GEOLOGY AND PALEONTOLOGY OF CANAL ZONE


Oligocene assignment fitted the occurrence of these fossils in southeastern United States (Vaughan, 1919a, p. 199-203, 207). The Eulepidina fauna is found in the Chickasawhay limestone of Alabama, the Suwannee limestone of Florida, aced the Flint River formation of Georgia usage, and the Antiguan coral fauna in the Flint River formation. These three formations are now assigned to the upper Oligocene (Cooke, Gardner, and Woodring, 1943, chart; MacNeil, 1944, fig. 1; Cole, 1952 (1953), p. 6). A middle Oligocene age is not unreasonable, if the American deposits that are essentially the equivalent of the European Aquitanian stage are assigned to the upper Oligocene. In the present report, however, they are referred to the lower Miocene. The age range of the Eulepidina-Antiguan coral fauna cannot be assumed to be narrowly restricted. It is a reef fauna and may eventually be found to have a considerable time range, like the Lower Cretaceous Urgovian reef fauna. In the Canal Zone and Panamd the Eulepidina fauna occurs in the upper part of the Bohio formation and at one locality in the Caimito formation (locality 51), and the closest approach to an Antiguan coral fauna is in the Caimito formation (localities 52 and 57). The time span between the deposition of the upper part of the Bohio in the Pacific coastal area and deposition of the Caimito at localities 51, 52, and 57 in the Gatun Lake area doubtless is not great, but it would be rash to correlate these deposits closely.
 CAIMITO FORMATION EXCLUSIVE OF MADDEN BASIN AND PACIFIC COASTAL AREA
 The Gatuncillo and Bohio formations are widespread, but are fairly uniform lithologically despite their widespread distribution. The Caimito formation is the youngest widely distributed formation. It is, however, lithologically heterogeneous. Deposits in too many areas have perhaps been included in the Caimito, but within the type region the formation is heterogeneous. In all the areas where the formation is identified it is marine, or at least mainly marine, and contains much volcanic debris.
 The name for the Caimito formation was proposed by MacDonald (1913, p. 569). He did not properly define the name then or later, and lee specified no type region. In the absence of evidence to the contrary, it may be assumed that the type region was intended to be the region that furnished the name. Caimito, or Caimito Junction, was located on the present alinement of the Panama Railroad near Darien in the Gatun Lake area (MacDonald, 1913, pl. 68). On the basis of accepting the Darien region as the type region, the Caimito formation consists of the strata, mostly tuffaceous, overlying the Bohio formation. MacDonald included in the Caimito formation strata now assigned to the La Boca marine member of the Panama formation and his


representation of the Caicito as overlying the Enperador limestone (now assigned member rank in the Culebra formation) was based on misidentification of both Emperador and Caimito (MeacDonald, 1913, pl. 68).
 In complete sections the Caimito formation overlies the Bohio formation or volcanic rocks that are thought to include the equivalent of the Bohio. Though the actual contact has not been observed, it evidentiv represents a discontinuity. In the northeastern part of the Gatun Lake area the Caimito seems to overlap the Bohio and Gatuncillo formations, directly overlying the basement complex, and in the northern part of Madden basin the Caimito overlaps the Bohio and rests on the Gatuncillo. In the southeastern part of the Gatun Lake area the lower part of the Caimito (or perhaps the entire formation), appears to grade into the Las Cascadas agglomerate. Wherever the Caimito formation is dated it is of late Oligocene age, except in Madden basin, where it includes both upper Oligocene and lower Miocene deposits. The Caimito of that area is described under the heading "Oligocene and Miocene series.' The Caimito of the Pacific coastal area, which is entirely of Oligocene age, appears to be continuous with the Oligocene part in Madden basin and is discussed under the same heading.
 According to estimates, the thickness of the Caimito ranges from 250 to 400 meters.
 STRATIGRAPHY AND LITHOLOGY
 Gatun Lake area.-Three members of the Caicmito formation were recognized by Jones in the Gatun Lake area: lower, middle and upper (Jones, 1950, p. 900-901), which correspond, respectively, to the basal, lower, and upper members of his former usage (Woodring and Thompson, 1949, p. 232-233). According to Jones, the lower member is made up of conglomerate and tuffaceous sandstone. The conglomerate resembles conglomerate of the underlying Bohio formation, but includes pebbles of tuff. The sandstone and the matrix of the conglomerate contain acidic tuff. The lower member is correlated by Jones with the Las Cascadas agglomerate, but perhaps the entire formation is to be correlated with the Las Cascadas. The lower member of the Caimito is recognized only locally. Its absence elsewhere may indicate discontinuity or lateral gradation into deposits grouped with the middle member.
 The middle member consists chiefly of tuffaceous sandstone, some of which is calcareous, and lenticular limestone, mostly algal limestone. Tuff, tuffaceous calcareous siltstone, and conglomerate are other constitutents. Agglomerate and poorly sorted, coarsegrained, tuffaceous, nonmcarine sandstone in the northeastern part of Barro Colorado Island evidently are the equivalent of marine strata in the middle member. Foraminiferal soft limestone, such as that at locality


28