326 The American Geologist. May, i897 
ject to such changes saturate a bed of clay containing dissem- 
inated particles of calcium carbonate, there will be periods of 
solution and deposition of calcium carbonate corresponding 
to the changes in the character of the solvent waters. 
It is, however, difficult to see why this should cause the seg- 
i-egation into concretions of the previously scattered particles, 
for it is to be expected that deposition of material would take 
place upon the separate particles in the same ratio as solution, 
so that their relative sizes would be preserved even though 
each increas'e or diminish in weight. At first thought it may 
seem that some particles may be more favorably situated than 
others, but here also those most favorably situated to receive 
material during periods of deposition will be most favorably 
situated to lose material during periods of solution. 
If alternate solution and deposition by the solvent waters 
were alone sufficient to cause segregation and the formation 
of concretions, then the calcium carbonate in all calcareous 
clays and shales should be in the form of concretions. There 
is, however, an abundance of occurren(;es of calcareous clay 
and shale where the calcite yet remains disseminated. There 
are other conditions necessary for the formation of concre- 
tions. 
With the exception of the limonite balls which require sep- 
arate consideration, all concretions of common occurrence are 
crystalline.* 
Consequently c/*^/^/ ''''^^"6* f'^^ to be regarded as erystollfne 
afff/regates whose grov'tli has been modified by the large a- 
mount of foreign matter (ctai/) present, and the conditions of 
their formation are the same as those ichich wotdd in more 
favorable positions lead to the formation of an aggregate of 
calcite crystals of the more usual form. 
Therefore the modern theories of saturated solutions should 
be applied to the problem in the way that the chemists have 
already applied it to the growth of crystals. There are two 
conditions which in the light of these theories appear favor- 
able to the formation of concretions. These are: 
(1) The presence of ((rragonite with the disseminated calcite 
of the clay beds\ (2) the presence of the unstable humas acids. 
*Flint, although amorphous in ai>pearance. is in reality partially 
cryptocrystalline. 
