1927 — Page 240

The user wants me to proofread OCR output of a historical record from Hong Kong. The text appears to be a table of prison staff (warders) with columns: Office, Name, Date of Appointment, Authority, Annual Salary, House/Quarters/Allowances, Absence from Colony during 1927, Date of First Appointment.

The OCR is messy with broken lines, misaligned columns, garbled text. I need to reconstruct the table properly in Markdown.

First, I need to understand the structure. The header row is:

OFFICE. NAME. Date of Appointment. Authority. Annual Salary. House or Quarters, and Allowances for Rent, Entertainment, Personal, or for any other purpose. Absence from the Colony during 1927. Date of First Appointment.

Then there's a page number "238 (J 154) PRISON,-Continued." Then rows for Warders.

Let me parse each row.

The OCR text lines:

"Warder, Don Do. Zardad. Kader Bux. Emam Deen. 1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance. * 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance. Don Rahimat Khan. Do.. Chanan Singh. 17th January, 1921, 22nd September, 1922. No. 2891 of 1921. No, 2981 of 1922. 420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do., Share Jung. Isher Singh II. Khame Singh. Plumman. 1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese, Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925. Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995. Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925. 27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925. 10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do. Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese). 24th January, 1910. 12th December, 1917. 416 September, 1903. 30th October, 1913. 13th February, 1914. - 12th May, 1916. 24th January, 1916. 15th September, 1917. 7th March, 1910. 21st October, 1919, 5th March, 1910. 28th October, 1919. 13th March, 1920. 25th November, 1914. n 1 month and 12 days. a"

This is a mess. I need to reconstruct each row.

Let me identify each warder entry. The first column is "Warder" for all. Then Name, Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

The OCR seems to have merged multiple rows. Let's separate by names.

Names appear: Don, Zardad, Kader Bux, Emam Deen, Rahimat Khan, Chanan Singh, Share Jung, Isher Singh II, Khame Singh, Plumman, Share Shah, Pala Singh, Ahmed Noor, Yakub, Harzura Singh, Faiz Ali, Ali Asghar.

But the OCR has "Don" repeated? Actually first row: "Warder, Don Do. Zardad. Kader Bux. Emam Deen." That might be four warders: Don, Zardad, Kader Bux, Emam Deen. But "Don" might be a name? Or "Don" could be "Don" as a name? Or maybe "Don" is a typo for "Do." meaning "Ditto"? The header says "Office. Name." So first column is Office (Warder), second is Name. The OCR shows "Warder, Don Do. Zardad. Kader Bux. Emam Deen." Could be that "Don" is a name, then "Do." is ditto for Warder? But then "Zardad" is next name? Actually the pattern: each row starts with "Warder" then name. The OCR might have lost line breaks.

Let's look at the original structure: It's a table with columns. The OCR read line by line but the table spans multiple lines per row.

Better approach: The text includes "Page 240 Page 241" at end. The page number "238 (J 154)" at top.

I need to produce a Markdown table with the data.

Let me try to parse each warder entry sequentially.

I'll list the data fields for each warder:

1. Name: Don? But "Don" might be "Don" as a name? However later there is "Don Rahimat Khan". So "Don" might be a title? Actually "Don" could be "Don" as a name? But "Don" appears alone first. Then "Zardad", "Kader Bux", "Emam Deen". Then later "Don Rahimat Khan". So maybe the first entry is "Don" (name), then "Zardad", "Kader Bux", "Emam Deen". But the OCR shows "Warder, Don Do. Zardad. Kader Bux. Emam Deen." The "Do." might be "Do." meaning ditto for Warder? But the office column is already Warder. In such tables, they often write "Do." for ditto in the office column for subsequent rows. So the first row: Office: Warder, Name: Don. Then next row: Office: Do. (ditto), Name: Zardad. Next: Do., Name: Kader Bux. Next: Do., Name: Emam Deen.

That makes sense. So the first four warders: Don, Zardad, Kader Bux, Emam Deen.

Then after that, we have "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance." This likely corresponds to the first warder (Don). Then next line: "* 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance." This might be for Zardad? But the pattern: each warder has Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

Let's map columns:

Columns:

1. Office
2. Name
3. Date of Appointment
4. Authority
5. Annual Salary
6. House or Quarters, and Allowances...
7. Absence from the Colony during 1927
8. Date of First Appointment

The OCR text after the names seems to list these fields for each warder in order.

But the OCR is jumbled. Let's try to split by warder.

I'll go through the text sequentially and assign fields.

Text after "Emam Deen.":

"1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance."

This looks like: Date of Appointment: 1st July, 1920. Authority: No. 3469 of 1912. Annual Salary: $420? But there is "Do. Do." maybe for Annual Salary? Actually "Do." might be ditto for Authority? Wait, Authority column might have "No. 3469 of 1912" for first, then "Do." for second? But there are two "Do."? Let's see: "No. 3469 of 1912. Do. Do." Could be: Authority: No. 3469 of 1912. Annual Salary: Do. (ditto? but salary is $420). Actually the column "Annual Salary" might be $420. The "Do." might be for something else.

Better to look at the original table layout. The header: "OFFICE. NAME. Date of Appointment. Authority. Annual Salary. House or Quarters, and Allowances... Absence from the Colony during 1927. Date of First Appointment."

So after Name, we have Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

The OCR for first warder (Don) might be:

Date of Appointment: 1st July, 1920

Authority: No. 3469 of 1912

Annual Salary: $420 (but written as "Do. Do."? Actually "Do. Do." might be "Do." for Authority? No, Authority is already given. Maybe "Do." is for Annual Salary? But Annual Salary is a number. Could be that the Annual Salary column is "Do." meaning same as previous? But there is no previous. Hmm.

Let's look at the next warder (Zardad) data: "* 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance."

This seems mixed. The "*" might be a marker. Then "420 Free Rations, $48 Language Allowance..." that looks like Allowances column. Then "3rd September, 1907" could be Date of First Appointment? But Date of First Appointment is last column. "7 months and 26 days" could be Absence. "8th March, 1910" could be Date of Appointment? Then "❤ Do. No. 3075 of 1920" maybe Authority? Then "420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance." That looks like Allowances again.

This is confusing. Perhaps the OCR has merged two rows.

Let's consider that the table might have two pages: Page 240 and 241. The OCR might have read columns incorrectly.

Given the complexity, I might need to reconstruct the table as best as possible, preserving the data as it appears, but aligning into columns.

Since the user wants proofreading of OCR output, they expect me to output the corrected text in Markdown, with tables. I should not guess missing data but represent what's there.

I'll create a Markdown table with the columns as per header. I'll parse each warder row by row based on the sequence of names.

Let's list all names in order as they appear:

1. Don
2. Zardad
3. Kader Bux
4. Emam Deen
5. Rahimat Khan (with "Don" prefix? "Don Rahimat Khan")
6. Chanan Singh
7. Share Jung
8. Isher Singh II
9. Khame Singh
10. Plumman
11. Share Shah
12. Pala Singh
13. Ahmed Noor
14. Yakub
15. Harzura Singh
16. Faiz Ali
17. Ali Asghar

That's 17 warders.

Now, for each, we need to extract the 7 data fields (Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment). The Office is "Warder" for all.

The OCR text after the names contains these fields interspersed. I'll try to segment the text after "Emam Deen." into chunks per warder.

The text after "Emam Deen.":

"1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance. * 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance. Don Rahimat Khan. Do.. Chanan Singh. 17th January, 1921, 22nd September, 1922. No. 2891 of 1921. No, 2981 of 1922. 420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do., Share Jung. Isher Singh II. Khame Singh. Plumman. 1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese, Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925. Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995. Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925. 27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925. 10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do. Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese). 24th January, 1910. 12th December, 1917. 416 September, 1903. 30th October, 1913. 13th February, 1914. - 12th May, 1916. 24th January, 1916. 15th September, 1917. 7th March, 1910. 21st October, 1919, 5th March, 1910. 28th October, 1919. 13th March, 1920. 25th November, 1914. n 1 month and 12 days. a"

This is a mess. It seems the OCR read the table column by column? Or the table spans two pages and the OCR merged them.

Given the difficulty, perhaps the best is to present the data as a table with the information as it appears, but I need to make it readable.

Given the instruction: "Reconstruct Sentences and Paragraphs" but this is a table. The rule: "Tables — if data is clearly in table format, reconstruct the table using Markdown table syntax."

So I must produce a Markdown table.

I'll attempt to parse each warder's data by looking for patterns.

Let's note that the Annual Salary appears often as "420" or "360". The Allowances column contains "Free Rations", "Quarters", "Language Allowance", etc.

The Date of Appointment and Authority appear as dates and "No. xxxx of year".

The Absence column has entries like "7 months and 26 days", "1 month and 12 days", or blank.

The Date of First Appointment is a date.

The OCR seems to have lost the row boundaries. However, the names are listed in the second column. In the OCR, names appear at various points: "Don Rahimat Khan", "Chanan Singh", "Share Jung", "Isher Singh II", "Khame Singh", "Plumman", "Share Shah", "Pala Singh", "Ahmed Noor", "Yakub", "Harzura Singh", "Faiz Ali", "Ali Asghar". Also the first four: "Don", "Zardad", "Kader Bux", "Emam Deen".

But in the OCR text, after "Emam Deen." we have data for first warder? Then "* 420 Free Rations..." might be for second warder? Then "Don Rahimat Khan" appears later. So maybe the first four warders' data are in the first block.

Let's assume the table rows are in order of names as they appear in the Name column. The OCR might have read the Name column first, then the other columns? But the text shows names interspersed.

Another approach: The original document might be a list of warders with details. The OCR output is a linearized version. I need to reconstruct the table by aligning columns.

Given the time, I'll produce a table with the data I can extract, but I must not invent data. I'll include all the information as rows, even if some columns are empty.

I'll create a Markdown table with columns: Office, Name, Date of Appointment, Authority, Annual Salary, Allowances, Absence from Colony during 1927, Date of First Appointment.

I'll go through the text and assign each piece to a column for each warder sequentially.

Let's number the warders in order of appearance in the text (including the first four). The text starts with "Warder, Don Do. Zardad. Kader Bux. Emam Deen." So that's four names. Then later "Don Rahimat Khan", "Chanan Singh", "Share Jung", "Isher Singh II", "Khame Singh", "Plumman", "Share Shah", "Pala Singh", "Ahmed Noor", "Yakub", "Harzura Singh", "Faiz Ali", "Ali Asghar". That's 13 more, total 17.

Now, the data fields appear in the text in some order. Perhaps the OCR read the table row by row, but each row spans multiple lines. The text after the first four names might be the data for those four, then the next names appear with their data.

Let's try to split the text after "Emam Deen." into 4 chunks for the first four warders.

But the text after "Emam Deen." is: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance. * 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance."

This could be two warders' data? The first chunk up to "" might be for Don. The second chunk from "" to before "Don Rahimat Khan" might be for Zardad? But there are four warders.

Maybe the "Do." indicates ditto for the previous column. In the first line: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations..." The "Do. Do." might be for Annual Salary and Allowances? But Annual Salary is a number, not "Do.".

Wait, the header: "Date of Appointment. Authority. Annual Salary. House or Quarters, and Allowances..." So after Authority, we have Annual Salary, then Allowances. The first line: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations..." That would be: Date of Appointment: 1st July, 1920. Authority: No. 3469 of 1912. Annual Salary: Do. (ditto? but no previous). Allowances: Do. (ditto?) then "$420 Free Rations..." That doesn't fit.

Maybe the "Do." is for the Office column? But Office is already Warder.

Let's look at the original image mentally: These are colonial Hong Kong civil service lists. They often have "Do." for ditto in the Office column for subsequent rows. But here the Office column is only shown once at top? Actually the table might have Office column only for the first entry, then "Do." for the rest. But the OCR captured "Warder, Don Do. Zardad. Kader Bux. Emam Deen." That suggests the first row: Office: Warder, Name: Don. Second row: Office: Do., Name: Zardad. Third: Do., Name: Kader Bux. Fourth: Do., Name: Emam Deen.

Then the subsequent columns for each row follow. But the OCR then continues with the data for the first row? Or the data for all rows are listed column by column? The text after the names might be the Date of Appointment column for all four, then Authority column for all four, etc. But the text doesn't look like that.

Given the difficulty, I think the best is to output the raw text cleaned up, but in a table format with the data as it appears in the OCR, but I need to make it a proper table.

Given the user is expecting a proofread version, they might have the original image and want the text corrected. I should try to reconstruct the table as accurately as possible.

I'll attempt to parse each warder's full record by using the names as delimiters.

The text includes names at various points. Let's split the text by names (but names appear also in allowances? No).

List of names in order of appearance in the text (including the first line):

1. Don (from "Warder, Don")
2. Zardad
3. Kader Bux
4. Emam Deen
5. Don Rahimat Khan (appears later)
6. Chanan Singh
7. Share Jung
8. Isher Singh II
9. Khame Singh
10. Plumman
11. Share Shah
12. Pala Singh
13. Ahmed Noor
14. Yakub
15. Harzura Singh
16. Faiz Ali
17. Ali Asghar

Now, the text between these names likely contains the data for the previous name.

So, for each name, the data following it until the next name belongs to that warder.

But the first four names are in a line: "Warder, Don Do. Zardad. Kader Bux. Emam Deen." So the data for Don might start after "Don"? But the text after "Don" is "Do. Zardad." That's not data.

Actually, the first line might be the Office and Name columns for the first four rows. Then the subsequent lines are the other columns for those rows, but the OCR has linearized them.

Let's consider the possibility that the OCR read the table column by column: first the Office column (all "Warder" or "Do."), then Name column, then Date of Appointment column, etc. But the text doesn't show that.

Given the time constraints, I'll produce a table with the data as I can best guess, but I must not guess. The instruction: "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."

So I should correct OCR errors (like "Rutions" -> "Rations", "Cbniese" -> "Chinese", "unge" -> "Allowance", "Lang-" -> "Language", etc.) and fix line breaks, but keep the structure.

Perhaps the best is to present the text as a series of paragraphs for each warder, but the user asked for Markdown table.

I'll create a table with the columns and fill in the data as per the OCR text, aligning each warder's information in a row. I'll use the order of names as they appear in the Name column (which I assume is the order in the text). I'll extract the data for each from the text segments.

Let's manually parse the text segment by segment.

I'll copy the text into a variable and analyze.

Text:

"OFFICE. NAME. Date of Appointment. Authority. Annual Salary. House or Quarters, and Allowances for Rent, Entertainment, Personal, or for any other purpose. Absence from the Colony during 1927. Date of First Appointment. 238 ( J 154 ) PRISON,-Continued. Warder, Don Do. Zardad. Kader Bux. Emam Deen. 1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance. * 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance. Don Rahimat Khan. Do.. Chanan Singh. 17th January, 1921, 22nd September, 1922. No. 2891 of 1921. No, 2981 of 1922. 420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do., Share Jung. Isher Singh II. Khame Singh. Plumman. 1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese, Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925. Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995. Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925. 27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925. 10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do. Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese). 24th January, 1910. 12th December, 1917. 416 September, 1903. 30th October, 1913. 13th February, 1914. - 12th May, 1916. 24th January, 1916. 15th September, 1917. 7th March, 1910. 21st October, 1919, 5th March, 1910. 28th October, 1919. 13th March, 1920. 25th November, 1914. n 1 month and 12 days. a Page 240 Page 241"

This is a single block. I need to split into rows.

Observing that the text includes "Page 240 Page 241" at the end, and "238 (J 154)" at top. The table likely spans pages 238-241.

Given the complexity, I'll assume the table rows are as follows (based on typical Hong Kong Blue Book entries):

Each row: Office, Name, Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

I'll try to match each name with the data that follows until the next name.

But the first four names are given in a line without intervening data. Then data appears. Then "Don Rahimat Khan" appears, then "Chanan Singh", then "Share Jung", "Isher Singh II", "Khame Singh", "Plumman", "Share Shah", "Pala Singh", "Ahmed Noor", "Yakub", "Harzura Singh", "Faiz Ali", "Ali Asghar".

The data between "Emam Deen." and "Don Rahimat Khan" likely belongs to the first four warders. But there are four warders and the data block seems to have multiple entries.

Let's count the data entries in that block:

• "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance."
• "* 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance."

That's two distinct blocks. But there are four warders. Maybe the first block is for Don, second for Zardad, and the other two have data later? But then "Don Rahimat Khan" appears.

Maybe the first block is for Don, and the second block is for Zardad, Kader Bux, Emam Deen combined? But the second block has "3rd September, 1907" which could be Date of First Appointment for Zardad? And "7 months and 26 days" absence, "8th March, 1910" Date of Appointment? Then "Do. No. 3075 of 1920" Authority? Then "420 Free Rations..." Allowances.

Then after that, "Don Rahimat Khan" appears, which is a new name.

Then "Do.. Chanan Singh." So "Do." for Office, Name: Chanan Singh.

Then "17th January, 1921, 22nd September, 1922." Could be Date of Appointment for Rahimat Khan and Chanan Singh? Two dates.

Then "No. 2891 of 1921. No, 2981 of 1922." Authorities.

Then "420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do.," This is messy.

Then "Share Jung. Isher Singh II. Khame Singh. Plumman." Names.

Then "1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese,"

Then "Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925. Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995. Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925. 27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925. 10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do. Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese). 24th January, 1910. 12th December, 1917. 416 September, 1903. 30th October, 1913. 13th February, 1914. - 12th May, 1916. 24th January, 1916. 15th September, 1917. 7th March, 1910. 21st October, 1919, 5th March, 1910. 28th October, 1919. 13th March, 1920. 25th November, 1914. n 1 month and 12 days. a"

This is extremely messy. It seems the OCR has completely garbled the table structure.

Given the instruction to proofread OCR output, perhaps the expected output is the corrected text in the same linear format but with fixed words and line breaks? But the user said "Return the proofread text in standard Markdown: Use #, ##, ### for headers found in the original. Use Markdown table syntax to reconstruct tabular data."

So they want a table.

I think the best I can do is to create a table with the columns and fill each row with the data as it appears in the text for each name, even if some columns are empty or merged.

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

Let's list the names in the order they appear in the text (including the first line). I'll consider the first line as four separate rows.

Row 1: Office: Warder, Name: Don

Row 2: Office: Do., Name: Zardad

Row 3: Office: Do., Name: Kader Bux

Row 4: Office: Do., Name: Emam Deen

Then the text continues with data. Perhaps the data for these four rows are given in the subsequent lines in column order? But the text doesn't have clear separation.

Maybe the original table has the first four rows on page 238, and then the next rows on page 240-241. The OCR might have combined two pages.

Given the difficulty, I'll output a table with all the names and the data I can extract, but I'll note that the OCR is garbled.

But the user might be testing my ability to reconstruct. I need to do my best.

Let's try to parse the text after the first four names as the data for those four, assuming the data is listed in the order of the columns for each row sequentially.

The columns after Name: Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

For Row 1 (Don): The next tokens: "1st July, 1920." -> Date of Appointment. "No. 3469 of 1912." -> Authority. "Do." -> Annual Salary? But "Do." might mean ditto for Annual Salary? But no previous. "Do." -> Allowances? Then "$420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance." That looks like Allowances. So maybe the two "Do." are for Annual Salary and Allowances? But then the Allowances are given explicitly. So perhaps the "Do." are for something else.

Maybe the table has sub-columns? Or the "Do." are for the Office column? But Office already given.

Let's look at the header: "OFFICE. NAME. Date of Appointment. Authority. Annual Salary. House or Quarters, and Allowances... Absence from the Colony during 1927. Date of First Appointment."

There are 8 columns. The first line "Warder, Don Do. Zardad. Kader Bux. Emam Deen." covers Office and Name for four rows.

Then the next line "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance." might be the Date of Appointment, Authority, Annual Salary, Allowances for the first row? But there are two "Do.".

If we assume the first row's Date of Appointment = 1st July 1920, Authority = No. 3469 of 1912, Annual Salary = $420 (but written as "Do."? No), Allowances = $420 Free Rations... But the text says "Do. Do. $420 Free Rations..." So maybe Annual Salary = Do. (meaning same as previous? but no previous), Allowances = Do. (same as previous?), then the actual allowances are listed? That doesn't make sense.

Perhaps the "Do." are for the Absence and Date of First Appointment? But those come later.

Let's check the second block: "* 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance."

This has allowances, then a date (3rd September 1907), then absence (7 months 26 days), then date (8th March 1910), then "Do. No. 3075 of 1920", then allowances again.

This suggests that the data for each row might be interleaved.

Given the time, I'll produce a table with the raw data for each name as a single cell? No.

I think the best is to output the corrected text as a series of paragraphs for each warder, but the user explicitly asked for Markdown table for tabular data.

I'll create a table with the columns and fill in the data as per the OCR text, but I'll keep the text as it appears in the OCR for each column, correcting obvious OCR errors.

I'll assume each warder's data is in the order of the names listed. I'll manually assign the data chunks to each warder.

Let's list the names in order of appearance in the text (including the first line). I'll consider the first line as four names. Then the next names appear later: "Don Rahimat Khan", "Chanan Singh", "Share Jung", "Isher Singh II", "Khame Singh", "Plumman", "Share Shah", "Pala Singh", "Ahmed Noor", "Yakub", "Harzura Singh", "Faiz Ali", "Ali Asghar". That's 13 names. Total 17.

Now, the text after the first four names until "Don Rahimat Khan" contains data for the first four? Let's split that segment into four parts.

Segment: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance. * 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance."

There are two distinct paragraphs separated by "*". Maybe each paragraph corresponds to two warders? But there are four warders.

Perhaps the first paragraph is for Don and Zardad? But it's one paragraph.

Let's look at the original image: The table might have two columns per page? No.

I think I need to accept that I cannot perfectly reconstruct. However, as an AI, I can try to produce a plausible table by aligning the data with the names based on typical patterns.

Given the instruction to only correct obvious OCR errors and preserve content, I might output the text as a cleaned-up version of the OCR output, but in a table format with each row containing the text for that warder as a single string? That would not be a proper table.

The user said: "Use Markdown table syntax to reconstruct tabular data." So they expect a table with columns.

I'll create a table with the 8 columns. I'll fill each row with the data I can extract for each name, leaving blank if not clear.

I'll go through the text and for each name, collect the subsequent text until the next name, and then try to split that into the 6 data fields (Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment). But the text for each name is not clearly separated.

Let's try to split the entire text after the header into lines based on the names.

The names appear as: "Don", "Zardad", "Kader Bux", "Emam Deen", "Don Rahimat Khan", "Chanan Singh", "Share Jung", "Isher Singh II", "Khame Singh", "Plumman", "Share Shah", "Pala Singh", "Ahmed Noor", "Yakub", "Harzura Singh", "Faiz Ali", "Ali Asghar".

But in the text, "Don" appears at the beginning, then "Don Rahimat Khan" later. "Zardad", "Kader Bux", "Emam Deen" appear in the first line. Then "Don Rahimat Khan" appears after the first data block. Then "Chanan Singh" appears after "Do..". Then "Share Jung", "Isher Singh II", "Khame Singh", "Plumman" appear together. Then "Share Shah" appears. Then "Pala Singh" appears. Then "Ahmed Noor" appears. Then "Yakub" appears. Then "Harzura Singh" appears. Then "Faiz Ali" appears. Then "Ali Asghar" appears.

So the names are not evenly spaced.

Given the difficulty, I'll output a table with the header and then rows for each name, with the data fields as they appear in the text in sequence, but I'll have to make judgment calls.

Perhaps the best is to present the data as a list of records in a table, with each record containing the raw text for that warder from the OCR, but that's not a table.

I think I need to do a more systematic parsing. Let's write a pseudo-code in my mind.

I'll copy the text into a string and split by the names (with regex). But the names are not unique markers.

Given the time, I'll produce a Markdown table with the columns and fill in the data as per the OCR text for the first few warders, and for the rest, I'll put the remaining text in a single cell? No.

I'll try to reconstruct the table as it might have been: The Hong Kong Blue Book for 1927? The page 238, 240, 241.

I can search my memory: The Hong Kong Civil Service List often has such tables. The columns are standard.

But I must not add external knowledge.

Given the constraints, I'll output the corrected OCR text in a linear format but with Markdown table for the header and then each row as a row in the table, using the data as it appears in the OCR, but I'll separate the columns by the delimiters that seem to be present: dates, "No.", "$", etc.

I'll attempt to parse the text after the first four names as a list of values for the columns for each of the four warders in order.

Assume the data for the four warders are given in column-major order? That is, all Dates of Appointment for the four, then all Authorities, then all Annual Salaries, then all Allowances, then all Absences, then all Dates of First Appointment. But the text doesn't look like that.

The text: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance." This could be the first warder's full record.

Then "* 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance." This could be the second warder's record.

Then the next names appear: "Don Rahimat Khan. Do.. Chanan Singh." So the third and fourth warders might be Don Rahimat Khan and Chanan Singh? But we already have four warders from the first line. So maybe the first line is not four warders, but the first warder is "Don", and "Do. Zardad. Kader Bux. Emam Deen" are actually the Date of Appointment, Authority, etc? No, "Do." is ditto.

Let's read the first line: "Warder, Don Do. Zardad. Kader Bux. Emam Deen." Could be: Office: Warder, Name: Don. Then the next columns: Date of Appointment: Do. (ditto? but no), Authority: Zardad? That doesn't make sense.

Maybe the OCR misread the lines. The original might have:

Warder | Don | 1st July 1920 | No. 3469 of 1912 | $420 | Free Rations, $84 Language Allowance... | | 3rd September 1907

Do. | Zardad | 8th March 1910 | No. 3075 of 1920 | $420 | Free Rations, $64 Quarters... | 7 months 26 days | 3rd September 1907

Do. | Kader Bux | ... etc.

But the OCR read the first row's columns as a continuous line.

Given the difficulty, I'll produce a table with the data as I can best interpret, but I'll include a note that the OCR is garbled? The instruction says "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 the following: Reconstruct the table as a Markdown table with the header row. Then for each warder, I'll create a row with the information that seems to belong to them, based on the sequence of names and data. I'll correct obvious OCR errors (e.g., "Rutions" -> "Rations", "Cbniese" -> "Chinese", "Lang-" -> "Language", "unge" -> "Allowance", "Ratious" -> "Rations", "Al-" -> "Allowance", "Lowance" -> "Allowance", "Do." -> "Do." (keep as ditto), "Dom" -> "Do."? "Dom" might be "Do." for ditto. "10." might be "Do."? "❤" might be a bullet. "n" and "a" at end might be artifacts.

I'll also fix the file references: "No. 3469 of 1912" -> "No.3469 of 1912"? The rule: "do not add or leave spacing inside parentheses in file references. E.g., XCR (85)72 → XCR(85)72, GR 1178/1922/32 (III) → GR1178/1922/32(III)." But these are "No. 3469 of 1912" which is not a file reference with parentheses. It's a reference number. I'll keep as is.

I'll also fix the page numbers: "Page 240 Page 241" at end.

Now, let's attempt to create rows.

I'll list the warders in the order they appear in the Name column (as per the first line and subsequent names). I'll assume the first line gives four warders: Don, Zardad, Kader Bux, Emam Deen. Then the next warders are: Rahimat Khan, Chanan Singh, Share Jung, Isher Singh II, Khame Singh, Plumman, Share Shah, Pala Singh, Ahmed Noor, Yakub, Harzura Singh, Faiz Ali, Ali Asghar.

But the text includes "Don Rahimat Khan" as a name. So "Don" might be a title? Actually "Don" could be "Don" as a name, but "Don Rahimat Khan" suggests "Don" is part of the name? Or "Don" is a separate warder, and "Rahimat Khan" is another? The text says "Don Rahimat Khan." with a period. Then "Do.. Chanan Singh." So "Don Rahimat Khan" is one name.

Thus the first line "Warder, Don Do. Zardad. Kader Bux. Emam Deen." might be: Warder, Don; Do., Zardad; Do., Kader Bux; Do., Emam Deen. So four warders: Don, Zardad, Kader Bux, Emam Deen.

Then later "Don Rahimat Khan" is a fifth warder.

So total warders: 1. Don, 2. Zardad, 3. Kader Bux, 4. Emam Deen, 5. Don Rahimat Khan, 6. Chanan Singh, 7. Share Jung, 8. Isher Singh II, 9. Khame Singh, 10. Plumman, 11. Share Shah, 12. Pala Singh, 13. Ahmed Noor, 14. Yakub, 15. Harzura Singh, 16. Faiz Ali, 17. Ali Asghar.

Now, I need to assign data to each.

The data appears in the text in chunks. Let's split the text after the first line into chunks separated by the names (5-17). But the names 5-17 appear in the text. So the text between the first line and "Don Rahimat Khan" belongs to the first four warders. The text between "Don Rahimat Khan" and "Chanan Singh" belongs to Don Rahimat Khan. Between "Chanan Singh" and "Share Jung" belongs to Chanan Singh, etc.

But the text has "Do.. Chanan Singh." So "Do.." might be the Office column for Chanan Singh (ditto). So the name is "Chanan Singh". The preceding "Don Rahimat Khan." ends with a period. So the chunk for Don Rahimat Khan is from after "Don Rahimat Khan." to before "Do.. Chanan Singh."? But the text shows "Don Rahimat Khan. Do.. Chanan Singh. 17th January, 1921, 22nd September, 1922. No. 2891 of 1921. No, 2981 of 1922. 420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do., Share Jung." So the data for Don Rahimat Khan and Chanan Singh are interleaved.

This is too messy.

Given the time, I'll output a table with the header and then each row containing the name and the raw text chunk for that warder, but that's not columnar.

I think the user expects me to use my best judgment to reconstruct the table. I'll do that.

I'll create a table with 8 columns. I'll fill the rows as follows, based on typical patterns and the data snippets.

Row 1: Don

• Date of Appointment: 1st July, 1920
• Authority: No. 3469 of 1912
• Annual Salary: $420
• Allowances: Free Rations, $84 Language Allowance (English and Chinese), $180 House Allowance, $21 Personal Allowance
• Absence: (blank)
• Date of First Appointment: 3rd September, 1907? But that appears in the second chunk.

Row 2: Zardad

• Date of Appointment: 8th March, 1910
• Authority: No. 3075 of 1920
• Annual Salary: $420
• Allowances: Free Rations, $64 Quarters, Language Allowance, $60 Old Offenders Allowance
• Absence: 7 months and 26 days
• Date of First Appointment: 3rd September, 1907

Row 3: Kader Bux

• Data not clear. Maybe from later.

Row 4: Emam Deen

• Data not clear.

Then Don Rahimat Khan:

• Date of Appointment: 17th January, 1921
• Authority: No. 2891 of 1921
• Annual Salary: $420
• Allowances: Free Rations, Quarters, $36 Language Allowance? (from "420 Free Rutions. Quarters. $36 Lang-")
• Absence: ?
• Date of First Appointment: ?

Chanan Singh:

• Date of Appointment: 22nd September, 1922
• Authority: No. 2981 of 1922
• Annual Salary: $420
• Allowances: similar
• Absence: ?
• Date of First Appointment: ?

Then Share Jung, Isher Singh II, Khame Singh, Plumman: they appear together. Then data: "1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese,"

This might be for Share Jung, Isher Singh II, Khame Singh, Plumman respectively.

Then Share Shah: "Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925."

Then Pala Singh: "Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995." (1995 is likely 1925)

Then Ahmed Noor: "Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925."

Then Yakub: "27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925."

Then Harzura Singh: "10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do."

Then Faiz Ali: "Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese)."

Then a list of dates at the end: "24th January, 1910. 12th December, 1917. 416 September, 1903. 30th October, 1913. 13th February, 1914. - 12th May, 1916. 24th January, 1916. 15th September, 1917. 7th March, 1910. 21st October, 1919, 5th March, 1910. 28th October, 1919. 13th March, 1920. 25th November, 1914. n 1 month and 12 days. a"

These might be Dates of First Appointment for the warders.

Given the complexity, I'll produce a table with the data I can extract, but I'll keep the text as close to the OCR as possible, correcting only obvious OCR errors.

I'll write the Markdown table with the header and then rows for each warder, filling the columns with the text snippets that seem to belong.

I'll make a best-effort reconstruction.

Let's start.

First, the header row.

Then row for Don:

• Office: Warder
• Name: Don
• Date of Appointment: 1st July, 1920
• Authority: No. 3469 of 1912
• Annual Salary: $420
• Allowances: Free Rations, $84 Language Allowance (English and Chinese), $180 House Allowance, $21 Personal Allowance
• Absence: (blank)
• Date of First Appointment: 3rd September, 1907

But the Absence for Don might be blank. The "7 months and 26 days" appears in the second chunk, which I assigned to Zardad.

Row for Zardad:

• Office: Do.
• Name: Zardad
• Date of Appointment: 8th March, 1910
• Authority: No. 3075 of 1920
• Annual Salary: $420
• Allowances: Free Rations, $64 Quarters, Language Allowance, $60 Old Offenders Allowance
• Absence: 7 months and 26 days
• Date of First Appointment: 3rd September, 1907? But that's same as Don? Actually the second chunk has "3rd September, 1907" before "7 months and 26 days". That might be Date of First Appointment for Zardad. But Don also has 3rd September 1907? That seems unlikely. Maybe the "3rd September, 1907" is the Date of First Appointment for Zardad, and Don's Date of First Appointment is not given? The first chunk doesn't have a Date of First Appointment. The second chunk has "3rd September, 1907" then "7 months and 26 days" then "8th March, 1910". So likely: Date of First Appointment: 3rd September 1907, Absence: 7 months 26 days, Date of Appointment: 8th March 1910. But the column order is Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment. So the Date of Appointment is 8th March 1910, Authority: No. 3075 of 1920, Annual Salary: $420, Allowances: ..., Absence: 7 months 26 days, Date of First Appointment: 3rd September 1907. That matches.

For Don, the first chunk: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance." This has Date of Appointment: 1st July 1920, Authority: No. 3469 of 1912, then "Do. Do." which might be Annual Salary and Allowances? But then it gives allowances. So perhaps Annual Salary: $420 (implied), Allowances: as listed. Absence: not given. Date of First Appointment: not given in this chunk. Maybe it's in the later list of dates.

The later list of dates at the end might be the Dates of First Appointment for all warders. There are many dates. Let's count: 24th January 1910, 12th December 1917, 416 September 1903 (probably 16th September 1903), 30th October 1913, 13th February 1914, 12th May 1916, 24th January 1916, 15th September 1917, 7th March 1910, 21st October 1919, 5th March 1910, 28th October 1919, 13th March 1920, 25th November 1914. That's 14 dates. Plus "1 month and 12 days" might be an absence. We have 17 warders. Not matching.

Given the difficulty, I'll produce a table with the data as it appears in the OCR, but organized by the names in the order they appear, and for each name, I'll put the subsequent text until the next name in the "Allowances" column? No.

I think the only way to satisfy the user is to output a cleaned-up version of the OCR text in a linear format, but with Markdown table for the header and then each line as a row? But the user said "reconstruct tabular data".

I'll output a Markdown table with the columns and then each row as a warder with the data I can parse, leaving blanks where uncertain.

I'll do it now.

I'll create the table in Markdown.

First, the header:

Then rows.

I'll include the page header "238 (J 154) PRISON,-Continued." as a caption or note? The instruction: "Page numbering — if the original scan has page number markers (e.g., "Page 363" appearing at top and bottom), keep them as metadata in the proofread text." So I should include the page number. I'll put it as a header above the table: "Page 238 (J 154) PRISON,-Continued."

Also at the end "Page 240 Page 241". I'll include at the end.

Now, rows.

I'll list the warders as per the names in the first line and subsequent names.

Row 1: Don

Row 2: Zardad

Row 3: Kader Bux

Row 4: Emam Deen

Row 5: Don Rahimat Khan

Row 6: Chanan Singh

Row 7: Share Jung

Row 8: Isher Singh II

Row 9: Khame Singh

Row 10: Plumman

Row 11: Share Shah

Row 12: Pala Singh

Row 13: Ahmed Noor

Row 14: Yakub

Row 15: Harzura Singh

Row 16: Faiz Ali

Row 17: Ali Asghar

Now, I'll fill each row with data from the text.

I'll go through the text and assign.

Text after header: "Warder, Don Do. Zardad. Kader Bux. Emam Deen. 1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance. * 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance. Don Rahimat Khan. Do.. Chanan Singh. 17th January, 1921, 22nd September, 1922. No. 2891 of 1921. No, 2981 of 1922. 420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do., Share Jung. Isher Singh II. Khame Singh. Plumman. 1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese, Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925. Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995. Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925. 27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925. 10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do. Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese). 24th January, 1910. 12th December, 1917. 416 September, 1903. 30th October, 1913. 13th February, 1914. - 12th May, 1916. 24th January, 1916. 15th September, 1917. 7th March, 1910. 21st October, 1919, 5th March, 1910. 28th October, 1919. 13th March, 1920. 25th November, 1914. n 1 month and 12 days. a Page 240 Page 241"

I'll now assign data to each row by splitting at the names.

But the names are not clearly delimited. I'll use the following approach: The first four names are in the first sentence. Then the next name is "Don Rahimat Khan". Then "Chanan Singh". Then "Share Jung", "Isher Singh II", "Khame Singh", "Plumman". Then "Share Shah". Then "Pala Singh". Then "Ahmed Noor". Then "Yakub". Then "Harzura Singh". Then "Faiz Ali". Then "Ali Asghar".

The text between these names belongs to the previous name.

So:

• For Don: text from after "Don" to before "Zardad"? But the first line is "Warder, Don Do. Zardad. Kader Bux. Emam Deen." So the text after "Don" is "Do. Zardad." That's not data. So the data for the first four starts after "Emam Deen.".

Thus the data block after "Emam Deen." until "Don Rahimat Khan" belongs to the first four warders collectively. But we need to split among four.

Given the two paragraphs separated by "*", maybe the first paragraph is for Don and Zardad, second for Kader Bux and Emam Deen? But there are two paragraphs for four warders.

Paragraph 1: "1st July, 1920. No. 3469 of 1912. Do. Do. $420 Free Rations. $84 Language Al- Lowance (English and Chinese), $180 House Allowance, and $21 Personal Allowance."

Paragraph 2: "* 420 Free Rations, $48 Language Al- lowance (English and Chinese), $180 House Allowance and $21 Personal Allowance. 3rd September, 1907. 7 months and 26 days. 8th March, 1910. ❤ Do. No. 3075 of 1920. 420 Free Rations. $64 Quarters. Language Allowance. $60 Old Offenders Allowance."

Paragraph 2 has more data. It includes a date of first appointment (3rd September 1907), absence (7 months 26 days), date of appointment (8th March 1910), authority (No. 3075 of 1920), allowances.

Paragraph 1 has date of appointment (1st July 1920), authority (No. 3469 of 1912), and allowances.

So perhaps Paragraph 1 is for Don, Paragraph 2 is for Zardad. Then Kader Bux and Emam Deen have no data in this block? But then the next names are Don Rahimat Khan and Chanan Singh, which are new warders.

Maybe the first four warders are Don, Zardad, Kader Bux, Emam Deen, but only Don and Zardad are listed on this page? The page is 238, and the table continues on page 240. The OCR might have captured only part.

Given the instruction to preserve original content, I should not omit warders. I'll include all names that appear.

I'll create rows for all 17 names, and for each, I'll put the data that seems to follow them in the text.

Let's do a linear scan:

Initialize list of warders with names in order of appearance in the text (including the first line). But the first line has four names. Then the text continues. The next name encountered is "Don Rahimat Khan". Then "Chanan Singh". Then "Share Jung", "Isher Singh II", "Khame Singh", "Plumman". Then "Share Shah". Then "Pala Singh". Then "Ahmed Noor". Then "Yakub". Then "Harzura Singh". Then "Faiz Ali". Then "Ali Asghar".

So the sequence of names in the text (as they appear) is: Don, Zardad, Kader Bux, Emam Deen, Don Rahimat Khan, Chanan Singh, Share Jung, Isher Singh II, Khame Singh, Plumman, Share Shah, Pala Singh, Ahmed Noor, Yakub, Harzura Singh, Faiz Ali, Ali Asghar.

Now, the text between these names (i.e., after each name until the next name) is the data for that name.

But the first four names are in a single line with "Do." between them. The text after "Emam Deen." is the data for Emam Deen? Or for the group?

Let's treat the first line as: "Warder, Don Do. Zardad. Kader Bux. Emam Deen." This might be a single line in the OCR. The names are separated by periods. So the text after "Emam Deen." is the data for Emam Deen? But then the data for Don, Zardad, Kader Bux would be before? There is no text before.

Thus, perhaps the table is read row by row, and the OCR output each row as a line. But the OCR merged the first four rows into one line because they have "Do." for Office.

If each row is: Office, Name, Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

Then the first row: Warder, Don, 1st July 1920, No. 3469 of 1912, $420, Free Rations..., (absence), (date of first appointment).

Second row: Do., Zardad, 8th March 1910, No. 3075 of 1920, $420, Free Rations..., 7 months 26 days, 3rd September 1907.

Third row: Do., Kader Bux, ... data not in this block? Maybe the data for Kader Bux and Emam Deen are on the next page (240). The OCR then jumps to page 240 with "Don Rahimat Khan" etc.

The text after the second paragraph ends with "Old Offenders Allowance." Then "Don Rahimat Khan." So that suggests the first two rows are done.

Then "Don Rahimat Khan. Do.. Chanan Singh." That might be two rows: Don Rahimat Khan and Chanan Singh.

Then "17th January, 1921, 22nd September, 1922." Two dates of appointment.

"No. 2891 of 1921. No, 2981 of 1922." Two authorities.

"420 Do. 420 Free Rutions. Quarters. $36 Lang- Do., Do., Do.," This is messy.

Then "Share Jung. Isher Singh II. Khame Singh. Plumman." Four names.

Then "1st January, 1923. No. 2891 of 1923. 420 Free Ratious. Quarters. $72 Lang- Do. Do. 420 Free Rations. $84 Language Al- lowance, $21 Personal Allowance. Do. 19th November, 1923. 28th June, 1924. No. 2890 of 1923. 420 Free Rations. Quarters. $84 Lang- uage Allowance. 420 Quarters. Free Rations $72 Lang- uage Allowance, (English and Cbniese,"

Then "Share Shah. 21st September, No. 3140 of 1925. 420 Free Rations. Quarters. $81 Lang- Do., uage Allowance. 1925."

Then "Dom Pala Singh. (1) 1st May, Do. 420 Free Rations. Quarters, $24 Lang- uage Allowance, 1995."

Then "Ahmed Noor 1. 3rd July, Do. 420 Free Rations. Quarters. $84 Lang- unge Allowance. Do., 1925."

Then "27th July, Do. 420 Free Rations. Quarters. $36 Lang- Do., Yakub. unge Allowance. 1925."

Then "10., Harzura Singh. 16th September, Do. 420 Free Rations. Quarters. $72. Lang- │unge Allowance, 1925, Do. Do. 420 Do."

Then "Faiz Ali. Do., Dom Ali Asghar. 20th April, 1927. No. 3821 of 1927. 360 Free Rations: Quarters. $84 Lang- uage Allowance. uage Allowance $30 Personal Allowance. uage Allowance, (English and Chinese)."

Then a list of dates.

This suggests that the table continues with many warders.

Given the complexity, I'll produce a table with the data as it appears in the text for each name, but I'll have to make the columns align.

I'll write a Markdown table with the columns and for each name, I'll put the relevant text snippets in the appropriate columns based on the order of columns.

I'll do my best to parse each warder's data from the text chunks.

Let's define a function in my mind: For each warder, the data appears in the order: Date of Appointment, Authority, Annual Salary, Allowances, Absence, Date of First Appointment.

I'll go through the text after the first line and extract values sequentially for each warder.

But the text is not separated per warder.

Given the time, I'll output a table with the header and then each row containing the name and the raw text chunk that follows it until the next name, but that's not columnar.

I think the user might be okay with a table that has the columns and the data filled as per the OCR, even if not perfect.

I'll create the table now.

I'll start with the page header.

Then the table.

I'll include 17 rows.

I'll fill the columns as follows:

For Don:

• Office: Warder
• Name: Don
• Date of Appointment: 1st July, 1920
• Authority: No. 3469 of 1912
• Annual Salary: $420
• Allowances: Free Rations, $84 Language Allowance (English and Chinese), $180 House Allowance, $21 Personal Allowance
• Absence:
• Date of First Appointment:

For Zardad:

• Office: Do.
• Name: Zardad
• Date of Appointment: 8th March, 1910
• Authority: No. 3075 of 1920
• Annual Salary: $420
• Allowances: Free Rations, $64 Quarters, Language Allowance, $60 Old Offenders Allowance
• Absence: 7 months and 26 days
• Date of First Appointment: 3rd September, 1907

For Kader Bux:

• Office: Do.
• Name: Kader Bux
• Date of Appointment:
• Authority:
• Annual Salary:
• Allowances:
• Absence:
• Date of First Appointment:

For Emam Deen:

• Office: Do.
• Name: Emam Deen
• Date of Appointment:
• Authority:
• Annual Salary:
• Allowances:
• Absence:
• Date of First Appointment:

For Don Rahimat Khan:

• Office: Do.
• Name: Don Rahimat Khan
• Date of Appointment: 17th January, 1921
• Authority: No. 2891 of 1921
• Annual Salary: $420
• Allowances: Free Rations, Quarters, $36 Language Allowance
• Absence:
• Date of First Appointment:

For Chanan Singh:

• Office: Do.
• Name: Chanan Singh
• Date of Appointment: 22nd September, 1922
• Authority: No. 2981 of 1922
• Annual Salary: $420
• Allowances: Free Rations, Quarters, $36 Language Allowance (from "Do., Do., Do."?)
• Absence:
• Date of First Appointment:

For Share Jung:

• Office: Do.
• Name: Share Jung
• Date of Appointment: 1st January, 1923
• Authority