40 



KNOWLEDGE. 



jANl-AKV, 1911. 



The moveable circle i.s adjusted in accordance with the 

 little table. The number given for the 1st of the month is 

 brought opposite the pointer, which represents the top of the 

 imaginary celestial dial at the Pole. The circle is then moved 

 on. counter clockwise, through the fraction of the ne.xt two- 

 hour interval w'hich corresponds to the fraction of the month 

 elapsed. 



In direct comparison the dial is held up with its back 

 towards the Pole, with the pointer at the top. The line joining 

 Polaris and 13 Ursae Minoris is compared with it ; and an 

 imaginary line is drawn across the centre of the dial parallel 

 to that line. This imaginary line shows the time according to 

 the hours marked on the dial.'' 



If it is desired to study the method so as to be able to work 

 without the dial, the student may keep the dial by him. 

 adjusting it every few days. He should study the dial, the 

 adjustment of which changes slowly, and remember it. so as 

 to be able to work without it at night i using it, if necessary, 

 for verification. The changes run through a year, sometimes 

 the one star and sometimes the other being uppermost. After 

 a year's practice the changes should be sufficiently engrax ed 

 on the memory to admit of finding the time roughly without 

 the direct use of the dial. But the scheme is a little 

 complicated ; and it will be con\enient to have the dial at 

 hand to refresh the memory. 



The principle is not difficult. .\11 star phenomena occur 

 about four minutes earlier for every day that passes. This 

 arises from there being one more sidereal day in the year than 

 the nmnber of solar days. The resulting change of twenty- 

 four hours is distributed o\er the year in a maimer strictly 

 proportional to mean time. Thus, if the months were of 

 exactly equal length, the change in each month would be 

 exactly two hours. As it is, a small irregularity arises from 

 the varying lengths of the months, which may be neglected 

 for our purpose. 



It is desirable to take stars always visible. We take the 

 Pole-star and /i Ursae Minoris. The latter is the nearer to 

 the Pole of the two conspicuous stars ji and y Ursae Minoris, 

 which lie near together at distances of about sixteen degrees 

 and eighteen degrees from the Pole' respectively. 



The times indicated by the line joining any two stars near 

 the Pole can be determined by a single accurate observation. 

 This observation should be made when the line is either 

 \ertical or horizontal, or preferably, obserxation should be 

 made in both positions. In other positions the observation is 

 much less certain. In the present case the observations were 

 made a day or two before and after 1st December. 1910. when 

 the vertical position occurred at ten o'clock, and the horizontal 

 at four. This determines the table of adjustments. 



In estimating angles it is sometimes of use to note that the 

 stars -i and 7 Ursae Minoris subtend just half an hour at the 

 Pole ; and we may take it that the angle at Polaris is sensibly 

 the same. 



In one of Hardy's novels (I think " Far from the Madding 

 Crowd") the shepherd is said to be able to tell the time from 

 the stars. I have often speculated whether this is possible for 

 an uneducated man. It appears to be possible, though 

 improbable, given the outdoor habit at night, a clear head. 

 and either a good memor}', or a habit of making notes. 



.Any observant person much out at night must notice that 

 the star phenomena change their times, being earlier every 

 night. Also he might notice that they recur after a year. 

 The deduction of the two hours change per month seems to 

 present the greatest difficulty from this point of view, though 

 it is quite simple when explained. Then, one good observation 

 only is required to determine the time by any line joining two 

 stars near the Pole. But it would want a clear head to keep 

 the changing scheme of the hours always in view. 



The same moveable disc would do for any pair of stars 

 near the Pole, the setting table only being determined by 

 observation. 



This method is not susceptible of considerable accuracy. 

 It is a recognised principle in naked eye astronomy that the 

 roughest measurement is better than the best eye estini.ate. 

 Here there also appears to come into play, for positions 

 intermediate between the vertical and horizontal, one of those 

 illusions which depend on the structure of the eye. I myself 

 see the bisector according to time between the horizontal and 

 xertical positions decidedly nearer the vertical than the 

 horizontal. This is in the sky, not in the dial, which subtends 

 a comparativeh- small angle at the eye. The resulting error 

 may be either fast or slow, according to the position. I 

 believe it depends on the fact that, in a figure subtending a 

 considerable angle at the eye, the eye estimates vertical 

 distances on a larger scale than horizontal ones. This may 

 arise from the retina being slightly flatter vertically than 

 horizontally. If one is able to estimate the amount of this 

 error for one's ow-n eye. one may allow for it roughly. The 

 vertical and horizontal positions admit of fairly accurate 

 observation. ^ R H \[ H 



ECLIPSES. — The two following questions have not been 

 numbered, as they are here answered. 



Does a total eclipse of the sun occur, on an average, ten 

 times in one hundred years — of course, astronomically ? 



A total eclipse of the sun may occur on the earth more than 

 once in a year. From the year 1851 forty total eclipses have 

 occurred, but may have not been observed (astronomically) 

 owing to inaccessibility (near the polesi, to being only visible 

 at sea. or to cloud interference. Total eclipses of the sun 

 occur in cycles, the path of total phase gradually moving 

 southward or northward over the earth's surface and taking 

 over one hundred years to return to a similar point ; and a 

 cycle of about fifty-four years, which may be divided into 

 three periods of eighteen years, the area of total phase 

 moving east to west ; so that every fifty-four years the 

 eclipse is at the same longitude, though gradually moving 

 to greater or less latitude, according to the other cycle 

 movement. 



Was a total eclipse of the sun visible at Greenwich in the 

 year 1900? 



No total phase was visible at Greenwich in 1900: no total 

 eclipse had been visible in Great Britain for about one 

 hundred-and-fifty years : there will be a momentary one about 

 1927, and again in 1999. and possibly for a second or so in 

 one or two other vears this century. 



NOTICES. 



X-K.AVS. — The Sanitas Electrical Company, of 61 New- 

 Cavendish Street, W., have sent us a booklet containing a 

 large number of testimonials from workers in X-rays and 

 electro-therapeutics, as well as physicists, upon the results 

 obtained with their apparatus, including " Sanax " X-ray 

 Outfits, Intensified Induction Coils, and the Motor Mercury 

 Interrupter ; as well as their " Multostat" Universal Apparatus. 

 In addition to illustrations of the apparatus, reproductions of 

 two remarkably fine radiographs are gi\'en. 



EXPOSURE RECORD.— "The Wellcome Photographic 

 Exposure Record and Diary for 1911 " has also reached us. 

 It contains a great many hints for the benefit of photographers, 

 in addition to a neatly arranged series of tables in which 

 the owner may record all the details with regard to his 

 exposures. There is, besides a diary, an exposure calculator 

 and Several interesting illustrations, including the reproduction 

 of a colour photograph developed with tabloid photographic 

 chemicals. The price is one shilling. 



Note. — The motion round the Pole looking N. appears counter clockwise. Looking S. the motions seen appear clockwise. 



