As three of you immediately identified in your comments, last week’s crocodile mystery was the fastening in the center of a volvelle, holding the various layers in place and allowing them to turn:
Volvelles are paper wheels that are fastened to a leaf so that the discs spin independently. Some of the earliest volvelles were used for prognostication; Ramon Llull is credited with bringing the volvelle to the West in the late thirteenth century for use in his Ars Magna. Suzanne Karr describes the system of two discs of letters on top of a third layer as one that did not simply aid memory but produced new knowledge: “When properly used, the triple layer of combinations of nine letters . . . answered questions about all creation and even the future, as well as inquiries intended to settle religious debates.” ((Suzanne Karr, “Constructions Both Sacred and Profane: Serpents, Angels, and Pointing Fingers in Renaissance Books with Moving Parts” Yale University Library Gazette 78:3/4 (April 2004), p. 103. http://www.jstor.org/stable/40859568)) Although volvelles could also be used as memory aids, it is their calculating capacity that has been the most influential, and Llull is often seen as a forefather of ars combinatoria and computing. ((For a quick introduction to Llull and his volvelles, see Brooke Palmieri’s post, “An Introduction to Paper Computing.”))
The volvelle is probably most familiar to us as a scientific instrument and, specifically, as a device used in astronomical calculations. The source of the volvelle I featured falls into that category: Martín Cortés’s Breve compendio de la sphera y de la aite de navegar, printed in Seville in 1551, was one of the foundational texts for oceanic exploration. This volvelle is intended to help the user determine the position of the sun and the moon—turn the roundels so that their indexes (one for the sun and one for the moon) point to the appropriate days and the resulting information will provide their locations.
Breve compendio features six volvelles of varying complexity (and beauty). The other five, in the order in which they appear in the book, are below:
You might notice that some of them look a bit odd. The third one (leaf 73r) seems to be incomplete and the index doesn’t spin, and some of them (especially the last one, leaf 113v) look as if the roundel might be a facsimile. I haven’t closely investigated them to see if that’s the case (one might seem some clues in a close examination of the paper). One of the things you might notice, too, is that the fastening is different—only the crocodile has the button fastener, while the fourth image above (leaf 83v) stands out for not being sewn in the same manner as the others.
It wouldn’t be surprising if some of those featured replacement parts. Volvelles, like any other moving part of a book, are susceptible to wear and tear; in some cases, they are removed and repurposed. The following two images show examples from 1584 and the 1596 editions of the English translation of this work, one with the volvelle and one without (click on the images to enlarge in a new window).
Personally, I find it amazing when any volvelles survive—they can be fragile and they’re designed to be spun, both conditions that make them likely to be used up. If you’d like to find out some more about volvelles, a good place to start is with their entry in Architectures of the Book; there’s also a partial bibliography of extant volvelles; you can also browse some more images of volvelles in Jim Kuhn’s media group on Luna. Finally, if you’d like to think more about how to connect volvelles with other forms of cut-ups and digital texts, Whitney Trettien’s “Computures, Cut-ups, and Combinatory Volvelles: An Archaeology of Text-Generating Mechanisms” explores the subject in an appropriately combinatory way. ((And many thanks to Whitney and to Chad Black for help in researching volvelles and in translating Cortés’s Spanish!))