ABSTRACT
There is a property of the least squares adjustment called the principal property. It was first formulated by J. M. Tienstra. It states that each adjustment can be subdivided into phases where the corrected observations of a particular phase accompanied by their derived weight coefficients are used as "quasi" new or as indirect observations to be adjusted in the subsequent phase. There are however many special cases of elementary survey adjustments, which are subdivided into phases, and where the adjusted observations of a preceding phase are not correlated, and may even have equal precision. An example of this is the use of the mean value of a number of observations of an angle or distance in further adjustments. The fact that the weighted mean of observations is the result of a least squares adjustment is often ignored.
