CHAP. III. ] TENERIFFE TO SOMBRERO. 225 
the chart as red clay or gray ooze are in all cases entirely, or 
even chiefly, produced by the more or less complete decompo- 
sition of the shells of surface animals with calcareous shells, 
and the mixture of their ash in varying proportions with the 
tests of Radiolarians. That they are so toa great extent, I fully 
believe; but they always contain a certain amount of mineral 
matter: very usually, perhaps universally, small particles of a 
substance which has nearly the composition of “wad,” and 
which often occurs, as I have already mentioned, in the form 
of large concretionary nodules and cakes; and very frequently 
crystals and groups of crystals of sulphate of lime. Everywhere 
near land—for example, at Station 25, near St. Thomas, where 
the depth was 8875 fathoms—the deposit is colored grayish 
with foreign matter; and everywhere in volcanic regions, and 
notably over nearly the whole area of the Pacific, the red and 
gray clays owe a considerable portion of their material to the 
disintegration of pumice, which appears to be drifted about and 
distributed by currents until it becomes water-logged, when it 
falls to the bottom and undergoes slow decomposition. 
The serial temperature soundings of this section (Appendix 
A), taken with extreme care, bear internal evidence of accura- 
cy in the way in which, in almost all cases, the curves or lines 
representing a number of observations indicate permanence or 
change in perfect harmony. Thus on Plate V. the isotherm of 
11° C., although the combined result of twelve separate obser- 
vations, is almost straight, while at every station in the section 
the aggregate of the isothermal lines either spread out consist- 
ently, or gather together, as if by a common impulse. 
The thermometers have been read with very great care, and 
the corrections for pressure required for each individual ther- 
mometer have been in all eases calculated ; but it is difficult to 
read a thermometer of the ordinary construction to tenths of a 
Fahrenheit degree, and it is possible that when reduced to a 
diagrammatic form, as in Plate V. and Figs. 57 and 58, slight 
