One thing to add is that there is still plenty of debate about whether particulate quantity or mass is the most important parameter. A large number of small particles has no correspondence in total mass to a smaller number of larger particle distribution. From a toxicology standpoint, which is more relevant -- quantity inhaled into the body or mass? -- is, I believe, still mostly an open question.
For our purposes we just need to be aware that when we see graphs like that which wxman posted above, that we interpret it with an appropriate understanding that there is another side to the story. Also, I must comment that the comparison of PM emissions of a gasser engine against a DPF-equipped Diesel naturally puts the latter in a particularly beneficial light. Since the number of DPF-equipped Diesels globally is dwarfed by the cumulative number of non-DPF equipped Diesels that are operating, it would be enlightening to readers to see the distribution for one example.