692 A NEW TYPE OF SUNDIAL. 



It should be noted in graduating the correction-plate that 

 the plus side must be to the left and the minus side to the 

 right in the Southern Hemisphere. As regards the subdivision 

 of the dial-plate itself, there is one point to be noticed, namely, 

 that since the upper edge of the shadow is to be used (for the 

 sake of sharpness of reading), whether it is fore- 

 noon or afternoon, it is necessary to leave a space 

 at 12 o'clock to represent the thickness of the shadow, 

 i.e., both edges of the shadow must read 12 o'clock at noon. The 

 graduations for the forenoon hours must work to the left, 

 starting at the left edge of the noon shadow, whilst the after- 

 noon graduations start from its right edge, and go to the right- 

 Since the shadow is narrower than the wire which causes it, the 

 amount of space to leave at noon must be ascertained by observa- 

 tion. It is convenient to make the dial exactly 18 inches in 

 diameter, and the axial wire not less than ^^5 inch. In this case 

 the divisions on the perimeter work out at exactly i mm. per 

 minute of time, and the space at noon is about 4 mm., so that a 

 graduated steel tape can be curved round the pernneter and 

 used for finding the position of the hour marks and their sub- 

 divisions : e.g., 12 cm. to left of the left edge of the noon mark 

 gives the position of the 10 a.m. mark. 



The correct mounting of the dial in its stand presents some 

 difficulties. An approximate north is first found by slowly 

 rotating a horizontal compass until the marked end of the needle 

 lies on 340°, when the N. and S. marks (360° and 180°) on its 

 dial will be in a line with the geographical north and south. 

 This can be verified and improved on at night by finding the 

 pole-star <r Octantis with field-glasses. The axial wire of the dial 

 has to lie in this line, yet not horizontally, but so that the north 

 end is lower, the angle of slope being the latitude of the place. 

 For Johannesburg and the Reef we have the Coincidence that this 

 angle is tan ~H, so that when correctly adjusted (with a spirit-level 

 below) the vertical height of any point on the wire is half the 

 horizontal distance along the level. 



The axial wire must, of course, project from both sides of 

 the dial, partly for support and partly because in winter the sun 

 is below the level of the equatorial dial-plate, and consequently 

 it is the lower half of the axial wire which casts the shadow on 

 the dial in winter, whereas the upper half acts in summer. It 

 is desirable, however, for the sake of strength and rigidity, to 

 have the wire as short as possible, but with an 18" dial the 

 minimum length that can be used is 10", i.e., 5" on each side of 

 the dial-plate. The plate used to rivet the wire into the dial must 

 be thin, otherwise in March and September (when the sun is 

 nearly in line with the plate) it will itself cast a shadow on the 

 dial and prevent the time from being read. 



The main drawback of an equatorial dial is this very fact that 

 it is somewhat troublesome to read in winter, when the shadow 

 is on the under-side of the dial-plate. There are, however, three 



