112 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
lize in rhombohedral crystals, yet occasionally a rock is met with in 
which the crystals are much more acute and consist probably of scale- 
nohedra. This is more particularly true when the crystals are deposited 
from solution and line and project into cavities in the rock. This 
type of crystallization is seen in Figure 6, which represents a limestone 
from the central depression of Makatea. 
The ground mass of the rock consists of “mud,” some of which has 
been partially clarified in the process of recrystallization. The large © 
cavity is seen to be lined with scalenohedral crystals of calcite, whose 
acute terminations project into the center of the cavity. One interest- 
ing and uncommon feature of these crystals is the marked zoning seen in 
many of them. This is probably due to the invasion of mud during 
crystallization, and while not uncommon in calcite from mineral veins is 
very rarely met with in crystals seen in sections of limestones. The 
above changes represent the normal structural and mineralogical altera- 
tions observed in these rocks, but there occur occasionally certain appar- 
ently anomalous features in the limestones. It has been noticed above 
that “mud” filling cavities in corals under certain conditions is con- 
verted into aragonite, but usually into calcite. 
This is certainly true, but it is often noticed that the “mud” in a 
rock resists change after most of the organisms have recrystallized to 
calcite and even after another alteration, presently to be described, 
has still further obliterated the organic structure of the rocks. Where, 
however, the “ mud” or other matrix in a rock becomes recrystallized it 
does not always happen that the organisms are also similarly altered at 
the same time. On the contrary, it has been occasionally noticed that 
the “ mud” filling the cavities of a coral has become crystalline calcite, 
while the original aragonitic stereoplasm of the coral has broken down 
into a gray structureless “ silt.” 
Many limestones, owing to the percolation of water containing carbon 
dioxide and to other causes, have passed or may pass through a series 
of changes such as has been described above. Whether or no a par- 
ticular limestone, either unaltered or recrystallized, will undergo an 
important modification of composition and structure about to be de- 
scribed depends upon certain circumstances in the history of the rock 
about which very little is known. 
This change consists in the introduction of magnesium into the lime- 
stone, and the partial replacement of calcium by magnesium carbonate. 
This replacement may go on in a rock until the amount of magnesium 
carbonate reaches as much as 15 per cent, and yet no sign of any min- 
” 
