534 



NA TURE 



[April 9, 1896 



pulling, and a connected inner ring pulling or pushing. I do 

 not myself see that the existence of acceleration— in the strictly 

 mathematical sense — in a ring rotating with uniform angular 

 velocity is any real disadvantage ; but if Mr. Wilberforce thinks 

 it is, then as an accomplished and not too materialistic mathe- 

 matician, he will, I hope, be satisfied with the case of a ring 

 formed of matter repelling as, say, the inverse square of the 

 distance. 



Judging by what has passed, I think an attempt to put clearly 

 the problem as it exists in a magnetised ring may be of service. 

 Suppose the diagram on the preceding page to represent 3 out of, 

 let us say, 360 — or any larger number deemed requisite to justify 

 treating arc and chord as practically identical — equal elements 

 of a ring. When unmagnetised suppose each to be of length 

 /, and to be separated from its neighbours by extremely small 

 air gaps of width A,. The elements may be supposed to lie on a 

 smooth horizontal table. When uniformly and equally magnetised 

 each element will change in length, let us suppose lengthen, by 



The Sacred Tree of Kum-Bum. 



Referring to the letter of Mr. W. T. Thiselt(m-Dyer, you 

 have published under the title " The Sacred Tree of Kum-Bum " 

 (March 5, 1896, p. 412), I may add that Dr. Kanitz (as I am 

 told by Dr. S. Bretschneider) identifies the tree as the Ligjtstrina 

 aniiirentis. With the permission of the Vice-President of the 

 Imperial Russian Geographical Society, I send you herewith a 

 photograph (natural size, untouched) of about two-thirds of the 

 piece of the stem of that tree in possession of the I.R.G.S., 

 which received the piece from a lama who visited Kum-Bum 

 late in the autumn of 1894. 



The wood is very light, the bark thin (the diameter of the 

 piece being 46 x 47 on one end, 45 x 46 millimetre on the 

 other ; the thickness of the back does not exceed i millimetre, 

 being the third of it in some places of the circumference), its 

 surface is somewhat rugged, as from desiccation (to be seen on 

 the photo), the outer tissue is grey, translucent, the colour of the 

 bark reddish, the signs in it of a golden hue, and can be dis- 



Natural size in breadth, about two-thirds in length, of the piece in possession of the Imperial Russian Geographical Society, St. Petersburg. 



the same amount 5/. This lengthening is due presumably to j 

 several causes ; of which one is the tension which experiments | 

 by Dr. Taylor Jones and others have shown to be nearly B^/Stt, ' 

 so long as the permeability is large. For shortness, I shall treat I 

 this tension as actually B"/8ir up to actual contact. The contri- 

 bution to 5/ due to this stress may be taken as BV/8irE, where I 

 E is an elastic modulus, which, strictly speaking, varies with B. ' 

 Suppose now B so chosen that 5/ = A, so that the width of the 

 air gaps reduces to zero. Then Mr. Wilberforce's view would 

 seem to be that at^the instant the gaps close the stress producing I 

 the lengthening B-//8irE becomes zero, and comes into existence I 

 again only when the consequent shortening of the elements ; 

 reopens the gaps ; in this way a species of oscillation would 

 ensue. Prof. Ewing, I rather fancy, would suppose the stress i 

 to exist until there is absolute continuity of matter. To deal I 

 with either view, suppose that at the first instant of contact, by | 

 some process of welding which leaves the material unaltered, ' 

 the 360 elements transform into a soHd ring. Call the state just 

 before welding M, and that when equilibrium has ensued after i 

 welding N ; then the following views seem to be or have been 

 held : 1 



(1) (By Mr. Shelford Bid well. Dr. More, and Prof. Ewing | 

 [originally]) that the ring in N is shorter than my reasoning '• 

 makes it in M by 2 x (360 B2//87rE). ' 



(2) (By Prof. Ewing [now] and Mr. Wilberforce) that the 

 ring in N is shorter than in M by i x (360 B-//8wE). I 



(3) (By, I believe, Prof. J. J. Thomson, Dr. Taylor Jones, 

 and myself) that the ring in N, if shorter than in M, is so by a ' 

 less amount than i x (360 B^Z/S^rE). Charles Chree. : 



March 23. 



NO. 1380, VOL. 53] 



cerned through the epiderm, but are best seen when the latter is 

 pulled off. 



A. Grigoriev, 

 Secretary of the Imperial Russian Geographical Society, 

 St. Petersburg. 



A Jamaica Drift Fruit. 



In connection with the article by Dr. Morris on a Jamaica 

 drift fruit, which appeared in Nature of November 21, 

 1895, I ^™ ^ble to supplement it by a record of the tree in 

 Trinidad. On March 10, after a personal visit to the vicinity, I 

 despatched Mr. Lunt, my assistant — who may be remembered 

 as being attached to Mr. Bent's expedition to the Hadramaut — 

 to search for the tree or trees stated to have been seen by Messrs. 

 Crueger and Devenish, I am glad to say that I have now Mr. 

 Lunt's report on the trip before me, and still better, I have 

 specimens taken from the trees, which show that Saccoglottis 

 amazotiica may still be regarded as one of our forest trees. Mr. 

 Lunt followed as directed the course of one of the rivers, and 

 found the tree in more than one place. He was able to bring 

 fresh fruits, leaves and buds, but no flowers were to be had. 

 Mr. Lunt notes that the specimens brought show that the fleshy 

 outer layer or sarcocarp is not, as has been supposed, worn away 

 by water, but owing to its palatable character forms the food of 

 numerous fruit-eating animals, and that on falling to the ground 

 it is further cleaned by ants. 



It appears that after the sarcocarp is eaten away the seeds rest 

 upon the ground until the occurrence of hea\'y floods, when they 

 are washed away by the currents. 



