FOR 64 YEARS, most of the Baltimore Museum of Art's 34 mosaics from ancient Antioch have remained on the beige stone walls of its central court.
They blend into the background so well as to be taken for granted by museum-goers circling to the latest show.
Actually, the mosaics are not from Antioch proper. The earthquake that demolished that great city in 526 A.D. probably made its floors unrecoverable.
Archaeologists would have had to wreck an unremarkable but populated modern town atop the ruins to find out, and did not.
This stuff comes from the ancient suburbs of Antioch. To a large extent, these mosaics were the floors of private houses. They were the better suburbs.
The exhibition of "Lost" Antioch, at the Baltimore Museum of Art through Dec. 30, on one level displays the rich arts and life of that cosmopolitan center of the Hellenist and Roman worlds. On another, it commemorates the vast dig in the 1930s that deposited a great collection of Greco-Roman mosaics on Baltimore walls.
As such, it is partly about Robert Garrett, the Baltimore financier who had explored the Middle East in his youth and was a trustee of Princeton University and the Baltimore Museum of Art.
When Syria, under French mandate, allowed archaeologists to seek the site of the ancient city known only through literature, the Louvre Museum in Paris, Princeton University, the BMA and the Worcester Art Museum in Massachusetts teamed up to do it. Garrett got the BMA involved.
The dig began in 1932. Harvard University and Dumbarton Oaks joined in 1936. Local ferment produced a plebiscite in 1938 that resulted in the district's transfer to independent Turkey in 1939, when the digging stopped.
It has never resumed. More mosaics may remain outside modern-day Antakya, as it is called today, but no one may ever know.
This exhibition, organized by the Worcester Museum and already seen in Worcester and Cleveland, brings together mosaics from the original sponsors and many objects from other collections. After this year, they might never be seen together again.