Archaeologists digging within the sacred precinct of the Temple of Artemis at Ephesus found a small stone-built grave in an abandoned Roman latrine channel. Inside lay two tiny skeletons, side by side, in direct contact with one another. They were premature infants, possibly twins, buried there some twelve centuries ago.
Finding infant remains in a sewer or latrine usually means one thing to archaeologists, and it is not a happy conclusion. Such contexts are conventionally read as evidence of abandonment, disposal, or the selective discarding of unwanted newborns. The new study of this grave argues the opposite, and the evidence is difficult to dispute.
The research, by Caroline Partiot and Lilli Zabrana of the Austrian Academy of Sciences and its Austrian Archaeological Institute, was published in the journal Childhood in the Past.
Layers of a very old place
The setting is a stack of centuries. The famous Temple of Artemis at Ephesus, first built around 550 BC and counted among the Seven Wonders of the ancient world, stood at the center of a sacred precinct. Within that precinct, in the first half of the first century AD, the Romans built an odeion, a roofed hall for musical and poetic performances. Sometime in the second or third century, a latrine was added to the structure, a public facility with capacity for roughly thirty users at once, served by a drainage channel.
Centuries later, long after the odeion had fallen out of use and the channel had gone dry, someone chose that abandoned space for a burial. Working inside one of the odeion’s substructure chambers, they built a small cist grave of brick and stone within the old channel and laid two infants in it. Radiocarbon dating places the burial in the eighth or ninth century AD, the Middle Byzantine period. Archaeologists uncovered it during excavations in 2010, though the full analysis has only now been published.
Two who arrived too early
Judging primarily by the size of the bones, Partiot and Zabrana estimate that both infants were born prematurely, at roughly thirty-three to thirty-four weeks of gestation, about seven and a half months, against a full term of thirty-nine to forty weeks.
The case for twins rests on two observations, that the two died at almost exactly the same developmental age, and that they were buried in direct physical contact with each other. Partiot is careful about the limits of that inference, noting that only future DNA analysis could confirm whether they were genuinely twins rather than two infants who happened to die close together.
Neither skeleton showed signs of trauma or disease. One, however, carried something unusual, a cervical rib, an extra rib growing from a vertebra at the base of the neck. The trait is hereditary and occurs in roughly one percent of people, where it usually causes no problems at all. In infants, though, Partiot’s earlier research has linked cervical ribs to anomalies in the Hox genes, the family of genes that governs how body segments develop in the embryo, and those anomalies are associated with stillbirth, premature birth, and sudden infant death. As the researchers note, no case of twins with cervical ribs had previously been documented in any archaeological context.
Why there, of all places
The obvious question is why a family would bury their children in a disused latrine rather than in one of the city’s cemeteries. The researchers suggest an explanation rooted in the religious situation of the time. Born prematurely and dying almost immediately, the infants would not have been baptized, which in Byzantine Christian practice complicated their eligibility for burial in consecrated ground.
What the grave represents, in the researchers’ reading, is a practical compromise, a way of reconciling Christian burial norms with a family’s need to lay their children somewhere meaningful. Zabrana has described the setting as unexpected and exceptional, noting that previously known graves at Ephesus were placed in a dedicated cemetery outside the boundaries of the old temple precinct. This one was inside it.
No trace of the twins’ mother has been found nearby or in the Ephesus cemetery. Partiot points out that she may have died at the same time or may not have, and that the question is difficult to resolve, since both infant and maternal mortality ran very high in this period.
Care, not disposal
The strongest argument against reading this as a discarding of unwanted infants is the grave itself. Whoever buried these children did not simply place them in the channel. They constructed a protective stone cist around them, and it worked, preserving both skeletons intact for more than a thousand years. Zabrana takes that construction as clear evidence of a respectful burial rather than the disposal of human remains.
The researchers describe what the grave shows as a family’s intention to care for two dead infants and to acknowledge the brief place they held in their community. It is a small conclusion drawn from a small grave, but it carries a broader implication for the discipline, that archaeologists should be slower to assume the worst when infant remains turn up in sanitary contexts. Sometimes an abandoned drain was simply the most sheltered structure available, and the care is visible in how the burial was made.
Further work by the team will look at how the people of early medieval Ephesus reused abandoned sacred spaces, both for everyday needs and for spiritual ones, of which this quiet grave beneath a ruined concert hall is a particularly moving example.
Sources: Live Science (July 24, 2026). Partiot, C., and Zabrana, L. (2026). "Peace in Unexpected Places, The Middle Byzantine Burial of Two Perinates, Possibly Twins, in a Latrine Canal of the Odeion in the Artemision of Ephesos." Childhood in the Past, 1 to 21. doi.org/10.1080/17585716.2026.2682804