In March 2024, a curator at the Boston Historical Society made a discovery that would unravel a mystery hidden for more than a century. Sarah Mitchell, a specialist in nineteenth-century American photography, was reviewing a donation of old photographs when she came across a formal family portrait from 1892. The image, taken by the J. Morrison Studio in Boston, showed a bearded man in his late thirties seated in an ornate wooden chair, a severe-looking woman standing beside him, and three children arranged stiffly around them, as was the custom of the era.
Sarah had examined hundreds of such portraits, but something about this one made her pause. The image was unusually sharp, the details remarkably clear for a photograph more than 130 years old. Then her eyes drifted to the floor beside the man’s chair. Tucked beneath the hem of a small decorative rug was an object that did not belong in a formal Victorian portrait: a small cloth doll.
Roughly handcrafted, worn and clearly loved, it had been deliberately placed where it would be barely visible, present in the photograph without dominating the composition. In the 1890s, portrait photography was expensive and carefully staged. Every element was arranged to convey status and character. A worn child’s toy lying casually on the studio floor was not something a family would have allowed by accident.
Sarah consulted her reference books on Victorian photography customs and found no other example of a toy placed on the floor of a formal portrait. Someone had wanted the doll in the picture, but had not wanted it to be obvious. She began searching the city records. The Morrison Studio’s appointment book for 1892 revealed that the portrait had been taken on November 14th for Thomas and Elizabeth Harper, a family of five subjects.
The payment was recorded in full: three dollars and fifty cents. But when Sarah searched birth records for the Harper children, she found only two sons, James, born in 1884, and Robert, born in 1886. There was no daughter registered to the family at all. The girl in the photograph, the youngest child, stood with her hand resting in Elizabeth Harper’s, yet she did not exist in any official record.
Sarah next turned to Thomas Harper’s employment records at the Waltham Textile Mill, where he had worked since 1881 and risen to supervisor. Tucked between quarterly reports from 1891, she found a newspaper clipping filed with his records. The headline read: Tragedy at Waltham Mill. Five dead in boiler room fire.
The article described a boiler explosion on January 18th, 1891, that had triggered a fire in the mill’s basement. Five workers had died, trapped when a stairwell collapsed. Thomas Harper, who had been supervising the morning shift, had entered the building repeatedly to search for trapped workers. Near the end of the article was a brief mention: Mrs.
Sullivan, one of the victims, left behind a daughter, age four, whose care remained uncertain. Sarah connected the pieces. Margaret Sullivan, a widow and textile worker, had died in the fire, leaving a young daughter in January 1891. The Harper portrait had been taken nearly two years later.
The little girl in the photograph would have been about five or six years old. Further research showed that Margaret Sullivan’s husband had died of pneumonia in 1889, leaving her to support herself and her child alone. She had worked the early morning shift at the mill, which explained why her daughter had not been with her during the fire. Sarah searched orphanage records from the months following the fire and found no admission matching the Sullivan girl.
The child had not gone to an institution. Someone had taken her in privately. Someone who had witnessed the tragedy first-hand. She found the confirmation she needed in school enrollment records for the Cedar Street Primary School.
The register for the 1892 to 1893 school year listed three children from the Harper household: James, age eight, Robert, age six, and Anna Harper, age five. The name Anna appeared nowhere in birth records or census data before 1892. The Harpers had taken in Margaret Sullivan’s daughter and were raising her as their own. Sarah contacted the Waltham Historical Society, where a volunteer coordinator named Dorothy shared an oral history collection containing recorded interviews with descendants of mill workers.
In one transcript from 1979, a woman named Helen, born in 1895, recalled the Harpers. She described them as living two houses down on Cedar Street, with three children, two boys and a girl. Her mother had told her the girl was not theirs by birth, but they loved her just the same. After her real mother died in the fire, Mrs.
Harper’s heart broke for the child. They took her in when others would not. Everyone knew, Helen said, and everyone respected them for it. With the name Anna and confirmation that she was Margaret Sullivan’s daughter, Sarah traced the child forward.
The 1900 census listed Anna Harper, age thirteen, as the Harpers’ daughter without qualification. The 1910 census showed her still living with Thomas and Elizabeth, working as a seamstress. In 1912, she married Joseph Mitchell at St. Mary’s Church in Waltham.
The record listed her father as Thomas Harper, with no reference to Margaret Sullivan. Anna and Joseph settled in Cambridge, where she gave birth to four children between 1913 and 1921. She appeared in the following census records like millions of other ordinary lives, marriages, addresses, occupations, children. Her death certificate from 1967 listed her parents as Thomas and Elizabeth Harper.
She had died at age eighty, surrounded by the family that had grown from a single act of compassion in 1891. Sarah constructed Anna’s family tree and identified six of her grandchildren still living, now in their seventies and eighties. She wrote letters explaining what she had found and asking if any would speak with her. Two weeks later, a woman named Patricia, Anna’s granddaughter, called.
She brought to their meeting a small cloth doll with yarn hair and a simple dress, wrapped carefully in tissue paper. Anna had kept it on a shelf in her bedroom for as long as Patricia could remember. When asked as a child, Anna had said only that it was very special and had belonged to someone she loved very much. Comparing the doll to the one in the photograph, they were identical, the same construction, the same yarn hair, even the same small tear in the dress.
Patricia organized a family meeting the following month. Seventeen descendants gathered in Cambridge, grandchildren, great-grandchildren, and two great-great-grandchildren. Sarah presented her findings, displaying the 1892 portrait on a large screen. She told them that Anna Sullivan was born around 1887, that her mother died in the boiler room fire when she was four years old, and that Thomas and Elizabeth Harper took her in informally, a private arrangement that was never a legal adoption.
They enrolled her in school as their daughter, raised her alongside their own sons, gave her away at her wedding, and never attempted to hide where she came from. That cloth doll, placed deliberately beside the chair, was proof of both truths: that she belonged to the Harper family, and that her first mother had loved her too. The exhibit opened at the Boston Historical Society three months later, titled simply Beside the Chair, One Family’s Act of Compassion. The Harper family portrait was enlarged to show every detail, with the worn cloth doll displayed in a case nearby.
Anna’s descendants stood together before the photograph, studying the faces frozen in time on that November day in 1892. That portrait was proof that she belonged, that she was part of the family, and the doll beside the chair was proof of where she came from. Anna had been loved twice, once by the mother who made her a doll, and again by the family who let her keep it.
The photograph, and the small worn toy barely visible beside the chair, had finally revealed the story that had remained hidden for 132 years.