330 
THE VOYAGE OF H.M.S. CHALLENGEE, 
spherules from the deposits. It is easy to indicate the origin of this coating if we grant 
the rapidity with which meteorites penetrate the atmosphere. This would determine a 
superficial fusion, and the formation of a coating of magnetic oxide, as in the case of these 
spherules. Their formation may, indeed, be compared to what is observed in the little 
particles of iron that fiy away from the anvil under the stroke of the hammer, and are 
transformed in piirt or entirely into magnetic oxide. The non -oxidised nucleus being 
placed under protection by the layer of magnetic iron remains in a metallic condition, 
and in this way we may account for the presence of these unoxidised metallic particles at 
the bottom of the sea.* It is the same phenomenon as takes place with iron in industrial 
processes by the coating of Barflf. The superficial fusion and oxidation of the external 
coating thus probably took place in the atmosphere at a very high temperature, and on 
account of their small dimensions the particles at once assumed a spherical form. The 
contraction of the superficial crust on cooling would lead to the formation of the cupule. 
Thus the composition of the nucleus, the formation of the black coating and the cupule, 
the form, and, in short, all the peculiarities of these spherules, lead us to regard them as 
cosmic bodies that must be grouped with the holosiderites. 
(6.) Brown-coloured Spherules or Chondres. 
If we now turn to the spherules with a crystalline structure, there are many reasons 
for l)elieving that they, too, have probably a cosmic origin. It is well known that 
clioncb’es are more or less spherical concretions, and are characteristic elements of a great 
group of meteorites — the chondrites. Tschermak considers them as drops of matter of 
cosmic origin, in fusion, that have become solidified. Chondritic globules have never, 
moreover, in spite of all the researches that have taken place, been found in eruptive 
rocks, nor, indeed, in any rocks of terrestrial origin. 
The distinguishing characters of these globules of silicates from the deep-sea deposits, 
and their relations to the chondres of meteorites, may now be referred to in detail. In 
the fii-st ])lace, they present ]>rofound analogies in external aspect with the chondres of 
meteorites, although, as will be presently pointed out, they differ from them in some of 
their cr}’stallographic details. These brown-coloured spherules are either yellowish or 
brown, with a pronounced bronze lustre. Under the microscope, in reflected light, this 
metallic lustre is seen to be due to a finely lamellated structure ; their surface, in place 
of being smooth as in the black spherules, is seen to be striated. Their diameter 
rarely attains a millimetre, and their mean diameter may be about 0'5 mm. They arc 
not regularly spherical. The cupule, when it exists, is not very deep, but rather 
* It may l>e well to recall here that Home meteoric ironn, c.q.^ the meteoric iron of Santa-Catarina (Brazil), do not 
ozidiAc under the action of water ; this is the case when the iron contains a relatively large amount of nickel (see 
Boniwingault, CompUt Itmdxu, tom. Ixxxvi. p. 513, 1878). 
I 
ii 
