Dating the Undated: Newspaper Radio Logs as Primary Evidence for a Quarter-Million Broadcast Recordings

Abstract

Many old American radio recordings survive today. Collectors trade them as sound files. But a large number of these recordings carry no date. The file name is often the only information about a recording. File names do not last. People write them again at every copy. Computers cut them short. Someone chooses a date when a name is hard to read. So the date is lost, and many of the dates that remain are wrong.

This paper describes a method for finding the true date of such a recording. The main evidence is the newspaper radio log — the daily list of radio programs that American newspapers printed. The method compares logs from several separate newspapers. It also uses trade magazines and evidence inside the recording itself.

The paper reports the results of this method at The Archive of American Radio. The archive holds 259,915 broadcast records across 6,114 programs. Twenty-eight percent of them still have no date.

The paper also proposes a way to mark how strong the evidence is for each date. This separates a date proved by several independent listings from a date that was only chosen, or one that arrived from another collection and was never tested.

One example shows the method at work. Six independent listings move one recording's date by almost seven years. The day of the week cannot settle the question, because both possible dates fall on a Sunday, and the program aired on Sundays.

The paper reports one limit as well. The number of listings available for a broadcast rises sharply with the year across the 1930s in the sample studied. The measure of this rise is a correlation of 0.899 across fourteen records. Evidence is not equally thick across the whole period. Any dating system must show this, and not hide it behind a flat claim of fact.

Keywords: old-time radio, discography, broadcast dating, newspaper radio logs, provenance, audio archives, metadata

---

1. The problem: a body of evidence that lost its dates

Old American radio survives mainly because collectors saved it. People recorded the programs on transcription discs and later on tape. People copied the copies. They traded them, and in time they made them into computer files. This went on for about seventy years. No library and no archive set a rule for the information that traveled with each recording. The result is a large body of valuable history with almost no reliable record keeping.

The clearest failure is the date. A recording without a date is much weaker as evidence. You cannot place it against the news of its week. You cannot connect it to an advertiser's campaign. You cannot say where it falls inside a program's run. You cannot use it to show what a network really broadcast on a given night. Radio was live and scheduled, and then it was gone. So the date is not one detail among many. The date is what turns a sound file into a historical document.

Dates disappear in several different ways. It helps to separate them, because each one needs a different answer.

File names break down. In trading networks, the file name carries the information. People write the name again at every transfer to fit their own system. Computers cut long names short. Some letters are not legal on some systems, so people remove them. Computer programs for cataloging change others. A date written as xx-xx-xx may survive one copy. It does not survive three.

Wrong dates are passed along. When a date is present, it is often passed along rather than proved. A collector chooses a date that looks reasonable. Everyone who receives the file copies that date forward. The date starts to look official over time, only because so many people repeat it. Nothing in a trading network ever removes a wrong date.

Programs are dated, but recordings are not. Many holdings carry only the years the program ran, such as 1952 to 1961. That is a fact about the program. It is not a fact about the recording.

Guesses are stored as facts. When a cataloger is not sure, the chosen date goes into the same field, in the same form, as a proved date. Nothing marks the difference. So the difference disappears at the first copy.

This paper treats the last of these as the main problem, because a careful method can repair it. The Association for Recorded Sound Collections states the principle clearly in its Guidelines for Discographies: "estimates are preferable to omitting the information. Discographers should explain the rationale behind estimates, and clearly distinguish them from 'solid' data."¹ The rule is not that every date must be sure. The rule is that the reader must be able to see which dates are sure, and why.

1.1 What existing note books do, and what they cannot do

The field already has good note books. John Dunning's On the Air: The Encyclopedia of Old-Time Radio describes about 1,500 programs. It gives their broadcast histories, their times, their networks, their advertisers, and their main workers.² The Handbook of Old-Time Radio, by Jon Swartz and Robert Reinehr, serves collectors in a similar way.³ The two guide books establish the record at the level of the program, and establish it well. The method described here depends on that work. Step 2 of the procedure in section 4 uses exactly this kind of source to learn a program's usual day and hour.

But these books do not date a single surviving recording, and they never set out to do so. Their subject is the program and its run. The problem here is the single item — one sound file, in trading now, whose place inside that run is not known. A note book can tell a cataloger that Greatest Story Ever Told was a Sunday program, and that it ran in 1948 and also in 1955. It cannot say which Sunday produced the recording they hold. The program-level record in section 5.4 fits the correct date and the wrong date equally well.

So the hole is built into the problem. It is not a failure of the existing books. A cataloger dating a single item needs evidence about that single item. The newspaper radio log is the only published record for this period that works at that level, day by day. This paper argues that the radio log should be treated as the primary source for dating these recordings. It also argues that each record should show which listings were consulted, and what they proved.

2. The collection

The Archive of American Radio is a public, not for profit catalog of American broadcast recordings. The archive published 259,915 separate broadcast records across 6,114 programs in September 2026.⁴ Every broadcast record has a fixed web address in the form /series/{program}/episodes/{item}. Each record, where the evidence allows, also carries a formal citation, a set of newspaper listings, and a set of trade magazine listings.

The collection did not arrive all at once. It gathers material from the main old-time radio distribution channels, so it carries every dating problem described above, at full strength. Much of the cataloging work replaces a date that was passed along and never tested with a date proved against published evidence from the time.

Two facts about the published catalog can be measured directly. Both matter for the method.

The evidence is nearly complete for the classic period. The archive holds sixty records, chosen at random, whose item name carries a date between 1930 and 1959. Fifty-nine of them — 98 percent — carry at least one newspaper or trade magazine page image.⁵ The evidence this method needs is almost always there for the years when the American radio log was a daily feature.

Old dates survive beside corrected ones. Of the 259,915 records, 44,566 carry a date in the item name, in the form YYYY-MM-DD, passed along from the file name of a source collection. Two records in a random sample of forty from the whole collection — 5 percent — carry a name date that does not agree with the broadcast date the record itself states. One record in sixty did not agree inside the 1930 to 1959 window.⁶ Both samples are small, and the numbers should be read only as a rough size. The two samples together suggest that one to two thousand records still carry an old date in the name after the catalog has corrected the date itself. Section 5.4 examines one of them closely.

More than a quarter of the collection has no date at all. In a random sample of 150 records from the whole catalog, 108 — 72 percent — carry a broadcast date. Forty-two — 28 percent — carry none.⁷ Forty-two in 150 points to about seventy-two thousand items held with no date across 259,915 records.

The link between dating and evidence in that sample is the more useful result. 101 of the 108 records with dates — 93.5 percent — carry at least one page image. Two of the 42 records without dates carry one. So records without dates are not, as a rule, records where someone looked for evidence and failed to find it. They are records where no one has looked yet. Dating a record and gathering its evidence are nearly the same job in this collection. The forty records in the sample with no date and no evidence stand for work still to be done, not for a limit on what the method can reach.

2.1 The same broadcast, held more than once

The published catalog shows a connected problem. The archive holds three separate records for Winston Churchill's speech on Anglo-American unity, which he gave at Harvard on 6 September 1943. Two of them carry the date Monday, 6 September 1943, and six page images each. The third carries no date and no evidence at all.⁸

This is normal for a collection built from trading. The same broadcast arrives by several routes, and each copy carries a different amount of information. Removing the copies is itself a dating problem. The record without a date is a weak record whose answer the archive already holds, completely proved, in records it cataloged separately. Sorting out these groups is among the most useful work a cataloger can do, because the evidence has already been found.

3. Kinds of evidence

Three kinds of evidence can fix the date of a surviving broadcast recording. They are different in how often they exist, in how exact they are, and in the kind of mistake each one invites. A method that treats them as equal will produce sure wrong answers.

3.1 Newspaper radio logs

This is the main kind. Large American daily newspapers printed a day-by-day radio schedule from the early 1930s, called the radio log. It listed the station, the time, and the program name for the coming day. These logs are the closest thing the period produced to a complete published record of what went on the air.

Four newspapers carry this method: the New York Times, the Washington Post, the Chicago Daily Tribune, and the Los Angeles Times. Their value comes less from any one of them than from the fact that they are separate. Each paper built its own log, from information supplied separately by networks and stations, for a different city and a different time zone. A program name and hour that appears in all four is confirmed in a way that a single listing can never be.

Radio logs have three weaknesses, and each one must be answered openly. First, a log records the schedule — what the network meant to broadcast — and not the broadcast itself. A news bulletin going out instead breaks the agreement. Second, the log went to the printer before the broadcast day began, so it shows an earlier deadline. Third, the logs are printed in very small print in narrow columns. Computers read small print badly, so mistakes enter easily.

3.2 Trade magazines

Broadcasting, the industry's weekly magazine, and magazines like it, recorded changes to schedules, advertiser assignments, network line-ups, and special events. They give a kind of detail the daily newspapers never tried to give. Trade coverage is weekly, not daily, so it covers fewer dates. But it is much richer in the detail that separates one broadcast from another inside a run. It is the natural second source when a newspaper log proves that a program aired, but not which episode aired.

3.3 Evidence inside the recording

The recording itself contains information that can be dated. A note to the news sets an earliest possible date. An advertiser's words, and the special campaign named in them, can be placed inside a season. The announcer's name, the network's words, and the wording of station identification all changed over time, and those changes can be dated. A broadcast sometimes names a current event. It may give an election result, a war report, or the result of a game. The recording can then often be placed to the exact day from what it says. No outside listing is needed.

This is the only kind of evidence that speaks to the broadcast rather than the schedule. So it decides the question when the two do not agree. It is also the kind most easily mixed up by a later repeat, a later copy, or an Armed Forces Radio Service reissue of the same program.

4. The procedure, and the strength of evidence

The procedure below is written so a reader can follow the same steps. It begins with a recording and whatever information arrived with it. It ends with a date, a mark showing how strong the evidence is, and the citations that support both.

Step 1 — Set the date that came with it aside. Write down the date that arrived with the item. Note where it came from: the file name, an index from the source collection, or an earlier catalog. Mark it as not tested. Do not use it as evidence at any later step. People leave out this step more than any other, and leaving out it is what turns a chosen date into a seeming fact. A cataloger who starts from the passed-along date will find support for it, because the search has become a search for support.

Step 2 — Identify the program and its usual time. Find the program, and from published schedules across its run, find its usual day and hour. This narrows the possible dates. It never proves one by itself. Section 5.4 shows two possible dates, seven years from each other, that both fit the program's usual time.

Step 3 — Get possible dates from the recording itself. Listen for content that can be dated: notes to the news, advertising campaigns, the announcer's name, network words, station identification, and any spoken date. References to the news give the strongest limit, often to the exact week. Record each one as a range, not as a single date.

Step 4 — Test each possible date against the radio logs. For every possible date, read the radio log of each of the four newspapers for that date and city. Write down each agree with as the newspaper, the date, and the kind of listing. A date that survives in two or more separately produced logs is confirmed. A date that survives in only one is supported.

Step 5 — Check the trade magazines. Read Broadcasting for the same week, for schedule changes, advertiser assignments, and special events. Trade evidence is what separates one broadcast from another inside a run, when the daily logs give only the program name.

Step 6 — Test for a repeat or a reissue. Before you assign the date, ask if the recording is an Armed Forces Radio Service reissue, a regional repeat, or a later copy of the same program. Evidence inside the recording decides this. Whether advertising is there or not is often enough, because reissues for the armed forces had the advertising removed. A recording that agrees with a listing has not thereby been shown to be the broadcast that listing describes.

Step 7 — Mark the strength, and cite the sources. Give the record the strongest mark the evidence supports. Cite every listing you used. Say so where the date is an estimate, explain the reasoning, and give a range instead of a single day.

Step 8 — Compare with the other copies. Where the collection holds other records of the same broadcast, settle them together. A record without a date takes the evidence of a confirmed record, not simply its answer. See section 2.1.

The procedure produces a date together with a mark for the strength of its evidence. The marks are:

Mark A — Confirmed. Two or more separate published listings from the time support the date, and nothing inside the recording contradicts them. The listings must be separate: two newspapers printing the same network schedule are one source, not two.

Mark B — Supported. One published listing from the time supports the date, and nothing inside the recording contradicts it.

Mark C — Reasoned. No listing states the date directly. But it follows from evidence inside the recording together with the program's known schedule. For example, a word about the news fixes the week, and the program is known to have aired on Tuesday evenings.

Mark D — Estimated. A limited estimate, given as a range and not as a single day, with the reasoning stated. As ARSC says, an explained estimate is better than no information.

Mark E — Inherited, not tested. A date carried forward from a source collection that the archive has not yet tested against published evidence. This mark exists so that not tested dates stay visible as not tested, instead of quietly becoming facts. It is the mark that cataloging work moves records up from.

The most important feature of this system is simply that Mark E can be recorded at all. A system in which every date is only a date cannot show the difference between a date the archive has proved and a date it has only received. Such a system will give a false picture of its own evidence, no matter how careful the work behind it has been.

5. Examples

5.1 A broadcast with a great deal of evidence: 1 September 1939

The BBC announcer Alvar Lidell reported the German invasion of Poland. The archive holds the report as a recording of forty-five seconds.⁹ The catalog record carries ten separate page images. They include radio logs from the Chicago Daily Tribune, the New York Times, and the Los Angeles Times. They also include two different listings from the Washington Post. All of these are for Friday 1 September 1939. The matching issue of Broadcasting completes the set.¹⁰

This is a Mark A date by a wide margin. It shows the best the method can do. A broadcast that mattered around the world, on a day that mattered around the world, is listed everywhere, and confirming it is easy.

5.2 The same day, a different recording

The archive catalogs Ignacy Jan Paderewski's speech about the coming German invasion separately, on the same day. It rests on the same group of evidence: a Chicago Daily Tribune radio log for Friday 1 September 1939, and the same issue of Broadcasting.¹¹ The two is useful for what it shows about method. One group of listings for one day can support several separate broadcast records. The part of evidence is the printed page, not the recording.

5.3 A broadcast with little evidence: 1932

Now compare a recording from 1932. The archive's record for Franz von Papen's speech as chancellor, on 13 August 1932, carries three page images. They come from the New York Times and the Washington Post only.¹² The Chicago Daily Tribune and Los Angeles Times listings that support the 1939 records are missing. The date is supported, but the confirmation behind it is much less. A system that reported both records the same way would give a false picture of the second one.

5.4 A case that corrects a passed-along date by seven years

The value of the method lies in correction, not in repetition. A clear case is the Greatest Story Ever Told episode "They Did Not Receive Him."

The recording is traded under a file name holding the date 13 February 1955. That date also survives in the archive's own item name, passed along from the source collection. But the archive's catalog record gives the broadcast date as Sunday, 29 February 1948 — an error of almost seven years in the traded date.¹³

Six separate listings from the time support the 1948 date. They are radio logs in the New York Times, the Washington Post, the Los Angeles Times, and the Chicago Daily Tribune. A separate "Radio Highlights" item in the Chicago Daily Tribune is the fifth. The matching issue of Broadcasting is the sixth. All are for Sunday, 29 February 1948.¹⁴ This is a Mark A date.

Two features make the case worth studying.

First, the day of the week cannot settle it. Greatest Story Ever Told was a Sunday program, and both possible dates fall on a Sunday. Both 13 February 1955 and 29 February 1948 were Sundays. A method that relied on the program's known day would have accepted the passed-along date without trouble. Only the supporting listings separate the two. This is the reason for requiring published evidence instead of reasoning from the schedule.

Second, the correct date is a leap day. 29 February 1948 exists only because 1948 was a leap year. Leap days are a known source of passed along error. A cataloger putting a date back together from a damaged or partial file name will usually to change an strange 02-29 into a date that looks usual. The error here runs in exactly the direction that explanation predicts.

The archive's own citation for this record already points to a web address without the wrong date in it. So the corrected date is what the catalog states and cites. The old 1955 text survives only in the item name — which is itself an example of the file-name problem described in section 1, now repeated inside a catalog that has otherwise corrected it.

6. The evidence is not equally thick across the period

The three examples above suggest a way, and the way can be measured. The archive holds fourteen cataloged broadcasts from one news program, covering 1932 to 1939. The number of separate page images attached to a record rises strongly with the year of the broadcast. The correlation is 0.899 across fourteen records.¹⁵ The average number of page images by year:

Two reasons are possible, and these measurements cannot separate them. The first is that newspaper radio logs really did grow across the 1930s, as radio became a mass medium and the log grew from a new thing into a fixed daily feature. The second is that the archive's own collecting of newspaper images is simply more complete for the later 1930s. The shape describes the archive and not the period if the second explanation is right.

The difference matters for how the marks should be read. Under the first explanation, a Mark B date from 1932 is as strong as the evidence of 1932 lets. The same date is only a record the archive has not finished working on under the second. Telling them from each other would require looking at a newspaper index that is independent of the archive's own holdings. That has not been done.

The finding is limited in another way as well. It rests on fourteen records, all from a single program of international news. That is exactly the kind of program most likely to appear in every city's radio log. A general statement about the whole collection would need a sample spread across program kinds and networks.

7. Limits

A schedule is not a broadcast. Every date this method takes from a newspaper proves what was scheduled. Programs were pulled, above all during news events — and news events are exactly when this collection is richest and this reasoning is weakest.

Repeats and reissues. A recording that agrees with a listing may be an Armed Forces Radio Service reissue, a regional repeat, or a later copy of the same program. Only evidence inside the recording can tell the difference, and it is not always present.

Text reading errors. Radio logs are printed in dense, small kind. Computers read them badly. A mistake in the evidence passes with no sign into the catalog.

Small samples. The finding in section 6 rests on fourteen records from one program. The collection-wide numbers in section 2 come from the published catalog, and completing them needs the archive's own cataloging history.

Copying the evidence. For this period, the newspaper pages that provide the evidence are still under copyright. This paper cites listings by newspaper, date, and kind of listing. It does not copy the pages.¹⁶

8. Publishing the reasoning, not only the answer

A mark for the strength of evidence solves nothing if it lives only in a cataloger's private notes. The failure described in section 1 is exactly that careful decisions flatten into simple claims as they pass from person to person. The mark must be published with the record to survive the trading networks that will pass this material on, and attached to it firmly.

Three rules follow. The archive described here follows all three.

The citation belongs to the record, not to a report. Each broadcast record carries its formatted citation and its list of supporting listings beside the date itself, at the same address. A reader who reaches the record reaches its evidence. This is what makes the undated copy in section 2.1 visible as a problem instead of invisible: two records state a date with six listings each, a third states nothing, and anyone can see the difference.

The name must not carry a competing claim. Section 5.4 is a warning. The catalog holds the corrected date, cites six listings for it, and points its citation at an address without the wrong date. Yet the item name still reads 1955-02-13, and that text is what another system reading web addresses will collect. A name that contains a date makes a claim, if or not anyone meant it to. The estimate in section 2 suggests one to two thousand records are in this state. Dates belong in fields, not in names.

Missing information must be recordable. Twenty-eight percent of the collection has no date, and the catalog says so rather than supplying a likely year. This is the same rule ARSC asks of discographers in the passage quoted in section 1. The other choice, for a date with no support, is not an empty space to fill in. It is a clear statement that the question is still open. Mark E holds that statement. The number of two in forty-two — records without dates almost never carrying page images — is what makes it useful, because it identifies the undated records as work waiting to be done.

9. Conclusion

The dating problem in old-time radio is not mainly a problem of missing evidence. The evidence largely exists. It sits in the radio logs of four large daily newspapers and in the trade press, and most of it is now on computers. The problem is that the information kept with these recordings has no way to show the difference between a date proved against that evidence and a date passed along without it. So the two can no longer be told from each other.

The answer offered here is simple. Attach a mark for the strength of evidence to every date. Keep a visible category for dates that were passed along and never tested. Cite the exact published listing behind every confirmed date. Applied across 259,915 records, this turns a trading collection into one that scholars can cite.

---

Notes

1. Association for Recorded Sound Collections, Guidelines for Discographies. ↩
2. John Dunning, On the Air: The Encyclopedia of Old-Time Radio (Oxford University Press, 1998). ↩
3. Jon D. Swartz and Robert C. Reinehr, Handbook of Old-Time Radio: A Comprehensive Guide to Golden Age Radio Listening and Collecting (Scarecrow Press, 1993). ↩
4. Counted from the published sitemap of radioindex.org, 1 September 2026: 259,915 episode addresses across 123 sitemap files, and 6,114 series addresses. One file (episodes-96.xml) returns a valid but empty list; it is counted among the files and excluded from the record count. ↩
5. Random sample of sixty cataloged broadcasts drawn from the 40,727 item names carrying a date between 1930 and 1959; retrieved 1 September 2026. Fifty-nine carried one or more page images. ↩
6. Random samples of forty records from all 44,566 date-carrying item names, and sixty from the 1930 to 1959 group; retrieved 1 September 2026. A disagreement means the date in the item name differs from the broadcast date the catalog record states. ↩
7. Random sample of 150 records from the whole catalog; retrieved 1 September 2026. ↩
8. "Winston Churchill on Anglo-American Unity," Radio News; and "Anglo-American Unity" (two records), Winston Churchill Speeches and Radio Broadcasts. Radio Index, retrieved 1 September 2026. The first two carry the date Monday, 6 September 1943 and six page images each; the third carries neither date nor image. ↩
9. "Alvar Liddell Reports the German Invasion of Poland." WWII News. Radio broadcast, 1 September 1939. Radio Index. https://www.radioindex.org/series/wwii-news/episodes/alvar-liddell-reports-the-german-invasion-of-poland. ↩
10. Radio logs, Chicago Daily Tribune, New York Times, Los Angeles Times, and Washington Post, all Friday, 1 September 1939; and Broadcasting, 1 September 1939.¹⁷ ↩
11. "Ignacy Jan Paderewski on the Nazi Invasion Looming over Poland." WWII News. Radio broadcast, 1 September 1939. Radio Index. https://www.radioindex.org/series/wwii-news/episodes/ignacy-jan-paderewski-on-the-nazi-invasion-looming-over-poland. ↩
12. "Chancellor Franz von Papen." WWII News. Radio broadcast, 13 August 1932. Radio Index. Supporting listings: New York Times and Washington Post radio logs, Saturday, 13 August 1932. ↩
13. "They Did Not Receive Him." Greatest Story Ever Told. Radio broadcast, 29 February 1948. Radio Index. https://www.radioindex.org/series/greatest-story-ever-told/episodes/they-did-not-receive-him. The item name keeps the wrong inherited date 1955-02-13. ↩
14. Radio logs, New York Times, Washington Post, Los Angeles Times, and Chicago Daily Tribune; "Radio Highlights," Chicago Daily Tribune; and Broadcasting — all Sunday, 29 February 1948. ↩
15. Sample of fourteen cataloged broadcasts from WWII News, retrieved 1 September 2026; page-image counts taken from each record's newspaper and periodical sections. ↩
16. On the copyright of the recordings themselves: United States law, at 17 U.S.C. section 1401, brought sound recordings made before 1972 under federal copyright. Recordings first published between 1926 and 1946 are protected for 100 years, and those from 1947 to 1956 for 110 years. Unpublished recordings made before 1972 are protected until 15 February 2067. There is no general rule that material from before 1972 is free of copyright. ↩
17. Following the Chicago Manual of Style, 18th edition, page numbers are left out of newspaper citations throughout this paper. A newspaper may print several editions in one day, and items move or disappear between them. The newspaper, the city, the date, and the kind of listing identify the item; the page number does not do so reliably. ↩

---

Discography

Following the Chicago Manual of Style, 18th edition, sound recordings are listed here and not in a bibliography. The Archive of American Radio holds these recordings. Each is cited by its permanent web address. The dates given are the broadcast dates established by the method in section 4, with the strength mark in parentheses.

"Alvar Liddell Reports the German Invasion of Poland." WWII News. Radio broadcast, 1 September 1939 (Mark A). Radio Index. https://www.radioindex.org/series/wwii-news/episodes/alvar-liddell-reports-the-german-invasion-of-poland.

"Anglo-American Unity." Winston Churchill Speeches and Radio Broadcasts. Radio broadcast, 6 September 1943 (Mark A); and a second record of the same broadcast carrying neither a date nor a supporting listing (Mark E). Radio Index.

"Chancellor Franz von Papen." WWII News. Radio broadcast, 13 August 1932 (Mark A). Radio Index.

"Ignacy Jan Paderewski on the Nazi Invasion Looming over Poland." WWII News. Radio broadcast, 1 September 1939 (Mark A). Radio Index. https://www.radioindex.org/series/wwii-news/episodes/ignacy-jan-paderewski-on-the-nazi-invasion-looming-over-poland.

"They Did Not Receive Him." Greatest Story Ever Told. Radio broadcast, 29 February 1948 (Mark A). Radio Index. https://www.radioindex.org/series/greatest-story-ever-told/episodes/they-did-not-receive-him. Circulates under the wrong inherited date 13 February 1955; see section 5.4.

"Winston Churchill on Anglo-American Unity." Radio News. Radio broadcast, 6 September 1943 (Mark A). Radio Index.

The twenty WWII News records examined for section 6 appear with their supporting sources in the Appendix, and are not repeated here.

---

Bibliography

Dunning, John. On the Air: The Encyclopedia of Old-Time Radio. Oxford University Press, 1998.

Swartz, Jon D., and Robert C. Reinehr. Handbook of Old-Time Radio: A Comprehensive Guide to Golden Age Radio Listening and Collecting. Scarecrow Press, 1993.

Newspaper radio logs and trade magazine listings are cited in the Notes and are not repeated here, following the Chicago Manual of Style, 18th edition, which leaves newspaper items out of the bibliography.

---

Sources for the discography

This section states the sources the method rests on, what each one covers, and where the coverage fails, so that a reader can follow the same steps described in section 4.

Newspaper radio logs. Four large daily newspapers carry this method: the New York Times (New York), the Washington Post (Washington), the Chicago Daily Tribune (Chicago), and the Los Angeles Times (Los Angeles). Each built its log from information supplied for its own city and time zone. That is why agreement between them counts as confirmation rather than repetition. Two kinds of listing are counted separately where a newspaper printed both: the daily radio log, which is a complete schedule, and radio highlights, which is a selected column written by an editor. The Chicago Daily Tribune record for 29 February 1948, discussed in section 5.4, provides one of each, and they count as two listings because they were produced separately.

Coverage is not even. The number of listings available for a broadcast rises across the 1930s within the archive's holdings, from an average of 3.0 in 1932 to 7.8 in 1939 in the sample studied (section 6). Ninety-eight percent of the cataloged broadcasts sampled for 1930 to 1959 carry at least one page image. The number is 68.7 percent across the whole collection. Almost all of that difference comes from records with no date, of which only two in forty-two carry any page image at all.

The source of the newspaper images. The catalog presents each newspaper listing as a page image, labeled with the newspaper, the city, the date, and the kind of listing. It does not record where the scan came from. The published version of this paper should name that source. It should say whether each run was read in a library newspaper database, on microfilm, or in a scan collection made by someone else. It should also say what is known about how complete that source is. Where the source is a private collection rather than a library, the paper should say so plainly. It should give what is known about the collection's history and about any rights it claims. A reader cannot follow steps whose starting point is unnamed, and recording it is the single most useful addition this apparatus still needs.

Trade magazines. Broadcasting, the industry weekly, is read through WorldRadioHistory, which the catalog names and cites as the source of the scan. Trade coverage is weekly rather than daily. It is used for schedule changes, advertiser assignments, network line-ups, and special events — that is, for separating one broadcast from another inside a run when the daily logs give only a program name. Broadcasting appears for the better documented broadcasts in the sample studied, and is missing for the thinner ones. It adds to the daily logs. It does not replace them.

Evidence inside the recordings. The recordings are the archive's own holdings. Each has a permanent web address in the form /series/{program}/episodes/{item}, and each carries a formal citation. Evidence inside a recording, where this paper cites it, means sound that can be heard in the archive's copy.

The points where coverage fails. Three failures are known, and any reader following this work should expect them. Radio logs record the schedule and not the broadcast, so a replaced program breaks the match — and that happens most often during news events, which is when this collection is richest. Logs went to press before the broadcast day began. And they are printed in dense, small type that computers read badly, so mistakes in the evidence pass silently into the catalog. Adding more newspapers does not fix any of these, because all four newspapers share them.

---

Appendix: Dated log

The log below covers the WWII News recordings examined for this paper, in order of broadcast date. It follows the usual form for a discography: the date, the title, the running time, the number of separate supporting listings, those listings by short name, and the strength mark from section 4.

Short names: NYT New York Times; WP Washington Post; CDT Chicago Daily Tribune; LAT Los Angeles Times; Bcstg Broadcasting. Both are counted where a newspaper printed a radio log and a separate radio highlights item for the same date, because they were produced separately.

Every record in this log carries Mark A. Each is supported by two or more separately produced listings from the time. That sameness is a feature of the program sampled, not of the whole collection. WWII News is international news, the kind of program most reliably printed in every city's radio log. A log drawn from local daytime drama would show a very different spread of marks, and section 6 should be read with that in mind.