We recently reported that catalytic amounts of F- cause rapid rearrangement of silsesquioxane (SQ) T-8 cages in THF at room temperature. These rearrangements lead primarily to the T-10 and T-12 cages as long as the F- is trapped for example using CaCl2 to form insoluble CaF2 in situ. In this report, we use this approach to make di- and triaminophenyl, phenyl silsesquioxane T-10 and T-12 mixtures: (NH2Ph)(2x)Ph10/12-x(SiO1.5)(10/12). Thereafter, we isolate the fraction that consists primarily of x = 2-3 via column chromatography. We then explore the reaction of this fraction (fraction 3) with the diglycidyl ether of bisphenol A (DGEBA) to form a soluble epoxy resin wherein the cage SQs are contained in the chain backbone. We call these products beads on a chain (BoC) polymers. The use of 1:1 mole ratios of [epoxide]:[NH2Ph] of DGEBA and (NH2Ph)(2x)Ph10/12-x(SiO1.5)(10/12) leads to insoluble products. However, the reaction of 0.8:1 mole ratio of [epoxide]:[NH2Ph] provides a soluble but bimodal distribution of products (after 24 h reaction at 90 degrees C in THF). The low MW component is much more soluble in THF than the higher MW component, allowing simple separation. The low MW polymer has a M-n of 6000 Da with a PDI of 1.6. The higher MW material has an M-n of 21 000 000 Da with a PDI of 21. Both components offer similar ceramic yields (to SiO2) of 32 +/- 1%, indicating that the relative contents of cages are identical. They also exhibit very similar FTIR spectra, suggesting that they are the same material. Two explanations are possible for the bimodal size distribution. One is that the positioning of the NH2 groups on the cage leads to cyclic trimers given that the average monomer unit will mass approximate to 1700-1900 Da, which represents the low MW fraction. Alternately and more likely, the higher molecular weight component consists of di- and triamino functionalities such that in the growing chain, once the first two amine groups on the cage react, the third group is likely sterically protected. Thus, only late in the reaction will the quantities of the third group exceed those of the diamino components, allowing a slow tertiary functionalization and likely some degree of cross-linking (or internal cyclization) to occur. That this happens is suggested by the much poorer solubility of the higher MW materials as well as the very large PDI. These studies serve as a model for other possible studies targeting BoC polymers.