The nearly constant value of the Raman peak corroborates our assumption about the tube sliding within the bundle. The paper by Salvetat et al., [12] in which SWNT bundles are loaded with an AFM, also supports this hypothesis. As the tube bundles grow in diameter, both the axial modulus and shear modulus decrease dramatically, suggesting that individual SWNTs are slipping within the bundle and decreasing the load required to deform the bundle. [12] We stress that it is the low-modulus features of the bundles, and not the axial modulus of individual tubes, that control the mechanical stability and strength of the SWNT±polymer composite. Our experiments show strong evidence to this effect. When nanotube applications as reinforcements are considered, it is important to consider the most effective ways of obtaining strengthening: for example, by breaking bundles into individual tube fragments (say, fullerene pipes [14] ) and dispersing these segments in the matrix. On the other hand, strengthening could also come from reinforcing the bundles themselves by crosslinking the tubes within bundles (e.g., via irradiation or chemical treatments) to increase the bundle rigidity and eliminate tube slippage. Even if this is achieved, one will have to worry about a strong tube±polymer interface. The formation of strong interfaces in carbon nanotube±polymer composites could be challenging but could be done through functionalization of the tube fragment ends and tube body, which then can be chemically bonded to the polymer chains.Nanotube reinforcements, however, will increase the toughness of the composites by absorbing energy because of their highly flexible elastic behavior during loading. Other interesting applications could be in creating nanotube-filled adhesive polymer films where adhesion between nanotube-filled polymer surfaces could be greatly enhanced by some kind of reversible Velcro effect at the nanoscale, due to the attractive van der Waals forces between the nanotubes present on each of the surfaces, as was observed in Figure 1a. As we stated earlier in our work on MWNTs, the low density of the nanotubes will clearly be an advantage for composites. Nanotubes would also offer multifunctionality; for example, conducting polymer structures may be constructed at low loadings of nanotube fillers due to lower percolation thresholds needed for the high aspect ratio nanotube structures. Nature 1999, 399, 761. [11] The shear modulus of an isolated SWNT is much larger and can be related to the elastic constants of graphite as 0.5 (C 11 ±C 12 ) = 440 GPa. This is not the magnitude relevant for the present experimental studies. In the case of the nanotube rope, its shear modulus is related to that of graphite by the ratio of the distance between tube±tube layers divided by the graphitic interplanar distance. This comes out as approximately, G rope~5 G graphite (see [12] for details of this relationship), suggesting that for a nanotube rope (assuming disorder) the lower limit for G will be~1 GPa.[12] J. , Phys. Rev. Lett. 19...