A b s tra c t. -Optical emission line velocities in NGC 1365 are presented along 35 slit positions, of which 18 have not been published earlier. Diagrams of emission line velocities as function of distance along the slits are presented, and separate velocity fields are derived for the disc as given by the low excitation lines, combined with H i 21 cm VLA data, and for the nuclear outflow cone as given mainly by the high excitation lines. An algorithm for creating a velocity field and contour maps based on measurements along randomly oriented spectrograph slits is given in the Appendix. Detailed discussions and interpretations of these velocity fields are given elsewhere. The combined optical and 21 cm disc velocity field is available by an anonymous ftp.

No abstract

No abstract

LETTERS TO THE EDITORSThe Editors do not hold themselves responsible for opinions expressed by their correspondents. No notice is taken of anonymous communications. Radio Observations of the CometArend-Roland THE observation of radio emission from the Comet Arend-Roland has been reported 1 at 600 Mc./s. and 2 at 27 •6 Mc./s. The reported flux densities are of the order of 5 x 10-23 (private communication) and 5 X 10-22 W.m.-2 (c./s.

The theory of the Nolc birefringent filter is complicated and the general formulae describing its optical properties do not invite to any simple physical picture. Due to these formal difficulties, many of its inherent possibilities have not been fully appreciated. In this paper we point out some new possibilities for filters based on this general design.The detailed shape of the transmission profile is a function of the angles of the crystal plate optic axes. It is shown how unwanted transmission sidelobes can be suppressed to any desired level by altering the distribution of plate angles. By the same means, the transmission band can be split into two symmetrically placed replicas, and the distance between the two bands can be varied. In this way the filter can easily be tuned over half the free spectral range.Some of the error sources that are important to filter performance are discussed. An expression is given for the amount of parasitic light introduced by errors in the orientation of the plate optic axes and it is found that error s as large as 0.~ can be tolerated. A laboratory experiment with 16 birefringent plates has shown that accuracies an order of magnitude better than this figure can easily be achieved. Manufacturing errors in the thickness of the plates can be compensated for by assembling the pile of plates in a particular way. The very strict tolerances usually quoted can therefore be considerably relaxed.