Comparison of the Synthetic and Actual Distribution of Galaxies on a Photographic Plate.

The present work was undertaken with the following two questions in mind: (i) If it is true that galaxies occur only in clusters, then why is there such difficulty in identifying these clusters so that only a few of them are sufficiently conspicuous to be studied per se? (ii) Would a visual comparison between the actual photographic plate and the implications of the recent theory, by Neyman, Scott, and Shane, of clustering of galaxies provide indications for modifying the probabilistic model underlying this theory? In order to answer these questions, the specificity of the probabilistic model was used to constrnct a "synthetic" photographic plate. Following the postulates of the model a series of random sampling experiments was performed leading to the determination of positions of galaxies in a fiO X 60 square representing a photographic plate, without the images of stars. This synthetic plate is then compared with the chart of a real photographic plate with the images of stars removed. The general appearance of the two plates is very similar. In particular, the clusters of galaxies which compose the entire distribution on the synthetic plate appear superimposed and mixed together and are unidentifiable. This result answers question i. As to question il, the galaxies on the actual plate appear somewhat more tightly clustered than on the synthetic plate, and this visual impression is confirmed by an examination of the counts in 10' X 10' cells. If this difference is not due to chance variation, then it implies the necessity of a change in the specialization of the probabilistic model, presumably affecting the details of the hypothesis regarding the internal strncture of clusters.