Intraperitoneal injection has been described as the most sensitive way in which carcinogenicity of elongate mineral particles (EMPs) can be determined. Accordingly, a number of such animal experiments have been reported, but in general, the materials injected have not been appropriately characterized, and dosage has been measured in terms of mass, rather than the number of fibers. In one study conducted in 1991, the tremolite samples used were characterized by scanning electron microscopy (SEM). The resolution and image contrast that could be achieved by scanning electron microscopes in 1991 were limited, however, and in the light of more recent work, improved measurement accuracy for the thinnest EMPs would be desirable. In an animal study reported in 2014, characterization of the tremolite and anthophyllite used in the study was limited to optical microscopy. This study could not be fully interpreted without accurate measurement of the size distributions. The respirable fractions of the bulk materials used in both of these animal studies have been separated by water elutriation. The size distributions of the respirable fractions were measured by transmission electron microscopy (TEM) and these results were compared with the mesothelioma frequencies reported by the authors of the animal studies. The results show major discrepancies between the two studies. In particular, the numbers of tremolite EMPs injected that had dimensions associated with mesothelioma differed by a factor of 33.8 for the same biological outcome. The authors of the 2014 study suggested that the different strain of rat used in their study may explain discrepancies between their experiments and some of their earlier work on anthophyllite. The TEM results reported for the tremolite used in the 2014 study support this suggestion.