No, it is not a baby chicken. It's a common netspinner pupa, Hydropsyche/Ceratopsyche sp. And it's been so long that I've seen one that I wasn't sure what it was until I got close enough to take pictures. We shouldn't be surprised. We see the larvae all winter long, and I guess it's time for them to hatch. This one had just broken its way out of its pupal dome and was -- I assume -- on its way to the surface.
Like all "retreat-making" caddisflies, the common netspinner builds a dome out of pebbles when it's ready to pupate. The cocoon is encased in the dome which is then attached to a rock. The process is described in detail -- with illustrations -- by Wiggins on pp. 62-64 of his wonderful book, Caddisflies: The Underwater Architects (Toronto, 2004). (Also, see the entry posted on 12/24/11.)
But how do we establish the genus ID? The only source I have for making pupa ID is the comprehensive An Introduction to the Aquatic Insects of North America (4th edition, 2008), which is edited by R.W. Merritt, K.W. Cummins, and M.B. Berg. The Hydropsychidae pupa is described on p. 460, and Hydropsyche sp. is illustrated on p. 467. The key to the ID is the number of "hook plates" on the abdominal segments (dorsal view). I quote: "Abdominal segments III and IV each with two pairs of hook plates; segment V with only one pair of hook plates." And that's what we've got.
How about the species? I have no way of knowing for sure. But, I've only seen 3 netspinner species in this river -- the Rapidan -- Ceratopsyche/Hydropsyche sparna, Ceratopsyche/Hydropsyche alhedra, and Ceratopsyche/Hydropsyche slossonae.
Not the only strange thing that I found yesterday. I collected a lot of these unwelcomed diptera: Tabanidae, horse flies!