Royal Society. 303 



they are insufficient in number and frequency, and, the earlier ones 

 particularly, in the required accuracy, to enable us to determine 

 whether the annual decrease has been uniform or otherwise. 



The author having taken much pains to obtain a correct deter- 

 mination of the dip in the Regent's Park in August 1821 (published 

 in the Philosophical Transactions for 1822), repeated his observa- 

 tions in August 1828, at the expiration of seven years from the 

 former determination, — an interval which he considered sufficient 

 to throw light on the rate at which the dip is at present diminishing. 

 In consequence of the increase of buildings in the Regent's Park, 

 he was induced to change the place of observation to the Horticul- 

 tural Society's garden at Chiswick : the distance apart is about five 

 miles, but the direction is as nearly as possible that of the line of 

 equal dip. 



The apparatus, modes of observing, and needles employed, are 

 fully described. The needles were four in number — one of the or- 

 dinary construction ; a second fitted with Professor Meyer's appa- 

 ratus for avoiding the errors arising from the non-coincidence of the 

 centres of gravity and motion; a third, having a cross of wires at- 

 tached to the axis, on the well-known plan of Dr. Mitchell; and a 

 fourth, devised by Mr. Dollond, the middle of which is a cube per- 

 forated at right angles, so that the axis may be inserted in eight dif- 

 ferent ways. 



In addition to his own apparatus and needles, the author obtained 

 from the Colonial Department the use of a smaller apparatus, with 

 a needle on Professor Meyer's plan, the same which was used by 

 Capt. Franklin on his last land expedition. The observations with 

 this apparatus were made by Mr. David Douglas, of the Horticul- 

 tural Society. The results were as follow : — 



With the ordinary needle 69° 46' 1 



With Meyer's needle 69 47-4' 



With the needle having an adjustable axis 69 .SS*3 



With Mr. Dollond's needle 69 51-7 



With the smaller apparatus 69 51 •4' 



Dip in London in August 1828 69 47 N. 



From the observations of 1821 and 1828, the author finds a de- 

 crease in the dip in London of 17'*5 in seven years, or an annual 

 decrease of 2'"5. 



The average annual decrease for the century preceding 1821 ap- 

 pears, from the most authentic observations, to have exceeded 3'. 

 On examining the scries of observations made on the dip in Paris 

 since 1798, by MM. Humboldt, Gay Lussac, and Arago, the author 

 finds a corresponding indication of a recent diminution in the yearly 

 decrease of the dip ; it appearing, by those observations, that the ave- 

 rage yearly decrease in the first half of the period between 1798 and 

 1 828 exceeded 4'-75, and in the second half fell sliort of 3'. He con- 

 cludes by remarking that a repetition of the observations in London, 

 at the expiration of another seven years, and a continuation of those 

 at Paris, will probably afford a decisive indication on this point; and 



notices, 



