2S8 Remarks on JFalkcr's new Transit Instrument. 



affect the gnomon. It is therefore necessary to make this 

 part of the irxslrument adjustable, which may be done in tht 

 following manner : 



Let a triangular stone, resting on three foot screws, be 

 placed with one side facing the north and the opposite angle . 

 pointing to the south ; near'the middle of this side the gno- 

 mon must be fixed to stand directly over the meridian line. 

 Then, by means of these screws and a plumb-line suspended 

 fron> the top of the gnomon, the right line which passes 

 through the centres of the perforations may be readily set 

 perpendicular to the horizon. The plumb-line must be sus- 

 pended on the south side of the gnomon to pass over two 

 <lots, one made near the top of it, and the other near the 

 bottom, in a line parallel to the line of perforations ; and, 

 as the plumb-line will not be in the way wlicn observations 

 are taken, it may aUvays remain attached to the instrument. 



No light should be suffered to enter the instrument from 

 t1»e south further than the gnomon, except those rays of the 

 sun which are transmitted through one of the small aper- 

 tures made in it ; but at the north a small door must be 

 made in the roof, to be opened when the observations are 

 taken. 



The precise time that the limb of the solar image comes 

 in contact with a line drawn on a plane surface, cannot be 

 exactly observed, in consequence of the penumbra which 

 surrounds ; but I am of opinion that this inconvenience may 

 be avoided by using a different mode of observing. 



Tl'.e eve is capable of judging with considerable exactness 

 when a rie;ht line divides a circle into two equal parts : thus 

 in (\^. 1. Plate V'll, it is easy to see that the segment on the 

 riiht hand side of the line is greater than tlie other ; in fig. k. 

 it appears that the circle is divided into two equal parts, but 

 in firr. 3. the greater segment is on the left side of the right 

 line. Hence I am inclined to think, that if an observation 

 be taken when the line bisects th? solar image, it will be 

 nearer the truth than when the two limbs are observed. If 

 five lines be drawn at proper distances, five observations 

 may be taken this way in less time than six minutes. 



Bt'.t a still better method v ould be to observe with three 

 fines it '* lime, as represented in figures A, b, and 6, as th? 



