30 THE BERMUDA ISLANDS. 



north side of St. George's Causeway also failed to show much 

 evidence of coral growth, although shells and millepore frag- 

 ments were packed in endless quantities; the tests, perfect and 

 imperfect, of the foraminifer genus Orbiculina were also very 

 ahundant. I do not wish to be understood as saying that the 

 islands are not really of coral formation ; that a coral funda- 

 ment exists, needs no further demonstration than is furnished 

 by the rich growth of Diploria and Mseaiidrina within the 

 reef- waters, and by the coral fragments and masses that are 

 inclosed by the beach formation. I wish merely to emphasize 

 the important part which organisms other than corals have 

 taken toward the shaping and the making of the rocks, 

 especially of the superficial parts which have lent themselves 

 to wind-action. 



The true relations of the Bertnudian rock were first clearly 

 established by Nelson.* With remarkable sagacity this ob- 

 server read the history of the discordant layers, here horizon- 

 tal, there steeply inclined, now arched in one direction, then 

 in another, which appear in all the sections, and he was not 

 slow to point out that they were the result of wind-drift mere 

 shifting (calcareous) sands which had been thrown about 

 promiscuously by the winds, and had solidified in layers in 

 the positions where they had been finally dropped. This in- 

 terpretation stands to-day as firmly established as it stood 

 when it was first enunciated upwards of a half century ago. 

 The thin knife-sheets which are so characteristic of this drift- 

 rock build up massive beds, which are thrown together in 

 most irregular confusion dove-tailed, apparently faulted, 

 lenticulated, and otherwise. No more interesting exposures can 

 be had than the faces of the road-walls, both in the city and in 

 the country, where synclines, anticlines, slopes, and horizontals 

 appear sometimes in the space of a few yards. At other places 

 no bedding, beyond the thin lamination, is apparent, and the 

 whole mass rests concordantly either in straight or undulating 

 lines. 



*Trans. Geol. Soc. of London, 2d Ser., vol. v, Part I, pp. 103-123. 



