Near the west side of Yorke's Peninsula, the corrected east variation was 3° less than on the east side, although the places of observation were not more than forty-eight miles asunder: the uncorrected observations differed 6°.
Upon the east side of the high hills behind Memory Cove, the east variation was 1° 40′ greater than at the granitic summit of the same hills.
In Shoal-water Bay, at anchor near the eastern shore, the corrected east variation was 1° 25′ less than near the west shore; at Broad Sound also, it was 1° less on the east than on the west side. These effects were correspondent to the former, though the expression of the situations be unavoidably different.
In the Investigator's Road, Gulph of Carpentaria, the east variation was full 1° more on the east side of Bentinck's than on the west side of Sweers' Island. The rock here is partly iron ore.
Near the east side of Pellew's Group, the east variation appeared from the bearings to be increased 2° from what it had been at a further distance, though in regular course it should have diminished; and at stations on the east sides of the different islands, I found it necessary to allow 1° degree more than on the west sides.
There were several other examples where the south end of the magnetic needle was drawn towards the nearest land; but only two where the contrary attraction seemed to have been exerted. These were both on shore, and probably might not be exceptions to the rule if all the circumstances were known; for although the body of an island may lie to the west, a single block of stone near the theodolite on the other side might do more than counteract the opposite attraction.
To arrive at the cause why both the iron in the ship and also the land should preferably attract the south end of the compass needle in the southern hemisphere, it seems necessary to refer to the direction of the dipping needle; for the unobstructed magnetism in a mass of iron will lie as nearly in that direction as the form of the mass will admit. Where the south end of the needle dips 60°, the north end will necessarily be no more than 30° from standing perpendicularly upright; and it is to be supposed, that the upper ends or parts of the different pieces of iron in the ship will possess the same attraction as the north end of the dipping