I30 



NATURE 



[yuneS, 1876 



service both to teachers and to students. The best proof 

 of knowledge of any branch of physics, and the most 

 practical result of the study of any such branch is the 

 acquisition of the power of applying numerical calcula- 

 tion to every question where a numerical result can be 

 obtained. The student knows that he understands a sub- 

 ject thoroughly when he can write down numbers to 

 express definitely the amount of every effect observed and 

 measured by experiment. The importance of numerical 

 calculations in absolute measure is becoming daily more 

 and more appreciated : and in the best English text-books 

 numbers expressing quantities in absolute measure are 

 now to be found, instead of the relative numbers that were 

 alone obtainable from the text-books of only a few years 

 ago. Mr. Day's book brings very fairly together such 

 questions as are likely to present themselves to the 

 student of electricity and magnetism. Anyone who has 

 acquired sufficient knowledge to work through a consider- 

 able part of the exercises cannot fail to find them extremely 

 useful. 



We have observed some slips that ought to be cor- 

 rected in future editions. Among them may be men- 

 tioned some of his exercises on the tangent galvanometer. 

 No practical experimenter would think of using the 

 tangent galvanometer in such a way as to bring the 

 deflection to 89° 30', as Mr. Day does in Ex. 23, p. 47, or 

 to the high numbers that he refers to elsewhere. We 

 find readings of Thomson's reflecting galvanometer given 

 in degrees, minutes, and seconds. This seems rather 

 absurd, to say the least of it. In a few of the exercises, 

 as in Ex. 9, p. 33, the data are insufficient. 



A few more definitions would, we think, be found useful. 

 Some of the terms employed are uncommon, and some 

 appear to be used somewhat ambiguously. Thus in Ex. 

 2, p. 17, the word density is wrongly used for quantity. 

 Again density of an electric current is a term so unusual 

 that some explanations regarding it seem all but necessary. 

 The definition, given in Ex. 2, p. 72, as Bunsen's defi- 

 nition, appears a very incomplete one. According to it, 

 a current of nnit density is a current of unit streti^th 

 passing throus;h a voltameter between two electrodes each 

 one square millimetre in diameter ; and^from this it would 

 follow, we presume, that the so-called density of the current 

 is the same at every part of the voltameter ar.d independent 

 of the form of the voltameter. If so we cannot think of 

 any possible use of such a name. The terms Farad and 

 Weber, given by some of the practical electricians seem 

 to be used indifferently in more senses than one. It is 

 simply unpardonable, in the present state of the science, 

 to introduce ambiguities of language. 



On the whole, however, we are much pleased with 

 Mr. Day's little book, and can warmly recommend it 

 both to teachers and to those who are studying electricity 

 and magnetism without the aid of a teacher. 



Geological Survey of Victoria. Prodromus of the Palce- 

 ontology of Victoria. Decade 3. By Frederick 

 McCoy. (Melbourne. — London : Triibner and Co., 

 1876.) 



We are glad to find that in spite of the unpromising news 

 which has recently reached England concerning the pre- 

 sent condition of the Geological Survey of Victoria, the 

 palaeontological work, which is in the hands of such a 

 well-tried and indefatigable naturalist as Prof. McCoy, 

 continues to make satisfactory progress. The present 

 decade of the Prodromus is of more than local interest, 

 containing as it does interesting new details concerning 

 Owen's marsupial lion, the Thylacoleo carnifex. The re- 

 sult of Prof. McCoy's examination of more perfect speci- 

 mens than those on which the first description species was 

 based, is to suggest modifications in some of the views 

 published by Prof. Owen, but to add confirmation to 

 that author's main position concerning the carnivorous 

 habits of the animal, a conclusion which was called in 



question by Dr. Falconer and Prof. Flower. Scarcely less 

 interesting at the present time is the illustration and 

 description of a species belonging to the sub-genus of 

 Nautilus, known as Aturia. A similar form has been 

 found by Dr. Hector in New Zealand, but in rocks of far 

 older date, and the facts which have already come to light 

 concerning the distribution in space and time of this 

 remarkable genus are such as to invest it with the very 

 highest interest both to the geologist and biologist. 



On similar grounds the new species of Tertiary Tri- 

 gonia and Plcurotomaria — genera which were so abundant 

 during earlier periods of the earth's history, but which, 

 except in Australia, appear to have become almost wholly 

 extinct at the close of the Mesozoic epoch — are especially 

 worthy of the attention of the paleontologist. The ether 

 new forms illustrated in this dtcade, including a number 

 of Trilobites and Tertiary Mollusca, do not call for any 

 special remark. Prof. McCoy's scientific descriptions are 

 admirably clear and exact, and his general remarks on 

 the relationships and distribution of the species very valu- 

 able and suggestive. The engraving and printing of these 

 decades afford evidence alike of the progress made by 

 our Austrahan colonies and the liberality with which 

 scientific research is supported in them. The plan of 

 publication by decades, illustrating the palasontology of 

 the countries geologically surveyed, was commenced in 

 the United Kingdom by Sir Henry de la Beche, and has 

 been followed both in Canada and India. The decades 

 of the Victoria Survey are quite worthy to take rank, both 

 as regards matter and form, with those of either of the 

 older surveys we have mentioned ; and higher praise than 

 this it would scarcely be possible to add. 



LETTERS TO THE EDITOR 



[ The Editor does not hold himself responsible for opinions expressed 

 by his cotrespondents. Neither can kt undertake to return, 

 or to correspond with the writers of, rejected manuscripts. 

 No notice is taken of anonymous communications. \ 



Scientific Poisoning' 



For giving instruction to one person in the art of poisoning 

 without detection, the medical student, Vance, is lusderyoing 

 the very lenient punishment of eighteen months' imprisonment. 

 What would hi the appropriate penally to indict upon the 

 responsible editors of nevvspapeis who initiate the ptd)lic generally 

 into Vance's secret ? Chemist 



Pyrology — Quantitative Analysis by the Blowpipe 



The estimation of constituents in compounds by the blowpipe 

 has leen hitherto, as is well known, limited to the process 

 of metallic (or, in the case of cobalt, srsenidal) reduction of 

 oxides, &c., and that with regard to a very few metals only. 

 I now piopose to inaugurate a new plan, by which this rapid, 

 elegant, and accurate method of analysis may (apparently) be 

 applied far more generally, and, as I hope, successfully. In my 

 published work ''Pyrology, or Fire Chemistry," I have, with 

 the exception of a few indications (as in the case of the insoluble 

 balls formed by lime in boric acid), confined myself to qualita- 

 tive research only, but many methods will suggest themselves to 

 the attentive student of that book, by which qualitative may be 

 readily extended to quantitative examination. 



I propose to proceed more in the direction of a kind of volu' 

 metric analysis than of analysis by means of the successive sepa- 

 ration of constituents, as in the " wet way," and I trust that the 

 consideration usually accorded to novelty and the difficulties 

 always inseparable from useful novelties will not now be refused 

 by scientific Englishmen to my feeble initiatory researches, espe- 

 cially as I am (1 believe) the first Englishman who has published 

 much original matter on this subject. It seems likely that the 

 operator who can, by reason of the rapidity of his methods, 

 obtain the 7nean 0/ a number of approximate analyses of a parti- 

 cular substance in the same or less time than that required by 

 the employer of an abstractedly more correct but practically more 



