642 



NATURE 



\Oct. 31, 1889 



extension of the spectroscopic field into the ultra-violet and 

 infra-red. In the old tables the limits of the spectra were 

 practically 394 and 670, whereas in the new ones the lines 

 range from 204 to 770. Spectroscopy has also advanced 

 in another direction. It was formerly believed that each 

 substance had its own characteristic spectrum, from 

 which there was no departure ; but subsequent researches 

 have shown that the spectrum does not entirely depend 

 upon the substance under examination, but also upon 

 the conditions of temperature and pressure. In the 

 old tables, for example, only one spectrum of oxygen 

 was recorded, but now no less than three are given. 

 Hydrogen, again, has now two spectra recorded, and 

 nitrogen three, including Hasselberg's important obser- 

 vations. 



The wave-lengths given in Angstrom's " Spectre Nor- 

 mal du Soleil," with a few small corrections, are still taken 

 as the standards for reduction. The tables printed in 

 the Reports of the British Association Committee form the 

 basis of the new edition, but there are also many im- 

 portant additions. One new feature is the addition of 

 a column giving oscillation frequencies, in number of 

 waves per centimetre in vacuo, which will no doubt be 

 appreciated most by investigators of the molecular origin 

 of spectra. Tables of the spectra of various compounds, 

 such as ammonia, alumina, and other oxides, chlorides, 

 iodides, &c., and water, are also given. The different 

 substances are arranged alphabetically as in the old 

 edition, and at the head of each there are full references 

 to books and memoirs. The introductory matter has also 

 been considerably expanded, and now forms an excellent 

 guide to spectroscopic scales and methods of mapping. 

 The use of a lens to throw an image of the light source 

 on to the slit, a method which has yielded many valuable 

 results, is, however, unfortunately omitted. The book 

 will be heartily welcomed by all who are engaged in 

 spectroscopic work, and no recommendation of ours is 

 necessary. 



A Text-book on Steam and S team-Engines. By Prof. 

 Andrew Jamieson, M.Inst.C.E. (London : Chas. Griffin 

 and Co., 1889.) 



We welcome with pleasure the fifth edition of this work. 

 Few engineering text-books are intelligible to the average 

 student. Many writers, in dealing with even the simplest 

 engine or mechanical contrivance, completely fog the 

 reader's understanding by the undue use of mathematics 

 and abstruse formulae. The volume before us is the best 

 yet pubhshed for use in the engineering classes at our 

 schools and colleges. Prof. Jamieson has treated the 

 subject in a sensible and useful manner ; his examples are 

 worked out as simply as possible ; and the descriptions 

 throughout the work are those of a practical man who 

 knows his business. 



The new edition contains many extensive and important 

 additions both to the text and illustrations. The chapter 

 on locomotives has been considerably enlarged and im- 

 proved. An express-engine built by Messrs. Dubs and 

 Co., the eminent Glasgow locomotive builders, is taken as 

 an example, and many well-executed scale- drawings are 

 given as illustrations. Even with these additions the 

 chapter does not do justice to this important branch of 

 engineering, and Prof. Jamieson must not overlook the 

 fact that he has many locomotive engineer apprentices at- 

 tending his Glasgow classes. The few paragraphs on the 

 compound locomotive are decidedly weak. Mr. Webb's 

 compound locomotive " The Experiment" is excellent 

 ancient history, no doubt ; but why not describe the more 

 recent Webb compounds, or, better still, the Worsdell and 

 Von Borries two-cylinder compounds, now doing such 

 good work on the North-Eastern and many foreign 

 railways ? 



N. J. L. 



LETTERS TO THE EDITOR. 



[ The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can he undertake 

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

 manuscripts intended for this or any other part of Nature,. 

 No notice is taken of anonymous communications, "l 



An Unusual Geological Sequence, 



In a late expedition to the north-west coast I have come upon 

 evidence of a fact which was quite new to me — namely, that the 

 well-known Cambrian red sandstones of Ross and Sutherland do 

 not always rest upon the Archsean gneiss, but occasionally on 

 dark blue stratified rocks with which the sandstones are perfectly 

 conformable. For many years I have been familiar with the 

 ordinary sequence, according to which the Cambrian or 

 "Torridon" sandstones rest unconformably on the Archaean 

 gneiss with nothing interposed between them. Nowhere in 

 Sutherland, or in Loch Torridon, so far as I have observed, is 

 there any variation in this order, and I have stood on some hills 

 in Sutherland where the Cambrian sandstones are represented by 

 only a few remaining cakes of conglomerate which lie bedded 

 almost horizontally upon highly unconformable gneissic strata, 

 I was therefore much surprised to see in a little creek on the 

 eastern shore of the Island of Raasay, a low precipice of the red 

 sandstone terminating in conformable beds of a rock of very 

 dark colour, and with a texture but little crystalline. The 

 sudden and violent change of colour at once attracted my atten- 

 tion, and on landing and obtaining specimens I found there could 

 be no mistake that the Cambrian sandstones here rest upon some 

 older rock totally different in mineral character from the Archsean 

 gneiss, and equally different from themselves. 



Pursuing this (to me) discovery, I examined the eastern face of 

 the same island, where its precipices include fine escarpments 

 both of the Secondary and of the older rocks. There, at one 

 point, I found the same unusual sequence beautifully distinct. 

 The sandstones are represented by a bed of strong conglomerate, 

 and this'bed rests conformably upon well stratified rocks of a blue, 

 or dark blackish-blue colour, with a fracture far less crystalline 

 than most of our Silurian slates on the mainland of Argyllshire, 



Following up the same clue, I found that on the western 

 shores of the Island of Scalpa, these blue rocks underlie in great 

 thickness the red sandstones which form the bulk of the island, 

 and which are exclusively seen by all who appi^oach it from the 

 eastern and northern sides. 



I now understand that this fact has been for some time known 

 to Dr. Geikie, and that the officers of the Survey under him 

 have come across it with equal surprise, in certain parts of Ross- 

 shire. But, so far as I know, it has not been published, and 

 is not generally known. 



In one specimen which I obtained on Scalpa there are obscure 

 indications of Annelid borings, together with calcareous cavities, 

 which are very suggestive of an organic origin. 



If these rocks really belong to the Cambrian series, as this 

 complete conformability would imply, and if they have been 

 wholly removed in all but a few spots, before the Torridon sand- 

 stones were laid down, the fact gives one a good deal to think 

 of both as regards the intervals of time which they represent, 

 and as regards the agencies of change which must have been 

 at work. 



To what horizon do these blue rocks belong ? The Suther- 

 land fossils from Durness are thought to be among the very 

 oldest Silurian forms. Below these come the great white 

 quartzites of the same county. Below them, again, uncon- 

 formably, come the Torridon sandstones, and lowest of all come 

 these subsequent blue beds — not at all metamorphosed — less 

 crystalline than many of the secondary rocks. Yet they must be 

 amongst the very oldest sedimentary rocks known to us. 



I may add that I found by actual experiment that in a deposit 

 now forming here, of the same blue colour, Annelid burrows 

 develop precisely the same ferruginous stains which I find in the 

 Scalpa specimen before referred to. Argyll. 



Inveraray, Argyllshire. 



Mr. Galton on Natural Inheritance. 



Mr. Galton's recent ingenious book on natural inheritance 

 suggests some remarks on the value of his method and results. 

 In the first place, it is plain that the method of probable error, 

 which he uses, is only applicable with any certainty to cases 



