The Man from the Train

by Bill James

  CHAPTER XXXVII

  The Zoos and the Hubbells

  Between the murders of the Hardy family and the Bernhardt family are the murders of the Zoos and Hubbell families, stories that I mentioned, but did not detail, during our first chronological pass through this time period.

  The Zoos family was murdered with an axe on September 20, 1910, near Byers, Pennsylvania. Byers was a little railroad village west of Philadelphia and a mining town, much like many of our other crime scenes. John Zoos, a Polish immigrant, worked in a plumbago mine, got home about dusk, found his family murdered. (The word plumbago is now used only for a flowering plant. A hundred years ago, however, it was primarily used for graphite.) There is a low probability that the crime is related because it occurred in daylight and may have been a robbery. The crime, which involved the deaths of three people, appears never to have been seriously investigated and was absolutely and completely forgotten within a year.

  * * *

  The family of Oda (Odell) Hubbell was murdered exactly two months later, on Sunday night, November 20, 1910. On February 9, 1911, a man named Hez Rasco was convicted of murdering Oda Hubbell. As was the custom of the time, Rasco was charged with only one of the murders as a hedge against double jeopardy; had he been acquitted of that murder, he could still have been charged with any of the other three, as was done to Ray Pfanschmidt. Rasco was convicted, sentenced to death, and appealed his conviction to the circuit court.

  The circuit court, in their ruling, did a marvelous job of summarizing the case, and I am going to quote from their ruling at length here, not only so you get to hear a voice other than mine telling you about this crime, but also because there are discrepancies between the court’s summary of the facts and what is in the newspapers, and we can assume the court’s opinion to be accurate:

  The deceased, Oda Hubbell, 30 years old, lived with his wife and two small children in a cottage about two miles distant from the town of Guilford, in Nodaway County. On Sunday evening, November 20, 1910, this cottage was burned, and with it the bodies of the deceased, his wife and two children. The children were nearly consumed; nothing being found save some of their bones. The deceased and wife were partly burned. He had been shot in the head, and the wife’s head had been crushed by a blow.

  The state gave in evidence tending to prove the following facts: During Saturday night, November 19th, preceding the homicide, the deceased, the defendant, and one Wallace were playing poker for money in a box car in Guilford. Wallace went home at midnight. Deceased and defendant continued the game until 6:30 Sunday morning, soon after which they separated; deceased going to his home. On this occasion deceased exhibited to defendant a large roll of bills, amounting to $300 or $400. Deceased spent a portion of Sunday visiting his father, who lived nearby, exhibiting there the roll of bills and also a handful of silver money. That evening, at about 6:30 or 7:00, two gunshots, in quick succession, were heard in the vicinity of the home of the deceased, followed shortly by two more, muffled in sound. Later, about 10 o’clock that Sunday night, neighbors saw the cottage in flames. Those first on the ground saw, through the window, the body of deceased lying on the floor, burning fiercely. Later, an empty coal oil jug was found near it. Outside, at one corner of the house, was a pool of blood, and indications showing that a body had been dragged from thence into the house. Near the pool was found an empty shotgun shell. In the ruins of the house were found the remains of Mrs. Hubbell and the two children. The hands of Mrs. Hubbell were burned off. Around one arm was the telephone wire, and lying near, the telephone receiver. Her head had been crushed in by a blow from a blunt instrument. The back bones and some rib bones only of the children were found. A charge of shot had entered the head of the deceased, and some of the shot and the wadding of the shell were extracted therefrom. No silver money was found.

  The defendant, on the morning of that day, borrowed a gun from one Cayton, also some cartridges which corresponded exactly with the one found at the pool of blood. The gun was found a few days later in the hayloft at defendant’s home, with some foreign substance dried upon the stock, which substance contained blood, but not shown to be human blood. On Sunday afternoon defendant was seen near the Hubbell home, and seeking to avoid observation. At the edge of the pool of blood alluded to was observed a heel print in the mud, made by the right heel of a shoe. The heel print faced away from the house, was sunk in the mud, and indicated, it was thought, that it was made by a person who was braced and pulling toward the house. The mark indicated that the heel had three tacks or nails on each side and one in the middle, also that the inside corner of the heel was worn down, and that it was the heel of the right shoe. The next day, Monday, about noon, two bloodhounds, in charge of their master, arrived, and were put upon the scent from this heel track. Proceeding from this heel mark the trail, although broken at one point, led the dogs to the home of defendant and into his bedroom, where were found a pair of his shoes. The heel of the right shoe corresponded to the heel mark described above, being worn down at the corner, and showing the seven nails arranged as shown in the track. There was also found in the room a pair of overalls, which the defendant admitted to the sheriff, as the latter testified, were worn by him on Sunday night. A chemical examination showed that certain spots on the overalls were human blood.

  The defendant, testifying in his own behalf, admitted the card playing in the box car. His story is that on Sunday morning he borrowed the shotgun and shells from Cayton, went hunting, returned about 11 o’clock in the forenoon, attempted to return the gun, knocked at Cayton’s door, and, no one responding, he left the gun on the porch. The Caytons testify that they were at home at the time defendant claimed to have called, but heard no knock and saw no one, although the upper part of the door was of glass, and that they could have both seen and heard defendant had he been there. Defendant says that after this he went by train, about noon, to Ravenwood, a town 12 or 15 miles distant, and returned home about 3 o’clock on Monday morning, spending the interval in Ravenwood. He produced no substantial evidence to corroborate this story, and was contradicted in several particulars by witnesses for the state. He claimed that he went to Ravenwood to collect a poker debt from a man whom he failed to find, and waited about the station until he took the return train. He said that the blood found on the overalls came from his nose during the card game. He admitted former convictions for murder in the second degree and grand larceny. Other facts will appear in the opinion.

  OK, that’s a pretty solid case against Hez Rasco, real name Hezekiah, right? It is. Sometimes prosecutors become convinced that they know what happened when they don’t, but I don’t fundamentally question that Rasco murdered the Hubbell family. But there are nine things about this case that bug me, and I’ve got to tell you about them. The big one is number nine, so stay with me a moment.

  First, the image of two men and only two sitting for six and a half hours in an empty box car on a side rail in the middle of the night in November playing poker for money seems to be so out of the ordinary that I find it difficult to give it complete credence. I’ve been involved in a lot of poker games; I’ve never known two guys who would do that, I don’t think. Four guys or five guys, maybe, but not two.

  Second, accepting that this happened, I can’t get out of my head the image of The Man from the Train, watching them unseen from a dark box car nearby, and following them at dawn when they go back to their houses.

  Third, Hez Rasco was Oda Hubbell’s best friend. He had been to the Hubbell house many times; thus, the bloodhounds finding a trail from Hubbell’s house to Rasco’s is not (in itself) evidence.

  Fourth, while I accept that this is an unrelated crime, for an unrelated crime it sure has a lot of odd similarities to our cases—the crime happening on a Sunday night, the house being set on fire at the conclusion of the crime, the use of coal oil (kerosene) to build the fire, the dragging of the man’s body into the house for no obvious reason. There are other crimes in our series in which the m
an of the house was shot, while the women and children were bludgeoned to death, as seems to have happened here. One of the victims in this crime was an eight-year-old girl named Jessie Hubbell.

  Fifth, less than three weeks later and about 115 miles due south of Guilford, the Bernhardt family was murdered. The Man from the Train at this time of the year would be heading south.

  Sixth—acknowledging that points five and six overlap and are not entirely separate—there are a set of four crimes that form a perfect hand-in-glove, tongue-and-groove chronological and geographic pattern: the Hardy family murders, the Hubbells, the Bernhardts, and the Casaways. If you take the dates of those murders and put pushpins in a map representing where the crimes occurred, it could not possibly fit together any better.

  The murder of an entire family is a rare crime. It just seems odd to me that this family murder (a) happened so close in time and place to an unsolved crime, and (b) fits so perfectly into the time-and-space pattern formed by the other three crimes.

  Seventh, this crime is often described as occurring in Maryville, Missouri, where Rasco was tried and convicted; the crime actually happened near another small town.

  Maryville, Missouri, is closely connected to Villisca. From Maryville to Villisca is fifty miles almost straight north. The bloodhounds who followed a trail to Rasco’s house were the very same bloodhounds who were called to Villisca, the Nofzinger bloodhounds of Beatrice, Nebraska. The river that runs through Villisca, Iowa, is the Nodaway River; this crime happened in Nodaway County, Missouri. A railroad line ran from Maryville to Villisca. I don’t mean that it ran through Maryville and Villisca; I mean that it started in Maryville and ended in Villisca, or started in Villisca and ended in Maryville, whichever way you want to look at it. To the people of Villisca it was known as the Maryville train; to the people of Maryville it was known as the Iowa train or the Villisca train. I find that an odd coincidence under the circumstances.

  Eighth, in what was described as an irony, but might alternatively be described as a conflict of interest, the judge who presided at Rasco’s trial was the lawyer who had defended him when he was on trial for the earlier murder.

  Ninth, the sheriff who investigated the Hubbell family murders was a man named W. R. Tilson, who would lose his campaign for reelection as sheriff. He then ran for and won election as Nodaway County treasurer.

  And who would you guess should turn up in the middle of the fraudulent prosecution of William Mansfield, in Villisca, but W. R. Tilson, county treasurer of Nodaway County, Missouri? Tilson popped up in 1916 in Villisca, telling a story that was so weird and convoluted that nobody believed it. His story was that someone, whom he believed to be William Mansfield, had come to the county treasurer’s office in Maryville, unannounced, asking if someone from Villisca had left an envelope of money for him with the county treasurer. Villisca in 1916 was a weird and confusing place, and lots of people believed lots of weird and confusing stories, but nobody believed that one.

  Tilson had become one of J. N. Wilkerson’s stooges. But if Tilson was willing to participate in the fraudulent prosecution of Mansfield and Jones in Villisca, don’t we have to worry about the integrity of the case that he created against Hezekiah Rasco? Isn’t it possible that what happened here was the same thing that happened to Howard Little, that the investigation focused on him immediately because of his prior murder conviction, and that the rest of the case against him was manufactured?

  In order to believe that Hez Rasco was innocent, one has to believe that he was deliberately framed by the police. There is sufficient evidence to convince us that Rasco is guilty, unless that evidence was manufactured and planted by the police. While police often make damaging mistakes, I think we can generally reject the notion that a small-town sheriff has deliberately manufactured evidence—but Tilson’s participation in J. N. Wilkerson’s Villisca fraud makes that more believable than it otherwise would be. But in the end, I think we have to accept that the murder of the Hubbells is an unrelated case.

  The appeals court rejected Rasco’s motion for a new trial, and he was executed by the state of Missouri on March 26, 1912. Winifred Bonfils (Winifred Black), a famous reporter for the Hearst newspapers, interviewed Rasco in his jail cell and wrote the following:

  Young, slim, erect, square shouldered, sleek haired, as well groomed apparently as any club man, his hands carefully manicured . . . as vibrant with life as an electric engine, and as full of subtle, quick agility and grace as a cat.

  Where in the world did he, a farmer’s boy, born almost in the cornfields and brought up almost at the tail of a plow, ever get the quick, springing tread of an Indian and the square shoulders and set of the head of a West Pointer?

  Rasco was a voracious reader; he explained his educated appearance by the fact that he had read so many books. His favorite book was Victor Hugo’s Les Misérables, which he said that he had read many times. Rasco insisted until the moment he was executed that he was innocent. He was baptized the day before his execution. The minister wanted to bring a water tank into the prison for the baptism, but the sheriff, fearing that Rasco would drown himself in the tank, refused to allow it, so Rasco was baptized in a bathtub in the women’s section of the prison at Maryville, where he was to be hanged. As he completed the baptism, Rasco said he forgave those who had falsely accused him of the murder.

  Rasco left a letter, to be opened by his minister after his death. It was widely expected that this would include a posthumous confession, but it did not; instead, Rasco thanked the minister and others for their comfort to him as a condemned man, and warned young men to stay away from gambling and liquor.

  * * *

  On June 10, 1913, Arthur Kellar and his daughter, Margaret, were murdered, probably with an axe, in Harrisonville, Missouri. The authors are satisfied that this murder had nothing to do with our case.

  CHAPTER XXXVIII

  Clementine Barnabet

  Eleven black families were murdered in Texas, Louisiana, and Mississippi in 1911 and 1912, mostly—or entirely—with an axe. We are not going to tell you about all of those crimes in any detail, because the last few are not relevant to our story, but it is a remarkable number. In this era there were about eight families (or substantial portions of families) murdered per year in the United States. Prorate that to three states and two years, it comes out to one expected event.

  These murders are known as the Clementine Barnabet Murders or the Church of Sacrifice Murders. We wrote earlier (chapter V) about the New Orleans Axeman—also in Louisiana, also active in 1911 and 1912. But it is easy to see that the murders committed by The Man from the Train and the New Orleans Axeman are different and not related. With the Clementine Barnabet Murders, it is not as easy to dismiss the possibility of some confusion between the two. It is our opinion that at least two and possibly as many as four of the Texas/Louisiana murders were actually committed by The Man from the Train, although most of the crimes were committed by other persons.

  November 12, 1909, was a Friday. About 1:00 a.m. on November 13, neighbors heard screams coming from the house of Edna Opelousas in Rayne, Louisiana. Running to her defense, they found Ms. Opelousas dead from a blow by an axe. Her three children (aged four to nine) were still alive, but all had been mortally wounded. The murderer(s) had escaped.

  (Edna’s actual name may have been Edmee. Census records show an Edmee Opelousas, born in Louisiana in April 1882, who had three younger sisters. If this is she, then she had probably never been married. We’re going to call her Edmee going forward.)

  A man named George Washington was arrested on the morning of the murders on suspicion of having some connection to the crime, and, later on, his wife and daughter were arrested as well. His release was not noted by as many newspapers, although he must have been released. The Rayne city marshal told the press that

  • The victims had been stabbed with a knife as well as hit in the head with an axe,

  • He (the marshal) was in possession of the knife,
and

  • He was headed to Crowley, where he believed the knife had come from, to try to identify the knife’s owner.

  Ho—Kay. Not sure how you can tell where a knife came from like that. The main point there is the multiple weapons. Multiple weapons almost certainly indicate multiple murderers.

  Crowley and Rayne were just a few miles apart, almost walking distance. In the 1880s the towns were about the same size and had been rivals in pursuit of the position of the parish seat of Acadia Parish. Crowley won the battle, and since then had grown to be the larger town. A man named Houston Goodwill was arrested the following week. Goodwill was married to one of Edmee’s younger sisters and had been kicked out of the house a couple of weeks before in a domestic dispute.

  A chilling fact for you. The Man from the Train’s previous murder, on Halloween 1909, was in Beckley, West Virginia. His next murder, in March of 1910, was in Houston Heights, Texas. If you go on a mapping service today and ask for directions from Beckley, West Virginia, to Houston Heights, Texas, the route will send you directly through Rayne, Louisiana. Look at it this way: the mapping service will not send you through 99.9 percent of the little towns in America. It will send you through Rayne, Louisiana, as well as Crowley, Lafayette, and Lake Charles. Do the math—and the time frame is about right. The murders in Rayne were committed twelve days after the crime in West Virginia.

  In spite of this, the indications that this was not The Man from the Train seem to outweigh the indications that it was. If there are multiple murderers here, then it isn’t him. No one ever heard any of The Man from the Train’s victims scream, in part because he preferred isolated houses, but mostly because he didn’t stab people with knives before he hit them in the head with an axe. If you stab someone with a knife, they will scream; if you hit them in the head with an axe while they are asleep, they won’t. Clementine Barnabet would later claim this crime as her own, but we will get to that later.