15 September 1995 OPTICS COMMUNICATIONS ELSEVIER Optics Communications 119 ( 1995) 693-{}95 Comment Fundamental reduction of the observation volume in far-field light microscopy by detection orthogonal to the illumination axis: Confocal theta microscopy (Optics Comm. 111 (1994) 536) C.J.R. Sheppard Physical Optics Department, School of Physics, University of Sydney, NSW 2006, Australia Received 30 December 1994; revised version received 4 April 1995 Abstract In the paper of Stelzer and Lindek on confocal theta microscopy they have assumed identical numerical apertures for both con focal and con focal theta microscopy. In practice a higher numerical aperture can be used in ordinary confocal microscopy, in which case the relative performance is reversed. In a recent paper Stelzer and Lindek [1] described the effects on the resolution of confocal fluorescence microscopy of off-setting in angle the axis of the detec- tion lens. They claim that substantial improvement can result. Their interesting paper is exhaustive in its con- sideration of various different microscope geometries. Advantages of the off-set geometry include an almost spherically-symmetric point spread function, and a use- ful combination of long working distance and resolu- tion. The practice of using an angularly off-set detection lens is well known in a number of areas, notably in cytometry, Raman spectroscopy and luminescence microscopy. A particular advantage of the arrangement in these cases is that the incident beam does not pass into the collection lens, thus alleviating filtering prob- lems. An example of a confocal system with off-set detection lenses was illustrated by Wilson and Shep- pard [2, p. 116]. A limitation of confocal theta micros- copy is the geometrical problem of fitting in two high aperture microscope objectives. Thus in practice Stel- zer and Lindek have used water immersion objectives 0030-4018/95/$09.50 © 1995 Elsevier Science B.Y. All rights reserved SSDI0030-4018(95)00367-3 of 0.75 NA, and even then they have to be off-set slightly from the orthogonal arrangement to fit them in [3]. However, in their paper they choose to compare the various arrangements with the same numerical aper- ture lenses in each case, whilst in practice in confocal (non-theta) microscopy it is possible to use commercial water immersion objectives of 1.22 NA. In practice it should be feasible to design water immer- sion objectives of 1.26 NA. We thus consider here imaging in the confocal theta arrangement of 0.75 NA and compare it with the stan- dard confocal arrangement of 1.22 NA. For simplicity we assume a non-vector "pseudo-paraxial" theory [4] for imaging, which has been shown to give reasonably accurate results for high numerical apertures. Accord- ing to this theory the intensity in the focal plane of a lens is given by the standard Airy disc, ( 211 (km Sina))2 l(r) = , kmsina (I)