38S 



NATURE 



[February 23, 1S93 



tabular key based chiefly on neuration and antennae. 

 Mr. Hampson admits thirty-four families of Indian moths, 

 of which the first twenty-three, including 1158 species, 

 are dealt with in the volume before us. The earlier 

 families of moths are, however, much less numerous in 

 species than the later ones, and it must not be supposed 

 that Mr. Hampson has dealt with anything like half the 

 Indian species in his first volume, which comprises the 

 series of families usually classed under Sphinges and 

 Bombyces, extending, according to the author's classifica-' 

 lion, from SaturniidcB to HypsidcE. The important 

 Bombycide idLXm\\&'s,,Arctiidce,AgarisiidcE,?i's\6. UranitdcB, 

 are, however, relegated to the second volume, while 

 several famihes of more or less doubtful position find a 

 place in vol. i., such as the Cymatophoridce, Thyridida, 

 SesiidcE, and Tinccgeriidce. We observe that Mr. 

 Hampson closes the series of moths with the Tinezd(z, 

 Pterophorida, and Alucitidcs, and in this adopts the 

 usual classification, though in the main he has struck out 

 an entirely new classification of his own, and the very 

 first innovation which meets the eye is the novelty of 

 commencing the moths with the Saftirnndcs. 



We hope that Mr. Hampson will take an opportunity 

 of discussing the various systems of classification of 

 moths which have been proposed by Guenee, Herrich- 

 Schaffer, Plotz, and other entomologists, not forgetting 

 the strange system proposed by Zebrawski, in his work 

 on the Lepidoptera of Cracow, in which the butterflies 

 are placed in the middle instead of the beginning of the 

 series of Lepidoptera. Such a discussion would be un 

 suitable in the present work, but if published elsewhere 

 might be very useful. 



Long descriptions of genera and species in a work of 

 this character would have been out of place, and we are 

 glad to find that they have been avoided. Each family 

 or subfamily is succinctly characterised, and usually 

 illustrated by a figure of the larva. This is followed by 

 a tabular key to the genera, and then by a notice of the 

 genera and species. The notice of each genus con- 

 sists of synonymy, type, range, and a brief indi- 

 cation of the principal characters. That of the species 

 includes synonymy, description, including both sexes, 

 and transformations when necessary, range and expanse. 

 An excellent woodcut is usually given of one represen- 

 tative of each genus, showing the wings and body on 

 one side, and the neuration on the other, extra figures 

 of antennae and legs being sometimes added. 



No book, however useful or carefully compiled, can be 

 free from errors, but these cannot be detected at a glance, 

 and the only technical mistake of importance which we 

 have noticed in turning over Mr. Hampson's work is 

 that the broad-bordered Australian Macroglossum kingii, 

 Macl., is included among the synonyms of the narrow- 

 bordered Cephonodes hylas, Linn. 



Much, no doubt, remains to be said about Mr. Hamp- 

 son's classification, his use of generic names, and his 

 placing together insects regarded as distinct by other 

 authors as synonyms. But these are all points admitting 

 of great difference of opinion, and we do not propose to 

 discuss them further in the present notice. 



We should add that various new families, besides many 

 new genera and species, are described by Mr. Hampson 

 for the first time. W. F. K. 



NO. 12 17, VOL. 47] 



OUR BOOK SHELF. 



The Year-Book of Science (for 1892). Edited by Prof. T. 

 G. Bonney, D.Sc, LL.D., F.R.S. (London: Cassell 

 and Co., 1893.) 



All interested in scientific progress will welcome the 

 appearance of the second volume of this useful year-book. 

 The staff of contributors includes such names as Dr. 

 Ramsay, Prof. Seeley, Mr. Botting Hemsley, &c., and the 

 accuracy of the summaries of the year's developments 

 may therefore be thoroughly relied upon. The plan of 

 the volume follows closely on the lines of its predecessor, 

 but it has been extended so as to include geographical 

 and anthropological matters, and zoology has received 

 more complete treatment. If one may judge of the activity 

 in different departments of science by the space required 

 for the account of their progress, electricity and organic 

 chemistry would appear to take the lead. As in the last 

 volume, no attempt has been made to present a complete 

 catalogue of papers. The object has been simply to select 

 the memoirs of exceptional interest ; and so far as we 

 have been able to judge, the selections have been judicious. 

 An excellent index of subjects, and one of authors, com- 

 plete what will no doubt be found a very useful volume. 



Treatise on Thermodynamics. By Peter Alexander, M. A. 



Pp. xii, 203. (Longmans, Green, and Co., 1892.) 

 This is in many respects a singular work. Whole pages, 

 we may almost say whole sheets, are devoted to the 

 multiplication of elaborate proofs of intrinsically simple 

 theorems for which a i&w lines would be ample allowance, 

 while some of the real difficulties of the subject are but 

 lightly touched on. The other special characteristics, so 

 far as we have seen, are three in number. First, and most 

 prominent, the extraordinary proportion of formulae to 

 text, which gives the whole the look of a treatise on 

 Partial Differential Coefficients rather than on a branch 

 of Physics. Second, the fearful and wonderful collection 

 of names for special cycles, eg. Isotherjneniropicycle, 

 Isobarymegacycle, Isenergentropicycle, &c. Finally, the 

 expressions of doubt or hesitancy with which many steps, 

 universally recognised as valid, are introduced. In the 

 first and second of these characteristics the author far 

 transcends the results of the licence willingly allowed to 

 pioneers like Clausius and Rankine. But these have 

 been (at least in great part) long since discarded, and can 

 never be reintroduced. The third characteristic is, to say 

 the least, not precisely one to be desiderated in a text- 

 book, where we naturally expect to find some slight trace 

 of " Sir Oracle." 



MedicBval Lore : an Epitome of the Science, Geography, 

 Animal and Plant Folk- Lore and Myth of the Middle 

 Ages. Being Classified Gleanings from the Encyclo- 

 paedia of Bartholomew Anglicus on the Properties 

 of Things. Edited by Robert Steele. (London : 

 Elliot Stock, 1893.) 

 The original work of which parts are translated in the 

 present volume, may be said to have a place of its own 

 in the history of European literature. It was written in 

 the thirteenth century, and the Latin text was soon widely 

 appreciated, while in the course of the fourteenth cen- 

 tury it was translated into French, Spanish, Dutch, and 

 English. The book is full of interest, for it presents a 

 summary of all that was known in the Middle Ages 

 about man and the world. The change which has been 

 gradually effected by the use of modern scientific methods 

 is, of course, incalculable ; but some readers will pro- 

 bably be surprised to find to how large an extent Bar- 

 tholomew mingles the results of shrewd and accurate 

 observation with quaint fancies and unverified judgments. 

 The present volume consists of selections from the 

 edition of Berthelet, 1535 ; and the good style of the 

 translator adds greatly to the charm of the author's 



