CONSTITUTION AND OKIGIN OF v SPHERULITES. 427 
busliy growth of the compound spherulites represented by 
Fig. 1, Plate 6, they are uniformly negative, but those of 
the second, coarser growth between these arms are all posi¬ 
tive. The feldspars of the residual and enveloping forms 
are variable in character, being positive in some cases and 
negative in others. In certain enveloping spherulites the 
needles of one portion are all negative, and of other parts all 
positive, the change from one character to the other taking 
place most frequently on some concentric zone of growth. 
Again, positive and negative fibers are mingled so intimately 
as to show that for these the optical sign is probably ±. 
All of these characters are explainable on the assumption 
that the needles of positive optical character are feldspar 
elongated parallel to the vertical axis, and that they possess 
the so-called abnormal optical orientation, the optic axes 
lying in the plane of symmetry. The needles whose char¬ 
acter is ± are elongated parallel to the vertical axis, but 
possess the usual optical orientation, and those of constant 
negative character are doubtless parallel to the clinoaxis and 
may have either optical orientation. The so-called ab¬ 
normal position of the axes of elasticity is found most fre¬ 
quently in sanidines of acid lavas, and that these needles of 
somewhat peculiar conditions of growth should be thus 
different from the free microlites is nothing strange. 
The cross sections of positive needles are sometimes 
rhombic, the angles corresponding to those of the feldspar 
prism, and in other cases they are apparently hexagonal 
through probable development of the clinopinacoid; but in 
many growths the faces of the prismatic zones are curved or 
otherwise irregular in development. Cross sections of positive 
needles polarize much more strongly than those of negative 
fibers, the latter being sometimes sensibly isotropic, a differ¬ 
ence which is a natural consequence of the optical orienta¬ 
tion. The larger negative fibers are frequently Manebach 
twins, and there are reasons to suppose that they grow char¬ 
acteristically with the reentrant angle oo Poo Aoo Poo out¬ 
wards. They fork at larger angles than the positive needles, 
55—Bull. Phil. Soc., Wash., Vol. 11. 
