NOTES 



ASTRONOMY. 



By A. C. D. Crommelin. B.A.. D.Sc, F.R.A.S. 



WIRELESS TELEGRAPHY FOR TIME SIGNALS.— 

 L'Astrniioiiiic for August gives au interesting account of the 

 manner in which the Time Signals of the Paris Observatory 

 are distributed by wireless telegraphv. 



The standard clocl;s are kept in the " Catacombs," ninetv feet 

 under ground, where there is a practically constant temperature 

 of 11°-S C. the variations in several years being under 0"-02. 

 The sidereal and solar clocks can be compared at a distance, 

 their beats being rendered audible by microphones. 



\\'ireless telegraphy was first used for distributing time 

 signals on 23rd May, 1910. It is probably only a question of 

 time before these or similar signals are available to ships in 

 any part of the world, which will completely solve the problem 

 of longitude at sea, with an accuracy formerly undreamt of. 

 This and the gyro-compass are two re\-olutionary impro\-e- 

 ments in navigational methods during the last fifteen months. 



The mean solar clock is put right, as at Greenwich, by an 

 electro-magnet that can either aid or oppose the action of ■ 

 gravity on the pendulum. Several warning signals are sent by 

 hand in a pre-arranged manner to give notice of the actual 

 signal, which is sent automatically by the clock, and goes 

 through a relay to the Eift'el Tower, whence it proceeds by 

 wireless telegraphy. A note in the Observatory recently 

 stated that a clockmaker at Canterbury regulates his clocks bv 

 the Paris signals, which were recently changed to accord with 

 Greenwich Time. Germany has also a system of signals 

 despatched from Norddeich, and Rio de Janeiro is about to 

 follow suit. The French propose to use the method for the 

 accurate determination of the difference in longitude between 

 Paris. Bizerta (Tunis). Athens and possibly Lake Tchad. 



MR. BARTRUM'S THIRD OUERV,— In my sohition 

 given last month, the full value of the constant k" should 

 have been given, as this varies with e and so alters the result. 

 Professor Adams showed that to get the numerical \alue of 

 the acceleration correctly it was necessary to introduce the 

 variation of e into the differential equations, and not merely 

 put in its rate of variation at the end. He in this way brought 

 down the numerical \-alue to about half what had been found 

 before. The publication of his result e.xcited a lively discus- 

 sion among mathematicians, and it was some time before 

 Adams' result won uni\'ersal acceptance. 



STELLAR MOTION AND SPECTRAL TYPE. — The 

 study of the systematic motions of the stars has engrossed 

 much attention in recent years. Professor Kapteyn taking the 

 lead with his announcement of the two general drifts which 

 could be detected when the motions were analysed. This 

 announcement was confirmed and e.xtended by the work of 

 Mr. Eddington. Professor Schw.arzschild, and others. Last 

 year Mr. Hough and Dr. Halm examined the radial motions 

 of the stars, derived from spectroscopic observations, and 

 found that these too showed anomalies such as the two- 

 stream theory would lead us to expect. Dr. Halm has 

 recently (Month. Xot. R.A.S.. 1911. June) returned to the 

 subject and arranged the evidence in a manner which makes 

 the conclusions more evident. After eliminating the solar 

 motion he finds preponderance of motion in the direction of 

 the vertices of the two drifts. The average motion taken 

 without regard to sign is twenty-six kilometres per second in 

 these regions, and goes down to fifteen kilometres per second 

 at intermediate points. Dr. Halm's paper also contains 

 further confirmation of the result already announced by 

 Kapteyn and others that a star's velocity increases as its 

 spectral type grows more advanced. Thus the Orion type, 

 which is supposed to consist of very large and heavy stars, in 



a \-ery early stage of de\elopment. has a mean motion of six 

 kilometres per second in the line of sight ; the Sirian type has 

 eleven kilometres, while for later types the mean motion goes 

 up to eighteen kilometres. The advantage of the spectroscopic 

 method is that it is independent of distance ; it is particularly 

 valuable in the case of the C~)rion stars, which are so remote 

 that their transverse velocity is difficult to ascertain. 



Mr. Eddington. in a paper read before the British Association, 

 reprinted in the Observatory for October last, refers to the 

 above result as "one of the most startling in modern 

 astronomy." He proceeds: — "For the last forty years astro- 

 physicists have been studying the spectra and forming their 

 systems by which they arrange the stars in order of evolution. 

 . . . If this result is right we have a totally different 

 criterion by which the stars are arranged in the same order. 

 If it is really true that the mean motion of a class of stars 

 measures its progress along the path of evolution, we have a 

 new and most powerful aid to the understanding of the steps 

 of stellar de\elopnient," 



BROOKS' COMl'T, — This was a most conspicuous object 

 in the morning sky at the beginning of November, The tail 

 w-as plainly visible for 20° or more. Rev. T. E. R. Phillips 

 noted that its type appeared to have altered from the earlier 

 one of a bunch of straight rays diverging from the head, to a 

 parabolic envelope enclosing the head. There are good 

 reasons for supposing that this latter type arises for comets 

 when their distance from the Sun is small. 



BORRELLY'S PERIODIC COMET will be visible in 

 small telescopes during December. An ephemeris is given in 

 "The Face of the Sky for January." 



MARS has been a very brilliant object in Taurus, far out- 

 shining Aldebaran. the lucida of that constellation. From the 

 reports of MM. .-Vntoniadi and Jarry-Desloges it would appear 

 that veiling by mist or cloud of portions of the dusky regions 

 has taken place to an imusual degree at this opposition. 



BOTANY. 



By Professor F. Cavers, D.Sc. F,L.S, 



STRUCTURE OF FLOWER IN CRUCIFERAE,— 

 Arising from the preceding note, there may be considered some 

 points in the morphology of the flower in the Cruciferae. The 

 peculiar structure of the Cruciferous flower has given rise to a 

 good deal of discussion, and the question is still open. As is 

 well known, the flower shows isobilateral symmetry. The 

 calyx is in two whorls, each consisting of two sepals ; the 

 corolla in one whorl, alternating with the calyx as a whole, 

 with the four petals placed diagonally in a cross. The stamens 

 are also regarded as being in two whorls, an outer whorl of 

 two short lateral stamens, and an inner of four stamens in 

 two pairs placed back and front. The pistil is apparently 

 composed of two carpels placed transversely (laterally) ; the 

 ovary is divided into two chambers by a partition joining the 

 two placentas, the latter being placed back and front — this 

 partition arises as an outgrowth of the placentas, the two out- 

 growths meeting in the middle of the ovary cavity ; the two 

 stigmas are placed above the placentas — a somewhat unusual 

 position (the stigmas usually alternate with the placentas) but 

 found also in the Poppy family (Papaveraceae). On the bases 

 of the stamens are nectaries, the honey being secreted into 

 the bases of the sepals. 



From comparison witli the Papaveraceae, which have two 

 sepals, and from the fact that in development the four inner 

 stamens appear to arise in pairs by branching of an outgrowth 

 .it first simple in structure, it has been supposed that the 



47S 



