Ai'uil 5, 1906J 



NA TURE 



545 



log A Bright- 

 0-3763 ... 0-20 



o 3943 ■•■ ° -I 7 

 0-4114 ... o 15 



04275 ... 013 



04428 ... on 



An observation at Strassburg on March 17 gave correc- 

 tions of —us. and — 3'-g to this ephemeris. The comet 

 was; pale, with no certain nucleus, and the ill-defined 

 nebulosity was about 2' in diameter; total magnitude 

 about 11-5. 



This faint object is now apparently traversing the con- 

 stellation Taurus towards the Pleiades, and will be some 

 3 south of that asterism on April 23. It sets, a little to 

 the N. of W., at about -9 p.m. 



A Systematic Study of Faint Stars. — Apropos of Prof. 

 Kapteyn's plan (or studying faint stars, Prof. Pickering 

 gives a detailed account of how similar work is being- 

 performed at Harvard College Observatory in Circular 

 No. 108. 



It is impossible to describe the whole work here, but 

 both photometric and photographic methods are being 

 employed, and by using the two 24-inch reflectors Prof. 

 Pickering hopes to extend the survey to stars fainter than 

 Phoebe. 



In studying the spectra, the 8-inch Draper and Bache 

 telescopes have been employed, and stars down to the 

 eleventh magnitude have been observed successfully. 

 Using the 24-inch reflectors, Prof. Pickering hopes to 

 photograph the spectra of much fainter stars. 



By following the stars precisely, so that the resulting 

 spectrum is merely a line, thirteenth magnitude stars have 

 been dealt with, and, on a photograph obtained with the 

 S-inch Draper telescope, the spectrum of a star of mag. 13-3 

 is sufficiently clear to be classified ; the same plate shows 

 the spectra of no stars within 1° of the North Pole. 



Stars having Peculiar Spectra. — In Circular No. in 

 of the Harvard College Observatory, Prof. Pickering gives 

 a list and details of twenty-four stars which, from a study 

 of the Henrv Draper memorial photographs, Mrs. Fleming 

 has found to have " peculiar " spectra. 



One or two of the objects call for special remark. The 

 star D.M.-r-2i° 1609 is identical with N.G.C. 2392, which 

 was found to have a continuous spectrum, with three 

 bright lines, by Wenlock and Peirce on January 7, 1S69. 

 D Arrest, also, found it to be gaseous. Photographs taken 

 at Harvard on November 21, 1900, and November 27, 

 1005, show nn traces of the bright lines characteristic of 

 gaseous nebula:', but that its spectrum is of the 

 fourth type. This star is in Gemini, its approximate 

 right ascension and declination (for 1900) being 7b. 23-3111. 

 and +21 7' respectively. 



In the spectrum of the star D.M.-f36° 3907 the hydrogen 

 line H$ appears as a fine bright line centrally superposed 

 mi a dark line, on photographs taken on July 4 and 

 V vember 4, 1005. 



A spectrogram of the variable star R Cygni, obtained 

 on November 19, 1890, showed the hydrogen lines II7 and 

 H5 bright, but a photograph taken on December 7, 1904, 

 with the same instrument, shows a spectrum of the fourth 

 type containing no trace of bright hydrogen lines. 



The Lunar Eclitse of February S. — The total eclipse 

 of the moon which took place on February 8 this year was 

 fully observed at the Goodsell Observatory, Northfield, 

 Minn. (U.S.A.), and an account of the observations is 

 given in No. ;, vol. xiv., of Popular Astronomy. 



Dr. Wilson especially remarks on the brightness of the 

 eclipsed moon, and on the remarkable contrasts of colour 

 seen on the darkened surface. 



Five photographs were secured, and the last one, taken 

 with a small camera attached to the telescope, exposure 

 5m., shows the details of the lunar surface and its unequal 

 colouring very plainly. 



Similar observations were made by M. Quenisset at the 

 Nanterre Observatorv, and four of his photographs are 

 reproduced in the March number of the Bulletin de la 

 Socidt4 astronomique de France. He records the shadow 

 as " very transparent." 



NO. ItjOI, VOL. 73] 



PARALLEL RUNNING OF ALTERNATE 

 CI 'RRENT GENERATORS. 

 THE Bulletin des Seances de la SociiU francaise de 

 Physique lor the second quarter of last year contains 

 an interesting article by M. Boucherot on the general prin- 

 ciples which enter into the design and construction of 

 alternating current generators. 



After a brief description of the present methods of con- 

 struction, the author passes on to consider, first, the 

 wave form; secondly, the regulation; thirdly, parallel 

 running ; and, finally, methods of compounding. 



The most important part of the article is that devoted 

 to the question of parallel running, and, although the 

 mathematical treatment is not verv clearly explained, the 

 conclusions arrived at are interesting. 



The question is considered under two headings. In each 

 the subject of inquiry is the influence of the fly-wheel, or 

 the fly-wheel effect of the rotating parts, on the oscillations 

 about a state of uniform angular velocity ; but under the 

 first heading the oscillations .are caused by the variations 

 of turning effort of the prime mover during a single revolu- 

 tion, and under the second heading the oscillations are 

 produced by the action of the engine governor. 



In dealing with the first of these there are two quanti- 

 ties which are of prime importance in the calculations viz. 

 the energy stored in the rotating parts, W( = 3jn : ), and 

 the elastic couple or restoring force, C, . which is defined 

 as the couple which tends to restore the rotating parts to 

 phase coincidence with the network to which the armature 

 is connected, when the deviation is one radian. Then it 

 is shown that the natural period of the system is 

 2tv'J/C i <, and the restoring force C, = C*pk, where C» is 

 the full load torque, p the number of pairs of poles, and k 

 the ratio of the short-circuit current to the normal current. 



Next, the analysis of the turning moment is given for 

 single-cylinder and for multiple-cylinder engines as 

 follows : — 



Single cylinder 



Order of ha 



Multiple cylinder. 



the mean constant turning moment being reckoned as 

 unity. 



Each harmonic produces its own oscillation in the rotating 

 system, the amplitude of which is proportional to the value 



of the harmonic multiplied by — — (not YV YV — YY„ as 



r - W - W„ 



stated in the original), where n is the order of the harmonic 

 and \\ '„ = C,. , 2» . If the total fly-wheel effect happens to 

 be such that YY' = \Y„, resonance will occur, and this ex- 

 pression will enable the designer to proportion the rotating 

 parts so as to avoid serious trouble. 



Mi ire interesting is that part of the article devoted to the 

 effect of the engine governor on parallel running and hunt- 

 ing. The subject is confessedly a difficult one, and at 

 present there is no accepted theory. The suggestion pul 

 forward by the author is somewhat compressed and difficult 

 to follow ; and, even so, only the chief points of the argu- 

 ment can be mentioned here. 



The generating set with its governor is treated as being 

 composed of two interdependent oscillating systems, each 

 with its own natural period and its own coefficient of 

 damping. In addition to these four quantities, two others 

 are of great importance, viz. K, the percentage variation 

 of speed between no load and full load, and '{,,, the time 

 lag el the governor. The latter quantity is defined as the 

 time which elapses between the governor reaching its 

 extreme position and the turning moment of the engine 

 taking up its corresponding value. This time lag is greater 

 in compound and triple expansion than in simple engines, 

 due to the passage of the steam through the cylinders. 



Considering, first, the case of a generating set connected 

 to an external network assumed to be of infinite capacity, it 



