Kay, Marshall

views updated


(b. Paisley, Ontario, Canada. 10 November 1904; d. Englewood, New Jersey, 3 September 1975)


Marshall Kay was the older of two sons born to George Frederick and Bethea Kay, both natives of Ontario, but of Presbyterian Scottish and Yorkshire derivation. The father, who was educated at Toronto and Chicago and became a distinguished geologist. moved his family to the United States in 1904 when he accepted a position on the faculty of the University of Kansas. In 1907 he transferred to the University of lowa as professor of Pleistocene geology, but soon became department head and state geologist (1911) and then dean of liberal arts (1917–1941). Marshall became a United States citizen at the age of five and grew up in lowa City, but he retained a lifelong affection for Canada and Scotland. Indeed. he did considerable research in eastern Canada. He was originally christened “George Marshall Kay” and in his early years used “G. Marshall,” but he later dropped “George” entirely to minimize professional confusion with his father.

Besides the model of an illustrious geologist father. Marshall Kay was influenced as a student by such other lowa faculty members as J. J. Runner, A. O. Thomas, A. C. Trowbridge, and C. K. Wentworth. He was also in the company of several exceptionally promising geology classmates, such as R. E. King, P. B. King H. S. Ladd, A. Pabst, and M. Stainbrook. These nascent young scientists amused themselves by solving mathematical puzzles and practicing memory training with railroad timetables and baseball statistics. Marshall received the B.S. degree cum laude from Iowa in 1942 and the M.S. in 1925. He shared the Lowden Prize with P.B. King and received a special citation for his M.S. thesis. In 1924 he attended a University of Chicago field geology course taught at Ste. Genevieve. Missouri, which complemented his 1923 University of Iowa field course conducted at Baraboo, Wisconsin. Kay always considered himself to be a field geologist. and he continued to do active field research up to his death.

Kay chose Columbia University for further postgraduate study, and after receiving the Ph.D. in 1929 commenced a forty-four-year teaching career at that institution. He began as lecturer and then instructor at Barnard College (1929–1931), soon became instructor at Columbia College (1931–1937), and subsequently advanced through the professorial ranks to his ultimate appointment as Newberry Professor of Geology in 1967. Marshall frequently represented Columbia on the faculty of the University of Wyoming’s Summer Science Field Camp, where he met his wife-to-be, Inez Clark, who was a botany student from the University of Michigan. They were married in 1935 and raised four children, three of whom studied earth science.

Geology at Columbia University had its beginning in 1792 with the appointment of Samuel L. Mitchill as professor of chemistry and natural history in Columbia College. Edinburgh-trained Mitchill, who was a well-known contemporary of William Maclure, comte de Volney, and Amos eaton, was one of the first persons to teach geology in an American university. Stratigraphic geology, which was Kay’s specialty, was not identifiable as a distinct study at Columbia until the creation of the department of geology and paleontology and the appointment of John S. Newberry as its first professor (1866–1892). Stratigraphy continued under the guidance of Amadeus W. Grabau (1901–1919) and J. J. Galloway (1919–1931). Kay became the fourth in this distinguished lineage when he took the reins in 1931 from his mentor, Galloway. who had resigned to move to Indiana.

Kay’s midwestern roots led to an early emphasis upon Paleozoic stratigraphic paleontology. His M. S. thesis director at Iowa. A. O. Thomas, had himself been a student of paleontologist stuart Weller at the University of Chicago. Weller’s approach emphasized faunal assemblages more than phylogeny, and it stressed the ability to identify fossils in the field. The Weller stamp was clearly transferred by Thomas, for Kay both practiced and taught this approach at Columbia. Kay’s thesis research on the stratigraphy and paleontology of the Ordovician Decorah Formation in Iowa began his lifelong love affair with lower Paleozoic strata and their invertebrate fossils; only rarely did he stray very far from the Ordovician.

Stratigraphic paleontology or biostratigraphy began in America a century before Kay’s student days with Lardner Vanuxem (1792–1848). who introduced the concept of index fossils and faunal correlation developed first in Europe by William Smith and Georges Cuvier. This field flourished in the latter part of that century in the hands of James Hall (1811–1898), one of Vanuxem’s many protégés and younger colleagues on the newly organized New York Geological Survey. By the 1920’s and 1930’s, stratigraphic paleontology was a maturing specialty in which heated debates often developed over issues that assumed a degree of refinement of the fossil record not always justified by the available evidence. Some of these exchanges were heated and personal.

Young Kay was himself involved in some of the acrimony, especially with the opinionated E. O. UI rich, who had proposed in 1911 a whole new geologic system, the Ordovician strata of the uppermost Cambrain and lower Ordovician strata of the conventional time scale. Ulrich discounted Kay’s early work on the Ordovician because it did not support his own prejudices. This was not to be the last time that Marshall would be involved in controversy.

Kay’s own research interests and teaching mirrored the evolution of stratigraphy in America during the middle twentieth century. Although he never abandoned his early training in biostratigraphy, already during the 1930’s he was becoming increasingly interested in physical aspects of the subject, that is, the fossil-containing strata themselves and their complex variations both laterally and vertically. He acknowledged that it was J. J. Galloway who had inspired his own interest as a graduate student by “taking the dry bones of stratigraphy and giving them flesh and life.” Kay, likewise, gave gave life to a subject that, in the hands of less skilled teachers. could be the dullest of all geology courses. He stress the distribution of strata in time and space as records of tectonic evolution and used the stratigraphic record to illustrate general principles rather than to present it descriptively as an end in itself. The book Stratigraphy and Life History (1965), coauthored with E. H. Colbert, provides a hint of the approach. One of his pedagogical techniques was to insist that students extract many of the regional relationship themselves from stratigraphic data provided on countless posterlike diagrams drawn from the literature for teaching aids.

During the 1940’s and 1950’s the preeminence of the classic biostratigraphic emphasis was being supplemented by a new, rapidly evolving physical approach to stratigraphy, which received great impetus from both the needs of and the subsurface data from the petroleum industry. Together with other Americans such as L. L. Sloss, W. C.Krumbein, E. C. Dapples, H. E. Wheeler, and A. I. Levorsen, Kay emphasized the importance of lithofacies as well as biofacies, stressed the inherent limitations of index fossils, and refined the concepts of chrono-(or time-) stratigraphy, Following World War II, stratigraphy became increasingly concerned with the analysis of regional patterns of thickness, facies, and unconformities, as well as the relationships of these to major tectonic elements. Marshall pursued such analyses primarily in New York. Vermont, Ontario pennsylvania, and Virginia (1930’s-1940’s). Nevada and Utah (1950’s)and Newfoundland (1960’s1970’s), but he was always very well versed in other regions through the literature. His papers “Analysis of Stratigraphy” and “Isolith, Isopach, and Palinspastic Maps” were harbingers of the new physical stratigraphy.1Other workers, including Sloss and Wheeler, broke more completely with the old stratigraphy, however, and developed regional stratigraphic analysis even further. Present-day emphasis on sequence stratigraphy and the importance of global (eustatic) sea-level changes as a major determiner of the sequences is the latest outgrowth of the “new” stratigraphy.

Beginning early in his career, Kay emphasized that one of the ultimate goals of stratigraphy should be the refinement of paleogeographic restorations. His 1945 paper “Paleogeographic and palinspastic Maps.” although received at first with skepticism soon proved to be a fundamental step forward for the study of complexly deformed orogenic belts.2 Although Kay was always more stratigrapher than petrologist, he urged attention to such details as conglomerate pebble compositions as an important clue to the histories of orogenic belts. In this he anticipated in some measure an important phase of sedimentologic studies that was to mature in the 1960’s and 1970’s namely detailed attention to sedimentary petrography and paleocurrent measurements in paleogeographic studies. This approach was pioneered in America primarily by sedimentary petrologists P.D, Krynine and F. J. Pettijohn.

After coming to Columbia University, Marshall had became increasingly concerned with tectonics, which led to the contributions for which he was to be most famous, because of Columbia’s location within the classical Appalachian orogenic belt, development of this interest might seem inevitable, but its seeds actually were sown back on the midwestern prairies before he arrived in New York City. In his early lowa work Kay had encountered an Ordovician altered volcanic ash layer and wondered about its origin. In two papers he argued that this Hounsfield Metabentonite extended from New York State all the way to Iowa and Minnesota.3 Because the Hounsfield thickened toward the southeast and because more such Ordovician ash layers appeared in the Appalachian region. Kay inferred that a volcanic source lay somewhere in the southeastern United States. On a paleogeographic map published in 1935 he showed such a source superimposed upon the orthodox Appalachia borderland.4 This was the first step in an assault upon conventional wisdom about the evolution of mountain belts, which already for more than half a century had been regarded as the locus of generation of continental crust, and so was of truly fundamental global importance.

An enormously influential concept of borderlands, that is, long-lived highlands lying along the edges of continents and composed of Precambrian crystalline rocks, had dominated American thought beginning with publications in the 1870’s by J. D. Dana of Yale University. Dana’s ideas, in turn, had been stimulated by James Hall’s celebrated but incomplete theory of mountains (1857–1859). Erosion of the borderlands for hundreds of millions of years was thought by Dana to have supplied great volumes of sediments to adjacent, long, subsiding great volumes geosynclines to Dana. These subsiding belts lay between the presumed borderlands and the more stable continental interior. After hundreds of millions continents.

Dana’s venerable scheme of Precambrian borderlands was threatened by the 1934 discovery by M. P. Billings in New Hampshire of marine Devonian fossils in metamorphosed sedimentary and volcanic rocks within the very heart of the presumed Pre-Cambrian borderland. Kay’s next paleogeographic map, drawn in 1935 but published in 1937, now showed a marine trough in New England rather than the old borderland.5 This was the first major break with orthodoxy. In 1936 Kay saw for himself the important New Hampshire rocks with Billings and in 1937, on an International Geological Congress field trip in the Ural Mountains, he again observed volcanic rocks associated with Paleozoic sedimentary ones.

For his graduate course in stratigraphy Kay compiled the distribution of major rock types on continental-scale maps for different segments of geologic time. From his compilations as well as the critical field observations noted above, he was beginning to see in the late 1930’s a consistent pattern of distribution of ancient volcanic rocks along the continental margins.

Kay’s illumination was bolstered by an encounter with the important writings of H. Stille and H. H. Hess. Hans Stille contrasted geosynclines with stable continental interiors, which he called “cratons.” He also recognized two parallel subdivisions within what had been termed geosynclines since Dana’s time. To formalize his distinction Stille coined the term orthogeosyncline (straight geosyncline) for the entire linear zone of thick strata first denoted by Hall and Dana. This he then subdivided into a belt of thick strata with associated volcanic and other igneous rocks, which he named eugeosyncline (truly or wholly geosyncline), and a complementary nonvolcanic zone, which he named miogeosyncline (lesser geosyncline).

Because Stille’s distinction fit the pattern that Kay himself had discovered independently. Kay adopted the terminology and subsequently added several new categories. Meanwhile Hess had begun in 1939 to draw the attention of Americans to analogies between modern volcanic island archipelagoes and ancient mountain belts, a comparison that had begun in Europe half a century earlier. In particular Hess had noted that zones of serpentine rocks in ancient mountain belts occupied positions similar to such rocks in modern volcanic arcs. Kay was impressed with this comparison as well as the Alpine geologists’ comparison of their ophiolite suite (consisting of serpentine, gabbro, basalt, and chert) with oceanic volcanic rocks. Moreover, he noted that a number of mountain belts located along continental margins pass oceanward directly into volcanic archipelagoes, his favorite example being the Aleutian are extending into the North Pacific from southern Alaska.

Marshall Kay’s most important contribution to the understanding of ancient mountain or orogenic belts developed during the 1940’s and culminated in the publication in 1951 of his famous Geological Society of America Memoir 48. North American Geosynclines. Kay concluded that the nature of strata of geosynclines depends upon the complex interrelationships of uplift, adjacent subsidence, sedimentation, and presence or absence of contemporary volcanism. Most important, geosynclines, though generally subsiding, have at different times and places contained ancient volcanic arcs or tectonic lands, or both, raised within them. Kay argued that such internal lands were the major sources of clastic geosynclinal sediments rather than hypothetical borderlands external to the geosyncline. His tectonic lands were a special innovation that allowed for cannibalism of much geosynclinal sediment from but-slightly-older rocks of varied types. Thus was completed the overturn of the paleogeography and tectonic history of mountain belts. By the late 1940’s A. J. Eardley (who had studied with Hess at Princeton) and others were also comparing ancient volcanic belts with modern arcs. Kay was always conscious of great complexity and ambiguity in rocks, especially in rocks, especially in those of deformed mountain belts. Although a master of synthesis and generalization himself, he regarded the principle of simplicity as a dangerous tool when carelessly applied. “Much of nature is not simple,” he frequently observed. It was characteristic for he frequently observed. It was characteristic for him to draw attention to differences of pattern and exceptions to simplifications, In fact it was to emphasize regional differences of stratigraphic and tectonic patterns that he proposed in the 1940’s his well-known and much criticized classification of geosynclines. Given his own extensive early training in stratigraphic paleontology and his admiration for Stille’s insight, it seems natural that Kay would propose additional taxonomic categories of depositional basins besides Stille’s orthogeosyncline and its two subdivisions. It was also natural that he would extend Stille’s binomial approach by coining a Greek prefix appended to the suffix geosyncline for each new category, as in the following list:

Orthogeosyncline (from Stille) 
EugeosynclineVolcanic-bearing zone
MiogeosynclineNonvolcanic zone
EpieugeosynclineNonvolcanic troughs on former eugeosynclines
TaphrogeosynclineFault-bounded troughs
ParaliageosynclineCoastal-plain wedges of sediments
AutogeosynclineSubsidence without complementary highlands
ZeugogeosynclineSubsidence yoked to complementary highlands
ExogeosynclineSubsidence with sediment derived from beyond the craton (that is, from an uplifted orthogeosyncline)

One may wonder why, during a period of still great taxonomic emphasis in geology, this classification aroused so much antipathy. Two common criticisms suggest a partial answer. First. it was confusing to have “geosyncline” now applied both to a first-order feature (Stille’s orthogeosyncline) and also to subdivisions thereof (euand miogeosyncline), and, second, Kay’s extension of “geosyncline” to include cratonic basins blurred the whole geosynclinal concept, which until then had been linked with major mountain belts. Attention to the important patterns that Kay wished to emphasize by establishing his categories was obscured by semantics and emotional reaction for more than two decades. Indeed, his classification was a frequent butt of jokes. Ironically it was overlooked that “miogeocline,” which was coined by others in the 1960’s and quickly received wide currency, was a redundant synonym of Kay’s “paraliageosyncline.” With the advent of the sea-floor spreading hypothesis in the early 1960’s and the revolutionary theory of plate tectonics in the late 1960’s the significance of tectonic distinctions that Kay had emphasized finally came to be appreciated, but his terms remained generally unpopular. While a few workers strained to fit his terminology to the new paradigm, Kay recognized that the older terminology was no longer very appropriate. It had served its purpose by drawing attention to diverse tectonic patterns that could now at last be mostly explained by the new theory. He put the situation in perspective in 1967 when he wrote, “The concern about the terms has resulted in more penetrating analyses of the history in the rocks: the very endeavour to classify has been rewarding.”6

Marshall Kay was also active in several significant investigations of lesser scope than global tectonics. For example, beginning with early work on Lower Paleozoic rocks along the New York-New England boundary, he developed a long-standing interest in overthrust faulting, especially where different sedimentary facies have been juxtaposed. He carried this interest to Nevada in the 1950’s and still later was among the first to suggest large-scale overthrusting in western Newfoundland. He also was early to recognize the significance of exotic carbonate-rock boulders contained in Ordovician shales in Quebec and Newfoundland. And with characteristic insight he recognized in his later years, in light of plate tectonics, that the OttawaBonnechere graben of southeastern Ontario and adjacent Quebec, which he had mapped in the late 1930’s, represented an aulacogen.

Theyear 1967 was a landmark in geology, for it was the year that the theory of plate tectonics was first announced, that Kay was named Newberry professor, and that he convened the International Gander Conference on Stratigraphy and Structure Bearing on the Origin of the North Atlantic. Even in his sixty-third year the man was still at the forefront of his field with a perfect sense of timing to bring together in Gander, Newfoundland, over 100 leading authorities on Western European and eastern North American geology. Unlike most symposia, which are soon forgotten, this conference had exceptional impact. At a critical time in the history of geology, as nothing before it focused attention upon the compelling evidence for continental separation across the North Atlantic ocean basin. Many fruitful collaborations by workers from both sides of the Atlantic were forged here, and a momentous symposium volume of 1, 082 pages appeared two years later, thanks to the careful and efficient attention of its editor, Marshall Kay.

It is poignant that the American Association of Petroleum Geologists, which published the Gander volume, had also sponsored in 1928 a famous symposium in New York City that was very uncomplimentary to continental drift, in keeping with the characteristic American antipathy for that idea. Kay was then a Ph.D. candidate at Columbia University and was thoroughly steeped in the American tradition of permanently fixed continents. Thirty-nine years later, however, his open-mindedness and firsthand knowledge of Lower Paleozoic rocks in both eastern North America and the British Isles allowed him to see clearly the merit of continental separation across the Atlantic in spite of that intellectual legacy.

Marshall Kay’s many honors included election to Phi Beta Kappa in 1924, receipt of the Kunz Prize of the New York Academy of Sciences in 1941, and election as honorary foreign member of the Geological Society of London in 1964 and honorary correspondent of the Geological Society of Stockholm in 1968. He received the prestigious Penrose Medal of the Geological Society of America in 1971, the Distinguished Service Award of the University of lowa in 1971, and an honorary degree from Middlebury College in 1974. He was honored by a special’ Conference on Modern and Ancient Geosynclinal Sedimentation’ in 1972 and another on “Paleozoic Margins of Paleo-American and PaleoEurafricank Plates” in 1975.

Kay was a member of many professional societies, including the Geological Society of America, Paleontological Society, Geological Association of Canada, American Association of Petroleum Geologists, New York Academy of Sciences, Society of Economic Paleontologists and Mineralogists. American Association for the Advancement of Science, American Geophysical Union, Paleontological Society of Japan, Paleontological Association, and the lowa Academy. He was a member of the board of managers of the New York Botanical Garden, the American Commission of Stratigraphic Nomenclature, and the International Commission on Stratigraphy, and was a delegate to four international congresses. During 1966 and 1967 he was invited to lecture in the U.S.S.R., Sweden, Canada, and Great Britain.


1.In Bulletin of the American Association of Petroleum Geologists, 31 (1947), 162–168, and 38 (1954), 916–917.

2.Ibid., 29 (1945), 426–450.

3.“Age of the Hounsfield Bentonite,” in Science, 72 (1930), 365. and “Stratigraphy of the Ordovician Hounsfield Metabentonite,” in Journal of Geology, 39 (1931), 361–376.

4. In “Distribution of Ordovician Altered Volcanic Materials and Related Clays”, in Bulletin of the Geological Society of America, 46 (1935), 243.

5. In “Stratigraphy of the Trenton Group,” ibid., 48 (1937), pl. 7.

6. “On Geosynclinal Nomenclature”, in Geological Magazine. 104 (1967). 315.


I. Original Works. Kay wrote almost 200 publications, the most important of which are listed in the Geological Society of America’s Memorial Volume, 7 (1977). A dozen or so more papers on stratigraphy and paleontology published in the 1930’s first established his reputation in America. The 1945, 1947, and 1954 papers cited in the article were major contributions to stratigraphic concepts. Papers published in 1930 and 1931 on the Hounsfield metabentonite and papers on Ordovician paleogeography in 1935 and 1937 (also cited in the article) were the precursors of his contributions on geosynclines and mountain belts. “Geosynclines in Continental Development,” in Science, 99 (1944), 461–462, was his first major paper on this theme. It was followed by “Geosynclinal Nomenclature and the Craton,” in Bulletin of the American Association of Petroleum Geologists, 31 (1947), 1289– 1293; and his single most important publication, North American Geosynclines. Geological Society of America Memoir 48 (1951).

In 1955 he was invited to write “The Origin of Continents” for Scientific American, 193 (September 1955), 62–66. The Gander conference symposium volume. North Atlantic: Geology and Continental Drift (Tulsa, Oklahoma, 1969), was edited by Kay and contains five articles by him. This was his last major contribution on large-scale tectonics, but the following two papers provide important retrospective insights by Kay on his involvement with geosynclines and tectonics: “On Geosynclinal Nomenclature,” in Geological Magazine, 104 (1967). 311–316; and “Reflections: Geosynclines, Flysch, and Melanges.” in Modern and Ancient Geosynclinal Sedimentation (Tulsa, Oklahoma, 1974), a symposium volume dedicated to Kay.

While all of these contributions to the literature on tectonics were appearing, Kay continued to publish many papers dealing with stratigraphy, paleontology, and structure of the regions in which he was doing fieldwork. Many of these were important contributions even though of more restricted geographic interest. An example is his last publication. “Dunnage Melange and Subduction of the Protacadic Ocean, Northeast Newfoundland,” in Geological Society of America Special Paper 175 (1976). Another major contribution of still a different sort was the textbook Stratigraphy and Life History (New York, 1965), written with E. H. Colbert.

II. Secondary Literature. Short biographical treatments of Kay have been published by P. A. Chenoweth in Bulletin of the American Association of Petroleum Geologists, 60 (1976), 1129–1130; and R. H. Dott, Jr., provides an overview of changing ideas about geosynclines from Hall and Dana to plate tectonics and contains an assessment of Kay’s contributions. K. J. Hsü, “The Odyssey of Geosyncline,” in Robert N. Ginsburg, ed., Evolving Concepts in Sedimentology (Baltimore, 1973), provides a similar overview from a different perspective. Dott presented further historical analysis illustrated by reproductions of several paleogeographic maps (including three of Kay’s) in’ Tectonics and Sedimentation a Century Later, ’ in Earth-Science Reviews, 14 (1978), 1–34.

R. H. Dott, Jr.