J. Cell Sci. 14, 551-559 (i974) 551 Printed in Great Britain THE HANDS OF HELICAL LATTICES IN FLAGELLAR DOUBLET MICROTUBULES R. W. LINCK* AND LINDA A. AMOS MRC Laboratory of Molecular Biology, Hills Road, Cambridge CBz 2QH, U.K. SUMMARY Negatively stained arrays of flagellar doublet microtubules have been tilted about an axis perpendicular to the tubule axes and imaged at angles of ± 20°. The images have been filtered to enhance the 4-nm periodicities in the tubulin lattice, and the resulting filtered images interpreted by comparison with simulated projections of tilted helical lattices. The different appearances of the edges of filtered doublet tubules for positive and negative tilt angles suggest that the 4-nm lattices of both subfibres are of the same hand, with the 3-start helices in each being left-handed. The information has been incorporated in a model proposed for the arrange- ment of subunits in doublet microtubules, which is described in the preceding paper. INTRODUCTION The helical surface lattice of tubulin subunits in both A- and B-subfibres of flagellar outer doublet tubules has been determined from electron micrographs of negatively stained specimens, as described in the preceding paper (Amos & Klug, 1974). It was not possible to determine the absolute hand of the helical arrangement in either sub- fibre from the data used in that study, nor even the relative hands of the 2 subfibres. The latter piece of information is particularly important since it is required for the interpretation of the filtered images of doublet tubules, such as those shown in fig. 11 of the preceding paper. Without knowledge of the relative hands one does not know whether each filtered image represents the same side (near or far) of both sub- fibres or the near side of one subfibre juxtaposed alongside the far side of the other subfibre. The absolute hand of a helical structure seen in the electron microscope cannot be found from a single view taken at right angles to the helix axis. Since the electron- microscope image is effectively a plane projection of the entire structure, the near and far sides of the image cannot be distinguished. It can only be done by tilting the particle in a known way and observing the relative movement of the lattices of sub- units on the 2 sides. The simplest cases are for tilts about axes perpendicular and parallel to the helix axis. The results of tilting about an axis at right angles to the helix appear to be generally more successful than those of tilting about the helix axis itself (see Finch, 1972). This is probably because the changes in appearance are more dis- tinct and do not repeat as they do for rotation about the helix axis. Also a single view * Present address: Department of Anatomy, Harvard Medical School, Boston, Massachu- setts 02115, U.S.A.