590 



NA TURE 



[April 20, 1^93 



The general plan was lo observe every slar ihree times, and out 

 of Ihe total number of stars in the catalogue (3415) 289 slar> 

 were observed less than tiiis number of limes, while 1048, 491, 

 and 194 stars were observed four, five, ai d six times respectively, 

 and the rest seven times or more. The various differences of 

 brightness were estimated by Argelander's method of step- 

 estimations, each sequence comprising ten, five, or twenty stars 

 according to the number of -tars in the vicinity observed. 

 Commencing in the year 1882, Mr, Sawyer says that nearly half 

 of the whole work wa> done in that time, an opera glass being 

 extensively used fur fainter sequences, such as those in which the 

 stars were of the 6th or fainter magnitude a field glass was em- 

 ployed. During the years 1883 and 1885 the observations, as he 

 tells us, were wholly discontinued, "oa ing to the injury lo the eyes 

 from the trying nature of the work." In the method of reduction 

 the magnitudes were deduced by plotting out the sequences, 

 graphically using the Uranometria Argentina magnitudes as 

 ordinate?, and the observed differences ol brightness, expressed 

 in steps, as abscissa^. The arrangement of the catalogue itself 

 is as follows :— The columns give successively the catalogue 

 ciirrent number of the star, U. A. cataloj;ue number, constellation. 

 Right Ascensions and Declinations for mean equinox 1875.0, 

 number of observations, mean magnitude deduced, U.A. 

 magnitude, and the three last the separate dates of the obser- 

 vations and magnitudes. 



Comparing the average differences between the magniiudes 

 here assigned and those given by Gould, it is found that 

 ± 0"o88m. about represents it, w hile the average error of a single 

 determination, assuring equal degree of precision and ir.cluding 

 besides accidental errors, the effect of systematic ditTerence is 

 given as ± 0059m. 



While the work was in hand eight variables were discovered, 

 which were as follows :— U Ophiuchi (1881), U Ceti (1885), U 

 Aquilse and Y Sagiitarii (1886), R Canis Majoris (1887), Y 

 Ophiuchi and W Hydroe (1888), and (?) Leporis (1891), and in 

 addition several large discordances were noticed in many values 

 obtained (the catalogue number of these are here given), render- 

 ing these stars worthy of Sfiecial attention. The volume con- 

 cludes with notes, in which several suspicious cases of variables, 

 &c., are recorded. 



A New Table of Standard Wavelengths. — Under this 

 title Prof. H. A. Rowland contributes to Asb'jnoniy atid Astro- 

 physics for April (No. 114) the new measurements of several 

 metallic lines to be used as standards. The actual measures were 

 made by Mr. L. E. Jewell, the probable error of one setting 

 amounting to I pait of 5,coo,coo of the wave-length, and the 

 reductions of the reading by Prof. Rowland himself. The 

 measurements were obtained with a new machine, supplied with 

 a screw specially made after Prof. Rowland's process. The 

 standard wave-length of D used was the mean of the determina- 

 tions of Angstrom, Midler and Kempf, Kurlbaum, Pierce, and 

 Bell, and was 5896" 156, different weights being given to these 

 separate values. This value was utilised in six different ways, 

 and the resulting table of wavelengths from 2100 to 7700 was 

 obtained, the accuracy of which might, as he says, be estimated 

 as follows : — " Distribute less than j^^ division of Angstrom 

 properly throughout the table as a correction, and it will be per- 

 fect within the limits 2400 and 7000." 



Meteor Showers. — Among the principal meteor showers 

 for the current year, a list of which is given in the Companion 

 to the Obseivatory, the following two occur this week, the former 

 of which is described by Denning as " one of the most briliant 

 showers." The radiant points are : — 



WOLSINGHAM OBSERVATORY, CIRCULAR No. 35. —A plate 



ta'<en with the Compton 8-inch photo-telescope, April ii, com- 

 pared with a photo by Max Wolf, 1891, shows that the two 

 stars 



Es-Birm 545 i8h. 28-9m. + 36° 55' (1900) 

 ,, 561 i8h. 39"4ai. 36° 52' ,, 



are var., the photo differences being approximately 9*1, ir4 ; 

 8-8, io'2. 



NO. 1225, ^OL. 47] 



GEOGRAPHICAL NOTES. 



Letters dated March 9 have been received from the Ant 

 arclic whaling vessels'confirming and extending the brief tele- 

 graphic infoimation already published. The .ships did not 

 proceed farther south than 67" latitude, and discovered no signs 

 of the existence of the Greenland whale, although whales ol 

 several other species were common, and there were great nuuj- 

 bers of grampuses. In default of whaling, the energy of ihe 

 crews was devoted to sealing, and the four vessels secured 

 between iheiii about 16,000 skins and a full cargo of oil. The 

 seals were of several varieties, but until the return of the ships 

 their species cannot be determined, nor their commercial value 

 known. The weather throughout ths whole stay in Antarciic 

 waters was severe, and the formation of ice compelled the 

 vessels to return at an earlier date than was at first intended. 

 Flat icebergs of enormous size were seen, one being reported as 

 fifty miles in length. The facilities afforded for scientific woik 

 were disappointing. 



The Delcommune expedition (p. 474) has returned to 

 Europe, and M. Delcommune was received with great enthusiasm 

 in Brussels. The expedition, together with the others sent 

 out by the Katanga Company, has to a large extent completed 

 the work of Livingstone and his successors in the Congo Basin, 

 and in the main confirms the accepted geography of the region. 

 One point of some interest which has been established is that 

 the Lake Lanji, marked from Arab reports at the junction of 

 the Lukuga and the Lualaba, has no existence. 



The new number of Peterviann's Mitteilungen contains a 

 short paper by Pi of. Kiiimmel on recent Russian oceanographical 

 woik in the north Pacific. This is accompanied by a map of 

 the salir.ily of the surface water, which extends, and in a general 

 way confirms, Mr. Buchanan's map founded on the Challenger 

 work. The centre of maximum salinity lies between 20° and 

 30° N., and has its centre about 170° W, A tongue of con- 

 siderably fresher water stretches nearly across the ocean, about 

 10° N.and sweeps round the coasts of America and Asia. The 

 diminution of salinity northward is very interesting, the curves 

 of equal salinity s'.veeping through Bering Sea without regard 

 to the line of Aleutian Islands, thus showing that so far as 

 regards surface water, Bering Sea is simply part of the Pacific 

 ocean, standing in very marked contrast to the Sea of Okhotsk, 

 a fact of some interest during the present international con- 

 troversy, 



Mr. T. H. Hatto.n-Richards read a paper on British 

 New Guinea at the last meeting of the Royal Colonial Institute. 

 While giving an account of the climate discouraging lo would- 

 be white settlers, Mr. Richards describes the native Papuans 

 from personal experience as a fine race, possessing a keen 

 sense of justice, and most laborious and successful as agricul- 

 tuiists. I 



RECENT INNOVATIONS IN VECTOR 

 THEORY} 



C\^ late years there has arisen a clique of vecior analysts who 

 ^^ refuse to admit the quaternion to the glorious company of 

 vectors. Their high priest is Prof. Willaid Gibbs. His reasons 

 for developing a vector analysis devoid of the quaternion are 

 given with tolerable fullness in Nature, vol. xliii. p. 511. His 

 own vector analysis is presented in a pami hiet, " Elements of 

 Vector Analysis, arranged for the Use of Students in Physics, 

 not Published" (1881-84). Mr. Oliver Meaviside, in a series 

 of papers published recently in the Electrician and in an 

 elaborate memoir in the Philosophical Iransactions, supports 

 some of Gibbs's contentions and cannot say hard enough things 

 about the quaternion as a quantity which no physicist wants. 

 Prof. Macfarlane, of Texas University, has added to the 

 literature of the subject, and without altogether agreung with 

 Gibbs takes umbiage at a most fundamental principle of 

 quaternions and developes a pseudoquaternionic s)sttm of 

 vector algebra which is non-associative in i;s products ! 



Between the years 1846-52, just at the time when Hamilton 

 was developing the quaternion calculus, a series of papers was 

 published by the Rev. M. O'Brien, Professoi in Kings College, 

 London. The system developed by O'Brien is essentially that 



1 Abstract of a p.iper by Prof. C. G. Knott, read before the Royal Society 

 of Edinburgh, on Monday, December 19, ijgj. 



