Although only a few examples of actual collapse can be blamed on temperature and shrinkage loadings, a concrete structure will crack and spall if it does not have freedom to modify its dimensions with temperature changes and with aging. Unfortunately, the dimension changes are not entirely linear. Rotational displacements result and seriously affect brittle masonry surfacing, windows and door frames. Although adding uniformly distributed and continuous reinforcement will help the concrete mass resist thermal changes, reinforcing will not prevent shrinkage. Actually, shrinkage is more likely to occur when larger amounts of reinforcement are used because the steel will not shrink and this causes internal bond failures. One of the reasons concrete cracks more frequently as it grows older may be because of expansion and contraction. A German publication stated that the average coefficient for expansion of dry concrete specimens 90 days old is .0000069 and for contraction is .0000056. If this can be substantiated, it would explain the tendency of concrete structures to grow, even where aggregate reactions are not a factor in volume change. A length of 400 feet would show shrinkage cracks of .31 inches as the residual of a 50 degree F. temperature fluctuation. Recent experience indicates that delayed shrinkage, at ages of about 2 years, is more critical with lightweight aggregate concrete than it is with normal weight concrete. This situation is usually attributed to the saturated state of the aggregate when it is proportioned in the mix. Attempts have been made to correct this condition by mixing the lightweight aggregates dry, but this results in a concrete that is "bony" and very difficult to properly place and finish. As a result, water is added to the mix in the field, with resultant low strengths and increased delayed shrinkages.