May 5, 1893.] 



SCIENCE. 



249 



pie, thanks to its contained arsenic, was long; regarded as poison- 

 ous, until being produced in a state of purity, its entire harmless- 

 ness was demonstrated. There is a general review of the laws 

 regulating the use of poisonous colors, and then, verbatim, the 

 enactments of Germany under date of July 5, 1887. In 1888 

 there were appended to the said enactments regulations as to the 

 examination of colors, fabrics, fruit jellies, liquids, etc., for 

 arsenic and tin, and these Dr. Weyl has given in full. The 

 methods are interesting and exact, though not original. The laws 

 of other countries than Germany are given in some detail, and 

 then we pass to the experimental part, the method to be followed 

 being first described. As it was out of the question to test all, or 

 « ven the greater portion, of the numberless coal-tar derivatives, Dr. 

 Weyl selected such as were suspicious or had already been regarded 

 as poisonous and endeavored to take those in most general use. 

 Of the nitroso colors, we have dinitrosoresorcinol and naphthol 

 green, B. The nitro colors include picric acid, saffron-substitute, 

 Martins' yellow, naphthol yellowS, brilliant yellow, and aurantia, 

 and of these only the sulphonated colors, naphthol yellow, and 

 Martin's yellow were found to be harmless. The azo- colors aie 

 discussed at some length from both a technical and toxicological 

 standpoint, but of the twenty-three colors examined only two, 

 menatil yellow, and orange II., produced distinctly poisonous 

 €ffects when administered by the stomach. Many, however, de- 

 veloped a slight albuminuria, and one at least was plainly poison- 

 ous when introduced into the subcutaneous cellular tissue. 



It is highly gratifying to remark the comparative harmlessness 

 of by far the greater number of the coal-tar colors, and even in 

 those colors which are indicated as poisonous such large does are 

 necessary in order to produce toxic effect as to render accidental 

 poisoning from the same a practical impossibility. 



Much honor is due Dr. Leff maun for his part in giving to the 

 English-reading public this book, the first on the subject in our 

 language, — but the hearty reception it has met with from chem- 

 ist, medico-legal expert, and medical practitioner alike, bespeaks 

 sufficiently its worth and opportune appearance. 



Charles Platt. 



Alternating Currents. By Frederick Bedell, Ph.D., and A. 

 C. Crehore, Ph.D., Instructors in Physics, Cornell Univer- 

 sity. New York, W. J. Johnson Co. 

 The Johnson Co. is to be congratulated upon the appearance 

 and make-up of this volume. The large, clear print, good paper, 

 and well-drawn figures, make it one of the best books, 

 from a mechanical standpoint, which has ever been 

 published. On cai'eful examination there does not ap- 

 pear to be a single misprint, or a single error in the math- 

 ematical formulae, in marked contrast to the slipshod English 

 and errata which disfigure almost every page of Fleming's 

 book. No less are the authors to be congratulated on their work, 

 for this book will probably be for years a standard text-book on 

 the subject. Whatever one may find to criticise, it will not be 

 the manner in which the subject is treated, nor mistakes in the 

 treatment. 



The subject is developed in a logical and simple manner. In 

 Part I., which contains the analytical methods, we have, after 

 an introduction on the elementary notions of the magnetic 

 field, current flow, and harmonic motion, the general equation 

 for circuits with resistance and self-induction; then the solu- 

 tion to this equation, and its application to the different cases 

 possible. The constants of the equation are determined in each 

 case, and curves plotted from actual values of the resistance 

 and self-induction. Next in order come the general equations 

 for circuits with capacity and resistance, and circuits with resis- 

 tance, capacity, and self-induction. These are treated in the 

 same manner. All possible cases are considered, the constants 

 determined and curves drawn to illustrate the solutions. 



Chapters xii. and xiii. treat of circuits with distributed capac- 

 ity and self-induction, a subject of the utmost importance in 

 these days of long-distance telephoning and telegraphy. 



Part II. contains the graphical treatment. The analytical re- 

 sults obtained in Part I. are made use of as a foundation for the 

 graphical methods. In addition to the cases considered in Part 



I. we have cases of circuits, in series and parallel, containing dif- 

 ferent voltages, resistances, self-inductions and capacities, and 

 the results of variations of the latter in such circuits. At the 

 end of the book is given a table of mechanical and electrical anal- 

 ogies, amplified from that previously given by other writers. 

 The consistent notation used throughout the book gives an added 

 pleasure to its perusal. 



There are some things omitted which might have been treated 

 of with advantage. For instance, though the graphical solution 

 of problems concerning divided circuits is given, the analytical 

 is not. If Lord Rayleigh's method were the only one known, 

 there might be a reason for this, but those who are readers of La 

 Liimiere Electrique and L^ Electrician, will call to mind various 

 neat and simple methods of treating the subject, and the latter 

 is too important, practically, to be able to do without any thing 

 which can add to our information. 



We understand that the authors have underway a volume on 

 alternating circuits containing iron. With Kennelly's and 

 Steinmetz's laws, we may expect from the analytical treatment 

 much that is new and important with regard to the best size 

 and dimensions of transformers for given efficiency and output, 

 etc. 



This work has been adopted as a text-book by a number of 

 American universities, Cornell, Purdue, University of Califor- 

 nia, and others. R. A. F. 



Comparative Philology of the Old and New Worlds with Reference 

 to Archaic Speech. By R. P. Greg, F.S.A., F.G.S., etc. 

 1 Vol. LXXII. 355 p. Royal 8°. London, Kegan Paul, 

 Trench, Trubner & Co., 1893. 

 It it a painful duty for a reviewer to take up a work which is 

 honest in intention and laborious in executioji, but hopelessly 

 deficient in method ; and such is the one before us. To issue its 

 considerably more than four hundred large pages must have cost 

 the author a great deal of work and of money ; yet for all scien- 

 tific purposes the results he reaches must be estimated as scarcely 

 above zero. 



The judgment may seem harsh, but let us see what he sets out 

 to prove and what methods he adopts. He writes to support the 

 hypothesis of an original unity of language, of an original com- 

 mon tongue, an archaic speech of great simplicity, composed of 

 differentiated emotional and imitative utterances, fragments of 

 which can be traced in all the languages of the world, bringing 

 them, therefore, into a genetic relationship. To prove this, he 

 devotes over 350 pages to "Tables of Accordances," lists of words 

 which he believes to be from the same root in the most diverse 

 tongues. The hypothesis is by no means a novel one, nor does he 

 claim it as such, but perhaps it has not before been urged with 

 such abundance of illustration. 



Whatever one thinks of the hypothesis, all will agree that a 

 competent knowledge of linguistics should be asked in its sup- 

 porters, if they claim a hearing before the scientific public; and 

 just here Mr. Greg is strangely deficient. His introduction 

 begins with a survey of American languages, and as these figure 

 largely in the tables, they will serve as a test of his work in gen- 

 eral. 



His authorities at once awake astonishment. Ignatius Don- 

 nelly's "Atlantis," the second-hand reports of Bancroft, Canon 

 Cook, Hyde Clark, and Bradford, the tracts of Professor Camp- 

 bell, and Vincente Lopez, and a few unimportant and de- 

 fective vocabularies, such as these of Marcoy and Parry, are the 

 books that figure most prominently in his "list of authorities '' 

 What he has learned from them is on a par with their value. He 

 sfieaks (p. x.) of "the ancient Nahua and Aztec languages of 

 Mexico, ' unaware that these words are merely different names 

 for the same language. On the same page he refers to the 

 " Californian " language, as if any such existed; and attributes 

 to Schoolcraft (instead of Lieber) the term holophrastic, as applied 

 to American idioms Who "Dr. Daniel Whitney, the well- 

 known American philologist," may be, will certainly puzzle 

 readers, as he is surelj' not known on this side of the Atlantic. 



When it comes to the tables of accordances, all American lan- 

 guages are conveniently divided into northern, central, and 



