A new method for the separation of the constant and the variable fluorescence of chlorophyll a in in vivo.

The constant fluorescence of chlorophyll a in alga Dunaliella tertiolecta was estimated by a method of the least square regression applied to the linearized form of the equation y = axb. The value of the constant fluorescence obtained by this method was compared by the values estimated with simple linear extrapolation. Constant fluorescence, evaluated by the simple linear extrapolation of 10 ms of the initial variable fluorescence, was 50% higher than the value obtained by the method of the least square regression. We demonstrated that the estimation of constant fluorescence by the least square regression is a more correct method and provides a better comparison of results from different laboratories. This method offers a simple way to determine and separate constant fluorescence from variable fluorescence in the total yield of chlorophyll a fluorescence in "in vivo" conditions. Furthermore it facilitates the interpretation of the variable fluorescence phenomena in "in vivo".