May 22, 1896.] 



SCIENCE. 



785 



" To complete a series, any specimen is better 

 than none." 



"A copy, model or picture of a good thing is 

 often more useful than an actual specimen of a 

 poor one." 



"Restorations made in such a manner that the 

 part restored is not at once distinguishable are 

 unpardonable. ' ' 



' ' A label (in the exhibition series) should an- 

 swer all the questions which are likely to arise 

 in the minds of the persons examining the ob- 

 ject to which it is attached." 



Dr. Goode's critical notes on various mu- 

 seums, introduced as illustrations of the princi- 

 ples discussed, are so interesting as to suggest 

 that the author would render the scientific pub- 

 lic a further service, if he could find time to 

 expand this little pamphlet into a moderate- 

 sized treatise on the museums of the world and 

 their administration. 



Wm. North Eice. 



Spectrum Analysis. De. John Landaxjee. 



Brunswick, Fred. Vieweg & Sohn. 1896. 



This handbook of some 175 pages is sub- 

 stantially a reprint of the author's article upon 

 Spectrum Analysis, which appeared in the 

 ' Handbook of Chemistry ' of Drs. Fehling and 

 Hell. Though now somewhat enlarged, it still 

 treats more particularly of the chemical appli- 

 cations of the subject. A brief historical intro- 

 duction, covering the time from Melville to the 

 present day, is followed by tolerably complete 

 descriptions of instruments for obtaining and 

 examining the various spectra. No attempt is 

 made to develop the theory of any of the in- 

 struments considered. The conditions affecting 

 the character of emission and absorption spectra, 

 and the empirical formulae which have been 

 suggested to express the relation between the 

 lines and groups in the spectra of diflferent ele- 

 ments are also touched upon, and then follow 

 tables of wave-lengths of various metalic spectra. 

 These embody the recent work of Kayser & 

 Eunge, Rowland and others, and all wave- 

 lengths are expressed in Rowland's scale. 

 Rowland's (1892) table of solar wave-lengths is 

 also given, and the principal astronomical ap- 

 plications of spectroscopy are briefly treated in 

 some fifteen pages at their end. Throughout 



the book copious references are given to origi- 

 nal papers, etc. , the whole forming a fairly com- 

 plete resume. The English student will find the 

 German unusually clear and concise. 



C. E. M. 



SCIENTIFIC JOURNALS. 



JOUENAL OF GEOLOGY, APEIL-MAY. 



The Magmatic Alteration of Hornblende and Bio- 

 tite: By Heney S. Washington. 

 It is well known to petrographers that these 

 minerals, under some conditions, tend to alter 

 into a granular mass of augite and magnetite. 

 The causes of this alteration are here discussed. 

 After reviewing current theories, including that 

 of Zirkel, the author proceeds to develop his 

 own views. He finds that this alteration is 

 most common in the intermediate group of vol- 

 canic rocks. He also finds it rare in the plu- 

 tonic rocks. From the latter fact he infers that 

 conditions of slight pressure are favorable to 

 the changes. The theory proposed is that 

 hornblende and biotite crystals are formed at 

 an early (intratelluric) stage of eruption under 

 conditions of great pressure, and probably in 

 presence of mineralizers. As they approach 

 the surface in the course of an eruption the 

 pressure diminishes, leaving the temperature 

 still high until a point is i-eached where the sub- 

 stance is no longer stable. Here a molecular 

 change is begun which induces a molar change, 

 so that the chemically and physically homoge- 

 neous hornblende or biotite becomes the hetero- 

 geneous granular aggregate of augite and mag- 

 netite. The origin of the augite andesites is 

 then discussed in the light of this theory. 



On the Origin of the Chouteau Fauna : By Henry 



Shalee Williams. 



In a former number of the Journal of Geology 

 the origin of this fauna was discussed by Stuart 

 Weller. In the present paper the author dis- 

 sents from two opinions therein expressed (1) 

 that the Chouteau fauna was contemporaneous 

 with the Chemung fauna of New York, and (2) 

 that it arose by the mingling of a fauna which 

 in the Devonian was represented by the Hamil- 

 ton in New York and the general Devonian 

 fauna of Europe represented by the Middle 

 Devonian of Iowa and British America. Three 



