April 13, 1905J 



NATURE 



573 



The new regulations recently issued by the War Office, 

 under which commissions in the Army may be obtained by 

 university candidates, provide that commissions shall be 

 allotted each half-year to the University of London. To 

 satisfy the requirements of the regulations, the Senate has 

 appointed a nomination board for military commissions 

 which will nominate qualified students for commissions, and 

 arrangements have been made for the instruction of candi- 

 dates in military subjects. To be eligible for a commission, 

 a candidate must have graduated as an internal student, 

 and this involves three years' study at one or more of the 

 schools of the university. Before a student can be nomin- 

 ated for a commission he must, as a rule, have attended 

 the various courses of instruction in military subjects in 

 the university, and he must have been attached for two 

 periods of six weeks, or for one period of twelve weeks, to 

 a regular unit. Courses of lectures in military subjects 

 are being given at the University of London by Colonel 

 H. A. Sawyer, P.S.C., and Lieut.-Colonel F. N. Maude, 

 P.S.C, late R.E. 



SOCIETIES AND ACADEMIES. 

 London. 



Royal Society, February 23. — "Two Cases of Trichromic 

 \ision." By Dr. F. W. Edridge-Green. Communicated 

 by Dr. F. W. Mott, F.R.S. 



One case (Prof. J. J. Thomson) sees only three colours 

 in the bright spectrum — red, green, and violet. He can 

 distinguish nothing of the nature of pure yellow, like the 

 sensation given him by the sodium flame, in the spectrum. 

 There is no definite colour to him at the portion of the 

 spectrum where the normal sighted see pure blue. Reddish- 

 green would describe the orange and yellow regions and 

 greenish-violet the blue. \ 5950 (orange-yellow) is the 

 point which differs most from red and green. There was 

 no shortening of either end of the spectrum. 



Difference of Hue Perception. — The author then tested 

 him with his apparatus for ascertaining the size of different 

 parts of the spectrum which appear monochromatic, and 

 found him defective in distinguishing differences of hue. 



Colour Mixtures. — Tested with Rayleigh's apparatus for 

 matching spectral yellow by a mixture of red and green, 

 the mi.xed colour of his match always appeared green to 

 the author. 



Classification Test. — Only a few colours were selected in 

 each case. On being asked to pick out all the yellows he 

 chose those with orange in them. He had considerable 

 difficulty in matching the colours. In common with the 

 cases previously observed, the effects of simultaneous 

 contrast were much more marked than in the normal 

 sighted. Two wools changed colour to him on being con- 

 trasted, when no change was evident to the author. 



Lantern Test. — He correctly named the red, green, and 

 violet with and without the neutral glasses, and saw them 

 at the normal distance. He had difficulty with yellow 

 and blue. He called pure yellow " greenish yellow." 



The other case is that of Mr. P. S. Barlow, a 

 research student in the Cavendish Laboratory, and was 

 similar in most respects to the above. 



The author uses the term trichromic as a statement of the 

 fact that persons having this vision see only three colours 

 in the bright spectrum, whilst the normal sighted see six, 

 and may, therefore, be designated hexachromic. It is 

 probable that the appearance of the bright spectrum to the 

 trichromic is very similar to that of a spectrum of feeble 

 luminosity to the normal sighted, in which only three 

 colours — red, green, and violet — are seen. The defective 

 difference perception which is found in these cases accounts 

 for most of the facts. Both these cases are bordering on 

 the tetrachromic, as the sodium flame appears to give rise 

 to a distinct sensation. 



March 2. — "Atmospheric Electricity in High Latitudes." 

 By George C. Simpson, B.Sc. Communicated by Arthur 

 Schuster, F.R.S. 



This paper is an account of a year's work on atmo- 

 spheric electricity undertaken at Karasjok, Norway, from 

 October, 1903, to October, 1904, with the results of a 

 month's observations on atmospheric radio-activity made 

 .nt Hammcrfest. 



NO. 1850, VOL. 71] 



Karasjok is situated well within the Arctic Circle 

 (69° 17' N.), and during the winter has a severe .\rctic 

 climate, so that it is well situated for finding the influence 

 of meteorological elements and the presence or absence of 

 direct sunlight on the electrical conditions of the atmo- 

 sphere. 



The observations were limited to determinations of the 

 potential gradient, electrical dissipation, atmospheric 

 ionisation, and atmospheric radio-activity. A continuous 

 record of the potential gradient was obtained by means of 

 a Benndorf self-registering electrometer, and measure- 

 ments of the dissipation and ionisation were made three 

 times each day unless the weather made it impossible to 

 use the instruments. Measurements of the radio-activity 

 were made between the hours of 10 to 12 a.m., 3 to 5 p.m., 

 and 8.30 to 10.30 p.m. on 253 days, and in addition 42 

 measurements were made between 3 and 5 a.m. The re- 

 sults of the work are shortly as follows : — 



Ye.irly Vari.'vtio.n-. — Potential Gradient. — The yearly 

 course was found to be in accordance with the general 

 rule for the northern hemisphere — rising rapidly from 

 October to February, when it reaches a maximum, then 

 falling more rapidly until the end of May, after which it 

 remains constant until the winter sets in again during 

 October. Dissipation. — The yearly course is exactly 

 opposite to that of the potential gradient, the curves re- 

 presenting the two being almost mirror images of one 

 another. Ionisation. — The course of the ionisation consists 

 of a nearly linear si.x months' fall from the beginning of 

 September to the end of February, followed by a similar 

 six months' rise from March to the end of .'\ugust. 



D.\iLV Variation. — Potential Gradient. — The daily course 

 for the whole year consists of a single period having a 

 minimum about 5 a.m. and a maximum about 9 p.m. 

 Dissipation. — For the whole year the dissipation is slightly 

 higher at midday than earlier in the morning, while the 

 evening observations show the lowest dissipation of the 

 three. Ionisation. — The daily period of the ionisation is 

 not so pronounced as that of the dissipation, but the 

 ionisation is slightly lower in the evening than in the 

 morning or at midday during the whole year. 



Relation between the Meteorological and Electrical 

 Conditions of the .Atmosphere. — Wind. — As is to be 

 expected, the dissipation increases greatly with the wind 

 strength. Temperature. — Both the ionisation and dissipa- 

 tion become much less as the temperature goes down. 

 With temperatures between 10° C. and 15° C. the dissipa- 

 tion is 495 per cent, and the ionisation 044 per cent., while 

 with temperatures below —20° C. these become 083 per 

 cent, and 017 per cent, respectively. The potential 

 gradient increases as the temperature falls. Relative 

 Humidity. — With rising relative humidity the dissipa- 

 tion falls rapidly, and the ratio of negative to positive 

 dissipation increases. When the whole year is taken 

 into account, the same result is found for the ionisation ; 

 but for the winter and summer six months, taken separ- 

 ately, the effect of the humidity of the air on the ionisa- 

 tion is not apparent. 



Interrelation of Electrical Factors. — Both the dissi- 

 pation and ionisation greatly influence the potential 

 gradient. Low values of ionisation and dissipation are 

 accompanied by high values of the potential gradient, and 

 vice versd. The dissipation increases with the ionisation. 



The .Aurora and the Electrical Condition of the 

 Atmosphere. — \o relation whatever could be detected be- 

 tween the aurora and the electrical conditions of the 

 atmosphere. The most careful watching of the electro- 

 meter needle revealed no variation of the potential gradient 

 with variations of the aurora. 



Radio-activity. — Measurements of the radio-activity were 

 made by Elster and Geitel's method, and their arbitrary 

 unit was used in expressing the results. A most distinct 

 yearly course of the radio-activity was found, the maxi- 

 mum, 129 (mean for month), falling in December, and 

 the minimum, 47, in June. The radio-activity has also 

 a very pronounced daily course, the maximum, 162 (mean 

 for year), falling in the early hours of the morning, and 

 the minimum, 58, about midday. 



There is a distinct connection between the radio-activity 

 and the meteorological conditions of the atmosphere ; the 

 radio-activity increases as the temperature falls, rises as 



