Eclipses or maximum brightness are predicted by published 'light elements' or ephemerides. These comprise the epoch or time the first of these features was observed, together with the period taken for one cycle. The form is as shown in the illustration. Often the formula concludes with E, which merely indicates that the period must be multilied by the number of cycles since the epoch date.

Another term arising from the epoch and period of a star is O-C which stands for Observed-Computed. This is used to compare what is happening to the period of a star. An example is shown on the Cepheids page where the observed epochs of maximum of l Carinae now depart quite markedly from where they should be if the old ephemeris was still valid. It shows a change of period instead.

Similar light curves can be determined for Cepheids by observation. These pulsating stars are brightest just after minimum radius, when their surface temperature is at its greatest.

In order to use these colours effectively they must be transformed to the standard system - which describes how the star would look outside the atmosphere. This entails the removal of effects introduced by the observing system of telescope, filters and the detector. As well, there are more general deviations introduced by the atmosphere above the observer. These latter fall into two types, the primary extinction caused by the opacity of the air through which the observer is measuring the object, and a secondary extinction which causes a change in the colour of the object, invariably making it look a little redder than it really is. These are discussed in the section dealing with transformation to the standard system.