582 



NATURE 



[October 20, 1892 



(tome vii, of the Recueil) in 1827, the year of his death, and 

 this is the paper which contains Fresnel's latest develop- 

 ments of his theory. In it he suppresses entirely his 

 method of generalization and develops that mechanical 

 theory by which, to quote his biographer Verdet, " he 

 endeavoured to rediscover truths which a profound in- 

 tuition had first revealed to him." The truths remain, 

 and though Fresnel's methods and their historical 

 development are clearly given in his collected works and 

 in M. Verdet's most admirable introduction to them, 

 Mr. Fletcher has done good service to science in call- 

 ing fresh attention to these earlier papers and in making 

 a modification of this method of Fresnel's the foundation 

 of his work. 



In his note to the first memoir, when discussing the 

 inexact reasoning by which Fresnel afterwards supported 

 his mechanical theory, Verdet writes (" CEuvres de 

 Fresnel," ii., p. 327) : — 



" II pouvait sembler singulier que le resultat d^finitif 

 d'un raisonnement incomplet et inexact en deux points 

 flit une des lois de la nature dont I'expdrience a le mieux 

 confirm^ la veritd. On a vu au contraire que cette loi 

 s'dtait manifestee a Fresnel comme le resultat d'une 

 generalisation toute semblable aux generalisations qui ont 

 amen^ la plupart des grandes d^couvertes. Lors qu'il a 

 voulu ensuite se rendre compte de la loi par une thdorie 

 mccanique il n'est pas ^tonnant qu'il ait, peutctre k son 

 insu, conduit cette thdorie vers le but qu'il connaissait 

 d'avance et qu'il ait dte determine dans le choix des 

 hypotheses auxiliaires moins par leur vraisemblance 

 intrinseque que par leur accord avec ce qu'il etait en 

 droit de considdrer comme la vdritd." 



True though this may be, the fact remains that until 

 Lord Kelvin developed his 'theory of a contractile ether 

 a few years ago, no one of the distinguished men who 

 have followed in Fresnel's steps had discovered a satis- 

 factory mechanical basis for Fresnel's great generaliza- 

 tion. But to return to the book before us. As has been 

 mentioned, Mr. Fletcher's method of development differs 

 somewhat from that indicated by Fresnel. He finds it 

 more convenient to work with rays than with wave- 

 normals or wave-fronts, and the construction he adopts 

 is the following: — Draw a normal at any point of the 

 ellipsoid of elasticity — the indicatrix in Mr. Fletcher's 

 language. From the centre of the eUipsoid draw a line 

 perpendicular to this normal, and consider a ray travelling 

 in the direction of this perpendicular. Then the reci- 

 procal of the intercept on the normal between the surface 

 and the ray measures the velocity of propagation along 

 the ray, and the plane of polarization of the ray touches 

 the indicatrix at the point at which the normal is drawn. 

 According to Fresnel's theory the radius vector drawn 

 from the centre to this point is the direction of vibration 

 in the ray, while according to the most recent modifica- 

 tion of the theory, the motion takes place along the normal 

 itself. From this simple construction the form of the 

 wave-surface and all the known laws of the propagation 

 of light in crystals are deduced in a strict and skilful 

 manner. At the same time, while giving Mr. Fletcher 

 the fullest credit for his originality, we are at times in- 

 clined to wish he had adhered more closely to Fresnel's 

 method. He admits of course himself that a single ray 

 cannot be propagated through the ether. We may hope 

 that some of those who read his book will go on to study 

 NO. I 199, VOL. 46] 



the mechanical theory of double refraction. Then they 

 must deal with waves and not with rays, and they would 

 find it an advantage to have had the one idea to guide them 

 throughout. Again, the new method leads to a multi- 

 plicity of names for one and the same thing, and this is 

 a disadvantage. We have ray-surface used for wave- 

 surface, although the two are identical, nor is it easy at 

 first to recognize the optic bi-normals and the optic 

 bi-radials as the optic axes and the lines of single ray 

 velocity respectively ; but these are small points when 

 compared with the main object of the book, which well 

 deserves attention and careful study. The last chapter 

 deals with the problem in a more general way, but space 

 forbids us to follow Mr. Fletcher into the questions he 

 there raises ; it must suffice to call the reader's attention 

 to it, and especially to the fallacy discussed in Section 17. 



R. T. G. 



THE PROGRESS OF HORTICULTURE. 

 Contributions to Horticultural Literature. By William 



Paul,] F.L.S. (Waltham Cross, Herts: W. Paul and 



Son, 1892.) 

 For about half a century Mr. Paul has been labouring at 

 the work of horticulture alike in the garden and at the 

 desk. As, a business man he has not confined himself 

 simply to commercial routine. As an observer and an 

 experimenter he has not been hedged in by the dogmas 

 and prejudices of any particular school of science, and as 

 a writer his aim has always been to record truthfully and 

 instruct faithfully. It is a matter of congratulation, 

 therefore, that the author should have gathered together 

 in a convenient form some records of a lifetime's work. 



Certain portions we should have eliminated as of past 

 or of personal interest only ; certain others as of relatively 

 minor importance ; but Mr. Paul is addressing a mixed 

 audience with varied sympathies and interests, and it may- 

 be that the paragraphs we should mark for deletion 

 would be those which others would best care to preserve. 



Mr. Paul groups his writings, as here collected, under 

 the three heads of (i) roses, (2) trees and plants, and 

 (3) fruit culture and miscellanea. They would fall equally 

 well under other categories, such as the commercial and 

 practical, the sesthetic and the biological. In this notice 

 we must confine ourselves to Mr. Paul's writings as a 

 naturalist. Such, however, is the interdependence among 

 various branches of inquiry, that it is almost impossible, 

 in this connection to isolate any special subject, even if it 

 were desirable to do so. From this point of view Mr. Paul's 

 book is, though undesignedly, an apt representation of 

 the present condition of horticulture. On the one hand, the 

 relations of that art to the perception of and to the canons 

 of beauty are obvious. Equally clear are its bearings on 

 routine practice. On the other hand, its connection with 

 biological science, in spite of the teaching and example of 

 Darwin, is not yet adequately recognized ; nor has the 

 statesman as yet grasped the truth that progress in 

 agriculture must follow to a large extent on the lines 

 familiar to horticulturists. Of the many remedies pro- 

 posed to mitigate and clear away the depression under 

 which agriculture is suffering, none is more likely 

 to be serviceable than the adoption, so far as 



