The value of flow cytometry for quantifying unreduced (2N) pollen production in plants is well recognized; however, the approach has been limited by technical obstacles to obtaining high quality nuclei fluorescence histograms and the difficulty in distinguishing 2N nuclei from 1N doublets. Here, we use mathematical arguments and observations of fluorescence properties of angiosperm pollen nuclei to generate guidelines for applying pulse analysis to correct for doublets in pollen nuclei data. We show that the theoretical requirements for applying pulse analysis for doublet correction are met when nucleus fluorescence height and/or width measures in the unreduced gamete (2C DNA content) region exhibit bimodality (reflecting singlets and doublets) in combination with unimodal distributions of the same parameters in the reduced gamete (1C) region. These conditions are regularly met in the family Brassicaceae but not in the Asteraceae and Poaceae. We further show that when these requirements are met, pulse analysis estimates of doublet proportions are well correlated with estimates obtained with microscopy. We propose guidelines for doublet correction when estimating frequencies of unreduced male gametes. V C 2015 International Society for Advancement of Cytometry Key terms unreduced gametes; flow cytometry; pulse analysis; doublet discrimination; pollen POLYPLOIDY has played an important role in angiosperm evolution (1,2) and is believed to arise through the union of unreduced gametes (3-7). Despite the important role unreduced gamete production may play in plant evolution, there is much that we still do not know about it, including its frequency and distribution among individuals in natural populations and the factors that affect its occurrence (5,7,8). The limits on our knowledge are due in part to the time-consuming nature of measuring unreduced gamete production, which has traditionally involved microscopic examination of individual gametes or gametophytes or crossing experiments (6,7). More recently, high throughput methods have been developed that promise greater sampling intensity, at least for male gametes, including pollen volumetric measures (9) and flow cytometry.Since the first flow cytometric measures of DNA content of pollen nuclei, researchers have recognized the method could be used to estimate unreduced male gamete production (10,11). In fact, one of the main uses of pollen flow cytometry has been to qualitatively identify individuals with high unreduced gamete production (12). Attempts to quantify unreduced gamete production in pollen more precisely and to test statistically for small differences among individuals are less common (but see Refs. 10 and 13). Studies of this kind have been limited by the difficulties in obtaining good quality histograms with nuclei from large numbers of pollen grains, but recent methodological improvements have removed some of these