58 



SCIENCE 



[N. S. Vol. XXX. No. 758 



distinct and clean-cut line, while the northern 

 border was every~where indefinite, gradually 

 dying out at about 30° to the north of the 

 zenith. During the course of the evening the 

 luminous area varied in width from 10° to 

 about 35°. 



The aurora was remarkable on account of 

 its unusual position, its rapid changes in bril- 

 liancy, and its varying shape. The color was 

 a pale bluish-gray, no iridescence having been 

 seen at any time. Moreover, there was no sug- 

 gestion of streamers or rapidly-moving irides- 

 cent patches, often referred to as "merry- 

 dancers." When the aurora was at its maxi- 

 mum brilliancy, only the stars of brightest 

 magnitude could be seen in that region of the 

 sky, and the " milky way " was rendered en- 

 tirely invisible. Over nine tenths of the sky 

 was cloudless throughout the evening, a trace 

 of alto-stratus cloud having been visible above 

 the northern horizon, and an equal amount of 

 cumulo-nimbus cloud with distant lightning 

 having been seen far off to the west. 



Andrew H. Palmer 



Blub Hill Obseevatoey, 

 May 17, 1909 



SCIENTIFIC BOOKS 

 Elements of Optical Mineralogy, An Intro- 

 duction to Microscopic Petrography. By 

 N. H. WiNCHELL and Alexander N. Win- 

 CHELL. New York, D. Van Nostrand Com- 

 pany. 1908. Pp. 502; 350 figures; 4 

 plates. Price, $3.50. 



During the last few years several excellent 

 treatises on optical mineralogy by Americans 

 have appeared, namely, those by Luquer, Idd- 

 ings and Johannsen. According to the au- 

 thors, none of these contains a concise and 

 clear exposition of the principles, methods and 

 data of optical mineralogy. It is to supply 

 this want that this text was written. 



The book is divided into three parts, as fol- 

 lows ; (1) Principles and Methods, (2) De- 

 scription of Minerals and (3) Analytical 

 Tables. 



In part one, seventeen pages are devoted to 

 some of the phenomena of light, twenty-two 

 pages to the elements of mineralogy, and fifty- 



nine pages to the application of polarized light 

 to crystalline substances. Part two contains a 

 systematic description of all the rock-forming 

 minerals concerning which there is sufficient 

 data to permit their being determined by 

 means of the microscope. These descriptions 

 occupy 310 pages. Part three is made up of 

 exhaustive analytical tables for deter- 

 mination, microscopically, of rock-forming 

 minerals. The tables extend over fifty-seven 

 pages. 



There are also three appendices, as fol- 

 lows: (1) Optical Study of Opaque Minerals 

 — ten pages, (2) Microchemical Methods — ^nine 

 pages, and (3) A Partial Bibliography- — two 

 pages. 



There will be, undoubtedly, much difference 

 of opinion among petrographers and physical 

 crystallographers as to whether the authors 

 have succeeded in presenting the principles 

 and methods, fundamental to a clear under- 

 standing of the physical properties of rock- 

 forming minerals, " concisely and clearly." 

 To be sure, they have been treated concisely, 

 but only in a few cases clearly. Thus, the 

 description of the nicol prism, a thorough 

 understanding of which is absolutely essential, 

 is intelligible only to those who have had some 

 previous knowledge of it and know what to 

 expect. Certainly a beginner can obtain no 

 clear conception of it. Furthermore, the de- 

 scription is not entirely accurate, as is shown 

 by the following sentence, lines 7 to 10, page 

 16 : " The cut faces, after polishing, are 

 cemented together again in their original posi- 

 tion by Canada balsam — which has nearly the 

 same index of refraction as the Iceland spar" 

 (the italics are the reviewer's). 



The statement, page 8, that the most exact 

 method of applying total reflection to the de- 

 termination of the index of refraction is the 

 Kohlrausch method, is misleading, as any one 

 who has had experience in applying it knows. 

 Line 9, page 9, should read "the axis of the 

 observing telescope OT is the line of the re- 

 flected ray," instead of the incident ray. Fig. 

 3c, accompanying this description, is poorly 

 executed, the line ON being by no means 

 normal to the plate OD. In fact, many of the 



