FIFTH REPORT— 1835. 
On comparing the view which he had thus formed with Dr. 
Halley’s magnetic chart of 1701, M. Hansteen found a ge¬ 
neral accordance. The principal differences were in regard 
to the situation of the Siberian pole, and to the motion of the 
axes, for both which Dr. Halley had very insufficient data. 
M. Hansteen considers therefore that Dr. Halley was the first 
person to discover the true magnetic arrangement of the globe, 
and that his deductions were fully as precise as the observations 
made in his time permitted. A century having since elapsed 
of observations with more perfect instruments and methods, 
M. Hansteen deemed that the time had arrived when their col¬ 
lection in one view and their careful examination might justify 
a far more complete and confident deduction. In 1811 the 
Royal Society of Sciences at Copenhagen proposed the following 
prize question: u In order to explain the magnetic phenomena 
of the earth, is one magnetic axis sufficient, or must we assume 
more ?” The principal part of M. Hansteen’s work was com¬ 
posed to meet this question, and received the prize : it was com¬ 
pleted and published in the German language in 1819. 
In an Appendix of 148 quarto pages are collected the various 
magnetic observations which had been made from the earliest 
times to the year 1817, and which were previously scattered 
in voyages and travels and in the works of philosophers and 
systematic writers. These are arranged in appropriate tables, 
and are the materials from which M. Hansteen has constructed 
maps of the variation corresponding to the years 1600, 1700 , 
1710, 1720, 1730, 1744, 1756, 1787, and 1800; and of the dip 
for the years 1600, 1700, and 1780. In collecting these obser¬ 
vations and in arranging them in maps M. Hansteen has ren¬ 
dered a great service to all who desire to acquaint themselves 
with the facts regarding terrestrial magnetism that observation 
has made known. 
In the first chapter, entitled “ Of the Lines of Variation, and 
of their changes between the years 1600 and 1800,” the autho¬ 
rities for the several maps of variation are discussed, particu¬ 
larly those on which the map of 1600 is founded; these are 
examined in considerable detail, and are shown to be fully de¬ 
serving of confidence, and sufficiently exact for the purpose, not¬ 
withstanding the early period at which they were made. 
In comparing the maps of 1600 and 1700 (PI. I.), the difference 
appears at first sight so great that we can hardly imagine how one 
series of lines can have passed into the other. The interval of 
a century is certainly a longer one than is desirable, and it is 
greatly to be regretted that sufficient materials do not exist for 
an intermediate map. Aided, however, by the light thrown on 
