May 9, 1895] 



NA TURE 



This applies more especially, perhaps, to the sections 

 treating on the examination of chrome-iron, silicon-iron, 

 nickel alloys, &c. 



The volume is particularly valuable as embodying 

 the results of an extensive experience in the ex- 

 iimination of certain iron alloys which are bound to 

 lu-iomc of special importance in the near future ; most 

 steel works analysts will cordialK- appreciate this 

 portion. 



As the results of my own practice, 1 can confirm the 

 accuracy and efficiency of most of the selected methods ; 

 more especially as applied to the assay of fcrro-chrome, 

 ferro-aluminium, silicon, nickel, &c. 



In regard to the assay of ferro-chrome or steels, 

 (ialbraith s method is to be preferred, if the precautions 

 given are adopted. The original process did not always 

 give concordant results. The gravimetric methods are, 

 liowe\er, on the whole most trustworthy. Results are 

 apt to be low unless great care is taken ; no doubt 

 for the reasons shown at page 207. The estimation of 

 small quantities of aluminium presents difficulties not 

 easily overcome ; indeed, simpler and less complicated 

 methods are required : a remark which applies to most 

 of the methods now practised. 



The assay processes for sulphur and plu)s|)h()rus are 

 clearly set forth, leaving practically nothing to be desired. 

 For the former element, certainly, gravimetric estima- 

 tions are best ; but it is nearly impossible to obtain the 

 necessary acids quite free from sulphur compounds : 

 this constitutes a serious drawback, and entails the 

 necessity of a blank experiment, which should be avoided 

 when possible. The evolution methods give only relative 

 results, agreeing pretty closely amongst themselves, but 

 somewhat under those obtained gravimetrically. The 

 author's colour test is a good one, but somewhat compli- 

 cated. A more simple modification of the colour test 

 consists in passing the evolved H^S through 50 c.c. of a 

 very dilute lead acetate solution f,'jj grm. in litre H.,0) 

 contained in a long test-tube. This is compared with a 

 standard steel, treated in the same manner, containing a 

 known percentage of sulphur. No precipitate is formed, 

 and a clear brown tint is obtained, which lasts for smnc 

 time, and is easily compared with the standard. 



The processes advocated for phosphorus (pp. 110-115; 

 are complete, but the necessary manipulative skill required 

 to carry them out can only be acquired b\' constant 

 practice. 1 find, however, that the addition of a little 

 HCl to the nitric acid solution assists the precipitation of 

 phosphorus when precipitating with ammonium molyb- 

 date. Kurther, I agree with the author that in ordinary 

 steels the presence of silicic acid may be ignored : « ith 

 regard to time, fifteen or twenty minutes is ample ; if 

 longer, molybdic acid is precipitated. In addition, even if 

 this does not occur, the precipitate may rcdissolve to a 

 notable extent. The dried phospho-moKbd.itc- precipitate 

 is distinctly soluble in dilute nitric acid. 



The author's method of precipitating arsenic with ll.,S 

 is good, but no others are given. The process with 

 modifications gives good results, but the ordinary method 

 is preferable when it is desired to estimate this element. 

 I or the mere elimination of arsenic from the phosphoric 

 acid, in order to determine the latter, the boiling or 

 ilistillation process is useful. 



NO. 1332, VOL. 52] 



It is to be regretted that no trustworthy process has 

 been given for the determination of oxygen in steel. A 

 thorough examination of the whole work, however, reveals 

 the pains taken by the author, not only as regards the 

 portions mentioned in the foregoing, but also in the some- 

 what less important sections dealing with fuel and other 

 materials. There can be little question that Prof .Arnold 

 has rendered steel-works analysts a decided senice by 

 the publication of his work. John P.\rrv. 



OUR BOOK SHELF. 



Wayside aiui Woodland /iiossoiiis. A Pocket Guide to 

 British Wild Flo-wcrs foi- the Country Rambler. By 

 Edward Step. With coloured figures of 156 species, 

 black and white plates of 22 species, and clear descrip- 

 tions of 400 species. (London: Frederick Warne and 

 Co., 1895.) 



M.\NV persons who admire the beautiful flowers that 

 adorn our woods and pastures would fain know their 

 names, with a view to further knowledg'c of them ; but for 

 various reasons they are unable to use the ordinarj- 

 "Flora," however simply compiled. Here is a little book 

 that will meet the wants of such persons, and do more, 

 we belie\e, to lay the foundation of a sound know ledge of 

 plants than the form in which " life-histories " are taught in 

 ordinary schools and classes for the purpose of passing an 

 examination. In spite of all that is said to the contraiy, to 

 knowalargenumberof plants, animals, or minerals bysight, 

 is of more value, to begin with, than a more detailed know-- 

 ledge of a single, or few, organisms or objects ; especially 

 when this detailed knowledge is gained by rote, and not 

 by observation. We therefore commend this little book 

 to the notice of those interested in, and believing in, small 

 beginnings, though the kind of information it contains is 

 not exactly what the examiner demands. The coloured 

 figures are well drawn, and the colouring, although a little 

 crude, is good enough to enable one to recognise the 

 plants the figures are intended to represent. The 

 majority of the common and prominent plants of our 

 native flora are figured. Many of them are drawn of the 

 natural size, whilst others are reduced and a few enlarged, 

 without indications of the reduction or enlargement. 

 These things should be explained for a beginner. The 

 dcscriptiv e and explanatory letterpress is instructive, and 

 free from pedantry, b)- which we iiiean the display of 

 technical terms only used by ''teachers" of botany ;'not 

 by botanists. There are some inconsistencies in the choice 

 of subjects for illustration. For example, the exceedingly 

 rare Holosleuni iiinlicllatuni is represented, whereas the 

 allied genus Cerastii/in. found in every county, and perhaps 

 in every parish and field in the kingdom, is left out. 

 There is also an uncxplainable absence of characteristic 

 sea-side plants. The bUu k and white figures mentioned 

 in the title represent native trees and some of the com- 

 monly-planted exotic species. .\n omission here is the 

 common yew, which might well have taken the place of 

 the \ery poor figure of .lilantu.t. In spite of the short- 

 comings indicated, we strongly recommend this little 

 pocket-book to those in search of some practical 

 knowledge of common wild [ilants. W. B. H. 



Tlie Lepidoptera of the Hritish Islands: a Descripti-e 

 Aeeount of the h'aniilies, Geiieni, and .Species indigenous 

 to Great Britain and Ireland^ their I'reparatorv States 

 Habits, and Localities. By Charles G. Barrett, F.E.S. 

 \'ol. ii. Heteroccra, Sphinges, Bombyces. (London : 

 I.. Reeve and Co., 1895.) 



.Mk. li.\KKKTT's great work on British Lepidoptera is 

 making steady progress, and we are glad to find that the 

 second volume which includes the Sphinges and the first 



