1923 — Page 55

The user wants me to proofread OCR output of a historical document from Hong Kong. This appears to be a comparative yearly statement of expenditure for 1922 and 1928 (though the header says 1922 and 1928, but the data seems to be for 1922 and 1923? Wait, the header says "Expenditure, 1922. Expenditure, 1928." but the data shows 1922 and 1923? Let me check: The OCR shows "Expenditure, 1922." and "Expenditure, 1928." but the numbers look like they might be for 1922 and 1923? Actually, the document title says "COMPARATIVE YEARLY STATEMENT OF THE EXPENDITURE." and the columns are "Expenditure, 1922." and "Expenditure, 1928." but the data includes "Japanese Earthquake $250,000, British Empire Exhibition $250,000." which suggests 1923 (Great Kanto Earthquake was 1923, British Empire Exhibition was 1924-25). Hmm, maybe it's 1922 and 1923? But the header says 1928. Could be a typo in OCR. I'll keep as OCR shows: 1922 and 1928.

The OCR text is messy. I need to reconstruct the table and the "Cause of Increase or Decrease" column. The OCR seems to have merged the cause column with the table rows. Let me parse.

The original appears to be a table with columns:

1. Cause of Increase or Decrease (description)
2. Expenditure, 1922
3. Expenditure, 1928 (or 1923?)
4. Increase
5. Decrease

But the OCR has the cause descriptions interleaved with the department rows. Actually, looking at the OCR: It starts with "Cause of Increase or Decrease." then "COMPARATIVE YEARLY STATEMENT OF THE EXPENDITURE." then "Specify each separate Head of Expenditure according to the classi- fication employed in the Estimates," then column headers: "Expenditure, 1922. Expenditure, 1928. Increuse. Decrease." (Increase misspelled).

Then there are lines like:

"Promotions and stipulated increments.

Increased personnel.

Lapsing salaries.

New stamping maeltine.

New posts and typhoon damage.

More opium purchased, lower exchange, New Motor Truck,

Japanese Earthquake $250,000, British Empire Exhibition $250,000.

Increased personnel, more passages,

More patients,

Increased personnel, Vaccination, New Motor Vehicles.

Removal and formation of nursery for Kowloon Development Scheme. Increased personnel and grants.

Increase in revenue on which Defence Contribution is payable. Increased personnel, transport and travelling expenses.

More work lone.

Do.

Transit charges for 1920/1921 recovered during 1923.

Suspension of through traffic, certain works not proceeded with, Lower exchange.

More pensioners,

6.

Governor,

94.287.62

94,100.92

C.

C.

186.70

Cadet Service,--

280,567.64

300,211.07

19,643.51

Colonial Secretary's Department and

Legislature,.....

56,454.63

60,759.08

4,304.45

Secretariat for Chinese Affairs,

21,115,67

28.795.63

7,679.96

Audit Department,...

51,060.35

48.196.15

2,864.20

Тгеннигу,

68,367.64

70,480,33

2,512.69

Harbour Master's Department,.

310,717,36

316,994,95

Imports and Exports Department,

631,733.97

908,095.52

Royal Observatory, .....

38,350.10

38,374.11

Miscellaneous Services..........

1,617,305.35

1,828,078.38

Indicial and Legal Departments,

282,989.15

285,009,31

Police,.....

1,533,757.51

1,633,832.14

6.277,59 276,361.55 24.01 210,773.03 2,020.16 100,074.63

Firo Brigade,

177,799,07

166,783.43

11,015.64

Prison Department,

324,175.12

324.698.26

Medical Departments,

371,781.92

402,761.76

523.14 30,979.84

Sanitary Department,

548,824.201

620,390.31

71,566.11

Hotanical and Forestry Department,

71,223.47

77,157.40

5,933.93

Education,

Military Expenditure,

Public Works Department,

728,133,89 3,230,779.89

856,367,04

128,213.15

4,033,500.36

802.720.47

4

821,674.89

901,782.71

Do.

Recurrent,

1,074,646,30

Do. Extraordinary,

8,575,635.19

1,424,532.80 4,716,602.94

80,107.82 349,886,50

1,140,967.75

Post Office..................

309.042.61

123,492,33

Kowloon-Canton Railway,

922,855.75

757,030.11

185,550,28 165,325,64

Charge on account of Public Debt,

914,040.12

964,810,00

50,769.88

Pensions,...

429,361.04

486,167.76

56,806.72

Charitable Services,

76,802.56

102,499.94

25,697.38

Total..............

18,563,002.91

21,371,904.72

3,373,844.27

364,942.46

(07)

53"

This is a mess. The cause descriptions likely correspond to each department row. But the OCR has dumped all cause descriptions at the top, then the table rows. Actually, the original document probably had a column "Cause of Increase or Decrease" on the left, then the department name, then the figures. The OCR read the left column first (all causes), then the department names and figures. But the cause descriptions are not aligned with departments in the OCR output. I need to reconstruct the table by matching each department with its cause.

Let's list the departments in order as they appear in the table rows:

1. Governor
2. Cadet Service
3. Colonial Secretary's Department and Legislature
4. Secretariat for Chinese Affairs
5. Audit Department
6. Treasury (Тгеннигу is likely "Treasury")
7. Harbour Master's Department
8. Imports and Exports Department
9. Royal Observatory
10. Miscellaneous Services
11. Judicial and Legal Departments (Indicial -> Judicial)
12. Police
13. Fire Brigade (Firo -> Fire)
14. Prison Department
15. Medical Departments
16. Sanitary Department
17. Botanical and Forestry Department (Hotanical -> Botanical)
18. Education
19. Military Expenditure
20. Public Works Department
21. Post Office
22. Kowloon-Canton Railway
23. Charge on account of Public Debt
24. Pensions
25. Charitable Services
26. Total

Now the cause descriptions (in order as they appear in OCR):

1. Promotions and stipulated increments.
2. Increased personnel.
3. Lapsing salaries.
4. New stamping maeltine (machine).
5. New posts and typhoon damage.
6. More opium purchased, lower exchange, New Motor Truck,
7. Japanese Earthquake $250,000, British Empire Exhibition $250,000.
8. Increased personnel, more passages,
9. More patients,
10. Increased personnel, Vaccination, New Motor Vehicles.
11. Removal and formation of nursery for Kowloon Development Scheme. Increased personnel and grants.
12. Increase in revenue on which Defence Contribution is payable. Increased personnel, transport and travelling expenses.
13. More work lone. (likely "More work done.")
14. Do. (ditto)
15. Transit charges for 1920/1921 recovered during 1923.
16. Suspension of through traffic, certain works not proceeded with, Lower exchange.
17. More pensioners,
18. (then "6." appears, maybe a page number or item number)

There are 17 cause descriptions before the "6." but we have 26 departments. Some departments may have no cause listed? Or the cause descriptions correspond to the first 17 departments? Let's see: The first department is Governor, cause maybe "Promotions and stipulated increments." Second: Cadet Service -> "Increased personnel." Third: Colonial Secretary's Department -> "Lapsing salaries." Fourth: Secretariat for Chinese Affairs -> "New stamping machine." Fifth: Audit Department -> "New posts and typhoon damage." Sixth: Treasury -> "More opium purchased, lower exchange, New Motor Truck," Seventh: Harbour Master's Department -> "Japanese Earthquake $250,000, British Empire Exhibition $250,000." Eighth: Imports and Exports Department -> "Increased personnel, more passages," Ninth: Royal Observatory -> "More patients," Tenth: Miscellaneous Services -> "Increased personnel, Vaccination, New Motor Vehicles." Eleventh: Judicial and Legal Departments -> "Removal and formation of nursery for Kowloon Development Scheme. Increased personnel and grants." Twelfth: Police -> "Increase in revenue on which Defence Contribution is payable. Increased personnel, transport and travelling expenses." Thirteenth: Fire Brigade -> "More work lone." Fourteenth: Prison Department -> "Do." Fifteenth: Medical Departments -> "Transit charges for 1920/1921 recovered during 1923." Sixteenth: Sanitary Department -> "Suspension of through traffic, certain works not proceeded with, Lower exchange." Seventeenth: Botanical and Forestry Department -> "More pensioners," Then the rest (Education, Military, Public Works, Post Office, Railway, Public Debt, Pensions, Charitable Services) have no cause listed? Possibly the cause column continues but OCR didn't capture? Or maybe the cause descriptions are only for those with changes? The table includes Increase and Decrease columns, so each row has either increase or decrease. The cause column explains the change.

Given the OCR output, the cause descriptions are listed separately at the top. In the original document, they were likely in a column beside each department. Since the OCR read column by column, it read the entire cause column first, then the department column, then the figures. So we need to recombine them row by row.

Thus, I should produce a Markdown table with columns: Cause of Increase or Decrease, Head of Expenditure, Expenditure 1922, Expenditure 1928, Increase, Decrease.

But the OCR shows numbers with dots as thousand separators? Actually, numbers like "94.287.62" likely means 94,287.62 (using dot as thousand separator and comma as decimal? But then "94,100.92" uses comma as thousand separator and dot as decimal. Inconsistent. Probably the original uses commas for thousands and dots for decimals. The OCR misread some commas as dots. I'll standardize to commas for thousands and dots for decimals.

Also, the "Increase" and "Decrease" columns: some rows have values in both? Actually, each row should have either increase or decrease, not both. The OCR shows for Governor: "C." and "C." and "186.70" - maybe "C." stands for "Decrease"? Actually, "C." might be a dash or something. The numbers: 94,287.62 and 94,100.92, difference is -186.70 (decrease). So the Increase column blank, Decrease column 186.70. The OCR shows "C." "C." "186.70". Might be that "C." is a placeholder for zero? Or maybe "C." stands for "Decrease"? Not sure. I'll interpret: For Governor, Expenditure 1922: 94,287.62; 1928: 94,100.92; Increase: -; Decrease: 186.70.

Similarly, Cadet Service: 280,567.64 -> 300,211.07; Increase: 19,643.51; Decrease: -.

Colonial Secretary: 56,454.63 -> 60,759.08; Increase: 4,304.45.

Secretariat for Chinese Affairs: 21,115.67 -> 28,795.63; Increase: 7,679.96.

Audit Department: 51,060.35 -> 48,196.15; Decrease: 2,864.20.

Treasury: 68,367.64 -> 70,480.33; Increase: 2,512.69.

Harbour Master: 310,717.36 -> 316,994.95; Increase: 6,277.59? Wait, the OCR shows "6.277,59 276,361.55 24.01 210,773.03 2,020.16 100,074.63" after Police? That line seems garbled. Let's parse carefully.

After "Police,..... 1,533,757.51 1,633,832.14" then "6.277,59 276,361.55 24.01 210,773.03 2,020.16 100,074.63". That might be the Increase/Decrease for Police and Fire Brigade? Actually, Police increase: 1,633,832.14 - 1,533,757.51 = 100,074.63. That matches the last number 100,074.63. The other numbers: 6,277.59? 276,361.55? 24.01? 210,773.03? 2,020.16? Not sure.

Maybe the OCR merged multiple columns. Let's compute differences for each department to verify.

I'll compute expected increases/decreases:

1. Governor: 94,287.62 -> 94,100.92 = -186.70 (Decrease 186.70)
2. Cadet Service: 280,567.64 -> 300,211.07 = +19,643.43 (approx 19,643.51)
3. Colonial Secretary: 56,454.63 -> 60,759.08 = +4,304.45
4. Secretariat for Chinese Affairs: 21,115.67 -> 28,795.63 = +7,679.96
5. Audit: 51,060.35 -> 48,196.15 = -2,864.20
6. Treasury: 68,367.64 -> 70,480.33 = +2,112.69? Wait 70,480.33 - 68,367.64 = 2,112.69. But OCR says 2,512.69. Slight difference. Maybe numbers are 68,367.64 and 70,880.33? But OCR says 70,480,33. Could be 70,480.33. Difference 2,112.69. But OCR shows 2,512.69. Might be a misread. I'll use the OCR's increase figure as given: 2,512.69.
7. Harbour Master: 310,717.36 -> 316,994.95 = +6,277.59. That matches the first number in the garbled line: 6.277,59 (i.e., 6,277.59). Good.
8. Imports and Exports: 631,733.97 -> 908,095.52 = +276,361.55. That matches the second number: 276,361.55.
9. Royal Observatory: 38,350.10 -> 38,374.11 = +24.01. Matches third number: 24.01.
10. Miscellaneous Services: 1,617,305.35 -> 1,828,078.38 = +210,773.03. Matches fourth number: 210,773.03.
11. Judicial and Legal: 282,989.15 -> 285,009.31 = +2,020.16. Matches fifth number: 2,020.16.
12. Police: 1,533,757.51 -> 1,633,832.14 = +100,074.63. Matches sixth number: 100,074.63.

Great! So the garbled line "6.277,59 276,361.55 24.01 210,773.03 2,020.16 100,074.63" are the Increase amounts for departments 7 through 12 (Harbour Master to Police). The Decrease column for those is blank (or zero). So the table has separate Increase and Decrease columns. For each row, only one of them is filled.

Then next: Fire Brigade: 177,799.07 -> 166,783.43 = -11,015.64 (Decrease 11,015.64). OCR shows "11,015.64" after Fire Brigade line? Actually OCR: "Firo Brigade, 177,799,07 166,783.43 11,015.64". That's the Decrease amount.

Prison Department: 324,175.12 -> 324,698.26 = +523.14. OCR shows "523.14 30,979.84" after Medical? Wait: "Prison Department, 324,175.12 324.698.26" then "Medical Departments, 371,781.92 402,761.76 523.14 30,979.84". Actually the OCR lines:

"Prison Department,

324,175.12

324.698.26

Medical Departments,

371,781.92

402,761.76

523.14 30,979.84"

It seems the Increase for Prison (523.14) got placed on the Medical line. And Medical increase: 402,761.76 - 371,781.92 = 30,979.84. So the numbers "523.14 30,979.84" correspond to Prison increase and Medical increase. But the OCR placed them after Medical. We'll need to assign correctly.

Sanitary Department: 548,824.201 -> 620,390.31 = +71,566.11. OCR shows "71,566.11" after Sanitary.

Botanical and Forestry: 71,223.47 -> 77,157.40 = +5,933.93. OCR shows "5,933.93".

Education: No figures given? OCR shows "Education," then "Military Expenditure," then "Public Works Department," then numbers. Let's see: After Botanical, OCR: "Education,

Military Expenditure,

Public Works Department,

728,133,89 3,230,779.89

856,367,04

128,213.15

4,033,500.36

802.720.47

4

821,674.89

901,782.71

Do.

Recurrent,

1,074,646,30

Do. Extraordinary,

8,575,635.19

1,424,532.80 4,716,602.94

80,107.82 349,886,50

1,140,967.75"

This is messy. It seems Public Works Department has sub-items: Recurrent and Extraordinary. The numbers:

• Public Works Department: maybe 728,133.89? and 3,230,779.89? Actually, the original might have two lines: "Public Works Department, Recurrent" and "Public Works Department, Extraordinary". The OCR shows "Public Works Department, 728,133,89 3,230,779.89" then "856,367,04" then "128,213.15" then "4,033,500.36" then "802.720.47" then "4" then "821,674.89" then "901,782.71" then "Do." then "Recurrent, 1,074,646,30" then "Do. Extraordinary, 8,575,635.19" then "1,424,532.80 4,716,602.94" then "80,107.82 349,886,50" then "1,140,967.75".

Let's compute differences for Public Works. The total for Public Works might be split into Recurrent and Extraordinary. The OCR shows "Public Works Department," then two numbers: 728,133.89 and 3,230,779.89? That might be 1922 and 1928 for Recurrent? But then "856,367,04" and "128,213.15" maybe for Extraordinary? Not sure.

Better to look at the total at the end: Total 18,563,002.91 and 21,371,904.72. The sum of all departments should match. We'll need to reconstruct the table as best as possible.

Given the complexity, I think the user expects me to output a cleaned-up Markdown version of the table, with the cause column aligned with each department. Since the OCR separated the cause column, I'll match them by order.

There are 17 cause descriptions for the first 17 departments (up to Botanical and Forestry). The remaining departments (Education, Military Expenditure, Public Works Department, Post Office, Kowloon-Canton Railway, Charge on account of Public Debt, Pensions, Charitable Services) have no cause listed in the OCR. Possibly the cause column continues but OCR didn't capture? Or maybe they have no cause because they are not classified under "Cause of Increase or Decrease"? But the table includes them. The cause column might have entries for them too, but the OCR only captured the first part of the cause column (maybe the column was split across pages). The OCR shows "(07) 53" at the end, likely page numbers.

Given the instructions: "Preserve original content — do not add or remove substantive words. Do not rephrase, rewrite, or change the author's style, tone, or word choice. Only correct what is clearly wrong." I should not invent causes for the later departments. I'll leave the cause column blank for those rows.

Also, the OCR has "Do." which means "ditto". For Fire Brigade cause "More work lone." and Prison "Do." meaning same cause? Actually "Do." likely repeats the previous cause "More work lone." But the cause for Prison is "Transit charges for 1920/1921 recovered during 1923." according to the list. Wait, the cause list: after "More work lone." comes "Do." then "Transit charges..." So "Do." corresponds to Prison? But the cause list has 17 items. Let's index:

Cause list lines (as they appear in OCR):

1. Promotions and stipulated increments.
2. Increased personnel.
3. Lapsing salaries.
4. New stamping maeltine.
5. New posts and typhoon damage.
6. More opium purchased, lower exchange, New Motor Truck,
7. Japanese Earthquake $250,000, British Empire Exhibition $250,000.
8. Increased personnel, more passages,
9. More patients,
10. Increased personnel, Vaccination, New Motor Vehicles.
11. Removal and formation of nursery for Kowloon Development Scheme. Increased personnel and grants.
12. Increase in revenue on which Defence Contribution is payable. Increased personnel, transport and travelling expenses.
13. More work lone.
14. Do.
15. Transit charges for 1920/1921 recovered during 1923.
16. Suspension of through traffic, certain works not proceeded with, Lower exchange.
17. More pensioners,

Now departments 1-17:

1. Governor
2. Cadet Service
3. Colonial Secretary's Department and Legislature
4. Secretariat for Chinese Affairs
5. Audit Department
6. Treasury
7. Harbour Master's Department
8. Imports and Exports Department
9. Royal Observatory
10. Miscellaneous Services
11. Judicial and Legal Departments
12. Police
13. Fire Brigade
14. Prison Department
15. Medical Departments
16. Sanitary Department
17. Botanical and Forestry Department

That's 17 departments. Perfect. So each cause matches each department in order.

Now departments 18-25:

1. Education
2. Military Expenditure
3. Public Works Department
4. Post Office
5. Kowloon-Canton Railway
6. Charge on account of Public Debt
7. Pensions
8. Charitable Services

No causes provided. I'll leave cause blank.

Now, the figures for each department: I need to extract 1922, 1928, Increase, Decrease.

Let's compile data from OCR.

I'll go through the OCR text sequentially after the cause list.

The OCR after "More pensioners," shows "6." then "Governor," then numbers. The "6." might be a page number or item number. I'll ignore.

Now, for each department, the OCR gives:

• Department name
• Expenditure 1922
• Expenditure 1928
• Increase (if any)
• Decrease (if any)

But the OCR sometimes puts Increase/Decrease on same line or next lines.

Let's parse systematically.

I'll write a script in my mind to parse.

Text after "6.":

"Governor,

94.287.62

94,100.92

C.

C.

186.70"

Interpretation:

1922: 94,287.62 (since 94.287.62 likely 94,287.62)

1928: 94,100.92

Increase: C. (maybe blank)

Decrease: C. (maybe blank) but then 186.70? Actually there are two "C." lines then 186.70. Could be that Increase column shows "C." (meaning no increase), Decrease column shows "C." (meaning no decrease?) but then 186.70 is the decrease amount? But there are two C's. Maybe the table has three columns after the expenditures: Increase, Decrease, and something else? Or the "C." stands for "Decrease" in some notation? Actually, "C." might be a misread of "—" (dash). The OCR might have read a dash as "C."? Or maybe it's "Cr." for credit? Not sure.

Given the total decrease column at the bottom is 364,942.46, and total increase 3,373,844.27. The sum of decreases for rows with decrease should match.

Let's compute decreases from obvious rows:

• Governor: decrease 186.70
• Audit: decrease 2,864.20
• Fire Brigade: decrease 11,015.64
• Prison? Actually Prison increase 523.14, so no decrease.
• Medical increase.
• Sanitary increase.
• Botanical increase.
• Kowloon-Canton Railway: 922,855.75 -> 757,030.11 = decrease 165,825.64? OCR shows "185,550,28 165,325,64" maybe increase and decrease? Actually 922,855.75 - 757,030.11 = 165,825.64. OCR shows "185,550,28 165,325,64" - not matching. Might be two numbers: increase and decrease? But it's a decrease overall. The OCR shows two numbers: 185,550.28 and 165,325.64? Not sure.
• Pensions: increase 56,806.72.
• Charitable Services: increase 25,697.38.
• Public Debt: increase 50,769.88.
• Post Office: 309,042.61 -> 123,492.33? That's a huge decrease. But OCR shows "309.042.61 123,492,33". That's a decrease of 185,550.28. That matches the first number in the Railway line? Actually 309,042.61 - 123,492.33 = 185,550.28. And the Railway decrease is 165,825.64. The OCR line "185,550,28 165,325,64" might be the decreases for Post Office and Railway. But they appear after Railway in OCR. Let's see order: After Public Works, we have "Post Office.................. 309.042.61 123,492,33 Kowloon-Canton Railway, 922,855.75 757,030.11 185,550,28 165,325,64". So the two numbers after Railway are likely the decrease for Post Office (185,550.28) and decrease for Railway (165,325.64). But the Railway decrease computed is 165,825.64, close to 165,325.64 (difference 500). Might be rounding.

Then "Charge on account of Public Debt, 914,040.12 964,810,00 50,769.88" -> increase 50,769.88.

"Pensions,... 429,361.04 486,167.76 56,806.72" -> increase 56,806.72.

"Charitable Services, 76,802.56 102,499.94 25,697.38" -> increase 25,697.38.

Now for Public Works Department: The OCR shows multiple numbers. Let's parse that section carefully.

After Botanical and Forestry, we have:

"Education,

Military Expenditure,

Public Works Department,

728,133,89 3,230,779.89

856,367,04

128,213.15

4,033,500.36

802.720.47

4

821,674.89

901,782.71

Do.

Recurrent,

1,074,646,30

Do. Extraordinary,

8,575,635.19

1,424,532.80 4,716,602.94

80,107.82 349,886,50

1,140,967.75"

This is very messy. It seems the Public Works Department has two sub-heads: Recurrent and Extraordinary. The numbers might be:

For Recurrent:

1922: 728,133.89?

1928: 3,230,779.89? That's a huge increase. But then "856,367,04" and "128,213.15" maybe for Extraordinary? Not sure.

Alternatively, the table might have columns: Department, 1922, 1928, Increase, Decrease. For Public Works, it might be split into two rows: "Public Works Department, Recurrent" and "Public Works Department, Extraordinary". The OCR shows "Do." meaning ditto for department name.

Let's try to reconstruct from the total. The total expenditure 1922: 18,563,002.91. 1928: 21,371,904.72.

We can sum the known department figures to see what's missing.

But given the time, I think the best is to produce a Markdown table that reflects the data as clearly as possible, aligning causes with departments, and cleaning up numbers.

I'll create a table with columns: Cause of Increase or Decrease | Head of Expenditure | Expenditure 1922 | Expenditure 1928 | Increase | Decrease

I'll use the cause list for first 17 rows, blank for rest.

For numbers, I'll standardize formatting: use commas for thousands, dot for decimal. Fix obvious OCR errors (e.g., "94.287.62" -> "94,287.62", "28.795.63" -> "28,795.63", "48.196.15" -> "48,196.15", "70,480,33" -> "70,480.33", "316,994,95" -> "316,994.95", "285,009,31" -> "285,009.31", "324.698.26" -> "324,698.26", "548,824.201" -> "548,824.20"? Actually 548,824.201 has three decimals, maybe it's 548,824.20? But the increase 71,566.11 suggests 548,824.20 + 71,566.11 = 620,390.31, matches. So 548,824.20 (two decimals). I'll use 548,824.20.

For Public Works, I'll try to interpret the two sub-heads.

From the OCR: "Public Works Department, 728,133,89 3,230,779.89" then "856,367,04 128,213.15" then "4,033,500.36 802.720.47" then "4" then "821,674.89 901,782.71" then "Do. Recurrent, 1,074,646,30" then "Do. Extraordinary, 8,575,635.19" then "1,424,532.80 4,716,602.94" then "80,107.82 349,886,50" then "1,140,967.75".

This looks like multiple lines of data. Perhaps the original table had columns for "Recurrent" and "Extraordinary" under Public Works, and the OCR read them in a messy order.

Let's search for known Hong Kong historical data: This might be from the Hong Kong Blue Book or Estimates. The Public Works Department often has "Recurrent" and "Extraordinary" expenditure. The figures might be:

Recurrent:

1922: 728,133.89

1928: 1,074,646.30? (from "Do. Recurrent, 1,074,646,30")

Increase: 346,512.41? Not shown.

Extraordinary:

1922: 3,230,779.89?

1928: 8,575,635.19? (from "Do. Extraordinary, 8,575,635.19")

Increase: 5,344,855.30?

But then there are other numbers: 856,367.04, 128,213.15, 4,033,500.36, 802,720.47, 821,674.89, 901,782.71, 1,424,532.80, 4,716,602.94, 80,107.82, 349,886.50, 1,140,967.75.

Maybe the table has more columns like "Increase" and "Decrease" for each sub-head. The OCR might have captured the Increase/Decrease columns for each.

Given the complexity, and the instruction to "reconstruct the table using Markdown table syntax", I should produce a table that matches the original structure as closely as possible. Since the original had a "Cause of Increase or Decrease" column, and then the department, then the figures, I'll create a table with those columns.

For Public Works, I'll include two rows: "Public Works Department (Recurrent)" and "Public Works Department (Extraordinary)" with the figures I can deduce.

But the OCR also shows "Education" and "Military Expenditure" with no figures. Possibly they are zero or not applicable? Or the figures are missing in OCR. I'll leave them blank or with zeros? The total includes them, so they must have figures. The OCR might have omitted them. I'll check if there are numbers for Education and Military later. The OCR after Botanical goes straight to Education, Military, Public Works. No numbers for Education and Military. Then Public Works numbers. Then Post Office, etc. So maybe Education and Military have no expenditure? Unlikely. Could be that the OCR didn't capture their numbers because they were in a different column. I'll look at the total: sum of all department expenditures should equal total. Let's quickly sum the known 1922 figures (in thousands) to see if Education and Military are missing.

Known 1922 figures (from OCR, cleaned):

Governor: 94,287.62

Cadet Service: 280,567.64

Colonial Secretary: 56,454.63

Secretariat Chinese Affairs: 21,115.67

Audit: 51,060.35

Treasury: 68,367.64

Harbour Master: 310,717.36

Imports/Exports: 631,733.97

Royal Observatory: 38,350.10

Miscellaneous: 1,617,305.35

Judicial/Legal: 282,989.15

Police: 1,533,757.51

Fire Brigade: 177,799.07

Prison: 324,175.12

Medical: 371,781.92

Sanitary: 548,824.20

Botanical/Forestry: 71,223.47

Public Works: ?

Post Office: 309,042.61

Kowloon-Canton Railway: 922,855.75

Public Debt: 914,040.12

Pensions: 429,361.04

Charitable: 76,802.56

Sum these (excluding Public Works, Education, Military). Let's approximate:

94,288 + 280,568 = 374,856

+56,455 = 431,311

+21,116 = 452,427

+51,060 = 503,487

+68,368 = 571,855

+310,717 = 882,572

+631,734 = 1,514,306

+38,350 = 1,552,656

+1,617,305 = 3,169,961

+282,989 = 3,452,950

+1,533,758 = 4,986,708

+177,799 = 5,164,507

+324,175 = 5,488,682

+371,782 = 5,860,464

+548,824 = 6,409,288

+71,223 = 6,480,511

+309,043 = 6,789,554

+922,856 = 7,712,410

+914,040 = 8,626,450

+429,361 = 9,055,811

+76,803 = 9,132,614

Total 1922 is 18,563,002.91. So missing about 9,430,389. That's a huge amount, likely Public Works (which includes Recurrent and Extraordinary) plus Education and Military. Public Works Extraordinary alone could be millions. The OCR shows "3,230,779.89" and "8,575,635.19" etc. So Public Works is large.

Thus, Education and Military might be included in Public Works? Or they are separate but figures not captured. I'll include rows for Education and Military with blank figures, noting missing data.

Given the instructions: "Missing text — if a word or phrase is clearly missing due to OCR damage, insert ... in its place." So for missing numbers, I can insert ....

But the user might expect a clean table with all rows. I'll do my best to reconstruct.

Let's parse the Public Works section more carefully.

The OCR lines after "Public Works Department,":

"728,133,89 3,230,779.89" -> likely two numbers: 728,133.89 and 3,230,779.89. Could be 1922 and 1928 for Recurrent? But then "856,367,04" and "128,213.15" -> maybe 1922 and 1928 for Extraordinary? But 856,367.04 and 128,213.15? That would be a decrease. But then "4,033,500.36" and "802.720.47" -> maybe Increase and Decrease? Then "4" -> maybe a page number. Then "821,674.89" and "901,782.71" -> maybe 1922 and 1928 for something else. Then "Do." (ditto) "Recurrent," "1,074,646,30" -> 1,074,646.30. Then "Do. Extraordinary," "8,575,635.19". Then "1,424,532.80 4,716,602.94" -> maybe Increase and Decrease for Recurrent? Then "80,107.82 349,886,50" -> for Extraordinary? Then "1,140,967.75" -> maybe total increase for Public Works?

This is too ambiguous. Perhaps the original table had a sub-table for Public Works with multiple columns. Since the instruction says "if data is clearly in table format, reconstruct the table using Markdown table syntax", I should reconstruct the main table, and for Public Works, maybe include the sub-rows as they appear.

Given the time, I'll produce a Markdown table for the main expenditure statement, with the cause column, department, and the four numeric columns. For Public Works, I'll include two rows: "Public Works Department (Recurrent)" and "Public Works Department (Extraordinary)" with the figures that seem most plausible:

• Recurrent: 1922: 728,133.89; 1928: 1,074,646.30; Increase: 346,512.41; Decrease: -
• Extraordinary: 1922: 3,230,779.89; 1928: 8,575,635.19; Increase: 5,344,855.30; Decrease: -

But the OCR shows "1,424,532.80 4,716,602.94" which might be the increase and decrease for something else. Actually, the total increase for the whole statement is 3,373,844.27. The sum of increases from other departments (excluding Public Works) is:

Cadet: 19,643.51

Colonial Sec: 4,304.45

Chinese Affairs: 7,679.96

Treasury: 2,512.69

Harbour Master: 6,277.59

Imports/Exports: 276,361.55

Royal Obs: 24.01

Miscellaneous: 210,773.03

Judicial: 2,020.16

Police: 100,074.63

Prison: 523.14

Medical: 30,979.84

Sanitary: 71,566.11

Botanical: 5,933.93

Public Debt: 50,769.88

Pensions: 56,806.72

Charitable: 25,697.38

Total increases (excluding Public Works, Education, Military, Post Office, Railway) = let's sum:

19,643.51+4,304.45=23,947.96

+7,679.96=31,627.92

+2,512.69=34,140.61

+6,277.59=40,418.20

+276,361.55=316,779.75

+24.01=316,803.76

+210,773.03=527,576.79

+2,020.16=529,596.95

+100,074.63=629,671.58

+523.14=630,194.72

+30,979.84=661,174.56

+71,566.11=732,740.67

+5,933.93=738,674.60

+50,769.88=789,444.48

+56,806.72=846,251.20

+25,697.38=871,948.58

Total increase overall is 3,373,844.27. So Public Works + Education + Military + Post Office + Railway must contribute the rest: 3,373,844.27 - 871,948.58 = 2,501,895.69.

Post Office shows a decrease (185,550.28), Railway decrease (165,325.64). So they contribute negative. So Public Works, Education, Military must have large increases.

If Public Works Recurrent increase ~346,512 and Extraordinary increase ~5,344,855, that's 5,691,367, too high. So my guess is wrong.

Maybe the figures for Public Works are:

1922: 728,133.89 (Recurrent) + 3,230,779.89 (Extraordinary) = 3,958,913.78

1928: 856,367.04 (Recurrent) + 128,213.15 (Extraordinary) = 984,580.19? That would be a decrease. But then "4,033,500.36" and "802,720.47" might be something else.

Alternatively, the numbers "728,133,89 3,230,779.89" could be 1922 and 1928 for the whole Public Works? Then "856,367,04 128,213.15" could be Increase and Decrease? But 3,230,779.89 - 728,133.89 = 2,502,646.00 increase. That matches the needed increase ~2,501,895.69 closely! So perhaps Public Works total 1922: 728,133.89; 1928: 3,230,779.89; Increase: 2,502,646.00. Then the other numbers are for sub-items? But the OCR shows "856,367,04 128,213.15" next. Could be for Education and Military? But Education and Military are separate heads.

Let's check: If Public Works 1922 = 728,133.89, 1928 = 3,230,779.89, increase = 2,502,646.00. Then the remaining increase needed from Education and Military is 2,501,895.69 - 2,502,646.00 = -750.31, close to zero. So Education and Military might have small net changes.

But the OCR shows "Education, Military Expenditure," with no numbers. Then "Public Works Department," then numbers. So maybe the numbers "728,133,89 3,230,779.89" belong to Education? No, the line says "Public Works Department," then those numbers.

Let's read the OCR sequence:

"Education,

Military Expenditure,

Public Works Department,

728,133,89 3,230,779.89

856,367,04

128,213.15

4,033,500.36

802.720.47

4

821,674.89

901,782.71

Do.

Recurrent,

1,074,646,30

Do. Extraordinary,

8,575,635.19

1,424,532.80 4,716,602.94

80,107.82 349,886,50

1,140,967.75"

It seems the OCR read the table row by row but the columns are misaligned. The original table likely had multiple columns for each department: maybe 1922, 1928, Increase, Decrease. For Public Works, there might be two sub-rows (Recurrent and Extraordinary) each with those four columns. The OCR captured them in a linear fashion.

Given the difficulty, I'll reconstruct the main table as a single table with all departments, using the cause descriptions for the first 17, and for the rest, I'll leave cause blank. For the numeric columns, I'll use the figures that appear in the OCR for each department in order.

I'll go through the OCR sequentially and assign numbers to each department as they appear.

Let's list the OCR text after the cause list, with line breaks as they appear:

"6.

Governor,

94.287.62

94,100.92

C.

C.

186.70

Cadet Service,--

280,567.64

300,211.07

19,643.51

Colonial Secretary's Department and

Legislature,.....

56,454.63

60,759.08

4,304.45

Secretariat for Chinese Affairs,

21,115,67

28.795.63

7,679.96

Audit Department,...

51,060.35

48.196.15

2,864.20

Тгеннигу,

68,367.64

70,480,33

2,512.69

Harbour Master's Department,.

310,717,36

316,994,95

Imports and Exports Department,

631,733.97

908,095.52

Royal Observatory, .....

38,350.10

38,374.11

Miscellaneous Services..........

1,617,305.35

1,828,078.38

Indicial and Legal Departments,

282,989.15

285,009,31

Police,.....

1,533,757.51

1,633,832.14

6.277,59 276,361.55 24.01 210,773.03 2,020.16 100,074.63

Firo Brigade,

177,799,07

166,783.43

11,015.64

Prison Department,

324,175.12

324.698.26

Medical Departments,

371,781.92

402,761.76

523.14 30,979.84

Sanitary Department,

548,824.201

620,390.31

71,566.11

Hotanical and Forestry Department,

71,223.47

77,157.40

5,933.93

Education,

Military Expenditure,

Public Works Department,

728,133,89 3,230,779.89

856,367,04

128,213.15

4,033,500.36

802.720.47

4

821,674.89

901,782.71

Do.

Recurrent,

1,074,646,30

Do. Extraordinary,

8,575,635.19

1,424,532.80 4,716,602.94

80,107.82 349,886,50

1,140,967.75

Post Office..................

309.042.61

123,492,33

Kowloon-Canton Railway,

922,855.75

757,030.11

185,550,28 165,325,64

Charge on account of Public Debt,

914,040.12

964,810,00

50,769.88

Pensions,...

429,361.04

486,167.76

56,806.72

Charitable Services,

76,802.56

102,499.94

25,697.38

Total..............

18,563,002.91

21,371,904.72

3,373,844.27

364,942.46

(07)

53"

Now, I'll parse department by department.

1. Governor:

1922: 94.287.62 -> 94,287.62

1928: 94,100.92

Increase: C. (maybe blank)

Decrease: C. (maybe blank) but then 186.70 appears. Actually there are two "C." lines then 186.70. Could be that the Increase column is "C." (meaning no increase), Decrease column is "C." (meaning no decrease?) but then 186.70 is the decrease amount? But why two C's? Maybe the table has three columns after the expenditures: "Increase", "Decrease", and "Net"? Or the "C." is a misread of "—" for both increase and decrease, and the 186.70 is the net decrease? But the total decrease column at bottom is 364,942.46, which is sum of decreases. So each row should have a decrease value if it decreased. For Governor, decrease = 186.70. So I'll put Decrease = 186.70, Increase = 0 (or blank). The "C." might be artifacts. I'll ignore "C.".

1. Cadet Service:

1922: 280,567.64

1928: 300,211.07

Increase: 19,643.51

Decrease: (blank)

1. Colonial Secretary's Department and Legislature:

1922: 56,454.63

1928: 60,759.08

Increase: 4,304.45

1. Secretariat for Chinese Affairs:

1922: 21,115.67 (21,115,67)

1928: 28,795.63 (28.795.63)

Increase: 7,679.96

1. Audit Department:

1922: 51,060.35

1928: 48,196.15 (48.196.15)

Decrease: 2,864.20 (since 1922 > 1928)

Increase: blank

1. Treasury (Тгеннигу):

1922: 68,367.64

1928: 70,480.33 (70,480,33)

Increase: 2,512.69

1. Harbour Master's Department:

1922: 310,717.36 (310,717,36)

1928: 316,994.95 (316,994,95)

Increase: 6,277.59 (from the garbled line)

1. Imports and Exports Department:

1922: 631,733.97

1928: 908,095.52

Increase: 276,361.55

1. Royal Observatory:

1922: 38,350.10

1928: 38,374.11

Increase: 24.01

1. Miscellaneous Services:

1922: 1,617,305.35

1928: 1,828,078.38

Increase: 210,773.03

1. Judicial and Legal Departments (Indicial):

1922: 282,989.15

1928: 285,009.31 (285,009,31)

Increase: 2,020.16

1. Police:

1922: 1,533,757.51

1928: 1,633,832.14

Increase: 100,074.63

1. Fire Brigade (Firo):

1922: 177,799.07 (177,799,07)

1928: 166,783.43

Decrease: 11,015.64

1. Prison Department:

1922: 324,175.12

1928: 324,698.26 (324.698.26)

Increase: 523.14 (from next line)

1. Medical Departments:

1922: 371,781.92

1928: 402,761.76

Increase: 30,979.84

1. Sanitary Department:

1922: 548,824.20 (548,824.201)

1928: 620,390.31

Increase: 71,566.11

1. Botanical and Forestry Department (Hotanical):

1922: 71,223.47

1928: 77,157.40

Increase: 5,933.93

1. Education: No numbers given. Possibly the next numbers belong to Education? But the next line is "Military Expenditure," then "Public Works Department," then numbers. So Education and Military might have no expenditure? Or the numbers for them are missing. I'll insert ... for their figures.

1. Military Expenditure: same.

1. Public Works Department: This is complex. The OCR shows multiple numbers. I need to decide how to represent. Since the table likely has a single row for Public Works Department (maybe with sub-rows), but the OCR shows "Public Works Department," then a bunch of numbers. Then "Do. Recurrent," and "Do. Extraordinary,". This suggests that the main Public Works Department row might be a total, and then there are two sub-rows for Recurrent and Extraordinary. In the original table, they might be indented. I'll include three rows: "Public Works Department" (total), "Public Works Department (Recurrent)", "Public Works Department (Extraordinary)". But the numbers for the total might be the first pair: 728,133.89 and 3,230,779.89? But that seems low for total Public Works. The Recurrent and Extraordinary figures later are larger: Recurrent 1,074,646.30 and Extraordinary 8,575,635.19. Those might be the 1928 figures? Actually, "Do. Recurrent, 1,074,646,30" likely means 1928 Recurrent = 1,074,646.30. "Do. Extraordinary, 8,575,635.19" likely 1928 Extraordinary = 8,575,635.19. Then the 1922 figures for Recurrent and Extraordinary might be the earlier numbers: 728,133.89 and 3,230,779.89? But 728,133.89 + 3,230,779.89 = 3,958,913.78 for 1922 total. 1,074,646.30 + 8,575,635.19 = 9,650,281.49 for 1928 total. Increase = 5,691,367.71. That would make the total increase much larger than 3,373,844. So that can't be.

Maybe the first pair (728,133.89 and 3,230,779.89) are for Education and Military? But the text says "Public Works Department," before them. Could be that the OCR misordered: The numbers for Education and Military are missing, and the numbers for Public Works are the later ones.

Let's look at the total 1922: 18,563,002.91. Sum of all departments except Public Works, Education, Military = 9,132,614 (from earlier). So Public Works + Education + Military = 9,430,389. If Public Works total 1922 is 3,958,914, then Education + Military = 5,471,475. That's possible. But then 1928 total for Public Works + Education + Military = 21,371,904.72 - (sum of other 1928 figures). Let's compute other 1928 figures:

Other 1928 figures (excluding Public Works, Education, Military):

Governor: 94,100.92

Cadet: 300,211.07

Colonial Sec: 60,759.08

Chinese Affairs: 28,795.63

Audit: 48,196.15

Treasury: 70,480.33

Harbour Master: 316,994.95

Imports/Exports: 908,095.52

Royal Obs: 38,374.11

Miscellaneous: 1,828,078.38

Judicial: 285,009.31

Police: 1,633,832.14

Fire Brigade: 166,783.43

Prison: 324,698.26

Medical: 402,761.76

Sanitary: 620,390.31

Botanical: 77,157.40

Post Office: 123,492.33

Railway: 757,030.11

Public Debt: 964,810.00

Pensions: 486,167.76

Charitable: 102,499.94

Sum these: Let's approximate:

94,101 + 300,211 = 394,312

+60,759 = 455,071

+28,796 = 483,867

+48,196 = 532,063

+70,480 = 602,543

+316,995 = 919,538

+908,096 = 1,827,634

+38,374 = 1,866,008

+1,828,078 = 3,694,086

+285,009 = 3,979,095

+1,633,832 = 5,612,927

+166,783 = 5,779,710

+324,698 = 6,104,408

+402,762 = 6,507,170

+620,390 = 7,127,560

+77,157 = 7,204,717

+123,492 = 7,328,209

+757,030 = 8,085,239

+964,810 = 9,050,049

+486,168 = 9,536,217

+102,500 = 9,638,717

So other 1928 total = 9,638,717. Then Public Works + Education + Military 1928 = 21,371,905 - 9,638,717 = 11,733,188.

If Public Works 1922 = 3,958,914, then Education+Military 1922 = 9,430,389 - 3,958,914 = 5,471,475.

If Public Works 1928 = 9,650,281 (from Recurrent+Extraordinary), then Education+Military 1928 = 11,733,188 - 9,650,281 = 2,082,907. That would be a decrease for Education+Military. Possible.

But the OCR shows "Education, Military Expenditure," with no numbers. Then "Public Works Department," then numbers. It's ambiguous.

Given the instructions to not add or remove substantive content, I should present the data as it appears in the OCR, but structured. Since the OCR has a linear sequence, I'll create a table that follows the sequence of departments as they appear, with the numbers that follow each department name.

From the OCR, after "Botanical and Forestry Department," we have "Education," then "Military Expenditure," then "Public Works Department," then a block of numbers. Then "Post Office," etc. The numbers after Public Works Department likely belong to Public Works Department (maybe multiple sub-heads). The "Do. Recurrent," and "Do. Extraordinary," are likely sub-heads under Public Works.

Thus, I'll treat "Public Works Department" as a heading, and then include sub-rows for "Recurrent" and "Extraordinary" with the numbers that appear near them.

Let's extract numbers for Public Works section:

After "Public Works Department," the numbers:

728,133,89 3,230,779.89

856,367,04

128,213.15

4,033,500.36

802.720.47

4

821,674.89

901,782.71

Do.

Recurrent,

1,074,646,30

Do. Extraordinary,

8,575,635.19

1,424,532.80 4,716,602.94

80,107.82 349,886,50

1,140,967.75

This is too many numbers. Perhaps the table has columns: 1922, 1928, Increase, Decrease for each sub-head. For Recurrent: 1922: 728,133.89? 1928: 1,074,646.30? Increase: 346,512.41? Decrease: 0? For Extraordinary: 1922: 3,230,779.89? 1928: 8,575,635.19? Increase: 5,344,855.30? But then what are the other numbers? 856,367.04, 128,213.15, 4,033,500.36, 802,720.47, 821,674.89, 901,782.71, 1,424,532.80, 4,716,602.94, 80,107.82, 349,886.50, 1,140,967.75.

Maybe the table has more columns like "1922 Recurrent", "1928 Recurrent", "1922 Extraordinary", "1928 Extraordinary", "Increase Recurrent", "Decrease Recurrent", etc. But the main table only has four numeric columns.

Given the complexity, I'll simplify: In the main table, I'll include a row for "Public Works Department" with the total figures that match the overall totals. But I don't have the total figures explicitly. The total row at the bottom gives the grand total. The Public Works total can be derived by subtracting all other departments from the grand total. But that's too much.

Alternatively, I'll replicate the OCR's linear structure in a table with columns: Cause, Head, 1922, 1928, Increase, Decrease, and for Public Works, I'll put the first two numbers as 1922 and 1928, and the next two as Increase and Decrease? But there are many numbers.

Given the time, I'll produce a Markdown table that includes all the rows as they appear in the OCR, with the cause column filled for the first 17 rows, and for the rest, cause blank. For the numeric columns, I'll parse the numbers sequentially for each department.

I'll write a script-like parsing in my response? No, I need to output the final Markdown.

I'll manually create the table rows.

Let's list departments in order of appearance in the OCR after the cause list:

1. Governor
2. Cadet Service
3. Colonial Secretary's Department and Legislature
4. Secretariat for Chinese Affairs
5. Audit Department
6. Treasury
7. Harbour Master's Department
8. Imports and Exports Department
9. Royal Observatory
10. Miscellaneous Services
11. Judicial and Legal Departments
12. Police
13. Fire Brigade
14. Prison Department
15. Medical Departments
16. Sanitary Department
17. Botanical and Forestry Department
18. Education
19. Military Expenditure
20. Public Works Department
21. Public Works Department (Recurrent) [from "Do. Recurrent"]
22. Public Works Department (Extraordinary) [from "Do. Extraordinary"]
23. Post Office
24. Kowloon-Canton Railway
25. Charge on account of Public Debt
26. Pensions
27. Charitable Services
28. Total

Now, for each, I'll assign the numeric values that appear after them in the OCR.

I'll go through the OCR text and assign numbers.

I'll write a parsing algorithm in my mind:

Initialize a list of departments in order. For each department, the OCR provides some numbers. The numbers are separated by newlines or spaces. The pattern: Department name, then 1922, then 1928, then Increase, then Decrease? But not consistent.

From the first few, it's clear: Department, 1922, 1928, Increase (or Decrease). For Governor, there are extra "C." lines. For Cadet Service: Department, 1922, 1928, Increase. For Colonial Secretary: Department, 1922, 1928, Increase. For Secretariat: Department, 1922, 1928, Increase. For Audit: Department, 1922, 1928, Decrease. For Treasury: Department, 1922, 1928, Increase. For Harbour Master: Department, 1922, 1928, (Increase appears later in a batch). For Imports/Exports: Department, 1922, 1928, (Increase in batch). For Royal Observatory: Department, 1922, 1928, (Increase in batch). For Miscellaneous: Department, 1922, 1928, (Increase in batch). For Judicial: Department, 1922, 1928, (Increase in batch). For Police: Department, 1922, 1928, (Increase in batch). Then a batch of six numbers: 6.277,59 276,361.55 24.01 210,773.03 2,020.16 100,074.63. These correspond to Increase for Harbour Master, Imports/Exports, Royal Observatory, Miscellaneous, Judicial, Police. Good.

Then Fire Brigade: Department, 1922, 1928, Decrease (11,015.64).

Prison Department: Department, 1922, 1928, (Increase appears next