456 



NATURE 



[June io, 1920 



/>■(-'•)' 



Number of ordinate^ used. 



4 6 : 



Semicircle (a: - .r'-)* 0*3927 0*4000 o'3949 0*3939 0*3934 



Quadrant (i -jr-')* 0-7854 07898 07868 07862 0*7859 



Parabola jr^ 0-33^3 0*3400 0*3350 03341 0*33 ?7 



sin.r 04597 04580 0*4593 0*4596 o*45g& 



log(i+x) 03863 0*3850 o'3859 0*3861 0*3862 



^ i7'83 17234 I7'97 17187 17186 



- o*693[ 0*6945 0*6937 06934 0*6933 



(3) \iY{x) = a + bx + cx^ + dx\ 



/W=s[f(-;'^).k(4).f(-;'^)] 



= J[F(o_*i464) + F(i) + F(o*8536)J. . {b) 

 A simple three-ordinate rule is therefore 



F(.r)^ = MF(l) + F(A) + F(y)]. 



\, 



In practice this is not quite so convenient as the 

 application of the two-ordinate rule. 



A. F. DUFTON. 



Trinity CoUei^e, Cambridge, May 20. 



P.S. — 1 thank Mr. C. F. Merchant for pointing out 

 in Nature of June 3 that the four-ordinate rule is 

 alread}' in use, and for giving a reference to Tcheby- 

 cheff 's, rules, with which 1 was unacquainted. The 

 positions of Tchebycheff's ordinates, as in (a) and (b) 

 'above, are inconvenient, and the rules obtained by 

 taking neighbouring ordinates attain simplicity with- 

 out great loss of accuracy. A. F. D. 



June 5. 



The Cost of Laboratory Fittings. 



It is evident from the correspondence which, has 

 followed the publication of the letter from me on 

 the subject of laboratory fittings that I must again 

 ask leave to trespass on your space in order to explain 

 that my remarks referred solely to fixed fittings, as 

 stated, embracing working benches, lecture tables, and 

 the like. I have no doubt that questions of actual 

 instrMments and ■ apparatus are of much greater im- 

 portance, but of these I have no right to speak. 



Perhaps I may be allowed to make myself clear 

 by reference to one or two specific directions in 

 which research on fixed fittings might possibly prove 

 useful. The present price of teak as bought in bulk 

 from a merchant is 30s. per cubic foot, and if impreg- 

 nated soft vi'ood could be substituted for bench and 

 table tops much saving would result. This impreg- 

 nation might be effected by precipitation, electrolysis, 

 oxidation (oils), or evaporation {e.g. silica solutions). 

 Again, bituminous materials with perhaps barytes 

 rolled into them might be investigated for use as a 

 thin layer on wood or concrete.- Soapstone is much 

 used in America and lave emaill^e in France, but not 

 as yet in this country. There are, further, certain 

 hard flooring plasters which should be very inert 

 chemically. An investigation is much needed into the 

 proper composition of bituminous materials for coat- 

 ing laboratory drains. Drains executed in wood thus 

 coated are in many cases much cheaper than glazed 

 ware drains. 



For repetition work such as locker doors and even 

 drawers pulped and stamped material might prove 

 economical if some standard could be agreed upon. It 

 should not be a very expensive matter to set on foot 

 some researches of this nature, and any effective re- 

 sults would, I imagine, be very welcome to institutions 

 at present faced with additions to their material 

 equipment. Alan E. Munby. 



9 Old Square, Lincotn's Inn, W.C.2. 

 NO. 2641, VOL. 105] 



The First Act of a Young Thrush. 



Since observations of the first acts of wild birds 

 immediately after hatching are very difficult, the 

 following may be worth recording. Last week I 

 went to look at a thrush's nest which I had found a 

 fortnight before, with four eggs in It. Two eggs 

 were hatched and two were not. As I was watching 

 the young birds, one of the two remaining 'eggs 

 cracked right across, and I saw the bird wriggle out 

 and toss the two halves of the shell out of the nest 

 by a convulsive movement of its back; but the 

 curious thing was that, before the bird was properly 

 free from the shell, it opened its beak — as if fo'r 

 food. I dug up a worm near by and offered it to 

 the bird, which swallowed it eagerly. I purposely 

 dug for the worm in a place from wHich I could see 

 the nest, and I feel sure" that the parent bird did not 

 come and feed the nestling meanwhile. A few minutes 

 later the other egg hatched, and the bird behaved 

 just as in the former case, opening its beak before 

 it was out of the shell. 



Now the question is : Was the opening of the bird's 

 beak a reflex or an "instinctive" act? If it were 

 reflex, it would presumably have been induced by 

 sudden exposure to tl\e new environment of open 

 air; and, obviously, such a reflex act would serve 

 the purpose of an "instinctive" one in this case. 

 Moreover, is it not a question whether anv " instinc- 

 tive" act at so early a stage can be anvthing more 

 than a reflex act thus adaptable to survival purposes 

 — by natural selection if need be? 



Honor M. M. Perrycoste. 



Polperro, Cornwall, May 30. 



Marat and the Deflection of Light. 



Carlyle's vivid portraiture of Marat as "horse- 

 leech " and savage revolutionist has rather obscured 

 the fact that this "friend of the people" was a learned 

 doctor of medicine, a physicist, and a physiologist. 

 It is true that Carlyle refers to him as " Renovator 

 of Human Science, Lecturer on Optics," but the mis- 

 take about the "horse-leech" is repeated in the same 

 passage. 



In Marat's "Notions elementaires d'optique " (1784), 

 p. 16, the following statement is made : 



"II est hors de doute, que les rayons de lumiere 

 changent toujours de direction dans le meme milieu, 

 lorsqu'ils passent k certaine distance d'un corps. Se 

 trouvent-ils dans la sphere d'attraction ? ils se repliant 

 jusqu'^ certain point a sa circonference, et se pro- 

 longent ensuite en droite ligne." 



This at first glance may appear a remarkable 

 anticipation of recent discoveries in physics, but in 

 reality the conclusion is based on wholly false 

 premises, as further reading of the pamphlet will 

 disclose. W. A. Osborne. 



University of Melbourne, April 22. 



British and Metric Systems of Weights and Measures. 



On p. 355 of Nature of May 20 Mr. M. E. Yeat- 

 man in a letter on the above subject says : " It seems 

 that the advantage of any given system of weights or 

 measures lies largelv in the facilities that it offers for the 

 division of a sum or quantity into equal parts"; and 

 I have seen " facility of factorisation " claimed before 

 as one of the merits of the British system. As an 

 engineer who "figures frequently," I fail to appre- 

 ciate this fetish of factorisation. One uses a slide- 

 rule and logs, and never worries about factors. Will 

 Mr. Yeatman, or someone else, demonstrate the use 

 of factors in practical calculations, bearing in mind 

 the use of slide-rules, calculating machines, and logs? 



The metric svstem seems to be gaining ground in 

 spite of the lack of factors disclaimed for it. 



Alfred S. E. .Ackermann. 



