December io, 1891] 



NA TURE 



m 



traces of iodine. The phosphide of boron thus obtained is a 

 brown powder, very light in texture, and insoluble in every 

 solvent which has yet been tried. In contact with oxygen the 

 compound ignites at a temperature about 200°, and burns with a 

 very brilliant flame, forming boric and phosphoric anhydrides. 

 With chlorine gas it inflames at the ordinary temperature, pro- 

 ducing boron trichloride and phosphorus pentachloride. Vapour 

 of sulphur converts it into sulphides of boron and phosphorus. 

 When thrown into a little fused nitre instant incandescence and 

 deflagration occur. Its behaviour with nitric acid is charac- 

 teristic ; it immediately becomes incandescent, and moves.rapidly 

 to and fro over the surface of the acid, all the while burning 

 with a most dazzling flame. It reduces concentrated sulphuric 

 acid to sulphur dioxide. Fused potash decomposes it with 

 evolution of phosphoretted hydrogen and formation of potassium 

 borate. Sodium or potassium, in an atmosphere of hydrogen, 

 react upon warming with great energy, the mass becoming white- 

 hot. Magnesium, heated with the phosphide to 500", also reacts 

 with incandescence. Even silver and copper react violently 

 upon heating with phosphide of boron. Vapour of water 

 decomposes it at 400°, with production of boric acid and phos- 

 phoretted hydrogen. Heated to 300° in ammonia gas it takes 

 fire, and burns with formation of nitride of boron and deposition 

 of phosphorus. 



The second compound of boron and phosphorus, PsP's, was 

 obtained by M. Moissan by heating the compound PB just 

 •described in a current of hydrogen to a temperature near 1000°. 

 Under these circumstances a portion of the phosphorus is 

 eliminated, and condenses in drops in the colder part of the 

 tube, leaving the P3B5 in the form of a light brown powder, 

 which is distinguished from the normal phosphide BP by its 

 indifference to chlorine and nitric acid. It is much more stable 

 than the normal phosphide, but is, like the latter compound, 

 decomposed with incandescence by fused nitre. 



The additions to the Zoological Society's Gardens during the 

 past week include a Formosan Fruit Bat {Pteropus formosus) 

 from Formosa, presented by Mr. Thomas Perkins, F.Z.S. ; a 

 Patagonian Cavy {Lohchotis patachonica) from Patagonia, pre- 

 sented by Mr. H. H. Sharland, F.Z.S. ; a Blotched Genet 

 {Genetta tignna) from South Africa, presented by Mr. Edmund 

 R. Boyle ; a Grey Ichneumon {Herpestes griseus) from India, 

 presented by Mr. G. F. Hawker ; a Little Grebe ( lachybaptes 

 fluviatilis), British, presented by Mr, T. E. Gunn ; a Tuatera 

 Lizard (Sphenodon punctatus) from New Zealand, presented by 

 Mr. W. King ; a Brush-tailed Kangaroo {Petrogale penicillata) 

 from New South Wales, purchased ; three Carpet Snakes 

 {Aforelia variegata) from New South Wales, received in 

 exchange. 



OUR ASTRONOMICAL COLUMN. 



The Secular Acceleration of the Moon and the 

 1 ENGTH OF THE SiDEREAL Day.— Laplace showed that the 

 secular diminution of the eccentricity of the earth's orbit ought 

 to produce in the longitude of the moon a term proportional to 

 the square of the time, and which he determined as -+- lo"/-, 

 where t is expressed in centuries. Adams and Delaunay have 

 reduced this term to + tl'-\\t-. From a discussion of eclipses Airy 

 concluded that the coefficient of acceleration is as much as 12" 

 or 13" ; and accepting this, the question arises as to the cause, 

 other than that indicated by Laplace, which will account for the 

 difference of 6Y-. This forms the subject of a paper by M. 

 Tisserand in Comptes rendus, No. 20, 189 1. Prof. Darwin 

 found that the tidal action between the earth and the moon was 

 sufficient to furnish an apparent acceleration equivalent to the 

 required complement. The accompanying decrease in the earth's 

 rotational velocity produces an apparent acceleration of 3""8/* in 

 the case of Mercury, an amount which may make the longitude 

 of the planet vary by as much as 15" in a couple of hundred 



years. Since the observed transits of Mercury extend over more 

 than two centuries, M. Tisserand has discussed them with the 

 idea of determining whether the term 3" •8/'- is really indicated by 

 them. He finds, however, that the extreme transits are not so 

 well represented with the new term as without it, although the 

 difference is not very great. Thi-; result, therefore, is unfavour- 

 able to the idea as to the variability of the sidereal day, or at 

 least to a variation sufficient to reconcile the result of Airy's 

 research with the calculations of Adams and Delaunay. This 

 being so, it is concluded that the increase in the length of the 

 day, produced by tidal action, has nearly the same value as the 

 diminution which results from the contraction of the earth caused 

 by secular cooling, and that, on account of the compensating 

 action of the two effects, the length of the sidereal day remains 

 very nearly invariable. 



State of Solar Activity.— Prof. Tacchini gives, in 

 Comptes rendus for November 30, a rfj-«wt'ofsolnr observations 

 made at the Royal Observatory of the Roman College during 

 July, August, and September of this year. The number of days 

 of observation were 31 in July, 31 in August, and 19 in 

 September, and the results obtained are as follows : — 



Relative frequency Relaiive magnitude 



July ... 



August ... 

 September 



of spots. 



18-65 

 884 

 17-52 



of days 

 without 

 spotp. 

 000 

 006 



o-oo 



of spots. of facula 



76-20 

 49 06 

 114-45 



82-03 

 70-81 



6no 



A comparison of these numbers with those determined in the 

 preceding quarter shows that solar activity has sensibly increased, 

 for the spotted surface has twice the area. It will be seen that 

 the minimum magnitude of faculse occurred at the time of a 

 maximum of spots. The following are the results obtained for 

 prominences : — 



Number of Prominen ces. __ 



'891. days of Mean Mean Mean ' 



observation, number. heigiit. extension. 



July 30 ... 8-37 ... 40-2 ... VA, 



August 30 ... 6-77 ... 41-0 ... 1-9 



September 23 ... 9-26 ... 41-4 ... 2-2 



The number of prominences recorded is greater than during the 

 preceding three months. The highest prominence (142") was 

 observed in August. 



Observations of y. Cephei. — Mr. J. E. Gore made some 

 observations of the variable star /* Cephei, the " garnet star " of 

 Sir William Herschel, between January 1888 and December 

 1890, which show that the variation of light is very irregular, 

 and that the star sometimes remains for several months with 

 little or no perceptible change of magnitude ( Proc. Royal Irish 

 Acad., January 26, 1891). 



Astronotnische Nachricliten, No. 3067, contains an account of 

 the investigation, carried out by Herr Dr. Walter Wislicenus 

 on the " Influence of Ring and Disk Blinds in Micrometic 

 Measurements," in order to account for the following pheno- 

 menon. If one lifts a transit off its pillars and places it so that 

 it does not interfere with the line of sight of the collimators, 

 and then brings the central wires of each collimator exactly in 

 coincidence, it is found that, by putting the meridional circle 

 back again, and placing it in its vertical position with the 

 apertures in the central cube open, coincidence of the wires no 

 longer exists, but a slight displacement is noticed. It may be 

 remembered that this question was raised at Greenwich as early 

 as the year 1868, while in the two following years, from 

 observations made in that interval, a correction of o"'48 and 

 o"-58 was found for the difference of reading. In 1874 this 

 discrepancy was accepted as real, and corrections for it were made, 

 but no real origin for it was assigned. Mr. Turner, in the year 

 1886, also investigated this difference of reading, employing the 

 collimators of the transit circle at Greenwich, and the numerical 

 results obtained were given in vol. xlvi. p. 329, of the Monthly 

 Notices. By using a vooden model of the central cube of the 

 transit, he got essentially the same results as those given 

 by the cube in the ordinary manner, but both were in 

 discordance wiih the readings taken when nothing was inter- 

 posed. To account for the difference he says : " The discrepancy 

 is due to a real difl^erence between the lines of collimation of 

 the central and eccentric portions of the object-glasses of the 

 collimators." 



NO. 



154, VOL. 45] 



