320 PROCEEDINGS OF THE AMERICAN ACADEMY. 



twinning stnicture of the latter with that of the well known case of 

 leucite. It is true, so far as the writer can ascertain, that the potash 

 member of such microperthites, where this has been determined care- 

 fully, is microcline either with the usual gitter structure, or with the 

 albite twinning alone, or even without twinning. 



The theory above put forth does not contradict the conclusion of 

 Vogt that many, perhaps most microperthites and cryptoperthites 

 are essentially eutectic mixtures. It supplements it l)y way of adding 

 something to the theory of feldspar mixtures so far as the alkalie 

 members are concerned, and it is believed may lead to somewhat bet- 

 ter understanding of their relations in igneous rocks generally. The 

 approximation to constancy in the composition of the cryptoperthite 

 and microperthite feldspars as shown by Vogt, is a strong argument in 

 favor of their being eutectic in nature. Their peculiar structure 

 is also favorable to this view. For, if we reason from the analogy of 

 the well known eutectic structures in alloys and laboratory salts in 

 which one of the characteristics is an intimate mechanical or crystallo- 

 graphic mixture of the two phases forming the eutectic, we should 

 expect to find in feldspar mixtures, which are of nearly or exactly 

 eutectic composition, a corresponding tendency to form an intimate 

 and curious intergrowth. In view of the well recognized sluggishness 

 associated with transformations in the solid state, we should expect 

 that the unmixing at a transition point, such as the one suggested 

 above, would in many, perhaps most cases be more or less incomplete, 

 and in mixtures which originally departed more or less widely from 

 eutectic proportions, it is probable that the resulting crypto- or mi- 

 croperthite would, owing to the incomplete separation of the two 

 phases, be only an approximate eutectic ("anchi-eutectic" of Vogt). 

 To the sluggishness of transformations in the solid state must be added 

 in the case of the feldspars a further, and doubtless consideraV)le, im- 

 pediment to the progress of the readjustment caused by the extraor- 

 dinarily viscous character of the feldspar substance. This may be 

 inferred from the thermal properties of the feldspars as revealed by 

 the brilliant work of Dr. A. L. Day ^^ and his associates at the Geo- 

 physical Laboratory in Washington, D. C On account of their 

 peculiar thermal properties it is to be expected that except under the 

 most favorable conditions, critical points on their cooling curves will 

 be readily passed by through supercooling, and that we should rather 

 speak of critical intervals than points. For the same reasons, while 



87 The Thermal Properties of the Feldspars, Carnegie Institution, Washing- 

 ton, D. C, (1905), and American Journal of Science, 14 (Feb., 1905). 



