April 19, 1894] 



NA TURE 



575 



maps, their scale and mode of colouring being, it is 

 stated, based on a definite system. The work is one of 

 Philip's Geographical Series, which is edited jointly by 

 Mr. Ravenstein, Mr. J. Scott Keltie, and Mr. Mackinder. 

 An admirable discussion of projections, scales, and 

 measurement on maps with diagrams, is the most 

 interesting part of the introductory text, and these 

 difficult questions are handled with rare conciseness and 

 clearness. The name of the projection and the natural 

 scale are given for every map— a most useful innovation. 

 An original diagram, showing the relative heights, 

 depths, and curvature of the surface of the globe along 

 the equator, and the parallels 30'^, 45^, and 60"^ N. and S. 

 gives a very striking view of the vertical relief of the earth. 

 Ten plates are devoted to general physical geography, 

 and although the scale is small, and the features conse- 

 quently much generalised, they are clear and satis- 

 factory. The rainfall maps on plate 5 are particularly 

 interesting, including two which are entirely new, repre- 

 senting the average number of rainy days in different 

 parts of the world, and the relative humidity of the 

 atmosphere. Equal praise cannot be given to the maps 

 dealing with the distribution of plant and animal life, 

 the names on which are frequently puzzling, and the 

 species selected for treatment curiously unequal. 



The maps of continents and countries, which make up 

 the bulk of the atlas, show configuration by a combina- 

 tion of contour lines and tints with hill-shading. Here 

 we regret that Mr. Ravenstein had not courage to dis- 

 regard the conventional strong green tint for the low- 

 lands, and to adopt an unbroken system of deepening 

 shades of brown. It is noticeable, also, that the applica- 

 tion of the green tint is unsystematic, extending to 300 

 feet in some maps, to 600 in others, and in one at least to 

 3000 feet above sea-level. A similar break in the 

 system of indicating density of population is sure to give 

 rise to confusion ; the same shade being used to repre- 

 sent regions of over 512 inhabitants to the square mile in 

 Europe, those over 256 in Asia, over 64 in Africa, and 

 over 32 in America. There are several minor defects 

 visible in the maps, such as the omission of links in the 

 through railway system, and difference in the representa- 

 tion of county-boundaries in the maps of England and 

 Scotland, but these are not more frequent than in atlases 

 of much greater pretensions, and will of course disappear 

 in a new edition. The colour-printing of many of the 

 maps may also be improved. 



A great feature of the work is the number of inset 

 maps, garnishing the margins of larger plates with en- 

 largements of regions of special interest, or small-scale 

 general maps showing geology, climate, vegetation, 

 race, language, or density of population. In this way 

 the fact that there are as many maps as there are i 

 sciences involving distributions is kept to the front, and 

 the teacher or scholar using this atlas is led to see that 

 geography is no haphazard agglomeration of discon- 

 nected details, but a shapely system incorporating and 

 elucidating the results of all departments of nature- 

 study. We know of no atlas in any language in which 

 the systematic plan has been more successfully elaborated, 

 the exceptions we have noted above being thrown into 

 undue prominence by the general excellence of the 

 whole. 



NO. 1277, VOL. 49] 



OUR BOOK SHELF. 



Life and Rock ; a Collection of Zoological and Geoloi^ical 

 Essays. By R. Lydekker, B.A., F.G.S., &c. Know- 

 ledge ^^x\qs. (London: Universal Press, 1894.) 

 Mr. Lydekker's name is well known as that of a 

 popular exponent of the results of zoological and pahton- 

 tological research. The essays which he collects under 

 the title of " Life and Rock," and which deal with a wide 

 range of zoological subjects, from elephants to what he 

 terms " forams," will serve to sustain his reputation 

 in this respect. Mr. Lydekker's chief aim is to convey 

 information which shall not be couched in terms so 

 technical as to discourage those who have neither the 

 inclination nor the desire to become serious students. 

 In this he is eminently successful. His language is simple 

 clear and direct, without any attempt at a distinctive 

 style ; and he shows good judgment in drawing the line 

 beyond which his readers would not be able to follow 

 him, nor care to make any effort to do so. Above all he 

 is accurate, and his information may be relied on as up 

 to date. We question, however, his wisdom of speaking 

 in a book of this popular kind of "those writers who ex- 

 plain evolution by some mode of what they are pleased to 

 call natural selection " ; and of " those who put their faith 

 in a mode of evolution dependent only upon so-called 

 natural causes." Let us by all means have the best and 

 most forcible available criticism on natural selection as 

 a valid explanation of the observed facts of organic 

 nature and on the evolution of living and extinct beings 

 by processes which are termed natural. Such criticism 

 is the very life of science. But the words we have 

 italicised are mere side-thrusts, which do the reader no 

 good, and the writer no credit. 



Disease and Race. By Jadroo. (London : Swan 

 Sonnenschein and Co., 1894.) 



" To show some continuity in disease, to evolve a little 

 order out of the existing chaos," these were the objects 

 which led the author to write this book. He collects in- 

 oculable diseases, and arranges them in a genealogical 

 tab]e,which is suggestive, to say the least of it. Tuberculosis 

 is shown to be descended directly from scrofula, and 

 scrofula from syphilis, which in turn is regarded as 

 having descended from leprosy. Other diseases are 

 supposed to be connected in a similar manner, though 

 the grounds upon which the supposition is based are not 

 very firm in many cases. The chief point to which the 

 arguments lead is that "every contagious or infectious 

 disease, by either the formation of a hybrid, or by here- 

 ditary transmission of the individual modification, tends 

 to eradicate itself." The author is strongly in favour of 

 the establishment of institutes for the purpose of research 

 into the bacteriology, etiology, epidemicity, and sequence 

 of diseases. As there is a dearth of human subjects 

 upon which to experiment, he suggests that convicted 

 murderers should be given the option of death or the 

 probability of leading a comparatively comfortable exist- 

 ence in a bacteriological institute. It has been said that 

 "the worst use you can put a man to is to hang him" ; 

 and certainly, if a murderer were used as a medium for 

 the cultivation of bacteria, he would expiate his offence 

 in a very suitable manner. 



LETTERS TO THE EDLTOR. 



\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 

 antiscripts intended for this or any other part <?/Nature, 

 o notice is taken of anonymous co>ntnunications.'\ 



The Mass of the Earth. 

 This month's Philosophical Magazine contains an interest- 

 ing, instructive, and suggestive, criticism of Prof. Poynting's 



