December i i, 1902] 



NA TURE 



127 



The Paradox of the Piano Player. 



When a number of notes in different parts of the keyboard 

 of a pianoforte are struck by means of levers actuated by a 

 common pneumatic pressure, it appears to be the universally 

 prevailing belief that the only variations possible are those in 

 which the whole chord is made to sound louder or softer by 

 increasing or decreasing the pressure. It is commonly regarded 

 as an impossibility to vary the relative intensities of the sounds 

 produced by the various notes so as to make, e.g., the bass parts 

 sound louder and the treble softer, or vice vtrsA. 



On the other hand, dynamical considerations suggest that the 

 intensities of the sounds excited in the different strings of the 

 piano depend, not only on the total pressure applied to the 

 mechanism, but also on the way in which this pressure is made 

 to vary during at least part of the interval from the instant at 

 which the key is first touched to the instant at which the hammer 

 leaves the strings. A short, sharp impulse suddenly cut off should 

 produce its greatest effect on the notes of higher pitch, while a 

 heavy, sustained or increasing pressure should make its eftect most 

 marked on the lower notes of the instrument. During the last few 

 months, I have given considerable attention to the practical 

 application of this theory, and the effects which I find it possible 

 to produce, provided that the accent nation is performed at exactly 

 the right instant of time, are most remarkable. The treble or 

 bass parts may be made to stand out in so conspicuous a way 

 as to make it difficult to believe that different notes of the chords 

 are not struck by different human fingers. The matter opens up 

 a wide field of discussion, and suggests considerable possibilities 

 in the way of quantitative laboratory measurements. For the 

 present, it may be sufficient to suggest that those of your readers 

 who possess the new musical instruments of the twentieth 

 century suitable for the purpose should, if they have not already 

 done so, perform the experiment for themselves : they will soon 

 be rewarded by being able to enjoy their music in a way they 

 have never enjoyed it previously. G. H. Bryan. 



Cost of Scientific Education in Germany and 

 England. 



I notice, in the issue of Nature of December 4, that you 

 quote Mr. Holzapfel's letter to the Times on the cost of scientific 

 education in Germany and England. Although, unfortunately, 

 there can be no dispute as to the great difference between the fees 

 charged in' Germany and in England, I think it right that the fees 

 of King's College should be correctly stated. The sum quoted 

 by Mr. Holzapfel represents the charge made for chemistry and 

 physics ; for chemistry only it was 34/. lis. for the year. I have 

 no knowledge of the amount of instruction which the other son 

 obtained for 7/. at Aachen. Walter Smith, Secretary. 



King's College, London, W.C, December 9. 



THE REPRODUCTIOX OF COLOURS BY 

 PHOTOGRAPHY. 



THE services which photography has rendered to 

 science are now well recognised, and its value for 

 purposes both of observation and record is well known and 

 admitted. It is probably not so well known that methods 

 now exist by which not only the form, but the colour, of 

 natural objects can be represented with approximate 

 fidelity. We are fortunate in being able to illustrate 

 this fact by a plate giving some excellent reproductions 

 of birds' eggs, produced under the superintendence of 

 Mr. H. E. Dresser, entirely by photographic methods, 

 and without the intervention of an artist. 



There is no need to dwell on the value of such 

 work. For many scientific purposes it is as important 

 to record colour as shape, and if this can be done in a 

 trustworthy manner, a new and useful power is placed at 

 the disposal of the teacher of science and of the writer 

 of scientific books. The difficulty about the three-colour 

 process of photography is that it is extremely difficult to 

 make certain that the colours are reproduced with 

 sufficient accuracy for scientific work. Accuracy 

 enough for pictorial purposes is easily attained, but 



NO. 1728, VOL. 67] 



absolute truth to nature is quite another thing. The 

 reasons for this are various. The photographic gradation 

 of light intensities, in the case of both white light and 

 of its various components, is generally different from the 

 visual gradation, and even if accuracy is ensured in a nar- 

 row range of tones, it is hardly possible to make certain of 

 its being secured in wider ranges. Another difficulty is that 

 pigments have to be employed, and such pigments can 

 never, of course, give pure colours. The consequence of 

 this is that in the production of the picture it is necessary 

 to vary the intensity of the different colouring agents 

 employed until a satisfactory result is obtained. There 

 is thus considerable room for judgment and dexterity, 

 and the final result is not automatic, but depends on 

 the artistic skill of the person who produces the picture. 

 The whole process is. it must be admitted, of the 

 character of a makeshift, but at the same time, when 

 carefully employed it is a makeshift of considerable 

 practical use. 



Mr. Dresser, in the article printed below, deals only 

 with the representation of natural objects for purposes of 

 book illustration. An equally valuable application of the 

 process is for the production of lantern slides for purposes 

 of demonstration, and, as many of our readers are well 

 aware, the process is beginning to be largely used for 

 such purposes. A lantern slide coloured by hand is at 

 best but a poor thing, and though a few very skilful 

 operators— such as Mr. Cyril Davenport, of the British 

 Museum — have by a combination of microscopic sight 

 and great deftness of manipulation succeeded in 

 producing some remarkable results, even these will 

 hardly stand the large amount of magnification required 

 by the lantern. Now a slide made by the three-colour 

 process will stand as much enlarging as any ordinary 

 photographic slide, and will give a reasonably close 

 approximation to the natural colours of the subject. 

 The process is applicable to any specimen which can be 

 photographed. Excellent reproductions of microscopic 

 objects have thus been produced ; botanical specimens, 

 birds, beetles and butterflies have all been rendered with 

 great beauty and with really close accuracy to nature. 

 Those who were present at Prof. Poulton's lectures at 

 the Royal Institution last session had the opportunity of 

 admiring the exquisitely coloured pictures he showed of 

 insects, all produced by the process, which, first practi- 

 cally demonstrated by Mr. F. E. Ives, has since been 

 further developed by Mr. Sanger Shepherd and others 

 in this country. 



Although, as said above, absolute accuracy is very 

 difficult, or even impossible, to ensure— certainly not by 

 automatic means — it is not too much to say that any 

 photographer ought, after a very little practice, to be 

 able to produce useful and serviceable illustrations for 

 lecture purposes if he is content with something which, 

 though not perhaps the best possible, is infinitely su- 

 perior to anything which can be produced by painting 

 an ordinary monochrome lantern slide. 



Mr. Dresser in his remarks places, perhaps, needless 

 stress on the difficulties of the process, and we are not 

 quite disposed to agree with him as to its unsuitability 

 for many purposes which he mentions. Although the 

 exposures he gives may have been necessary by reason 

 of the conditions under which his pictures were produced 

 —namely, the photographing of the objects life size 

 through a ruled screen and by the use of daylight at a 

 time of the year when the light is not very good— it 

 is a very different matter when it is required to 

 produce illustrations for the lantern. In an ordinary 

 studio, the exposure may take from, say, three minutes 

 to a quarter of an hour through the red screen, which of 

 course takes the longest time, while for out-of-door views 

 in bright sunshine, with a moderate aperture of the lens, 

 it is a matter of seconds only. 



