4i8 



NATURE 



■ [November 25, 1920 



dye is increased by the introduction of heavier 

 radicals, and they are able to draw other inferences 

 that may prove useful guides in seeking for new sensi- 

 tisers. Of these twenty derivatives, which include 

 ethyl red and pinachrome, it appears that none is so 

 generally advantageous as sensitol green. The second 

 communication treats of the "carbocyanines," and 

 describes the preparation and action of eighteen 

 different derivatives. One of the diethylcarbocyanine 

 iodides is pinacyanol or sensitol red, and it is found 

 that the sensitising power for gelatino-silver bromide 

 is far less in the dimethyl derivative, and that it sinks 

 gradually on passing to the dipropyl and dibutyl com- 

 pounds. In the ixocyanines the two quinoline residues 

 are linked by the group ICH-, while the joining 

 group in the carbocyanines becomes ICH-CHICH-. 

 The lengthening of this linking chain is accompanied 

 by an extension of the extra sensitisation far into the 

 red region of the spectrum, and the authors suggest 

 that if methods were available for still further 

 lengthening this chain it would be possible to produce 

 compounds that would sensitise still further into the 

 infra-red. Inferences are also drawn with regard to 



the effects of the positions of the substituting radicals 

 and other matters. 



The Research Defence Society has lately published 

 four pamphlets of general interest, namely : (i) Vac- 

 cination, by Dr. Mary Scharlieb ; (2) The Prevention of 

 Tetanus during the Great War by the Use of Anti- 

 tetanic Serum, by Sir David Bruce; (3) The Work 

 of the Medical Research Committee, by Sir Walter 

 Fletcher ; and (4) The Value of Experiments on 

 Animals : Notes of Personal Experience, bv Sir 

 Leonard Rogers. The set, price 2S., may be obtained 

 from the society's Secretary, 11 Chandos Street, 

 Cavendish Square, London, W. i. 



It is eight years since the first edition of Prof. 



F. Soddy's volume, "The Interpretation of Radium," 



was published, and during that momentous period the 



necessity for enlarging and revising the original version 



has been proved. Mr. Murray announces that the 



author has been at work, and, wfth due compressions 



i and the right additions, particularly those that bear 



upon the problem of the constitution of the atom, 



I has brought the volume as closely up to date as is 



j possible with a large and rapidly extending subject. 



Our Astronomical Column. 



Longitude by Wireless. — The scheme for linking 

 up the observatories of the world by utilising wireless 

 time-signals was referred to in Nature for May 20 

 last (vol. cv., p. 370). It must be understood that no 

 appreciable increase of accuracy over the older method 

 by cable signalling is claimed ; indeed, where the 

 observers are not interchanged the precision is less. 

 But the gain in convenience, expense, and wide dis- 

 tribution of signals is considerable, and it is known 

 that where the travelling-wire method of observing 

 transits is adopted, personality is greatly reduced ; 

 what remains is of the same order as the small local 

 deflections of gravity, which can be eliminated only by 

 extensive geodetical operations. 



Mr. Dodwell, the director of Adelaide Observatory, 

 has communicated the longitude which he deduces by 

 the reception of the Lyons and .Annapolis signals at 

 Adelaide and Greenwich. It is gh. 14m. 19-955. using 

 Lyons signals, and ig-ySs. using Annapolis ones. The 

 Nautical Almanac value is 20-305. Allowance has 

 been made for time of transmission, assuming a speed 

 equal to that of light. 



Many of the Australian boundaries are defined as 

 meridians east of Greenwich by a specified number of 

 degrees. Thev were determined by lunar observations 

 and are known to be in error by some miles. It is 

 not, however, expected that any change will now be 

 made in them. 



An Apparent Earth-Effect on the Distribution 

 OF Solar Facul.«. — The Monthly Notices for June 

 contains a paper on this subject by Mr. E. W. 

 Maunder, who acknowledges important help from 

 several others in discussing the material, which con- 

 sists of the Greenwich photographs from 1878 to 1916. 

 The research was undertaken to test the result 

 announced by Mrs. Maunder in 1907 that there was 

 a preponderance of spots on the eastern half of the 

 visible disc. Suggestions were made that this might 

 arise from the spots sloping backwards or from the 

 surface being heaped up behind the spot, thus avoid- 

 ing the necessity of invoking an "earth-effect." The 



NO. 2665, VOL. I06I 



faculae, however, since they are evidently above the 

 surrounding surface, could scarcely be affected in 

 either of these ways, so that an eastern preponderance 

 seems very hard to interpret otherwise than as an 

 earth-effect. Such a preponderance is, in fact, shown 

 for the greater part of the period under discussion. 

 The average excess for thirty years is about 3 per 

 cent. — a quantity of the same order as that found by 

 Mrs. Maunder for spots, and later for prominences. 

 The northern and southern hemispheres of the sun 

 are plotted, and show a general accordance, with 

 differences of detail. There is fairly clear evidence 

 that the eastern excess varies with the progress of 

 the sun-spot cycle, being least marked during the 

 increase of solar activity and most marked during its 

 decline. In other words, regarding the earth-effect as 

 a damping influence on the spot activity, then the 

 solar resistance to this damping is greatest at the 

 time of increasing activity. 



The Densities of Binary Stars. — In a paper in 

 Mem. della Soc. degli Spett. Ital. (vol. viii., 

 Ott., Nov., Die, 1919) Dr. G. .-Xbetti discusses 

 the densities of several binaries of which the 

 relative masses and parallaxes are known. The dia- 

 meters are inferred from the absolute magnitude, and 

 surface brightness is inferred from the spectrum. The 

 extreme values of density are 0002 for e Hydrae A 

 and 1-87 for e Hydrae B. On plotting the mean 

 densities as functions of spectral type, there is a 

 slow but steady decline from o-6o for A5 and 0-55 for 

 F5 to 0-45 for Go and 02 for Ko. This agrees with 

 Prof. H. N. Russell's hypothesis on the assumption 

 that the stars in question are in the giant stage, pass- 

 ing from an early diffused condition (type Ko) to 

 one more condensed through the types G, F, and A. 

 Plotting mass as a function of absolute magnitude, all 

 masses above 1-5 have about the same absolute magni- 

 tude, while all the fainter absolute magnitudes have 

 about the same mass, the mean for these being about 

 0-3. The number of stars discussed is too small to 

 lay great stress on the results. 



