July 20, 1905J 



NA TURE 



267 



finds it extrenielv difficult to dver-reacli these wary 

 lilrds, and in some places one may see them in 

 hundreds of thousands for several weeks at a stretch 

 without the possibility of securing a single specimen. 

 This is especially the case in thickly populated 

 regions, where the geese already know that danger 

 may threaten them." 



Bv this we know that the writer has toiled and 

 suffered many disappointments. In certain British 

 w.iters wliere for three seasons Brent geese were 

 abundant we never obtained more than one good shot 

 with the punt g'un in a season. This was generally 

 at the commencement, when the birds arrived in late 

 October. .After this date we could only " look " and 

 "long." Other species are equally cunning. 



The key to genera, species, and subspecies with 

 which the author furnishes us is an excellent com- 

 pilation, although he does not make clear the differ- 

 ence between species and subspecies. For instance, it 

 appears that full specific rank is accorded to Branta 

 boniicla, Branta bernicla glaucogastcr, and Branta 

 hcrnida nigricans, the three varieties of the Brent goose 

 which visit our shores. If those which are furnished 

 with trinomial names are intended to be subspecies, 

 and it is a very doubtful point if they deserve even this 

 distinction, the author should say so in his table, 

 l^ersonallv we do not think that there is any reason 

 for separating these three well marked varieties. \A'e 

 have killed all three from one flock, and visitors to 

 the northern breeding-places of these birds have also 

 found all three, as well as intermediate forms, breed- 

 ing together on the same ground. If such splitting 

 were to come into general use, endless new subspecies 

 must be created amongst the goldfinches, crows, 

 skuas, S:c.. and many other birds we could mention 

 the slight local peculiarities of which afford small 

 points of distinction. Neither is the author consistent 

 in this respect, for he refuses to recognise " two geo- 

 graphical races, much less two species," of grev 

 geese living in eastern and western areas, and also the 

 .\merican and European forms of the white-fronted 

 goose as distinct. With regard to the bean goose, 

 Mr. .-\lpheraky recognises three distinct races, 

 .1. segetum, the common bean goose, .1. arvensis, 

 which possesses white feathers at the base of the bill, 

 and the eastern bean goose, .4. serrirostris, a bird 

 described by .Swinhoe, which is larger, dis- 

 tinguished by its more massive bill, .-\nothcr species 

 closely allied to the last named, namelv, .4. mcntalis, 

 but which was first described bv Przewalski in 1876, 

 seems to be of very doubtful rank, and may be onlv 

 a large form of the .Siberian bean goose. 



In this excellent monograph the author gives us 

 all we wish to know about the difference of sexes, 

 gradual growth from nestling upwards, plumage 

 variation, moulting, local names, chase, and colour 

 of the soft parts, the latter, perhaps, the most im- 

 portant point of all in the determination of species. 

 Many excellent outline figures of tlic bills are also 

 given, so that the reader has no difficulty in recog- 

 nising the differences of the various races even if he 

 feels inclined, as he must sometimes do, to question 

 the necessitv of specific separation. 

 NO. 1864, VOL. 72] 



To the oologist, too, the table and descriptions to be 

 found on pp. 185-190, furnished by Mr. G. F. Gobel, 

 are of the most exact and comprehensive nature, and 

 the book is one that every working naturalist or wild- 

 fowler should possess in his library, for it is by far 

 the best worii that has as yet appe.-ired on this 

 interesting family of birds. .[. G. M. 



THE ELECTRIC FURNACE. 

 Lc Four Electriqiie: son Origine, ses Transformations 

 ct scs Applications. By Adolphe Minet. ler Fasci- 

 cule. Pp. 76. (Paris : Librairie Scientifique, 

 .-\. Hermann, 1905.) Price 5 francs. 



THE application of electric heating to various 

 metallurgical and other industries has of late 

 been making very rapid progress. The time seems, 

 therefore, to be well chosen for examining the various 

 stages of development which the electric furnace has 

 passed through. 



M. Minet has taken great pains to collect together 

 as much as possible of the available information, and 

 has certainly succeeded in producing an interesting 

 studv. Chronologically, he divides his subject into 

 three periods: — (i) laboratory furnaces (1808-18S6); 

 (2) industrial furnaces (1886-1890); (3) development of 

 the industrial applications of the electric furnace from 

 1S30 to the present day. The furnaces themselves 

 are classified in nine groups, according to the function 

 of the current and the method of its application. 

 Any historical treatment of such a subject as this, 

 which expects to be generally recognised as authori- 

 tative, demands very great care and judgment in its 

 preparation. The present review certainly promises 

 to be the most complete which the electric furnace 

 has yet received. 



It is, however, not so clear that the author has 

 succeeded in accentuating just those developments 

 which have been of the greatest influence to the 

 general progress. There are no doubt difficulties in 

 deciding between two such different claims as those 

 of a brilliant invention and of a painstaking scientific 

 investigation. The successful historian must, how- 

 ever, accurately estimate the value of each and decide 

 on the relative merit according to the influence 

 exerted by each upon subsequent development. 



The classification of electric furnace processes is 

 complicated, not only by the large number of separate 

 cases which have to be considered, but more especially 

 bv the very different purposes for which the electric 

 current is applied. In the first place it is necessary 

 to distinguish between the purely electrothermal and 

 the electrolytic functions of the current. The latter 

 case embraces all such electrolytic methods as are 

 carried out at a moderately high temperature. Here 

 the electric current serves the double function of main- 

 taining the necessary temperature and separating by 

 electrolytic decomposition one or more of the con- 

 stituents of the materials treated in the furnace. 



During recent years the most extensive develop- 

 ments in electric furnace work have centred around 

 the production and application of extremely high 

 temperatures. The direct results of the scientific and 



