SKEATS: CHEMICAL COMPOSITION OF LIMESTONES. 1 Ps 
In 1898 Mr. Stanley Gardiner’ attributed the dolomitization of coral 
islands to the greater insolubility, in water containing carbon dioxide, of 
the double carbonate of calcium and magnesium over that of either cal- — 
cium or Magnesium alone. He believes this causes a concentration of the 
magnesium carbonate in certain limestones, and that this concentration 
may be aided at times by the evaporation of the spray from sea-water. 
It is clear, in view of the distribution of dolomite in coral islands 
that the early views of Von Buch, Haidinger, Von Morlot, and Marignac 
are quite untenable if applied to the dolomitization of such structures. 
A study of the results of analyses from different localities shows that 
there is no necessary relation between dolomitization and the presence 
of volcanic rocks. Although Christmas Island, Mango, and other islands 
which are partly volcanic also contain dolomitic rocks, yet, in most cases, 
the volcanic rocks are immediately associated, not with dolomite but 
with a non-magnesian limestone. Again, no dolomitization has been 
discovered in the volcanic island of Guam. On the other hand, islands 
such as Vatu Vara and N¢gillangillah, in which no volcanic rocks 
have been found, are dolomitized from top to bottom. The views of 
T. Sterry Hunt are open to several objections. The formation of gyp- 
sum, associated with dolomite, appears to militate against his first view, 
since gypsum has only been recorded from two or three coral islands, 
and never, so far as I am aware, in association with dolomite. The 
gypsum from raised atolls is, however, derived from sea-water by simple 
concentration, while that from the double decomposition of magnesium 
sulphate and calcium carbonate might be kept in solution in sea-water 
by means of its greater solubility (1 in 400) than that of dolomite. His 
second explanation involves the mixing of river-waters, containing sodium 
bicarbonate in solution, with sea-water, a condition which is not realized 
in the neighborhood of coral islands. 
Hardman’s views were advanced to explain the local dolomitization of 
the Carboniferous limestone of Ireland and could not apply to islands 
such as Vatu Vara and Ngillangillah where apparently the whole islands 
are dolomitized. At the end of his paper he remarks that it is probable 
that deposits like the magnesian limestone may be due to evaporation 
and assimilation of magnesium carbonate by animals in the lagoon and 
subsequent alteration to dolomite ; and that only in this way could large 
masses of dolomite be formed. Stanley Gardiner’s views appear to be 
very similar to Hardman’s, and can only have a very local application. 
Most limestones contain not more than two or three per cent of mag- 
1 Proc. Camb, Phil. Soc., 1898, Vol. IX., Part VIII, pp. 417-503. 
