AT THE ROYAL OBSERVATORY, CAPE OF GOOD HOPE. 469 



eters in computing the refractions, so that the refractions correspond to altitudes of the mer- 

 curial column, indicated by the standard barometer of the Royal Society." 



(7.) There is nothing in this account to indicate that the capillarity correction was applied to 

 the Cape standard by Jones. Mr. Henderson's astronomical observations are in England ; and 

 the numbers may be found therein which would settle the question. In the hope of obtaining 

 information by another channel here, I compared Jones's with the Magnetic Observatory stand- 

 ard by Newman. The comparison agrees with Capt. Wilmot's in showing that Jones's is lower 

 than the Magnetic Observatory barometer by 0.013 inch no correction for capillarity being 

 applied to either. But the Magnetic Observatory barometer is not the one referred to in the 

 report of the meteorological committee of the Royal Society. That barometer was broken in 

 its passage to the Cape. The present one was sent out in 1840 or 1841. I have no document 

 or register of its comparison at Somerset House. I believe it was compared there, and no doubt 

 the register can be got at. The diameter of the Magnetic Observatory barometer's tube is 0.508 

 inch ; consequently the capillarity correction is 0.003. Omitting this correction to either of the 

 Cape barometers, the Magnetic Observatory's barometer is higher/ than the Eoyal Society's 

 Daniel by 0.008 + 0.013 = 0.021 inch ; the capillarity correction being applied to Daniel, 

 but not to either of the others. 



(8.) 'Mr. Bailey has shown, (Philosophical Transactions, 1837, part 2, page 436 et seq.) that 

 Daniel's Royal Society's standard, between the years 1827 and 1837, had been corrected for tem- 

 perature on the hypothesis that the distance between the surfaces of the mercury in the tube 

 and cistern was measured on glass ; whereas the case is wood, and the scale is on a slip of brass 

 attached to the wood the numbers for expansion being .0000857339. These limits include 

 Henderson's comparisons. Sir John Herschel's being of brass (6), would be at all times com- 

 parable with Jones's. The latter, therefore, should be further corrected for (expansion of 

 glass the expansion of brass) x into the difference in temperature between the comparisons at 

 Somerset House and the Cape. The tendency of this correction, assuming the temperature at 

 Somerset House to be the lowest, is to shorten the scale of Daniel, which increases the reading 

 for the height of the column and diminishes the difference between it and Jones's, or to de- 

 stroy 0.008 altogether. 



(9.) If Henderson entered the original readings in his journal both at Somerset House and 

 the Cape, the examination of these numbers will clear up whether the capillarity correction was 

 applied to Jones or not. They will also furnish the elements for the correction pointed out by 

 Bailey for Daniel. 



The comparison of the Magnetic Observatory standard with (?) the crown and flint glass 

 barometers at Somerset House, (Philosophical Transactions, 1837,) will (7) furnish a second 

 determination. 



(10.) The journals at the Cape Magnetic Observatory are silent with regard to any barometric 

 index error ; none appears to have been applied in any of the returns to Woolwich. The mer- 

 curial column is, and has been, corrected for the expansion of the mercury and brass alone. 



(11.) The original barometric readings from Newman are given with the equatorial observa- 

 tions on Mars, and the original readings from Jones are given with the circle observations 

 without any correction whatever to either. Paragraph (3) furnishes their common difference. 



THOS. MACLEAR. 



