By convention the angles on base and pinacoids are (+) when the direction of extinction has apparently moved as the hands of a watch, with reference to the upper right hand edge (between base and pinacoid) of the crystal. When the reverse is true the angles are (−), see Fig. 73. (2) The statistical method of _Michel Lévy_ and others is often applicable, especially in the following case: Sections at right angles to the brachy pinacoid (010) and hence showing _Albite_ twinning.—These sections, as nearly perpendicular to the lamellæ as possible, are known by the sharp dividing lines, by the extinction angles on each side of the trace of the twinning plane being approximately equal and by the fact that the two adjacent lamellæ are of the same color when the trace of the twinning plane is parallel to the plane of vibration of either nicol. Also in the 45° position the lamellæ are exactly the same color and the dividing lines disappear. Measure the extinction angles in as many sections thus selected as possible and take the maximum value.[111] This should be very close to the maximum extinction angle, which is a constant for each kind of feldspar. [Illustration: FIG. 73.—Showing conventional signs of extinction angles. ] MAXIMUM EXTINCTION ANGLES IN SECTIONS PERPENDICULAR TO ALBITE TWINS. Albite 16° Oligoclase, Ab_{4}An_{1} 5° Labradorite, Ab_{1}An_{1} 27° Anorthite 53° In the determination of rod-like microlites,[112] oligoclase extinguishes almost parallel to its length, while anorthite may show extinction angles of over 27°. When these microlites show _Albite_ twinning use the method just described. (3) _Fouqué’s_ method[113] can be used when the optical orientation of the section is known (as the result of a test with convergent light). The extinction angles of these known sections are of great diagnostic importance. The best sections are those at right angles to the two bisectrices, and these may be obtained by rapidly testing those sections, in the rock, which show an interference color about half as high as the maximum color in the rock section, in this way avoiding the sections parallel to the optic axes. Having found such a section, test it with a gypsum or ¼ undulation mica plate to prove whether the bisectrix is ⟂ a or c. If ⟂ a (these sections show sharp twinning striations) measure extinction angle between trace of axial plane and _albite_ twinning; if ⟂ c measure extinction angle between trace of axial plane and basal cleavage cracks.