March 22, 1894J 



NA TURE 



479 



commutator to produce even wear ; although with gauze 

 brushes this would seem an unnecessary refinement — the 

 wear of the commutator surface being in this case so 

 small and even. With regard to " end play " a further 

 word should be added. In the case of steam-dynamos, a 

 fly-wheel is often placed between the dynamo and engine 

 to steady the running. In such cases the fly-wheel 

 becomes magnetised, and as a consequence the armature 

 is pulled away from the engine, thus bringing into play 

 a considerable pressure on the bearings. This is usually 

 remedied by placing the armature core slightly out of 

 symmetry (longitudinally) with regard to the magnet 

 limbs. 



The subject of insulation material for commutators is 

 discussed — this being a point of great importance in 

 armature building. Useful data under this head have 

 been collected and tabulated. 



Chapters xii. and xiii., on the construction of commu- 

 tators and the manner of connecting the segments to the 

 armature winding, are very good. Descriptions, amply 

 illustrated by woodcuts, are given of the principal methods 

 of construction adopted by makers, and the whole subject 

 of commutators is well treated. On p. Ii6 a method is 

 mentioned for preventing what is termed " slewing ' of 

 the segments caused by the washer going round with 

 the nut (when not specially prevented) and taking the 

 ends of the segments with it. This is a point in the 

 construction of commutators worthy of attention. 



In treating of the sparking at commutators, the author 

 starts by giving an elementary theory of electric sparks, 

 as introductory to the main subject ; and after a descrip- 

 tion of what he terms " elementary planes through j 

 commutator and armature," he leads up to the elementary ' 

 consideration of the brush itself 



A chapter is devoted to carbon brushes, and another to 

 causes of sparking exterior to the machine. On the 

 subject of armature reaction a good deal of information 

 is given, and Bayer's system of winding, in which what 

 are termed "commutator coils" are interwound with 

 the main winding of the armature, is very fully described. 



The book concludes with a chapter on armature defects, 

 and a few practical hints on the taper of commutator 

 segments. E. Wilson. 



BRITIS H MOSSES. 



Illustrated Guide to Byitish Mosses. B\- H. G. 

 Jameson, M.A. (Eastbourne: Published by the 

 Author, 6, College Road.) 



AT R. JAMESON has proiluced a very useful treatise 

 ■^^ ^ on British mosses, wel calculated to aid the stu- 

 dent of theirsystematic classification. The book consists 

 of an introduction giving clear and valuable information 

 on the structure of the mosses, with a useful section on 

 the practical examination of specimens ; a key to the 

 genera ; a short account of each genus, followed by a 

 key to its species ; and a series of 59 plates, containing 

 over 2400 figures, all of which appear to be clear, and 

 some of which are admirably drawn and lithographed. 

 The figures are for the most part not those of the entire 

 plant, but of those parts which are especially useful in 

 distinguishing the species. As the book contains no de- 

 NO. \ 273. VOL. 49] 



119 



scription of each species, but of the genera only, the 

 student must mainly rely on two things — on the key and 

 on the plates. The result is to throw a great burthen 

 on the key, which is framed on the familiar dichoto- 

 mous arrangement, by which the student is continually 

 presented with one or other of two courses, so that the 

 success of his search depends on his taking the right one 

 of the two alternatives before him at each successive 

 step. A single false choice sends him off in the wrong 

 direction, and all his labour is wasted. It follows that 

 the choice put before the student at each step should be 

 between two alternatives perfectly true, perfectly distinct, 

 and, if possible, indubitable. Now, here Mr. Jameson 

 seems to us sometimes to fail. We repeat that failure 

 at a single point may be fatal to the student's course. 

 For instance, when the student has got to No. 119 

 in the key, he finds himself in face of these alterna- 

 tives : — 



i Stem evidently pinnate or bipinnate, or with numerous 



\ divergent branches. 



( Stem not pinnate, not or sparingly branched. 



Now, suppose our student has before him a specimen 

 not pinnate, but with branches, and even with divergent 

 branches, it may come under either of the two alternatives; 

 and everything turns thenon theantithesis of the two words 

 "numerous'' and "sparingly"; and neither of these 

 words has any exact meaning — i.e. both of them are only 

 expressive of degree — and no standard is given us, or 

 could be given us, by which to tell whether the branches 

 on a given stem are to be called numerous or sparing. 

 Take again No. 186; the two alternatives before the 

 student are thus stated : — 



r,r i Leaves curled up or merely flexuose when dry. 

 \ Leaves crisped and twisted when dry. 



Now here the point to be settled, and a point on which 

 the whole future of the hunt depends, may be whether 

 a dry leaf is curled up or crisped and twisted ; and 

 surely the language used in the two cases does not, at 

 least to us, state a clear antithesis. Again, there is to 

 be found in many cases in these keys the use of adjec- 

 tives in the comparative degree, which always seem to 

 us bad in such a connection, because before you can find 

 the plant to which it is applied you must find and make 

 out the plant of which the positive degree is affirmed. 

 Under '" Fissidens," we are offered, for instance, these 

 alternatives — ''plant small" and "plant larger," " cells 

 obscure" and "cells clearer." Again, we can well 

 imagine a specimen in respect of which anyone would 

 be puzzled whether to accept the description " leaves 

 acute, yellowish" — not yellow, be it observed — or "leaves 

 rather obtuse, nerve and border orange, cells smaller." 

 Some of the characteristics given are open, also, to this 

 objection, that they are certain to vary as to their obvious- 

 ness, or even their accuracy, with the differences of the 

 individuals, as regards age, nourishment, and habit. 

 Linnseus was very right when he wrote, " Magnitudo 

 species non distinguit. Magnitudo mutatur a loco, solo, 

 climate : mutatur a copia alimenti in plantis, non minus 

 quam in animalibus." But let us be fair to Mr. Jameson : 

 he is not a sinner in these respects above many others 

 who have gone before him, and have worried generations 

 of students by the want of precision in the alternatives 

 which they present to his choice. In one respect his 



