1927 — Page 344

The user wants me to proofread OCR output of a historical document about widows and orphans' pensions in Hong Kong. The text appears to be a table with columns: Name of Pensioner, No. of Pension Register, Amount of Pension in Sterling per annum, Amount of Pension in Dollars per annum, Date from which Pension has been paid, Particulars, Present Age of Pensioner.

I need to reconstruct this as a Markdown table. The OCR has some issues: broken lines, misaligned columns, some garbled text. I'll need to parse the data carefully.

Let me analyze the text line by line:

First lines are headers:

"Name of l'ensioner." -> "Name of Pensioner."

"No. of Pension Register."

"Amount of the Pension in Sterling, per annun." -> "per annum."

"Amount of the Pension in Dollars, per anum." -> "per annum."

"Date from which the Pension has been paid,"

"Particulars."

"Present Age of Pensioner."

Then a row: "342" maybe a page number? Or a register number? Actually "342" appears alone, then "£ $. d." and "C." maybe column headers for currency? Actually the table likely has columns: Register No., Name, Sterling (£ s. d.), Dollars (C.), Date, Particulars, Age.

But the OCR shows:

"342

£ $. d.

C.

Brought forward,......

3,330 15 4

30,036.67"

This seems like a summary row: "Brought forward" with totals: £3,330 15s 4d? and $30,036.67.

Then individual entries:

"Lui Ying,

1,834

49.73

Matilda M. Angus,

3

70 15 10

Lau Yat Kin,.

1,474

171.82

Ebrahim Khan,

1,408

207.10

27th June, 1925.

6th July, 1925.

22nd July, 1925,

Do.

Widow of Chung Ju, late of Imports and Exports Department,..................

Widow of Peter Augus, late of Police Department,

42

54

Orphan of Lan Ka Lok, late of Police Department,

Orphan of Nawab Khan, late of Post Office,

4

Yim Sz,

1,598

251.21

22nd August, 1925.

Widow of Fan Tsung Hing, late of Harbour Department,

16

51

Cheung Sau In,

499

99.15

12th November, 1925. | Widow of Kwok Yuk Tin, late of District Office South,

63

(Kwok Cheung Shi)..........

Ho Chiu Shi,

179

58,79

3rd December, 1925.

Widow of Ho Fung Cheung, late of Education Department,.

544

68

Catherine Johnston An-

derson Kynoch,

1,012

91 3 5

---

Theodora Maria Wilkinson,

487

268.47

6th February, 1926.

22nd February, 1926.

Widow of James Henry Kynoch, late of Public Works Department, ..

Widow of William Josiah Wilkinson, late of Prison Department,...

41

52

Tang Yue,

1,202

281.80

1st March, 1926.

Widow of Fung Kam Tin, late of Kowloon Canton Railway,

3333

(Fung Tang Shi), ......

Chan So.................................

1,979

Lilian Howard Wheal,

488

85.01

372.29

Mary Hall Muchart Reid,

1,735

64 0 4

10th March, 1926.

29th March, 1926.

7th May, 1926,

Widow of Kwong In, late of Sanitary Department,..........

40

Widow of Isaac Alfred Wheal, inte of Public Works Department,...

Widow of Alexander Reid, late of Police Department,

50

32

Wong Tso Pik,

1,397

277.44

7th July, 1926.

Widow of Chan Shiu To, late of Education Departmeut,

28

( L 38)"

This is messy. I need to reconstruct rows. Each pensioner entry likely has: Register No., Name, Sterling amount (£ s. d.), Dollar amount, Date, Particulars, Age.

But the OCR doesn't align well. Let's try to parse sequentially.

After "Brought forward" totals, we have:

1. Lui Ying, Register 1,834? Actually "1,834" might be register number. Then "49.73" dollar amount. No sterling amount shown? Maybe sterling column is blank or zero? Or maybe the sterling column is in £ s. d. format but not shown for some.

1. Matilda M. Angus, Register 3? "3" then "70 15 10" likely £70 15s 10d. Then dollar amount? Not shown? Actually next line "Lau Yat Kin, 1,474 171.82" so maybe Matilda's dollar amount missing? Or maybe the columns are: Register No., Name, Sterling, Dollars, Date, Particulars, Age.

But the data seems interleaved: names, register numbers, amounts, dates, particulars, ages.

Let's list all lines in order:

• 342 (maybe page number)
• £ $. d.
• C.
• Brought forward,......
• 3,330 15 4
• 30,036.67
• Lui Ying,
• 1,834
• 49.73
• Matilda M. Angus,
• 3
• 70 15 10
• Lau Yat Kin,.
• 1,474
• 171.82
• Ebrahim Khan,
• 1,408
• 207.10
• 27th June, 1925.
• 6th July, 1925.
• 22nd July, 1925,
• Do.
• Widow of Chung Ju, late of Imports and Exports Department,..................
• Widow of Peter Augus, late of Police Department,
• 42
• 54
• Orphan of Lan Ka Lok, late of Police Department,
• Orphan of Nawab Khan, late of Post Office,
• 4
• Yim Sz,
• 1,598
• 251.21
• 22nd August, 1925.
• Widow of Fan Tsung Hing, late of Harbour Department,
• 16
• 51
• Cheung Sau In,
• 499
• 99.15
• 12th November, 1925. | Widow of Kwok Yuk Tin, late of District Office South,
• 63
• (Kwok Cheung Shi)..........
• Ho Chiu Shi,
• 179
• 58,79
• 3rd December, 1925.
• Widow of Ho Fung Cheung, late of Education Department,.
• 544
• 68
• Catherine Johnston An-
• derson Kynoch,
• 1,012
• 91 3 5
• ---
• Theodora Maria Wilkinson,
• 487
• 268.47
• 6th February, 1926.
• 22nd February, 1926.
• Widow of James Henry Kynoch, late of Public Works Department, ..
• Widow of William Josiah Wilkinson, late of Prison Department,...
• 41
• 52
• Tang Yue,
• 1,202
• 281.80
• 1st March, 1926.
• Widow of Fung Kam Tin, late of Kowloon Canton Railway,
• 3333
• (Fung Tang Shi), ......
• Chan So.................................
• 1,979
• Lilian Howard Wheal,
• 488
• 85.01
• 372.29
• Mary Hall Muchart Reid,
• 1,735
• 64 0 4
• 10th March, 1926.
• 29th March, 1926.
• 7th May, 1926,
• Widow of Kwong In, late of Sanitary Department,..........
• 40
• Widow of Isaac Alfred Wheal, inte of Public Works Department,...
• Widow of Alexander Reid, late of Police Department,
• 50
• 32
• Wong Tso Pik,
• 1,397
• 277.44
• 7th July, 1926.
• Widow of Chan Shiu To, late of Education Departmeut,
• 28
• ( L 38)

We need to group into rows. The pattern: each pensioner has a name, register number, sterling amount (maybe), dollar amount, date, particulars, age.

But the OCR seems to have lost column alignment. Let's try to infer from known data: The header says "No. of Pension Register." So each pensioner has a register number. Then "Amount of the Pension in Sterling, per annum." Then "Amount of the Pension in Dollars, per annum." Then "Date from which the Pension has been paid," Then "Particulars." Then "Present Age of Pensioner."

Thus each row should have 7 columns.

Now, the "Brought forward" row likely has register number blank, name "Brought forward", sterling total, dollar total, date blank, particulars blank, age blank.

Then each subsequent pensioner.

Let's try to parse sequentially, assuming the OCR read row by row but lines broken.

I'll write a script mentally? Better to manually group.

Look at the first few entries after Brought forward:

"Lui Ying,

1,834

49.73"

That's three lines. Could be: Name: Lui Ying, Register: 1,834, Dollar: 49.73. Sterling missing? Maybe sterling is 0 or not applicable. But the header includes sterling. Perhaps some pensions are only in dollars? Or maybe the sterling column is for those paid in sterling, dollars for those paid in dollars. But the table has both columns.

Next: "Matilda M. Angus,

3

70 15 10"

Name: Matilda M. Angus, Register: 3, Sterling: 70 15 10 (i.e., £70 15s 10d). Dollar amount missing? Next line "Lau Yat Kin,.

1,474

171.82" So maybe Matilda's dollar amount is not given, or maybe the dollar amount is on the same line but OCR split.

Next: "Ebrahim Khan,

1,408

207.10"

Then dates: "27th June, 1925.

6th July, 1925.

22nd July, 1925,

Do."

These likely correspond to the four pensioners above: Lui Ying, Matilda M. Angus, Lau Yat Kin, Ebrahim Khan. So four dates.

Then "Widow of Chung Ju, late of Imports and Exports Department,..................

Widow of Peter Augus, late of Police Department,

42

54"

These are particulars and ages for the first two? Actually "Widow of Chung Ju..." likely for Lui Ying? But Lui Ying is a Chinese name, could be widow of Chung Ju. "Widow of Peter Augus" for Matilda M. Angus? But Matilda M. Angus sounds like a widow of Peter Angus. Then ages 42 and 54.

Then "Orphan of Lan Ka Lok, late of Police Department,

Orphan of Nawab Khan, late of Post Office,

4"

These for Lau Yat Kin and Ebrahim Khan? Lau Yat Kin is orphan of Lan Ka Lok? Age 4? And Ebrahim Khan orphan of Nawab Khan, age? Not given? Actually "4" might be age for first orphan, second orphan age missing.

Then "Yim Sz,

1,598

251.21

22nd August, 1925.

Widow of Fan Tsung Hing, late of Harbour Department,

16

51"

This seems like a new pensioner: Yim Sz, register 1,598, dollar 251.21, date 22nd Aug 1925, particulars: Widow of Fan Tsung Hing, age 16? Wait "16" and "51" two numbers. Maybe age 16? But 51 also appears. Could be two pensioners? Actually "16" and "51" might be ages for two entries? But only one name Yim Sz. Let's see: after "Widow of Fan Tsung Hing..." there is "16" then "51". Then next "Cheung Sau In, 499 99.15 12th November, 1925. | Widow of Kwok Yuk Tin, late of District Office South, 63 (Kwok Cheung Shi).........."

So maybe "16" is age for Yim Sz? But 16 seems young for a widow. Could be "16" is something else. Or maybe the data is misaligned.

Let's step back. Perhaps the OCR has merged two columns: the "Particulars" column includes the relationship and department, and the "Present Age" column is separate. The numbers 42, 54, 4, 16, 51, 63, etc. are ages.

But we have many numbers. Let's list all numbers that appear after particulars:

After "Widow of Peter Augus, late of Police Department," we have "42" and "54" on separate lines. So two ages: 42 and 54.

After "Orphan of Nawab Khan, late of Post Office," we have "4". Only one age? Maybe the other orphan's age is missing.

After "Widow of Fan Tsung Hing, late of Harbour Department," we have "16" and "51". Two ages? But only one pensioner Yim Sz. Unless there are two pensioners: Yim Sz and another? But the text shows "Yim Sz, 1,598 251.21 22nd August, 1925. Widow of Fan Tsung Hing, late of Harbour Department, 16 51". Could be that "16" is register number? No, register is 1,598. "16" might be age of Yim Sz? But 16 is too young. Maybe "16" is the age of the orphan? But Yim Sz is a widow. Hmm.

Then "Cheung Sau In, 499 99.15 12th November, 1925. | Widow of Kwok Yuk Tin, late of District Office South, 63 (Kwok Cheung Shi).........." So Cheung Sau In, register 499, dollar 99.15, date 12 Nov 1925, particulars: Widow of Kwok Yuk Tin, age 63.

Then "Ho Chiu Shi, 179 58,79 3rd December, 1925. Widow of Ho Fung Cheung, late of Education Department,. 544 68"

Ho Chiu Shi, register 179, dollar 58.79? (58,79 with comma), date 3 Dec 1925, particulars: Widow of Ho Fung Cheung, then "544 68" - maybe two numbers: 544 and 68. Could be sterling amount? 544? But sterling format is £ s. d. 544 could be pounds? 68 could be age? But age 68 plausible.

Then "Catherine Johnston An- derson Kynoch, 1,012 91 3 5 --- Theodora Maria Wilkinson, 487 268.47 6th February, 1926. 22nd February, 1926. Widow of James Henry Kynoch, late of Public Works Department, .. Widow of William Josiah Wilkinson, late of Prison Department,... 41 52"

Here we have two pensioners: Catherine Johnston Anderson Kynoch (register 1,012, sterling 91 3 5? i.e., £91 3s 5d), and Theodora Maria Wilkinson (register 487, dollar 268.47). Two dates: 6 Feb 1926 and 22 Feb 1926. Two particulars: Widow of James Henry Kynoch, Widow of William Josiah Wilkinson. Two ages: 41 and 52.

Then "Tang Yue, 1,202 281.80 1st March, 1926. Widow of Fung Kam Tin, late of Kowloon Canton Railway, 3333 (Fung Tang Shi), ...... Chan So................................. 1,979 Lilian Howard Wheal, 488 85.01 372.29 Mary Hall Muchart Reid, 1,735 64 0 4 10th March, 1926. 29th March, 1926. 7th May, 1926, Widow of Kwong In, late of Sanitary Department,.......... 40 Widow of Isaac Alfred Wheal, inte of Public Works Department,... Widow of Alexander Reid, late of Police Department, 50 32 Wong Tso Pik, 1,397 277.44 7th July, 1926. Widow of Chan Shiu To, late of Education Departmeut, 28 ( L 38)"

This is a block of several pensioners.

Let's try to parse systematically. I'll create a table with columns: Register No., Name, Sterling (£ s. d.), Dollars, Date, Particulars, Age.

We need to decide which currency column applies. Some entries have sterling amounts (like 70 15 10, 91 3 5, 64 0 4), others have dollar amounts (49.73, 171.82, 207.10, 251.21, 99.15, 58.79, 268.47, 281.80, 85.01, 372.29, 277.44). Some have both? For example, Matilda M. Angus has sterling 70 15 10 but no dollar shown. Lui Ying has dollar 49.73 but no sterling. Lau Yat Kin has dollar 171.82. Ebrahim Khan dollar 207.10. Yim Sz dollar 251.21. Cheung Sau In dollar 99.15. Ho Chiu Shi dollar 58.79. Catherine Kynoch sterling 91 3 5. Theodora Wilkinson dollar 268.47. Tang Yue dollar 281.80. Lilian Wheal dollar 85.01 and also 372.29? Wait "Lilian Howard Wheal, 488 85.01 372.29" two dollar amounts? Or maybe 85.01 is sterling? But sterling format is £ s. d. 85.01 could be dollars. 372.29 also dollars. Mary Hall Muchart Reid: 1,735 64 0 4 (sterling £64 0s 4d). Then three dates: 10th March, 29th March, 7th May. Three particulars: Widow of Kwong In, Widow of Isaac Alfred Wheal, Widow of Alexander Reid. Three ages: 40, 50, 32. But only two names? Actually Lilian Howard Wheal and Mary Hall Muchart Reid are two names, but three entries? Maybe there are three pensioners: Lilian Howard Wheal (widow of Isaac Alfred Wheal), Mary Hall Muchart Reid (widow of Alexander Reid), and another widow of Kwong In (name not given? Or maybe "Widow of Kwong In" is a third pensioner with name missing? The OCR shows "Widow of Kwong In, late of Sanitary Department,.......... 40" then "Widow of Isaac Alfred Wheal, inte of Public Works Department,... 50" then "Widow of Alexander Reid, late of Police Department, 32". So three widows, but only two names given (Lilian Howard Wheal and Mary Hall Muchart Reid). The first widow of Kwong In might be someone else, maybe name missing? Or maybe the name is "Wong Tso Pik"? But Wong Tso Pik appears later with register 1,397, dollar 277.44, date 7th July 1926, particulars: Widow of Chan Shiu To, age 28.

So the block from "Tang Yue" to "Wong Tso Pik" includes multiple entries.

Let's list all distinct pensioner names from the text:

1. Lui Ying
2. Matilda M. Angus
3. Lau Yat Kin
4. Ebrahim Khan
5. Yim Sz
6. Cheung Sau In
7. Ho Chiu Shi
8. Catherine Johnston Anderson Kynoch
9. Theodora Maria Wilkinson
10. Tang Yue
11. Chan So? (appears as "Chan So................................. 1,979" maybe a name)
12. Lilian Howard Wheal
13. Mary Hall Muchart Reid
14. Wong Tso Pik

Also "Widow of Kwong In" unnamed? Possibly the name is missing.

And "Widow of Fan Tsung Hing" is Yim Sz? Actually Yim Sz is the widow of Fan Tsung Hing.

"Widow of Kwok Yuk Tin" is Cheung Sau In.

"Widow of Ho Fung Cheung" is Ho Chiu Shi.

"Widow of James Henry Kynoch" is Catherine Johnston Anderson Kynoch.

"Widow of William Josiah Wilkinson" is Theodora Maria Wilkinson.

"Widow of Fung Kam Tin" is Tang Yue (Fung Tang Shi).

"Widow of Kwong In" unnamed.

"Widow of Isaac Alfred Wheal" is Lilian Howard Wheal.

"Widow of Alexander Reid" is Mary Hall Muchart Reid.

"Widow of Chan Shiu To" is Wong Tso Pik.

Also orphans: Orphan of Lan Ka Lok (Lau Yat Kin), Orphan of Nawab Khan (Ebrahim Khan).

Now, register numbers:

• Lui Ying: 1,834
• Matilda M. Angus: 3
• Lau Yat Kin: 1,474
• Ebrahim Khan: 1,408
• Yim Sz: 1,598
• Cheung Sau In: 499
• Ho Chiu Shi: 179
• Catherine Kynoch: 1,012
• Theodora Wilkinson: 487
• Tang Yue: 1,202
• Chan So: 1,979
• Lilian Wheal: 488
• Mary Reid: 1,735
• Wong Tso Pik: 1,397

Also "3333" appears after "Widow of Fung Kam Tin, late of Kowloon Canton Railway," maybe a register number for Tang Yue? But Tang Yue already has 1,202. 3333 might be something else.

Now sterling amounts:

• Matilda M. Angus: 70 15 10
• Catherine Kynoch: 91 3 5
• Mary Reid: 64 0 4
• Possibly others have sterling but not shown.

Dollar amounts:

• Lui Ying: 49.73
• Lau Yat Kin: 171.82
• Ebrahim Khan: 207.10
• Yim Sz: 251.21
• Cheung Sau In: 99.15
• Ho Chiu Shi: 58.79
• Theodora Wilkinson: 268.47
• Tang Yue: 281.80
• Lilian Wheal: 85.01 and 372.29? Two amounts? Maybe 85.01 is sterling? But 85.01 not in £ s. d. format. Could be 85.01 dollars and 372.29 dollars for two different pensions? But Lilian Wheal is one person. Maybe the 372.29 belongs to Mary Reid? But Mary Reid has sterling 64 0 4. Or maybe the line "Lilian Howard Wheal, 488 85.01 372.29" means register 488, dollar 85.01, and then next pensioner Mary Hall Muchart Reid has register 1,735 and sterling 64 0 4? But 372.29 is left hanging. Could be that Lilian Wheal has two pensions? Or maybe 372.29 is the dollar equivalent of Mary Reid's sterling? But unlikely.

Let's examine the original OCR lines around that:

"Chan So.................................

1,979

Lilian Howard Wheal,

488

85.01

372.29

Mary Hall Muchart Reid,

1,735

64 0 4

10th March, 1926.

29th March, 1926.

7th May, 1926,

Widow of Kwong In, late of Sanitary Department,..........

40

Widow of Isaac Alfred Wheal, inte of Public Works Department,...

Widow of Alexander Reid, late of Police Department,

50

32"

So after Chan So (register 1,979), we have Lilian Howard Wheal (register 488), then two numbers: 85.01 and 372.29. Then Mary Hall Muchart Reid (register 1,735), then 64 0 4 (sterling). Then three dates. Then three particulars. Then three ages: 40, 50, 32.

Thus there are three pensioners in this group:

• Chan So (register 1,979) - but no name? Actually "Chan So" might be a name. Particulars? Not given. Maybe Chan So is the widow of Kwong In? But the particulars list "Widow of Kwong In" first. So Chan So could be the widow of Kwong In. Age 40.
• Lilian Howard Wheal (register 488) - widow of Isaac Alfred Wheal. Age 50.
• Mary Hall Muchart Reid (register 1,735) - widow of Alexander Reid. Age 32.

But then what are the amounts? For Chan So: maybe dollar amount 85.01? But 85.01 appears after Lilian's register. For Lilian: maybe dollar 372.29? For Mary: sterling 64 0 4.

But the dates: 10th March, 29th March, 7th May. Which corresponds to which? Possibly:

• Chan So: 10th March 1926
• Lilian Wheal: 29th March 1926
• Mary Reid: 7th May 1926

But the particulars order: Widow of Kwong In (Sanitary Dept), Widow of Isaac Alfred Wheal (Public Works), Widow of Alexander Reid (Police). That matches the three names.

Thus we have three pensioners with registers: Chan So (1,979), Lilian Howard Wheal (488), Mary Hall Muchart Reid (1,735). Amounts: Chan So: $85.01? Lilian: $372.29? Mary: £64 0s 4d.

But the OCR shows "Lilian Howard Wheal, 488 85.01 372.29". That could be read as: Name: Lilian Howard Wheal, Register: 488, Dollar: 85.01, and then the next number 372.29 might be for Chan So? But Chan So appears before Lilian. The line "Chan So................................. 1,979" then "Lilian Howard Wheal, 488 85.01 372.29". So maybe Chan So's register is 1,979, and her dollar amount is 85.01? But 85.01 is after Lilian's register. Could be that the OCR misordered: actually the table might have columns: Name, Register, Sterling, Dollars. For Chan So, maybe sterling blank, dollars 85.01. For Lilian, register 488, dollars 372.29. For Mary, register 1,735, sterling 64 0 4.

But the text "Lilian Howard Wheal, 488 85.01 372.29" suggests two numbers after 488. Could be that 85.01 is sterling? But sterling format is £ s. d. 85.01 is not that. Could be 85.01 dollars and 372.29 dollars for two different pensions? But Lilian is one person.

Maybe the table has two dollar columns? Unlikely.

Let's look at the header again: "Amount of the Pension in Sterling, per annum. Amount of the Pension in Dollars, per annum." So two amount columns. Some pensioners may have both sterling and dollar amounts? But usually a pension is paid in one currency. However, some might be paid in sterling but converted to dollars for record. The "Brought forward" shows both sterling and dollar totals.

Thus each row might have both sterling and dollar amounts. For example, Matilda M. Angus: sterling 70 15 10, dollar maybe not shown. Lui Ying: dollar 49.73, sterling maybe 0. But the OCR might have omitted zero values.

Given the complexity, I think the best approach is to reconstruct the table as faithfully as possible from the OCR text, preserving the order and grouping as they appear, and note uncertainties with ... if needed.

But the instruction: "Reconstruct the table using Markdown table syntax." So I need to produce a Markdown table with the columns as per header.

I'll create a table with columns: Register No., Name, Sterling (£ s. d.), Dollars, Date, Particulars, Age.

I'll parse the text sequentially, grouping each pensioner's data.

Let's go through the text again, but this time with an eye for pattern: The OCR likely read the table row by row, but each row's cells may be split across lines. The original table probably had 7 columns. The OCR output seems to list cell contents in reading order (left to right, top to bottom). So the sequence of cells is: for each row, the 7 cells in order.

But the header row: "Name of Pensioner.", "No. of Pension Register.", "Amount of the Pension in Sterling, per annum.", "Amount of the Pension in Dollars, per annum.", "Date from which the Pension has been paid,", "Particulars.", "Present Age of Pensioner."

Then a row: "Brought forward" maybe in Name column, then register no. blank, then sterling total, dollar total, date blank, particulars blank, age blank.

Then subsequent rows.

If we assume the OCR reads row by row, then the sequence after header should be: Row1 cell1, cell2, cell3, cell4, cell5, cell6, cell7; Row2 cell1, cell2, etc.

But the OCR text we have is not clearly separated by rows. However, we can try to segment by noticing that each pensioner has a name, then a register number, then amounts, then date, then particulars, then age.

But the OCR might have merged multiple rows because of line breaks.

Let's attempt to simulate the table reading order.

I'll write down the tokens in order as they appear:

1. "342" (maybe page number, not part of table)
2. "£ $. d." (sub-header for sterling)
3. "C." (sub-header for dollars)
4. "Brought forward,......"
5. "3,330 15 4"
6. "30,036.67"
7. "Lui Ying,"
8. "1,834"
9. "49.73"
10. "Matilda M. Angus,"
11. "3"
12. "70 15 10"
13. "Lau Yat Kin,."
14. "1,474"
15. "171.82"
16. "Ebrahim Khan,"
17. "1,408"
18. "207.10"
19. "27th June, 1925."
20. "6th July, 1925."
21. "22nd July, 1925,"
22. "Do."
23. "Widow of Chung Ju, late of Imports and Exports Department,.................."
24. "Widow of Peter Augus, late of Police Department,"
25. "42"
26. "54"
27. "Orphan of Lan Ka Lok, late of Police Department,"
28. "Orphan of Nawab Khan, late of Post Office,"
29. "4"
30. "Yim Sz,"
31. "1,598"
32. "251.21"
33. "22nd August, 1925."
34. "Widow of Fan Tsung Hing, late of Harbour Department,"
35. "16"
36. "51"
37. "Cheung Sau In,"
38. "499"
39. "99.15"
40. "12th November, 1925. | Widow of Kwok Yuk Tin, late of District Office South,"
41. "63"
42. "(Kwok Cheung Shi).........."
43. "Ho Chiu Shi,"
44. "179"
45. "58,79"
46. "3rd December, 1925."
47. "Widow of Ho Fung Cheung, late of Education Department,."
48. "544"
49. "68"
50. "Catherine Johnston An-"
51. "derson Kynoch,"
52. "1,012"
53. "91 3 5"
54. "---"
55. "Theodora Maria Wilkinson,"
56. "487"
57. "268.47"
58. "6th February, 1926."
59. "22nd February, 1926."
60. "Widow of James Henry Kynoch, late of Public Works Department, .."
61. "Widow of William Josiah Wilkinson, late of Prison Department,..."
62. "41"
63. "52"
64. "Tang Yue,"
65. "1,202"
66. "281.80"
67. "1st March, 1926."
68. "Widow of Fung Kam Tin, late of Kowloon Canton Railway,"
69. "3333"
70. "(Fung Tang Shi), ......"
71. "Chan So................................."
72. "1,979"
73. "Lilian Howard Wheal,"
74. "488"
75. "85.01"
76. "372.29"
77. "Mary Hall Muchart Reid,"
78. "1,735"
79. "64 0 4"
80. "10th March, 1926."
81. "29th March, 1926."
82. "7th May, 1926,"
83. "Widow of Kwong In, late of Sanitary Department,.........."
84. "40"
85. "Widow of Isaac Alfred Wheal, inte of Public Works Department,..."
86. "Widow of Alexander Reid, late of Police Department,"
87. "50"
88. "32"
89. "Wong Tso Pik,"
90. "1,397"
91. "277.44"
92. "7th July, 1926."
93. "Widow of Chan Shiu To, late of Education Departmeut,"
94. "28"
95. "( L 38)"

Now, if the table has 7 columns per row, then after header, the first row (Brought forward) would take 7 tokens: tokens 4-10? But token 4 is "Brought forward,......", token5 "3,330 15 4", token6 "30,036.67", token7 "Lui Ying," - that would be name of next row. So Brought forward row: Name: Brought forward, Register: (blank?), Sterling: 3,330 15 4, Dollars: 30,036.67, Date: (blank), Particulars: (blank), Age: (blank). But then the next token "Lui Ying," would be the Name of next row. That seems plausible.

But then the next row (Lui Ying) would need 7 tokens: Name: Lui Ying, Register: 1,834, Sterling: ?, Dollars: 49.73, Date: ?, Particulars: ?, Age: ?. But we have tokens: 7 Lui Ying, 8 1,834, 9 49.73, 10 Matilda M. Angus, ... So after 49.73, the next token is Matilda M. Angus which is a name. So Lui Ying row would have only 3 tokens before next name. That suggests that the sterling column for Lui Ying is empty (or zero), and the date, particulars, age are not yet listed; they appear later in the list (tokens 19-26). This indicates the OCR did not read row by row but column by column? Or the table is split across pages? Actually, the OCR might have read the table column by column: first all names, then all register numbers, then all sterling amounts, then all dollar amounts, then all dates, then all particulars, then all ages. That would explain the grouping: first a list of names, then register numbers, then sterling amounts, then dollar amounts, then dates, then particulars, then ages.

Let's test this hypothesis. Look at the tokens: After Brought forward totals, we have a series of names: Lui Ying, Matilda M. Angus, Lau Yat Kin, Ebrahim Khan. Then register numbers: 1,834, 3, 1,474, 1,408. Then sterling amounts: for Matilda: 70 15 10 (only one sterling amount shown? Actually token12 is "70 15 10" after Matilda's register 3. But Lui Ying's sterling not shown. Lau Yat Kin's sterling not shown. Ebrahim Khan's sterling not shown. Then dollar amounts: 49.73 (Lui Ying), 171.82 (Lau Yat Kin), 207.10 (Ebrahim Khan). But Matilda's dollar amount not shown. Then dates: 27th June, 6th July, 22nd July, Do. (four dates for four pensioners). Then particulars: Widow of Chung Ju, Widow of Peter Augus, Orphan of Lan Ka Lok, Orphan of Nawab Khan. Then ages: 42, 54, 4, (maybe missing for fourth). Then next group: Yim Sz, Cheung Sau In, Ho Chiu Shi, Catherine..., Theodora..., Tang Yue, Chan So, Lilian..., Mary..., Wong Tso Pik. That's 10 names? Let's count: Yim Sz, Cheung Sau In, Ho Chiu Shi, Catherine Johnston Anderson Kynoch, Theodora Maria Wilkinson, Tang Yue, Chan So, Lilian Howard Wheal, Mary Hall Muchart Reid, Wong Tso Pik. That's 10 names. Then register numbers: 1,598, 499, 179, 1,012, 487, 1,202, 1,979, 488, 1,735, 1,397. That's 10 register numbers. Then sterling/dollar amounts: 251.21 (Yim Sz), 99.15 (Cheung Sau In), 58,79 (Ho Chiu Shi), 91 3 5 (Catherine), 268.47 (Theodora), 281.80 (Tang Yue), 85.01 (Chan So?), 372.29 (Lilian?), 64 0 4 (Mary), 277.44 (Wong Tso Pik). But there are 10 amounts, but some are sterling, some dollars. Then dates: 22nd August, 12th November, 3rd December, 6th February, 22nd February, 1st March, 10th March, 29th March, 7th May, 7th July. That's 10 dates. Then particulars: Widow of Fan Tsung Hing, Widow of Kwok Yuk Tin, Widow of Ho Fung Cheung, Widow of James Henry Kynoch, Widow of William Josiah Wilkinson, Widow of Fung Kam Tin, Widow of Kwong In, Widow of Isaac Alfred Wheal, Widow of Alexander Reid, Widow of Chan Shiu To. That's 10 particulars. Then ages: 16, 51, 63, 41, 52, 40, 50, 32, 28? Wait we have ages: after first group we had 42,54,4. Second group ages: tokens 35,36: 16,51; token41:63; tokens62,63:41,52; token84:40; tokens87,88:50,32; token94:28. That's 10 ages? Let's list: 16,51,63,41,52,40,50,32,28. That's 9 ages. Missing one? Maybe the age for Chan So is missing? Or for Tang Yue? Actually Tang Yue's particulars "Widow of Fung Kam Tin" then "3333 (Fung Tang Shi)" maybe age 3333? No.

But the pattern of column-wise reading seems plausible. The OCR likely read the table column by column because of the layout (maybe the table was split across pages or had vertical lines). The instruction says: "Restore column reading order — if text was originally in multiple columns but OCR read it in the wrong order, reorder it to make logical sense (left-to-right, top-to-bottom)." So we need to reorder to row-wise.

Thus we need to reconstruct rows by aligning columns.

We have 7 columns: Name, Register, Sterling, Dollars, Date, Particulars, Age.

We have a list of names (including Brought forward). Let's compile all names in order as they appear in the name column.

From tokens, the names appear at tokens: 4 (Brought forward), 7 (Lui Ying), 10 (Matilda M. Angus), 13 (Lau Yat Kin), 16 (Ebrahim Khan), 30 (Yim Sz), 37 (Cheung Sau In), 43 (Ho Chiu Shi), 50-51 (Catherine Johnston Anderson Kynoch), 55 (Theodora Maria Wilkinson), 64 (Tang Yue), 71 (Chan So), 73 (Lilian Howard Wheal), 77 (Mary Hall Muchart Reid), 89 (Wong Tso Pik). That's 15 names (including Brought forward). But Brought forward is a total row, not a pensioner. So 14 pensioners.

Now register numbers: tokens: 5? Actually Brought forward register? Not given. Then token8: 1,834 (Lui Ying), token11: 3 (Matilda), token14: 1,474 (Lau Yat Kin), token17: 1,408 (Ebrahim), token31: 1,598 (Yim Sz), token38: 499 (Cheung Sau In), token44: 179 (Ho Chiu Shi), token52: 1,012 (Catherine), token56: 487 (Theodora), token65: 1,202 (Tang Yue), token72: 1,979 (Chan So), token74: 488 (Lilian), token78: 1,735 (Mary), token90: 1,397 (Wong). That's 14 register numbers (excluding Brought forward). Good.

Now sterling amounts: We need to identify which tokens are sterling. Sterling format: £ s. d. like "70 15 10", "91 3 5", "64 0 4". Also "3,330 15 4" for Brought forward. Others are dollar amounts (decimal). So we have sterling for: Brought forward: 3,330 15 4; Matilda: 70 15 10; Catherine: 91 3 5; Mary: 64 0 4. Possibly others have sterling but not shown (maybe zero). The OCR might have only captured non-zero sterling amounts. But in column-wise reading, the sterling column would have entries for each row, but blank for 14 rows. However, we only see 4 sterling amounts (plus Brought forward). That suggests that the sterling column might have been read but many entries are empty or zero and OCR didn't capture them. But the token list doesn't have placeholders for empty cells. So the column-wise reading might have omitted empty cells. That complicates alignment.

Similarly, dollar amounts: tokens: 6: 30,036.67 (Brought forward), 9: 49.73 (Lui Ying), 15: 171.82 (Lau Yat Kin), 18: 207.10 (Ebrahim), 32: 251.21 (Yim Sz), 39: 99.15 (Cheung Sau In), 45: 58,79 (Ho Chiu Shi), 57: 268.47 (Theodora), 66: 281.80 (Tang Yue), 75: 85.01 (Chan So?), 76: 372.29 (Lilian?), 91: 277.44 (Wong). That's 12 dollar amounts (including Brought forward). But we have 14 pensioners. Missing dollar amounts for Matilda, Catherine, Mary? But they have sterling. So maybe each pensioner has either sterling or dollar, not both. The table might have two separate amount columns, but only one is filled per row. In column-wise reading, the dollar column would have entries for those paid in dollars, and blank for others. But again, blanks not captured.

Dates: tokens: 19: 27th June, 1925; 20: 6th July, 1925; 21: 22nd July, 1925; 22: Do. (ditto, meaning same as previous? Actually "Do." likely means same date as previous? But there are four dates for first four pensioners). Then 33: 22nd August, 1925; 40: 12th November, 1925; 46: 3rd December, 1925; 58: 6th February, 1926; 59: 22nd February, 1926; 67: 1st March, 1926; 80: 10th March, 1926; 81: 29th March, 1926; 82: 7th May, 1926; 92: 7th July, 1926. That's 14 dates (including the "Do." as a date for the fourth pensioner). Good.

Particulars: tokens: 23: Widow of Chung Ju...; 24: Widow of Peter Augus...; 27: Orphan of Lan Ka Lok...; 28: Orphan of Nawab Khan...; 34: Widow of Fan Tsung Hing...; 40 (part): Widow of Kwok Yuk Tin...; 47: Widow of Ho Fung Cheung...; 60: Widow of James Henry Kynoch...; 61: Widow of William Josiah Wilkinson...; 68: Widow of Fung Kam Tin...; 83: Widow of Kwong In...; 85: Widow of Isaac Alfred Wheal...; 86: Widow of Alexander Reid...; 93: Widow of Chan Shiu To... That's 14 particulars.

Ages: tokens: 25: 42; 26: 54; 29: 4; 35: 16; 36: 51; 41: 63; 62: 41; 63: 52; 84: 40; 87: 50; 88: 32; 94: 28. That's 12 ages. Missing ages for two pensioners: maybe Chan So and Tang Yue? But we have 14 pensioners, 12 ages. Token 69: "3333" might be an age? But 3333 is unrealistic. Token 70: "(Fung Tang Shi)" not age. Token 76: 372.29 is amount. So two ages missing.

But note: the first group had 4 pensioners, ages: 42, 54, 4, and missing for Ebrahim Khan? Actually we have 4 ages for first group? Tokens 25,26,29: 42,54,4. That's three ages for four pensioners. The fourth pensioner (Ebrahim Khan) age missing. Second group: 10 pensioners, ages: 16,51,63,41,52,40,50,32,28 = 9 ages. Missing one age. So total 12 ages for 14 pensioners.

Now, we need to match each pensioner with their data across columns. Since the OCR read column by column, the order within each column should correspond to the same row order. So the first name (after Brought forward) corresponds to first register, first sterling (if any), first dollar (if any), first date, first particulars, first age. But because some columns have missing entries (blanks), the alignment might be off if we don't account for blanks.

However, the typical OCR of a table with empty cells might still output something for each cell? But here it seems the OCR only output non-empty cells. But the order of non-empty cells in each column should still follow the row order. For example, the sterling column has non-empty entries for rows: Brought forward (row0), Matilda (row2?), Catherine (row8?), Mary (row13?). But we need to know which rows have sterling entries. The dollar column has non-empty entries for rows: Brought forward, Lui Ying (row1), Lau Yat Kin (row3), Ebrahim (row4), Yim Sz (row5), Cheung Sau In (row6), Ho Chiu Shi (row7), Theodora (row9), Tang Yue (row10), Chan So (row11), Lilian (row12), Wong (row14). That's 12 entries for 14 rows (excluding Brought forward). The missing dollar entries are for Matilda, Catherine, Mary (who have sterling). So that matches: rows with sterling have no dollar, rows with dollar have no sterling. Good.

Thus we can align by assuming each row has either a sterling or a dollar amount (or both? but not both in this data). The Brought forward row has both.

Now, the dates column has 14 entries (one per pensioner). The particulars column has 14 entries. The ages column has 12 entries (two missing). The register column has 14 entries.

Thus we can reconstruct the table by taking the 14 pensioners in the order of names (excluding Brought forward). The order of names as they appear in the name column:

1. Lui Ying
2. Matilda M. Angus
3. Lau Yat Kin
4. Ebrahim Khan
5. Yim Sz
6. Cheung Sau In
7. Ho Chiu Shi
8. Catherine Johnston Anderson Kynoch
9. Theodora Maria Wilkinson
10. Tang Yue
11. Chan So
12. Lilian Howard Wheal
13. Mary Hall Muchart Reid
14. Wong Tso Pik

Now assign registers in same order:

1. 1,834
2. 3
3. 1,474
4. 1,408
5. 1,598
6. 499
7. 179? Wait check: register list: 1,834; 3; 1,474; 1,408; 1,598; 499; 179; 1,012; 487; 1,202; 1,979; 488; 1,735; 1,397. That's 14 registers. So:
8. 1,834
9. 3
10. 1,474
11. 1,408
12. 1,598
13. 499
14. 179
15. 1,012
16. 487
17. 1,202
18. 1,979
19. 488
20. 1,735
21. 1,397

Now sterling amounts: non-empty in order of rows? The sterling column non-empty entries (excluding Brought forward) appear at tokens: 12 (70 15 10) after Matilda's register? But in column-wise reading, the sterling column would be read top to bottom. The first pensioner (Lui Ying) likely has no sterling (blank). Second (Matilda) has 70 15 10. Third (Lau Yat Kin) blank. Fourth (Ebrahim) blank. Fifth (Yim Sz) blank. Sixth (Cheung Sau In) blank. Seventh (Ho Chiu Shi) blank. Eighth (Catherine) has 91 3 5. Ninth (Theodora) blank. Tenth (Tang Yue) blank. Eleventh (Chan So) blank. Twelfth (Lilian) blank. Thirteenth (Mary) has 64 0 4. Fourteenth (Wong) blank. That matches the tokens: we have sterling amounts at positions corresponding to rows 2, 8, 13. Good.

Dollar amounts: non-empty in order of rows: row1 (Lui Ying) 49.73; row2 (Matilda) blank; row3 (Lau Yat Kin) 171.82; row4 (Ebrahim) 207.10; row5 (Yim Sz) 251.21; row6 (Cheung Sau In) 99.15; row7 (Ho Chiu Shi) 58.79; row8 (Catherine) blank; row9 (Theodora) 268.47; row10 (Tang Yue) 281.80; row11 (Chan So) 85.01; row12 (Lilian) 372.29; row13 (Mary) blank; row14 (Wong) 277.44. That matches the dollar tokens in order: 49.73, 171.82, 207.10, 251.21, 99.15, 58.79, 268.47, 281.80, 85.01, 372.29, 277.44. But wait, the token list has 49.73 (token9), then 171.82 (token15), 207.10 (token18), 251.21 (token32), 99.15 (token39), 58.79 (token45), 268.47 (token57), 281.80 (token66), 85.01 (token75), 372.29 (token76), 277.44 (token91). That's 11 dollar amounts for 14 rows (excluding Brought forward). But we predicted 11 rows with dollar amounts (all except Matilda, Catherine, Mary). That's 11. Good.

Dates: 14 dates in order:

1. 27th June, 1925
2. 6th July, 1925
3. 22nd July, 1925
4. Do. (meaning same as previous? But "Do." likely means "Ditto" for 22nd July? Actually "Do." might be for the fourth pensioner, meaning same date as third? But the third is 22nd July, so fourth also 22nd July? Or maybe "Do." refers to the date column being same as above? In tables, "Do." often means "same as above". But here each pensioner has a distinct date? The first three have distinct dates, the fourth might be same as third? But the token "Do." appears after three dates. Could be that the fourth pensioner's date is also 22nd July? But then we have 14 dates for 14 pensioners. Let's list the date tokens in order: token19: 27th June, 1925; token20: 6th July, 1925; token21: 22nd July, 1925; token22: Do.; token33: 22nd August, 1925; token40: 12th November, 1925; token46: 3rd December, 1925; token58: 6th February, 1926; token59: 22nd February, 1926; token67: 1st March, 1926; token80: 10th March, 1926; token81: 29th March, 1926; token82: 7th May, 1926; token92: 7th July, 1926. That's 14 date entries. The "Do." is the fourth entry. So we'll treat "Do." as "22nd July, 1925" (same as previous). But maybe it's a separate date? Could be "Do." means "Ditto" for the date column, but the date column might have "Do." for multiple rows? However, only one "Do." appears. So we'll assign the fourth pensioner (Ebrahim Khan) the date "22nd July, 1925" (same as third). But the third is Lau Yat Kin with date 22nd July. So both Lau Yat Kin and Ebrahim Khan have same date? Possible.

But note: the first four pensioners are Lui Ying, Matilda, Lau Yat Kin, Ebrahim. Their dates: 27th June, 6th July, 22nd July, 22nd July (Do.). That seems plausible.

Now particulars: 14 entries in order:

1. Widow of Chung Ju, late of Imports and Exports Department
2. Widow of Peter Augus, late of Police Department
3. Orphan of Lan Ka Lok, late of Police Department
4. Orphan of Nawab Khan, late of Post Office
5. Widow of Fan Tsung Hing, late of Harbour Department
6. Widow of Kwok Yuk Tin, late of District Office South (Kwok Cheung Shi)
7. Widow of Ho Fung Cheung, late of Education Department
8. Widow of James Henry Kynoch, late of Public Works Department
9. Widow of William Josiah Wilkinson, late of Prison Department
10. Widow of Fung Kam Tin, late of Kowloon Canton Railway (Fung Tang Shi)
11. Widow of Kwong In, late of Sanitary Department
12. Widow of Isaac Alfred Wheal, late of Public Works Department
13. Widow of Alexander Reid, late of Police Department
14. Widow of Chan Shiu To, late of Education Department

Ages: 12 entries in order (but missing two). The age tokens in order: 42, 54, 4, 16, 51, 63, 41, 52, 40, 50, 32, 28. That's 12 ages. Which rows have ages? Likely the first 12 rows have ages, last two missing? But we have 14 rows. Let's see: The age tokens appear after particulars in the token stream. The first three ages (42,54,4) correspond to first three pensioners? But there are four pensioners in first group. The fourth pensioner (Ebrahim) age missing. Then next ages: 16,51,63 correspond to pensioners 5,6,7? But pensioner 5 (Yim Sz) age 16? 16 seems young for a widow. Pensioner 6 (Cheung Sau In) age 51? Pensioner 7 (Ho Chiu Shi) age 63? Then ages 41,52 for pensioners 8 and 9 (Catherine and Theodora). Then 40,50,32 for pensioners 11,12,13? But pensioner 10 (Tang Yue) age missing? Then 28 for pensioner 14 (Wong Tso Pik). And pensioner 11 (Chan So) age 40? But Chan So is pensioner 11, age 40. Pensioner 12 (Lilian) age 50. Pensioner 13 (Mary) age 32. Pensioner 14 (Wong) age 28. That leaves pensioner 10 (Tang Yue) without age. Also pensioner 4 (Ebrahim) without age. So two missing ages: Ebrahim Khan and Tang Yue.

But the age token "4" is for pensioner 3 (Lau Yat Kin) who is an orphan, age 4. Good. "16" for pensioner 5 (Yim Sz) - but Yim Sz is a widow, age 16? That seems very young. Could be that "16" is not age but something else? But it's in the age column. Maybe the age column includes both age and something else? Or maybe the order of ages is not exactly aligned because of missing entries. Let's check the token positions: After particulars for first group, we have ages 42,54,4. Then after Yim Sz's particulars (token34), we have tokens 35:16, 36:51. Then after Cheung Sau In's particulars (token40), token41:63. Then after Ho Chiu Shi's particulars (token47), tokens48:544, 49:68. Wait tokens48 and 49 are "544" and "68". Those are not ages? They appear before Catherine's name. Actually token48: "544", token49: "68". These might be sterling/dollar amounts for Ho Chiu Shi? But we already assigned dollar 58.79 to Ho Chiu Shi. 544 and 68 could be something else. Let's re-examine tokens 544 could be a register number? But register for Ho Chiu Shi is 179. 544 might be a sterling amount? £544? But not in £ s. d. format. 68 could be age. But then age for Ho Chiu Shi would be 68. But we have age 63 from token41? Token41 is after Cheung Sau In's particulars. The sequence: token37: Cheung Sau In, token38: 499, token39: 99.15, token40: 12th November, 1925. | Widow of Kwok Yuk Tin..., token41: 63, token42: (Kwok Cheung Shi).........., token43: Ho Chiu Shi, token44: 179, token45: 58,79, token46: 3rd December, 1925., token47: Widow of Ho Fung Cheung..., token48: 544, token49: 68, token50: Catherine Johnston An-...

So token41: 63 is after Cheung Sau In's particulars. That could be age for Cheung Sau In (63). Token48: 544 and token49: 68 appear after Ho Chiu Shi's particulars. 544 might be a sterling amount for Ho Chiu Shi? But Ho Chiu Shi already has dollar 58.79. Could be that Ho Chiu Shi has both sterling and dollar? 544 could be £544? But no shillings/pence. 68 could be age. Then Catherine's sterling is 91 3 5 (token53). So maybe the sterling column for Ho Chiu Shi is 544? But 544 is not in £ s. d. format. Could be 544 dollars? But dollar column already has 58.79. Hmm.

Alternatively, the OCR might have misread the table lines. The "544 68" might be part of the next row? But Catherine's register is 1,012. 544 is not that.

Let's look at the original OCR lines around that:

"Ho Chiu Shi,

179

58,79

3rd December, 1925.

Widow of Ho Fung Cheung, late of Education Department,.

544

68

Catherine Johnston An-

derson Kynoch,

1,012

91 3 5

---

Theodora Maria Wilkinson,

487

268.47"

The "544" and "68" are on separate lines. Could be that "544" is the register number for Catherine? But Catherine's register is 1,012. Or maybe "544" is a page number? But "68" could be age for Ho Chiu Shi. But then why "544"?

Maybe the table has a column for "Amount of the Pension in Sterling" and "Amount of the Pension in Dollars". For Ho Chiu Shi, the dollar amount is 58.79. The sterling amount might be blank. But "544" could be the sterling amount for the next pensioner? But Catherine's sterling is 91 3 5.

Wait, the token "---" appears after Catherine's sterling. That might be a separator.

Given the complexity, I think the column-wise alignment is the most reliable. The OCR read each column top to bottom. The columns are: Name, Register, Sterling, Dollars, Date, Particulars, Age. The tokens we have for each column (in order) are:

Name column (14 pensioners + Brought forward):

1. Brought forward
2. Lui Ying
3. Matilda M. Angus
4. Lau Yat Kin
5. Ebrahim Khan
6. Yim Sz
7. Cheung Sau In
8. Ho Chiu Shi
9. Catherine Johnston Anderson Kynoch
10. Theodora Maria Wilkinson
11. Tang Yue
12. Chan So
13. Lilian Howard Wheal
14. Mary Hall Muchart Reid
15. Wong Tso Pik

Register column (14 entries):

1. 1,834
2. 3
3. 1,474
4. 1,408
5. 1,598
6. 499
7. 179
8. 1,012
9. 487
10. 1,202
11. 1,979
12. 488
13. 1,735
14. 1,397

Sterling column (14 entries, but only some non-empty): We need to infer which rows have sterling. From tokens, non-empty sterling amounts appear in the token stream at positions that likely correspond to rows 2, 8, 13 (Matilda, Catherine, Mary). Also Brought forward has sterling. So for the 14 pensioners, sterling amounts: row1: blank, row2: £70 15s 10d, row3: blank, row4: blank, row5: blank, row6: blank, row7: blank, row8: £91 3s 5d, row9: blank, row10: blank, row11: blank, row12: blank, row13: £64 0s 4d, row14: blank.

Dollar column (14 entries): non-empty for rows: 1: $49.73, 2: blank, 3: $171.82, 4: $207.10, 5: $251.21, 6: $99.15, 7: $58.79, 8: blank, 9: $268.47, 10: $281.80, 11: $85.01, 12: $372.29, 13: blank, 14: $277.44.

Date column (14 entries):

1. 27th June, 1925
2. 6th July, 1925
3. 22nd July, 1925
4. 22nd July, 1925 (Do.)
5. 22nd August, 1925
6. 12th November, 1925
7. 3rd December, 1925
8. 6th February, 1926
9. 22nd February, 1926
10. 1st March, 1926
11. 10th March, 1926
12. 29th March, 1926
13. 7th May, 1926
14. 7th July, 1926

Particulars column (14 entries): as listed above.

Age column (14 entries, but only 12 non-empty in tokens). The non-empty ages in order of rows: from tokens, the ages appear in the token stream after particulars. The first three ages (42,54,4) correspond to rows 1,2,3? But row4 missing. Then next ages: 16,51,63 correspond to rows 5,6,7? Then 41,52 correspond to rows 8,9. Then 40,50,32 correspond to rows 11,12,13? Then 28 corresponds to row14. Row10 (Tang Yue) missing. Row4 (Ebrahim) missing. So ages:

1. 42
2. 54
3. 4
4. (missing)
5. 16
6. 51
7. 63
8. 41
9. 52
10. (missing)
11. 40
12. 50
13. 32
14. 28

But check: Row5 is Yim Sz, age 16? That seems young. Row6 Cheung Sau In age 51. Row7 Ho Chiu Shi age 63. Row8 Catherine age 41. Row9 Theodora age 52. Row11 Chan So age 40. Row12 Lilian age 50. Row13 Mary age 32. Row14 Wong age 28. Row4 Ebrahim missing. Row10 Tang Yue missing.

But the token "544" and "68" appear after Ho Chiu Shi's particulars. In our column-wise alignment, the age for Ho Chiu Shi is 63 (token41). But token48 is 544, token49 is 68. Those are not in the age column if we have 14 ages. They might be extraneous or from another column. Could be that the sterling column for Ho Chiu Shi is 544? But 544 is not in £ s. d. format. Could be that the dollar column for Ho Chiu Shi is 58.79, and the sterling column is 544? But 544 what? £544? That would be huge. Unlikely.

Maybe the OCR captured some numbers from the "Particulars" column that look like numbers? For example, "District Office South, 63" - the 63 is part of the particulars (age of Kwok Cheung Shi?). Actually token40: "12th November, 1925. | Widow of Kwok Yuk Tin, late of District Office South," token41: "63" token42: "(Kwok Cheung Shi).........." So the 63 might be the age of the widow (Kwok Cheung Shi) and is part of the particulars. But in the table, age is a separate column. However, the particulars might include the age in parentheses. In the original table, the "Particulars" column might contain the description and the age in parentheses, and the "Present Age of Pensioner" column might be separate. But the OCR might have split them.

Look at the first group: Particulars: "Widow of Chung Ju, late of Imports and Exports Department,.................." then ages 42 and 54 are separate. For the second widow, "Widow of Peter Augus, late of Police Department," then ages 42 and 54? Actually two ages for two pensioners. So the age column is separate.

For Cheung Sau In, the particulars: "Widow of Kwok Yuk Tin, late of District Office South, 63 (Kwok Cheung Shi).........." Here the "63" appears in the particulars line, but then the age column might have 63 as well? But we have token41: 63, which we assigned as age for Cheung Sau In. That matches.

For Ho Chiu Shi: "Widow of Ho Fung Cheung, late of Education Department,. 544 68". The "544 68" might be two numbers: maybe 544 is a register number? But register is 179. Could be that the particulars include "544" and "68" as ages? But 544 is not an age. Could be a typo for "54 4"? But no.

Maybe the "544" is the sterling amount for Ho Chiu Shi? But the sterling column for Ho Chiu Shi is blank in our alignment. However, the token "544" appears after the particulars and before Catherine's name. In the column-wise reading, after the particulars column comes the age column. So the age column for Ho Chiu Shi would be the next token after her particulars. But we have two tokens: 544 and 68. Which one is age? 68 is plausible age. 544 might be something else, maybe a reference number. But the age column should have one entry per row. So perhaps the age for Ho Chiu Shi is 68, and 544 is a stray token from the sterling column? But the sterling column for Ho Chiu Shi would have been read earlier (in the sterling column pass). The sterling column pass would have produced tokens for each row. We only have three sterling tokens for pensioners (Matilda, Catherine, Mary). So Ho Chiu Shi's sterling is blank. So 544 is not from sterling column.

Maybe the OCR read the table row by row but with merged cells? This is getting too speculative.

Given the instructions, I should produce a Markdown table that represents the data as clearly as possible, based on the OCR text. I think the best is to present the table with the columns as per the header, and fill in the data for each pensioner as we can deduce, using the column-wise alignment for the main fields, and for the problematic ones, use the text as it appears.

I will create a table with 14 rows (plus a header row and a "Brought forward" row). I'll include the Brought forward row as the first row.

Let's construct the table.

Columns:

• Name of Pensioner
• No. of Pension Register
• Amount of Pension in Sterling (per annum)
• Amount of Pension in Dollars (per annum)
• Date from which Pension has been paid
• Particulars
• Present Age of Pensioner

Row 0: Brought forward

• Name: Brought forward
• Register: (blank)
• Sterling: £3,330 15s 4d
• Dollars: $30,036.67
• Date: (blank)
• Particulars: (blank)
• Age: (blank)

Row 1: Lui Ying

• Name: Lui Ying
• Register: 1,834
• Sterling: (blank)
• Dollars: $49.73
• Date: 27th June, 1925
• Particulars: Widow of Chung Ju, late of Imports and Exports Department
• Age: 42

Row 2: Matilda M. Angus

• Name: Matilda M. Angus
• Register: 3
• Sterling: £70 15s 10d
• Dollars: (blank)
• Date: 6th July, 1925
• Particulars: Widow of Peter Augus, late of Police Department
• Age: 54

Row 3: Lau Yat Kin

• Name: Lau Yat Kin
• Register: 1,474
• Sterling: (blank)
• Dollars: $171.82
• Date: 22nd July, 1925
• Particulars: Orphan of Lan Ka Lok, late of Police Department
• Age: 4

Row 4: Ebrahim Khan

• Name: Ebrahim Khan
• Register: 1,408
• Sterling: (blank)
• Dollars: $207.10
• Date: 22nd July, 1925 (Do.)
• Particulars: Orphan of Nawab Khan, late of Post Office
• Age: (missing) — but we have no age token for him. Could be missing. I'll leave blank or put "..."? The instruction: if missing, insert .... But we don't know if it's missing or just not captured. I'll put ....

Row 5: Yim Sz

• Name: Yim Sz
• Register: 1,598
• Sterling: (blank)
• Dollars: $251.21
• Date: 22nd August, 1925
• Particulars: Widow of Fan Tsung Hing, late of Harbour Department
• Age: 16 (but seems young; however, token says 16). I'll put 16.

Row 6: Cheung Sau In

• Name: Cheung Sau In
• Register: 499
• Sterling: (blank)
• Dollars: $99.15
• Date: 12th November, 1925
• Particulars: Widow of Kwok Yuk Tin, late of District Office South (Kwok Cheung Shi)
• Age: 51? Wait token41 is 63, but we have two ages 16 and 51 for Yim Sz? Actually tokens 35:16, 36:51. Which belongs to whom? Yim Sz is row5, Cheung Sau In row6. The ages in order: after Yim Sz's particulars, we have 16 and 51. But there are two numbers. Could be that Yim Sz age 16, Cheung Sau In age 51. But then token41:63 appears after Cheung Sau In's particulars. That would be a third age. But we have only 14 rows. Let's see the age tokens in order as they appear in the OCR after each particulars? The OCR might have listed ages in a separate column, but the tokens are interleaved. The column-wise approach says ages column has 14 entries in order. The age tokens we extracted in order from the token stream (assuming they appear in the age column) are: 42,54,4,16,51,63,41,52,40,50,32,28. That's 12 entries. If we map them to rows 1-14 sequentially, skipping rows where age might be missing? But we have 14 rows. The first three ages (42,54,4) for rows1-3. Row4 missing. Then next three (16,51,63) for rows5-7. Then next two (41,52) for rows8-9. Row10 missing. Then next three (40,50,32) for rows11-13. Then last (28) for row14. That gives ages for rows1,2,3,5,6,7,8,9,11,12,13,14. Rows4 and 10 missing. That seems plausible.

Thus:

Row5 (Yim Sz) age 16

Row6 (Cheung Sau In) age 51

Row7 (Ho Chiu Shi) age 63

Row8 (Catherine) age 41

Row9 (Theodora) age 52

Row10 (Tang Yue) missing

Row11 (Chan So) age 40

Row12 (Lilian) age 50

Row13 (Mary) age 32

Row14 (Wong) age 28

But wait, row7 Ho Chiu