I encourage readers to check out Charles E. Hill’s article at the Bible and Interpretation. Hill is a well respected expert on the canonization of the gospels as one can see from his publications on his faculty page and his recent book Who Wrote the Gospels: Probing the Great Gospel Conspiracy (Oxford University Press, 2010) (see various blog reviews, also here, here, here, short note here). I hope to carefully read through this book in the Fall when I have a bit more time, but I did find his article ”What Papias Said about John (and Luke): A ‘New’ Papian Fragment” (JTS 49 [1998]: 582-629) to be an impressive and well-argued case that Eusebius preserved Papias’ remarks in H.E. 3.24.5-13 though I myself am not quite persuaded that Papias knew a fourfold gospel canon. But one of his parenthetical asides caught my attention: “Judging only from the numbers, it already seems tenuous to claim that other Gospels were about as popular as Matthew, Luke, or John (Mark could be a different story), or that the competition between them was quite close” (italics added). In comparison to the other three canonical gospels, Mark was the odd one out in the canon. In fact, this is one reason why the source critical discussions are so important, as it was only the establishment of Markan priority as a viable solution to the Synoptic Problem by scholars such as C.H. Weiss (1838) and H.J. Holtzman (1863) that helped Mark emerge out of the shadows of the other gospels (especially Matthew) and led to a huge increase in scholarly interest as can be observed in the continuing stream of commentaries on Mark that have been written. However, the situation for Mark throughout most of church history has been one of neglect. This may be due to Mark’s brevity, the absences of the birth stories, much of the sayings material, resurrection appearances, or perceived doctrinal or literary/rhetorical deficiencies. An excellent book on this is by Brenda Deen Schildgen, Power and Prejudice: The Reception of the Gospel of Mark (Detroit: Wayne State University Press, 1999) (RBL reviews here) and here is a list of some of her observations on pp. 39-41, 50-52:
- We have no commentaries on Mark (though Jerome left 10 sermons on Mark) before Victor of Antioch in the 5th century who compiled one from the passing comments of Origen, Titus of Bostra, Theodore of Mopsuestia, Chrysostom and Cyril of Alexandria.*
- From the Biblia Patristica, up to Clement of Alexandria and Tertullian there are roughly1400 plus citations/allusions to Mark versus 2000 of John, 3300 of Luke and 3900 of Matthew. In the 3rd century, there are roughly 250 of Mark versus 3600 for Matthew, 1000 for Luke and 1600 for John. Origen references Mark roughly 650 times, versus almost 8000 of Matthew, 5000 of John and 3000 of Luke. In Augustine’s extant sermons, he has 250 references to Matthew, 170 to John, 150 to Luke and 15 to Mark.**
- Schildgen argues for some impact of Mark on lectionary readings (the Easter account in Mark 16 may have influence on western rites, Mark may have had some special status in Alexandria), but she still concludes that Mark was only read in the lectionary once per every 16 readings of John or Matthew
- She could have added that we have little early manuscript attestation – our oldest manuscript evidence is Chester Beatty papyri (p45) from the third century that has fragments of all 4 gospels, and then not until the fourth century in Sinaiticus and Vaticanus.
* Note: Adela Collins in her Hermeneia commentary on Mark argues that Origen might have compiled a commentary on Mark based on the way he exegetes Markan passages in his commentaries on Matthew and John (p. 105), but if he did there is no more trace of it.
** I counted 1463 references to Mark in the first volume of the Biblia Patristica (though many of these are concentrated in figures such as Irenaeus [237 times], Tertullian [347 times] or Clement of Alexandria [270 times]), 245 in the next volume and 646 in the volume on Origen. However, my eyes started to drift alot as I sat in a library trying to count all the references so I wouldn’t mind if someone wanted to double check my count 
. But one thing I noticed from a brief glance is that there was an extremely loose definition of a citation or allusion to Mark’s text and I think the numbers are actually much lower as many of them could be from shared material in the triple tradition (so from a Matthew or Luke parallel rather than Mark), a gospel harmony or continuing oral tradition. Other bloggers have also noted very different statistics: see Stephen Carlson or Peter M. Head (and comments from Hill below).
