The codex known as P46 actually exists today divided between two collections. Fifty-six leaves of P46 constitute Papyrus 2 of the Cheaster Beatty Collection in Dublin, Ireland, while another thirty leaves make up inventory number 6238 of the University of Michigan Papyrus Collection in Ann Arbor, Michigan. How these two collections came to possess separate portions of the same codex, and how all the different parts came to be discovered and eventually published together, is a convoluted story.
It is a sad fact that such an important and well preserved text as this should have fallen victim to the practices of early 20th century antiquities dealers. The fact that this papyrus was purchased on the antiquities market, rather than unearthed as part of a scientific excavation, presents many problems to modern scholars. Questions about the origin, use, and age of this papyrus are difficult to answer because of the lack of an archaeological context for this papyrus. For example, the stratigraphy of the site could have provided information useful for dating the papyrus. Also, the geographic location of the site, as well as the type of site (residence, monastery, e.g.) could have provided useful information about how early New Testament texts were used and circulated throughout Egypt.
While questions such as those above still linger today, there were even more questions to be asked in 1931 when the first fragments of this codex began to appear. At that time a batch of New Testament papyri, including ten leaves of P46, was acquired by Chester Beatty and published by Sir Frederick Kenyon of the British Museum. However, since a large portion of the contents was missing, Papyrus 2, as it was called, was not considered to be one of the best specimens from the group which Mr. Beatty had obtained.
Over the next few years, however, more leaves from the codex began to come to light, and today this codex is among the most important examples of early NT manuscripts. Unfortunately it took some time before the complete codex, which had been divided up and sold off by the dealers who found it, was finally reassembled. Following Kenyon's publication of the original ten leaves in 1934, it was discovered that the University of Michigan had thirty additional leaves of the same codex (six of which had been bought in 1931, and the remaining twenty-four in 1933). With this discovery, now forty leaves had been discovered, and it was hoped that soon the remaining leaves might be found. However, after two years with no new discoveries, Henry Sanders, a papyrologist at the University of Michigan, published the 30 Michigan leaves along with the 10 Beatty leaves already published. Only then, shortly after his publication, was it announced that Chester Beatty had managed to acquire 46 additional leaves from the same codex. These new leaves were published by Kenyon, along with the previously published leaves, in 1936. This publication turned out to be the last, and these 86 leaves are now all that survive from the original 104-leaf codex.
If the codex had not been mutilated by the dealers who sold it, this confusion surrounding its publication could have been avoided. Nevertheless, perhaps one can find a silver lining in this story; because the text now resides in separate collections on both sides of the Atlantic ocean, it can be enjoyed by a much wider audience of researchers and visitors.