July 9, 1896] 



A\~l TURE 



225 



be in sunn; way tlfi-iemlcnl on ihe laci thai any varielyut .siil|iluii- 

 increases in conductivity with rise of temperature. Unless it is 

 suggested that we did not know one temperature from another, 

 I fail to understand this criticism. Ru HARD Threlfai.l. 

 University, Sydney, N.S.W. 



I si;e no reason to recall what I have said regarding tlie 

 general attitude of chemists and physicists on the question of 

 the influence of minute traces of impurity ; and when I come 

 across the remark in Messrs. Threlfall, Brearley and Allen's 

 paper in the Phil. Trans., that " it is not too much to say that 

 the electrical action of most bodies in a pure state is entirely 

 unknown at present," I feel there is not much difference of 

 opinion between us. 



Then, as to my being guilty of that unpardonable crime — 

 pedantry — it has always seemed to me that those of us who 

 undertake scientific work should also strive to be scientific, i.e. 

 exact, in their use of language. Those who had the great good 

 fortune to be present a year or so ago at the Nati'RE dinner, 

 and to hear I luxley's marvellous speech — almost, if not the last 

 he delivered — will recollect how strongly he insisted on the 

 importance of greater care being taken in the writing of papers 

 describing scientific inquiries. In a conversation I had with 

 him afterwards, he greatly lamented the careless manner in 

 which such work was too frequently done. 



Now if pure mean " free from mixture," a pure substance 

 must, as I have said, ever remain an ideal conception ; the 

 purist must ever regard all things as impure. Prof. Threlfall 

 tells us that "the word pure has no significance except with 

 respect to a definite state of the art of chemistry." I would 

 rather accept the meaning w hich is to be found in the dictionary, 

 paie Stas even : and would prefer to assert that the word too 

 frequently has no significance except with reference to an in- 

 definite state of the mind of the person — chemist or physicist — 

 using it. To my mind, there can only be degrees of impurity — 

 not of purity. 



Whatever time Prof. Threlfall and his colleagues may have 

 S]5ent in seeking to purify' sulphur, the fact remains that their 

 experiments were made with sulphur which they obtained by 

 chance, and that the only method of purifying it they adopted 

 was to distil it several times in vacuo, after filtering it while 

 molten through glass w'ool and platinum gauze, and then to 

 fuse it in vacuo — in order, they tell us, to get rid of gases 

 (probably water vapour, they say) given oft' even from the purest 

 samples. But distillation in vacuo, even when followed by 

 fusion in vacuo, can scarcely be regarded as a process which 

 "exhausts the resources of physics, including chemistry." 



"Chance" sulphur is prepared by burning sulphuretted 

 hydrogen. It is probably impossible to burn sulphur without 

 producing some sulphuric acid. Messrs. Threlfall, Brearley 

 and Allen, however, do not even refer to the possibility of its 

 presence, and apparently took no precaution whatever to 

 eliminate it, if present. 



They tell us that on breaking up .such sulphur after it had 

 been strained while molten through glass wool and platinum 

 gauze, it emitted a horrible smell of gas-lime, " which shows 

 that it requires to be distilled if sure results are to be obtained." 

 I imagine, therefore, that the sulphur they used initially was by 

 no means so remarkably " pure"; as they also state that gases 

 were given oflf even from the purest samples when fused in 

 vacuo after distillation, it may well be doubted whether so sim]ile 

 a process as mere fusion could suffice to efi'ect the necessary final 

 purification. 



Prof. Threlfall's statement that conducting "mixtures" were 

 caused to become non-conducting by annealing, is apparently a 

 good answer to my criticism ; but by no means finally disposes 

 of it. The structure of the two materials may have been very 

 different, and such in the one case as to allow an impurity 

 to act, which in the other case might be inoperative. By 

 my reference to the conductivity of sulphur at temperatures 

 above its melting point, I meant to imply that the behaviour 

 described afforded indication of the presence of impurity ; for I 

 do not believe that even molten sulphur is a conductor. Of 

 course, at present, this is but an opinion, but it may not be in- 

 appropriate to direct attention to the recent most remarkable 

 observations of Dewar and Fleming on bismuth, showing that 

 an amount of impurity altogether beyond detection by chemical 

 means may entirely alter electrical properties. 



I still, therefore, regretfully retain my opinion, and fear that, 



NO. 1393, VOL. 54] 



notwithstanding the great care lavished on the work of Prof. 

 Threlfall and his colleagues, it will be necessary to repeat it, 

 perhaps over and over again — a possibility which they 

 apparently themselves foresee in the introduction to their paper 

 — before so remarkable a conclusion as that they have arrived 

 at can be regarded as establi.shed. 



H. E. Armstrong. 



Increasing the Efficiency of Rontgen Ray Tubes. 



Mr. J. C. Porter, in a letter in Nature of June i8, 

 describes a method of increasing the efficiency of a Crookes' 

 tube. I have for some weeks used another very simple method 

 to obtain the same result. This consists in placing the flame of 

 a small glass spirit-lamp in the angle formed by the Crookes'' 

 tube and the wire passing to the kathode, and allowing a series 

 of small sparks to pass to the flame from the wire. 



Burnley, June 29. T. t!. Crump. 



THE POSITION OF SCIENCE AT OXFORD. 



"XX /"HILST the study of natural science has been pro- 

 ' ' grassing rapidly in other universities and colleges 

 during the last ten or fifteen years, it is a matter of 

 common knowledge that it has progressed very slowly 

 indeed in the University of Oxford. It would be in- 

 correct to say that it has not progressed, for there has 

 been during the last few years a steady, though very 

 gradual, increase in the numbers of men reading for 

 honours in the final school of natural science. In 1885 

 twenty-two men obtained honours in sciencce, in 1895 

 there w-ere forty-three names in the class list, and a 

 rather larger number in 1894. The school has just 

 doubled itself in ten years, but for all that the numbers 

 are still small, and out of all proportion to the provision 

 that has long existed for science teaching in the Univer- 

 sity. It must be understood at the outset that the 

 University, considered as a body separate from the 

 colleges which compose it, has not dealt ungenerously 

 with science. The staff of professors, and the emolu- 

 ments attached to their chairs, compare favourably with 

 those of any other unixersity in Great Britain ; and 

 Oxford actually set the example, at great cost to itself, 

 of building a museum and equipping laboratories for 

 educational purposes. Moreover, the opportunities of 

 scientific study in Oxford are greatly enhanced by the 

 existence within the precincts of the museum of a first- 

 rate scientific library, such as is not possessed by any 

 other college or university in the kingdom. It is a 

 strange thing that when it has so many advantages, 

 Oxford has allowed itself to be completely outstripped 

 in this particular path of intellectual progress. 



It is the purpose of the present article to discuss the 

 possible causes of comparative failure of the science 

 school at Oxford. A complete failure it is not, for, how- 

 ever poor its numerical results may be, it has long been 

 recognised that the attainments of the limited number 

 of scientific men which it turns out compare well with 

 those of men who have been educated in other places. 



It is commonly supposed that the prime cause of the 

 insignificant numerical result is the small encouragement 

 given to scientific study in the shape of fellowships and 

 scholarships ; and those who hold this opinion believe 

 that if the colleges were to do what is conceived to be 

 their duty in this respect, the science school would 

 progress by leaps and bounds. 



With respect to scholarships and exhibitions, it is 

 apparent, from an analysis of the figures, that science 

 does not get what may rightly be held to be its due. 

 The Utiivcrsily Calendar for 1896 shows that there are 

 in O.xford some 500 scholarships of an annual value of 

 £%o a-piece, and in addition some 225 exhibitions, the 

 annual \alue of each of which may be placed at £\o. 

 These figures apply only to college scholarships and 

 exhibitions, and so it appears that the colleges, apart 



