I once had a publication in Nature Genetics together with 78 other authors (I was on position 59, but the author list was in strict alphabetic order). The paper with me as the first author and largest number of co-authors was in Mammalian Genome (with 37 colleagues piggy-bagging on my research). I always thought that it is too difficult to handle a single publication with more than 5 people, each of them want to leave in their individual trace. But now I saw a single paper in the European Physical Journal that has been written by 2891 authors (all belonging to the ATLAS consortium). The paper with the title "Search for dark matter in events with heavy quarks and missing transverse momentum in p+p+ collisions with the ATLAS detector" is 22 pages long.Of these 22 pages, 10 are scientific text, and the remaining 12 pages just contain the list of authors and their affiliation. I feel a bit nostalgia for Tolstoi or Goethe: One single author, who wrote more then 1000 pages in one book.
Anyway, I like the Feynman diagrams on page 2 of the paper, they remind me of the schemes of foot pathes one see in mountain maps.