TRADERS IN CANOES: THE GREAT LAKES FUR TRADE IN. THE 18th CENTURY CONCEPTS or DEGIACIATION’ IN SOUTHERN NEW ENGLAND Two Research Papers for the Degree of M. A. MICHIGAN STATE UNIVERSITY MARK C. SULLIVAN .1972 . . . - — o , ....... TTTTT - ‘OZVJUHUVT TIINTTINATITTIITI E [Fifi-j Univctsi: m TRADERS IN CANOES: THE GREAT LAKES FUR TRADE IN THE 18th CENTURY By Mark C. Sullivan A RESEARCH PAPER Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Geography 1972 Approved There are two reasons why the Europeans came to the new world: to convert the "heathen" and to search for wealth. To the individual coun- tries the latter was the most important, as a means of financing this exploration and increasing their treasuries. Their efforts inthe acquisi- tion of wealth can be contrasted between the French and the Spanish. The Spanish looking for gold, sacked the Aztec cities and chased after mythical golden cities in the southwest. The French, however, found that the wealth of the new world was fur. As early as 1534 Cartier was trading with the Indians for furs, and it is believed that some trading occurred even earlier (Sauer, l97l). This trade increased to become a major source of wealth for the countries involved (principally France and England) which lead to exploration, settlement, and war. There are basically three areas or regions of fur-trade in North America. The Rocky Mountains, particularly in the United States; the areas adjacent to Hudson Bay; and the Great Lakes region. This paper will focus on the latter area, and will be directed toward the fur-trader and the ef- fects of the fur-trade on the region. The importance of the fur-trade cannot be overemphasized. As stated by Vandiveer (1929): The story of the fur-trade is very largely the story of pioneer America (sic). The fur-trade was the agency through which all the vast interior of the continent was explored and made known to the world. The trappers and the traders were the real pion- eers and trail blazers, while the rude trading posts they es- tablished at various strategic points were the genesis of many of our modern cities. The fur-trade, however, had very humble beginnings. It started with the l -2- early French explorers doing a little trading with the Indians while they were exploring the rivers and bays. Later, in the 16th Century, permanent settlements were established. These became centers for fur-trading, be- cause of the Indians who would bring their furs to the settlements to trade for metal and cloth goods. The Huron Indians developed into traders acting as middlemen for the more westerly tribes. They traveled the waterways to settlements such as Montreal, to trade. By the l7th Century certain Indian problems occurred, which dras- tically altered the Canadian fur-trade in the Great Lakes. The Huron's generally used the Ottawa River to travel to Montreal. This area, however, was controlled by their enemies the Iroquois. Hence, the fur-trade was periodically stopped by the Iroquois, who traded with the English. The second problem was that in I609, Champlain and the French earned the hatred of the Iroquois after a short battle with them at the site where Fort Ticon- deroga would be built (Vandiveer, l929). This increased the hostilities with the Algonquin tribes, who made an alliance with the French against the Iroquois. These factors tended to create a very unsteady supply of furs coming to Montreal. Toward the close of the l7th Century a class of pe0ple grew up in Canada known as coureurs de bois, rangers of the woods, or voyaguers into the unknown. These voyaguers, as thev were called, learned the Indian ways. They were skilled with the birch bark canoe, woolcraft, and survival in the woods. Many of the voyaguers married Indian women and had families. They were hardy and wild, and upon these men the fur-trade developed. With the development of the Indian problems, a few traders would travel into Indian country to trade with the Indians. These men would be gone for one or two years, return to sell their furs, buy supplies, and -3- generally raise hell. They were a wild and unruly lot, who ignored most of the rules of society they did not like. To control these men, as well as to increase and control the fur-trade, the French government issued licenses to traders to go into the Indian country. Since the licenses were expensive and hard to obtain, a trader would hire a number of voyag- uers who had no licenses to carry supplies and help in the trading. The trader bartered knives, beads, wampum, blankets, and other trinkets for furs worth many times more. Particularly the beaver skins were sought. The beaver fur was in great demand in Europe. Due to the nature of the fur, particularly the beaver trapped in winter, it could be de- tached from the hide and pounded into a felt-like substance. This material was used in the manufacture of hats. Other furs were also of value, and in lesser numbers, other animals killed were: martin, muskrat, lynx, fox, otter, wolf, bear, etc. . Neither the traders nor the trappers had much notion of ecology, so whole colonies of beaver were killed and dams broken. Other animals were hunted until the area could not even support the indigenous population. Vast numbers of animals were killed during the fur-trade years (Table I). And as the animals were eliminated, the fur-trade moved continually west, north, and then south. When the white man first came to the new world, he was impressed by the rugged nature of the topography. Travel, as was known in Europe, to say the least, was difficult. The terrain was full of streams, rivers, and lakes; which developed from the massive glaciation. To counteract this topography, the Indian invented the birch-bark canoe (the western Indians, having no birch trees, used pine bark and invented the Kootanay). The canoe was quickly adapted by the voyaguers and modified to suit their needs. As with the Indian, all of the materials needed for its construction came -4- from the forest. Planking was of cedar and covered with birch bark, tied with split roots and pitched with gum from pine trees. In a short period of time, the voyaguers surpassed the Indian in canoe making, becoming the undisputed leaders in river travel. TABLE I PELTS TAKEN IN 1798 BY NORTHWEST C0. Beaver l06,000 Deer l,950 Marten 32,000 Mink l,800 Muskrat l7,000 Fisher l,650 Lynx 6,000 Elk 700 Fox 5,500 Wolverine 600 Otter 4,600 Buffalo Robes 500 Wolf 3,800 Raccoon 100 Bear 2,lOO (Davidson, 1967, PC 2 70) Three types of canoes were used by the fur-traders. The largest was the Montreal canoe, 35—40 feet long and about six feet wide. This canoe Was used on the larger rivers as well as on the Great Lakes. It weighed about 300 pounds and could carry a load of five tons. The Montreal canoe carried a crew of ten men. The north canoe was about 25 feet long and carried three tons of cargo. This canoe was used in smaller streams, par- ticularly beyond Grand Portage, in the Northwest territories. The North canoe usually had a crew of four men. In between these two canoes was the Bastard canoe, with a crew of eight men. A 20-feet canoe, called a light -5- canoe, was sometimes used, but only by messengers of solitary traders (Nute, l93l). A licensee would hire numerous voyaguers and travel in Montreal canoes from Montreal to one of the major fur areas. The usual route was up the Ottawa River to Lake Nipissing, and then down the French River to Georgian Bay and Lake Huron. The lake was crossed hugging the coastline. Due to the very fragile nature of thecanoe, a storm on the lake would break up the canoe. Hence, a landing site had to be nearby at all times. From Lake Huron, the route would go down Lake Huron, North to Lake Super- ior, or West to Lake Michigan. An alternative route from Montreal was alone Lake Ontario, portaging Niagara Falls, and down Lake Erie. This route, however, was not as well used as the first (Map 1). Montreal with its outpost of Lachine, at the head of the rapids was a natural focus of water routes over which trade was circulated. Mackinac was some 700 miles away, a journey of a month or six weeks for experienced voyaguers. The fur-trade moved further and further west, and as it did the time it took to travel from Montreal to the fur-trading areas and to re— turn increased. At first, the traders would "winter" at the trading areas. They built a post (often just for one winters use) and traded all winter, returning in the spring. This, however, demanded a large supply of stores to be transported to the winter post. To overcome these difficulties, a number of licensees organized themselves into companies. These companies established a series of permanent posts through the Great Lakes region. The posts served two functions. First, they were collection points where the Indians could bring their furs to trade. Secondly, they were supply posts for the voyaguers. One group of voyaguers transported goods between b.01hwo 9 V." u.,' .1“, JV «5.3.237. o «\— imEzo-z gimme-r a I a w.¢(2 .whw h43_m 55(th IV wwhDOm ..m>_o w0(2.<¢0 mum; 3m cflumcmo on... *0 m.>m3-w.m>> £32 Law‘— o’u_m unuho ... m - -12- whether it be fighting, canoeing, or dog sledding. The Coureurs De Bois would brag about their accomplishments and ridicule those who fared poor- ly. This spirit of competition reached a climax between the Athabaskan voyaguers and those of the Great Lakes. The two groups would have mara- thon competitions, such as 48 hour canoe races, that is, 48 hours of paddling a loaded canoe. The voyaguer is best described in an old voy- aguers comment: Said one of these men, long past seventy years of age: 'I could carry, paddle, walk, and sing with any man I ever saw. I have been 24 years a canoe man, and 4l years in service; no portage was ever too long for me. Fifty songs could I sing. I have saved the lives of ten voyaguers. Have had twelve wives and six running dogs. I spent all my money in pleasure, were I young again, I should spend my life the same way over. There is no life so happy as a voyaguers life' (Nute, 1938). Looking at an average trip, from Montreal to Grand Portage, one can see the voyaguers life as it was. He was paid one-third of his pay in advance, and was given equipment as well. The equipment consisted of a blanket, a shirt, one pair of trousers, two hankerchiefs, and some tobacco. If he were a winterer he received about twice that amount. In the bottom of the canoe four "grand perch" poles were placed to protect the delicate canoe from the cargo. The cargo was loaded in sacks, each weighing exactly 90 pounds. In addition, each canoe was equipped with a sail and mast, and a shod oak pole for each man for poling the canoe up rapids. The crew consisted of a guide, a steersman, and eight paddlers. The guide was paid as much as four ordinary men and the steersman about half that amount. Each man would provide his own paddle, and barring the normal dissertions the trip would get underway (Daniells, I969). The men would paddle at a rapid rate, singing and paddling in mili- tary cadence, and would rest one pipe length (5-l0 minutes) per hour. The -13- guide would lead the singing, and was often hired with as mach respect to his voice as to his ability in a canoe. The song would vary as to the job being done. Allouette, with a four beat rhythm, was a poleing song, as poleing was a four movement job. When they came to a portage, the voyaguers used a leather harness (called a portage collar), and would carry 3-5 packs a short distance, put it down and return for another load. Most likely, it was advisable not to get to spread out in Indian country (Nute, 1938). The voyaguer not only traveled in the canoe, but at night lived under it. The canoe was probably his only roof. His meals, usually two daily, consisted of a quart of lyed corn or peas and an ounce or two of pork, grease, or bacon per day (Nute, I938). This was cooked in a common kettle and eaten from the kettle with a wooden spoon. This was, of course, agumented with what ever could be gathered in the wild. The voyaguers would rather paddle than portage, but at Kewenaw Peninsula, a choice had to be made. Here, a well worn portage trail shows an excep- tion to their general preference. When the voyaguers reached the wintering station, they immediately set to built a station. Log houses and a stockade were erected. During the winter, the voyaguers traded with the Indians, using dog sleds for transportation. Then, in the spring, they would take their furs to Mon- treal. Unlike the Great Lakes voyaguers, the Athabaskan group worked out of Grand Portage, not Montreal. From l7l3-l763 fur-trade from Montreal beyond the St. Lawrence drainage basin, and the Hudson's Bay drainage basin on the Saskatchewan increased. Competition from the British Hudson Bay Company became more -14- effective, costs of transportation increased, and profits declined. This, plus competition from the British in the south, were factors in the down- fall of the French regime in Canada (Innis, l956). Following the French and Indian War, the British government continued the fur-trade as the French had done, with the exception that a security deposit had to be given when a license was obtained. Commandants were placed in important posts and the traders had to follow their orders or lose their security deposits. These factors put ridged controls over the fur-trade. About this time there were several companies trading in the Great Lakes region. The American Fur Company, the Xy Company and the Northwest Company were the biggest. A number of independent traders form the Michi- limackinac Company, but it never gained much prominance. The Northwest Company dominated the trade in the Great Lakes. This Scottish firm, who's traders were known as Nor'westers or those master peddlers, was started in 1783. Their chief competition came from the Hudson Bay Company, as they moved into the Hudson's Bay drainage basin. In I784, five million dollars worth of furs were taken from the Great Lakes area. It cost 700 per canoe-full of goods for transportation from England to the Indian country. About one-third of the canoes were for the northwest, and two- thirds for the Green Bay, Lake Huron, and Lake Michigan areas. Transporta- tion from Montreal via the Ottawa River to Michilimackinac amounted to $6.40 per canoe. The Northwest Company built a chain of posts in the northwest and some in the United States, but the latter were abandon due to American pressure. Although the company was successful, it finally merged with the Hudson Bay Company in l82l. This merger closed about 50% of the northwest posts, and saved both companies a considerable amount of money. -15- The Honorable Company (Hudson's Bay Co.) was started in 1670 by the King of England. A very large grant of land was given to them: The whole trade of all those seas, streights and bays, rivers, lakes, creeks, and sounds. . . within the str- eights, commonly called Hudson's Streights together with all the lakes, countries, and territories upon the coasts and confines of the seas, streights, bays, lakes, rivers, creeks, and sounds aforesaid (Mackay, 1966, p. 75). Or, all of the Hudson Bay drainage basin. The company was content to set up posts (factories) on the coast and allow the Indians to come to them. In the late 18th Century, however, the company moved to compete with the Northwest Company (and others). After the great merger, the Honor- able Company lived up to its reputation, and resettled many voyaguers and half-breeds who were put out of work by the closing of posts. These people were placed on farm land, given tools and seeds, and some money. This helped open the Saskatchewan farming areas. The Hudson Bay Company is still doing a thriving business today. The Xy Company was started by Alexander MacKenzie in about 1798. It was known as Sir Alexander MacKenzie and'Company or the New North- west Company, but the Xy Company name stuck. The Xy Co. and the old Northwest Company were in bitter rivalry. McTavish, the President, re- organized the old Northwest Company in an attempt to compete with the new company. The Xy Company was quite successful, building posts on the Pease River, River Qu'Appelle, Red River, Salt River, and others. After the death of McTavish, the two companies united in 1802. John Jacob Astor started the American Fur Company in 1808, with hopes of controlling the North American fur-trade. Astor was no stranger to the Montreal fur-trade, having visited it almost annually since 1788. -15- Using his political "pull", Astor received consent to move into British occupied American territory. He made offers to the Northwest Company, becoming a partner in the east but a competitor in the west. He formed the Southwest Company for trading in the Columbia River region. Unlike the Northwest and Hudson Bay Companies, the American Fur Company was a monopoly. Clerks could not work their way up to being shareholders, neither were any winterers voting members of the company. The AMF Co. was enhanced by the building of national road systems and the Erie Canal in the east, but when the company spread to the Northwest and the American Rockies, much Canadian competition was eliminated. This company was in operation until the fall of the fur-trade in the Americas. It has been stated that after 1834 it was no longer profitable to trap for beaver (Saum, 1965). Assuming that this statement is true, it should have put the fur companies out of business. Fortunately for the Hudson Bay Company, it had other options to stay in business, such as selling land. The immediate effect of the decline of the fur-trade was to close down the Ottawa River transportation. During the early to mid 1800's lake travel came into its own. The Erie Canal increased American traffic on the lakes, and Montreal's need for river and lake transport declined. Hence, a new pattern of travel developed from Lake Erie to Michigan, Illinois, Ohio, and Wisconsin. For a long time Lake Superior was relatively unimportant, until locks were enlarged to move ships up the rapids to the lake. The voyaguers continued to work for Hudson's Bay Company, but their numbers necessarily declined. Many turned to farming in the present day Alberta and Saskatchewan areas of Canada. There, stories of open rebellion against the British are told of later generations. Still others started -17- small communities around the Great Lakes, some farming and others fishing. The voyaguers not only opened the region but settled it as well. Even today, voyaguers can be found around Hudson's Bay. Men still transport supplies by canoe, trap fur animals, and prospect for minerals. These men, Eskimo, Indian, White, or combinations of the three, may not be actual descendents of the original Coureurs De Bois but they are as much voyaguers as any of the past. The fur-trade cannot be discussed without mentioning the contribution of the Indians. At the time of the arrival of the white man, the Indian was self-sufficient, living off the land. The Indians eagerly accepted the white man's trade goods (knives, kettles, and guns) as a means to ease his difficult life. But for a long time the Indian still remained self-sufficient. The trader looked at the Indians, not as a threat, but as a valuable resource of the land; the Indians would work and trap for the fur-traders, as well as teach them the skills necessary for survival in the wilderness (Morton, 1937). The coureurs de bois were excellent pupils of the Indians. They learned how to make the birch-bark canoe, and in time, the voyaguers modified the Indian canoe to fit their needs. The Indians taught them to pole, paddle, portage, and haul the canoe over rapids. To survive in the woods, the coureurs de bois had to learn woodcraft, how and where to fish, hunt, and wilderness travel. This the white men learned, modi- fying the training and in some respects, surpassing the Indians; but they never became quite as well adapted to the wild as the Indians. The fur-traders hired great numbers of Indians. Not only for fur- trading but for hunting and transporting furs as well. Many Indian women were employed to prepare the furs, make snowshoes, moccasins, and other items necessary to the fur-trader. When the voyaguers moved up into -l8- Michilimackinac and beyond, it was Indian corn which fed them. The voyaguers adopted the Indian diet of corn and grease for subsistence. In the northwestern areas, the Indians supplied Pemmican (pounded dried meat mixed with fat) for the traders to eat. It is doubtful that the traders would ever have gotten as far as they did, were it not for their instructors (Fridley, 1967). The voyaguers did not look down on the Indians although later settlers often did, but had great respect for them. Many voyaguers married Indian women, producing a crop of half-breed children. These children formed a middle class, above the Indians and below the whites, and respected by both. Often educated by their fathers, they became very important to the fur-trade in the late 18th and early 19th Cen- turies. "In the Canadian furatrade . . . the only good Indian was not a dead one; he was on the contrary, a live one who would follow his trap line“ (Fridley: Morton, 1967, p. 12). The Northwest Company tended to treat the Indian more as an equal than did the Hudson's Bay Company. The latter would keep seg- regated from them except to trade. The idea that the English "scorned and neglected“ the Indians while the French "embraced and cherished him" is far from true (Saum, 1965). The English ran both Great Lakes and Hudson's Bay trading areas. The difference seems to be the company's policy. The Nor'westers utilized and learned from the Indians while the Hudson's Bay Company only mixed with them in the 19th Century. The effects of the fur-trade are both numerous and varied. The trade occurred in earnest for almost 200 years, and many things happened during that period. Several trading posts and forts developed into towns and finally cities: Detroit, Chicago, Buffalo, Green Bay, etc. . With -19- the increased shipping on the lake, farmers were attracted into the area, as they could now be supplied by the lake traffic. The wars, which occurred during this era, placed a naval fleet on the lakes. This was the time when the Great Lakes were growing up. Of greater consequence was the develOpment of western and northern Canada. All of the fur companies built trading posts throughout the area. In 1802, the Northwest Company had 117 posts (Davidson, 1916), and the Hudson's Bay Company in 1821 had 76 others (McKay, 1938). Men employed by the companies and major stockholders explored and settled areas of western Canada, men such as Selkirk and MacKenzie. All of this develop- ment which we take for granted, occurred only because of the fur trade. The effects of the fur-trade can best be summarized in three ways: 1. The commercial activity between England and France, and their eventual war. 2. The exploration of the new world (North America) by men such as MacKenzie, La Varenier, etc. 3. The beginnings of lake transportation and settlements along the lake; initially, by forts and posts, and finally by towns and cities. Without the fur-trade, it is difficult to tell how the lake region would have developed. Certainly it would have, but it probably would have taken many years longer without the early explorations and navigation. Vandi- veer's quote was accurate, "the story of the fur-trade is very largely the story of pioneer America." BIBLIOGRAPHY 8012, J. Arnold, 1963. Portage Into the Past. University of Minnesota Press, Minneapolis. Brown, Ralph H., 1948. Historical Geography of the U.S. Harcourt, Brace and Co., New York. Cuthbertson, George A., 1931. Fresh Water. MacMillan Co., New York. Daniells, Roy, 1969. Alexander MacKenzie and the North West. Barnes and Noble. Davidson, Gordon C., 1967. The Northwest Compapy. Russell and Russell, New York. Davis, Charles, 1964. Readings in the Gepgraphy_of Michigan. Ann Arbor Publishers, Ann AFbor. Fridley, Russell W. and June Holmquost, 1966. Aspects of the Fur Trade. Minnesota Historical Society, St. Paul. Hamil, Fred C., 1951. When Beaver Was King. Wayne University Press. Honic, Donald, 1967. Frontiers of Fortune: The Fur Trade. World's Work, Ltd., Kingswood, London. Innis, Harold A., 1956. The Fur-Trade in Canada. University of Toronto Press, Toronto, Canada: MacKay, Douglas, 1966. The Honorable Company. McClellan and Stewart Ltd., Montreal. Morse, Eric, 1969. Fur-Trade Canoe Routes of Canada/Then and Now. National Historic Parks’Branch,TIndian Affairs and—Northern Development. Morton, Arthur S., 1937. Under Western Skies. Thomas Nelson and Sons, Ltd., Toronto. Nute, Grace L., 1931. The Voyaguer. O. Appleton and Co., New York. Phillips, Paul C., 1961. The Fur Trade. University of Oklahoma. Quaife, M. M. and Sidney Glazer, 1948. Michigan. Prentice-Hall, Inc., New York. Robinson, H. M., 1879. The Great Fur Land. G. P. Putnam's Sons, New York. 20 -21- Sauer, Carl 0., 1971. Sixteenth Century_North America. University of California Pressj'Berkeley. Saum, Lewis 0., 1965. The Fur-Trader and the Indian. University of Washington Press. Skinner, Constance, 1933. Beaver, Kings, and Cabins. MacMillan Co., New York. Vandiveer, Clarence, 1929. The Fur Trade and Early Western Exploration. Clark Company, Cleveland. ‘ CONCEPTS 0F DEGLACIATION IN SOUTHERN NEW ENGLAND By Mark C. Sullivan A RESEARCH PAPER Submitted to Michigan State University in partial fulfillment of the requirements for the degree MASTER OF ARTS Department of Geography 1972 / , ‘7 , i .1" Approved Gag/k 454,212., TABLE OF CONTENTS Page INTRODUCTION ......................... 1 THE NEW ENGLAND PROBLEM ................... l PREVIOUS WORK ........................ 4 ORDERLY ICE MARGIN RETREAT .................. 7 LACUSTRINE DEPOSITION IN GLACIAL LAKES ............ 12 STAGNATION .......................... 15 SUMMARY ........................... 18 CONCLUSION .......................... 23 BIBLIOGRAPHY ......................... 25 Figure LIST OF ILLUSTRATIONS Map of Moraines in New England (Goldthwait) Map of Recessional Ice Border in Massachusetts (Taylor) Map of the Position of lee Edge During Retreat (Antevs) Map of Glacial Lakes in the Connecticut River Valley (Sullivan) Diagram of Terrace Formation (Flint) Diagram of Kame Formation from Active Glaciers (Lougee) Diagram of the Formation of Terraces with Er- ratics (Brown) Diagram of Stagnation Features (Rich) Page ll 19 20 INTRODUCTION During the Wisconsinan stage of glaciation, New England was covered by a continental ice mass. At the close of this glacial stage processes of post-glacial erosion began to modify the land- scape. When geomorphologists studied the resultant landscape, they recognized certain problems. One of the most significant of these was that the surface forms of New England were in some ways dissimi- lar to the glacial landforms of some classic areas in the central United States and Europe. The differences were believed to have been related to the deglaciation process, and various theories were de- vised to explain how the ice disappeared and to account for the nature of the topography. The purpose of this paper is to explain, compare, and evaluate several of the more important theories of de- glaciation, and to present a review and summary of several of the theories on deglaciation in southern New England, particularly along the Connecticut River. THE NEW ENGLAND PROBLEM New England was glaciated by ice moving south from the Lab- radorian source area in Canada. Much of New England, particularly southern New England, has significant topographic features which may have served as obstructions to this ice. To the north and northeast are the various sections of the White Mountains. In the north-northeast are the Green Mountains of Vermont. Extending south between parts of these topographic highs is the Connecticut River Valley. The Connec- ticut River extends 300 miles from the Canadian border to Long Island Sound, creating a fairly straight trough as its river valley. North- south-trending basalt ridges are located at various places within the valley in central Connecticut and Massachusetts. The ridges may re— strict east-west movement of the river, as well as affecting the loca- tion of smaller tributary streams. It is necessary to know the methods of glacial deposition when viewing glaciation. In an active glacier, the ice is generally moving from the point of origin to the terminus. The terminal zone, however, may be melted. The rate of melting may be increased along the glacial margin, away from the source area. If melting equals advance, the mar- gin temporarily remains fixed. The type of ice margin movement in part determines the type of deposition. There are no terminal moraines recognized by Goldthwait within the Connecticut River Valley (Map 1). The moraines generally recog- nized are located along the coast at Long Island Sound, and at Middle- town, Connecticut. Recessional moraines have also been identified by Taylor (1903) in western Massachusetts and by Lougee (1940) in the White Mountains, but these findings are open to debate. It has generally been concluded that a sequence of terminal and recessional moraines is lack- ing in the area extending north from southern Connecticut to Canada. The existence of terraces along the Connecticut River presents another problem in interpreting deglaciation. The terraces are found at low, medium, and high levels and do not always match up on both I I I I Vt N.H. I I I \ l l 1 Hudson Rim, / (ow ° 5 o - MP8 MA '3 .1 .5 I ‘ I.“ ‘ UWUIO“ \_/ \ Location Of Moraines In New England an moraine z—P N, sides of the river. The composition is mostly stratified sand and gravel and may include varved clays, erratic boulders, and till len- ses. Many of the terraces are marked with deltaic forset bedding, indicating formation in a relatively calm body of water. Associated with the terraces are ice contact features, the two most common being kettles and slump structures. Kettles are formed by a block (or blocks) of ice buried by alluvial deposits. When the buried ice melts, a depression or kettle is formed through collapse of the over- lying and adjacent sediments. Slump structures associated with the terraces indicate the terraces were supported, at least in part, by glacial ice. As the ice melted and the ice margin withdrew, the structure slumped (Figure 1). The presence of ice contact features, as well as other features, indicates the terraces were formed during deglaciation. Three major theories have been devised to explain the nature of glacial landforms and sediments in New England. These involve the following hypotheses: l. The hypothesis that the ice margin retreated in an orderly manner. 2. The hypothesis that lacustrine sediments were deposited in a series of glacial lakes existing in the river val— leys. 3. The hypothesis that the glacier stagnated on a local and/ or regional scale. PREVIOUS WORK The earliest study of the area was done in the mid-1800's. Smith (1832) wrote a physical description of the Connecticut River Valley and mentioned terraces along the river. The terraces, found in Connecticut O ‘0‘ b ‘0 0689‘ w o 9‘ k-j. ‘L c» .Q :i‘ . “.r'f'-",- , ILE o O. . 0 GO .0 o ‘0 TERRACE FORMATION AND THE ORIGIN ,3 OF VARVEO CLAY AND SAND RELATION ‘9 «so: ~~ . and Massachusetts, form three or four levels with surface heights of fifteen to two hundred feet above the river. In addition, he men- tioned floodplains and other features in the river valley. Smith gave no genetic interpretation of the features. Dana (1882) was one of the early glaciologists to write about this area. He claimed that the terraces formed from deposits that were left by the glacier and eroded by streams formed from the melt- ing ice. Shaler (1889) stated there were several ice advances with deposi- tion of stratified sand deposits, till, boulders, and various associated landforms. Some of the problems he recognized were: (1) moraines were not continuous and it was difficult to connect them: and (2) deposition, at least in part, was below sea level. The latter was thought to be associated with localized earth tilting due to the great weight of the ice mass. In the geology of Fishers Island, Wells (1890) found the possible termination of the Wisconsinan ice advance. The maximum position of the ice was partially determined by heavy surf breaking against the glacier. The deposits formed a breakwater, and it is represented by the island. Gulliver (1899), in reviewing Dana's work, stated that since eskers exist in the Connecticut River Valley at a lower level than the terraces, major flooding isn't possible. A major flood would wash away these delicate structures. Gulliver considered the terraces to be del- taic in origin. They were formed when the river valley was under water. The water may have been sea water, indicating tilting. It may also have freshwater dammed up by a blocking of the valley at New London by ice, rock or both. ORDERLY ICE MARGIN RETREAT The first of the three models postulates orderly ice-marginal retreat to explain the topography and sediments of the area. Taylor (1903) explained the apparent lack of recessional moraines by stating that in fact they existed, but had not been recognized by previous workers. He reconstructed a series of recessional borders in the Berkshire Mountains in northwestern Massachusetts. He recognized several ice border phenomena, including moraines, kames, eskers, and eroded river channels. Taylor, however, admitted that his evidence was sketchy. The remains of the ice borders in the Berkshires are mostly so faint and so fragmentary, and the fragments are so scattered about the valleys and on the hill- sides, with so little appearance of order or arrange- ment that a map showing those features alone is unin- telligible (Map 2) (Taylor, 1903, p. 345). Taylor further claimed that the fragmentary nature of the land- forms was attributed to the rugged topography, and that on flat terrain recessional moraines would be fully developed. Taylor's theory seems applicable for the mountainous regions, but there are no known moraines north of Middletown, Connecticut within the Connecticut River Valley. In 1922, Ernst Antevs described a study done on the Connecticut River Valley using varves to determine the retreat of the glacier. Varves are clay and silt layers that may be deposited in ice-marginal lakes. Each year when the ice melts, the clays are laid down, with the coarser being deposited first and the finer silts settling out of the lakes last. Thus the sediments of each year become clearly separa- ted from those of the preceding year (Antevs, 1911, p. 3). One lake bed can be correlated with another, by graphing the thickness of each varve layer in an old lake bed. Thus, the filling of each lake is 9 \ (l/ O ‘ I ‘.3__JL__-_:--- i Ir MASS. / j a . I .. I l I .° fl & B FLY. I § / 1 4' ° I ' J I " I K 00' l-‘k 1% ’MA I g a . . I ' I To: 0 fl . ”3%! , 6 "0 T 2 0y I ' , N‘ I - a ’I ’ ‘féfieflgo Q ' "' .’ v' 7944i," f O f "/ gi} Q‘ 5' '0 fl . , ‘ .p M), a"? Q ' ’4’ '1 .0..." 3 / . h..." I ’ .0 lllfllasl.“~‘ I '°. uncut A0 I . «r E‘. manual ’0’, ‘ : \‘ ‘I 'o y y u e o .w- I mgm ' Q j, ‘q%b& ' 0 Noon": . 0 : ‘Q g o ' .gl‘AINAO! % W ~ ‘5" . ‘ ' c 0' I x. :; a T a nun AT Ice I 0 R 0b - 1 ‘ . &rlont o ' V '0 a .\ 0 §/ . " L ‘ a . MASS I"D""‘-“‘ T‘.“’ ‘1?" dl Qov/ ”Q” C NN i D "' a' v0 s A 0.. MOP2 ’ | . Recessional Ice Borders In Berks hire, Mass. ~Tayler(l 903) H determined relative to another. From this information, the retreat of the glacial margin can be inferred, by following a series of filled lakes along the path of glacial retreat. Antevs concluded that orderly ice-margin retreat took place in the Connecticut River Valley (Map 3). He did not, however, interpret the genesis of land- forms in the area. In 1940, Lougee claimed that orderly ice margin retreat was capable of producing the existing landforms. To demonstrate this, he pointed out that active mountain glaciers produce the same land- forms in small scale that exist in New England. Lougee stated that kame terraces (the river terraces) are the aggraded beds of ice margin streams. He further concluded that seldom are kame terraces at equal elevations on opposite sides of a valley except by chance, unless they are associated with a single delta deposit. Deltas form in front of a retreating ice front either associated with lakes or with alluvial fans (Figure 2). These, then, may be dissected by a glacial stream when the ice front retreats to a position farther up the val- ley. Lougee claimed that the processes by which many glacial features were formed are poorly understood. Ice-contact features can be pro- duced by active ice. Ice blocks may break off the glacier's nose and be rapidly buried by the glacial debris. As the ice margin retreats, the rate of deposition on top of the ice blocks decreases. By the time the ice block melts and the overlying debris slumps into the cavity, there may not be enough debris being deposited to fill in the depression. Therefore, the kettle is created downstream from the IO ‘Maine — 00 ‘ I I \ I 41 ~ -.—- I ‘ NH ‘ 6%? . k,’ / I r’ a ‘ I’ I I”’ l‘ I "' “; /’/ . Vt —°°° , j" J I T'", 3 N I ’- ~ 5300’ 2 ’/ _""_"_- t -_ L———.____-_‘______‘3 I "“ :- ’ ::- Mass I -— ooo / :3:- ~ / =:__-E P /_ _ _ _ _ ""1503 _ _ __ a? I - _, — I \ _.-— I c I __.. R.l \\ E I \ °_- - l a Conn. 3 I . I I I n I I Position Of Retreating Ice-Margin For 100 Year Intervals (Anton . 1022) I‘ll Figure 2 KANE TERRACE FORMATION FROM ACTIVE GLACIERS-lOUGEEIIU4II Iii 12 glacial terminus. Lougee further stated that the lack of moraines is possible with orderly ice-marginal retreat. Moraines would only be produced if ablation was equal to glacial advance. A constantly re- treating glacial margin would deposit no moraines. In conclusion, however, the concept of orderly ice-margin re- treat has not been widely accepted to explain deglaciation in New England. The major reasons for this were the complexity of the con- cept when applied to a regional area and the interpretations made by Lougee. His [Lougee's] interpretation of stratified drift permitted him to identify extensive marine or lac- ustrine water planes (e.g., Lougee and Vander Pyl, 1951), where others find only the deposits of glacial streams or small lakes. This led him to a complicated chronology of uplift and steep tilting (Lougee, 1953) and to other conclusions that failed to win wide acceptance (Wright, 1965, p. 120). In addition, Lougee's attitudes and personality were quite forceful, whereas the cooler logic of Flint tended to command more respect and confidence. Antevs' varve study was stated as inconclusive as a measure of orderly ice-margin retreat by Flint. Finally, other "evi- dence" of retreat is quite speculative (e.g., Taylor's work). LACUSTRINE DEPOSITION IN GLACIAL LAKES Late in the 1800's, it was suggested that a series of four glacial lakes were formed in the Connecticut River Valley. These lakes extended from Middletown, Connecticut to Hanover, New Hampshire. The lakes were supposedly created by the formation of a morainal dam at Middletown, and it extended across the valley between the basaltic cliffs. The first three lakes, located northward from the dam to Greenfield, Mass- l3 achusetts, were collectively (Springfield, Hadley and Montague) called Glacial Lake Hitchcock. The fourth lake is Glacial Lake Upham (Map 4). Emerson (1898) was one of the early proponents of the lake hypothesis. He claimed that the glacier retreated from the Middle- town dam and deposited a large quantity of glacio-fluvial debris. The debris nearly filled a chain of lakes, but as the rate of melt- ing of the ice increased, a river eroded much of the debris. The 1 of the river, which remaining deposits are the terraces or benches were dissected and notched by tributary streams. This interpretation explains the apparent lack of connection between the terraces. The benches have no relation to the present river. At most, the river has undermined them, causing cave-ins, which creates an escarpment of bedrock. According to Flint, the Middletown dam was eroded several hun- dred years after the formation of the glacial lakes. The erosion was such that the lake was lowered in a series of successively lower levels. The final lowering occurred at about 10,650 i 320 BP.2 During the lowerings, lake currents caused finer sediments to be distributed over the bottom of the lake, with the exception of the 1Since terraces often infer ordinary river erosional terraces, Emerson used the term "benches" to indicate depositional terraces. 2The destruction of the glacial lakes is given at a minimum of 60 years and a maximum of 720 years due to the range of radio-carbon dating. This evidence was obtained from dating a log buried in the alluvial fan produced during the breaking of the dam. In addition, 320 feet of varve layers have been found, indicating a long lake exis- tence (Flint, 1953). T4 Glacial lakas ln‘lha Connecticut liver VaIIay --- Glacial [aka N7 15 very center. Here the current was the strongest, preventing accumula- tion of the fine sedimentary material. In this interpretation, these sediments formed the middle depositional terraces. Emerson further stated that the lowest terraces were the result of the present-day Connecticut River and represent its floodplain. To explain the ice contact features, Emerson stated that ice was present in the valley during its construction. This was believed to be mostly floating ice, grounded on shallow terraces and buried. Burke (1956) added that eskers, crevasse fillings, kames, and kame terraces were all formed by flowing water originating from the glacier at various times during ice disintegration. The reason this lacustrine concept never gained much popularity is not completely clear. Flint (1932) voiced disapproval due to the lack of evidence of wave erosion. He stated that in the Finger Lakes it is abundant, but not in New England. Another problem the concept faces is the lack of similar features in present-day active glaciers. While the lakes associated with glaciers may be present, terraces are not as evident as would be expected if this concept were accurate. Finally, it was overshadowed by the stagnation concept, which has had great acceptance. STAGNATION The concept of stagnation maintains that the ice melted in place. The scale of stagnation varied with the different interpreta- tions of the various individuals working on the problem. These ranged from regional to local stagnation (valley tongues or large ice blocks). 16 This concept, at its inception, was seriously doubted, due to its radical departure from classic deglaciation theory. It did, however, appear to explain some of the questions of the surface features of the Connecticut River Valley. One of the first proponents of this concept in New England was Clapp (1904), who adopted an idea of regional stagnation. Clapp concluded that at some time the continental ice mass stopped its forward motion, and both marginal and surface melting occurred. Since all of the land surface was covered with ice, melting was roughly similar in similar environments. In time, however, the tops of mountains (nunataks) protruded from the ice mass and heated faster, causing the ice to melt faster along the margins of the land. This is because of the radiation of heat from the land and the concentra- tion of drainage streams along the contact between land and ice. The ice continued to melt at the greatest rate adjacent to the ex- posed land masses. Eventually the glacier was reduced to a remnant of its former configuration and remained last in the deeper valleys of the area. This differential melting sometimes resulted in the lower areas being covered with debris originating from the protruding land masses and deposited by superglacial streams. The debris pro- duced an insulating blanket on the ice, reducing the rate of melting of the lowered area. The ice, therefore, melted almost simultaneously from all portions of the area. Flint (1929), a leading proponent of the stagnation concept, stated that the Labradorian ice sheet thinned through surface abla- tion, until it lacked the thickness to continue its southerly movement 17 over the northern barrier of mountains. There remained several thou- sand feet of ice in parts of southern New England which became cut off from a supply of ice to the north. Down melting occurred as Clapp (1904) had suggested, to the point where only remnants of the glacier existed. According to Flint, streams and narrow lakes developed between the ice and emerging land, which resulted in the deposition of fluvial and lacustrine sediments.. These streams flowed south. being confined by the ice mass, often at elevations of up to several hundred feet above the present river elevation. As down- wasting continued, the ice margin melted toward the center of the valley. The streams it confined moved to a lower level, and the fluvial-lacustrine deposits deposited against the ice collapsed and were left high above the new stream. Sometimes the successive ice margin position would not support a lake, but instead would allow the stream to drain under the ice, becoming a subglacial stream. When this occurred, no sediment would be deposited, and no terrace would be created. This explains why terraces may be absent from a series. If both sides of the valley were connected by an open crevasse, the two adjacent lakes would be at the same level, as would the terraces they formed. This, however, rarely occurred. The lowest terraces were created by the present-day Connecticut River, after deglaciation. This was concluded from the even pairing and the rare occurrence of deltaic bedding (Flint, 1930). In order to explain the deltaic bedding of the middle terraces, Brown (1932) showed that the delta was formed from drainage from the 18 land into the ice-margin lakes. As the ice moved, the delta often slumped. He also recognized the deposition of large erratic boulders and concluded that they originated on adjacent ice margins (Figure 3). Brown's explanation appears particularly feasible in glacial lakes located near large hills or mountains. Rich (1943) considered buried stagnant ice as an important factor in the creation of kame terraces. He reasoned that as the margin retreated, isolated parts of the ice sheet were left within the valleys and were covered with debris. As the buried ice slowly melted, features such as eskers and kame terraces developed (Figure 4). Of the three concepts of deglaciation, stagnation is probably the most widely accepted. The concept explains the lack of moraines, by eliminating the forward movement of the ice which causes them. Since the ice disappears by down melting, glacial debris is fairly evenly distributed over the landscape. The terraces are exceptions and are due to the damming of the ice in both sides of the valley. This concept explains the presence of more features, more easily, and more simply than the other concepts. SUMMARY While all three concepts claim to explain the deglaciation of southern New England, they do not do it with equal success. There are many reasons for this, and to explain them it is necessary to compare the concepts. By examining the arguments which have resulted from these different concepts, one can see the strengths and weaknesses of each. 19 - ....... sue rum lCE , STAGNANT ICE v" I .o Figure 3 FORMATION OF TERRACES WITH ERRATlCS-BROWNlflaathwfi —_ 20 1 E ! LONOITUDINAL .ECTION LIVE ICE MIA D h-v' "' ‘ ' '-‘.' 7 o'o.'rvp"t71 l'sl . ' ‘ .EOROCK CRORR RECTION CR“. .ECTION AFTER .OU'NO .URIAL FINAL 'ELTINO IURIAI. OF ICE IV OUTWAOH ORAVEL AND CREARINO PRODUCING OTAONATION OF THE IURIED ICE LIVE ICE PROTECTION or out!!! MARGIN or THE Tc: IV AlLAi’lou lonAme mourn“) FigureA m 21 The concept of lacustrine deposits in glacial lakes failed to account for several topographic features. Flint (l929) discounted large-scale erosion by glacial lake Hitchcock, due to the lack of wave-cut features along the shore. An additional criticism, leveled by Brown (l930), was that it does not account for extensive deposits of sand and gravel in both longitudinal and transverse ridges near the middle of the valley. These ridges also have ice-contact slopes on all sides. In addition, large and small boulders mixed with fines, small faults, folds, and the absence of sand and gravel from certain lake beds are not easily explained by the concepts. Brown further states that the sand and gravel plains of central Massachusetts which have frequently been referred to as delta plains show the typical internal structure of kame terraces. The concept of orderly ice-margin retreat faced the same type of criticism as did lacustrine deposition. Some of the objections made about the orderly ice margin retreat were (Flint, 1929): l. The lack of recessional moraines and outwash plains with fosse relationships at their heads. 2. Lack of proglacial or postglacial erosion of the glacial deposits in the low as well as the upper reaches of the valleys. These criticisms were countered by Lougee (194l), who stated that re- cession moraines did exist. Here the question of interpretation arises. Lougee gave a number of examples of moraines located in New England. He claimed that criticism of this concept is based largely on misiden- tification of or lack of distinction between such features as deltas built in water and alluvial fans built on land. The lack of erosion 22 is implied by the formation of open lakes, similar to Lake Hitchcock, protecting the glacial features from river erosion. The orderly ice-marginal retreat may explain all of the existing landforms assum- ing certain interpretations are correct. Unfortunately not much work was done along the Connecticut River in relating this concept to the landscape to see if it is indeed appropriate. The concept of stagnation faced a serious problem in the con- clusions reached by Antevs (l922), in his varve count done along the Connecticut River. In l922, Flint explained that the varves were not correctly correlated between exposures. Antevs, according to Flint, misunderstood the succession of terraces. In each terrace, the bottom layer of clay was deposited first; thus it was the oldest. The top layer was deposited last and was thus younger. Antevs concluded that the varves were laid down in one continuous succession from the lowest to the highest. Hence the top layer was younger than the bottom layer, located near the present-day Connecticut River. This was his reason- ing when he traced the succession of varves northward. However, Flint explained that in reality the top terrace was deposited first (Figure l). As the ice melted and the margin withdrew, a second lake formed at a lower level forming the middle terrace. The lowest terrace (not the varve layers in it) is the youngest and the highest the oldest. Flint (1932) further explained that in a single transverse section across a valley, the same bed of clay may locally underlie more than one terrace. Therefore, more than one interpretation is possible with this relationship. In short, Antevs' conclusions were inconclusive. 23 In addition to the problem of varve sequence, it was found that the 4,100-year varve series does not correlate with radiocarbon chronology (Wright, l965, p. 120). Antevs (l962), however, stated that the radio- carbon chronology of the northeastern North America is faulty and that his varve count was accurate. The stagnation concept might also be questioned if the existence of recessional moraines, such as those in the Berkshires, can be demon- strated on an extensive basis. However, here the problem of interpreta- tion arises once more. Taylor (l903) had pieced together from frag- ments a series of recessional moraines. They are, to say the least, sketchy. Goldthwait (1938) had originally claimed to have found several moraines in New Hampshire, but he later revised his conclusions, stating that they were really not moraines. Lougee (l940), however, was still convinced that Goldthwait's original interpretation was correct and that his revised interpretation was wrong. And so the argument continues. CONCLUSION In reviewing the deglaciation concepts of the Connecticut River Valley, one sees a number of complex ideas. All of the con- cepts are plausible, but they cannot all be correct. It is possible that none of them are accurate or that a combination of these concepts is responsible for deglaciation. The main problem for all of the con— cepts seems to be a matter of interpretation. As the various workers would see a landform, their interpretation would be in relation to the concept they believed in. Each worker tried to prove his own point of view. This becomes a particularly difficult problem when only bits and pieces of landforms exist (e.g., moraines). 24 It is this writer's Opinion that the stagnation concept best explains the features in southern New England. Others would disagree, using some of the same arguments stated in this paper. The problem is far from being solved, and more work is needed. The controversy of deglaciation is not restricted to New England; many of the "classic" areas are being reviewed with these concepts in mind. Solving the deglaciation problem in New England would do much to answer questions arising in other parts of the world. BIBLIOGRAPHY Alden, N. C., 1924, "The Physical Features of Central Massachusetts," U. S. G. S. Bulletin, No. 760, pp. 13-105. Antevs, Ernest, 1922, "The Recession of the Last Ice Sheet from New England," Amer. Gegg. 593, Research Series, No. 11. , l962, ”Transatlantic Climate Agreement VS C-14 Dates," Jour. 6e99,, Vol. 70, pp. 194-205. Bloom, A. L. and Minze Stuiver, 1963, "Submergence of the Connecticut Coast," Science, Vol. 139, pp. 332-34. Brown, T. C., 1931, "Kames and Kame Terraces of Central Massachusetts," Geol. Soc. Amer. Bulletin, No. 42, pp. 467-79. , 1932, "Evidence of Stagnation During Deglaciation of the Nashua Valley," Amer. Jour. Sci., Vol. 19 (Ser. 5), pp. 359-67. Burke, M., 1956, "Pleistocene Geology of the Connecticut Valley," Compass, Vol. 33, pp. 317-26. Chapman, D. H., 1937, "Late Glacial and Post Glacial History of the Champlain Valley," Amer. Jour. Sci., Vol. 34 (Ser. 5), pp. 89-124. Clapp, F. G., 1904, ”Relations of Gravel Deposits in the Northern Part of Glacial Lake Charles, Massachusetts," Jour. Geol., Vol. 12, pp. 198-214. , l904(b), "Ice Retreat in Glacial Lake Neponset," Jour. Geol., Vol. 12, pp. 181-97. , 1907, "Complexity of the Glacial Period in North East New England," Geol. Soc. Amer. Bulletin, Vol. 18, pp. 505-56. Currier, L. H., 1941, "Disappearance of the Last Ice Sheet in Massa- chusetts by Stagnation Zone Retreat (abst)," Geol. Soc. Amer. Bulletin, Vol. 52, p. 1895. Daly, R. A., 1936, The Changing World of the Ice Age (New Haven, Conn.: Yale University Press). 25 26 Dana, J. D., 1882, ”The Flood of the Connecticut River from the Melting of the Quaternary Glacier," Amer. Jour. Sci., Vol. 23, pp. 87- 98, 179-202, 360-73; Vol. 24, pp. 98-104. Donner, J. J., 1964, "Pleistocene_Geology ongastern Long Island, New York," Amer. Jour. Sci., Vol. 262, pp. 355-76. Emerson, B. K., 1898, "Geology of Old Hampshire County, Massachusetts,” U: S. G. S. Monograph, No. 29. Flint, R. F., 1929, "The Stagnation and Dissipation of the Last Ice Sheet,” Geog. Review, Vol. 19, pp. 256-89. , 1930(a), ”The Glacial Geology of Connecticut," Connecti- cut State Geol. Nat. Hist. SurveygBulletin, No. 47, 294 pp. , 1930(b), "The Classification of Glacial Deposits," Amer. Jour. Sci., Vol. 19 (Ser. 5), pp. 169-76. , 1931, “Terraces in the Connecticut Valley,‘l Science, Vol. 74, pp. 368-69. , 1932, "Deglaciation of the Connecticut Valley," Amer. Jour. Sci., Vol. 24 (Ser. 5), pp. 152-56. , 1942, ”Glacier Thinning During Deglaciation, Part II; Glacial Thinning Inferred from Geologic Data," Amer. Jour. S513, Vol. 240, pp. 133-36. , 1953, ”Probable Wisconsin Substages and Late Wisconsin Events in Northeastern U.S. and Southeastern Canada,” Geol. Soc. Amer. Bulletin, Vol. 64, pp. 897-919. , 1957, Glacial and Pleistocene Geology(New York: John Wiley and Sons), 553 pp. . 1963, "Altitude, Lithology, and the Fall Zone in Connecti- cut," Jour. Geol., Vol. 71, pp. 683-97, Goldsmith, Richard, 1960, ”A Post Harbor Hill-Charlestown Moraine in Southeastern Connecticut,” Amer. Jour. Sci., Vol. 258, pp. 740-743. Goldthwait, J. W., 1938, ”The Uncovering of New Hampshire by the Last Ice Sheet,” Amer. Jour. Sci., Vol. 36 (Ser. 5), pp. 345-72. Gulliver, F. P., 1899, ”Thames River Terraces in Connecticut (abst)," Geol. Soc. Amer. Bulletin, Vol. 10, pp. 492—95. Jahns, R. H. and Max E. Willard, 1942, "Late Pleistocene and Recent Deposits in the Connecticut Valley, Massachusetts," Amer. Jour. Sci., Vol. 240, pp. 161-71, 265-83. 27 Lougee, R. J., 1940, "Deglaciation of New England,” Jour. Geomorph,, Vol. 3, pp. 189-217. Lougee, R. J. and A. W. Vander Pyl, 1951, "Glacial Water Levels in the Thames-Willimantic River Valleys," Science Monthly, Vol. 73, pp. 275-83. Rich, J. L., 1943, “Buried Stagnant Ice As a Normal Product of a Progressively Retreating Glacier in a Hilly Region," Amer. Jour. Sci., Vol. 241, pp. 95-100. Shaler, G. H., 1889, ”The Geology of Nantucket," U. S. G. S. Bulletin, No. 53, pp. 1—55. Smith, A., 1832, "On the Water Courses and the Alluvial and Rock Formations of the Connecticut River Valley," Amer. Jour. Sci., Vol. 22, pp. 205-31. Stewart, 0. P. and Paul MacClintock, 1964, “The Wisconsin Stratigraphy of Northern Vermont," Amer. Jour. Sci., Vol. 262, pp. 1089-97. Taylor, F. B., 1903, "The Correlation and Reconstruction of Reces- sional Ice Borders in Berkshire County, Massachusetts," Journ. Geol., Vol. 11, pp. 323-64. Wells, 0. A., 1890, "Evidences of Glacial Action in Southeastern Connecticut," Pop. Sci. Monthly, Vol. 37, pp. l96-201. White, S. E., 1947, ”Two Tills and the Development of Glacial Drainage in the Vicinity of Stafford Springs,Connecticut," Amer. Jour. §gi,, Vol. 245, pp. 754-78. Wright and Fry, 1965, The Quaternary of the U.S. (Princeton, New Jersey: Princeton University Press). HICHIGQN STQTE UNIV LIBRQ ll!" "WW (IN! I WI [1 06289585