The thickness of spun films is often represented as inversely proportional to some power of the spin speed. For most materials, this power is found to be one-half. However, polyamic acid solutions show a much stronger dependence, with exponents ranging from about 0.5 to almost 1. A simple model of spin-coating is used to explain this difference. Using coating data for both polyamic acid and photoresist solutions, it is shown that in many experiments the coating process has not reached steady state. When steady state is attained, the speed-power is one-half; at shorter times, the power is greater. For some materials, the time required to reach steady state may be 10 min or more.