Mark,
To be a bit contrarian: this looks like bad science to me. I think it is part assumption error, and part a relic of cavity super-radiant emission (Dicke-Preparata) by nanotubes, which emission is partially focused on a good blackbody emitter. If fact most of the effect could be measurement error due to the way the nanotubes are laid down - with a preferential vector for emission (the open end of the tubes) combined with the lack of appreciation for the known nanotube anomaly with Kirchoff laws. In fact most of the error probably derives from the assumption that temperature measurement of the carbon side would be a blackbody spectra, when it is well known that Kirchoff's laws are not only violated by nanotube emission, but almost irrelevant. Plus cavity radiation - if a preferred vector is provided, can transmit photonic radiation as if it were semi-coherent (super-radiant). Since the authors never mention Dicke, or super-radiance, or cavity blackbody emission, or the known violation of Kirchoff (at least in the PR blip) this seems more like poorly done research and a premature announcement of what is already known - than good science. From: Mark Iverson FYI: http://phys.org/news/2012-04-carbon-nanotubes-weird-world-remote.html "This is a new phenomenon we're observing, exclusively at the nanoscale, and it is completely contrary to our intuition and knowledge of Joule heating at larger scales-for example, in things like your toaster," says first author Kamal Baloch,