a Se 
THE ROXBURY CONGLOMERATE SERIES. 49 
the banding, therefore, tell only part of the story. As has already been sug- 
gested in the discussion of the clays (p. 36), ideal conditions might exist in 
which an almost complete record might be preserved of the alternations. 
As to the cause of these extremely fine alternations, the difference in the 
flow of water of day and night might possibly explain them. (See Plate 16, 
fig. 4). Occasional thicker layers of coarse material such as are sometimes 
found might indicate storms of rain or unusually hot weather. In such cases 
many of the finer layers, due possibly to the day and night cause, might be eroded. 
It is interesting to note that the sections showing the most complete record 
of the finest layers have also the largest number, and these are in a more perfect 
condition of preservation than in those sections where erosion has greatly 
destroyed the regular order. This statement is made guardedly, as too few 
sections with comparatively undisturbed layers have been found to make general- 
izations. The quieter and deeper the water, therefore, up to a certain distance 
from the ice front, the thicker and more perfect should be the components 
of the banding, and the finer the sediment. Beyond the point of maximum 
deposition in deep, quiet water, the sediment should be fine but the layers 
thinner. 
A characteristic of the first 100 feet of the slate is the formation of den- 
dritic forms of limonite along thin sandy layers of the slate. These sandy 
but very thin layers usually come near the top of the coarse component in the 
banding, although they may occur in the fine components as well. The average 
size of grain of this coarse layer is slightly coarser than the average size of grain 
in the regular coarse component. I have thought that water percolating along 
these thin layers has deoxidized the iron in the material and deposited the oxide 
above and below in the finer material in dendritic forms of limonite. Den- 
dritic forms occur in the coarse layer itself, leaving most of the material without 
iron. This dendritic material gives the effect of a very dark band, and although 
fairly regular in its occurrence, is not to be confused with the dark layers com- 
posing the fine components. 
The reason for the occurrence of this fine layer of coarser material in the 
midst of the coarse component might indicate the warmest part of a season with 
resulting high water in the glacial streams. If there had been land without 
ice or snow near the ice front high winds might have blown fine sand in appreci- 
able amounts into the water. Some of the grains are too large to be classed as 
clay or silt. (David and Priestley, 1909, p. 305). As observed above, the 
layers in question usually but not always come near the top of the coarse com- 
