62 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
ing all the compounds likely to be present in the rocks, and also to obtain 
an idea of the proportions of the different constituents. It was found 
that the most important constituent was calcium carbonate, that many 
of the rocks contained a considerable amount of magnesium carbonate, 
and that calcium phosphate was always present, but never in large 
quantity. 
It was also noticed that some of the less altered rocks contained organic 
matter, that the amount of insoluble residue was as a rule almost inap- 
preciable, and that traces of ferric oxide, alumina, and silica were 
sometimes found, together with occasional slight reactions for chlorides 
and sulphates. 
These preliminary results serve to show that the calcium and mag- 
nesium carbonates constitute practically the whole of the rocks, and this 
circumstance determined the methods of analysis eventually adopted. 
The two methods used were substantially identical with those which 
have been employed by Dr. Cullis, Mr. Hart-Smith, and myself in the 
chemical analysis of the Funafuti boring. The details of the gravi- 
metric method may be briefly stated : — 
A specimen was prepared for analysis by rejecting the external part 
which might contain adventitious material. A piece of the more central 
portion of the limestone, reduced to a powder in an agate mortar and 
carefully sampled, was washed with boiling water to remove any soluble 
organic or inorganic matter which might be present, but this amount 
was found to be negligible. The powder was then dried at 100° C., 
and for a complete analysis two portions were weighed out. 
1. In the first portion the proportions of calcium and magnesium 
carbonates were estimated. About .8 gram was taken, the calcium pre- 
cipitated in the ordinary way as oxalate, dissolved and reprecipitated to 
free from magnesium, and after the careful addition of a few drops of 
sulphuric acid, heated and weighed as calcium sulphate. 
The magnesium was determined from the filtrate by precipitation as 
phosphate, and was weighed as magnesium pyrophosphate. 
2. As insoluble matter and calcium phosphate were present in very 
small quantity, 10 grams of the limestone were usually taken, dissolved 
in dilute hydrochlorie acid, and filtered through a tared filter paper. 
The filtrate contained the phosphate, while the insoluble matter, organic 
and inorganic, remaining on the filter paper, was dried and weighed. By 
burning off the organic residue, the amount of inorganic insoluble matter 
was obtained. 
The phosphate in the solution was estimated by precipitating with 
