Although recent experiments on pattern formation during amphibian limb regeneration have relied on the use of triploid marked cells, the purity of the triploid donor has not been proven. To test the purity of the donor, blood (which contains noncycling nucleated red blood cells) was collected from the gills of 12 pressure-induced triploid axolotls from two unrelated spawnings. Blood was also collected from 12 untreated sibling animals to serve as diploid controls. The blood samples were placed into a propidium iodide stain solution that had been adjusted to amphibian tonicity by the addition of distilled water. Samples were analyzed on a n EPICS V (Coulter) flow cytometer. A minimum of 20,000 cells were analyzed in each sample at a flow rate that did not exceed 1,000 celldsecond. In the diploid group, 88.6% of the cells were collected in channels of the DNA histogram corresponding to a 2C Gl/Go DNA content (channel 71-r 10) and 0.9% were collected in channels corresponding to a 3C Gl/Go DNA content (channel 106 * 10). In the triploid group, 99.2% of the cells were collected in nondiploid channels and only 0.8% were collected in channel 71 f 10. Utilizing the average peak channel number for the diploid group of 12 animals an average haploid channel number of 35 was calculated. The triploid group of animals exhibited DNA ploidy that averaged one haploid channel number above that of the diploid group. Based on forward angle light scatter, nuclei obtained from triploid red blood cells were observed to be larger than nuclei from diploid blood. Preliminary data are presented which demonstrate that the ploidy characteristics of solid tissues (brain, cornea, spleen, muscle, and limb blastemal mesenchyme) agree with those presented for blood. veloped a n experimental protocol and a Ronald K. Charlton's present address is