USGS Logo Geological Survey Professional Paper 294—D
Stromatolites of the Belt Series in Glacier National Park and Vicinity, Montana



The term "stromatolite" (Kalkowski, 1908, p. 68) has been generally accepted as applying to laminated structures attributed or possibly attributable to the work of blue-green (or green) algae. Gurich (1906, p. 7) proposed the name Spongiostromidae [sic] for a family of laminated forms from the Visean limestone of the Namur Basin of Belgium. He considered the possibility of their algal affinity (1906, p. 31) but rejected the idea in favor of referring the forms to the Protozoa. Later writers have indicated that the Spongiostromidae are probably calcareous algae. Julius Pia (1927, p. 36) used the family name "Spongiostromata" as a major division of the class Schizophyceae (Cyanophyta) and divided the family into two parts. Under "Stromatolithi" he included all forms, such as the genera Collenia Walcott, Cryptozoon Hall, and Spongiostroma Gurich, that grew attached to the substratum. The second division he called "Oncolithi," and in it he included unattached forms, such as the genera Osagia Twenhofel, Pycnostroma Gurich and Ottonosia Twenhofel. This grouping appears to be valid and its retention seems to be desirable.

The present study discriminates between fossil algae and stromatolites. Fossil algae preserve recognizable organic microstructures that enable the examiner to determine their true biologic relationships. Such microstructures include cell walls, reproductive organs, and other parts. The stromatolites, however, rarely exhibit recognizable microstructures beyond a fine lamination. They are large headlike masses of sediment that have come into existence through the work of certain lowly forms of algae. The remains of the algae that built the structures are only exceptionally preserved (Tyler and Barghoorn, 1954, p. 606; Woodring, 1954, p. 222). All that is ordinarily left of the algae to attest to their original presence are the stromatolites.

There have been basic differences of opinion among authors regarding the use of generic and specific names for stromatolites. O. A. Høeg (1929, p. 8) states:

There are two facts rendering it preposterous to create names for these structures, however useful such names may be: We cannot be sure that identical coenoplases [colonies] are not farmed by systematically widely different organisms; and in most cases, the coenoplases will certainly come into existence through the cooperation of more [than one] species.

P. E. Cloud, Jr., (1942, p. 363, 366) concurs with Høeg and suggests that as a matter of convenience the familiar form names be retained and used in the vernacular sense. Paul Dorn (1953, p. 36) refuses to recognize any classification of stromatolites. R. B. Young (1933, p. 32), who has contributed greatly to the knowledge of stromatolites in South Africa, states:

Since stromatolites are as a whole devoid of organic structure, their position in any scheme of botanical classification can be justified only on the grounds of analogy with calcareous structures * * * produced by known algae. The generic and specific names sometimes applied to stromatolites have avowedly little or no systematic value.

As early as 1906, H. M. Seely (1906, p. 170), in discussing the genus Cryptozoon, writes:

Seeking a place for them [Cryptozoon] in classification they are found to fall outside of any of the described orders. * * * Yet they are very distinct. They have a character of their own and they properly demand recognition. In a system of classification a place must be sought for them.

C. D. Walcott (1914, p. 107) in his pioneer paper on the stromatolites of the Belt series, states:

All of the genera and species are based on variation in form * * *. In the absence of the identification of the actual algae that built up the structure found in the fossil state a purely artificial classification has been adopted.

J. H. Johnson (1946, p. 1089) states:

They [stromatolites] may comprise form genera and form species rather than true biologic genera and species. Each form genus and species may represent several biologic species, or even a number of genera and species that lived in constant close association. Such associations, however, permit recognition of the colonies over wide areas through appreciable lengths of geologic time. Different assemblages formed colonies sufficiently different to be separated by megascopic characteristics.

Authors who have used names for stromatolites without justifying their use include F. M. de Almeida (1944); Lucien Cahen, A. Jamotte, and G. Mortelmans (1946); Georges Choubert (1945, 1952); Georges Choubert, R. Du Dresnay, and J. Hindermeyer (1950); Riuji Endo and C. E. Resser (1937); A. Jamotte (1944a, 1944b); V. P. Maslov (1937a, 1937b, 1938, 1939a, 1939b); Nicolas Menchikoff (1946); A. Meyer (1954); Julius Pia (1927); M. R. S. Rao (1949); W. H. Twenhofel (1919); and others.

Alfred Riedel (1953, p. 674) suggests a compromise. He favors the retention of generic names, such as Collenia and Conophyton, to be used to designate geometric forms but to have no biological significance.

C. L. Fenton (1943, p. 105 and 106) comments:

For generations cryptogamic botanists have been naming genera and species of lichens despite the fact that every lichen consists of a fungus and an alga. Some 15,900 of these "improprieties" are now established in botanical literature. Surely they form ample precedent for paleobotanical species of comparable nature * * *. As has been pointed out, recognition of genera and species among stromatolites rests primarily on the stability of their characters. It is well displayed by generally similar forms that do not intergrade, by forms that persist through thick formations despite sedimentary variations, and by those that have wide geographic range in clearly delimited horizons.

Stromatolites offer great promise for interpreting ecological relationships in sediments that are otherwise lacking in fossils. Also it has been found (Maslov, 1939b; Riedel, 1953; Fenton and Fenton, 1937; Rezak, 1953; Cahen, Jamotte, and Mortelmans, 1946; and others) that some of these fossils are useful in local correlation. It is evident that names should be applied to these structures. The rules for paleobotanical nomenclature (Lanjouw, 1952, p. 64) define a form-genus as "one that is maintained for classifying fossil specimens that lack diagnostic characters indicative of natural affinity but which for practical reasons need to be provided with binary names." A plant taxonomist might validly object to the application of this principle to stromatolites on the basis that stromatolite species do not represent biologic entities.

In the present paper, I have used the names of form-genera and form-species with the understanding that the classification is purely artificial and is used as a matter of convenience. However, I am aware of the possibility of confusing these names with the names of biological entities. Perhaps a new method of classifying these structure should be devised. This new method should not be based upon specimens from restricted geographic localities or geologic range but should result from a comprehensive examination of all kinds of "algaloid" structures from all parts of the geologic column. Also, if we are to separate the stromatolites from biological nomenclature, their new names should not have resemblance to biologic names. Such a project is beyond the scope of the present paper. It would probably require several years to devise and perfect a new nomenclatural system.


Reasons for great differences of opinion on how best to classify stromatolites are (1) lack of agreement as to the diagnostic basis for classification, (2) the impractically detailed systematic breakdown of some units that are here recognized as single, and (3) the fragmentary nature of the types of some species.

The stromatolites of the Belt series have been classified on the basis of form because of the apparent lack of microscopic organic structures. However, no statement as to what constitutes a form-genus or form-species of stromatolite is to be found in the published record. Some authors simply state that they are dealing with form-species (Walcott, 1914, p. 104; Fenton and Fenton, 1937, p. 1941). This has led to wide variation in the number and nature of recognized stromatolites.

To complicate matters, many so-called varieties of species have been erected. In many instances it has been found, by visiting sites in the field and examining types in museums, that the reason for the description of varieties is the incorrect orientation of natural sections. A vertical section through a colony of Collenia frequens Walcott shows the form of the colony to be a cylindroid, whereas a diagonal section through the same colony gives the appearance of an expanding column or turbinate form, such as Cryptozoon occidentale Dawson. If the two sections are found in two closely associated colonies, an observer may arrive at the erroneous conclusion that the expanding column is a variety of the columnar form. In other examples the reason for erecting varieties has been the discovery of a different form in rock units that are made up almost entirely of one species (Fenton and Fenton, 1933b, p. 1142). Even though the different form answers to the description of a previously described species, it has been described as a variety of the species with which it is associated. The creation of these varieties causes the interpretation of many descriptions of species to be purely a matter of personal preference, and the geologist who attempts to identify the species in the field experiences great difficulties.

The fragmentary nature of the types has proved to be a major stumbling block in comparison of specimens. Gross form of colonies is of major importance in distinguishing species, yet it is often impractical to collect complete colonies for type specimens. Some colonies attain diameters of 15 to 20 feet and weigh several tons. As a consequence, some of the types do not exhibit the diagnostic characteristics of the species. Therefore it is important that a photograph of the complete colony as it appears in the field be published in addition to the photograph of the holotype.



Two genera of stromatolites from the Belt series of northwestern Montana that have been described by Walcott (1914, p. 104, 110) are Collenia and Newlandia. He describes Collenia as "More or less irregular dome shaped, turbinate or massive, laminated bodies that grew with the arched surface uppermost." In describing the species Newlandia concentrica, Walcott states that it includes "Semispherical bodies built up of concentric layers of irregular thickness that appear to be attached at the base of each cup-shaped concentric layer." He comments further (1914, p. 111): "The resemblance between the structure of Collenia and Cryptozoon is marked in the hand specimens * * * but when we compare the mode of growth we find that Collenia has an encrusting-like growth that forms a dome-shaped body with the edges lamellae pointing downward, while Cryptozoon grows in a cup-shaped form with the edges of the lamellae on the upper surface."

J. H. Jolnson (1952, p. 54) does not agree with Walcott's comment. He observes: "Walcott's statement that 'Cryptozoon grows in a cup-shaped form with the edges of the lamellae on the upper surface' is not true except in a few rare cases. Usually in both genera the laminae are thickest on the top and curve downward. In other words they are arched." This leaves the reader wondering just how it is possible to distinguish between these genera. Walcott's statements regarding Newlandia and Cryptozoon are puzzling, and Johnson's comment creates doubt as to the validity of Walcott's genera. Walcott's concept of the mode of growth of Cryptozoon was apparently based upon observations of bedding surfaces in which the upper portions of the colonies had been planed off.

James Hall (1883, pl. 6), in his original description of the genus Cryptozoon, states: "A further examination shows the entire form of these masses to be hemispheric or turbinate, with the broadest face exposed upon the upper surface of the limestone layer, [and shows] that their growth has begun from a point below and rapidly expanding upwards, has often extended one or two feet in diameter." (See also Goldring, 1937, p. 530.)

Walcott (1914, p. 113) describes the genotype Collenia undosa as "laminated bodies that are usually concavoconvex. They appear very much as though the underside had been dug out or that the first encrusting calcareous deposit was made over a lump of mud. The interior of the body is made up of alternating fine and coarse laminations subparallel to the upper and lower surfaces of the body." Walcott's original description of Newlandia has already been cited.

It is now evident that Cryptozoon, Collenia, and Newlandia can be distinguished by their different modes of growth.

Mode of growth is the basis of generic distinction here adopted. Four distinct types of growth are recognized.

The first is typical of the genus Cryptozoon Hall. All species begin their growth from a point on the substratum and grow upward by the addition of convex upward laminae that increase in area as the colony develops.

The second type is exhibited by the genus Collenia Walcott. All the species begin as incrustations on a surface of the substratum. Growth is upward by the addition of convex upward laminae that do not increase greatly in surface area.

The third type is typical of the genus Newlandia Walcott. Here also the growth begins from a surface on the substratum, but it differs from Collenia by the addition of concave upward or saucer-shaped laminae.

The fourth type of growth is displayed by the genus Conophyton Maslov. Here the structure is cylindroidal, composed of laminae in the form of inverted nested cones. The cylindroids are attached to the substratum by the apex of the basal cone. The long axes of the cones are inclined at some angle to the bedding surfaces.

Because oncolites (unattached forms) are not known to exist in the Belt series, their classification is not discussed. Environments in which oncolites may develop are described in the section on modern environments.



Gross form of colony is of primary importance in determining species. The basic forms are cylindroidal, depressed spheroidal, and turbinate. However, combinations of these forms are recognized as distinct species. For example, Collenia multiflabella n. sp. is made up of two basic forms. The lower part of the colony consists of varying numbers of expanding columns. The upper part of the colony is dome shaped and completely covers the lower part. This gives the entire colony the form of a depressed hemispheroid. Cryptozoon occidentale Dawson is an example of the turbinate form. It develops into toplike structures in which the diameter is approximately equal to the height of the colony.


The nature of the laminae is an important although difficult characteristic to use in distinguishing species. Attempts have been made to measure the number of laminae per unit distance on several species, both on weathered surfaces and in oriented thin sections. However, these studies have not been encouraging and have been discontinued.

Ordinarily the shape of the laminae is reflected in the gross form of the colony. An example of this is Collenia symmetrica Fenton and Fenton, in which the laminae are parallel to the upper surface of the colony. There are, however, exceptions to the rule. Cryptozoon occidentale Dawson has a fan-shaped vertical cross section, but the laminae in the younger part of the colony are dome-shaped, becoming flattened as the colony increases in size. The flattened laminae are abruptly downcurved along their margins, and each one is but slightly larger than the one below it, giving rise to the fan-shaped cross section.

Laminae may also be smooth or crenulate. The regularity of the laminae seems to be consistent within each species. Conophyton inclinatum n. sp. and Cryptozoon occidentale Dawson exhibit smooth laminae, Collenia multiflabella n. sp. and Collenia symmetrica Fenton and Fenton have finely crenulate laminae, and Collenia undosa Walcott has coarsely crenulate laminae.


Size of colonies has been used by some authors as a specific characteristic, but, in my experience, size varies in each species and only form and laminae remain constant. In Glacier National Park, colonies of Conophyton inclinatum n. sp. range in size from a few inches to about 48 inches across the base of the cone. In other areas gigantic forms of Collenia symmetrica Fenton and Fenton have been observed. The largest colonies of this species in the Glacier National Park region are about 6 feet in diameter. In the Helena limestone, near Helena, Mont., the species attains diameters of 15 to 20 feet.


In the following descriptions of species, citations of museums and collections in which types and figured specimens are located are abbreviated. The abbreviations are as follows:

USNM—U. S. National Museum, Washington 25, D. C.

UCM—University of Cincinnati Museum, Cincinnati, Ohio.

PUM—Princeton University Museum, Princeton, N. J.

USGS—U. S. Geological Survey algae collection, Palentology and Stratigraphy Laboratory, Denver Federal Center, Denver, Colo.


The key that accompanies plate 19 (following p. 154) has been devised to serve as an aid to the identification of stromatolites in the Belt series. However, it is possible that it may be expanded to include all stromatolites. For this reason, the key has not been condensed but contains steps that may aid in the classification of some forms as yet unnamed. Plate 19 includes photographs of the seven species here recognized.

<<< Previous <<< Contents >>> Next >>>

Last Updated: 18-Jul-2008