Attention has been given to sustainable multicrop systems throughout the world such as the "taungya" systems of Indonesia and Central America, and the "forest farming" systems of Australia and New Zealand [87, 88].(1)  Much less attention has been given to describing or developing systems for northern, temperate regions such as the northeastern United States [23].

This paper will investigate some ecological aspects of the agroforestry systems used by northeastern American Indians, as inferred from reviews of the accounts of early Anglo-European explorers and settlers, with particular emphasis on those used on upland soils.  The assumption is made that any research on sustainable agro-ecosystems for a given region should include a careful examination of the systems that evolved over millennia of inhabitation by native peoples.

Since Anglo-European settlement, the agricultural, pastoral, and forestry practices of the Northeast have followed the models built by the settlers' forebears in the similar, but significantly different ecosystems of Great Britain and Western Europe.  These systems have been generally sustainable on level, alluvial soils, but less sustainable on upland soils [9, 22].  Plowing and cultivating on steep upland soils have caused severe soil erosion in some areas; overgrazing has compacted and depleted many upland soils; overcutting of fuelwood and timber have often negatively impacted the species compositions and productivities of forest communities [20].

Early accounts of the pre-settlement landscape are consistent in their descriptions of fertile soils, abundant wildlife, and healthy human populations.  Although reliable information is lacking on the number of people that presettlement agro-ecosystems were capable of supporting, it seems likely that numbers were much higher than generally supposed, especially in the upland areas [42, 71].  The Indians were apparently able to produce high quality food and fuel on lands that now generally produce fuel and fiber of only indifferent quality [23].  The following sections of this paper will show that the Indians' agro-ecosystems were in fact elegantly simple, yet highly productive agroforestry systems which were ideally suited to the ecology of the Northeast.

II. Descriptions of Indian Agroforestry Practices

As suggested in the definition of agroforestry cited above, trees may be integrated with agricultural crops and/or animals either simultaneously or sequentially.  Historical accounts [cited in 11, 24, 71] indicate that the Indians practiced sequential agroforestry systems on level, alluvial soils where they grew annual crops of corn, squash, and beans, as well as perennial crops of berries.  Simultaneous agroforestry systems were practiced on the upland, glacial till soils, where Indians grew perennial crops of berries and nuts, which were harvested by the Indians themselves as well as by the animals they hunted; upland soils also produced crops of fuelwood and other products important to humans and wildlife.

Sequential agroforestry practices, or shifting cultivation, in the Northeast entailed fallow periods for fields during which they would be invaded by grass, herb, shrub, and tree species.  Crops of corn, squash, and beans would be grown for 5-10 years [16, 83].(2) Then the soil would go uncultivated for 10-20 years.  Early accounts described these fallow areas as grassy, with berry bushes and small trees.  If the fields were burned, as was common [52], berries such as strawberry, raspberry, and blackberry would often follow the fires [6, 72, 75, 93].  Perennial grass species, particularly little bluestem, would also follow fires [11, 61, 79].  Nitrogen-fixing species such as wild indigo and lespedezas would likewise increase after fires [25, 53, 59, 79].  After a fire, the perennial grasses resprout vigorously from rhizomes which are protected by the soil above them; the hard coatings of leguminous seeds are scarified by the combination of heat and moisture, which promotes their germination.

Exposed mineral soils--resulting from cultivation or intense fires--would often be colonized by light-seeded tree species such as birches and white pine [4].  Day [24] noted that the Indians' name for alluvial areas (the "intervales" of the early records) meant the "places of little pines." Gordon [33] noted that pre-settlement white pine stands existed on areas that had previously been cleared for agriculture.  When the Indians decided to clear land for agriculture, they would girdle the trees with axe cuts or build fires at the bases of the trees to girdle them.  Corn would be planted among the dead trees, which would later be removed by felling and burning in place [20].  Land that had not grown up to trees could have been cleared by burning the grass and shrubs.  An intense fire in early summer would reduce the incidence of resprouting by shrubs and tree saplings [5, 13, 34].

Simultaneous agroforestry practices in the Northeast involved manipulating the forest environment to shift the species composition in the direction of more food-producing shrubs and trees, and to increase the productivities of these species.  The chief forest management tool used by the Indians was fire.  Bromley [11] and Day [24] cited many of the early accounts describing the presettlement landscape in advancing the theory that large areas of forest land were intentionally burned, chiefly during the months of November and April.  The following quotation from Wood [90, cited by Day] is typical of the early accounts:

. . . for it being the custom of the Indians to burn the wood in November, when the grass is withered, and leaves dried, it consumes all the Underwood, and rubbish, which otherwise would overgrow the country, making it unpassable, and spoil their much affected hunting . . .

The early accounts generally ascribed the purposes of improving travel and hunting to the Indians' burning of the woods.  Some later reviews of these accounts put emphasis on the additional purpose of favoring nut trees and berry plants [11, 12, 24, 81].  The early accounts are, in fact, fairly consistent in their indications of the predominance of nut trees in upland forests; the following citations from Day [24] are typical:

In 1607 Captain Gilbert described the trees at a point on the Maine coast--probably Point Elizabeth south of Casco Bay.  They were "the most part of them ocke and wallnutt (hickory) growing in a great space assoonder on from the other as in our parks in Ingland  and no thickett growing under them" [7].  Richmond Island nearby had "fine oaks and nut trees with cleared land and abundance of vines which in their season bear fine grapes." [19]

Figure 1.  A part of Green Hill Park in Worcester, MA.  This area has been burned every 1-3 years by local youth for decades.  Note the dead saplings in foreground from a fire in spring, 1988.  Overstory is red oak; ground cover is sprouts of red oak and red maple, huckleberry and various ferns and grasses.

The relatively thick barks of oak (especially chestnut, black, scarlet, and red oak), shagbark hickory, and American chestnut were able to withstand the heat of periodic low intensity burns.  The relatively thin-barked maples, birches, beech, and hemlock would have been limited to the less frequently burned, mesic sites [11, 61, 79].  Since the nut trees are also prolific sprouters from stumps, they would have assured their persistence on sites which were cut for fuelwood or burned intensively [43].  The decay resistance of white oak and American chestnut would have increased their resistance to basal wounds which would have inevitably resulted from periodic burning [E. Stone, personal communication].

Historical accounts and archaeological discoveries concur on the relative importance of various foods of the presettlement Indians.  In addition to the staples of corn and beans, Thomas [83] and Russell [71] emphasized the importance of small fruits and tree nuts in the Indian diet.  By floating the contents of refuse pits from an early 17th century Indian settlement in southwestern New Hampshire, Thomas was able to establish their extensive use of berries and nuts.  Furthermore, Thomas cited Wright [91] regarding the record of Indian deeds which reserved certain privileges (italics added--KD):

. . . that they (the Indians) shall have and enjoy all that cottinackeesh, or ground that is now planted; And have liberty to take Fish and Deer, ground nuts, walnuts, akornes and sasachiminesh or a kind of pease . . . (deed for a part of Agawam, MA)

. . . (the Indians) doth reserve liberty of fishing for ye Indians in ye Rivers of waters and free Liberty to hunt deere or other wild creatures and to gather walnuts, chesnuts and other nuts things &c on ye Commons . . . (deed for a part of Deerfield, MA)

Berries were an important source of vitamin C and sugars, not only in season, but throughout the year in dried form mixed with corn meal [16].  Nuts were an important source of protein and oil, particularly shagbark hickory ("walnuts") and butternuts.  Acorns and chestnuts were dried and stored; water mixed with alkaline maple wood ashes would be used to leach the tannic acid from the acorns [71].  Hickory nuts and butternuts would be husked and shelled, or crushed and then boiled to float off an oil emulsion that had a variety of uses [71].  Nut meats or oils would greatly enhance the food value of corn, which by itself is considered a nutritionally poor staple [14].

Thomas's [83] diagram of major Indian dietary constituents (Figure 1) shows a pattern that relied heavily upon field crops, plus nuts, berries, deer, bear, and moose through the fall and winter.  These fall and winter foods would have been amply provided by the sequential and simultaneous agroforestry systems outlined above.  The following two sections will discuss some important ecological aspects of these systems.

Figure 2. Reconstruction of Major Dietary Constituents in the Early Seventeenth Century (Indian) Subsistence Cycle (from Thomas [83]).

III.  Ecological Aspects of Shifting Cultivation

Although fish manures may have been used [18], the only nutrients known to have been added were the nitrogen fixed by beans, and the minerals released from crop residues by burning.  But since the Indians did not practice plowing or cultivating (except very locally in the corn "hills"), the modern problems of soil erosion would have been minimal.  Thus, a soil's fertility could have been restored relatively quickly by natural successional processes.

Carmean et al. [17] reported on the improvement in structure and fertility of a soil which had gone from cultivation (and consequent erosion) to pasture to young forest.  The improvements were attributed to the actions of root penetration and leaf litter accumulation.  Fisher and Stone [29] reported specifically on the increased availability of nitrogen and phosphorous under young conifer plantations:

Despite cumulative net withdrawal into the conifer biomass, soluble forms of both elements were higher near and under the trees.

The mechanism for increasing the available nutrients was not identified although they hypothesized that:

. . . the conifer rhizosphere can mineralize or otherwise extract some fraction of the soil N and P that was resistant to microbial activity under the previous vegetation of grasses and forbs.

Thus, the "little pines" of the intervals agricultural areas of the Indians would have served the purpose of improving soil fertility in two ways: by the unidentified action of the rhizosphere and by the accumulation of nutrients in the above-ground biomass which could be partially returned to the soil by felling and burning.  Young pine stands would also have served as cover for wildlife species such as ruffed grouse, turkey, deer, and bear [15].

Repeated burning of fallow, grassy areas would reduce diseases of the grasses and shrubs, particularly leaf spot fungi [22, 35].  Fire would also stimulate heavier fruiting of these plants [6].  Sharp [72] stated that grouse populations under fire management were 2-1/2 to 5 times higher than under timber-cutting management for the species, and that:

Evidence from various sources poses the hypothesis that the ruffed grouse is a fire climax species or one that benefits from recurrent fires in its habitats.  Its basic or key food plants are fire induced at crucial stages of development such as by stimulating seed germination and seedling establishment, by rejuvenating decadent plants or clones, or by controlling plant diseases.

Figure 3.  Research site at the Connecticut Arboretum in New London, CT.  This area has been burned annually since 1981.  Burning has slowed invasion by tree species.  Species which have increased due to burning are low-bush blueberry, viburnums, arrowhead, huckleberry and sweet fern.

Other birds, particularly the now extinct passenger pigeon and heath hen, would also have benefited from burning fallow Indian fields [8, 72, 841 for the reasons cited above, and because fire would also destroy some parasites and diseases of the birds [8].  Grouse, heath hen, passenger pigeon, turkey, and other fowl would have been important dietary items in early fall and early spring (Figure 1).

Some of the early accounts of Indian burning practices ascribed another purpose in addition to the ones cited in Section II, above--that of eliminating the "vermin" [24].  Specific vermin were not indicated, although one reason that the Indians often moved their villages was apparently to avoid fleas and lice which had come to infest a site [60].  Moist conditions are needed for the build-up of inocula for many plant diseases [35] and for the survival of many insects which live on or near the soil surface.  Burning would destroy inocula on grasses and shrubs, and would temporarily create drier conditions on and near the ground surface.  Some diseases would affect the plants as well as the animals which ate them.  Ergot and other fungal poisons affect animals (and humans) which eat infected grassland plants [35, 55]. 

Burning would also destroy insects and arthropods such as ticks, fleas, chiggers, and lice which live on the ground surface when not parasitizing an animal host.  These insects can be particularly damaging when they act as transmitters of internal parasites or diseases of wildlife and humans.  A recent outbreak of "Lyme disease" in southern Connecticut has been partially attributed to the increase in the population of deer ticks-which carry the disease and infect humans with it (R. Wyatt, personal communication).  Whether this disease existed in pre-settlement times is not known, although the transmitters of the disease existed then and certainly could have carried this or other diseases.

Fire affects organisms living below, as well as above, the soil surface.  Ahlgren [5] reviewed the research on how fire affects soil organisms.  Increased availability of mineral nutrients, which leach from the ash, improve conditions for the nitrogen-fixing bacteria of the Azotobacter and Clostridium species.  The overall pH of the soil increases for the same reason.  Stone [78] reported that these bacteria, as well as blue-green algae, were most likely the organisms responsible for increased levels of available nitrogen in frequently burned soils in the Southeast and in the Coastal Plain. 

These increases occurred in several studies, despite the volatization of much of the nitrogen in the accumulated organic layer.  Actinomycetes also increase following fires [5], but are not known to increase available nitrogen.  Other soil fauna generally decrease after fire (e.g., earthworms, snails, beetles, mites, collembolans, centipedes, and millipedes).  These organisms probably do not significantly affect soil fertility, but they do improve soil porosity and texture.

IV. Ecological Aspects of Controlled Forest Burning

The term "controlled burning" is used advisedly here, considering the lack of specific information in the early accounts about how the Indians actually managed the fires they set, although Martin [52] cited one account of Indians uprooting the grass around the circumference of an area they intended to burn in order to prevent the fire from running back.  Martin stated that "all other early references to Indian burning are unsubstantial" (i.e., unspecific), in making the point that the Indians were not the "irresponsible incendiaries" that some early colonists [20] and some historians [54] claimed them to be.  Indeed, it is likely that in their thousands of years of experience with fire the Indians would have learned as much as (if not more than) we know today about how meteorological, vegetational, and topographical factors interact to control the course of a fire [93].

If the forest fires were set in November or April, as stated in nearly all the early accounts, the soil would have been moist, since these months typically have increased precipitation in the Northeast [69].  Furthermore, during these months very little or no water is being withdrawn from the soil by transpiring vegetation.  Thus, a fire would have consumed only the dry grass and some or most of the dry leaves.  Intense fires which destroy the organic soil layer or reach the tree crowns would not normally have occurred at these times of year. 

Bodies of water such as streams, swamps, lakes, and rivers would have stopped a fire.  Rocky ridges or hilltops, where grass and leaves would have been sparse, would also probably have stopped most fires in November or April.  For these reasons, most fires probably would have been limited by natural factors to relatively small areas between streams and ridges.  If the Indians used methods such as the one described by Martin, above, the areas burned could have been even smaller.

There has been some objection to the general assumption that the Indians' burning in the Northeast was of wide extent [66, 70], although Raup [66] did not consider the evidence of the early accounts cited by Bromley [11] and Day [24], and Russell [70] only attempted to call their validity into question.  Neither writer produced a coherent theory as to how the pre-settlement landscape could have appeared as described without the agency of fire, particularly the extensive occurrence of grasslands and fire-resistant nut trees.  The following parts of this section will indicate ways in which periodic burning of the forest would have benefited the Indians by shifting the species composition towards more food-producing plants and increasing their productivities at the same time.  As will be shown, the benefits would have considerably outweighed the liabilities, thereby giving the Indians significant motivation to periodically burn the forests near their settlements.

The dietary constituents described by Thomas [83] in Figure 1 show many food products of the forest.  In addition to the berries and nuts mentioned in Section II, there were also deer, bear, moose, small mammals, and fowl.  All these animals have in common the fact that they depend on foods which are available on or near the ground level [8]: seed from grasses and herbs (especially fowl), berries and shoots from shrubs, nuts from trees, and sprouts from trees (especially deer and moose).  All of these foods would be more available in a periodically burned forest.  Bendell included the northeastern mammal species of moose, white-tailed deer, black bear, beaver, and hare in his list of fire-followers; on the other hand, marten, squirrel, and fisher were included in his list of species displaced by fire.  Of the birds, Bendell stated that turkey, pheasant, quail, grouse, and some water fowl would follow fire, while smaller birds would be displaced:

Many kinds of wildlife are adapted to living in environments created largely by fire.  Some of the features of flammable forests and grasslands that may affect wildlife are much growth of vegetation on or near the ground; growth of trees and shrubs in open stands with thick, often stubby branches and twigs; large fruits, seeds, and nuts that may be retained on the plant and protected in serotinous cones or thick shells. . .

Under these conditions we would expect birds and mammals that live as browsers and grazers.  There should be a tendency for large size, for this is of advantage where there are large variations in local climate and for other reasons.  Large wildlife are possible where they can feed from the ground and in shrubs and trees with open, strong branches.  Thus, in fire forest there may be deer and spruce grouse.  In relatively fireproof forest, such as hemlocks, there are birds . . . and mammals that are smaller in size.

One is tempted to add another mammal to the list of fire forest species: the human hunter, since he is so closely associated with the other animals of the fire forest.

It would be perfectly natural for a human culture which depended heavily upon hunting and trapping, as did the Indians, to manipulate their environment so as to increase the number of large animals and birds.  Larger targets would be easier to hit and would yield a higher return of food for each unit of energy expended in hunting and trapping.  In this context, Bendell's thesis would indicate that the Indians should have burned their environment.  The Indians, as well as all the other mammals and birds in the fire forest, would thereby have maximized their efficiency of food utilization according to the criteria of Bock and Lynch [10] who found that larger birds (or mammals) are less active than smaller ones and therefore utilize food more efficiently, and that the unit weight of birds (or mammals) is greater on a burned forest site.

One fundamental reason for more efficient and greater food utilization on burned sites is that some of the nutrients which have been tied up in the organic biomass--as live and dead trees, fallen dead branches, and fallen leaves--are released by fire and then become available to other plants which produce fruits and shoots that are eaten by wildlife.  Unburned oak-hickory forests in Tennessee and North Carolina have been shown to have amounts of nitrogen and calcium stored in the forest floor which are equal to or greater than the exchangeable amounts in the mineral soil [80].  Low intensity fires would convert most of these nutrients to exchangeable forms after leaching from ash, and they would also oxidize some of the tannins in the oak and hickory leaves which are known to inhibit nitrification [93].  However, some nutrients could be lost to groundwater or streamwater, particularly if burning occurred when the plants were not actively growing.  From this point of view, burning in April would have been preferable to November.

Over the long term, controlled burning could reduce the thickness of the soil organic layer, thereby exposing the mineral soil to erosive forces, especially where vegetative cover is less and slopes are steep [93].  On the other hand:

Low intensity fires which leave some litter and a large portion of the humus on the soil surface have little or no effect on surface runoff and erosion. . . . The most stable forest sites following logging and burning are on level terrain, have deep, well drained soils, have soils that are high in clay content, are on areas with low precipitation surplus, are seral forest types with a favorable growing season after treatment, and have a site history of periodic fires . . . [93]

Thus, it appears that low intensity fires on level or gently sloping sites in April would have significant benefits without harmful side effects.  November burns would be less desirable because of potentially higher losses of nutrients to stream water, although in the absence of a thick, insulating layer of organic litter, the ground would tend to freeze earlier, thereby reducing water infiltration and leaching of nutrients [93].

Increased availability of nutrients following fire would increase the productivities of all plants, especially those at or near ground level: little bluestem grassland productivity would increase 20% [59]; blueberry (and presumably other shrubs) would increase production at least a similar amount [86].  Tree productivity would also be increased.  Although no studies were found on oak or hickory productivity following fire, one study on red and white pine showed an 18% increase in volume production after 20 years of controlled burning in Connecticut [50].  But, since the Indians would have been interested chiefly in nut production, and since nut production does not necessarily follow trunk volume growth [32, 61], other ways that fire might affect tree fruiting should be investigated.

Widely spaced trees with full, deep crowns are generally known to produce more fruit (nuts) than trees in closed canopy forest stands [31, 45].  By eliminating or reducing the fire-susceptible birches, maples, beech, hemlock, and weak individuals of other species [12, 60] burning would have created the open stands of oak, hickory, and chestnut of the early accounts.  Trees of such an "oak-hickory savannah" (D. Olson, personal communication) would have had full, deep crowns for optimum nut production.  Furthermore, the blackened soil would have absorbed heat in the daytime and re-radiated it at night [82] which would have reduced spring frost damage to reproductive tissues: warm air temperatures in late April are known to be crucial to good acorn and hickory nut production [61, 73].  Burning would also have made available significant amounts of potassium [61, which is known to be very important to the nutrition of nut trees [74].  Finally, scorching of the bark would probably have served the same purpose as partial girdling in blocking the downward translocation of carbohydrates, which is known to be a cause of increased fruiting in some trees [28, 65, 77].

The forest animals which were most important in the Indians' diet (e.g., deer, bear, moose, turkey) are all heavy consumers of mast when it is available [15].  But these animals, as well as humans, experience intense competition from squirrels and other rodents, birds, insects, and fungi.  Ahlgren and Ahlgren [62], Bendell [8], and Wright and Bailey [93] stated that squirrels and other rodent populations are much lower in fire forest because removal of their cover (grass and shrubs for mice, voles, etc.) and escape routes (continuous tree canopy for squirrels) would make them more susceptible to predators.  However, large nut-eating birds such as jays and crows would probably increase following fire [8, 93].  Some nut tree insect larvae which overwinter in the leaf litter would be destroyed by fire, particularly the hickory shuckworm, oak twig pruner, oak twig girdler, solitary oak leaf miner, and walnut curculio [41].  Since fire would generally disfavor conditions for buildup of fungal inoculum [85], fallen nuts would probably be less affected in a fire forest.  Smoke is also known to have an antimicrobial effect [63].

The animals that the Indians hunted heavily are all browsers as well as berry and mast-eaters.  They all rely on large quantities of succulent grass, forb, and shrub shoots in the spring and early summer [93].  In the fall and winter they require (deer especially) shrub and tree sprouts for browsing.  Niering (personal communication) found that oak seedlings nearly doubled following burning, and hypothesized that persistent oak seedlings are stimulated by fire; Swan [79] also found vigorous sprouting from stem-killed oak and northern hardwood saplings.  Both Niering and Swan found increased production of succulent grasses and shrubs following fire.  Furthermore, Wright and Bailey [93] noted that the nutrient content of grasses, shrubs, and tree seedlings improves for several years following burns.

According to studies in Colorado and California, deer herds in burned forest are at least 2-4 times greater than those in unburned forest [93].  Increased mast, browse, and better nutrient status of these foods have been indicated as the causes.  Fire effects on bear, moose, or turkey have not been studied, nor have effects on total wildlife standing biomass been investigated.

It would appear likely that a sort of "coppice with standards" forest [75] would result from less frequent burning.  The grass species found on more frequently burned savannah sites would be displaced by tree sprouts and, eventually, saplings, while the overstory of large nut trees would remain.  Such a coppice with standards stand was in fact described by an early explorer in Maine:

And surely it did all resemble a stately park, wherein appears some old trees with high withered tops, and other flourishing with living green boughs . . . the wood in most places, especially on the East side, very thinne, chiefly oke and some small young birch. [68, cited in 24]

Figure 4.  Private property in Huntington, MA.  This mixed hardwood stand was severely burned in August, 1958.  Note the large red oaks which survived the fire and the small birches which have grown into the understory since the fire.

If allowed to grow out of the reach of deer, birch or other understory trees could eventually become a crop of fuelwood which could be killed by burning and then left to dry standing.  Because of its susceptibility to fire [60], a birch understory would have been especially appropriate for this type of coppice with standards forestry.  Furthermore, birch, like pine, would invade a disturbed site by seeding in from adjacent, undisturbed stands [2] and quickly become established.

The importance of fuelwood to the Indian subsistence strategies has been well documented [24, 71], although the actual means by which the Indians harvested the wood is unclear.  Stone axes were available, but whether they were extensively used for felling and splitting is unknown.  Considering the evidence produced by Niering et al. [60], it seems likely that fire-susceptible species like black birch could have been allowed to grow in old fields or nut groves near the villages as indicated above.  Fire killing would save the labor of felling and small fire-killed wood, once dried, could be broken into usable lengths by hand.  Larger trees, whether fire susceptible or not, would be harder to kill by fire [60, 79] and would require extra labor with tools of questionable effectiveness and durability in order to produce usable-sized pieces of wood.  In small to medium sized pole stands (10-20 cm diameter at breast height) of mixed species composition the nut trees would have developed fire resistance, while other trees would still be susceptible [60]; a controlled burn in such a mixed stand would accomplish the dual objective of releasing the nut trees from competition and providing a supply of dead wood for fuel.

Figure 5.  Research site at the Connecticut Arboretum in New London, CT.  This area has been burned bi-annually since 1981.  Note the dead black cherry trees in the right foreground and center of the photo.  Overstory is black oak; ground cover is sprouts of black cherry, black birch, red maple, plus andropogon, wintergreen and huckleberry.

Areas producing fuelwood as suggested above would have been but one more aspect of a very diverse landscape of fields in various stages of cultivation or fallow succession, plus forest stands variously affected by fire [20].  Diversity would have been imposed by physical features of the environment as well as the Indians' manipulation of it. Since burning in November or April would normally have been somewhat hampered by wet ground conditions, individual burns are likely to have been of limited extent, because of natural water courses (as suggested in Section IV, above) and because of seasonally moist soils which would have burned less than during summer or early fall.  A soil series map would easily indicate the areas having soils with poor drainage which would be less likely to burn in November or April.  Thus, a soils matrix overlain by a land use matrix could show a variety of types of vegetation in various stages of recovery from small burns.  The increased diversity of plant species following fire [59, 79] would add another dimension to this picture of diversity.

These factors of diversity--different types of vegetative covers, different ages of covers, and complexity of species within a type of cover--are all regarded as desirable characteristics in wildlife habitat management [15].  Not only do they provide an abundance of food and protection, they also provide different types of cover needed by birds and animals in their various life activities:

Generally a mosaic of seral and climax stages of vegetation that may be created by a series of fires over time in shrub and forest communities is the preferred habitat for the highest diversity and number of wildlife species.  Such habitat provides maximum "edge" for feeding, escape, loafing, and nesting [93]

Diversity and maximum edge would be natural results of periodic low-intensity fires, whereas no fire and high intensity fire would result in less diversity and edge.

To conclude the discussion of the effects of controlled burning of the forest, several off-site effects require mention.  First, any shift in species composition away from trees and toward grass would entail a loss of transpiring leaf surface area, and a consequent increase in water runoff [26].  Second, the amount of nutrients in water leaving a controlled burned watershed will generally be slightly greater than from unburned controls [27]; sediment loading of the streams may also be slightly higher.  Effects on stream flora and fauna should be positive from slightly increased nutrient levels [58], but negative from slightly increased sedimentation.  Third, the negative impact of smoke on air quality involves many complicated factors, but if a choice must be made between relatively frequent prescribed fires and infrequent wildfires (which eventually must occur if fuels are allowed to accumulate), then prescribed fires are preferable [85].

V. Discussion

The present day oak and white pine types of southern New England described by Bromley (Figure 2) can also be found in southern New Hampshire, southern Maine, New Jersey, southeastern New York, central New York, and eastern Pennsylvania [80].  These types can be characterized as being fire resistant, but not fire adapted [52].  The thick-barked black oak species and shagbark hickory (and in former times, American chestnut) are especially resistant.  Bromley [11] believed that in pre-settlement times the oak type also included most of the areas now covered by the pine type.  Studies by Oosting [62] and Little and Moore [48] would tend to confirm Bromley's hypothesis: low-intensity fires tend to favor regeneration of oak and hickory and discriminate against pine.  Thus, by the use of controlled burning, the Indians could have greatly expanded the extent of the areas that could have naturally tended toward the oak type [66].

Figure 2. The Principal Forest Regions of Southern New England (from Bromley [11]).

Fire rotation times, or the length of time between fires [92], have been hypothesized for the oak and pine types [47]:

. . . fires may have burned at intervals of 3 years or less on the drier forested sites, at 1-year intervals in the grasslands, and at long intervals--100 years or so--on the wetter forested sites, at which times severe fires could have developed during drought periods in certain types . . .

A three year fire rotation time in the oak-hickory savannah would probably have maintained the type indefinitely [21].  Similarly, a one year fire rotation time would probably have maintained the grasslands indefinitely.  If such were the case, the sequential agroforestry practices described in Section III would not have been very significant in extent, except perhaps on some of the poorly drained grassland soils which might not actually have burned each year.  Less frequent burning of the grasslands, or use of fire breaks as indicated by Martin [52], would have allowed for establishment of tree cover on old fields.  In any case, berry production and large bird habitat would have been byproducts of the grass and/or tree-covered fallow fields.

Moist forest sites on long fire rotation times would have added edge and wildlife shelter to the landscape.  These areas would have been refuges for the smaller birds and mammals which were not adapted to the grasslands or fire forest [8], and for larger birds and mammals in hot weather and during winters with deep snow.  The hemlock-pine stands typical of these areas [11] would have intercepted snowfall and allowed easier browsing.  Areas with deep soils in moist areas would undoubtedly have grown some butternut or black walnut [30]; nuts of the butternut were especially important to the diets of Indians in northern parts of the Northeast [14, 63].

We will probably never know the actual extent of the fire-adapted oak-hickory savannahs and grasslands that the Indians created where the present day oak and pine types are now found, but their existence cannot be seriously doubted in light of the historical accounts and of modern research on the effects of fire on ecosystems.  Periodic low intensity fires in November or April would have increased primary production of vegetable foods for humans and wildlife by 20-100% and would have concentrated the food production within easy feeding reach of large birds and mammals.  Competition for these foods from rodents, insects, and fungi would have been reduced.  Large birds and mammals (e.g., grouse and deer) would have increased 100 to 400% in the periodically burned forest.  The Indians were doubtless very much aware of these beneficial effects of fire, and managed their environment accordingly.

The sustainability of the Indian agroforestry systems is of course open to question.  But there seems to be no evidence that serious "leaks" would have developed, in the sense indicated by Harper [36]:

Agriculture deliberately channels energy and mineral resources out of one area into another and in addition it commonly creates "leaks" in otherwise self-maintaining systems.

Since soil would have been exposed to erosive forces only following very infrequent severe burns, and since nutrient losses would have been minimal or non-existent, no leaks of these resources should have occurred.  And, obviously, there would have been no inputs (losses) of fossil fuel energy in the form of fertilizers, herbicides, or machinery fuel [64].  The only leaks would have been in the form of the food crops which the Indians harvested from their agro-ecosystems. it is interesting to note in this regard that historical records indicate that the Indian fields were able to produce crops of corn equal to or greater than the settlers' fields [83]--without use of draft animals or plows; these same Indian fields also produced crops of beans and squash which the settlers' fields did not.

VI. Implications for Modern Agroforestry Systems

Morris [56], Smith [76], and MacDaniels [51] suggested that many tree species are capable of commercial nut production in the Northeast,(3) among them, in descending order of potential adaptability [23]: Korean pine (4), ginkgo, shagbark hickory, black walnut, Chinese chestnut, filbert/hazel, Carpathian walnut, butternut, and white oak.  Chinese chestnut, filbert, and Carpathian walnut are capable of producing the highest value crops, but are limited by diseases (e.g., chestnut blight, eastern filbert blight, walnut anthracnose, and witch's broom of walnut) and cold hardiness factors (i.e., winter kill and spring frost damage to reproductive tissues in colder areas).  Korean pine, ginkgo, shagbark hickory, and black walnut produce high value crops and are less affected by disease and cold hardiness problems [23].  Butternut and white oak produce crops of medium or low value and are little affected by diseases and cold temperatures.

Much research has been done on developing disease resistance in American chestnut, and on hybridizing American and Chinese chestnut for blight resistance [39], but the potential roles of fire or smoke in destroying the pathogen have not been investigated.  Considering the evidence presented by Hardison [35] and Parmeter and Uhrenholdt [63], such lines of investigation could be fruitful for reducing the incidence of chestnut blight, as well as eastern filbert blight and walnut anthracnose.

Since 1909, members of the Northern Nut Growers Association (NNGA) have done considerable breeding work to develop more cold-hardy strains of many nut tree species (NNGA Annual Reports 1-75).  Very little work has been done on modifying the soil nutrient status or microclimate in nut groves to minimize spring frost damage [38, 74].  Considering how fire increases the availability of nitrogen and potassium, which are known to be important in enhancing hardiness of nut trees [74], prescribed burning could be beneficial in nut groves if the trees were protected from fire.  Higher nighttime temperatures above blackened soil could also benefit many nut tree species by increasing fruit set in late April and early May.

Korean pine is a five needle pine very similar to eastern white pine in its growth habit, site requirements, and temperature requirements [44, 56, 47].  It should benefit associated vegetation by increasing nitrogen and phosphorus in the rhizospere in the same way other pines do [29, 67].  This nut-bearing pine could be planted as a multicrop with other nut trees and shrubs, particularly with hazels since pines and hazels associate naturally in northern climates [13].  Since older pines are fire resistant [48] and hazels are fire resilient [13], Korean pine and hazels (or perhaps filberts) could be ideal associates in a fire-maintained agroforestry system.

The precocity and productivity of apple and other tree species have been increased by grafting scions with good fruiting characteristics onto dwarfing rootstocks [82]. Little [46] ascribed the low growth habit of pitch pines in a frequently burned environment to genetic dwarfing.  Perhaps dwarfing rootstocks of pine and other species (e.g., shagbark hickory) could be obtained from areas which have experienced frequent fires over long periods.

Lowe et al. [49] stated that increased wildlife populations resulting from fires would also increase revenues from sales of hunting licenses.  Use of controlled burning in the Northeast could theoretically have the same benefits.  Use of fire in managing woodlands for wildlife would not necessarily seriously conflict with high quality timber production since northern nut tree species are known to compartmentalize infections resulting from basal wounds [37] and diameter growth rates are not reduced by fire wounds [40].

VII. Conclusions

The Indians' use of fire in manipulating successional trends in their environment enabled them to maintain sustainable agro-ecosystems that were very productive of the vegetable and animal foods that the Indians depended upon, particularly on the drier, upland soils.  Valuable information is available in the modern fire ecology literature which may help us in the future to develop sustainable agroforestry systems for the Northeast.  While future agroforestry systems for the region may be quite different from those of the Indians in many respects, they are likely to resemble the Indians' systems more than the upland agro-ecosystems imported from Great Britain and Western Europe, which have been plagued by very serious leaks in the form of soil and nutrient losses due to plowing, cultivating, overgrazing, and overcutting.

Endnotes

1.  Taungya systems incorporate fast-growing timber species (e.g., teak and eucalyptus) in a sparse overstory with understory food crops (e.g., maize, potato).  Forest farming systems integrate fast-growing timber species (i.e., radiata pine) with sheep or cattle grazing.

2.  Corn would be planted in "hills" of piled earth.  Later, beans would be planted which would then grow on the corn stalks.  Squash would also be interplanted with the corn.  The beans would fix nitrogen in the soil and the large squash leaves would act as a living mulch to control weeds.

3.  The commercial success of these nut trees depends upon the selection and propagation of high quality cultivars which produce nuts that crack out large kernels easily.  Many such cultivars are available from members of the Northern Nut Growers Association.

4.   Korean pine produces nutlets in cones which are very similar to pinyon pine or stone pine nuts.  Ginkgo produces a nut which is similar in shape to a pistachio but tastes like a chestnut.  Both nuts command high prices in gourmet and international food stores.  Shagbark hickory nuts have a unique taste that many people prefer to any other nut.  Black walnut kernels and shells have a variety of uses in baking and abrasion, respectively.

VIII. References

1. Agroforestry Systems, eds (1982) What is agroforestry?  Agroforestry Systems 1(1): 7-12.

2. Ahlgren CE (1959) Some effects of fire on forest reproduction in northeastern Minnesota.  J of  Forestry 57(3): 94-200.

3. Ahlgren CE (1973) Effects of fires on temperate forests: North-central United States.  In: Kozlowski TT, Ahlgren CE, eds. Fire and Ecosystems, pp 195-223.  New York: Academic.

4. Ahlgren CE (1976) Regeneration of red pine and white pine following wildfire and logging in northeastern Minnesota.  J of Forestry 74(3): 135-140.

5. Ahlgren IF (1973) The effect of fire on soil organisms. In: Kozlowski TT, Ahlgren CE, eds, Fire and Ecosystems, pp 47-72.  New York: Academic.

6. Ahlgren IF, Ahlgren CE (1960) Ecological effects of forest fire.  Botanical Rev 26: 483-533.

7. Anonymous (1880) Relation of a voyage to Sagadahoc, 1607-1608.  In: Original Narratives of Early American History 3(1906): 394-419.

8. Bendell JF (1973) Effects of fire on birds and mammals. In Kozlowski TT, Ahlgren          CE, eds, Fire and Ecosystems, pp 73-138.  New York: Academic.

9. Bidwell PW, Falconer JI (1941) History of Agriculture in the Northern United States.  New York: Carnegie Inst.

10. Bock CE, Lynch JF (1970) Breeding bird populations of burned and unburned conifer forests in the Sierra Nevada. Condor 72: 182-189.

11. Bromley SW (1935) The original forest types of southern New England.  Ecological Monographs 5: 61-89.

12. Bromley SW (1945) An Indian relict area.  Science 60: 153-154.

13. Buckman RE (1964) Effects of prescribed burning on hazel in Minnesota.  Ecology 45:626-629.

14. Bumsted MP (1980) VT-CH-94: Vermont's earliest known agricultural experiment station. Man in the Northeast 19: 73-82.

15. Burger GV (1973) Practical Wildlife Management.  NewYork:  Winchester.

16. Butler EL (1948) Algonkian culture and use of maize in southern New England.  Bull of the Archaeological Soc of Connecticut 22: 2-39.

17. Carmean W, Clark FB, Williams RD, Hannah PR (1976) Hardwoods planted in old fields favored by prior tree cover.  USDA For Serv Res Pap NC-134.  St Paul, MN.

18. Ceci L (1975) Fish fertilizer: A native North American practice?  Science 188: 26-30.

19. Champlain S de (1613-32) The Works of Samuel de Champlain.  Bigger HP, ed.  Toronto: Champlain Society.

20. Cronon W (1983) Changes in the Land: Indians, Colonists, and the Ecology of New England. New York: Hill and Wang.

21. Curtis JT (1959) The Vegetation Types of Wisconsin. Madison:  U of Wisconsin Press.

22. Danhof CH (1969) Change in Agriculture: The Northern United States 1820-1870. Cambridge, MA: Harvard U Press.

23. Davies KM Jr (1984) A systematic approach for indicating potential new perennial crop systems for the Northeast.  Master of Professional Studies Project Report.  Dept of Natural Resources, Cornell U, Ithaca, NY.

24. Day GM (1953) The Indian as an ecological factor in the northeastern forest.  Ecology 34(2): 329-346.

25. DeSelm HR, Clebsch EEC, Nichols GM, Thor E (1973) Response of herbs, shrubs, and tree sprouts in prescribed burn hardwoods in Tennessee.  In: Proc Ann Tall Timbers Fire Ecology Conf 13: 331-345.  Tallahassee, FL.

26. Douglass JE, Swank WT (1975) Effects of management practices on water quality and quantity: Coweeta Hydrologic Laboratory, North Carolina.  In: Municipal Water.shed Management Symp Proc 2-13.  USDA For Serv Gen Tech Rep NE-13.  Broomall, PA.

27. Douglass JE, Van Lear DH (1983) Prescribed burning and water quality of ephemeral streams in the Piedmont of South Carolina.  Forest Science 29(l): 181-189.

28. Ebell LF (1971) Girdling: Its effect on carbohydrate status and on reproductive bud and cone development of Douglas Fir.  Canadian J of Botany 49: 453-466.

29. Fisher RF, Stone EL (1969) Increased availability of nitrogen and phosphorous in the root zone of conifers.  Soil Sci Soc of America Proc 33: 955-961.

30. Folwells HA (1965) Silvics of forest trees of the United States.  USDA For Serv Agri Handbook #271. Washington, DC.

31. Garrett HE, Kurtz WB (1982) Nut production--a valuable asset in black walnut production.            In: Black Walnut for the Future: 53-56.  USDA For Serv Gen Tech Rep NC-74.  St Paul, MN.

32. Goodrum PD, Reid VH, Boyd CD (1971) Acorn yields, characteristics, and management criteria of oaks for wildlife.  J of Wildlife Management 35: 520-532.

33. Gordon RB (1940) The primeval forest types of southwestern New York.  NY State Museum Bull 321.

34. Hall IV (1955) Floristic changes following the cutting and burning of a woodlot for blueberry production. Canadian J of Agricultural Sci 35: 143-152.

35. Hardison JR (1976) Fire and disease.  In: Proc Ann Tall Timbers Fire Ecology Conf 15: 223- 234.  Tallahassee, FL.

36. Harper J (1974) Agricultural ecosystems.  Agro-Ecosystems 1: 1-6.

37. Hepting GH, Shigo AL (1972) Difference in decay rate following fire between oaks in North Carolina and Maine.  Plant Disease Rep 56: 406-407.

38. Hilton RJ (1974) Hardiness in nut-bearing plants.  Northern Nut Growers Assn Ann Rep 65:75-78.

39. Jaynes RA Selecting and breeding blight resistant chestnut trees.  In: Proc of the American Chestnut Symp: 4-6.  West Virginia U, Morgantown, WV.

40. Jemison GM (1944) The effect of basal wounding by forest fires on the diameter growth of some southern Appalachian hardwoods.  Duke U School of Forestry Bull 9: 1-63.

41. Johnson WT, Lyon HH (1976) Insects that Feed on Trees and Shrubs.  Ithaca, NY: Cornell U Press.

42. Judd S (1863) History of Hadley. Northampton, MA.

43. Keeley JE (1981) Reproductive cycles and fire regimes. In: Fire Regimes and Ecosystem Properties, Proc of the Conf at Honolulu, HI: 231-277.  USDA For Serv Gen Tech Rep WO-26.  Washington, DC.

44. Larsson HC, Jaciw P (1978) Prospects for growing Asian nut pines in the temperate and boreal forests of North America.  Northern Nut Growers Assn Ann Rep 69: 90-95.

45. Lassoie JP, Siebert SF, Lieberman AS, MacDaniels LH (1980) Acorns from natural woodlands and their utilization.  In: Tree Crops for Energy Co-Production on Farms: 33-52. US Dept of Energy.  National Tech Info Serv.  Springfield, VA.

46. Little S (1972) Growth of planted white pines and pitch seedlings in a south Jersey plains area.  Bull NJ Acad of Science 17: 18-23.

47. Little S (1974) Effects of fire on temperate forests: northeastern United States.  In: Kozlowski TT, Ahlgren CE, eds, Fire and Ecosystems: 225-250.  New York: Academic.

48. Little S, Moore EB (1949) The ecological role of prescribed burns in pine-oak forests of southern New Jersey.  Ecology 30: 223-233.

49. Lowe PO, Ffolliott PF, Dieterick JH, Patton DR (1978) Determining potential wildlife benefits from wildfire in Arizona ponderosa pine forests.  USDA For Serv Gen Tech Rep RM-52.  Fort Collins, CO.

50. Lunt HA (1951) Liming and twenty years of litter raking and burning under red (and white) pine.  Soil Science Soc of America Proc IS: 381-390.

51. MacDaniels LH (1981) Nut growing in the Northeast.  Coop Extension Info Bull 71.  Cornell U, Ithaca, NY.

52. Martin C (1973) Fire and forest structure in the aboriginal Eastern forest.  Indian Historian 6(4): 38-42, 54.

53. Martin RE, Cushwa CT (1966) Effects of heat and moisture on leguminous seed.  In: Proc Ann Tall Timbers Fire Ecology Conf 5: 159-175.  Tallahassee, FL.

54. Maxwell H (1910) The use and abuse of the forests by the Virginian Indians.  William and Mary Quarterly 19:3 3-103.

55. McDermott B (1975) Textbook of medicine (14th edition). Philadelphia. WB Saunders.

56. Morris RT (1924) Nut Growing.  New York: MacMillan.

57. Morris RT (1929) Notes on nut-bearing pine trees.  Northern Nut Growers Assn Ann Rep 20:15-122.

58. Neary DG, Currier JB (1982) Impact of wildfire and watershed restoration on water quality in South Carolina's Blue Ridge Mountains.  Southern J of Applied Forestry 6(2): 81-90.

59. Niering WA (1981) The role of fire management in altering ecosystems.  In: Fire Regimes and Ecosystem Properties, Proc of the Conf at Honolulu, HI: 489-510.  USDA For Serv Gen Tech Rep WO-26.  Washington, DC.

60. Niering WA, Goodwin RH, Taylor S (1970) Prescribed burning in southern New England: Introduction to long range studies.  In: Proc Ann Tall Timbers Fire Ecology Conf 10: 267- 286.  Tallahasse, FL.

61. Nixon CM, McClain MW, Hansen LP (,1980) Six years of hickory seed yields in southeastern Ohio.  J of Wildlife Management 44(2): 534-539.

62. Oosting HJ (1944) The comparative effect of surface and crown fire on the composition of a loblolly pine community.  Ecology 25: 61-69.

63. Parmeter JR Jr, Uhrenholdt B (1976) Effects of smoke on pathogens and other fungi.  In: Proc of Ann Tall Timbers Fire Ecology Conf 14: 295-305.  Tallahassee, FL.

64. Pimentel D, Terhune EC, Dyson-Hudson R, Rochereau S, Samis R, Smith EA, Denman D, Reifshneider D, Shepard M (1976) Land degradation: Effects on food and energy resources. Science 194: 149-155.

65. Pond JD (1936) Girdling for seed production.  J of Forestry 34: 78-79.

66. Raup HM (1937) Recent changes in climate and vegetation in southern New England and adjacent New York.  J of the Arnold Arboretum 18: 79-117.

67. Richards BN (1962) Increased supply of soil nitrogen brought about by Pinus.  Ecology 43:538-541.

68. Rosier J (1605) True relation of Waymouth's voyage, 1605.  In: Original Narratives of Early American History 3 (1906): 353-394.

69. Ruffner JA, Bair FE, eds (1981) The Eeather Almanac. Detroit: Gale Research Co.

70. Russell EWB (1983) Indian-set fires in the forests of the northeastern United States.  Ecology 64(l): 78-88.

71. Russel HS (1980) Indian New England Before the Mayflower.  Hanover, NH: Univ Press of New England.

72. Sharp WM (1970) The role of fire in ruffed grouse habitat management.  In: Proc Ann Tall Timbers Fire Ecology Conf 10: 47-61.  Tallahassee, FL.

73. Sharp WM, Sprague VG (1967) Flowering and fruiting in the white oaks: Pistillate flowering, acorn development, weather, and yields.  Ecology 48(2): 243-251.

74. Shear CB (1970) Importance of potassium in the nutrition of nut trees.  Northern Nut Growers Assn Ann Rep 57: 25-28.

75. Smith DM (1962) The Practice of Silviculture.  New York: J Wiley & Sons.

76. Smith JR (1950) Tree crops: A Permanent Agriculture.  New York: Harper Colophon edition (1978).

77. Stephens GR Jr (1964) Stimulation of flowering in eastern white pine.  Forest Science 10: 28-34.

78. Stone EL Jr (1971) Effects of prescribed burning on long-term productivity of Coastal Plain soils.  In: Prescribed Burning Symp Proc, held at Charleston, SC: 115-159.  USDA For Serv Southeast For Exp Sta; Duke U & Clemson U, Clemson, SC.

79. Swan FR (1970) Post-fire response of four plant communities in south-central New York State. Ecology 51(6):1074-1082.

80. Swank WT, Waide JB (1980) Interpretation of nutrient cycling research in a management context: Evaluating potential effects of alternative management strategies on site productivity. In: Forests: Fresh Perspectives from Ecosystem Analysis.  Proc of 40th Ann Biology Colloq: 137-157.  Oregon St U Press: Corvallis, OR.

81. Temple JH (1887) History of North Brookfield.  Boston.

82. Teskey BJE, Shoemaker JS (1972) Tree Fruit Production.  Westport, CT: Avi.

83. Thomas PA (1979) In the maelstrom of change: The Indian trade and cultural process in the middle Connecticut River Valley 1635-1665.  PhD Thesis, Anthropology Dept, U of Massachusetts, Amherst, MA.

84. Thompson DQ, Smith RH (1970) The forest primeval in the Northeast--a great myth?  In: Proc Ann Tall Timbers Fire Ecology Conf 10: 255-265.  Tallahassee, FL.

85. Tiedemann AR (1981) Regional impacts of fire.  In: Fire Regimes and Ecosystem Properties, Proc of the Conf at Honolulu, HI: 532-556.  USDA For Serv Gen Tech Rep WO-26. Washington, DC.

86. Trevett MF (1962) Nutrition and growth of lowbush blueberry.  Maine Agri Exp Sta Bull 605.  U of Maine, Orono, ME.

87. Tustin JR, Knowles RL, Klomp BK (1979) Forest farming: A multiple land-use production system in New Zealand.  Forest Ecology and Management 2: 169-189.

88. Weaver P (1979) Agri-silviculture in tropical America.  Unasylva 31(126): 2-12.

89. Westveld RH (1939) Applied Silviculture in the United States.  New York: John Wiley & Sons.

90. Wood W (1634) New England's Prospect.  Boston: Prince Society (1865).

91. Wright HA (1905) Indian Deeds of Hampden County.  Springfield, MA.

92. Wright HE Jr, Heinselman ML (1973) The ecological role of fire in natural conifer forests of western and northern North America.  Quaternary Res 3: 319-328.

93. Wright HA, Bailey AW (1982) Fire Ecology: United States and Southern Canada.  New York: Wiley & Sons.

IX. Common Names for Species and their Scientific Equivalents