Prof. Nordenskiold — Expedition to Greenland. 517 



considerable amount of oxygen, in consequence of whicli the mercury- 

 rises in the tube. 



In alcohol, the iron does not crumble. In water, it rusts, but does 

 not appear to fall to pieces. 



In air dried by sulphuric acid the crumbling process takes place 

 slowly. 



Varnishing does not fully protect these pieces of iron from weather- 

 ing, not even if immersed in warm copal-varnish. 1 thought at first 

 that the cracking was the result of the contracting and shrinking of 

 the mass, but this is not the case. On the contrary, the cracking 

 is caused by dilatation. With what force this operates may be 

 judged from the fact, that a piece of iron, on which chisel and saw 

 are without effect, is broken or bent by the decomposition of the mass. 

 In general, cracks first appear at right angles to the surface of the 

 stone; these diverge as from a centre, and at a depth of a few lines 

 below the surface of the stone meet a crack that runs parallel with 

 the surface, which, by the swelling of the overlying crust, is soon 

 formed into a little dome, sometimes an inch in height. In the 

 mean time the overlying crust is raised, doubled up and broken in a 

 manner which bears a striking likeness to the doubling of the 

 stratified rocks by the so-called eruptive forces, — ^that is, if one sup- 

 poses that the cracks, instead of being empty, are filled with detritus, 

 which gradually hardens to an "eruptive" rock. 



"When fragments of the largest stone weighing 134 gr. were 

 heated to redness, they parted with nearly two litres of gas, or about 

 100 times the volume of the iron, as also a considerable amount of 

 water, which, like the gas, had a bituminous smell. The gas was 

 clearly no primary constituent, but formed partly by the decomposition 

 of organic matter in the meteorite^ partly by the reducing operation 

 of compounds containing carbon on the oxide of iron in the meteorite, 

 which was found to be completely reduced at the termination 

 of the experiment. On the iron being dissolved in chloride of 

 mercury, only a trifling quantity of gas was emitted, probably 

 coming from the pores in the iron. In hydrochloric and nitric acid 

 the meteoric iron is dissolved, leaving in some cases a residue 

 containing much carbon, in others very little residue at all. The 

 gas that escapes during solution in hydrochloric acid has a most 

 penetrating smell, probably due to some hydrocarbon. On dissolving 

 Ovifak iron, which has been heated to redness, in air or oxygen, in 

 acid, there often remains a flocky, voluminous, brown material soluble 

 in warm, but hardly so in cold water, which in ammonia is very 

 easily dissolved, forming a dark brown, almost opaque fluid. The 

 same material is obtained from the carbon that remains after the 

 solution of the iron in acids. It can again be precipitated by means 

 of acids from the ammoniacal solution, though not quite completely, 

 so that the acid solution is also brown, but of a very light tint. 

 This material is a humus-like compound, which probably did not 

 originally exist in the meteorite, but arises from the solution of 

 the carboniferous iron in acids.^ This humus-like body can be broken 



1 A similar substance, obtained by dissolving iron containing- carbon, has been 



