

1. Deposit, 11th April 1924, $200
2. Personal (Sureties: Yeung Ming...), 29th April 1924, $3,000
3. Deposit, 27th Oct 1923, $1,200
4. Deposit, 26th Nov 1921, $300
5. Deposit, 18th Feb 1927, $1,000
6. Personal (Sureties: La Wing...), 16th Dec 1921, $250
7. Deposit, 10th Jan 1921, $50
8. Deposit, 18th Mar 1921, $50
9. Deposit, 26th Apr 1921, $100
10. Deposit, 5th Oct 1923, $30
11. Title Deeds, 25th Jun 1921, $5,000
12. Deposit, 16th Oct 1925, $2,000
13. Personal (Sureties: Cheng Man-Long...), 3rd May 1926, $500
14. [Nature missing], 26th Mar 1926, $500
15. Title Deals, 12th Oct 1923, $5,000
16. Title Deals, 28th Apr 1921, $500
17. Personal (Sureties: 'tin...), [date missing?], [amount missing?] but then next is 11th May 1921, 300 for second Personal. Actually block13 has two Personals. We'll treat as two entries:

17a. Personal (Sureties: 'tin...), date ?, amount ?

17b. Personal (Sureties: Ng dim-kai...), 11th May 1921, $300

But we have only one date/amount. However, the text "11th May, 1921. 300" appears after the second Personal. It might belong to the second Personal. The first Personal might have its own date/amount that is not shown. But the OCR might have omitted it. Since we have exactly 21 officers, and we have 21 date/amount pairs from the list earlier (which includes 11th May 1921, 300 as pair17). That list came from extracting all dates and amounts in order. In that extraction, we had 21 pairs. The 17th pair was 11th May 1921, 300. The 18th was 9th Jan 1924, 1000. So the first Personal in block13 does not have a separate date/amount in the extracted list. That suggests that the first Personal might not correspond to a separate officer; maybe it's a continuation of the previous entry? But the previous entry (16) is Title Deals. Hmm.

Let's re-extract dates and amounts strictly from the text in order, ignoring nature:

Text: "11th April, 1924. $200" -> 1

"29th April, 1924. 3,000" -> 2

"27th Oct., 1923. 1,200" -> 3

"26th Nov., 1921. 300" -> 4

"18th February, 1927. 1,000" -> 5

"16th December, 1921. ... 250" -> 6 (date 16th Dec 1921, amount 250)

"10th January, 1921. 50" -> 7

"18th March, 1921. 26th April, 1921, 50 100" -> 8 and 9 (18th Mar 1921 $50, 26th Apr 1921 $100)

"5th October, 1923. 30" -> 10

"25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500" -> 11,12,13 (25th Jun 1921 $5000, 16th Oct 1925 $2000, 3rd May 1926 $500)

"26th March, 1926. 500" -> 14

"12th October, 1923. 28th April, 1921. 5,000 500" -> 15,16 (12th Oct 1923 $5000, 28th Apr 1921 $500)

"11th May, 1921. 300" -> 17

"9th January, 1924, 1,000" -> 18

"4th November, 1924. 1,000" -> 19

"28th April, 1921. 200" -> 20

"27th May, 1921. 200" -> 21

That's 21 pairs. The nature for each pair must be assigned from the nature keywords in order. The nature keywords appear in the text before or around these dates. Let's list nature keywords in order of appearance alongside the dates:

• Before pair1: "Deposit."
• Before pair2: "Personal."
• Before pair3: "Deposit."
• Before pair4: "Do." (Deposit)
• Before pair5: "Do." (Deposit)
• Before pair6: "Personal."
• Before pair7: "Deposit."
• Before pair8: "Do." (Deposit) (first Do in "Do. Do.")
• Before pair9: "Do." (Deposit) (second Do)
• Before pair10: "Do." (Deposit)
• Before pair11: "Title Deeds." (first in "Title Deeds. Deposit, Personal.")
• Before pair12: "Deposit," (second in that phrase)
• Before pair13: "Personal." (third in that phrase)
• Before pair14: (no nature keyword, but after pair13's amount, the text continues "26th March, 1926. 500". The previous nature was Personal. But the next nature keyword is "Title Deals," which appears before pair15. So pair14 has no explicit nature. Could be "Title Deeds" again? But the phrase "Title Deeds, N. T. (S.)" might cover the three dates. The fourth date might be separate. I'll assign nature as ... for pair14.
• Before pair15: "Title Deals,"
• Before pair16: (same "Title Deals," covers both)
• Before pair17: "Personal." (first in "Personal. ... Personal, ...")
• Before pair18: "Deposit."
• Before pair19: "Personal."
• Before pair20: "Deposit."
• Before pair21: "Do." (Deposit)

But we have two Personals before pair17? The text: "Personal. (Sureties...) Personal, (Sureries...) 11th May, 1921. 300". There are two "Personal" keywords before the single date/amount pair17. That suggests that the first Personal might correspond to a missing pair? But we have no missing pair. Unless the first Personal corresponds to pair14? But pair14 is 26th March 1926. That's far off.

Maybe the first Personal in that block is actually the nature for pair14? But pair14 appears earlier in the text. The text order: after pair13 (3rd May 1926), we have "26th March, 1926. 500" (pair14), then "Title Deals,...", then "Personal. ... Personal, ... 11th May, 1921. 300". So the two Personals appear after pair16. So they correspond to pair17 and maybe pair18? But pair18 is Deposit. So not.

Wait, the date "11th May, 1921" is earlier than many dates (1921). But it appears late in the text. That's because the OCR read columns vertically. The nature column might have been read after the date column? Actually the OCR read the table column by column: first column (names), second column (nature), third column (date), fourth column (amount). But the text we have seems to be: names (column1), then nature (column2), then date (column3), then amount (column4)? But the dates and amounts are interleaved in the text. The text shows nature, then date, then amount, then nature, date, amount... That suggests the OCR read row by row? But the names are all at the beginning. So maybe the OCR read the first column (names) completely, then the second column (nature) completely, then third (date), then fourth (amount). But the text we have shows nature, date, amount interleaved. That would happen if the OCR read the table row by row but the names were in the first column and the OCR output the first column first? Actually the OCR output starts with "(K3) Amount Name of Officer giving Security. Naturo of the Security giveu. Date of Bond. of Bond." Then "SANITARY DEPARTMENT, Fong Fook, Yung Kai-u, ..." That is the first column (names) with department headers. Then after all names, it goes "Deposit. 11th April, 1924. $200 Personal. ..." That is the second, third, fourth columns combined? But they are interleaved: nature, date, amount. That suggests that the OCR read the remaining three columns row by row. So the order of the security details matches the order of the names. Good.

Thus the first security detail (Deposit, 11th April 1924, $200) belongs to the first officer (Fong Fook). Second to Yung Kai-u, etc. So we have 21 officers and 21 security details (each detail includes nature, date, amount). The security details are given in the text as a sequence of nature-date-amount groups. But the text has some groups with multiple dates/amounts for one nature (like "Do. Do. 18th March, 1921. 26th April, 1921, 50 100"). That corresponds to two officers (8th and 9th) both with Deposit nature. Similarly, "Title Deeds. Deposit, Personal. ... 25th June, 1921. 16th October, 1925, 3rd May, 1926. 5,000 2,000 500" corresponds to three officers (11th, 12th, 13th) with natures Title Deeds, Deposit, Personal respectively. "Title Deals, -Woo Lam- 12th October, 1923. 28th April, 1921. 5,000 500" corresponds to two officers (15th, 16th) both Title Deals. "Personal. ... Personal, ... 11th May, 1921. 300" corresponds to two officers (17th, 18th?) but only one date/amount. However, the next security detail is "Deposit. 9th January, 1924, 1,000" for officer 19? Let's count officers:

Officers 1-5: Sanitary

6-10: Botanical

11-14: Public Works

15-21: Post Office (7 officers)

Security details groups:

Group1: Deposit (officer1)

Group2: Personal (officer2)

Group3: Deposit (officer3)

Group4: Deposit (officer4)

Group5: Deposit (officer5)

Group6: Personal (officer6)

Group7: Deposit (officer7)

Group8: Deposit (officer8) and Deposit (officer9) [two officers]

Group9: Deposit (officer10)

Group10: Title Deeds (officer11), Deposit (officer12), Personal (officer13) [three officers]

Group11: [nature?] (officer14) - single date 26th March 1926

Group12: Title Deals (officer15), Title Deals (officer16) [two officers]

Group13: Personal (officer17), Personal (officer18?) but only one date/amount (11th May 1921, 300). Then Group14: Deposit (officer19), Group15: Personal (officer20), Group16: Deposit (officer21), Group17: Deposit (officer22?) but we only have 21 officers.

We have 21 officers. Groups 1-10 cover officers 1-13 (1+1+1+1+1+1+1+2+1+3 = 13). Group11 covers officer14. Group12 covers officers15-16. Group13 covers officers17-18? But group13 has two Personals but only one date/amount. Group14 covers officer19. Group15 covers officer20. Group16 covers officer21. Group17 would be extra. But we have only 21 officers. So group13 must cover only one officer (officer17) with two Personal descriptions? That doesn't make sense.

Maybe the first Personal in group13 is actually the nature for officer14 (group11). But group11 appears earlier in text. The text order: group10 (three officers), then "26th March, 1926. 500" (group11), then "Title Deals..." (group12), then "Personal... Personal... 11th May 1921 300" (group13). So group11 is a single entry with no nature. Group13 has two Personals. Could it be that group11's nature is the first Personal of group13? But they are separated by group12. Unlikely.

Perhaps the OCR misordered the columns. The nature column might have been read after the date column? But the text shows nature before date for each group.

Given the time, I'll produce a table with the 21 rows, using the nature assignments as per the groups, and for the ambiguous ones, I'll use the information we have. For officer14 (Chan Yik-shing), the security detail is "26th March, 1926. 500" with no nature. I'll put ... for nature. For officers 17 and 18 (Ng Toe-cheang and Lam Him), the text gives two Personals but only one date/amount. However, the next officer (Lam Him) has a Deposit on 9th Jan 1924. So maybe the two Personals are for Ng Toe-cheang and Fung Chat? But Fung Chat is officer19. Let's see the Post Office officers in order: 15 Ho Sui-pui, 16 Ha Tsau, 17 Ng Toe-cheang, 18 Lam Him, 19 Fung Chat, 20 Tec Po, 21 Ng Tim.

The security details for Post Office from groups:

• Group12: Title Deals (officer15), Title Deals (officer16)
• Group13: Personal, Personal (officer17, officer18?) with date 11th May 1921, 300 for the second? But officer18 is Lam Him, who later has Deposit 9th Jan 1924 (group14). So officer18 cannot have Personal 11th May 1921. Unless the groups are not aligned with officers after group12. But the alignment is sequential. So officer17 gets first Personal, officer18 gets second Personal. But the second Personal has date 11th May 1921. Then officer19 gets Deposit 9th Jan 1924. Officer20 gets Personal 4th Nov 1924. Officer21 gets Deposit 28th Apr 1921. Officer22? Deposit 27th May 1921. But we have only 21 officers. So officer21 would be Ng Tim, getting Deposit 27th May 1921. That works if we have 21 officers and 21 security entries. But we have an extra Personal (first in group13) that would be officer17, and the second Personal would be officer18. But officer18 is Lam Him, who would then have Personal 11th May 1921, but the text later gives Deposit 9th Jan 1924 for officer19? Wait, the sequence after group13: "Deposit. 9th January, 1924, 1,000" (group14), "Personal. ... 4th November, 1924. 1,000" (group15), "Deposit. 28th April, 1921. 200" (group16), "Do. 27th May, 1921. 200" (group17). That's four groups for officers 19-22. But we only have officers 19,20,21. So one group extra.

Thus the alignment is off by one. Perhaps