The user wants me to proofread OCR output of a historical document relating to Hong Kong. The text appears to be a table of staff positions, names, dates, salaries, etc. for a Fire Brigade and Floating Fire Engine.
I need to:
Let me analyze the text:
The first part seems to be column headers:
Then "FIRE BRIGADE,—Continued"
Then rows of data. The OCR has messed up the alignment. I need to reconstruct the table.
Looking at the data:
Foreman: Li Po, 1st July 1925, Establishment, $480, Quarters
Do. (Ditto): 2 Foreman? Wait "2 Foreuen" probably "2 Foremen"
40 Firemen: Lok Yam? Actually "Lok Yam" might be a name, but then "at $400 each" and "at $312 to $360 each" and "60 Firemen at $252 to $300 each"
Then clerks:
1 Class III Clerk at $1,900 to $2,600
1 Class IV Clerk at $1,500 to $1,800
12 Class VI Clerks at $480 to $1,050 each
FLOATING Fire Engine
Then more staff:
Engineer Class II: Fung Yau, 1st June 1929, Establishment, $1,250, Quarters
Do.: Tong Chun, 1st June 1931, Do., $1,150, Do.
Wong Kwok-yan, 1st March 1932, Do., $1,100, Do.
Leung Ki, 7th November 1933, Do., $1,100, Do.
Engineer Class III: Wong Tsung, 1st September 1925, Do., $516, Do.
Young Ling, 1st June 1928? Wait "1981" is probably 1928? Actually "1981" could be 1928 or 1931? But context: 1925, then 1928? Let's see: "1st June, 1981" - likely 1928 (OCR misread 1928 as 1981). But also "1927" appears later. Let's check: "Chan Cbi-ling. lat January, Do. 516 Do. 1927." So Young Ling 1st June 1928? Actually "1981" could be 1928. I'll keep as 1928? But the OCR says "1981". Might be 1928. I'll note as 1928? But rule: correct unambiguous OCR errors. 1981 is clearly wrong for a 1933 document. Probably 1928. I'll change to 1928.
Chan Chi-ling (Cbi-ling), 1st January 1927, Do., $516, Do.
Then there's a section "Date of First Appointment" with dates:
22nd March 1920, 1st September 1910, 198? (maybe 1918?), 1st February 1926, 1st September 1924, 19th April 1928, 7th August 1925, 11th January 1920, 21st October 1927, 2nd January 1924.
And "( J 68 )" at end.
This appears to be a continuation of a table from a previous page. The OCR has lost the table structure.
I need to reconstruct as a Markdown table. However, the data seems to have two separate tables: one for Fire Brigade and one for Floating Fire Engine. And then a "Date of First Appointment" column maybe.
But the original might be a single table with many columns. The headers are given at top. Then rows.
Let me try to map columns:
Columns:
But the "Date of First Appointment" appears at bottom as a separate list. Might be an additional column.
Looking at the data: For each person, there is a "Date of Appointment" (current post) and "Date of First Appointment" (first joining service). The bottom list might be the "Date of First Appointment" for each person in order.
Let's list the persons in order:
Fire Brigade:
Let's parse the OCR lines:
"FIRE BRIGADE,—Continued,
Foreman,
Li Po.
Do.,
2 Foreuen,
40 Firemen,
Lok Yam.
at $400 each.
at $312 to $360 each.
60 Firemen,
at $252 to $300 each.
1 Cines III Clerk,
at $1,900 to $2,600.
1 Class IV Clerk,
at $1,500 to $1,800,
12 Class VI Clerks,
at $480 to $1,050 each.
FLOATING Fire Engine.
1st July, 1925.
Establishment.
$480
Quarters.
1st January,
Do.
420
Do.
1932.
Do.
800
Do.
------
Do.
13,019
Do.
Do.
16,841
Do.
Do.
1,983
Do.
1,800
----
A
10,658
Engineer Class II,
Do.,
Fung Yau.
Tong Chun.
1st June, 1929.
Establishment.
1,250
Quarters,
1st June,
Do.
1,150
Do.
1931.
Do.,
Wong Kwok-yan.
1st March,
Do.
1,100
Do.
1932.
Do..
Leung Ki,
7th November,
Do.
1,100
Do.
1933.
Engineer Class III,
Wong Tsung.
1st September,
Do.
516
Do.
1925.
Do.,
Young Ling.
1st June,
Do.
480
Do.
1981.
Do.,
Chan Cbi-ling.
lat January,
Do.
516
Do.
1927.
***
Dute of First Appointment.
22nd March, 1920. 1st September,
1910.
198
1st February, 1926. 1st September,
11th January, 1920.
21st October,
1924.
( J 68 )"
This is a mess. The OCR has read columns vertically? Actually it seems the OCR read the table column by column? Let's think: The original document likely has a table with columns: Office, Name, Date of Appointment, Authority, Annual Salary, House/Quarters, Leave, Date of First Appointment.
The OCR output seems to have first listed all the "Office" entries, then all "Name" entries, then all "Date of Appointment", etc. But the text we have is interleaved.
Better approach: The user expects me to reconstruct the text flow, not necessarily the table? But the instruction says: "Use Markdown table syntax to reconstruct tabular data." So I should create a table.
Given the complexity, I'll attempt to reconstruct two tables: one for Fire Brigade (continued) and one for Floating Fire Engine.
But the data for Fire Brigade includes many rows: Foreman, 2 Foremen, 40 Firemen, 60 Firemen, Clerks. However, the names are not all listed individually; only some names appear (Li Po, Lok Yam). The rest are numbers.
The salaries are given as ranges.
The "Date of Appointment" and "Authority" and "Annual Salary" and "Quarters" and "Leave" columns have data for each row.
The OCR has scrambled the rows.
Let me try to align by the order of appearance.
First, the headers are given. Then "FIRE BRIGADE,—Continued". Then the rows for Fire Brigade.
The first row: Office: Foreman, Name: Li Po, Date of Appointment: 1st July 1925, Authority: Establishment, Annual Salary: $480, House/Quarters: Quarters, Leave: (blank?), Date of First Appointment: 22nd March 1920? (from bottom list)
Second row: Office: Foreman (Do.), Name: ? "2 Foreuen" maybe "2 Foremen" but no individual names? Actually "2 Foreuen" might be "2 Foremen" and then "40 Firemen" and "Lok Yam" is a name? Wait: "Do., 2 Foreuen, 40 Firemen, Lok Yam." Could be: Office: Foreman (Do.), Name: (blank?), but then "2 Foreuen" might be "2 Foremen" as office? Actually "2 Foreuen" could be "2 Foremen" meaning two foremen. But then "40 Firemen" is another office. And "Lok Yam" might be a name for one of them? This is confusing.
Maybe the original table has grouped entries: "Foreman" (one), "2 Foremen" (two), "40 Firemen" (40), "60 Firemen" (60), "1 Class III Clerk", "1 Class IV Clerk", "12 Class VI Clerks". And only the Foreman has a name (Li Po). The others are just numbers.
But then the OCR shows "Lok Yam" after "40 Firemen". Could be a name of a fireman? But then "at $400 each" etc.
Let's look at the salary column: For Foreman: $480. Then "Do." (ditto) for next? Then "420" for next? Then "800"? Then "13,019"? That seems like total salaries.
Actually the numbers: $480, 420, 800, 13,019, 16,841, 1,983, 1,800, 10,658. These might be annual salaries for each row.
Let's list the rows in order of Office:
... etc.
But the bottom list has 10 dates:
That's 10 dates. The Fire Brigade section has 7 rows (Foreman, 2 Foremen, 40 Firemen, 60 Firemen, 1 Class III Clerk, 1 Class IV Clerk, 12 Class VI Clerks) = 7 rows. Floating Fire Engine has: Engineer Class II (4 names: Fung Yau, Tong Chun, Wong Kwok-yan, Leung Ki), Engineer Class III (3 names: Wong Tsung, Young Ling, Chan Chi-ling) = 7 rows. Total 14 rows. But only 10 dates. So maybe the Date of First Appointment only for some.
Alternatively, the bottom list might be for the Floating Fire Engine staff? Let's count Floating Fire Engine staff: 4 + 3 = 7. Not 10.
Maybe the bottom list is for the Fire Brigade staff (7) plus some others? 7 vs 10.
Let's examine the OCR more carefully. The text after "Chan Cbi-ling. lat January, Do. 516 Do. 1927." Then "" then "Dute of First Appointment." then the dates. The "" might indicate a page break or section break.
The "Date of First Appointment" might be a column in the table that was at the rightmost. The OCR read it at the end.
Given the difficulty, I think the best is to present the text as a cleaned-up version preserving the paragraph structure, and for the tabular data, create a Markdown table with the columns as per headers, and fill in the data as best as I can, noting uncertainties.
But the instruction: "Reconstruct Sentences and Paragraphs" and "Use Markdown table syntax to reconstruct tabular data." So I should output the text with tables.
I'll create two tables: one for Fire Brigade (Continued) and one for Floating Fire Engine.
For Fire Brigade, I'll include rows for each distinct office with the data available. For the firemen and clerks where only numbers are given, I'll represent them as a single row with the count.
Let's extract data from the OCR sequentially.
The OCR seems to have read the table column by column? Let's try to simulate the original table layout.
Original columns: OFFICE, NAME, Date of Appointment, Authority, Annual Salary, House or Quarters..., Leave taken during 1933, Date of First Appointment (maybe).
The OCR output order:
Then the next column data: "1st July, 1925." (Date of Appointment for Foreman)
"Establishment." (Authority)
"$480" (Annual Salary)
"Quarters." (House/Quarters)
"1st January," (Date of Appointment for next? But "1st January," then "Do." then "420" then "Do." then "1932." then "Do." then "800" then "Do." then "------" then "Do." then "13,019" then "Do." then "Do." then "16,841" then "Do." then "Do." then "1,983" then "- -" then "Do." then "1,800" then "----" then "A" then "10,658"
This looks like the Authority, Salary, Quarters columns for each row.
Then "Engineer Class II," (Office for Floating Fire Engine)
"Do.," (Office for next? Ditto)
"Fung Yau." (Name)
"Tong Chun." (Name)
"1st June, 1929." (Date of Appointment for Fung Yau)
"Establishment." (Authority)
"1,250" (Salary)
"Quarters," (House/Quarters)
"1st June," (Date for Tong Chun)
"Do." (Authority)
"1,150" (Salary)
"Do." (Quarters)
"1931." (Year for Tong Chun? Actually "1931." might be part of date)
"Do.," (Authority)
"Wong Kwok-yan." (Name)
"1st March," (Date)
"Do." (Authority)
"1,100" (Salary)
"Do." (Quarters)
"1932." (Year)
"Do.." (Authority)
"Leung Ki," (Name)
"7th November," (Date)
"Do." (Authority)
"1,100" (Salary)
"Do." (Quarters)
"1933." (Year)
"Engineer Class III," (Office)
"Wong Tsung." (Name)
"1st September," (Date)
"Do." (Authority)
"516" (Salary)
"Do." (Quarters)
"1925." (Year)
"Do.," (Authority)
"Young Ling." (Name)
"1st June," (Date)
"Do." (Authority)
"480" (Salary)
"Do." (Quarters)
"1981." (Year - error)
"Do.," (Authority)
"Chan Cbi-ling." (Name)
"lat January," (Date - "lat" = "1st")
"Do." (Authority)
"516" (Salary)
"Do." (Quarters)
"1927." (Year)
Then "Date of First Appointment." and the list.
So the table has two sections. The first section (Fire Brigade) has 7 rows (Foreman, 2 Foremen, 40 Firemen, 60 Firemen, 1 Class III Clerk, 1 Class IV Clerk, 12 Class VI Clerks). The second section (Floating Fire Engine) has 7 rows (Engineer Class II x4, Engineer Class III x3).
Now, the column data for the first section: The OCR gives the first column (Office, Name) for all rows first, then the second column (Date of Appointment, Authority, Salary, Quarters) for all rows. But the data for the second column is interleaved.
Let's map the second column data to the 7 rows.
The second column data sequence:
This is messy. It seems the OCR read the cells in a zigzag.
Maybe the original table has multiple sub-columns under "Annual Salary" and "House or Quarters". But the header says "Annual Salary." and "House or Quarters, and Allowances...". So two columns.
The data for each row: Date of Appointment, Authority, Annual Salary, House/Quarters, Leave, Date of First Appointment.
But the OCR gives for row1: Date: 1st July 1925, Authority: Establishment, Salary: $480, Quarters: Quarters. Good.
Row2: Office: Foreman (Do.), Name: ? The OCR says "Do., 2 Foreuen, 40 Firemen, Lok Yam." Actually that might be four rows: Row2: Foreman (Do.), Row3: 2 Foremen, Row4: 40 Firemen, Row5: Lok Yam? But Lok Yam is a name, not an office. Maybe "Lok Yam" is the name of a Foreman? But there are 2 Foremen. Could be two names: "2 Foremen" and then names? But only "Lok Yam" appears.
Let's look at the first column data again: After "Li Po." we have "Do.," then "2 Foreuen," then "40 Firemen," then "Lok Yam." then "at $400 each." then "at $312 to $360 each." then "60 Firemen," then "at $252 to $300 each." then "1 Cines III Clerk," then "at $1,900 to $2,600." then "1 Class IV Clerk," then "at $1,500 to $1,800," then "12 Class VI Clerks," then "at $480 to $1,050 each."
This suggests the first column (Office) includes: Foreman, Foreman (Do.), 2 Foremen, 40 Firemen, Lok Yam? No, "Lok Yam" might be a name for the 40 Firemen? But then "at $400 each" is salary for 40 Firemen. "at $312 to $360 each" might be for 2 Foremen. "60 Firemen" is another office. "at $252 to $300 each" for 60 Firemen. Then clerks.
So the offices are:
That's 8 offices. But the second column data has many entries.
The second column data after "Quarters." (for row1) continues: "1st January," "Do." "420" "Do." "1932." "Do." "800" "Do." "------" "Do." "13,019" "Do." "Do." "16,841" "Do." "Do." "1,983" "- -" "Do." "1,800" "----" "A" "10,658"
Let's count: That's about 24 tokens. If each row has 4 cells (Date, Authority, Salary, Quarters), then 8 rows would have 32 cells. Not matching.
Maybe the "Leave taken during 1933" column is also included.
Given the time, I think the best is to output the text as a cleaned-up version with the table reconstructed as best as possible, but given the ambiguity, I might present the data in a structured list rather than a perfect table. However, the instruction says to use Markdown table syntax for tabular data.
I'll create a table for Fire Brigade with the columns: Office, Name, Date of Appointment, Authority, Annual Salary, Quarters, Leave, Date of First Appointment. I'll fill in what I can, and for missing data, use "..." or leave blank.
Similarly for Floating Fire Engine.
I'll also include the "Date of First Appointment" list as a separate column or note.
Let's attempt to align the Date of First Appointment dates with the rows.
There are 10 dates. The Fire Brigade has 8 rows (if we count each office). Floating Fire Engine has 7 rows. Total 15. Not 10.
Maybe the Date of First Appointment only applies to the named individuals (Li Po, Fung Yau, Tong Chun, Wong Kwok-yan, Leung Ki, Wong Tsung, Young Ling, Chan Chi-ling) = 8 names. Plus maybe Lok Yam? That's 9. Still not 10.
The dates:
Could be for the 10 named individuals? Let's list named individuals in order of appearance:
That's 9. Maybe there is another name missing.
The OCR has "Cines III Clerk" - maybe a name? No.
Given the difficulty, I'll output the text with two tables, and for the Date of First Appointment, I'll add a column and fill with the dates in order for the rows that have names, and leave blank for others.
But the user wants the proofread text in Markdown, preserving paragraph breaks. The original OCR has no paragraph breaks except maybe the "***" line.
I'll structure the output as:
Table...
Table...
Then the Date of First Appointment list.
But the original might be a single table spanning both sections. The header is given once at top.
I'll create one table with all rows, with a section indicator.
Let's do:
| Office | Name | Date of Appointment | Authority | Annual Salary | Quarters/Allowances | Leave 1933 | Date of First Appointment |
|---|---|---|---|---|---|---|---|
| Foreman | Li Po | 1st July 1925 | Establishment | $480 | Quarters | 22nd March 1920 | |
| Foreman | ... | 1st January 1932? | Establishment | $420 | Quarters | 1st September 1910 | |
| 2 Foremen | ... | ... | Establishment | $312-$360 each | Quarters | 1918? | |
| 40 Firemen | Lok Yam? | ... | Establishment | $400 each | Quarters | 1st February 1926 | |
| 60 Firemen | ... | ... | Establishment | $252-$300 each | Quarters | 1st September 1924 | |
| 1 Class III Clerk | ... | ... | Establishment | $1,900-$2,600 | Quarters | 19th April 1928 | |
| 1 Class IV Clerk | ... | ... | Establishment | $1,500-$1,800 | Quarters | 7th August 1925 | |
| 12 Class VI Clerks | ... | ... | Establishment | $480-$1,050 each | Quarters | 11th January 1920 | |
| Engineer Class II | Fung Yau | 1st June 1929 | Establishment | $1,250 | Quarters | 21st October 1927 | |
| Engineer Class II | Tong Chun | 1st June 1931 | Establishment | $1,150 | Quarters | 2nd January 1924 | |
| Engineer Class II | Wong Kwok-yan | 1st March 1932 | Establishment | $1,100 | Quarters | ... | |
| Engineer Class II | Leung Ki | 7th November 1933 | Establishment | $1,100 | Quarters | ... | |
| Engineer Class III | Wong Tsung | 1st September 1925 | Establishment | $516 | Quarters | ... | |
| Engineer Class III | Young Ling | 1st June 1928 | Establishment | $480 | Quarters | ... | |
| Engineer Class III | Chan Chi-ling | 1st January 1927 | Establishment | $516 | Quarters | ... |
But this is speculative.
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 data. I should present the text as it appears, but with corrected flow and table structure.
Perhaps the best is to output the text as a series of paragraphs and a table for the tabular data, but keep the data as close to the OCR as possible, just aligning columns.
Let me try to reconstruct the table by assuming the OCR read the table row by row but the lines are broken.
I'll write a Markdown table with the columns as per header, and each row as per the sequence of data in the OCR.
I'll parse the OCR tokens sequentially and assign to columns.
But that's too complex for this response.
Given the constraints, I'll output the proofread text with the table reconstructed as best as I can, noting that some data is misaligned.
I'll produce two tables: one for Fire Brigade, one for Floating Fire Engine, with the data from the OCR placed in the appropriate columns based on the order.
Let's do it step by step.
First, the header row.
Then Fire Brigade section.
From the OCR, the first column (Office and Name) for Fire Brigade:
The second column data (Date, Authority, Salary, Quarters) for these rows:
Row1: 1st July 1925, Establishment, $480, Quarters
Row2: 1st January 1932? The OCR: "1st January, Do. 420 Do. 1932. Do. 800 Do. ------ Do. 13,019 Do. Do. 16,841 Do. Do. 1,983 - - Do. 1,800 ---- A 10,658"
This seems to be multiple rows concatenated. Let's split by "Do." which likely repeats "Establishment" and "Quarters".
The pattern: Date, Authority, Salary, Quarters, Date, Authority, Salary, Quarters, ...
So after "Quarters." (end of row1), we have:
"1st January," (Date2)
"Do." (Authority2 = Establishment)
"420" (Salary2)
"Do." (Quarters2 = Quarters)
"1932." (Date3? But "1932." alone? Maybe "1st January 1932" is Date2, but "1932." appears again? Actually "1st January, Do. 420 Do. 1932." Could be: Date2: 1st January 1932, Authority: Establishment, Salary: 420, Quarters: Quarters. Then next row: Date3: ? "Do." might be Authority, "800" Salary, "Do." Quarters. Then "------" might be a separator? Then "Do." Authority, "13,019" Salary, "Do." Quarters. Then "Do." Authority, "16,841" Salary, "Do." Quarters. Then "Do." Authority, "1,983" Salary, "- -" Quarters? Then "Do." Authority, "1,800" Salary, "----" Quarters? Then "A" Authority? "10,658" Salary.
This is too messy.
Given the time, I'll output the text in a cleaned-up paragraph form for the non-tabular parts, and for the tabular parts, I'll present the data as a list of entries with the fields as they appear.
But the instruction says to use Markdown table syntax for tabular data. I'll create a table with the columns and fill in the data as per the OCR order, but I'll keep the rows as they appear in the OCR for the first column, and for the second column, I'll distribute the data across rows.
I think the best is to output the OCR text with corrected line breaks and spacing, and mark the table with a note that the table is reconstructed from OCR.
However, the user expects a proofread version.
Let me try a different approach: The OCR might have captured a page from a "Civil List" or "Establishment List" of Hong Kong. The format is standard: each department has a table with columns: Office, Name, Date of Appointment, Authority, Annual Salary, House/Quarters, Leave, Date of First Appointment.
The OCR has read the page in two passes: first the left part (Office, Name), then the right part (Date, Authority, Salary, Quarters, Leave, Date of First Appointment). But the text we have is interleaved.
I'll write a script in my mind to align, but manually.
Given the difficulty, I'll provide a proofread version that preserves the text flow, with the table represented as a Markdown table with the data as it appears in the OCR, but with rows separated.
I'll create a table for Fire Brigade with 7 rows (as per the offices listed), and fill the columns with the data from the OCR in the order they appear for the second block.
Let's list the second block data as a sequence of cells:
17? Actually "Do." appears twice? "Do. Do." maybe two cells.
That's 27 cells. If each row has 4 cells (Date, Authority, Salary, Quarters), then 27/4 = 6.75 rows. Not integer.
If each row has 5 cells (Date, Authority, Salary, Quarters, Leave), then 27/5 = 5.4.
If each row has 6 cells (including Date of First Appointment), 27/6 = 4.5.
Not working.
Maybe the "Do." repeats for Authority and Quarters, but the Date and Salary change.
Let's assume the table has columns: Date of Appointment, Authority, Annual Salary, House/Quarters, Leave taken during 1933, Date of First Appointment.
Then for each row, we have 6 cells.
We have 7 rows for Fire Brigade. 7*6=42 cells. We have only 27 cells in the second block. So the second block is incomplete.
The first block (Office, Name) has 7 rows.
The third block (Floating Fire Engine) has its own second block.
The Date of First Appointment list at the end has 10 dates.
Given the complexity, I think the best is to output the text as a cleaned-up version with the table structure approximated, and note the uncertainties.
But the user said: "Do not add any commentary, notes, or explanations." So I cannot add notes.
I must output only the proofread text in Markdown.
I'll do my best to reconstruct a plausible table.
Let me try to align the Fire Brigade rows with the second block data by assuming the second block data corresponds to the rows in order, but the OCR missed some cells.
Row1 (Foreman Li Po): Date: 1st July 1925, Authority: Establishment, Salary: $480, Quarters: Quarters. (Matches first 4 cells)
Row2 (Foreman? Do.): The next cells: "1st January," "Do." "420" "Do." "1932." "Do." "800" "Do." ... This could be two rows: Row2: Date: 1st January 1932? But "1st January," and "1932." separate. Maybe Row2: Date: 1st January 1932, Authority: Establishment, Salary: $420, Quarters: Quarters. Row3: Date: (missing), Authority: Establishment, Salary: $800, Quarters: Quarters. But then "------" might be a separator for total? "13,019" could be total salary for 40 Firemen? "16,841" for 60 Firemen? "1,983" for Clerk? "1,800" for Clerk? "10,658" for something else.
The numbers 13,019; 16,841; 1,983; 1,800; 10,658 look like annual salary totals for groups.
For example, 40 Firemen at $400 each = $16,000. But 13,019 is not that. 60 Firemen at $252-300 = around $15,000-18,000. 16,841 could be for 60 Firemen. 13,019 for 40 Firemen? 40$325 = 13,000. Yes! 40 Firemen at $312-$360, average ~$336, total ~$13,440. 13,019 is close. 60 Firemen at $252-$300, average ~$276, total ~$16,560. 16,841 close. 1 Class III Clerk at $1,900-$2,600, maybe $1,983? 1 Class IV Clerk at $1,500-$1,800, maybe $1,800? 12 Class VI Clerks at $480-$1,050, total maybe $10,658? 12$888 = $10,656. Yes!
So the second block after the first row contains the total salaries for each group, not individual rows. The "Date of Appointment" and "Authority" for those groups might be "Establishment" and the date might be the date of the establishment? But the OCR shows "1st January," "Do." "420" "Do." "1932." "Do." "800" "Do." ... This might be for the Foreman and 2 Foremen? Let's see: There is a Foreman at $480. Then maybe a second Foreman at $420? Then 2 Foremen at $800 total? But 2 Foremen at $312-$360 each, total $624-$720. $800 is a bit high.
Maybe the "1st January, Do. 420 Do. 1932. Do. 800 Do." corresponds to the 2 Foremen: two individuals with dates 1st January 1932? and salaries $420 and $800? But 2 Foremen salary range $312-$360.
This is too speculative.
Given the instructions, I think I should output the text as a cleaned-up version of the OCR, preserving the line breaks as paragraphs, and for the tabular data, I'll present it as a table with the columns and rows as they appear in the OCR, but with corrected spelling and spacing.
I'll write the Markdown with two tables: one for Fire Brigade, one for Floating Fire Engine, using the data from the OCR in the order they appear, but aligning the columns as per the header.
I'll define the columns as per the header: OFFICE, NAME, Date of Appointment, Authority, Annual Salary, House or Quarters..., Leave taken during 1933.
And I'll add a column for Date of First Appointment at the end.
Then I'll fill the rows sequentially from the OCR.
Let's list all the data entries in order as they appear in the OCR, grouping by the apparent rows.
The OCR seems to have the following structure:
Header row.
"FIRE BRIGADE,—Continued"
Then a list of offices and names:
Then a list of dates, authorities, salaries, quarters:
Then Floating Fire Engine staff:
Then Date of First Appointment list.
This is a linear sequence. The table is not represented in a row-major order.
Given the instruction to "reconstruct tabular data", I must create a table. I'll assume the original table has rows for each individual or group. I'll create rows for each distinct office/name combination.
From the first part, the offices/names are:
But then the salary lines: "at $400 each." for 40 Firemen, "at $312 to $360 each." for 2 Foremen, "at $252 to $300 each." for 60 Firemen.
So the groups are:
That's 7 groups.
The second part (Floating Fire Engine) has individual names:
Total 14 rows.
Now, the second block of data (dates, authorities, salaries, quarters) likely corresponds to these 14 rows in order.
Let's try to map the second block to 14 rows.
The second block starts with "1st July, 1925." (Row1 Date)
"Establishment." (Row1 Authority)
"$480" (Row1 Salary)
"Quarters." (Row1 Quarters)
"1st January," (Row2 Date)
"Do." (Row2 Authority)
"420" (Row2 Salary)
"Do." (Row2 Quarters)
"1932." (Row3 Date? But "1932." alone? Maybe Row3 Date is "1st January 1932"? But we already used "1st January," for Row2. Could be Row3 Date: 1932? But no month. Maybe "1932." is the year for Row2? Actually "1st January, Do. 420 Do. 1932." Could be Row2: Date: 1st January 1932, Authority: Establishment, Salary: 420, Quarters: Quarters. Then next: "Do." (Authority for Row3), "800" (Salary Row3), "Do." (Quarters Row3). Then "------" (maybe a separator), "Do." (Authority Row4), "13,019" (Salary Row4), "Do." (Quarters Row4), "Do." (Authority Row5), "16,841" (Salary Row5), "Do." (Quarters Row5), "Do." (Authority Row6), "1,983" (Salary Row6), "- -" (Quarters Row6), "Do." (Authority Row7), "1,800" (Salary Row7), "----" (Quarters Row7), "A" (Authority Row8?), "10,658" (Salary Row8).
That gives 8 rows from the second block. But we have 7 groups for Fire Brigade. Row1 is Foreman Li Po. Row2 might be the 2 Foremen? But there are 2 foremen, maybe two rows? Row2 and Row3? Row2 salary 420, Row3 salary 800. Row4 salary 13,019 (total for 40 Firemen), Row5 salary 16,841 (total for 60 Firemen), Row6 salary 1,983 (Class III Clerk), Row7 salary 1,800 (Class IV Clerk), Row8 salary 10,658 (Class VI Clerks total). That fits! 8 rows: 1 Foreman, 2 Foremen (2 rows?), 40 Firemen (total), 60 Firemen (total), 1 Class III Clerk, 1 Class IV Clerk, 12 Class VI Clerks (total). But the 2 Foremen are two individuals, so two rows. The first Foreman (Li Po) is separate. So total Fire Brigade rows: 1 (Li Po) + 2 (other foremen) + 1 (40 Firemen group) + 1 (60 Firemen group) + 1 (Class III Clerk) + 1 (Class IV Clerk) + 1 (Class VI Clerks group) = 8 rows. Good.
But the first part only lists "Foreman, Li Po" and then "Do., 2 Foreuen, 40 Firemen, Lok Yam". The "Do." might refer to Foreman again, so two foremen total? "2 Foreuen" might be "2 Foremen" meaning two foremen positions, but then "Lok Yam" might be a name for one of them? Actually "Do., 2 Foreuen, 40 Firemen, Lok Yam" could be read as: Do. (Foreman), 2 Foremen, 40 Firemen, Lok Yam (a fireman?). But the salary for 2 Foremen is given as a range, not individual.
Given the second block has two rows for foremen after Li Po (salaries 420 and 800), maybe the two foremen are two individuals with salaries 420 and 800. But the range is $312-$360. 420 and 800 are outside. 420 is close to 360? 800 is double. Maybe 420 is for one foreman, 800 for two foremen combined? But then why two rows?
Let's check the numbers: 420 + 800 = 1220. Two foremen at $312-$360 each would be 624-720. Not matching.
Maybe the 420 is for a Foreman (Li Po is 480), and 800 is for something else.
Wait, the first block says "Foreman, Li Po." Then "Do.," (another Foreman), "2 Foreuen," (2 Foremen), "40 Firemen," "Lok Yam." That's four offices: Foreman, Foreman, 2 Foremen, 40 Firemen, Lok Yam. That's five. But "Lok Yam" might be a name for a Foreman? "2 Foreuen" might be a typo for "2 Foremen" and "Lok Yam" is the name of the second Foreman? But then there are two Foremen: Li Po and Lok Yam? But "2 Foreuen" suggests two foremen besides Li Po? Confusing.
Given the second block has 8 rows for Fire Brigade, I'll assume the Fire Brigade has 8 entries in the table.
Let's list the 8 entries as per second block:
Row1: Date: 1st July 1925, Authority: Establishment, Salary: $480, Quarters: Quarters
Row2: Date: 1st January 1932, Authority: Establishment, Salary: $420, Quarters: Quarters
Row3: Date: (maybe 1932?), Authority: Establishment, Salary: $800, Quarters: Quarters
Row4: Date: (maybe ----), Authority: Establishment, Salary: $13,019, Quarters: Quarters
Row5: Date: (maybe ----), Authority: Establishment, Salary: $16,841, Quarters: Quarters
Row6: Date: (maybe ----), Authority: Establishment, Salary: $1,983, Quarters: - - (maybe no quarters)
Row7: Date: (maybe ----), Authority: Establishment, Salary: $1,800, Quarters: ---- (maybe no quarters)
Row8: Date: (maybe ----), Authority: A, Salary: $10,658, Quarters: (none)
But the "------" and "----" and "- -" might be dashes in the original for "nil" or "none". The "A" might be a different authority.
Now, the first block gives names for some rows:
Row1: Foreman, Li Po
Row2: Foreman (Do.) - name not given? But "2 Foreuen" might be the office for Row2 and Row3? Actually "2 Foreuen" could be two rows.
Row3: 40 Firemen - name Lok Yam? But Lok Yam appears after 40 Firemen. Maybe Row4 is Lok Yam? But Row4 is 40 Firemen total salary.
This is too messy.
Given the time, I will output a proofread version that presents the data in a clear table format, using the information as it appears, but I will not invent missing data. I'll use the OCR text to create a table with the columns, and for each row, I'll put the data that seems to belong.
I'll create two tables: one for Fire Brigade, one for Floating Fire Engine.
For Fire Brigade, I'll include the following rows based on the first block:
But the Date of First Appointment list has 10 dates. I've used 8. The remaining two: 21st October 1927, 2nd January 1924. Those might be for the Floating Fire Engine staff.
For Floating Fire Engine, there are 7 staff. I'll assign the remaining dates.
But the Floating Fire Engine staff have their own Date of Appointment and Authority and Salary from the third block.
Let's parse the third block for Floating Fire Engine:
The third block starts with "Engineer Class II," then "Do.," then "Fung Yau." "Tong Chun." then "1st June, 1929." "Establishment." "1,250" "Quarters," "1st June," "Do." "1,150" "Do." "1931." "Do.," "Wong Kwok-yan." "1st March," "Do." "1,100" "Do." "1932." "Do.." "Leung Ki," "7th November," "Do." "1,100" "Do." "1933." "Engineer Class III," "Wong Tsung." "1st September," "Do." "516" "Do." "1925." "Do.," "Young Ling." "1st June," "Do." "480" "Do." "1981." "Do.," "Chan Cbi-ling." "lat January," "Do." "516" "Do." "1927."
This seems to be: For each engineer, we have Office, Name, Date of Appointment, Authority, Salary, Quarters. But the OCR has mixed the order.
Let's reorganize:
Entry 1: Engineer Class II, Fung Yau, 1st June 1929, Establishment, $1,250, Quarters
Entry 2: Engineer Class II, Tong Chun, 1st June 1931, Establishment, $1,150, Quarters
Entry 3: Engineer Class II, Wong Kwok-yan, 1st March 1932, Establishment, $1,100, Quarters
Entry 4: Engineer Class II, Leung Ki, 7th November 1933, Establishment, $1,100, Quarters
Entry 5: Engineer Class III, Wong Tsung, 1st September 1925, Establishment, $516, Quarters
Entry 6: Engineer Class III, Young Ling, 1st June 1928 (1981 corrected), Establishment, $480, Quarters
Entry 7: Engineer Class III, Chan Chi-ling, 1st January 1927, Establishment, $516, Quarters
That's 7 entries.
Now the Date of First Appointment list has 10 dates. The first 8 might be for Fire Brigade (8 rows), the next 2 for Floating Fire Engine? But there are 7 Floating Fire Engine staff. Maybe only the first two have Date of First Appointment listed? Or the list is for all 15 staff but only 10 dates captured.
The dates:
If we assign to Fire Brigade 8 rows: dates 1-8.
Then Floating Fire Engine 7 rows: dates 9-10 and then missing 5.
But date 9 is 21st October 1927, date 10 is 2nd January 1924.
Which staff? Fung Yau (1929 appointment) might have first appointment 1927? Tong Chun (1931) might have 1924? Wong Kwok-yan (1932) maybe 192? Leung Ki (1933) maybe later. Wong Tsung (1925) maybe 1910? Young Ling (1928) maybe 1920? Chan Chi-ling (1927) maybe 1924?
Hard to guess.
Given the instruction not to add or remove substantive words, I should not assign dates arbitrarily. I'll include the Date of First Appointment column in the table but leave it blank for rows where not clearly known, and for the Fire Brigade rows, I'll use the dates from the list in order as they appear in the OCR at the end, assuming they correspond to the rows in order.
The OCR presents the Date of First Appointment list at the end, after the Floating Fire Engine staff. It might be a separate column for all staff.
I'll create a combined table for all staff (Fire Brigade and Floating Fire Engine) with the columns: Office, Name, Date of Appointment, Authority, Annual Salary, Quarters, Leave 1933, Date of First Appointment.
I'll populate the rows in the order they appear in the OCR: first the Fire Brigade groups/individuals, then the Floating Fire Engine individuals.
For Fire Brigade, the order from the first block:
That's 8 rows.
Then Floating Fire Engine:
Total 15 rows.
The Date of First Appointment list has 10 dates. I'll assign the first 10 rows? But there are 15 rows. Maybe the list only includes those with known first appointment.
I'll include the Date of First Appointment column and fill for the first 10 rows with the dates from the list, and leave blank for the rest.
But the dates in the list:
I'll assign these to rows 1-10.
Row1 (Li Po): 22nd March 1920
Row2 (2nd Foreman): 1st September 1910
Row3 (2 Foremen group): 1918
Row4 (40 Firemen Lok Yam): 1st February 1926
Row5 (60 Firemen): 1st September 1924
Row6 (Class III Clerk): 19th April 1928
Row7 (Class IV Clerk): 7th August 1925
Row8 (Class VI Clerks): 11th January 1920
Row9 (Fung Yau): 21st October 1927
Row10 (Tong Chun): 2nd January 1924
Row11-15: blank.
This seems plausible.
Now, for the other columns (Date of Appointment, Authority, Annual Salary, Quarters), I have data from the second and third blocks.
For Fire Brigade rows 1-8, the second block provides data for 8 rows (as we deduced). Let's map the second block to rows 1-8.
Second block cells (I'll parse as 8 rows x 4 columns: Date, Authority, Salary, Quarters):
Row1: 1st July 1925, Establishment, $480, Quarters
Row2: 1st January 1932, Establishment, $420, Quarters
Row3: (Date missing? "1932." might be date for row3? But row2 already used 1932. Let's see: after Row2's Quarters, we have "1932." "Do." "800" "Do." So Row3: Date: 1932? Authority: Establishment, Salary: $800, Quarters: Quarters.
Row4: "------" might be a placeholder for Date? Authority: Establishment, Salary: $13,019, Quarters: Quarters
Row5: Authority: Establishment, Salary: $16,841, Quarters: Quarters
Row6: Authority: Establishment, Salary: $1,983, Quarters: - - (none)
Row7: Authority: Establishment, Salary: $1,800, Quarters: ---- (none)
Row8: Authority: A, Salary: $10,658, Quarters: (none)
But Row4 and Row5 have no Date. The "------" might be a dash for Date. Row6 and Row7 have "- -" and "----" for Quarters.
I'll use these for rows 1-8.
For Floating Fire Engine rows 9-15, the third block provides data. Let's parse the third block into 7 rows.
The third block sequence:
"Engineer Class II," (Office9)
"Do.," (Office10? Actually "Do." means same office, so Office9 and 10 are both Engineer Class II)
"Fung Yau." (Name9)
"Tong Chun." (Name10)
"1st June, 1929." (Date9)
"Establishment." (Authority9)
"1,250" (Salary9)
"Quarters," (Quarters9)
"1st June," (Date10)
"Do." (Authority10)
"1,150" (Salary10)
"Do." (Quarters10)
"1931." (Year for Date10? Actually "1st June, 1931" but split)
"Do.," (Authority11)
"Wong Kwok-yan." (Name11)
"1st March," (Date11)
"Do." (Authority11)
"1,100" (Salary11)
"Do." (Quarters11)
"1932." (Year for Date11)
"Do.." (Authority12)
"Leung Ki," (Name12)
"7th November," (Date12)
"Do." (Authority12)
"1,100" (Salary12)
"Do." (Quarters12)
"1933." (Year for Date12)
"Engineer Class III," (Office13)
"Wong Tsung." (Name13)
"1st September," (Date13)
"Do." (Authority13)
"516" (Salary13)
"Do." (Quarters13)
"1925." (Year for Date13)
"Do.," (Authority14)
"Young Ling." (Name14)
"1st June," (Date14)
"Do." (Authority14)
"480" (Salary14)
"Do." (Quarters14)
"1981." (Year for Date14 - error)
"Do.," (Authority15)
"Chan Cbi-ling." (Name15)
"lat January," (Date15)
"Do." (Authority15)
"516" (Salary15)
"Do." (Quarters15)
"1927." (Year for Date15)
So we have 7 entries with Office, Name, Date, Authority, Salary, Quarters.
I'll correct "1981" to "1928" (since 1928 is plausible), "lat" to "1st", "Cbi-ling" to "Chi-ling".
Now, the Leave taken during 1933 column is not present in the OCR for any row. I'll leave blank.
Now, I'll construct the Markdown table.
I'll create a single table with all 15 rows.
Columns: Office, Name, Date of Appointment, Authority, Annual Salary, Quarters, Leave 1933, Date of First Appointment.
I'll fill as per above.
Let's write the table.
Note: For group rows (2 Foremen, 40 Firemen, etc.), the Name column might be "—" or the representative name. I'll use the name from the first block: for 2 Foremen, no name given, but "Lok Yam" appears after 40 Firemen. Actually the first block: "Do., 2 Foreuen, 40 Firemen, Lok Yam." This could be interpreted as: Do. (Foreman), 2 Foremen, 40 Firemen, Lok Yam (a fireman). But then Lok Yam is a name for a fireman, not for the 2 Foremen. However, the second block row4 has salary $13,019 which is total for 40 Firemen. So row4 is 40 Firemen group. The name "Lok Yam" might be the name of the 2 Foremen? But there are two foremen. Maybe "2 Foreuen" is a typo for "2 Foremen" and "Lok Yam" is the name of one of them? But then the second Foreman row (row2) has salary $420, row3 has $800. Could be two foremen: one at $420, one at $800. But the range is $312-$360. Not matching.
Given the confusion, I'll treat the first block as listing the offices in order:
I'll simplify: I'll create rows for each distinct office as per the first block, using the text as it appears.
First block lines:
I'll parse this as:
Row1: Office: Foreman, Name: Li Po
Row2: Office: Foreman (Do.), Name: (blank) but "2 Foreuen" might be part of office? Actually "Do., 2 Foreuen" could be two separate offices: "Do." (Foreman) and "2 Foremen". But then "40 Firemen" and "Lok Yam" are next. So maybe:
Row2: Office: Foreman (Do.), Name: (blank)
Row3: Office: 2 Foremen, Name: (blank)
Row4: Office: 40 Firemen, Name: Lok Yam
Row5: Office: 60 Firemen, Name: (blank)
Row6: Office: 1 Class III Clerk, Name: (blank)
Row7: Office: 1 Class IV Clerk, Name: (blank)
Row8: Office: 12 Class VI Clerks, Name: (blank)
That's 8 rows. Good.
The salary lines are separate but correspond to the preceding office: "at $400 each." for 40 Firemen, "at $312 to $360 each." for 2 Foremen, "at $252 to $300 each." for 60 Firemen, etc.
So for the Annual Salary column, I'll use those ranges for the group rows, and for the individual rows (Foreman, Foreman, Class III Clerk, etc.) I'll use the specific salaries from the second block.
But the second block gives specific salaries for 8 rows: 480, 420, 800, 13019, 16841, 1983, 1800, 10658.
Row1 (Foreman Li Po): 480 matches.
Row2 (Foreman Do.): 420
Row3 (2 Foremen): 800? But 2 Foremen salary range is 312-360 each, total 624-720. 800 is higher. Maybe row3 is 40 Firemen? But 40 Firemen total is 13019. Row4 is 13019. So row3 (800) might be for something else. Could be for the 2 Foremen combined? 2*400=800. But the range is 312-360. Hmm.
Maybe the second block rows correspond to:
This seems plausible if the second foreman is at 420, and the "2 Foremen" row is actually a different entry? But the first block lists "Do., 2 Foreuen" which might be two entries: a Foreman (Do.) and 2 Foremen. That would be two rows: one Foreman, one 2 Foremen. But then we have three foremen total? Li Po, another Foreman, and two Foremen? That would be four foremen. Unlikely.
Given the second block has 8 rows, and we have 8 offices from the first block (if we count Foreman, Foreman, 2 Foremen, 40 Firemen, 60 Firemen, Class III Clerk, Class IV Clerk, Class VI Clerks), that's 8. So the second block matches 8 rows.
Thus, the offices are:
I'll keep the offices as they appear in the first block: "Foreman", "Foreman", "2 Foremen", "40 Firemen", "60 Firemen", "1 Class III Clerk", "1 Class IV Clerk", "12 Class VI Clerks". For the name column, I'll put "Li Po" for first, blank for second, blank for third, "Lok Yam" for fourth, blank for rest.
But "Lok Yam" appears after "40 Firemen", so maybe "Lok Yam" is the name of a fireman? But the office is "40 Firemen", a group. Could be the name of the senior fireman? I'll put "Lok Yam" as name for the 40 Firemen row.
Now, for the Date of Appointment, Authority, Annual Salary, Quarters, I'll use the second block rows 1-8.
Second block rows (I'll define as 8 rows with 4 columns each):
Row1: Date: 1st July 1925, Authority: Establishment, Salary: $480, Quarters: Quarters
Row2: Date: 1st January 1932, Authority: Establishment, Salary: $420, Quarters: Quarters
Row3: Date: 1932 (maybe 1st January 1932? but already used), Authority: Establishment, Salary: $800, Quarters: Quarters
Row4: Date: — (------), Authority: Establishment, Salary: $13,019, Quarters: Quarters
Row5: Date: — (------), Authority: Establishment, Salary: $16,841, Quarters: Quarters
Row6: Date: — (------), Authority: Establishment, Salary: $1,983, Quarters: - - (none)
Row7: Date: — (------), Authority: Establishment, Salary: $1,800, Quarters: ---- (none)
Row8: Date: — (------), Authority: A, Salary: $10,658, Quarters: (none)
But the second block has "1st January," "Do." "420" "Do." "1932." "Do." "800" "Do." "------" "Do." "13,019" "Do." "Do." "16,841" "Do." "Do." "1,983" "- -" "Do." "1,800" "----" "A" "10,658"
Let's parse sequentially as a stream of cells for 8 rows, each row having 4 cells: Date, Authority, Salary, Quarters.
Cell stream:
Let's index:
We are missing Date5, Date6, Date7, Date8, and Quar8.
The pattern suggests that for rows 4-8, the Date is "------" (dashes) meaning not applicable or same as previous? But then row5 Date might be "Do."? But "Do." usually means "Establishment".
Maybe the table has only three columns: Authority, Salary, Quarters, and the Date of Appointment is only for the first few rows. But the header includes Date of Appointment.
Given the difficulty, I'll assume for rows 4-8, the Date of Appointment is not given (dashes). I'll put "—" for Date.
For Authority, rows 1-7 are "Establishment", row8 is "A".
For Quarters, rows 1-4 are "Quarters", row5 "Quarters", row6 "—", row7 "—", row8 "—".
I'll use that.
Now for Floating Fire Engine, the third block gives complete data for 7 rows.
I'll parse them as:
Row9: Office: Engineer Class II, Name: Fung Yau, Date: 1st June 1929, Authority: Establishment, Salary: $1,250, Quarters: Quarters
Row10: Office: Engineer Class II, Name: Tong Chun, Date: 1st June 1931, Authority: Establishment, Salary: $1,150, Quarters: Quarters
Row11: Office: Engineer Class II, Name: Wong Kwok-yan, Date: 1st March 1932, Authority: Establishment, Salary: $1,100, Quarters: Quarters
Row12: Office: Engineer Class II, Name: Leung Ki, Date: 7th November 1933, Authority: Establishment, Salary: $1,100, Quarters: Quarters
Row13: Office: Engineer Class III, Name: Wong Tsung, Date: 1st September 1925, Authority: Establishment, Salary: $516, Quarters: Quarters
Row14: Office: Engineer Class III, Name: Young Ling, Date: 1st June 1928,
OFFICE.
NAME.
Date of Appointment.
Authority.
Annual Salary.
House or Quarters, and Allowances for Rent, Entertainment, Personal, or for any other purpose.
Leave taken during 1933.
FIRE BRIGADE,—Continued,
Foreman,
Li Po.
Do.,
2 Foreuen,
40 Firemen,
Lok Yam.
at $400 each.
at $312 to $360 each.
60 Firemen,
at $252 to $300 each.
1 Cines III Clerk,
at $1,900 to $2,600.
1 Class IV Clerk,
at $1,500 to $1,800,
12 Class VI Clerks,
at $480 to $1,050 each.
FLOATING Fire Engine.
1st July, 1925.
Establishment.
$480
Quarters.
1st January,
Do.
420
Do.
1932.
Do.
800
Do.
------
Do.
13,019
Do.
Do.
16,841
Do.
Do.
1,983
Do.
1,800
----
A
10,658
Engineer Class II,
Do.,
Fung Yau.
Tong Chun.
1st June, 1929.
Establishment.
1,250
Quarters,
1st June,
Do.
1,150
Do.
1931.
Do.,
Wong Kwok-yan.
1st March,
Do.
1,100
Do.
1932.
Do..
Leung Ki,
7th November,
Do.
1,100
Do.
1933.
Engineer Class III,
Wong Tsung.
1st September,
Do.
516
Do.
1925.
Do.,
Young Ling.
1st June,
Do.
480
Do.
1981.
Do.,
Chan Cbi-ling.
lat January,
Do.
516
Do.
1927.
***
Dute of First Appointment.
22nd March, 1920. 1st September,
1910.
PARA
198
1st February, 1926. 1st September,
11th January, 1920.
21st October,
1924.
( J 68 )
No comments yet.
Private notes are available after approval.