An important mode of hydration of Ordinary Portland cement (OPC) is via the formation of a shell of hydration products around cement grains. It has been shown so far that shells form not only in OPC but also around C3A grains in a mixture of alite, C3A and CS and in more restricted form in pure alite. The shells are believed to be hollow and limited information of the chemistry of the phases that constitute the shells is known so far. From the work done in this thesis some crucial knowledge has been gained on this subject. It is shown that shells are not empty but filled with C-S-H which appears to have a lower density than the rest of the hydration products. At early ages, in all observed instances shells are seen to form only around polyphase grains. Further, the shells form only around the silicate part of the grain at all observed ages. The cement grains are seen to have uneven reactivity and the hydration seems to follow a reaction front, leaving striations ∼ 1 µm on the grains. The Ca:Si ratio of all the C-S-H in the matrix is almost the same (about 2) for all ages. The work done on novel mix binders involving OPC with minor additions of CAC and CS shows that the chemistry of the hydration products remain almost the same with these additions. The amount of expansion in the mixes depends on the added sulfate content, however, the effects of the time of demoulding of the mixes on the expansion are more drastic. This leaves many open questions concerning the phenomenon of expansion in cements. In expanded heat cured Portland cement pastes, fine ettringite needles are seen to grow from an adjacent monosulfate plate. The work seems to confirm the theory of crystallization pressures and their destructive nature.