GEOLOGY AND PALEONTOLOGY OF THE CANAL ZONE. 407 



brought to Washington about 100 specimens. A list of the stations 

 at which collected would })e almost a list of the exposures of the 

 Antigua formation examined. In Cuba, at station 7508, Ocujal 

 Spring, altitude 200 feet a. t., near Guantanamo, collected by 

 O. E. Moinzor, In Porto Rico, zone C, near Lares, collected by Bela 

 Hub])ard, of the Now Academy Porto Rican Explorations. Serro 

 Colorado, Arube, Dutch West Indies. 



As a slight variant from the typical form, it is common in the base 

 of the Chattahoochee formation along Flint River, near Bainbridge, 

 Decatur County, Georgia, and it is well represented in the silex bed 

 of the Tampa formation at Tampa, Florida. 



It is also found h} Anguilla, where I collected a single specimen at 

 station 6893, on the south side of Crocus Bay. 



In the State of Tamaulipas, Mexico, at the following localities: 

 One mile east of Salitre; Cerro del Aire, 7 miles southeast of Refugio; 

 1 mile east of San Jose de las Rusias; hill 4 miles east of San Rafael 

 (specimens submitted by Mr. E. T. Dumble). 



A specimen sent to the United States National Museum by Mr. 

 Philip Crutcher is reputed to come from Vicksburg, Mississippi; 

 subsequently collected by O. B. Hopkins at station 7463 in the Byram 

 calcareous marl, 4^ miles south of Vicksburg, Mississippi. 



In general, the species is abundant in the three formations men- 

 tioned, and is important in indicating an Oligocene horizon. It has 

 not yet been found in deposits younger than those of Tampa age. 



Prof. K. Martin, director of the Geologisch-Reichs Museum, 

 Leiden, submitted to me for determination some material from 

 Serro Colorado, Arube, that I referred to Orbicella tenuis (Duncan), 

 supposing at the time that Duncan's Astraea tenuis belonged to the 

 genus Orbicella} Subsequent study of additional collections has 

 shown that Duncan's Astraea tenuis is in reality a fungid coral. The 

 following are the notes I published on the Arube specimens in the 

 paper referred to in the footnote: 



The corallites are long; are close together, only a millimeter apart, and usually 

 are not round because of having been deformed by mutual pressure; the diameter 

 of the corallites is from 4 to 5 mm. The septa are thin, and crowded; the usual 

 arrangement being four complete cycles. The members of the first and second cycles 

 reach the columella; those of the third cycle are not so long; and those of the fourth 

 are still shorter. The members of the first and second cycles are of about the same 

 thickness, no constant difference in thickness according to cycles is discernible. 

 There is no marked difference in the thickness of any of the septa at the wall. The 

 members of the third and fourth cycles are slightly thinner. Endotheca is well devel- 

 oped . The exotheca has been destroyed in the process of f ossilization . The columella 

 is poorly developed, being formed by the loose fusion of the principal septa in the 

 axial space. 



I also pointed out in the paper cited the close resemblance of 

 the specimens described to "Orbicella'^ cellulosa (Duncan). I have 



1 Geolog. Reichs Mus. Leiden Samml., ser. 2, vol. 2, p. 33, 1901. 

 37149— 19— Bull. 103 15 



