October 27, 1892] 



NATURE 



toy 



wi to say that Bell's two roots (before 1824) were an 

 anterior " cerebral " root, subserving motion and sensa- 

 tion, and a posterior " cerebellar " root serving to govern 

 vital actions. The principle of localization in nerve- 

 roots, far more clearly stated by Walker in 1809 and the 

 facts demonstrated by Majendie in 1822, are not alluded 

 to. In the second lecture Kieinenberg's cells are figured 

 and described, and on the next page admitted to be 

 mythical ; thus Prof. Horsley is enabled legitimately 

 enough to utilize this time-honoured if anatomically in- 

 correct illustration to enforce the essentially correct 

 principle of differentiation. Lecture III. treats of jelly- 

 fish, star-fish, and cray-fish, with reference to rhythm, 

 "localization "and co-ordination of movements. " Local- 

 ization" is used as a term to denote a physiological pro- 

 perty or function (pp. 48-49) ; i.e., as used by psycholo- 

 gists to denote an act of the subject, rather than as used 

 by physiologists to indicate observed relations between 

 parts and functions. This use of the word is perfectly 

 legitimate, but it is rather apt to create confusion of 

 thought. " Localization" is sometimes used in a similar 

 sense in relation to brain-function, and with a similar 

 inconvenience ; " localization " by the brain in a psycho- 

 logical sense is properly localization by the subject, 

 localization in the brain is an object of physiological 

 experiment. No doubt it may be said that psychological 

 localization rests upon physiological differentiation 

 and localization ; none the less the use of the 

 term to denote a physiological property or function 

 is not advisable without very careful definition. ' 

 Lecture IV. deals with vertebrates— nerve-fibres, gullet 

 theory of canalis centralis, spinal cord, and nerve-roots, j 

 Lecture V. with ganglia. Here we must criticise. Look- \ 

 ing to the class of readers addressed, Fig. 26 may be 1 

 misleading as regards the anatomy of anterior and pos- I 

 terior roots. Fig. 28 (altered from Hirschfeld and ! 

 Leveilld) is very confusing, and the anatomy of the 

 brachial and lumbar plexuses is strange. A reader who 

 should gather his notions of the functions of spinal 

 ganglia from pp. 110-113 would have a very wrong idea 

 of the state of our physiological knowledge ; nor does ! 

 the odd expression, "the immense discovery by Claude \ 

 Bernard, of the so-called vaso-motor system of nerves," j 

 possess much justification as regards historical accuracy.^ 

 The four last lectures contain— necessarily mingled 

 with familiar elementary considerations— a statement of 

 the results arrived at by Professors Gotch and Horsley 

 from their electrical investigation of nerve-impulses in 

 afferent and efferent nerve-channels, and to the expert 

 form the most important part of the book. We begin, 

 therefore, to read more closely, still bearing in mind, 

 as indeed is suggested by the style, the requirements of 

 non-expert readers. Nothing arrests attention on the first 



' In point of t me Brown-S^quard is the true discoverer of vaso-motor 

 nerves. Bernard's experiments were made subsequently, and interpreted 

 otherwise. 



"D'aprfes ces experiences, il n'est done pas possible d'expliquer le 

 r ^chauffeinent des parties par une pritendue paralys'e des arteres, qui, k 

 raison d'un ^la-'gissement passif, laisseraient circuler une plus erande quan- 

 tity de sang 



"Si alors[f.«.. en galvan sant] les arteres. comme les veines, se resscrent 

 et reyiennent surelles-meraes. gela tient h ce qu'il n'y a plus de sang pour 

 les disiendre, mais ce nest pas du tout I'effet d'un resserrernent actif des 

 vaisseaux. 



..... " il ne peut venirk I'id^e de pers nne de penser i rapporter [le 

 phC-nomene c.rculatoire qui succfeHe k la section du nerf sympathique] i une 

 paralysie pure et simple d-s arteres." Bernard, AnnaUs des Sciences 

 .\atureUes, 1854, p. 198.) 



NO. 1200, VOL. 46] 



two pages. On p. 129 we pause at this sentence : — 

 " It is very interesting to see that the protoplasm of a 

 nerve-conductor has a distinctly longitudinal arrange- 

 ment, which, it is not going too far to suggest may, by 

 virtue of this fact, be more adapted for the polarization 

 of its molecules for the better transmission of nerve-im- 

 pulses." Having dissected out the possible meaning of 

 this sentence we proceed. Two pages further we are 

 stopped for a moment by a confusion between the local 

 excitability of nerve and its conductivity. On the next 

 page (p. 132) we demur to the assertion that "secondary 

 tentanus depends upon the electrotonic state of the first 

 preparation." On page 138 we find no reason to accept 

 the distinction that< " no doubt may reasonably exist that 

 active nerve yields products of oxidation, which doubt 

 certainly exists as to the acidification of nerve." Both 

 facts are possible but unproven ; no proof whatever has 

 been attempted of the first ; the second has been investi- 

 gated with positive and with negative results. Page 146 

 includes a figure in which the current is not shown as an 

 action current, but the reverse ; moreover, with the 

 instrument figured (capillary electrometer), no current is 

 under observation. 



But these twenty pages are enough, and we shall have 

 but little space to discuss what forms the main positive 

 differentia between Prof. Horsley's book and other books 

 of the same class, i.e., the conclusions derived from 

 electrical data. 



The conditions of criticism in this connection are alto- 

 gether different, and we need not stop to examine into the 

 accuracy of elementary points. Prof. Horsley is now 

 addressing himself to an expert audience ; his reasoning 

 and his data have yet to pass through the refining fires 

 of doubt and of objection, with, it is to be hoped, ultimate 

 confirmation. The principle of the method of investi- 

 gation is a well-established one ; we know that electrical 

 variations are indicators of functional variations ; in the 

 spinal cord, as elsewhere, functional activity may there- 

 fore be roughly gauged by galvanometer or by electro- 

 meter. Gotch and Horsley did this as regards efferent 

 channels and afferent channels ; as regards the first they 

 found by the electrometer that the character of discharge 

 in the pyramidal tract does not differ from its character 

 in motor nerves ; as regards the afferent tract they find 

 that impulses pass up the cord chiefly in the posterior 

 column of the same side. These conclusions may be 

 admitted without imprudence. But the conclusions that 

 may not safely be admitted without further experimental 

 elaboration, are those relating to the functional dis- 

 charges (inferred from electrical discharges) up and down 

 the anterior and posterior roots, and to the quantitative 

 distribution of centripetal impulses in the various columns 

 of the cord. As regards this second point the physical 

 conditions are not sufficiently analyzed (either in this 

 volume or in the original paper) for us to admit, e.g., that 

 average galvanometric swings of 60 and 20 indicate a 

 passage of afferent impulses in the proportions 60 and 

 20 per cent, in the posterior and in the lateral columns 

 respectively. That the deflection was proportional to 

 the number of fibres excited, is an assumption requiring 

 proof (p. 2i2,cf. also pp. 145, 159, 160). 



As regards the first point, it was found that electrical 

 discharges pass easily down as well as up the posterior 



