I agree with Leigh. The distribution you want is the Poisson distribution.
And I also agree with Leigh that you need to count for a longer time.
To further explain the second point, note that your data, as taken, appear
skewed toward the low counts. You had very high incidence of one count and
high incidence of no counts in 0.5 second. That means you would also expect
reasonably high incidence of one count and no count for a period of one
second, or a period of two seconds, etc. So your method of data acquisition
means you aren't seeing the "whole story." You have "sampled the
distribution" fairly well for the right side of the tail, but you have
insufficient sampling for the left side of the tail.
Michael D. Edmiston, Ph.D. Phone/voice-mail: 419-358-3270
Professor of Chemistry & Physics FAX: 419-358-3323
Chairman, Science Department E-Mail edmiston@bluffton.edu
Bluffton College
280 West College Avenue
Bluffton, OH 45817