584 



NATURE 



[February 29, 1912] 



auilicr i< i''ally seekiiu; tlir iriiili, .uid not a mere 

 .-•ladiatoiial victory or xoiiii^ "' |H)inls. 

 ** The .section most interesting to scientific workers 

 is that in which the author discusses vitalism and the 

 relation of mind to body. Quotinpr Bergson, Ward, 

 and Taylor, he expresses disapprobation of the theory 

 of "guidance." On this theory, mind and its world, 

 choice and action, become "utterly discontinuous." 

 The choosing unit or element is not a system of the 

 contents dealt with by choice. The "plan " is brought 

 to the material; it is not in it or elicited from it. 

 The view in question is a survival in principle of the 

 notion of matter ./)Zu5 miracle— the attitude of common 

 external teleology (p. 205 and foil.). Moreover, there 

 is the difficulty about energy. The guidance theory 

 tries to shade this down by analogies such as the 

 trigger, the ball or water-drop on a high divide, or 

 the spark which explodes the gas in a gas-engine. 

 In these cases a small variation in energy-expenditure 

 may cause huge differences in result. But some ex- 

 penditure there must be. On the analogy, the mind 

 must furnish energy without participation of the body. 

 "Views of this tvpe only escape manifest conflict with 

 common sense by restricting the amount of energy 

 so furnished to an amount below the possibility of 

 measurement " (xxvi.). 



Many readers who have studied with interest and 

 admiration the writings of Driesch, Bergson, and 

 Lodge on this point will feel that Dr. Bosanquet's objec- 

 tion is a formidable one; as is also his criticism of 

 Prof. Bergson 's startling contention that contempla- 

 tive and motor memory are radically different, the 

 former being independent of brain. It is true that 

 these are matters of science, and philosophers must 

 tread warilv in the foreign territory ; but their outlook 

 is wider— though with less perception of detail near at 

 hand— and their criticism is to be desired and welcomed. 



A Nature Calendar. By Gilbert White. Edited and 

 with an introduction by W'ilfred Mark W'ebb. Pp. 

 xii+62 + xiii-xx. (London : The Selborne Society, 

 1911.) Price 255. net. 

 This beautiful facsimile, published by the Selborne 

 Society, reproduces a record for the year 1766 of 

 botanical observations made chiefly at Selborne, with 

 an occasional note on birds or insects. This record, 

 of which the MS. is in the possession of Mr. Webb, 

 has never been before published, and is not to be 

 confused with the so-called " Naturalist's Calendar," 

 often printed at the end of the " Natural History of 

 Selborne." The printing, paper, and binding of this 

 large volume are all admirable, and the brief intro- 

 duction is adequate ; it is a superb volume to lie on a 

 drawing-room table and be admired by the chance 

 visitor, who will, it may be hoped, at least be struck 

 by the strong", firm, and legible handwriting of the 

 famous naturalist. White himself would be amazed 

 at the magnificent dress in which his humble notes 

 were destined eventually to appear ; no man could 

 know better than he that in no sense whatever could 

 they form even the material for a book. Yet Mr. 

 Webb claims that " now after an interval of a hundred 

 and twenty-three years a second book makes its ap- 

 pearance in the shape of the present volume." W'hite 

 published but one book, and that an incomparable 

 one. Mr. Webb publishes for him a second one, under 

 the auspices of the Selborne Society. Making all 

 allowance for enthusiasm, and for the carefulness of 

 the editing (of which the excellent index is perhaps 

 the best part), those who know how real books can 

 only be built up on a foundation of lengthy studies, 

 and how unwilling an author is to have such studies 

 exposed to the gaze of the curious, will feel some 

 regret that this rather meagre diary should have been 

 thus magnificently produced. W, W. F. 



NO. 2209, VOL. 88] 



■jjr 



[7, 



'IT OR. 



l>y his cor ' 

 . or to cv> ■ ; 

 >>hnni\' itpt: III!' ),'./ (/ fur litis or any other part of ^ 

 Au nuliic IS tuLcn oj aiwiiyryions rovniiunicattoti., 



Contour Diagrams of Human Crania. 



Has tkii Pii I lioiTipson got over the " l.i 



of ti\ii\ .11 111 I in the individual judgment 



volv. (1 ill siipi 1 ])(i-i!i^ lAo cr.'ini.-il cfjntours by selectii 

 cjiiiti: arliiirarilv , th<; vrlical axis of tli'j transverse sect: 

 as the Iciij^tli to be equalis'd in all such sections? M 

 1 sii)^<4ist tiiat he should try 'ciuaiising his auricu 

 distances, and taking his [)<rr(iitaj;5e differences on i. 

 vertical ordinates? 1 fancy he will then find that t! • 

 differences in form of two skulls will not even be "iiiu! - 

 sised at the same places as on his arbitrary schem' 



Again, in the case of the sagittal section, ther. 

 least half-a-dozcn fundamental lines any one of wl 

 might find justification in individual judgment as 

 standard for equalising size. A mathematician wouid 

 probably object to equalising any lines at all, but would 

 magnify up all his sections to be of equal area. !!• wimld 

 then be certain that the total area intercepted iiis 



superposed contours — however placed — was : :i 



would certainly mean that on any reasonable superposii 

 the contours would be very (In- together. In such < 

 for the transverse siMlimi, «■ should all probably suj)' - 

 pose the median lines. Inn, aijain, whether we should p * 

 the vertex on the veriux, or the auricular line on '■ 

 auricular line, or superpose neither, would be matter 

 discussion, if not for individual judgment. The width 

 individual judgment allowed in the case of the sagiti.il 

 section, having regard to such standard lines as eithT rh'- 

 " horizontal plane " provides or as join nasi a, 



lambda, inion, opisthion, and basion, is ^' .! 



Prof. D'Arcy Thompson's method would requin .1 i^i.i, 

 logical concordat before it could be put into practical for 

 even supposing we could agree on what should in this c 

 be the " area " of the section. 



Still another group of investigators might consider it 

 desirable to equalise, before superposition, not any 

 arbitrary lines or much more definite areas, but the 

 volumes of the two type crania as determined, say, by 

 average capacities or by the product, perhaps, of three 

 arbitrary diameters. Be this as it may, either an equali^-i- 

 tion of areas or of volumes seems to me a more reas 

 able preliminary to comparison of form than any equal: 

 tion of an arbitrary line. Yet such equalisations will <• 

 leave a " lack of fixity and precision " in our results, 

 wish to test how far our contours are similar and similarly 

 placed curves ; we ought to bring something approaching 

 a " centre of similitude " into superposition in both con- 

 tours ; the orientation in the case of the transverse and 

 horizontal sections will present no difficulty — in the case 

 of the sagittal it is much more questionable. The mathe- 

 matician would possibly select as his centres for testing 

 similitude the centroids of either the contours or of their 

 areas — if he were equalising areas, probably the latter. 



I would therefore suggest as a method to be compared 

 with Prof. D'Arcy Thompson's results, say, in the first 

 place, for the transverse contours : — (i) the equalisation of 

 areas ; (2) the superposition of centroids of areas ; (3) the 

 orientation by parallelism of median lines ; (4) the com- 

 parison along rays through this centroid. Thus the con- 

 tours themselves would be directly compared, and not 

 auxiliary curves. Lastly, if the superposed contours be 

 divided into equal angular elements a, and v be the mid- 

 distance of any element of the first contour from the 

 common centroid, v' the distance along the same ray to 

 the compared contour, then 



-Q^/^'''-''^'^\/c;'((r), 



■■^{m'}h 



where S denotes a summation for every element, would 

 be a fit measure of the degree of resemblance. 



Possibly some mathematician may be willing to under- 

 take the general theorem : Given two oval curves, the 

 shape of which must not be changed (but size is change- 



