“…We also consider the assumption of crystal defects resulting in a breakdown of the l = 2 n extinction as less probable as it has never been observed by SR diffraction experiments on a range of dragline silks. , The presence of the superlattice peak is therefore also difficult to reconcile with a different space group or a superstructure due to a repetitive sequence of other residues, such as serine residues in silk fibroin. ,, This also holds for the assumption of a more or less random distribution of β-sheet nuclei 34 in the amorphous matrix, which is incompatible with the meridional position of the superlattice reflection. We note, however, that the azimuthal width of the superlattice reflection corresponds to the broadening of the oriented equatorial halo due to short-range order as seen by WAXS, which can be separated from the 210 reflection by profile fits. , The azimuthal widths of both WAXS peaks show considerable variability during artificial drawing of dragline silk, which can be related to variations in drawing stress and possibly mechanical action within the spinnerets. , Dragline protein chain-folding models suggest an aggregation of alanine sequences through a backfolding of β-strands via serine−glycine or glycine−asparagine “hairpin loops”. ,,, We tentatively assume that the glycine-rich chains, which do not participate in the formation of a polyalanine block, are constrained into a smectic β-sheet structure, which is laterally disordered due to a mismatch of side groups. A larger orientation distribution of chain axes in the smectic phase would explain the increased azimuthal width of the superlattice reflection as compared to the crystalline domains 002 reflection.…”