734 



SCIENCE. 



[N. S. Vol. XIX. No. 488. 



of gold in the interior incited exploration, and 

 soon small collections of plants were brought 

 in, mainly from upper rocks of the Yukon. 

 As these agreed in Tertiary age with those 

 previously known from the coast region, it 

 came to be accepted that only Tertiary plants 

 occurred throughout this vast area. The U. S. 

 Geological Survey desired to establish a type 

 section, and in 1902 Mr. A. J. Collier was 

 delegated to make a trip down the Yukon, 

 studying the stratigraphy and collecting fossils 

 from as many points as possible. When the 

 plants were studied it was found that those 

 from above Kampart were Tertiary (so-called 

 Arctic Miocene), while below this point a very 

 different condition was found. Near Nahoc- 

 latilten Mr. Collier obtained collections which 

 appeared to be mixed, that is, a part of the 

 material seemed to be Tertiary and the re- 

 mainder Middle or Lower Cretaceous. Un- 

 doubted Upper Cretaceous plants were ob- 

 tained from a number of other localities, es- 

 pecially in the vicinity of Nulato, but so much 

 interest attached to the Nahoclatilten localities 

 that in 1903 Dr. Arthur Hollick was commis- 

 sioned to duplicate Mr. Collier's trip. From 

 the combination of the two collections it is 

 possible to decide with certainty that all points 

 above Rampart are Tertiary, while below the 

 plants indicate that the age is either un- 

 doubted Cretaceous or doubtful Tertiary. The 

 Cretaceous plants include cycads of several 

 genera, conifers and many dicotyledons, the 

 combination resembling mostly the Middle and 

 Upper Cretaceous flora of Bohemia. 



Mr. Marius R. Campbell then discussed 

 ' Glacial Erosion in Western New York,' de- 

 voting special attention to the origin of the 

 Finger Lakes. Mr. Campbell's paper is soon 

 to appear as a publication of the Geological 

 Society of America. 



The 154th meeting, held March 23, was de- 

 voted to a communication by Dr. Arthur L. 

 Day, entitled ' The Study of Minerals in the 

 Laboratory.' The follotving brief abstract 

 contains the more important conclusions. 



The paper is a summary of an extended ex- 

 perimental research by Dr. Day and Dr. E. 

 T. Allen upon a series of artificial feldspars 

 corresponding to albite (Ah) and anorthite 



(An) and the following plagioclases : AbjAn^ 

 AbjAn,, AbjAn,, Ab.An^, AbjAn^ and Ab,An,. 

 These were prepared with great care from the 

 purest chemicals, and, with the exception of 

 albite, which could not be crystallized on ac- 

 count of its extreme viscosity, were fully 

 identified microscopically. 



Melting-point determinations were obtained, 

 beginning with anorthite and continuing down 

 the series as far as AbjAn,. The viscosity was 

 found to increase enormously toward the albite 

 end of the series and appeared to veil the 

 melting point completely below AbjAn^. In a 

 specimen of natural albite (Mitchell County, 

 N. C.) the viscosity of the glass during melt- 

 ing was found to be of the same order of 

 magnitude as that of the crystallized portion, 

 and crystals of microscopic size sustained a 

 slow heating to a temperature of 150° above 

 where melting began, and preserved their 

 original orientation. It was also shown that 

 in viscous liquids, capable of considerable 

 undercooling, the solidifying point did not 

 coincide with the melting point or bear any 

 necessary relation to it. 



The chemical purity of the artificial feld- 

 spars enabled very accurate determinations of 

 the specific gravity to be made of both the 

 vitreous and the crystalline form. 



The following conclusions were offered as 

 tending to show that the soda-lime feldspars 

 form an isomorphous series and not a eutectic 

 mixture : 



1. In AbjAn^ the spherulites and the glass 

 matrix were analyzed separately and proved 

 to be identical in composition. In Ab,An„ a 

 portion was crystallized rapidly and the matrix 

 crystallized very slowly in a subsequent heat- 

 ing. The two sets of crystals, although very 

 different in size, proved to be identical in com- 

 position. 



If the relation had been that of a eutectic 

 mixture, the component in excess must have 

 crystallized out first. 



2. The curve of melting points of the pure 

 (artificial) feldspare of the series is a straight 

 line for some two thirds of the distance from 

 anorthite to albite (as far as it could be es- 

 tablished). There is no reason to suspect a 

 discontinuity in the lower portion of the curve. 



