The table on page 48 includes two of the most characteristic species of the Gatun formation: Turritella altilira and T. gatunensis. It also includes species that have living relatives in the Pacific Ocean but not in the Caribbean Sea, and species that survived in the Pacific but not in the Caribbean Sea (Trochita trochiformis, Neverita reclusiana, and N. helicoides).
 Echinoids.-Clypeaster gatuni, Encope annecteis, E. platytata, E. megatrema, and Schizaster panamensis w ere described by Jackson (1917).
 Ostracodes.-Strata in the lower part of the Gatun formation at Cativa, for which the name Cativa marl was casually used, yielded 18 species and varieties of ostracodes (Coryell and Fields, 1937).
 Age.-Age assignment of the Gatun formation was off to a good start when Douvilld (1891, p. 497-498) wrote that strata at Monkey Hill, Mindi, and Gatun are Miocene, and a few years later ventured the opinion that the strata at Monkey Hill are perhaps of Helvetian age, whereas those on the upper Chagres [Alhajuela sandstone member of Caimito formation] seem to be of Burdigalian age (Douvills, 1898, p. 592). Both age assignments are practically the same as those of the present time. Between publication of Douville's two reports, Dall beclouded the issue by maintaining that the strata at Gaten and Minidi are Eocene, and those at Monkey Hill Oligocene (Hill, 1898, p. 176, 180-181, 271-272, 273-274). As already outlined (Woodring and Thompson, 1949, p. 231), Dall was a victim of unfortunate circumstances so far as the Eocene part is concerned. One of Hill's collections was labelled Vacmos Vamos, though there is no doubt it was collected from the Gatun formation near Gatun (locality 158), apparently from strata near the base of the middle part; that is, it was labelled as though it represents the late Eocene or early Oligocene marine memberof the Bohio(?)formation. (See notationstcnder localities 40 and 158 on p. 115, 127.) Nevertheless Dall certainly would have been suspicious had lee arranged the collections according to Hill's field numbers, instead of arranging the real and alleged collections from Vamos Vamos biologically in one lot. The same mistake was made much later when an ill-advised early Miocene age was proposed for Hill's mixed fossils (Woodring, 1928, p. 76). All except a few of the alleged fossils from Vamos Vamcos have Hill's field number 17 on the specimens, in the vials, or on the labels written by Dall. Two lots bear the field number 18 (a Vamos Vamos collection, locality 40a), but 17 is written on the labels. One lot of Turritella altilira praecellens (USNM 135160) has 18 on the label and nothing but the catalog number in the vial. Regardless of numbering, the fossils from the Gatun formation may readily be sorted out, not onlyibecause the rock


cmatrix on microscopic examination is seen to be characteristic, but also because none of the species occurs at Vamos Vamos.
 Dall's assignment of part of the Gatun formation to the Oligocene was the result of his conviction, first published in 1896 (Guppy and Dall, 1896, p. 303-304), that the Mioceee of the Caribbean region (and all except the very latest of the Miocene in southeastern Uniited States) really is upper Oligocene. Toula justly protested against an early Tertiary age for the Gatun formation, but went too far in the opposite direction in claiming that the Gatun (that is, the middle part of the formation) is of late Miocene or even Pliocecne age (Toula, 1909, p. 737). For many years the Gatun formation has been considered middle Mioceee. A discussion of the age would be premature until the numerous mollusks are identified. Preliminary exacination suggests that in the Canal Zone the formation represents the entire span of middle Miocene timeessentially the equivalent of the Cercado and Gerabo formations of the Dominican Republic-and that the upper part in the western area, west of the Canal Zone, is late Mfiocene.
 Vf,10CENE SE RIFS
 CHANGES SANDSTONE, INCLUDING TORO LIMESTONE MEMBER
 The youngest Tertiary marine formation, the Chagres sandstone, overlies and partly overlaps the Gatun formation. The outcrop area lies entirely west of the canal, extending from the Canal Zone soctlwestward along the Caribbean coast to a locality between Rio Indio and Rio Migecel, about 45 kilometers southwest of Col'n (fig. 3). Meluch of the outcrop area, except in the Canal Zone and along the coast, still remains to be examined. Calcareous strata at the base of the formation throughout most of the outcrop area in the Canal Zone constitute the Toro limestone member.
 The name Chagres sandstone was proposed by MacDonald for the sandstone forming the hills that overlook the coast from Toro Point to the mouth of Rio Chagres (MacDonald, 1919, p. 532). The saccdstone is so massive that estimates of thickness are uncertain. MacDonald's estimate of 1,000 feet (300 meters) or more may be excessive.
 Tihe Toro limestone member also was named by MacDonald, who designated it a separate formation (MacDonald, 1915, p. 26). Toro Point was specified as the type locality. Earlier MacDoniald (1913, p. 570) used the informal name Caribbean limestone for this unit. The average thickness of the Toro is about 40 meters (Thompsoe, 1947a, p. 21).
 STRATIGRAPHY AND LITHOLOGY
 The Toro limestone member is a local basal calcareous deposit of variable thickness. It consists princi-


GEOLOGY


47