43 
Among the elements which must be considered as having the great- 
est influence upon the determination of phosphoric acid in soils are 
iron, aluminum, titanium, and silicon. Hence, in choosing soils 
for use in the present investigation, only those showing certain 
peculiarities were selected. Partial analyses of the soils are given in 
Table XVI. 
Table XVI . — Partial analysis of soils by hydrochloric acid (specific gravity extraction 
1.115). 
Soil 
No. 
1. 
Soil 
No. 
2. 
Soil 
No. 
3. 
Soil 
No. 
4. 
Soil 
No. 
5. 
Soil 
No. 
6. 
Soil 
No. 
7. 
Soil 
No. 
8. 
Soil 
No. 
9. 
Soil 
No. 
10. 
Soil 
No. 
11. 
Soil 
No. 
12. 
P.c. 
14.84 
18.20 
1.40 
35.21 
P.c. 
13.28 
26.25 
1.40 
30.52 
P.c. 
15.72 
24.78 
1.80 
29.64 
P.c. 
11.60 
26.01 
1.20 
28.57 
P.c. 
15.16 
19.55 
2.20 
29.22 
P.c. 
19.32 
21.76 
3.00 
30.57 
P.c. 
25.26 
7.03 
5.40 
33.54 
P.c. 
25.28 
17.43 
5.20 
29.20 
P.c. 
31.00 
P.c. 
25. 20 
P.c. 
30.56 
11.94 
5.20 
33.35 
P.c. 
30.84 
Aluminum oxid 
Titanium oxid 
15.81 !l9.C3 
5.00 4. 20 
28. 65 27. 03 
10.68 
6.20 
Insoluble residue 
Since the insoluble condition of phosphates in Hawaiian soils is 
partly due to physical influences, only soils of the red-clay types were 
used in this work. Soils Nos. 1 to 6, inclusive, represent the type of 
highly colloidal soil in which the aluminum content is in excess of the 
iron. In Nos. 7 to 12, inclusive, the opposite relation exists between 
the iron and aluminum. These relationships influence to a consider- 
able extent both the physical and chemical properties of the soil, and 
in the latter type, that is, those in which the iron is in excess, the 
formation of the white precipitate is more prevalent. It should also 
be mentioned that the latter are also higher in titanium than the 
former. An analysis of this white precipitate shows it to contain 
about 25 per cent of titanium oxid, 70 per cent of iron and aluminum 
oxids, small amounts of phosphoric acid, and no silica. Hence an 
error will result in the determination whether the precipitate is filtered 
off or weighed as magnesium pyrophosphate. The only solution of 
the problem is a prevention of the deposition of the white precipitate 
or a removal of the inhibiting factors. That titanium and iron, 
primarily the former, are the elements most active toward its forma- 
tion may be seen by reference to the table of analyses and to the 
table showing the weight oi white precipitate deposited. The greatest 
error occurs in those soils containing the highest percentage of iron 
and titanium, but the total elimination of these elements in the soil 
extract without removing traces of phosphate is a practical impos- 
sibility. 
A method for removing silica and titanium from the soil extract 
recommended by the Association of Official Agricultural Chemists is 
to evaporate the extract to dryness and take up in hydrochloric acid. 
This method fails utterly when applied to local soils and is never used 
in this laboratory. Instead, all soil analyses are made directly upon 
