Micro-particle formation in low pressure silane (SiH4) plasmas has been of technical interest and concern for at least 40 years. Negative ion plasma chemistry is a candidate for the initial nucleation, which has been extensively studied both experimentally and theoretically. Nevertheless, the nature and mechanisms of anion-neutral reactions remain uncertain in models of nucleation. In the pure silane, 0.1 mbar plasma of this work, silicon hydride anion mass spectra are interpreted in terms of iterative anion–silane reactions, involving release of one, or two, hydrogen molecules. Such 'multiple dehydrogenation' has been directly observed by Operti et al [2006 Rapid Commun. Mass Spectrom. 20 2696], where triple dehydrogenation also occurred. A statistical model, defined uniquely by a branching ratio of 68% for single hydrogenation, and 32% for double dehydrogenation, gives an accurate fit to the hydrogen distribution for all heavy silicon hydride anions (abbreviated here by 'silanions'), up to at least 800 atomic mass units. Most theoretical models of silanion-neutral nucleation consider dehydrogenation where no more than one H2 molecule is released. It is suggested that multiple dehydrogenation could improve the comparison of theory with experiments.