PALEOZOIC CORAL REEFS 249 



An examination of the matrix in which the Paleozoic reef-forming 

 corals are embedded gives information from another source bearing on 

 the depth at which they grew. In the accounts of the Silurian reefs of 

 Gotland, it is stated that "on the flanks of the reefs are found conglom- 

 erates and breccias of coral masses." The matrix of the Silurian reef 

 exposed in Anschiitz' quarry, Cedarburg, Wisconsin, has the structure 

 of a sandstone. The spaces between the coral heads of the Devonian 

 reef in the vicinity of Alpena, Michigan, "are filled with coral sand, 

 which generally consists of rather coarse fragments with a predominance 

 of crinoid joints. The sand filling the cavities of the reef is generally 

 much coarser than that forming the normal sediments on its flanks." 

 The Devonian reefs of the Traverse Bay region comprise "conglomeratic 

 beds, while within the thicker beds themselves the phenomena of contem- 

 poraneous erosion, of the wedging out of strata, and, occasionally, of 

 cross-bedding, and ripple-marks are met with. Indeed, all the phenom- 

 ena seen in heavy bedded sandstones are found in these fragmental de- 

 posits." In the Onondaga reef of Williamsville, New York, "the corals 

 of the bedded limestone in the neighborhood of the reef are fragmental 

 and indicate considerable wave activity around the margins of the reef." 

 From these data the conclusion is forced that the Silurian reefs of Got- 

 land and Cedarburg, Wisconsin; the Devonian reefs of Alpena, Michi- 

 gan, and Traverse Bay, Wisconsin, and the Onondaga reef of Williams- 

 ville, New York, were formed in water so shallow that they were within 

 the influence of surface waves. An examination of the types of organ- 

 isms composing these reveals that massive corals, such as Stromatopora, 

 Favosites, Acervularia, etcetera, are important. Professor Schuchert 

 informs me that he has seen in the Devonian reefs at Louisville, Ken- 

 tucky, heads of Cyathopliyllum "probably not less than 8 feet across," 

 and near Alpena, Michigan, Stromatopora heads "that were certainly not 

 less than 12 feet in diameter." As the reef -building organisms of Paleo- 

 zoic time consist of Stromatopora and its allies, of Favosites, of masses 

 of Alcyonarians, as Halysites, and massive Kugosa, the opinion seems 

 justified that all Paleozoic reefs were formed in very shallow water, as it 

 is probable that closely related organisms of the same facies lived under 

 similar conditions. 



The application of both the criteria derived from a study of Recent 

 coral reefs and from a study of the sediments in which the Paleozoic 

 coral reefs are inclosed leads to the same conclusion, which is, that 

 Paleozoic coral reefs were formed in shallow water, often or usually at a 

 depth not greater than that of the possibility of wave action. It seems 

 that 25 fathoms may be considered a safe maximum of the depth for 

 their formation. 



