C "7 ] 



ing glafs ; for although the furfaces of a glafs may 

 be parallel, yet there always arifes fome little con- 

 fufion from the double reflection. Neither will 

 the Moon appear too weak by two unfilvered re- 

 flexions, even when her crefcent is very fmall, 

 except me mould be hazy or clouded j and then 

 the light may be increafed by lowering the te- 

 lefcope fo as to take in part of the filvered re- 

 flection of the index-glafs, which in this cafe 

 mufi be uncovered : the fame is alfo to be under- 

 flood with refpect to the Sun, mould his light be 

 too much weakened by hazinefs or thin clouds. 

 The horizon- glaiTes mould be adjufted, or the error 

 of adjuftment found by the Sun or Moon ; the firfl 

 will be in general the belt object for the purpofe ; 

 and, as the Sun or Moon feen directly through the 

 unfilvered part of the horizon- glafs will be much 

 brighter than the image of the fame feen by two 

 unlilvered reflections, it rnuft be weakened by a dar- 

 kening glafs placed beyond the horizon-glafs, the 

 reflected image being farther weakened, if necefTary, 

 by a paler darkening glafs placed in the ufual man- 

 ner between the index-glafs and the horizon-glafs. 



If a quadrant was defigned principally for taking 

 the diftance of the Moon from the Sun and fixed ftars, 

 and was not wanted for obferving terreftrial ^angles, 

 it would be the bed way to have none of the glaiTes 

 filvered, but to leave the horizon glafTes intirely trans- 

 parent, and to put a red glafs for an index-glafs of 

 the fame matter with the darkening glaiTes, which 

 would reflect light from the fore-furface only. 



The Sun's altitude might alfo be obferved with this 

 inftrument, either by the fore or back-obfervation; 



and 



