Sept. I, 1887] 



NATURE 



411 



into the hands of students who have but little time to 

 spare and may not intend to become professional 

 chemists, a very wide analytical field is got over ; indeed 

 a little too much is attempted in the space, and sacrifices 

 have in nearly all cases to be made where "shortness and 

 simplicity " is the combined ruling idea. 



We fully agree with what the author says as to the 

 educational value of quantitative analysis. It is indeed 

 high time that our more elementary students should 

 have the long courses of qualitative analysis shortened, 

 and some more exact exercises substituted. 



In the course of the 127 pages of this book, including 

 six for tables, we are introduced to the balance, and it is 

 much to be regretted that more has not been said about it. 

 What is said is purely practical — -how to turn up the 

 handle and put on the weights. 



The first exercises are the determination of water in a 

 carbonate and the ash in several substances, after which 

 a couple of specific gravity methods are given, and then 

 we pass to " simple gravimetric analysis," iron, silver, 

 barium, lead, &c. In the silver exercise the factor 075276 

 is introduced to get the actual silver from the weight of 

 chloride found, and this "factor" is given in all other 

 analyses. It is not of much use any way, and for beginners 

 it is not advisable, as it binds them down to the book, 

 and no appreciable time is saved for ordinary analysis 

 calculations. 



The directions for volumetric analysis are very good, 

 and the exercises are well arranged in order of difficulty. 

 The separation exercises and miscellaneous examples 

 will need some attention from the teacher. 



In the description of organic analysis — combustion of 

 carbon compounds — the closed-tube process is well de- 

 scribed, and a student might be able to do a combustion 

 from the description only ; but we are not informed, 

 when the open tube is spoken of, whether the same 

 length, viz. 18 inches, will be sufficient or not. By infer- 

 ence it will. We venture to say that a very doubtful 

 analysis, especially of a volatile body, would result from 

 the use of an open tube only 18 inches long. The 

 description here is much too slight to work by. 



The tables at the end are sensible — only just those 

 wanted in the course of the work in the book itself. 



Qualitative Chemical Analysis. By Dr. C. Remigius 

 Fresenius. Tenth Edition. Translated and edited by 

 Charles E. Groves, F.R.S. (London : J. and A. 

 Churchill, 1887.) 



The fifteenth German edition of this well-known book 

 contains many emendations and additions, especially in 

 the concluding portions devoted to the reactions of the 

 alkaloids and the systematic methods of detecting them. 

 Of this edition of the original work the present edition of 

 the English translation is as nearly as possible an exact 

 reproduction, and much credit is due to the translator and 

 editor for the care with which he has accomplished a very 

 difficult task. Various styles of type and other typo- 

 graphical improvements have been introduced, in the 

 hope, as Mr. Groves explains, that the book may thereby 

 be rendered more handy and useful to students. 



Melting and Boiling Point Tables. Vol.11. By Thomas 

 Carnelley, D.Sc, and Professor of Chemistry in Uni- 

 versity College, Dundee. (Harrison and Sons, 1887.) 



The issue of vol. ii. of this important work completes it. 

 It is not too much to say that these two volumes will be 

 found in every laboratory. Their compilation represents 

 an amount of patient work from which most men would 

 have recoiled ; and the total result, which has cost ten 

 years of effort, reflects the highest credit upon Prof. 

 Carnelley. 



Part II., dealing with organic compounds, brings the 

 <lata down to 1885. 



Part III. deals with vapour tensions and boiling points 

 of simple substances, and freezing and melting points of 

 cryohydrates, including facts recorded in 1886. 



LETTERS TO THE EDITOR. 



\_The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he under- 

 take to return, or to correspond with the writers of, 

 rejected manuscripts. No notice is taken of anonymous 

 com m un ic at ions. 



[ The Editor urgently requests correspondents to keep their 

 letters as short as possible. The pressure on his space 

 is so great that it is impossible othenoise to insure the 

 appearance even of communications containing interesting 

 and novel facts. ^ 



The Law of Error. 



Everyone interested in the theory of statistics is aware how 

 strongly Queteiet was under the conviction that there is only one 

 law of error (or curve of facility, to use the correspDnding ex- 

 pression for the graphical representation of the law) prevalent 

 for the departure from the mean of a number of magnitudes or 

 measurements of any natural phenomenon. I have done what I 

 can to protest against this doctrine as a theoretic assumption ; 

 and recently Mr. F. Cilton and Mr. F. Y. Edgeworth have 

 shown in some very interesting and vahiable papers in the 

 Philosophical Magazine and elsewhere how untenable it is, and 

 how great is the importance of studying the properties of other 

 laws of error than the symmetrical binomial, and its limiting form 

 the exponential. 



I have been making some calculations recently, principally 

 in the field of meteorology, and I should be extremely glad of the 



judgment and criticism of any of your readers who maybe better 

 versed in this science than myself. It must be carefully under- 

 stood that the questions here raised are solely these : — (i) Do the 

 magnitudes, when arranged in order of their departure from the 

 mean, display a symmetrical arrangement? (2) If so, is this 

 arrangement in accordance with the binomial or exponential 

 law ? 



The first diagram represents the grouping, in respect of re- 

 lative frequency, of 4857 successive barometric heights. They 

 are from the observations of Mr. W. E. Pain, of Cambridge, 

 and show the readings at 9 a.m. on successive mornings for 

 about thirteen years from January i, 1865. They are the results 

 of the same instrument, which has required no correction or 

 alteration during that period. They are given to the first decimal 

 place. 



