BOTANICAL EVALUATION OF THE GOAT ISLAND COMPLEX, NIAGARA FALLS, NEW YORK
Following Charles Lyell's interpretation (1845) of the general outline of bedrock surfaces and subsequent deposition of sediments in the region about the Falls, there are:
a) first the bedrock strata themselves, all sedimentary: shales, sandstones, limestones and magnesium limestones (dolostones) all undergoing surface erosion by water at the beginning of the recent ice ages (of the Pleistocene). It was at this time that the escarpments and other cliff-features were formed in resistant layers over softer ones. Then,
b) the continental ice sheet stripped away all loose material, soil, boulders, the bedrock surface becoming "smoothed, polished and furrowed by glacial action" at the base of the ice mass;
c) as the ice retreated, the entire region including the lake floors, the table land above the escarpment, and the lake plain below it was covered with glacial drift: the deposition of sediments once suspended in the ice -"stratified and unstratified sand, gravel, and erratic blocks."
d) The land was underwater, such that the glacial drift was to some extent rearranged by new water currents, and marine drift became deposited upon it, including the shells of mollusks still alive in the lakes and river today.
e) "As soon as the table-land between Lakes Erie and Ontario emerged and was laid dry, the river Niagara came into existence."
Goat Island, like the surrounding territory, was covered with glacial drift, forming the basal layer of sediments. Carbon-14 dating of shells found on Goat Island indicate that of the 12,000 years since the area was liberated from glaciation, Goat Island has spent 9800 of those years as an island, that is, "unaffected by river erosion" (Brett in Tesmer, 1981), although theory based on geomophologic evidence has traditionally set the date to 2000 to 3000 years (Calkin & Brett, 1978).
Above this, unconsolidated sedimentary layers on the island and in terraces along the Niagara River indicate that the land now comprising Goat Island was submerged in a river with lake-like conditions to a depth of 100 feet (Porter, 16 Ann Rep Comm, 1900), with a northern boundary, or dam, at the situation of Lewiston, New York. Coarser sediments indicate increasing river turbulence with time. The height of this body of water decreased in elevation until, when the falls had cut its way past the Whirlpool, or at any rate, at "a point nearly a mile north of the present location of the Horseshoe Fall," the previously submerged surface of Goat Island, continuous with much land which now forms part of the mainland, became open to the air (Porter, 16 Ann Rep Comm, 1900). The channel of the American Falls had not yet been cut and Goat Island was still part of the eastern mainland, to which its sediments display great affinity (Kindle and Taylor, 1913). River gravel deposits of the same age and origin were mapped on Goat Island and Prospect Park, as well as in little cells on what is now the Queen Victoria Niagara Falls Park, in Ontario, by Kindle and Taylor (1913). This continuity of strata across what is now river bed was also noted by Agassiz (1850).
During this period the flora of the mainland would have established itself generally without interruption over these surfaces.
The mollusk layer below the surface of Goat Island has been dated at 9100 years B.P., suggesting deposition ceased at that time, and the land became exposed by lowered river levels. Spencer (1907, 1910) and Taylor (Kindle & Taylor, 1913; Taylor, 1933) give geomorphologic evidence that the Goat Island surface did not become exposed until 2000 to 3000 years B.P. (see Calkin & Brett, 1978). Calkin and Brett (1978) suggest "termination of deposition on Goat Island may have involved only shoaling rather than complete emergence of the surface."
Some centuries after the exposure of this land, water, which had been draining from the three upper Great Lakes through an outlet to the north was redirected through the St. Clair outlet around Detroit into Lake Erie as the eastern land mass, once depressed by the weight of the continental glacier, rebounded upward, and water began to drain more through Lake Erie and the Niagara River, than northern outlets. The result was that "the volume of the Niagara river was suddenly and enormously increased," raising the level of the river, and cutting across vulnerable low areas where the river changed course, flowing from east to west below Grand Island, and suddenly turning north at what is today the present position of the cataracts. The river "cut a corner" here, and created the American channel and cataract, eroding the glacial and riverine sediments formerly deposited there down to bedrock - its present condition (Porter, 16 Ann Rep Comm, 1900). Prior to this interpretation, both river channels flanking Goat Island were thought to have been created simultaneously with a decrease in the level of water in Lake Erie, and consequently in the Niagara River (Agassiz, 1850).
It is from this juncture that the destiny of the Goat Island flora would become separated from that of the mainland. It should be noted that in this scenario, the Three Sisters Islands, the islands in the American channel, Terrapin Rocks, Dufferin Islands, and all the land below or at the base of the little escarpment (the Niagara Moraine) in the Queen Victoria Park were still under water, inferred from geological relations depicted in the Kindle-Taylor map of 1913, although Porter inferred they emerged from the same event as that causing the emergence of Goat Island.
Lyell (1845) published two stratigraphic diagrams perhaps printed earlier by James Hall in a report of 1839 mentioned in Lyell's work. The diagrams represent the geological strata and their composition on Goat Island. The first diagram showed "freshwater strata on Goat Island, above 20 feet thick" beneath which lies an 80 foot thickness of limestone, and an 80 foot thickness of shale below that. The diagram indicated the identical origin of similar deposits on the American shore (mistakenly indicated as "east" when in fact it is north). These freshwater strata overlay glacial drift on the mainland banks, but no glacial strata in Lyell's map are indicated on Goat Island below the freshwater layers. Later interpretations suggest the lowest layers were in fact glacial drift deposits (Kindle & Taylor, 1913), that is, the basal sediments arose from the water flowing in the body of the glacial ice-mass and exiting from outlets in the base of the ice.
From examining a nine-meter section of the bank overlooking Luna Island, Calkin and Brett (1978) suggested the lowest section of sediment on Goat Island was the pebble-boulder gravel which extends down to the dolostone bedrock of the island, as it appears to do elsewhere in the gorge. The boulders are of dolostone (see Hall's "limestone of Black Rock" below). Above this lies glacial till, over that glaciolacustrine deposits, all capped by the "ancient mollusk-bearing gravels which form most of the flat surface of Goat Island."
Lyell's second diagram showed a cross section of Goat Island showing an east-west exposure that would parallel the river as indicated in the text (not a north-south one as indicated in the printed figure). The bluffs of the Island, corresponding with the brinks of the cataracts are depicted to the left where a line shows the land comes to an end, and the eastern, or upriver, end of Goat Island is represented as indeterminate, or the strata continue to the right of the diagram. The natural excavation of the calcareous caprock in the west end of the island was reported by Porter (16 Ann Rep Comm, 1900) to have been gouged or eroded, when it is probably interpreted today as the bed of a pre- or interglacial river. "In this excavated cavity, drift was deposited by the ice. Many of the boulders brought here in the ice age ... have been collected in this section and used in the construction of the handsome stone bridges that have been built on the Reservation, on the main shore opposite Goat Island" (Porter, 1900, see section on bridges below).
There is a canted, dotted line passing from left to right across the diagram indicating the "present surface of the river Niagara at the Rapids" showing the steep grade of the river as it descends some fifty feet from the upriver to the downriver end of the Island. The single limestone stratum shown in the 80 foot section in the first diagram is drawn with a layer atop it of the "upper thin-bedded portion of the Niagara limestone" above the "massive compact portion of the Niagara limestone."
For the purposes of reconstructing the original substrates of the Island at the time of the creation of the Reservation in 1885, it would appear that the eastern end of the Island displayed the same character as the Three Sisters do now: bare calcareous rock to the east, alluvium to the west following the trend of the river. "The ledges of the cascades pass under the drift on both sides of the river and on both sides of Goat Island on continuations of nearly the same lines that they follow in the rapids" (Kindle & Taylor, 1913). These ledges are obscured on the southeastern side of Goat Island by grading and ballast deposited on their surfaces, although they must have been more exposed, with a different kind of habitat than is presently seen (see section on the eastern meadow).
"The island is mainly a deposit of gravel upon the Niagara limestone which forms the bed of the river. The soil varies in thickness from forty feet at the western margin to a thin wedge at the eastern end ... where the rock comes to the surface. The eastern and southern margins of Goat Island are rock-bound; the western margin is a steep, rocky bluff. Along the southern margin the bed of the island is composed of layers of gravel, clay, gravel, quicksand and gravel, downward from the surface in the order named" (6 Ann Rep Comm, 1890). A specimen of the moss Hymenostylium recurvirostrum, which never grows on soil, but only on rock, was collected "on ground towards 3 Sister Islands" by Eugene A. Rau in 1886 during a foray of the annual meeting of the American Association for the Advancement of Science. The "ground" was probably exposed bedrock in this area. When the eight acres of land constituting the eastern meadow in 1911 was covered with soil transported from the mainland, much of the area was originally covered with around three inches of topsoil over bedrock. When the area was covered, more than "six hundred cubic yards of stone were picked off the eight acres and piled up for future use on the roads" (28 Ann Rep Comm, 1912). Perhaps this stone indicated recently exposed river bed, as stones were also removed from the bedrock exposed with lowering of water levels at the Horseshoe Falls and represent glacial debris not yet eroded away by the river current. Hall also mentioned the stratum of shell-bearing sediment which "thins entirely out at its southeastern extremity" (Hall, 1843).
In 1823, Douglas noted on "the south side of the island there is very good limestone and a good kind of gypsum" (Douglas, 1823). Certain portions of this area may have been naturally open, and not cleared for farming. Trees were cut down for planting to turnips (see section on land use), but it is unlikely topsoil was removed, except by some kind of erosion. Significant portions of this area may have never been forested. Indeed, when Superintendent Harries considered restoring this area to "the original appearance of this part of Goat Island," he felt that "holes must be blasted out of the solid rock and soil brought in before anything of greater growth than a sumac can be assured" (28 Ann Rep Comm, 1912) as though that area had once supported forest growth.
In Lyell's diagram, trees were drawn scattered not only on the loose sediment to the west, but on the thin soil or rock to the east. It is possible that the original vegetation of this area resembled that of the east end remnants visible today: Red Cedar (Juniperus virginiana) and Panicled Dogwood (Cornus racemosa), and to some extent, on the southeastern side, the vegetation on the eastern ends of the present Three Sisters Islands.
"Upon the southern side of this island, where there is an escarpment, the thickness of the superficial deposit is about twenty-five feet. The upper half consists of coarse gravel and sand, with abundance of ... shells," (Hall, 1843). Along the southern margin the bed of the island is composed of layers of gravel, clay, gravel, quicksand and gravel, downward from the surface in the order named" (6 Ann Rep Comm, 1890). Panton (1890) also mentioned areas of quicksand across the river in the Queen Victoria Niagara Falls Park, where in some places the "soil is a rich loam underlaid by quicksand" and the area "skirting the river below the terraces in front of Falls View ... is of a swampy nature with considerable quicksand."
The quicksand just mentioned, and which contributed to much early erosion on the south side of Goat Island may be similar to, if not produced by the same processes, as the quicksands in the forested Whirlpool Ravine on the Canadian shore in the gorge of the river facing Whirlpool State Park in New York. There one year Albert Tiplin discovered a "four foot thick river of red mud and sand" flowing over the tops of ice piled up on the shore. The sand-sized clasts in this sediment were "bead-like or rounded rather than angular, and prone to slippage, especially when saturated with water" (Tiplin, 1988). The clays at the Whirlpool were produced when the ground moraine which covered the area, including Goat Island, were modified during subsequent submergence beneath ancient Lake Lundy (Kindle & Taylor, 1913).
Lyell's diagram also showed where the river current touched the lowest layers of sediment, indicating where the chief erosional problems were in the Reservation's early years, especially when the river levels were high during storm surges. Some aspects of the early ground water relations on Goat Island may be suggested in this diagram - sedimentary strata touched by the river a century ago will not be affected now with the present lowered water levels in the Niagara River.
The lowest stratum of soil on Goat Island includes a clay, which "proves the first condition [of the area in which the island is situated] to have been that of a quiet lake," with little current (Hall, 1843). Day also referred to this clay: "in a single place upon the island there is to be seen a small quantity of clay, possibly deposited by the glacier where it is found, but more likely to have been brought by the current of the river along with the other materials which make up the soil." This clay is red "laminated glacial lake silt [which] overlies the till ... locally on Goat Island but has been eroded away prior to deposition of the overlying gravel along parts of the latter exposure" (Calkin & Brett, 1978). Barriers to water flow in the regional watershed causing impoundment were subsequently modified such that a current developed in the river with the corresponding energy to move larger and heavier sediment: pebbles and sand, which were deposited upon the clay (Hall, 1843). This layer is a "coarse pebble-cobble gravel, ... is very well bedded and sorted, and may in turn represent glaciolacustrine deposition following ... more vigorous melting during construction of the adjacent Niagara Falls Moraine" (Calkin & Brett, 1978). Above this is "brown, laminated lake clay and silt" representing final "ice retreat and glaciolacustrine deposition prior to development of the ancestral Niagara" (loc. cit.)
The Geologic Atlas of the United States of 1913 (Kindle & Taylor) displayed the distribution of surface sediment on Goat Island and correlated sediments on the mainlands of Ontario and New York. There are generally four sedimentary types depicted:
1. Sediments marked (SI) and Qnc correspond to the Three Sisters and adjacent land on Goat Island to the east of the shelter built near the Sisters of the time, and to all the islands in the American Channel, the mainland shoreline in the Reservation and very small areas in the vicinity of The Spring and at the Terrapin Rocks and Terrapin Point on Goat Island, and extensively along the mainland river shore at and just above Prospect Point. From research into the original aspect at Prospect Point, the Commissioners discovered that area was a "solid uneven ledge of rock sloping gradually back from the river for some distance. The next views, which were photographs, portray it as a natural slope of grass covered soil with an occasional tree or bush" (26 Ann Rep Comm, 1910).
-SI indicates bedrock outcrops of the Lockport dolomite (Silurian rocks).
-Qnc3-7 indicates Pleistocene and Recent sedimentary deposits: "Recently abandoned channel floors of Niagara River, 4, cut while falls retreated 1500 to 2000 feet south of Hubbard Point, 5, falls passing west end of Goat Island, 6 slightly later cutting."
The Second Sister has a mark Qnc6, Terrapin Point Qnc5, and Green Island Qnc4 indicating their relative ages, the lower numbers being the older deposits.
2. Sediments marked Qgmc and Qnc2 occur generally on the north and north east sections of Goat Island.
-Qgmc indicates ground moraine "made by and in conjunction with ice" during the Wisconsin stage of the Pleistocene, i.e. before the development of the subsequent glacial lakes Lundy, Tonawanda and Iroquois. This sediment is interpreted as ground moraine: "sheet of bowlder [sic] clay covering greater part of area" "forming floor of some abandoned channels"
Drainage in the northern part of Goat Island may be poorer than that on the southern due to higher fraction of clay-size sediments implied in the characterization "bowlder clay."
3. Sediments marked Qrg2 constitute the majority of the Island surfaces along the south and western boundaries of Goat Island, and correlate with Prospect Park sediments.
-Qrg2 sediments were deposited during the final stages of the Wisconsin stage of the Pleistocene, constituting river gravels "mostly coarse gravel with cobble deposited in bars or spread over channel floors, contains abundant fossil shells."
Because of the coarser sediment fraction described for these areas, it may be assumed that drainage was more efficient on the south and southwestern sides of the island.
4. A narrow strip of sediment marked Qdb4 on the north slope, Qdb5 on the south toward Terrapin Point, and Qdb6 upriver from there occur on the periphery of Goat Island, ending just downstream of the Three Sisters on the south and at the promontory near the old spring to the northeast.
-Qdb4 indicates "[glacial] drift bluffs forming banks of abandoned [river] channels."
Much of the sediment mantling the embankments or terraces above the crest of the Niagara Gorge, including the soils of Goat Island, are derived from riverine deposits, as is attested to by the inclusion of the remains of marine organisms in their matrix. Most sediments seem to derive from post glacial river deposition (Lyell, 1855, Hall, 1843) over the general sheet of moraine deposits characteristic of the region as a whole - note the numerous glacial erratics, constituting granitic boulders on the Three Sisters, west end, at Prospect Point, and possibly in the central woods. Peter Kalm (1770) noted that the "land about the falls is stony, and here and there a large bit of gneiss or granite is found" isolated out of the glacial drift. In 1898 the soils at the western end of Goat Island were undisturbed. Here "is the original drift, with the subsequent over lying alluvial deposits and accumulations, undisturbed by man" (Porter in 16 Ann Rep Comm, 1900).
As the geological strata, especially the resistant types, are higher in elevation at Lewiston than further south, at the falls, coupled with a gentle basin behind or south of the strata that form the present Niagara Escarpment, the orientation of the strata at Lewiston created a barrier to the river flow in early times, before the gorge had receded seven miles to its present locality. The river levels were correspondingly higher in the basin-like or lower elevations south of Lewiston. The brink of the falls became lower with recession, and the river channel in the gorge deepened in places with a corresponding lowering in the height of the river, exposing the old beaches with their shells at Goat Island and associated terraces on the Canadian and American mainlands (Tesmer, 1981).
The source of much of the stone in these terraces is "of the limestone of Black Rock [upstream at Buffalo, New York], and the harder layers of the Onondaga salt group, like the rock in place at the upper end of Grand island" (Hall, 1843). Day (1910) referred to Hall's comments when he wrote of terraces "largely composed of water-worn stones and materials, brought and deposited by the river itself from more southerly localities."
Three working gravel and/or clay pits were depicted on Goat Island in 1913 (Kindle and Taylor, 1913), the one described by Grabeau below (1901), one beside and just south of the American Falls shelter in the northwest side of the Island, and one toward the north east side inside the ring road to the east of the depicted structure (the Hermit's Hut?). The Porter's may have opened the large one on the island's southwest side, in order to surface the first circuit-way on the island (see section on land use below). Commissioner reports through the years do not specify which pit was the focus of their activities. Compost at one time was collected in "the gravel pit on Goat Island, and turned and mixed from time to time during the year" (16 Ann Rep Comm, 1900 and 17 Ann Rep Comm, 1901). In 1907 an ice house was built in the gravel pit on Goat Island (24 Ann Rep Comm, 1908). In 1912, "the gravel pit has received several thousand yards of fill this winter" (29 Ann Rep Comm, 1913). Later, "the old and useless structures that were in the old gravel pit have been removed and the pit is being filled in." "The old gravel pit near the spring has been graded and is now in condition to receive the barn and proposed labor centre" 29 Ann Rep Comm, 1913.
Grabeau (1901) mentioned that on the way to the Three Sisters from Terrapin Point, there is a spot "where a wood-road leads off to the left into the famous gravel pit of Goat Island, since there the shell-bearing gravels are exposed." This pit had been made famous by the study of its shells and the deduction of their geological significance made earlier by Sir Charles Lyell and James Hall, Geologist of the New York State Survey (Hall, 1843; Lyell, 1855).
"This fluviatile deposit is made by materials brought down by the current, and doubtless mingled with a large accumulation of shells of Unio, Melania, Anculotus, &c., as these shells of Uniones are constantly brought down the rapids during the summer season" (Hall, 1855).
A similar soil mass occurs on the American mainland shore, on a terrace "upon the east side of the river," that is, off the northeast boundary of Goat Island. A mastodon's tooth was discovered there. This terrace of fluviatile deposits continues north as far as the Whirlpool where fossilized shells may still be observed. Hall mentioned a corresponding terrace on the west or Canadian side.
Some soils are glacially derived in the Niagara Falls and Gorge area, most evident at Whirlpool Ravine, Ontario, where the slopes are composed of glacial till which has completely buried a river valley (the St. David's Gorge). Other sediments above these, mantling the walls of the Niagara Gorge and the terraces above the crest of the gorge, and the soils of Goat Island and Niagara Glen, are derived from riverine deposits, as is attested to by the inclusion of the remains of marine organisms in their matrix.
"The most notable" excavation in the old river-banks of the Niagara River "and the one longest known is on Goat Island, perhaps a quarter of a mile inland from the edge of the cliff, at the Biddle stairway. In the section opened here, most of the material is seen to be coarse and rudely stratified. The pebbles are subangular, often quite angular, while some appear to be scarcely worn at all. Blocks a foot or more in diameter are not infrequent, the material being generally limestone from adjoining ledges, though fragments of sandstone and of crystalline rocks are not uncommon. Occasionally a lens of fine sand occurs which shows cross-bedding structure, the laminae pointing in a northwesterly direction. The shells are found on the cross-bedding planes, conforming with them, and indicating that they were spread there by the current which moved the sand grains. Among the coarse material the shells are mixed indiscriminately. In many cases the gravels are of the loose type, with scarcely any sand between them, indicating deposition by a powerful current. Along these zones air and water have most readily penetrated, and a deposition of iron oxide has been formed which stains both pebbles and shells. The shells are generally very fragile, and commonly show signs of wear. Gastropods are most abundant in the Goat Island gravels" (Grabeau, 1901). Lyell (1855) reported finding shells of the genera Unio, Cyclas, Melania, Valvata, Limnea, Planorbis and Helix species in the "surficial deposit" of sediments on Goat Island, none of which were extinct. These shells indicate that Goat Island was once submerged.
Certain of these snails and clams are still living in the river, but some species and subspecies in these layers are presently extinct (Brett in Tesmer, 1981). This deposit:
"is of greatest thickness toward the fall, and thins entirely out at its southeastern extremity. In some places the lowest part of this deposit is of clay, which has been subsequently covered by gravel and sand, containing the freshwater shells. This proves the first condition to have been that of a quiet lake, while subsequently a current, transporting pebbles and sand, passed over the same bed, leaving the coarse deposit" (Hall, 1843).
Members of the Conchological Section of the Buffalo Society of Natural Sciences would conduct field trips to the base of Prospect Point, to Luna Island "from whose rocky edges we pick Goniobasis up to a few inches of where the waters make their destined plunge over the brink. Some terrestrial forms may be found on the Three Sisters Islands" (Robertson & Blakeslee, 1948). "In years past fossil forms of Goniobasis, Pleurocera, and other species could be found, sometimes in abundance, in the interglacial Pleistocene sands and gravels of Goat Island" but "laws forbid any digging unofficially and so the site is apparently forever sealed" (Robertson & Blakeslee, 1948). Fossil shells "were obtained from an open gravel pit supplying material for activities on the island, but a few years ago [prior to 1949] the pit was closed and filled in, and at the present time no opportunity exists for further collection" (Robertson & Blakeslee, 1949).
Actually, the rarest plants in the Goat Island complex occur in areas corresponding to Kindle and Taylor's Qnc sediments - the youngest in the complex (see above) at the edge of relatively newly formed soil mats which are still probably about a century old. Ice scour in winter with corresponding loss of soil and inhibition of the establishment of other, competing plants is probably essential to the continuation of these populations.