During valve formation of the siliceous frustules of diatoms, bulk uptake of silicic acid and its subsequent transport through the cell is required before it can be deposited in the silica deposition vesicle (SDV). It has been assumed that transport takes place via silicon transporters (SITs), but if that were the case a control mechanism would have to exist for stabilization of the large amounts of reactive silicon species during their passage through the cell on the way to the SDV. There is, however, no reason to assume that classical silica chemistry does not apply at elevated levels of silicic acid, and therefore autopolymerization could reasonably be expected to occur. In order to find alternative ways of Si transport that correspond with the high speed of valve formation at the earliest stages of cell division we followed 31Si(OH)4 uptake in synchronously dividing cells of the diatoms Coscinodiscus wailesii, Navicula pelliculosa, N. salinarum, and Pleurosira laevis. The results were related to systematically derived mathematical models for a compartmental analysis of 5 possible uptake/transport pathways, including one involving SITs and one involving (macro)pinocytosis-mediated uptake from the extracellular environment. Our study indicates that the uptake of radioactive silicic acid matches best with the model that describes macropinocytosis-mediated silicon uptake. This process is well in line with the observed ‘surge uptake’ at the start of valve formation when the demand for silicon is high; it infers that in diatoms a pathway of uptake and transport exists in which SITs are not involved.