+4 
ent knowledge of the mineral nutrition of plants such an assumption 
is, of course, only speculative. | 
The facts that neither a merely alkaline soil nor a soil containing 
much assimilable lime induces chlorosis, but that a soil which is at the 
same time alkaline and contains much easily available lime (as a cal- 
careous soil) does induce chlorosis, lend credence to the above view. 
In a soil alkaline with sodium carbonate there would be a depression 
in the availability of the iron, but not an increased absorption of lime. 
Plants grown on a soil containing gypsum absorb an increased amount 
of lime, but there is no depression of the availability of the iron on 
such soils. Plants grown on a soil containing calcium carbonate, 
however, absorb an unusual amount of lime, and because of the depres- 
sion of the availability of the iron absorb but a small amount of iron. 
That chlorotic plants contain less nitrogen and less peroxidase than 
green plants is probably because the nutrition has been disturbed by 
the increase of lime and lack of iron. The lower content of nitrogen 
and of the oxidizing enzyms are, then, not primary causes of the 
chlorosis, but rather results of the degeneration produced by the lack 
of iron. This view is strongly confirmed in the preceding work by 
the fact that treatment of chlorotic plants with iron increased the 
nitrogen and peroxidase content. 
Although but little attention has been paid to the iron requirements 
of plants, Molisch! has shown that iron exists in all plant organs, 
mostly in organic combination, and that seeds contain iron stored up 
in the globoid bodies of the aleurone grains. While iron is not a con- 
stituent of chlorophyll, it seems to be necessary for the formation of 
chlorophyll, since plants grown in iron-free solutions become chlorotic. 
It seems, then, that pineapples growing on calcareous soils absorb 
an excessive amount of lime and an insufficient amount of iron; that 
as a result there is an inability to form chlorophyll and degeneration 
of the plant follows, as is shown by a decrease in the content of 
peroxidase, nitrogen, and occasionally potash. 
SUMMARY. 
Pot experiments and a chemical survey of the pineapple soils of 
Porto Rico show that the failure of pimeapples, with the appearance 
of chlorosis, on certain areas is due to an excessive amount of car- 
bonate of lime in the soil. 
For ordinary sandy soils about 2 per cent of calcium carbonate 
renders them unsuitable for pineapples; smaller amounts than this 
do not appear to be injurious. 
Soils composed principally of organic matter may contain about 
40 per cent of calcium carbonate and still produce vigorous plants. 

1H. Molisch. Die Pflanze in ihren Beziehungen zum Eisen. Jena, 1892. 
[Bull. 11] 



