Ever since the existence of two types of herpes simplex virus (HSV) was brought to light [12,14,15], a number of biological markers for the type differentiation have been presented [7, 9-11, 13, 16-19]. However, most of the criteria thus far reported on which to differentiate the two types have been based on quantitative rather than qualitative differences, except the presence of tubular structures in type 2 herpes simplex virus (HSV-2)-infected cells [1] and density of viral DNA [3,6]. We have searched for a simple routine procedure to determine the types of HSV isolates and found that the chick embryo (CE) cell plaque marker [2,8] was suited for this purpose [5]. Yet this marker also measured a quantitative difference and showed a disadvantage that a strain experiencing serial egg or CE cell passages might possibly be mistyped. Aside from this, Wentworth and Zablotney [21] indicated some discrepancy between the type-dependent antigenic makeups and the designation of types based on the CE cell marker.While looking for a more qualitative difference between the two types of HSV, it was discovered that, in CE cells covered by a fluid overlay without antiserum, HSV-2 infection as evidenced by detachment of cells did not spread diffusely in the cell sheet but occurred within a limited area, perhaps because the main transmission took place via cell-to-cell spread. As a result plaques could be visualized in monolayers given an appropriate dilution of the seed virus. This was in contrast to type 1 herpes simplex virus (HSV-1), whose infection of CE cells either was absent or, when present, occurred in quite a different mode. Usually only the undiluted seed showed occurrence of a few plaques under the agar overlay in the case of unadapted virus, but even in such a case they showed a diffusely spreading cytopathic effect under the antiserum-free fluid overlay during the incubation. No similar typedependent difference in plaque morphology could be seen in any other cells tested.The following cell types were used: primary CE, L929 (a clone of mouse L cells), Vero and human embryo lung (HEL) cells. HEL cells were donated by Dr. Hondo, Department of Virus Infection, Institute of Medical Science, University of Tokyo. The technique of plaque titration was essentially the same as described by Taniguchi and Yoshino [20]. Monolayer sheets formed with the above cells in 60-mm glass dishes were washed once with phosphate-buffered saline and inoculated with serial decimal dilutions of HSV in the amount of 0.05 ml/dish. After incubation at 37 C for 1 hr, the dishes were divided into two groups. One group was overlaid with agar medium and the other with the same medium from which agar was omitted. All dishes were then held at 37 C in 5% CO2-air atmosphere for 4 days. In the agar overlay group, a second overlaying with neutral red agar was performed, and after a further incubation at 37 C for 1 day observed for plaque morphology. In the liquid overlay group, the medium was decanted and 70% ethanol was added. After being left standing ...