[ 2.54 ] 



XLVfl. Description of Mr. G. Atkins's Hylmmefer for 

 dtiennining ikt Spt.cijic Gravity of both Solids and Liquids. 



To Mr. TiUocli. 



SIR, 



-1 H prc-icnt improved state of chemistry ; its application 

 to ■<(> manv oi'our pnncipil inannfacfures, and the necessity 

 ol ('-'ermining the ?pec lie gravity ot the various substances 

 wiiic ' are used in them, or affiM-ding in all cases an an- 

 portaii nd :ali()n v. ith regard to their qualities, and being 

 in uia: y the only accurate ineasure oF their value, may 

 perhaps render the following descnption of an instrument 

 for thi> purpose not unacceptable to your numerous readers. 

 Bv giving it a place in yonr valuable Magazine you will 

 therciore oblige, Sir, your obedient .-ervant, 



Gko. Atkins. 



57, Porf.et-Street, F'.ect-Street, 

 Sept. 10, I80i<. 



The specific gr2_vity, or comparative weight of the ma- 

 jority of those substances which fall under the observation 

 of the manufacturer, the mineralogist, or the chemisi, hav- 

 ing always been considered as one of their most distinguish- 

 ing characteristics, a variety of methods have at different 

 periods been resorted to for ascertaining it. 



In point of accuracv, perhaps, the best mode of taking 

 the specific cravity of a body is by a very good hydrostatic 

 balance. This instrument, however, we may venture to 

 atfirm, can scarcely ever be obtauiLi sufficiently perfect to 

 be depended on for so nice a pur;;- -e. 



Persons who are in the habit ■■' idjusting bilances, and 

 those who use them with eons'O'i- ihle care, well know the 

 various .sources of error to whic'. • ley are liable. The c r- 

 cum^tarce ot the arms of a beam t 'ing in equilibrio, is no 

 proof of its correctness, unless i. . I remain so when eaher 

 loaded or unl laded, and with exi- nge of scale pans. The 

 necessity of having a piece ot stt.: f^^u- the beam which shall 

 be perlectly homo-: iieous ; the unc' i fainty with regard to 

 the exact equdli'y of the arms, in I'Otu weight and length j 

 and, even when very nicely adjusted, its liability to accu.i-e 



{•/oLiity, 



