The Sordes problem. Much ado has been made about the Sordes holotype with regard to pterosaur wing shape. On the left it appears as though some membrane makes a straight line toward the left ankle (Sharov 1971; Unwin and Bakhurina 1994; Elgin, Hone and Frey 2011). Nevermind that the line keeps going off to the right. Peters (2002) I thought that a geological fault was "at fault" for creating such a straight line on both sides of the leg. That's not the entire answer. The key to solving this problem was making sure all the other elements were present. Most were. Some were not.
What looked like the forearm was not. It was manual 4.1 bent toward the elbow exactly as if it was the forearm. The real antebrachium (radius + ulna) was missing on the left...or so it seemed. The first wing phalanx was bent in unnaturally and everyone thought the skull was damaged. Apparently it is largely missing too. In its place, or rather on top of it, is the left hand with metacarpal IV separated from I-III and the wrist unnaturally close to the elbow. The radius and ulna drifted posteriorly, between the wing and the leg. The ulna forms the straight line from the wing to the ankle and the radius crosses it. That's why Sordes seems to have a different wing shape than all the other pterosaurs, which you can see samples of here.
Note that the right wing is intact and the trailing edge extends to the elbow, not the ankle. The right iinner wing was torn from its mastings, folding much like one might expect a uropatagium to fold.
Everyone thought Sordes had uropatagia (or a single uropatagium!) much deeper than any other pterosaur. In fact the uropatagia were shallow and therefore were quite similar to those of other pterosaurs, as shown above in red. Other tissues were caught on the feet and ankles in the same current or event that dislodged the left antebrachium before burial. Those have never been otherwise explained.
This is also a fine example of wing ungual retention.
Unwin and Bakhurina (1994) also errored when they postulated that pedal digit V was involved with and controlled the uropatagium between them. There are no sister taxa that confirm this configuration and it would involve an untenable extension of skin from the tip of one lateral toe to the tip of the other lateral toe. This might have made more sense if the feet were oriented backward--but they were not. Again, this illusion was caused by the portion of the torn and dislodged left wing that came to rest across the ankles along with the ulna and radius. The cartoonish drawing presented by Unwin and Bakhurina (1994) shows a lack of attention to detail, despite the presence of the specimen in their lab. It has often been said that the so-called Photoshop method is intrinsically inferior to seeing the specimen first-hand. Well, not in this case, nor in Jeholopterus, the IVPP embryo and several other pterosaurs and other reptiles, like Vancleavea.