The nucleation processes in waterborne Maleic Anhydride-grafted-Polypropylene micro- and nano-droplet suspensions have been studied. Compared to a previous report on this topic, an extended set of samples in combination with improved particle size distribution data of the samples have been used, which are both essential for the advancement of the analysis. Self-nucleation was utilized to ensure that the observed lowered fractionated crystallization (peak) temperatures - down to the extremely low value of 34 degrees C - are due to a lack of seeds in the droplets, which seeds for the polypropylene system used are normally active at the heterogeneous crystallization temperature of approximately 110 degrees C. An unusual self-nucleation behavior was observed in case of samples having a large amount of small droplets, requiring an extremely low self-nucleation temperature in order to suppress all crystallization at the lowest temperatures. Such behavior was observed for block copolymers but has not been reported so far for droplets dispersed in an immiscible matrix, polymeric or not. Another unusual behavior was observed for some self-nucleation temperatures for which apparently two different populations of self-nuclei are created that are suggestive of the alpha(1) and alpha(2) crystal structures of isotactic polypropylene. Next, two new methods are presented to quantify various crucial parameters of the nucleation process: one estimates the density of nucleants acting at different temperatures from the combination of dynamic DSC data and particle size distribution (PSD) data, and the other one focuses on the nature of the nucleation mechanism using both isothermal DSC data and PSD data, quantifying the nucleation rate at different temperatures. For the present MA-g-PP dispersions the latter method leads to the conclusion that the lowest crystallization temperatures reflect sporadic nucleation, probably by way of volume (homogeneous) nucleation. In the field of polymer crystallization, polymer dispersions are usually treated as being monodisperse, even though that is rarely the case. This simplification is inadequate for the present calculations, which is why polydispersity has been taken into account in order to quantify the density of nucleants and the kinetics of nucleation. Though in the present study DSC data are used for the calculations, the methods developed can be easily adapted to other techniques like time-resolved X-ray, rheometry and dilatometry. (C) 2009 Elsevier Ltd. All rights reserved.