September 3, 1908] 



NA TURE 



413 



note tjoc. cit., August) that Prof. Morse's latest corrected 

 result (his own corrections) for the osmotic pressure at 

 0° C. of his i-o weight normal aqueous solution of cane 

 sugar is 24-45 atmospheres, while ours, by the method 

 under discussion, was 24-5 atmospheres. 



Evian les Bains. Berkeley. 



The Rotation of a Crystal of Tourmaline by Plane 

 Polarised Light. 



{Preliminary Note.) 



When a beam of plane-polarised light is incident normally 

 on a plate of tourmaline cut parallel to the optic axis, it 

 will be absorbed or transmitted depending upon whether 

 the axis of the tourmaline is parallel or perpendicular to 

 the plane of polarisation of the incident light. If the 

 arrangement is such that the light is absorbed, then in 

 a given time a definite amount of heat energy will have 

 passed from the source of light into the plate of tour- 

 maline, and if, as is necessarily true, the former is at a 

 high temperature while the latter is at a low temperature, 

 it is plain that the entropy of the system will have been 

 increased. The same increase in entropy would not have 

 taken place if the orientation of the tourmaline had been 

 such that the light had been transmitted. 



Since the entropy of a system always tends to increase, 

 it seemed to be of some interest to try what would happen 

 if a plate of tourmaline was suspended so as to be free 

 to rotate about an axis perpendicular to its fares, and then 

 allowing a beam of plane-polarised light to fall on it in 

 the direction of the axis of rotation, the arrangement 

 being such that the plane of polarisation should make an 

 angle of 45^^ with the optic axis of the tourmaline. 



The experiment was tried by the author during the 

 month of July, 1908, at the Rouss Physical Laboratory 

 of the University of Virginia, and the results, though not 

 absolutely conclusive, mdicate that a moment acts on the 

 tourmaline tending to set its optic axis parallel to the plane 

 of polarisation of the incident light. In other words, the 

 system tends to arrange itself so that as large a percentage 

 of the light as possible shall be absorbed. 



The apparatus used consisted of a fine plate of tour- 

 maline, I cm. square, and 2"96 mm. thick, weighing almost 

 exactly one gram. This was fastened to one end of a short 

 straight copper wire to which was also fastened a small 

 plane mirror. The system was suspended in a suitable 

 vessel having a plane glass top and bottom, which could be 

 exhausted, the suspension being such that the system had 

 a period of 29*5 seconds. From the period and the moment 

 of inertia of the system it was calculated that the moment 

 necessary to give a deflection of i cm. with the scale at 

 a distance of one metre was 2"5Xio-° dyne-cm. A beam 

 of approximately parallel light from an arc was reflected 

 in a vertical direction from a plane silver mirror, and was 

 rendered plane polarised by a Nicol before entering the 

 vessel. 



Owing to the fact that the suspended system was not 

 perfectly symmetrical with respect to its axis (A rotation, it 

 was found that the zero moved constantly in one direction 

 when the light was allowed to fall on the tourmaline. This 

 was undoubtedly due to radiometric action, and possibly 

 also to the pressure of light. The motion was, however, 

 much slower when the angle between the plane of polarisa- 

 tion and the optic axis of the crystal was 4-45° than when 

 it was —45°. An average of twenty trials gave iso 

 seconds as the time to get: a deflection of 30 cm. in the 

 first case, while the time required for the same deflection 

 when the angle was —45° was 90 seconds. These trials 

 were all made with as high a vacuum in the vessel as 

 could be obtained by means of the Gaede rotating mercury 

 pump. The experiment was repeated at various pressures 

 up to 10 or 15 cm. of mercury ; the results were qualita- 

 tively the same, that is, the rate of deflection was much 

 slower in every case when the angle mentioned above was 

 -(-45° than when it was —45°, but the motion of the zero 

 in one direction, although somewhat different at different 

 pressures, could not be avoided. 



If the cause of this motion of the zero is what was 

 slated above, then it ought to disappear when the sus- 



KO. 2027, VOL. 7S] 



pended system is made symmetrical about the axis of 

 rotation. 



These experiments were only preliminary, and during the 

 coming winter a more careful investigation of the question 

 will be carried out at the physical laboratory of the Johns 

 Hopkins University. John A. Anderson. 



Physical Laboratory, Johns Hopkins University, 

 August 15. 



Access to Chemical Woiks. 



Students of chemistry so often complain of the extreme 

 difficulty of obtaining access to chemical works that it 

 may be well to bear in mind that the universities and 

 technical schools of the United Kingdom have perhaps a 

 remedy for what is a serious obstacle to a proper study 

 of chemistry. The amount of chemicals now consumed in 

 educational laboratories must be enormous, and, as regards 

 ordinary materials, quite sufficient to employ a large 

 factory. A very large proportion comes from Germany. 

 Let tlie universities start their own cooperative factory for 

 the manufacture of the acids and salts they require. Let 

 chemical students both have free access to it and put in 

 part of their chemical course as workers there. The action 

 of several universities in running their own farms for the 

 benefit of agricultural students supplies a kind of prece- 

 dent. When motives of education, patriotism, and economy 

 point the same way, the scheme is worth consideration. 



Chemist. 



FLY FEVER IN AFRICA. 



IX the Times of June 27 there appeared an article 

 on " fly fever " in Africa, and the suggested 

 destruction of big game. It apDears that Prof. Koch 

 has lately suggested that the African big game should 

 be exterminated in order to destroy the principal 

 means of nourishment of the tsetse fly. As this fly 

 is the carrier of the infective agent of " fly fever," its 

 extinction would, in Koch's opinion, blot out the 

 disease. The members of the German Society for the 

 Preservation of Game in East Africa are, naturally, 

 opposing tooth and nail this proposed drastic measure 

 of Prof. Koch. They deny the truth of his conclu- 

 sions, and hold that the fly disease can exist where 

 there is no big game, and, further, that there are 

 other methods of getting rid of the fly without 

 destroying the game. 



This question is a complicated one, and perhaps if 

 I give a short sketch of the history of the " tsetse- 

 flv" disease " I shall best clear the way to its better 

 understanding. 



When I went to Zululand in 1894 at the request of 

 the Natal Government, to investigate an outbreak of 

 nagana among native cattle, I was unaware that 

 nagana and tsetse-fly disease were one and the same. 

 I'urther, I believed, with the rest of the world, that 

 the fly disease was caused by the poison of the tsetse 

 fly, just as an animal is killed by the poison of a 

 snake. Soon after arriving in Zululand a parasite, 

 the Trypanosoma Irucei, was discovered in the blood 

 of the affected animals, and it soon became apparent 

 that flv disease and nagana were one and the same. 

 By suitable experiments it was demonstrated that this 

 trvpanosome is the cause of the disease. The next 

 fact made out was that the trypanosome could be 

 conveyed from sick to healthy animals by the local 

 species of tsetse flv (Glossiiia worsitans).^ .\s this 

 tsetse flv onlv remained infective for forty-eight hours, 

 it was evident that it must get the parasite some- 

 where, and it seemed most probnble that this would 

 be from the wild animals living in the " fly countr>-." 

 This was found to be the case, and the Trypanosoma 

 hrucei was demonstrated in the blood of the buffalo, 

 wildebeeste, koodoo, and other big game, both micro- 

 r.cooicnllv and bv means of inoculation experiments. 



