410 Additional Notes. 



plate, \vith an aperture in it equal to the unsilvered part of 

 the glass 3 llie edge of the aperture toward H to be exactly- 

 straight, dividing between the silvered and unsilvered part of 

 the speculum, and standing in the line of the axis of the tele- 

 scope. This speculum is to be set at an angle of about 20 

 degrees, with the square of the piece AB, or at 110 degrees, with 

 the sides of it. Upon the piece AB, the telescope PQ is fixed, 

 of a good aperture and held, with the axis placed as above. 

 The limb is to be graduated by diagonals, or parallel circles, to 

 half degrees and half minutes, beginning from C, which are to be 

 numbered as whole ones. And if it be practicable to face wood 

 with brass without warping, the whole face should be so covered; 

 if not, then along tlie outward edge of the limb a narrow strip of 

 brass plate ma}- be let into the face of it, finely and equally in- 

 dented on the edge, to take a screw fitted to tliat tootliing to be 

 fixed on the moving index at L, as your instruments are made 

 that count by revolutions* and then, before this is used, it woiild 

 be proper to take the distance of the two objects first nearly by a 

 fore-staft^", and from thence accordingly to set the index. This 

 screw, at land, would be highly useful, but at sea it cannot be 

 wrought, while the instrument is directed by the same person ; 

 though, as the motion of the moon and variation of the angle is 

 but slow, it may be brought to exactness by several trials in the 

 intervals of direction. The instrument, as above described, will 

 not take an angle of much above 50 degrees, which, for the pur- 

 pose intended, may be fully sufficient. But if the speculum EF 

 be made to take oif and put on, and the end of the index at K be 

 so notched as to turn that speculum from its first perpendicu* 

 larity, to make an angle of about 25 degrees, it will then take 

 any distance to 100 degrees. 



By this description it may be thought that tlie utmost ac- 

 curacy will be required in making the instrument : yet, of all that 

 ever liave been invented of so curious a kind, it will probably 

 be found to demand the least ^ for, provided the speculums are 

 good, on which the whole depends, if the first EF be set truly 

 over the centre, the limb well graduated, and the other speculum 

 be also set perpendicular, there can, I think, be no other errour 

 but what the instrument itself will easily rectify: for if it be di- 

 rected to one jitar, and that be taken, at tlie same time, botli by a 

 direct ray through the glass GH, and by a refiexion from EF, 

 both exactly coinciding at O, it is evident that then the specu- 

 hims are tx.ictly parallel. And if this falls not precisely wbcfw 



