EARLIER REFERENCES TO AND PRESENT ASPECT OF THE DIKE. 351 



rock. Globular masses appear piled on each other like stones in a wall, and the 

 interstices are filled with the above-mentioned reddish sand. The external surface 

 of the greenstone frequently presents a rusty brown color, which arises from tlie 

 decomposition of the imbedded sulphuret of iron. The oxide of iron is found in 

 various states and of various colors in the same si)ecimen. Near the surface it is 

 yellow and 2:>ulverulent ; interiorly it is more compact and its color is reddish 

 brown and often of a bright vermillion. When by exposure to the atmos{)here 

 the sulphuret of iron is not only decomposed })ut removed, the surface of the 

 rock becomes cellular, and thus much resembles some varieties of lava. 8ucli 

 specimens are characterized by their difficult frangibility, toughness and by the 

 dark green color and crystalline structure of the feldspar, which are very evident 

 in the comj)act center of these masses." 



The general description as above given is sufficient!}'' detailed and 

 charactistic for our present purposes, though the cellular rock, due to 

 oxidation and removal of the pyrites, has not come under my observa- 

 tion. The color of the sand resulting from the decomposition is, further, 

 by no means due wholly, if at all, to the sulphuret of iron, but rather to 

 the decomposition of ferruginous silicates, such as mica and augite. 



Present Aspect of the Dhce. 



The general aspect of the dike, as recently exposed a few rods north of 

 the intersection of Mystic and Main streets in Medford village, is shown 

 in the accompanying plate (16) from a photograph, for which I am in- 

 debted to Professor W. 0. Crosb}^ as I am also for the samples of fresh 

 and disintegrated material utilized in the analyses and the general facts 

 regarding its occurrence. The still sound, boulder-like masses, some in 

 place and others rolled to the bottom of the cutting made in excavating, 

 are plainly evident, and the statements made by the Messrs Dana 78 

 years ago, though descriptive of another locality, are almost equally aj)- 

 plicable here. In the upper left portion of the view the disintegrated 

 rock is seen overlain by glacial till, a feature to which I shall refer again 

 later. 



Mechanical Analysis of disintegrated Rock. 



In order to first ascertain the physical changes which had taken 2)lace 

 in the breaking down, the resultant sand was submitted to a process of 

 sifting and washing, yielding the results given below, and for which I 

 am indebted to Professor Milton Whitney.* 



Of these separations, those represented by numbers 1 and 2 are plainly, 

 even to the unaided eye, of a compound nature, and easily recognizable 

 al^diabasic derivatives, though somewhat discolored by iron oxides. 



* In charge of the Division of Agricultural Soils, Department of Agricultui'e. 



