surface is fairly uniform at different depths, while in other cases it 
varies considerably. The same is true of the lateral variation. 
In some caves the material in different caverns or compartments 
is fairly uniform, while in others it is not. Generally the material 
is likely to be more uniform at different depths than in different 
parts of a cave. Analyses illustrating these points are afforded by 
samples Nos. 497 to 500, 772, 774, 775, 777 to 779, 788 to 791, 792 to 
794, 843 to 845, 848 to 850, and 856 to 859, as may be seen by re- 
ferring to a description of the samples in Table IV. 
CHEMICAL ANALYSES OF SAMPLES. 
METHODS OF ANALYSIS. 
In preparing the samples for analysis and for vegetation tests, all 
were passed through a 1 -millimeter sieve, except the fresh bat manure. 
This could be done by pulverizing without grinding. Occasional pieces 
of limestone or stony concretions not readily pulverized were dis- 
carded. In this way, a fair sample of the utilizable material was ob- 
tained, as the large lumps of nonpulverulent material have little 
fertilizing value and should be screened out before transporting the 
guano from the cave. 
The usual analytical methods of the Association of Official Agricul- 
tural Chemists were used when possible. Samples containing much 
organic matter were ignited with magnesium nitrate before deter- 
mining the total phosphoric acid. Lime was determined by the 
Glaser method as modified by Jones. 1 
The total nitrogen in most samples was determined by the Kjeldahl 
method modified for nitrates, as nearly all samples contained more 
or less nitrate. The nitrogen present as ammonium salts was deter- 
mined in a water solution of the guano by direct distillation with 
sodium hydroxid. Nitrates were then determined in the same solu- 
tion by distillation with the further addition of zinc and iron. The 
results for ammonia thus obtained may sometimes be slightly in 
excess of the true values, as the sodium hydroxid may have 
attacked organic nitrogenous compounds, but they are accurate 
enough for practical purposes. 
All percentages are calculated on a moisture-free basis. It should 
be borne in mind that the ordinary air-dried guanos contain 3 to 15 
per cent of moisture, and that the percentages calculated on an 
air-dry basis would therefore be somewhat lower. In the ordinary 
cave, the guano, before air-drying, contains a very high percentage 
of moisture, as may be seen from Table II, which gives the moisture 
content of different guanos from two caves of about average dryness. 
1 Wiley, H. W. Principles and Practice of Agricultural Analysis. Easton, Pa., 1908, vol. 2, 2. ed., 
p. 236. 
55016°— 18 2 
