SOLDADO, Formation

TERTIARY (Paleocene)

Soldado Rock, Trinidad, B . W. I.

Author of name: C. J. Maury, 1925.

Original reference: C. J. Maury, 1925c, p. 43.

Original description: ibid.

The section on Soldado rock really belongs with Trinidadian stratigraphy, but requires mention in the Venezuelan Lexicon, since formation names derived from there have been used frequently in Venezuela. Dr. H. G. Kugler, who worked out the complex structure of this tiny islet (1938), has described it as a huge block of Paleocene mixed with Wilcoxian rubble, in upper Eocene silt. It is the type locality for many species of macro and micro-fossils. "The Soldado formation" has been used with different meanings by different authors, due to imperfect understanding of the section and/or of the previous literature. In brief: the Soldado formation of Maury (1925) for part of the Paleocene; the homonymous Soldado formation of Liddle (1928), for the upper Eocene (Jacksonian); the "Soldado Rock" formation of Schuchert, 1935, for the entire section of the island; the Soldado formation of Kugler (1938), intended to be a more precise definition of Maury's Soldado, but which included a bed which Kugler now (1953) believes to be Wilcox, lower Eocene, in age and would exclude from the Soldado. This presumedly lower Eocene bed ("Bed 3" of Kugler, 1938) is the type locality of most of the "Soldado" foraminifera described by Cushman and Renz (1942), as well as several important larger foraminifera described by Vaughan and Cole (1941). The undersigned believes that it would be more convenient to retain this Bed 3 (which is only two meters thick) in the Soldado formation, while recognizing that it may constitute a diferent faunizone from the underlying beds.

An excellent historical summary of earlier publications was given by Kugler (1938). In the interest of brevity, therefore, we will suppress refence to various earlier publications bearing only on paleontology or correlation.

The first publication about Soldado is that of Maury (1912), who described fossils collected by A. C. Veatch. Veatch had distinguished 8 different beds, of which Nos. 2, 6 and 8 (in ascending order), proved to be fossiliferous. Maury (1912, p. 28-29) listed a large fauna from Bed 2 (her future Soldado formation). She considered this fauna of exceptional interest, as showing a mixture of forms characteristic of the North American Midwayan (then and for many years after, called "Lower Eocene"), and of forms hitherto known only from the Maria Farinha beds of Pernambuco, Brazil, which iiad been disputed between the Cretaceous and Tertiary. Bed 6 yielded practically nothing but foraminifera, which R. M. Bagg referred to the Eocene. Bed 8 (Maury's future Boca de Serpiente) also yielded a large molluscan fauna, which she then correlated with the "Lignitic" (Wilcox), but after 1925 recognized as upper Eocene. (See Boca de Serpiente formation.) Only very brief descriptions of the Ethology of the fossil beds were given by Maury, with no data on thickness.

In 1923, Kugler published the first geologic study of the islet. He distinguished three "formations", designated by letters: a lower limestone formation, called "A", which included Maury's Bed 2; a middle sandstonemarl formation, "B", with four members; and an upper limestone formation, "C". Kugler was uncertain whether "A" was "Lower Eocene" (as Maury thought) or middle Eocene (Lutetian) (as correlated by Douvillé, 1917). Formation "B" was assigned to the upper Eocene; B-1 as Auversian, B-4 (which equalled Maury's Bed 6) as Priabonian. The upper limestone formation "C" (which included Maury's Bed 8) was therefore thought to be Oligocene.

Maury published three papers in 1925 in which she mentioned the Soldado rock beds; in one (1925c, p. 43) she proposes the name Soldado formation for her Bed 2. In another (1925b, p. 412) she mentions that on Isla Margarita and on Toas island (misspelled Taos), she has found the same "Venericardia planicosta, smooth variety" (Venericardia (Venericor) toasensis Dusenbury, 1946) as in Bed 2, for which reason she judges the beds synchronous. (Same mention in 1925a, p. 159.)

Liddle (1928, p. 225), unaware that Maury had preoccupied the name Soldado, proposed the same name for the upper Eocene with type locality oii the island, extending the same name to upper Eocene beds in eastern Zulia and Falcón (Venezuela), and correlating with it the Río Caus and San Pedro formations (see). Liddle reproduced textually most of the description of the geology by Kugler (1923), but wrought havoc with Maury's paleontological data. He lumped together her collections from Bed 2 and Bed 8 (Liddle, 1928, p. 223-224), calling the whole a "tipical Lower and Middle Focene fauna"; this horizon he discussed under the general heading of "Aragua formation". He confused Maury's Bed 6 (a foraminiferal-algal bed) with her Bed 8 (Liddle, 1928, p. 232); whereas her extensive collections from her authentic Bed 8 were dismissed with the remark that "they could not have been collected" from that horizon (ibid.) and lumped, as noted, with the Paleocene fossils of Bed 2. (In this, obviously, he was influenced by Maury's unfortunate reference of the Bed 8 fauna to the Wilcox in 1912, unaware that in 1925 she had already corrected this assignment.)

Liddle's definition of the "Soldado formation" is not very well formulated; he begins:

"The Soldado formation receives its name from Isla Soldado... Though these deposits have been discussed at various times, they have generally been referred to as Eocene limestones." (Liddle, 1928, p. 225.)

But the fact that he heads the discussion "Upper Eocene" makes it clear that the lower beds were excluded, and from the quotations from Kugler (1923) which follow, it is clear that what Liddle considered "upper Eocene" was Kugler's "Formation B". It therefore did not include the important macrofauna of Maury's Bed 8. Kugler's upper limestone formation "C" was referred to the Oligocene.

Maury in 1929 called attention to Liddle's misuse of the name Soldado, and reasserted her priority in the name as applied to her Bed 2. She pointed out Liddle's paleontological confusions, and proposed the name "Boca de Serpiente" for her Bed 8, which she referred to the uppermost Eocene (Ludian). In 1935 she again had to defend her Soldado, against Shimer (1934) who had followed Liddle's misuse. (Fide Kugler, 1928.) Schuchert (1935, p. 701) used "Soldado Rock formation" to designate the entire sequence on the island:

"Eocene. — Soldado Rock formation (Maury, 1912, 1929). These strata, named for Soldado Rock in the Gulf of Paria, and about 1,000 feet thick, form the key sequence for the Eocene of the Antillean region. They have eight beds, but only Nos. 2, 6 and 8 have fossils."

However, in the more detailed discussion which follows, Schuchert notes that Maury regards Bed 2 "as the typical expression of the Soldado Rock formation" (loc. cit., p. 702) and that Bed 8 "is known as Boca de Serpiente". Schuchert does not use Soldado for any part of the Venezuelan succession (see his p. 685-687).

Kugler (1936, p. 1143) in a discussion of Trinidad geology, correctly used Soldado according to Maury's definition:

"The stratigraphical position of the Paleocene Soldado formation, with its glauconitic shell beds underlying the silty layers, above which is the Discocyclina limestone" (not defined, but in 1923 he had mentioned "Orthophragwina" in B-1) "is probably between the Tarouba formations and some Middle Eocene marls".

In his table of correlation, Kugler (1936, p. 1451) showed the Soldado represented as remnants in the Central Range, Naparima area and southwest Trinidad; also (as Marac limestone) in the Southern Range. (The correlation of the Marac quarry beds with the Paleocene Soldado had already been made by Harris in Waring, 1926, p. 99-100.)

In 1937, Mackenzie erroneously used the term Soldado referring to upper Eocene (following Liddle, 1928), in describing a formation from Barinas as the "Río Caus or Soldado". (Mackenzie, 1937, p. 280 and correlation table.)

In 1938, Kugler published a definite study of the geology of the Soldado Rock, with geologic map, sections, detailed measurements and description, and paleontological data. In this work, he distinguishes 12 different beds, of which his No 2 includes Maury's Bed 2; No 10 includes Maury's Bed 6; and No 11 includes Maury's Bed 8 (Boca de Serpiente of Maury, 1929); Beds 1, 2 and 3 are included in the Soldado formation (Paleocene); beds 4-12 in the "San Fernando or Mount Moriah" formation of Trinidad (bed 12 somewhat doubtfully included due to absence of fossils). Detailed description of the beds 1-3, in ascending order (slightly abbreviated and suppressing the fossil lists) are as follows:

Bed 1.—Massive layers of pale brown and white impure glauconitic limestone with scattered large oysters, with some pockets of comminuted shell fragments and echinoid fragments; upper part more glauconitic. Contains some of the same species as Maury's Bed 2 (list, Kugler, 1938, p. 213). Thickness about 20 meters. Exposed in south and southwest part of island.

Bed 2.—Two meters thick, called the "Venericardia limestone" in crosssections; begins with 40 cm. of highly glauconitic grey marl with layers of coquina, followed by leached calcareous sandstone with fossil casts; above 1.1 meters of hard glauconitic limestone full of fossils, and 20 cm. of a leached calcareous sandstone with fossil casts. Identified with Maury's Bed 2. Blocks of coquina redeposited in Bed 5 of the upper Eocene, or on the slope, are considered as derived from here. Rutsch identified a considerable fauna (list, Kugler, 1938, p. 214).

Bed 3.—Called in the geological sections, the "Pellatispirella" limestone (in modern nomenclature, it would be Miscellanea according to Vaughan, "Ranikothalia" of Caudri's use). Total thickness, 2.2 meters begins with about 70 cm. of fine-grained, well-bedded brown clays and sands, with occasional small pebbles of quartz; this passes upward into beds with an increasing number of calcareous lenses full of foraminifera, followed by 50 cm. of a hard dense limestone (the "Pellatispirella" limestone in a limited sense). Large and small foraminifera were cited (Kugler, 1938, p. 215).

The total thickness for these beds referred to the Soldado is thus 24.2 meters.

In the legend of the geologic map, Kugler indicates another unit, the "Discocyclina limestone", also referred to the "Midwayan". In the text, Kugler explains that Bed 4 is a "rubble bed" composed principally of fragments of the underlying "Pellatispirella" limestone, but that it also contains in places large blocks of an impure grey limestone which carries "Discocyclina soldadensis Vaughan and Cole, and which is therefore called the "Discocyclina limestone". This limestone is not represented in situ on the island; Kugler (1938, p. 217) says:

"It is assumed that this limestone either represents a facies of the Pellatispirella limestone itself, or is a younger zone of the Soldado formation overlying the Pellatispirella limestone".

However, although the individual blocks of Bed 4 are composed of these limestones, the bed as a whole is considered by Kugler to be a basal conglomerate of the overlying upper Eocene formation, comparable with a basal conglomerate in Mount Moriah which rests upon marls of possibly middle Eocene age.

Vaughan and Cole (1941) described Kugler's collections from Soldado Rock (as well as other Trinidad species). From the Soldado formation they identified the following: "Miscellanea antillea (Hanzawa), M. soldadensis Vaughan and Cole, Discocyclina (Discocyclina) barker V. & C., D. (D.) grimsdalei V. & C. and Pseudophragwina (Athecocyclina) soldadensis V. & C. The locality numbers show that the first three came from the upper part of Kugler's Bed 3 (his "Pellatispirella limestone"), the last from one of the blocks in Bed 4 (of Discocyclina limestone). The age was given as "Lower Eocene". The authors explain (p. 25-26) that they avoid the use of Paleocene, as the usage of the U.S. Geological Survey does not agree precisely with Vaughan's interpretation of the European Paleocene as including the Sparnacian. They conclude: "With reference to whether beds 3 and 4 of Dr. Kugler's Soldado Rock section should be referred to the Eocene or Paleocene, the horizon i5 close to that of" [two American species] "which occur in the upper part of the Nanafalia formation" of Alabama. The Nanafalia formation, according to the U. S. Geological Survey, would be lower Eocene; according to European usage, it would be Paleocene. (Vaughan and Cole, 1941, p. 26).

In 1942, Cushman and Renz described the smaller foraminifera of the Soldado, that is, of Bed 3 (Kugler's sample K. 2950). They found numerous species known from the Midwayan of Texas and/or Alabama, but also numerous new species and varieties. They noted that the fauna, although considered of Midwayan age, showed also a distinct relationship with that of the Wilcoxian lower Eocene from Salt Mountain and Ozark, Alabama, and that the same species which show this, are also closely related to, or identical with, species from the middle Eocene of north Africa. The same authors, in 1946, placed the fauna of Bed 3 (considered Midwayan) as representing a level above the upper zone of the Lizard Springs or at the base of the Navet. In 1948, (p. 2) they expressed the opinion that in some well sections in southern Trinidad, the equivalent of the Soldado formation might be represented in the basal part of the Navet formation, possibly by the Ramdat marl "which lacks the genus Hantkenina strongly represented throughout the rest of the Navet".

Liddle (1946) corrected the definition of the Soldado to agree with Kugler (1938) and Vaughan and Cole (1941). The Soldado was described in the part of the book dealing with Venezuela; the Marac Quarry horizon under Trinidad (p. 729-732).

Kugler (1953, p. 39) proposes to remove Bed 3 from the Soldado formation, on the ground that the Soldado was originally defined as Midwayan, and that Bed 3 represents Wilcoxian lower Eocene:

"In the midst of the glauconitic limestone of Marac we have succeeded in finding a lense of glauconitic marl with an assemblage of small arenaceous foraminifera which turned out to be completely different from any other fauna known from Trinidad. This completely excludes Bed 3 from the Soldado formation, which in its original definition represents the Midway stage of the Gulf Coast. Bed 3 is 70 cm. thick and contain,s in its lower part a fine-grained, well-bedded brown sand and silt with some quartz pebbles the size of a cherry stone and thus suggests a stratigraphic break. This break is indicated by a fauna of small foraminifera reminiscent of the Paleocene, to which percentage of Lizard Springs smaller foraminifera of an aspect that suggests reworking".

Kugler refers the "Pellatispirella" and "Discocyclina" limestones represented in Beds 3 and 4, to the lower Eocene s.s., citing the opinions of Vaughan and Cole (1941) regarding the comparison with the Nanafalia, and the comparisons by Cushman and Renz (1942) of the microfauna with the Wilcox. Ee concludes (1953, p. 44): "The foraminiferal fauna of Bed 3 of Soldado Rock has a Midwayan aspect, but in the writer's opinion, must be considered reworked".

Concerning the relationship of the Soldado with other formations, Kugler considers that it is older than the Chaudiere-Lizard Springs (referred by him to the Paleocene), which in turn are older than the Pointe-a-Pierre (given as Paleocene-Lower Eocene). He states:

We have no knowledge as to the relationship between the ChaudiereLizard Springs formations and the Soldado formation, but the writer assumes their superposition on the Soldado formation because Bed 3 of Soldado Rock contains a high percentage of Lizard Springs smaller foraminifera of an aspect that suggests reworking".

Concerning the nomenclature of Bed 3, Kugler (1953, p. 44) does not as yet make any formal recommendation:

"It is still hoped that a better and more normal section of the Wilcox Beds of Trinidad will be found, otherwise it may be necessary to introduce a term such as "Serpent formation" for Bed 3, or adopt a well defined stratigraphic term from Venezuela, but certainly not Guasare formation. On the other side one may look to Barbados where the Miscellanea limestone block of the Joe's River mudflow may ultimately be found to form a part of the Lower Scotland formation. Caudri (1948, p. 271) lists a fauna of larger foraminifera from the Lower Scotland formation which she considers to be of Lower Eocene age but which is similar to that of the Upper Scotland formation".

Kugler (1953, p. 38) compares the Soldado formation with the Guasare, but complains that this has not been adequately defined, especially as regards its relation with underlying beds. He mentions that Van Raadshoven (1951) has described large foraminifera from the upper Misoa river, which the latter compares with species from Beds 3 and 4 of Soldado; but Kugler refuses to accept the inclusion of these orbitoid beds in the Guasare, "since it was restricted by Liddle (1946, p. 301) to the more or less 100 m. thick limestone of the Guasare River" and "this limestone is a glauconitic coquina with a Midwayan fauna which is alien to the orbitoids described by Van Raadshoven". (Kugler, loc. cit., p. 39).

The writer wishes to make the following observations and recommendations:

(1) Authorities familiar with the eastern Venezuela-Trinidad sections find no evidence to support Kugler's assumption that the Soldado formation is older than the Chaudiere-Lizard Springs formations, but on the contrary, that the relations are as indicated by Cushman and Renz. Thus, the presence of reworked Lizard Springs foraminifera in Bed 3 in quite natural. The difference in age between Beds 1-2 and Bed 3, in this interpretation, is th,erefore much less than in Kugler's interpretation, and hence, there is much less reason for separating Bed 3 as a separate unit.

(2) The paleontological evidence for referring Bed 3 to the Wilcox stage is not convincing. No new evidence is presented by Kugler, and earlier investigators (Cushman and Renz, 1942) interpreted the microfauna as of Midwayan age, though with some Wilcoxian precursors. Vaughan and Cole's (1941) suggestion of a correlation with the Nanafalia beds of Alabama (referred by the U. S. G. S. to the Wilcox stage), as these authors noted, suggests not so much a lower Eocene age for this horizon as the desirability of revising the North American correlations to conform more closely with the definition of the Paleocene in Europe.

(3) The most important argument against the exclusion of Bed 3 from the Soldado formation is, that this separation is proposed solely on paleon tological grounds, and hence loses sight of the difference between lithogenetic units (formations) and biostratigraphic ones (faunal zones). In our interpretation, the arguments presented would, at the most, justify the recognition of Bed 3 as a faunizone (although, as indicated above, the evidence is not too convincing). As such, it should receive a name based on its faunal characteristics, not a geographic name. Even assuming that the stratigraphic break interpreted by Kugler as present at the base of Bed 3 is a sufficient physical criterion, the extremely limited thickness of Bed 3 (at most 2 meters) is extremely small for a formation considered as a mappable unit.

There remain also the problems, that if Bed 3 were separated from the Soldado formation, it would cause great confusion in the geologic literature (since all the foraminiferal species described from the Soldado formation come from this bed). In addition, there is an acute shortage of available geographic names; "Serpent formation" as suggested by Kugler, is nearly a homonym of Maury's Boca de Serpiente formation.

We recommend, therefore, leaving Bed 3 in the Soldado formation, whether or not it proves to represent a slightly higher faunal zone than Beds 1 and 2, a point which in our opinion requires further investigation. As a formational name, we suggest that Soldado formation be applied only on Soldado Rock, in view of the geographical isolation of this island. Maurenbrecher (1952) has suggested that Kugler's Bed 3 of the Soldado formation be taken as the type of a "Soldado stage", with, as characteristic fossils, the larger foraminifera described by Vaughan and Cole. This proposal seems to have been generally ignored. We suggest that it is somewhat premature, and that a stage should be defined ideally not merely on the basis of the microfauna, but of the macro-fauna also.

The macro-fauna of the Soldado formation has been described recently by Rutsch (1943).

Frances de Rivero