Istiodactylus latidens was first described in 1901 by Harry Govier Seely, under the name Ornithodesmus latidens, based upon material from the Early Cretaceous of southern England (this was changed after it was realized that the name Istiodactylus had already been assigned to a small Therapod Dinosaur). A reconstruction of the Pterodactyl was made by Reginald Hooley in 1913, which has served as the model for our understanding of the animal till the present; other reconstructions have been made in the intervening time, but all have essentially followed the pattern set by Hooley.
Reginald Hooley's 1913 interpretation of Istiodactylus latidens. Witton (2012).
This reconstruction of a weekly muscled, long snouted Pterodactyl with razor-like teeth has been hard to interpret. Three theories have been put forward, but none seems to completely fit with the available data. I has been suggested that Istiodactylus latidens could have been a dabbling filter-feeder, similar to a modern duck, a feeding technique that requires little strength, but ducks have short broad bills, whereas I. latidens has a long thin one, and no duck has razor sharp teeth. Alternatively I. latidens could have been a fish eater. Long thin shouts are good for catching fish, and do not necessarily need to be well muscled, but this hunting strategy normally requires pointed teeth, ideally curving backwards within the jaw, to catch and hold fish; razor blades would not be very useful, as they could hurt a fish, but not necessarily hold it. A third theory os that I. latidens could have been a carrion eater, as razor blade shaped teeth are good for cutting meat, but this requires a degree of physical strength that I. latidens apparently lacked.
In a paper published in the journal PLoS One on 21 March 2012, Mark Witton of the School of Earth and Environmental Sciences at the University of Portsmouth re-examines the original Istiodactylus latidens material to construct a new model of the skull, and is able to draw new conclusions about the animal's lifestyle from this.
Witton found that where Hooley had interpreted a gap in the centre part of the fossil as evidence for an elongate skull, the 'missing' piece of the skull was present; Hooley even described this, though he did not use it in his reconstruction. By reconstructing the skull with the missing piece in place, Witton came up with a shorter, stronger skull.
Istiodactylus latidens. (A-D) The fossil material. (E) Witton's reconstruction. Witton (2012).
Witton's shorter interpretation of the skull gives a stronger muscle attachment, at least for a shearing action. This strongly supports the carrion-feeding theory, as carrion feeders need to be able to cut through and ingest meat quickly (before anything else gets it) but do not need to deal with struggling prey, as with a carnivore. Witton also noted that Istiodactylus latidens lacked binocular vision, with eyes on the side of its head. This is strongly associated with carrion feeding in modern birds; carrion feeders need to be able to scan a lot of ground visually, but do not need to look for prey actively hiding or strike at prey trying to escape, both of which require good binocular vision. Based upon this Witton concludes that I. latidens was almost certainly a carrion feeder.
Reconstruction of Istiodactylus latidens as a carrion feeder. Witton (2012).