EEPOKT ON THE DEEP-SEA DEPOSITS. 
421 
found. This assumption^ however, is at once improbable and incapable of proof. It is at least 
equally probable that the cobalt, nickel, and thallium are all present as peroxides. 
Fluorine. — An attempt was made to determine the fluorine in about 6 grammes of the nodules. 
The calcium fluoride ultimately obtained weighed less than a milligramme (0’0006 per cent.), so that 
fluorine, although undoubtedly present, is so in quantity too small to be accurately determined by 
any of the recognised methods, at least without undertaking a special research. 
Estimation of Ammonia.-— Only one estimation was made. About 6 grammes of the nodules 
were distilled with strong caustic soda solution, the distillate collected in hydrochloric acid, and the 
ammonia determined gravimetrically by precipitation with platinum chloride. From the weight 
of metallic platinum obtained the percentage of (NH4)20 was calculated, and found to be 0'016 per 
cent, of the manganese nodules taken. 
Estimation of Carbonic Acid. — After addition of excess of ferrous sulphate and of silver 
sulphate (in order to prevent liberation of chlorine), weighed portions of the powdered nodules were 
boiled with dilute sulphuric acid in a flask connected with a Liebig’s condenser, chloride of calcium 
tubes, and Anally potash bulbs. During the operation a slow current of pure air was passed through 
the apparatus. 
Two estimations gave identical results, viz., 0'29 per cent. CO2. 
Estimation of Sulphuric Acid. — 2'0347 grammes of the nodules were fused with sodium 
carbonate, the fuse thoroughly extracted with water, and the sulphuric acid determined in the usual 
manner. 
0’83 per cent. SO3 was obtained. 
The sulphuric acid determined in an aliquot portion of the aqueous extract obtained by boiling 
46‘732 grammes of the nodules amounted only to 0‘36 per cent., so that, unlike the chlorine, the 
sulphuric acid is chiefly present in the nodules in insoluble combinations. 
The final results of the analysis are given in Tables I., II., and III. 
Column I., Table I., gives the percentage composition of the powdered nodules without making 
any deduction for hygroscopic moisture. 
Columns II. and III., Table I., give the percentages belonging to the residue insoluble in 
hydrochloric acid and to the aqueous extract. 
Table II. gives the percentage composition after deducting the water, the residue insoluble in 
hydrochloric acid, and the aqueous extract. 
Table III. gives the percentage composition of the residue insoluble in hydrochloric acid. 
In conclusion, I desire to acknowledge most gratefully the kind and valuable assistance which I 
have received from friends and students in the course of my analysis. The spectroscopic examination 
and the earlier qualitative analyses were carried out in conjunction with Mr. F. M. Gibson, B.Sc. 
The final quantitative analysis, down to the subdivision into aliquot portions of the filtrate from the 
sulphuretted hydrogen precipitate, was carried out in conjunction with Mr. J. S. Ford, to whom I 
am specially indebted for his skilful and painstaking assistance. My thanks are also due to Mr. A. 
King, Dr. T. R. Marshall, and Dr. J. Shields, for their kind assistance, more particularly in carrying 
out a number of control determinations. 
The investigation, by the kind permission of Professor Crum Brown, was carried out from first 
to last in the Chemical Laboratory of the University of Edinburgh. 
