NATURE 



[Feb, 2 1, 1889 



style in which it is printed and the excellent plates seem 

 to promise that the Transactions will be quite on a 

 level with any journal published in Europe. The illustra- 

 tions are, indeed, unnecessarily large ; but this cannot 

 be pointed to as a fault— at least by those who are not 

 responsible for the cost. 



Prof. Spencer's paper is of considerable interest, parti- 

 cularly that section which refers to the nephridia. Meg- 

 ascolidcs, like Perichata (as was first pointed out by 

 Beddard, not by Perrier, as Prof. Spencer asserts), pos- 

 sesses a ramifying network of nephridial tubes which 

 are continuous from segment to segment, and which open 

 on to the exterior by numerous pores ; connected with 

 these there are— in the posterior segments of the body— a 

 pair of large nephridial tubes in each segment, which open 

 internally by a funnel. It is from these latter that the 

 single pair of nephridia per segment of Lumbricus, &c., 

 are to be derived ; the network of minute tubules, which 

 represents the excretory system of the flatworms, has 

 disappeared in such forms as Lumbricus. 



Prof. Spencer discusses the much-vexed question of the 

 homologies of the sexual ducts, and concludes that they 

 are not derived from nephridia. 



Other points of interest cannot be touched upon in this 

 short notice. 



Lectures on Geography, delivered before the University of 

 Cambridge, during the Lent Term, 1888. By Lieut.- 

 General Strachey, R.E., C.S.I., President of the Royal 

 Geographical Society. (London : Macmillan and Co., 

 1888.) 



These lectures are published opportunely at a time when 

 it is most desirable that the now almost general effort 

 to further geographical education should be properly 

 directed. They form a short course introductory to the 

 work of the Lectureship on Geography now established in 

 Cambridge, and in them General Strachey describes the 

 aspects of the subject which he considers most suitable 

 for the instruction of students at the University. He thus 

 gives a complete summary of the aims and matter of 

 scientific geography — of geography as a natural science 

 related to other natural sciences, much as mathematics is to 

 physical science. He assumes that students, before going 

 to University, have acquired a general knowledge of 

 geography ; and, in passing, he points out that the 

 primary object of the school teaching of geography is to 

 impart an accurate knowledge of the main topographical 

 features of the entire earth, all trivial details being omitted, 

 and suitable instruction being given in the physical, eco- 

 nomical, and historical characteristics of important places. 



As material for the higher or University teaching of 

 geography, the author practically claims the various 

 branches of science which in recent years have been as- 

 sembled under the term " physiography " ; but he is most 

 successful in showing that the science is not a mere 

 patchwork, but a connected whole ; and he sees no rea- 

 son for abandoning the well-known name "geography." 

 Certainly from many points of view the introduction of 

 the new term has retarded the spread of a knowledge of 

 the science. 



An excellent epitome of the growth of our knowledge of 

 the astronomical relations of the earth, and a short 

 account of the methods of projection and orography, pre- 

 pare the way for the history of geographical discovery. 

 This department is reviewed in a manner at once interest- 

 ing and philosophical, indicating clearly the close con- 

 nection between the progress of discovery and the political 

 movements of the world. The influence of the form and 

 movements of the earth on terrestrial phenomena, terres- 

 trial magnetism, our knowledge of the interior of the 

 globe, and the relation of geology to geography, are in turn 

 shortly discussed. The sections on land, sea, and air, and 

 on the history of life and of man, indicate the results of 



recent investigation, and suggest many points which may 

 well receive much attention from students of geography. 

 The lectures are written throughout in an agreeable 

 and simple style, and will prove valuable to general 

 readers as an elementary epitome of scientific geography. 

 F. Grant Ogilvie. 



A Text-book of Elementary Metallurgy for the Use of 

 Students. By Arthur H. Hiorns. (London : Macmillan 

 and Co., 1888.) 

 We recently had occasion to notice a useful little work 

 on practical metallurgy by Mr. Hiorns. He has now 

 endeavoured to write a purely elementary treatise on 

 theoretical metallurgy, adapted to the capacity of be- 

 ginners. The attempt can scarcely be considered suc- 

 cessful. In 172 pages printed in large type he deals 

 with the whole of the wide field of metallurgy. This 

 necessitates a very fragmentary treatment. And besides 

 this, errors are so frequent as to render the book 

 quite unsuited for beginners. The following examples 

 may be cited : — The barrel method of amalgamation is 

 stated (p. 90) to be carried on at Freiberg, where it was 

 discontinued twenty-four years ago. One of the seven 

 methods of producing steel is stated (p. 74) to be " by 

 melting raw steel in crucibles." The Copp^e coke-oven 

 is described (p. 40) as being of the Appolt type. The 

 coke-oven described (p. 42) as the Simon-Carves is in 

 reality a Carves oven. The author appears to be ignorant 

 of the existence of the principality of Catalonia, for the 

 Catalan process is said (p. 54) to be carried on at " Catalan 

 in the Pyrenees." 



Altogether, the book compares very unfavourably with 

 the author's work on assaying, and appears to have been 

 hastily written. An illustration of the want of care dis- 

 played is afforded by the table of the specific gravities of 

 eighteen metals (p. 11), in which in nine cases the figures 

 differ from those given in the author's companion volume. 

 With a little care, the author could have avoided such 

 statements as — "An analogous compound,' Boghead ' of 

 Scotland, which is a bituminous schist, is richer in bitumen 

 than ordinary coal." Again, manganese, the author states 

 (p. 74), " prevents the separation of carbon in the form of 

 graphite, which is the opposite of silicon." The appendix 

 of examination questions, covering 65 pages, appears to 

 indicate that Mr. Hiorns's intention has been to write a 

 cram-book for the elementary stage of the Science and 

 Art Department's examination in metallurgy. It is, 

 however, doubtful whether a student who made such 

 blunders as occur in this book, would satisfy his examiners. 



B. H. B. 



LETTERS TO THE EDLTOR. 



yrhe 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. ] 



Weismann's Theory of Variation. 



According to Weismann ("Die Bedeutung der sexuellen 

 Fortpflanzung fiir die Selektions-Theorie," Jena, 1886), heredity 

 does not consist in the parent having the power to reproduce 

 offspring in its own likeness, but in the property of the germ 

 (ovum or spermatozoon) in each generation to develop into an 

 individual of a certain invariable type. He starts from the fact 

 that in development the germinal cells are separate from the 

 beginning, are portions separated off from the original fertilized 

 ovum. He distinguishes between actual and virtual differences. 

 Different individuals developed from successive remnants of a 

 given Keimplasma may show actual differences ; but these are 

 due to the action of conditions affecting the particular individual 

 during its development and life : these differences are not in- 

 herited, cannot possibly be transmitted to the offspring, because 



