QBARS - v16n4 An Ecological Interpretation of Rhododendron Colonies in Maine and New Hampshire

An Ecological Interpretation of Rhododendron Colonies
in Maine and New Hampshire
A. R. Hodgdon and Radcliffe Pike
Department of Botany and Department of Horticulture
University of New Hampshire, Durham, New Hampshire
Published with the approval of the Director of the New Hampshire Agricultural Experiment Station As Scientific Contribution No. 263

Over a large part of its natural range Rhododendron maximum is a local or rare species occupying certain unique habitats. Thus its disjunct distribution in the peripheral part of its range can be explained in part by the discontinuity of possible habitats. But after making extended observations of rhododendrons in central and northern New England we have seen that there are many, more apparently suitable places for the species to grow than there are colonies. One explanation of this interesting fact might be that wind dispersal results in the establishment of new colonies only here and there as the minute seeds are fortuitously blown into new areas.
Yet wind dispersal for great distances does not seem to warrant much consideration due first to the very protected and nearly windless sites in which the Maine and New Hampshire colonies are found and our failure to find seedling reproduction beyond one to two hundred feet away from mature plants, We might more reasonably account for the more disjunct colonies on the basis of their being relics of an earlier more continuous distribution of the species, the assumption here being that various disturbances have occurred to destroy them in many of the swampy areas that now seem to be entirely suitable for them, The recent fluctuations in size of a number of rhododendron colonies in Maine and New Hampshire as reported by us recently give some support to the contention that a great decline in both number and size of colonies could indeed have resulted from the climatic changes that are known to have occurred many times in the post glacial period, A recent review points up the nature of some of these climatic oscillations. The data presented in this paper make it seem likely that the limits of tolerance of rhododendron for certain climatic factors have been exceeded locally during some of the more severe climatic minima. At two periods in post glacial times climatic conditions were apparently such as to make possible the northward migration of warmer floras. These moderate periods were followed by decidedly colder more rigorous climatic conditions. For rhododendron to have persisted as a relic, one must postulate not only a once wider and more general distribution which could conceivably have occurred during these two warmer periods but also a continuity of suitable habitats from the time when the species was more or less continuously distributed.
There have been several papers on Rhododendron maximum emphasizing its distributional peculiarities and something of its ecology. Recently Iltis ³ has discussed interestingly and in some detail an outlying colony ^- in the Coastal Plain of Virginia below Fredericksburg. He found that the majority of common species of associated trees and shrubs there were also among the dominant species listed by Spencer ⁴ in his study of 36 New Jersey Colonies and by Griggs ⁵ for the Sugar Grove Region of Ohio. The Virginian Colony resembled the New Jersey stands also in having a strongly acid soil, pronounced shade and abundant water. Iltis agreed with Spencer's conclusion for New Jersey that the distribution of R. maximum in Virginia is governed by topography and not by climate. It should further be noted that the Virginia colony, all of those in New Jersey reported by Spencer and those in Ohio mentioned by Griggs (see references) are on north-facing slopes or, in the case of a few in New Jersey, in swamps.
During recent field work we have recorded for each of our 19 Maine and New Hampshire colonies, such facts as the associated woody species, the absence or presence of seedling rhododendrons, the nature of the habitat whether swamp or slope and, if so, the direction of slope as well as some other features, We are now in a position therefore to evaluate the ecological requirements of R. maximum by comparing our colonies with those discussed by Iltis, Spencer and Griggs. Since we had no reason to doubt that the soils in all colonies were definitely acid and moreover because of the mass of scientific evidence that shows the oxylophytic character of the species. we decided at the outset not to include pH determinations as part of the record. We did attempt to determine the area of each stand, however, to provide a basis for evaluating the changes that may occur in the future.

TABLE I shows 11 of our colonies to be chiefly in swamps, one other (Manchester (1) divided between a swamp and an adjoining east-facing slope, 4 colonies on north-facing slopes. 2 very small colonies in non-swampy woodland and 1 small stand on a south-facing slope. But it should he remarked that of the 11 swamp-colonies, 2 have excellent rhododendrons on their adjacent south-facing slopes and another has prolific plants on the west-facing adjacent slope. In contrast to the situation in New Jersey, the Coastal Plain of Virginia and in Ohio, swamps generally provide better habitats for rhododendrons in Maine and New Hampshire than do slopes. Also with us the direction of slope does not seem to be critical. We have pointed out elsewhere ⁶ that juxtaposition of swamps and adjacent slopes as alternative habitats for rhododendrons in many of the finest Maine and New Hampshire colonies may provide the species with a means of surviving environmental changes.
Seedlings were noted in all of the larger colonies in Maine and New Hampshire and in some of the smaller stands as well. Thirteen colonies (more than 68% ) had seedlings in contrast to the situation in New Jersey where Spencer reported seedlings in only 5 of his 36 stations (less than 14% ). Since Spencer's observations were made about 30 years ago we are led to speculate on the present condition of the New Jersey stands as a result of the tendency for temperatures in Northeastern United States to rise appreciably between the years 1930 and 1955. ⁷ Such temperature rise might easily further reduce areas suitable for seedling reproduction as our observations indicate that R. maximum seedling reproduction takes place only on mossy generally moist sites.
Tables IV and V. which compare the associated species of woody plants growing along with Rhododendron maximum in the 4 areas under consideration, point up certain similarities, notably the almost universal presence of Acer rubrum and the rather high frequency of Tsuga canadensis, Kalmia latifolia, Cornus florida and Quercus alba are the three other species that are present in more than half of the New Jersey stands and might therefore be regarded as being characteristic associates of rhododendron in that area. However C. florida is absent from all Maine and New Hampshire colonies, occupying instead only certain well drained and warm exposures often with Quercus velutina-- very different sites indeed from those of rhododendron. Mountain Laurel is similarly a disjunct species over most of its Maine and New Hampshire range; again it is adapted there to better drained habitats though occupying the same general range. Quercus alba no the other hand is a common forest tree in much of southern New Hampshire and southwestern Maine; yet it is met with in only 3 of the nineteen colonies; it too grows more commonly in drier places.
Of the remaining so-called "Common Species" Betula lutea is associated with rhododendron in 15 of our 19 stands though it is present in only 14 of the 36 New Jersey colonies and it is not mentioned by Griggs in Ohio nor is it present in the Virginian Colony studied by Iltis. Fagiis grandifolia seems to do a hit better being present in all 4 areas but occupying only 10 stands out of 36 in New Jersey and 11 of our 19. Of the other prevalent species in the Ohio, New Jersey and Virginia stands, Liriodendron tulipifera does not extend north of southern New England while Quercus Pri nus barely gets into southern New Hampshire and Maine.

Turning to the commonest Maine and New Hampshire associates, aside from those already mentioned, we find several species that would he found very rarely if at all in the other 3 areas. Such include Betula papyrifera, Pinus strobus, Viburnwn alnifolium, Picea rubens and Viburnum cassinoides.
In Braun's "Deciduous Forests of Eastern America" frequent mention is made of the types of forest communities in which Rhododendron maximum occurs, Tsuga canadensis being usually a dominant tree where rhododendron is found and Acer rubrum occurring commonly. The presence of other species of trees depends on the peculiarities of the habitat and the geographical area in which the rhododendrons are found. These also are the only two species that appear repeatedly in a majority of colonies under consideration in this paper. Of the 15 other commonest associated species of trees in Maine and New Hampshire 11 are absent in New Jersey, 11 in Ohio and 13 in the Virginian Coastal Plain, the remaining species having no high order of coincidence in these places either.
For local parts of the range of Rhododendron maximum there seems to be some usefulness in recognizing a characteristic rhododendron association, because the same group of associated species of plants is met with over and over again. But over the whole range the associates may change markedly; the nearly ubiquitous Tsuga may give way to Picea rubens at one extreme of climatic tolerance of rhododendron or to Chamaecyparis thyoi des at another.
Good ⁸ has stated that each species has its particular range of tolerance. As a corollary it may be assumed that no two species, unless they are mutually dependent, have precisely the same range of tolerance of environmental conditions. A consideration of the associated woody species growing with Rhododendron maximum in various parts of its range lends support to this contention. Obviously none of the associated species has precisely the same tolerance of environmental factors as R . maximum. Those that are most frequently associated with it over its entire range are those that presumably most closely approach it in tolerance. But many of the common associated species many any one part of the range have very different amplitudes of environmental tolerance from that of rhododendron and therefore will be absent from climatically different parts of the range. It seems preferable to interpret the facts of the Rhododendron maximum "association" in this way rather than to try to contrive a definite rhododendron association to embrace any considerable portion of the area that it occupies. On the other hand, it is to be expected that whenever environmental conditions are somewhat similar there will be essentially the same associates unless indeed these associates have had a different history and followed different migrational paths. It can be seen then that Tsuga canadensis and Acer rubrum have ranges of tolerance somewhat similar to that of R . maximum. But Betula lutea and other northern associated species have ranges of tolerance overlapping that of R. maximum only in the northern States.
We must conclude from this comparison of Rhododendron maximum colonies in 4 outlying parts of its range that:

1. There is no single physiographic situation to which the species is confined so long as an abundance of moisture is available.
2. There is no such thing as a predictable association of species with which Rhododendron maximum is constantly to be found.
3. Within any particular climatic zone the rhododendron association is usually composed of essentially the same dominant species and often occupies similar physiographic situations.
4. The suggestion is made that Good's concept of tolerance suggests a reasonable explanation of the observed diverse character of the rhododendron association .

¹ Hodgson, A, R, and Pike, R, "Recent Changes in Some Rhododendron Colonies in Maine and New Hampshire", Rhodora, 62:87-93, April 1960.
² Dorf, Erlig, Climatic Changes of the Past and Present, American Scientist, 48: 341-364, September 1960.
³ Iltis, Hugh H., Studies in Virginia Plants 11, Rhododendron maximum in the Virginia Coastal Plain and its Distribution in North America, Castanea, 21: 114-24, September 1956.
⁴ Spencer, Ernest L., "Natural distribution of Rhododendron maximum in New Jersey,", Bull. Torr. Bot. Club, 59: 401-14, 1932,
⁵ Griggs. R. F., A Botanical Survey of the Sugar Grove Region, Ohio State University Bull, 18(25): 273-75, 328, 1914.
⁶ Hodgson, A. R. and Pike, R., "Recent Changes in Some Rhododendron Colonies in Maine and New Hampshire.", Rhodora, 62: 87-93, April 1960.
⁷ Braun, E. L., Deciduous Forests of Eastern North America, Blakiston, 1950.
⁸ Good, R., A Theory of Plant Geography, New Phytologist, 30: 155, 1931.