A San Antonio railroad crossing where cars are reportedly pushed off the tracks
In this part of the broadcast, the host, whom a second speaker addresses on air as “Mr. Bill,” tells a short story that he presents as true. It concerns a train crossing in San Antonio, Texas. He says a school bus full of children stalled on the track and was struck broadside by a train, killing all or most of the children. Ever since, he says, cars left in neutral on the tracks are pushed off to safety. He adds that testers have put talcum powder on cars and afterward found children's handprints in the powder, and that this reportedly happens again and again.
This clip comes from the source recording. AI chose where it starts and ends, so it may start or end a little early or late. The word-for-word transcript is not published. You can hear the source recording on the Internet Archive (file “Ghost to Ghost 1998.mp3”, starting at 1:01:55). Is this you, or someone you know? Ask for a correction or removal.
The story as told
The host says he will tell [quote withheld] and states plainly that it [quote withheld] According to him, in San Antonio, Texas, a school bus full of children stalled on a train track. A train hit it broadside and killed all or the majority of the children on the bus. He calls this [quote withheld] He then says that ever since, at that particular intersection, if a car stalls, the car is [quote withheld] Asked implicitly how this is known, he says people go and test it. They drive their car onto the tracks, stop, and put the car in neutral. He adds that tests have been done with talcum powder on the back of the car, and that when the car reaches the other side of the tracks, [quote withheld] He states that this [quote withheld] and [quote withheld] A second speaker on air says [quote withheld] and the host answers by restating that it is a true story. The host closes by saying that a story “from the Enterprise” discussed earlier reminded him of this one. No date for the crash, no name for the crossing, and no named witnesses are given.
Reported events, in order
- 0:03
The host introduces the story as a “very quick story” and says it is true.
- 0:10
He states that a school bus full of children stalled on a train track in San Antonio, Texas.
- 0:17
He says a train hit the bus, killing all or the majority of the children.
- 0:31
He says that ever since, cars that stall at that intersection are reportedly moved to safety.
- 0:41
He describes people driving onto the tracks, stopping, and putting the car in neutral as a test.
- 0:47
He reports the car being “pushed or seemingly” moving by itself off the tracks onto the road on the other side.
- 0:59
He says cars were tested with talcum powder on the back.
- 1:11
He reports children's handprints being found in the talcum powder after the car crossed.
- 1:21
He says this can be repeated [quote withheld]
- 1:27
He links the story to one “from the Enterprise” that had been under discussion.
Claims
Each claim is labeled as something the person telling the story noticed (observation), what they made of it (interpretation), something others told them (hearsay), or something the host said (host statement).
- host statement· 0:03
The host says the story is true and asks the audience to believe it.
- host statement· 0:10
A school bus full of children stalled on a train track in San Antonio, Texas.
- host statement· 0:17
A train struck the bus broadside, killing all or most of the children aboard.
- host statement· 0:22
The crash killed all or the majority of the children on the bus.
- host statement· 0:31
Ever since the crash, cars that stall at that intersection are reportedly moved to safety.
- host statement· 0:41
People reportedly drive onto the tracks, stop, and put the car in neutral to test the story.
- observation· 0:47
The car is said to be pushed, or to seemingly move by itself, off the tracks onto the road on the other side.
- host statement· 0:59
Cars have reportedly been tested with talcum powder applied to the back of the car.
- observation· 1:11
Children's handprints are reportedly found in the talcum powder once the car reaches the other side of the tracks.
- host statement· 1:21
This is said to happen again and again.
- hearsay· 1:09
A second speaker on air says they have heard of the story. This shows they know it, not that they saw anything.
Scores, with reasons
An AI model gave these scores using a fixed set of scoring rules (scoring-v1). They are editorial judgments and rough guides, not the odds that anything paranormal happened. How scoring works.
The story is relayed, not witnessed. It names a city and describes a test, but gives no date, no named witnesses, and no examined evidence. It is a specific, coherent retelling of a circulated story, which puts it above vague hearsay but well below a firsthand story that can be checked.
Evidence for and against
- A named city is given: [quote withheld]
- The reported test is described concretely: drive onto the track, stop, [quote withheld] with talcum powder on the back.
- The sequence holds together, and the second speaker's interjection fits the story rather than contradicting it.
- The host allows for an ordinary reading of the car's movement: [quote withheld]
- The host is not a witness. The story is framed as something the audience may have heard before ([quote withheld]).
- No date, no name for the intersection, and no named witnesses or officials are given. The testers are “they” and “people.”
- The strongest claim rests on a repeated test result (powder handprints) that is asserted, not demonstrated.
- [quote withheld] hedges how it happens, even though the story's force comes from something unseen.
- The labels showing who is speaking switch mid-sentence, so which words belong to the host and which to the other speaker is partly uncertain.
- No date for the claimed crash is given, and the crossing is never named.
- 0:03
- 0:10
- 0:47
What the model inferred (not said on air): Closeness to events scores low because the host is recounting a story he has heard, not describing anything he experienced. The specifics and consistency scores reflect the concrete details he does give.
There is only the host's word, plus details that could be checked (a city and a described test) but were not. A physical trace is claimed but was not produced or examined. This places the story in the low range of evidence strength.
Evidence for and against
- A place that could be checked is named: San Antonio, Texas. A specific test that can be repeated is also described.
- The test result is claimed to be repeatable: [quote withheld]
- No documentation, recording, photograph, or physical sample is produced or examined in the transcript.
- No witness speaks on air about having taken part in a test. The second speaker only says they have heard of the story.
- The only “trace” claimed, handprints in powder, is not in the studio and is asserted secondhand.
- It is not stated whether the host has any documentation of the reported tests.
- The address “Mr. Bill” is presumably a mistake in the transcript for the host's name, so who made the remark that backs up the story is uncertain.
- 0:59
- 1:09
What the model inferred (not said on air): I treat the claimed talcum-powder handprints as an asserted result, not as physical evidence in hand, because the transcript shows no object being examined on air.
The story turns on the deaths of children at the crossing and on small handprints found on a tested car. That imagery is truly disturbing. But the reported effect is protective, not harmful, which keeps the story in the middle range.
Evidence for and against
- The premise is the death of a busload of children: “killed all the” / [quote withheld]
- The image of small handprints appearing in powder on a car at the site of a fatal crash is unsettling.
- The claim that it [quote withheld] gives it an unresolved, recurring quality.
- The reported effect is protective: the car is moved [quote withheld]
- Nothing malevolent, threatening, or physically harmful to living people is described.
- The story is told as a brief anecdote.
- The time of day of the tests is never stated, which affects how eerie the scene reads.
- 0:22
- 1:11
- 1:24
What the model inferred (not said on air): The score is a judgment about the story as told to a radio audience, not about whether anything happened.
The story is a familiar roadside-legend frame (“ghost tracks”) with one distinctive detail, the powder-and-handprints test. That places it around the middle of the scale.
Evidence for and against
- The specific detail of talcum powder on the car, with children's handprints found in it after the car crosses, is a distinctive flourish.
- The claimed effect is protective, not frightening, and is presented as repeatable on demand.
- Cars moving off ghost tracks and a fatal school-bus and train crash are familiar parts of a widely told American roadside legend.
- The structure, tragedy, then a repeatable protective strange event, then a physical trace, is a standard legend shape.
- There are no ghosts seen, no named spirit, and no strange event beyond the one trace.
- Originality compared with this archive cannot be judged meaningfully from a single story.
- 1:11
What the model inferred (not said on air): This is a provisional judgment based on content. With few stories in the archive, the comparison with the rest of the archive carries little weight, so the published originality score should be treated as provisional.
Scoring confidence 0.50 · analysis confidence 0.60 · checked against outside sources: no · evidence available: The host's retelling of a circulated story (no firsthand story from the host); No named witnesses; testers described only as “people” and “they”; A reported test with talcum powder and reported children's handprints, claimed but not produced or examined on air; On-air acknowledgment by a second speaker that they have heard of the story (not a witness story).
Ordinary explanations to consider
These are possibilities only. None is a finding about what happened.
- A car left in neutral on a crossing on a slight slope rolls off the rails by itself. · The host says testers [quote withheld] and that the car [quote withheld] A car left in neutral on a crossing with any slight slope would roll downhill. That could produce the reported movement without anything unseen.
- Handprints in talcum powder could come from ordinary handling during or after the test. · The only physical evidence described is powder on a car. Powder marks could be disturbed or added by whoever handles or inspects the car between the test and the observation, not by something unseen.
- The story is a local legend that has circulated, and its details may have piled up as it was passed along. · The host himself says [quote withheld] and the second speaker says they have heard of it. That points to a widely circulated story whose details may be embellished in retelling.
What stands out
This is not a caller's own experience. The host relays a local story that has circulated and says it is true ([quote withheld]). Its claimed evidence is a test anyone can repeat, talcum powder on a car producing children's handprints, not a sighting or a sound.
Related stories
Matched by shared recurring details, categories, and wording.