5'0 



NA TURE 



[March 31, i8y8 



illustrations; (3) description of the photographic apparatus 

 employed, and (4) the method of measuring the pictures 

 obtained. The two last chapters will be very valuable 

 for any one proposing to undertake the difficult task of 

 photographing the clouds, and of determining their 

 heights and movements. In referring to the various 

 attempts at cloud classification, we do not find any 

 mention of " Cloudland," by the late Rev. W. C. Ley. 



Proceedings of the London Mathematical Society. Vol. 

 xxviii. Demy 8vo, pp. 594. (London : Francis 

 Hodgson, 1897.) 



This collection of thirty-four original papers on every 

 branch of mathematics affords abundant evidence that 

 English mathematicians are not behindhand in moving 

 with the times. If proof be needed that the younger 

 generation of mathematicians are quite following in the 

 lines of those that have gone before them, it may be 

 sufficient to mention that at least six of the papers are 

 by men who have graduated at Cambridge since the year 

 1886. As might be expected, " Partitions " and " Groups " 

 occupy a prominent place, seven of the papers being de- 

 voted to them. The former of these two subjects is 

 ably introduced by Major MacMahon, F.R.S., in his 

 address on " Combinatory Analysis," delivered on re- 

 tiring from the office of president ; and the publication of 

 the outlines of seven lectures on the " Partitions of 

 Numbers," delivered by the late Prof. Sylvester at King's 

 College, London, in 1859, is another important feature. 

 On the other hand, " hyper-Euclidian geometry " is con- 

 spicuous by its absence, and applied mathematics is 

 represented by eight papers only. 



During the past year the London Mathematical Society 

 has lost two members in addition to the late Prof. 

 Sylvester : the Rev. Alexander Freeman, who died on 

 June 12, 1897, and Lieut.-Colonel John Robert Campbell, 

 who died on June 23. Colonel Campbell, besides serving 

 on the Council, was a benefactor to the Society, and we 

 understand that had it not been for his munificence it 

 would have been impossible for the Society to issue such 

 large and interesting volumes of Proceedings as the one 

 now before us. 



First Year oj Scientific Knowledge. By Paul Bert. 

 Translated by Madame Paul Bert. Revised and 

 partly re-written by Richard Wormell, D.Sc, M.A. ; 

 and Montagu Lubbock, M.D. Pp. vi -I- 417. (London : 

 Relfe Brothers, Ltd. Paris : Armand Colin and Co.) 



This is a revised edition of a work which has had a very 

 successful career, but is constructed upon a plan which 

 has little to commend it. The revision has consisted in 

 bringing the information into line with current scientific 

 knowledge, the plan of the book remaining as in the 

 original. The rudiments of zoology, botany, geology, 

 physics, chemistry, animal physiology, and vegetable 

 physiology are all described in the four hundred pages 

 which constitute the text, so that the book is com- 

 prehensive in its scope, if nothing else. The chief fault 

 we have to find is that far too many technical terms are 

 defined and used, so that the unfortunate pupils who are 

 introduced to natural history by this book will be given 

 the idea that science consists chiefly of words of Greek 

 origin, and an unpronounceable terminology. 



Who's Who, 1898. Edited by Douglas Sladen. Pp. 

 xviii + 846. (London : A. and C. Black, 1898.) 



This is undoubtedly the handiest biographical dic- 

 tionary and compendium of information, referring to 

 prominent persons and their doings, in existence. It 

 contains nearly . seven thousand biographies — mostly 

 autobiographies — of the leading men and women of the 

 day, and a large amount of information in addition. 

 Among the general contents of interest to men of science 

 is a list of Royal, National and learned societies, showing 



NO. 1483. VOL. 57] 



the addresses of the societies, secretaries' names, annual 

 subscriptions and other conditions of membership. We 

 notice also a table of university degrees, with the correct 

 explanation of each, a list of chairs and professors in the 

 great universities of the United Kingdom, arranged 

 alphabetically by their chairs, and a list of Fellows of the 

 Royal Society (most of whom appear among the bio- 

 graphies). The volume is one to be kept on the writing 

 table for ready reference ; and it possesses the merit of 

 including in its pages biographical details of more men of 

 science than usually figure in similar reference books,, 

 though even now some of the minor literary lights could 

 be struck out with advantage to make room for well- 

 known scientific men who have been omitted. 



LETTERS _ TO THE EDITOR 



[The Editor does not hold himself responsible for opinions eX' 

 pressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of NATURE. 

 No notice is taken of anonyrnous communications.'] 



Mechanism of Self-fertilisation in the Banana. 



I PROPOSE to describe here the mechanism of fertilisation inr 

 the banana plant. 



(i) Packing of the inflorescence. — The inflorescence in this 

 plant is packed air-tight in the large red-coloured bracts, whose 

 margins are secured in place by a sort of cement. If we re- 

 move one of these bracts and examine the buds within, we find 

 that the reproductive organs are also packed air-tight in the 

 perianth. A closer examination of this packing is necessary to- 

 understand its efficiency. 



The perianth consists of an outer whorl and an inner whorl ; 

 the outer one consists of three sepals, usually united into an 

 elongated concave hood. (Sometimes, instead of the three being 

 united together, only two are united and one is free, which in 

 the bud is partially overlapped by the other. Very rarely the 

 three are quite free. In about fifty examples I examined, I 



>/\^/ 



'K-ffri^ 



S t^.ttoi. 



F'J 



got only one flower with all the three sepals distinct.) The 

 margins of the hood are folded inwards, so that they overlap 

 each other. There is no fixed rule as to which is the outer, and 

 which the inner, sepal. Sometimes the right overlaps the left, 

 or vice versd. This overlapping is not complete throughout 

 their length, and cannot possibly be so. For, to ensure the 

 packing being air-tight, the pectinate inflorescence must neces- 

 sarily be concavo-convex, i.e. convex without and concave 

 within ; and obviously any cylindrical tube bent concavo- 

 convexly must necessarily leave a gap or a window at the 

 bottom (see Figs, i and 2). 



This window would be a very weak point in the packing, and 

 hence most accessible to insects or other injurious agencies if 

 the outer packing became loose by accident. This weak 



