PLMTS PUBLICATION 1088 JMBER I960 PRIME 30>f 9 /o^ o Research Branch CANADA DEPARTMENT OF AGRICULTURE C ORRECTIONS Title page: the date of publication was February 7, 1961. page 37, line 2; delete: 35; insert: 135 page 54, between lines 22 & 23; insert: PINACEAE page 74, line 5; delete: FUBETTE; insert FINETTE page 154, line 23; transfer: Map 408 to line 26, page 176, line 18; delete: Bailey), insert: Blanchard; R. canadensis var, pergratus (Blanchard) Bailey, Roland), page 176, line 19 & 20; delete: R, canadensis. . . .. .Roland). page 185, map 517, delete: var. linearis page 192, line 14; delete square brackets. page 194, line 5 from end; delete square brackets, page 210, line 28 page 220, line 22 page 234, line 11 page 236, line 13 page 254, line 23 delete comma from (DC.), Domin delete square brackets and apply to line 27, delete: southwestern; insert: southeastern delete: n, comb,, based; insert: Fern, ( remove square brackets; delete: if present; insert: West Prince NOMENCLATURAL INNOVATIONS page 190, Toxicodendron radicans (L.) Ktze9 var. rydbergii (Small) page 196, X Hudsonia intermedia (Peck) PUBLICATION 1088 DECEMBER 1960 THE PLANTS of PRINCE EDWARD ISLAND by David S. Erskine Plant Research Institute Research Branch CANADA DEPARTMENT OF AGRICULTURE Price $1.50 2M-27273- 12-60 89470-1 TO ORDER COPIES OF THIS PUBLICATION Make cheques or money orders payable to the Receiver General of Canada, enclose with your order and send to Publications Branch Queen's Printer, Ottawa. ROGER DUHAMEL, F.R.S.C. QUEEN'S PRINTER AND CONTROLLER OF STATIONERY OTTAWA, 1960 Price $1.50 Cat. No. A53-1088 CONTENTS Page INTRODUCTION 5 Procedure of Field Survey 5 History of Botanical Investigation 6 Acknowledgments 10 GENERAL FEATURES OF THE ISLAND AND ITS VEGETATION 11 Physical Features: climate, geology, soils, drainage 11 Vegetational Features: historical origin 14 VEGETATIONAL ASSOCIATIONS 15 Previous Knowledge of Vegetation 15 Associations 1 . Forest 17 2. Fresh-water communities 22 3. Bog communities 25 4. Maritime associations 26 5. Cultivated vegetation 29 PHYTOGEOGRAPHY 33 Absence and Presence 33 Local Distribution 35 Summarizing Statistics 37 REFERENCES 38 ANNOTATED LIST OF SPECIES 40 INDEX OF FAMILY NAMES 268 89470- VA THE PLANTS OF PRINCE EDWARD ISLAND INTRODUCTION An annotated list of the plants of Prince Edward Island, besides fulfilling a long-felt want, may serve many interests. Sources of information on the flora are at present limited to very incomplete, privately published lists without indication of the distribution of the species. "The Island," as inhabitants call the province, is a predominantly agricultural area; 87% of its land surface is farmland. Thus, the survey may serve the needs of those who want to know what wild plants are present that may serve as hosts to fungus or insect diseases of crop plants, or what noxious weeds are present that should be stamped out. At the same time, a survey of the plants serves the broader interests of plant geographers who wish to know what species occur and how far agriculture has modified the natural flora As only a dozen of the species present are not included in Roland's Flora of Nova Scotia (1945), and the Grasses of IS ova Scotia by Dore and Roland (1942), keys for identification and descriptive notes are superfluous. Victorin's Flore Laurentienne (1935) is also applicable to the area, but, however, it includes a smaller proportion of the Island's species. Comprehensive for the entire maritime area are the eighth edition of Gray's Manual of Botany (1950), and Gleason's edition of Britton and Brown's Illustrated Flora (1952), either of which must be resorted to for new additions to the flora, but which seldom give any indication of the presence or absence of a species from the Island itself. Procedure of Field Survey The survey was projected and initiated by the Botany Division, of the Canada Department of Agriculture, Ottawa, and carried out in the summers of 1952 and 1953 by the author under the guidance of W.G. Dore, and in 1953 with the field assistance of A.J. Smith of Macdonald College. A base of operations was set up at the Science Service Laboratory at Charlottetown, but limited transportation and the expanse of the Island made the establishment of field bases essential. These were set up for periods of two to seven days at towns and villages in the country: at Alberton, Hunter River, Wood Islands, Wellington, Souris (twice), Dalvay, O'Leary and Brackley Beach in the first year; at Tignish (twice), O'Leary (twice), Bideford, Wellington, DeSable, Sea View, Hunter River, Cavendish, Brackley Beach, Souris (twice) and Murray River, the second. It was planned as far as possible to cover each apparently significant region twice at different parts of the season. Although walking is the only means of progression permitting collection of plants, bus, train and car were used to expand the radius of exploration at each center; all-day trips from Charlottetown through the country or to a given locality often proved very profitable. On car or train, a list of sight records was kept in the manner described by Groh (1927) for his weed survey. At first, collecting of specimens took precedence; our objective was to obtain a complete set of the Island species in triplicate. As this phase approached completion, sight records and novelties became more important, the former filling in the natural distribution pattern of common species in the Island, the latter adding rare species. Collected specimens were made to conform to the size of the standard herbarium sheet, 17 by 11 inches. Plants of a herbaceous nature were excavated to show underground structures. The specimens were dried in newsprint folders between blotters in a plant press aerated by corrugates and set over infra-red bulbs on a collapsible aluminum support. This device did a thorough job of drying in 24 hours in all but the most succulent plants. Records as to locality, habitat, distinctive features and date of collection of the specimens were kept on numbered field labels. The sight records were kept by locality and transferred to taxonomic order by using initials of locality against a checklist. On the distribution maps, collections were indicated by solid discs, sight records by open circles. The prepared specimens resulting from the work of the survey, some 1800 collections in all, were distributed in sets as follows: first or "master" set, retained for preservation in the herbarium of the Botany Division, Ottawa; second set, to Acadia University, Wolfville, Nova Scotia.; third, fourth and fifth sets, less complete, to the University of Montreal, the British Museum of Natural History and the New York Botanic Garden. Small maps similar to those in this publication were reproduced photographically in five copies and affixed to the first five sets of specimens. A representative set (one of each species) was placed in the Science Service Laboratory, Charlottetown. Excess replicates have been dis- tributed variously. Previously, knowledge of the flora of Prince Edward Island was summarized from three sources, herbarium specimens, published lists and conversations with the local amateurs of botany. All herbaria of significance were consulted. The information, kept in loose-leaf files, recorded all data on locality, habitat and collection. These loose-leaf files are being preserved at Ottawa. The list at the end of the historical section presents the distribution of Prince Edward Island specimens in herbaria. History of Botanical Investigations The development of botanical knowledge of the Island has taken twc separate courses: that acquired by botanists of universities and museums outside the province and that by local amateurs. It began late and developed haltingly. The first native worker in botany was the general naturalist Francis Bain of York Point near Charlottetown, better known as a geologist and author of The Natural History of Prince Edward Island (1890), a minute school book which treats the plants in 44 pages, family by family, with occasional details of occurrence and habitat for conspicuous species and with entries such as "five asters" for less easily distinguished species. To discover which "five asters" were intended, one has to turn to the privately published list of the Island flora by John MacSwain and Francis Bain (1891). This list included 430 vascular plants. A supplement was printed in 1892 and another in 1894, the year of Bain's death. In 1888, John Macoun, Botanist to the Geological Survey, made the first major plant collection from the Island while based at Brackley Beach with his honeymooning son. He collected at Tignish, Charlottetown, Mount Stewart and Tracadie, East Point and Lake Verde, concentrating however at Brackley Point. The specimens are now at the National Museum, Ottawa. The published records of this collection are included in Macoun's Catalogue of Canadian Plants, Vol. 5, 1890, with the pteridophytes treated by Burgess who credited Bain with the Island records. The son, James M. Macoun, published additional notes on some of these specimens in the Canadian Journal of Science (1895). A few notable collections were made for the National Museum by the geologist W.J. Wilson, as assistant to Robert Chalmers on the first Pleistocene geology survey of the Island in 1892-3. The next chapter of exploration is delineated by the Island botanist Lawrence W. Watson, of Charlottetown. The interest in violets arising from the species-splitting of the American botanist E.L. Greene spread to Ottawa, to the younger Macoun and to James Fletcher of the Department of Agriculture, and so to Watson on whom they called for Maritime material. Live material of Viola adunca supplied to the experimental student of violets, Ezra Brainerd in Vermont, by Watson became the type of V. adunca var. glabra; an odd white form of a blue violet was named V. Watsoni by Greene, and the V. incognita of the Island was named V. nesiotica by Greene, all names since discarded. In 1901, the visit of the amateur J.R. Churchill of Massachusetts to Summerside and Tracadie included a collecting trip with Watson and added several species to the herbarium of the New England Botanical Club. In 1904 Fletcher stopped briefly in Charlottetown to collect a few violets. For two summers (1902 and 1903) Watson was employed by the National Museum as a collector of fossils and plants in the Island, but his projected flora of the Island was never prepared and his collections sent to Macoun for identification were discarded. Letters containing some of Macoun's replies are in the possession of H.A. Messervy of Charlottetown. Watson died about 1908. About this time, the vogue of Spotton's High School Botany in Toronto led to its adaptation for Island use by MacSwain (1907). Descriptions were taken over, plants not known from the Island were dropped from the MacSwain and Bain list and the few added by Watson included, to make a net total of 430 vascuiar plants. Copies of MacSwain' s list were kept and annotated by the amateurs who later were to revise it. Meanwhile, the rising tide of amateur botany in New England produced its synthesist in M.L. Fernald of Harvard University. The theory of persistence of plants through the later Ice Ages in the Gulf of St. Lawrence area on unglaciated "nunataks" or refugia was already forming in his mind when Robert Chalmers' conclusion that at least central Prince Edward Island had escaped glaciation drew his attention to this province. In 1912 and 1914 with Harold St. John as assistant and, for a brief space in 1912 with Long and Bartram of Philadelphia as companions, he made the most extensive collection for the Island, 713 duplicate specimens of which were deposited at the National Museum. It seems likely that the Island rather failed to fulfill Fernald's hopes of finding relict preglacial elements, for apart from incidental notes on certain coastal or marsh- dwelling species, no account of the trip was published. Thus, until the eighth edition of Gray's Manual appeared (1950), his finds were unknown to Island botanists. Blythe Hurst, the sage of Smelt Creek near Brackley Point, was the next naturalist of note. To him as the columnist Agricola of the Charlottetown Guarcfo'arc came the notes of the younger collectors: first, H.A. Messervy of Charlottetown, whose travels covered most of the Island, and R.R. Hurst, plant pathologist in charge of the Science Service Laboratory, recording many new introductions, then others. The MacSwain list, shorn of all descriptions and records of occurrence, was revised by Blythe Hurst to include these additions and "corrected" according to the nomenclature of Gray's Manual, seventh edition of 1907, as "A New Flora of Prince Edward Island", published in the Transactions of the Royal Canadian Institute (1933). Of particular significance to the Island botanists was the weed survey of the Island made by Herbert Groh of the federal Department of Agriculture in 1926. His survey, published in Scientific Agriculture for 1927, listed from sight records the occurrence of introduced, weedy or poisonous plants, by the number out of 55 Lots visited in which he had seen them. Some 200 specimens were added to the herbarium of the Department of Agriculture, Ottawa. At the same time, a visit from M.O. Malte, the botanist of the National Museum, added only specimens. Malte, like Fernald, projected a flora of the Gulf of St. Lawrence — Maritime region, but both abandoned them. John Adams of the Division of Botany, Ottawa, published in 1937 an article with additions to Hurst's 1933 list, increased by the sedges (records mostly taken from Mackenzie's monograph) and a few grasses. These groups remained the weakest links in Hurst's privately printed 1940 revision of his own list. He supplemented this last "New Flora" in 1941, and listed other new discoveries in his newspaper column, "Newsy Notes by Agricola," down to 1946. Blythe Hurst died in 1951. Specimens collected by Blythe Hurst, R.R. Hurst and Messervy which had not made their way to the Department of Agriculture herbarium at Ottawa were destroyed by the fire that razed the Science Service Laboratory at Charlottetown in that year. A revision, including as new these records and the results of a week's weed survey by J. Bassett in 1950, and with nomenclature corrected by the eighth edition of Gray's Manual (1950) was mimeographed by the Science Service Laboratory in 1952, and is referred to as "Campbell 1952" in the synonymy. In 1945, first Dore and Gorham in June collected 122 specimens in the Island during the course of a botanical investigation of pastures through the Maritimes, then in August Dore and Roland spent three intense days gathering 338 unicates in the eastern part of the province. Only six of these latter, kept 8 separate, escaped the fire which destroyed Roland's herbarium at Truro soon afterwards. Perhaps the impression gained of the inadequacy of Hurst's list coupled with this stroke of fate prompted the conception of the present project. Then Acadia University at Wolfville received Warren's early herbarium (1902-6) from North River built up with encouragement of Watson. Under E.C. Smith's curatorship, Acadia continued to receive specimens: of McGowan from Kilmuir, of Bruce from Heatherdale, of Aitken from Bay Fortune, etc. Some data on these and other preserved collections and their disposition in herbaria may be tabulated as follows: Adams, John (1936), Summerside and Brackley Beach; basis for additions to Hurst's 1933 list (DAO) 34 Aitken, H.E. (1950), Bay Fortune, Kings County, a student collection (AU) 33 Anderson, E.G. (1951), Charlottetown and Summerside; weed collection (DAO) 12 Bassett, J. (1950), a week's wide-ranging ragweed survey and general collection (DAO) 186 Blanchard, W.H. (1909), materials for a study of the Blackberries (Can) 7 Bruce, J.M. (1948-9), a general collection from southern Kings County (AU) 129 Campbell, J.E. (1952-3). additions to the Floristic Survey (DAO) 28 Campbell, Sterling (1938), wild plants (private collection) 18 Churchill, J.R. (1901), Summerside and Tracadie, general, mainly at Harvard University (DAO, Can) 7 Dore, W.G. (1945), two general collections, one of 122 specimens with E. Gorham, one of 338 with A.E. Roland, all but 6 of the latter lost by fire (DAO, Dal, AU, etc.) 128 Eastham, J.W. (1912), near Orwell, Queens County (DAO) 3 Erskine, D.S. (1951 to 1954, 1956), two large general collections, some of the first with W.G. Dore, most of the latter with A.J. Smith, the basis of the present report (DAO, AU, Mtl. NY, BM, PEI, NSMS, etc.) 1847 Erskine, J.S. (1955), two-day general collection (NSMS) 36 Fernald, M.L. (1912, 1914), two large general collections made with H. St. John and for part of 1912 with B. Long, the major part at Harvard University, many cited specimens (Can) 713 Fletcher, James (1904), casual collection made at Charlottetown, Brackley Beach and Summerside on official Trip (DAO) 13 Fyles, Faith (1915), small set of weedy or poisonous plants, mostly from Summerside as part of Government project (DAO) 24 Groh, Herbert (1926), 56 weed-survey collections and (1929, 1930, 1932, 1934, 1937, 1940) 24 small casual additions (DAO) 80 Haviland, Margaret (nee Grubbe) (1853), the first collection, casual (Kew) 30 Hurst, Blythe (1926-44), occasional specimens sent for determination (DAO) 21 Hurst, R.R. (1928-45), occasional specimens sent for determination (DAO) 9 Kerr, Robert (1949), student collection at Ellerslie (AU) 21 Leard, G.A. (1951), specimens from around Souris sent for identification (DAO) 7 Macoun, John (1888), first major general collection (Can) 395 Malte, M.O. (1°26), general collection made in preparation for flora of the Maritimes (Can) 99 McGowan, L.J. (1950), student collection from Kilmuir area (AU) 20 Messervy, H.A. (1938), orchids sent to Ottawa (DAO) 6 Perry, H.G. (1919), incidental to Fisheries work at large (AU) 6 Power, E.E. (1950), York and Bideford, incidental to nematode work (DAO) 26 Scott, R.V. (1948), student collection from Ellerslie, incidental to fisheries work (AU) 19 Smith, M.W. (1940, 1953), mainly aquatics, in preparation for fisheries ecology (DAO) 33 Taylor, A.R.A. (1948), small general teaching collection, mainly from near Ellerslie, incidental to fisheries work (UNB) 55 Warren, A. Emerson (1911), New Glasgow area (DAO) 7 Warren, G.C. (1902-06), North River, Queens County general herbarium (AU) 136 Watson, L.W. (1901-04), Charlottetown area and violets (Can) 9 Wilson, W.J. (1893), small set incidental to geological work (Can) 5 Miscellaneous: collected by visiting firemen, Science Service employees, or sent in for advice by farmers (DAO) 39 Total specimens available in Canadian herbaria 4241 Acknowledgments The list of acknowledgments can never be complete, but for the kindness of all those who helped in the project the author extends many thanks. To the systematic staff of the Botany Division at Ottawa thanks for the assistance with taxonomic questions, and to Doctors H.A. Senn and W.G. Dore for the continuance of the project when it seemed most unlikely to reach fruition (1954); to the staff of the National Museum for allowing me to catalogue and examine their set of Fernald's and Macoun' s Prince Edward Island specimens at weekends and Christmas in 1952; to Dr. Smith at Acadia and Dr. Taylor at the University of New Brunswick for access to their herbaria; and to Doctors Sampson and Wood for looking up records at Harvard for me. To the staff of the Science Service Laboratory of Charlottetown, most of all, thanks for their interest in the work, 10 such that there was no one who did not at one time or another assist in trans- porting us to the field, and for the use of their facilities, in particular to Mr. Hurst (whose photographs illustrate the report) and Mr. Campbell on whom the previous burden fell heaviest. To Mr. H.A. Messervy of Charlottetown, who gave me access to all his manuscript notes and to Watson's letters from Macoun, and drove us on several of our collecting trips; to Blythe Hurst, Jr., for giving me access to his father's manuscripts; the late Dr. G.C. Warren of Sydney, Nova Scotia, for the use of his MacSwain-Spotton with the MacSwain & Bain list, and for helpful correspondence; thanks. To Mr. G.A. Leard of Souris, Mr. & Mrs. Alton Webb of O'Leary, Mr. Sterling Campbell of Cavendish who drove and guided us to good localities and rare plants; to Mr. R. Found of the Fisheries Station at Ellerslie, Mr. V. Henderson of the Entomology Division, Mr. W. Profitt, the Assistant Provincial Forester, for assistance with transportation and exploration, many thanks. For the general guidance of the project, the preparation of the plot maps, selection of photographs and reading of the manuscript, Dr. Dore is deserving of much credit. GENERAL FEATURES OF THE ISLAND AND ITS VEGETATION Physical Features Prince Edward Island, a crescent-shaped island lying to the south of the Gulf of St. Lawrence, has a length of about 145 miles and an area of about 2100 square miles, and lies about 9 miles from New Brunswick and 14 from Nova Scotia at the nearest points. Its situation in latitudes from 45° 57'N to 47° gives a cool temperate climate with freezing winters and moderately warm summers, means ranging from 17°F for February to 67°F for July at the Experimental Farm, Charlottetown. Proximity to the sea modifies temperatures from the extremes of New Brunswick, but less so than if Nova Scotia were not interposed between it and open ocean. Drift ice accumulating in the Gulf circulates through the Northumberland Strait and, more especially, around the northern shores of the Island before finding its exit at Cabot Strait. This cooling delays the spring; the average date of the last frost at Charlottetown is May 13 but there is variation of as much as a fortnight either way. However, the sea also delays autumn frost till mid-October (average), giving the Island the longest frost-free season (180 days) of the Maritimes except the southwestern parts of Nova Scotia. Locally, climate varies from being more maritime in the east to more continental in the west. Summer heat is higher in the west. Rainfall (including snowfall reduced to equivalent rainfall) is fairly even throughout the year, least in spring and heaviest in fall (from 2.8 inches in April to 4.6 inches in December). The average precipitation of 43 inches is sufficient to class the climate as humid, but above-average summer temperatures make for some summer drought. Snow falls throughout the frost season, lying from the beginning of December to late March, but more or less complete thaws in January are of almost annual occurrence. 11 Geologically, the Island consists of a more or less uniform Permo- Carboniferous bedrock, slightly folded, eroded and overlain by deposits of Pleistocene (Ice) age which form the basis of the present soils. The bedrock is predominantly sandstone, but with interbedded shales and a very few small pockets of limestone (Park Corner, Crown Point, Miminegash). The shales are most conspicuous where the sandstone of a fold-produced dome has been eroded away to form Hillsborough Bay, so that a sandstone escarpment from Nine Mile Creek to Point Prim surrounds the shales of greater age. Many of the springs in the uplands are located at the contact of sandstone with impervious clay. A small Trias sic igneous rock area occurs at Hog Island off Malpeque. Pleistocene ice, though it hardly shaped the form of Prince Edward Island, brought foreign clay and stones, distributed local material into new forms and disrupted drainage. As the historical geology of this period in the Island is now being studied intensively by the Canadian Geological Survey, only tentative hypotheses accounting for the changes wrought by ice may be made. Ice from New Brunswick brought some calcareous and clay till from far beyond the Northumber- land Strait; ice possibly from Cape Breton and of late Wisconsin age may be responsible for foreign material in the till of the eastern coast. The major ice movement must have been from the west onto the Island; following the long axis of the Island, it would rapidly have become heavy with the soft easily eroded sandstone and shale of the Island, so that most till east of the Northumberland Strait slope of western Prince County is sandy and of local origin. Western Prince was part of an area depressed by the weight of ice which became submerged by the sea to heights of 75 feet above present sea level between the time melting ice left it bare and the time the earth's crust resumed its normal level. The present uplands there are thus flanked by terraces, often of beach gravels, while former barrier-beach ponds have become present day bogs (e.g. the large East Bideford Bog and the "Black Marsh" at North Point). Submerged peat of postglacial age is found along the southeastern coast at localities from Charlottetown round to Launching, and with the submerged forest stumps of the Queens Clay coast, indicates that for a time after the retreat of the ice the Strait was at least narrower and shallower by some 20 feet than at present. Assuming that the depression of western Prince by ice was accompanied by a corresponding rise in the eastern end of the island, there may have been a land bridge to Nova Scotia. In the section on phytogeography, we conclude that it perhaps had very slight significance in the formation of the present flora. The soils of the Island have developed since the Ice Age on these diverse geological types. On the Central Uplands and their outliers, deeper, more finely grained soils have been formed, the Alberry fine sandy loam and the Charlottetown fine sandy loam. The New Brunswick clayey till produced the O'Leary clay loam; the exposed clays of the Hillsborough Bay area, the Queens clay. On the kames and eskers, dry sandy soils developed (Dunstaffnage sandy loam), and on the postglacial beaches of western Prince, the Kildare sandy loam. Where the Cape Breton lobe scraped off the old weathered layer, the dry thin Culloden sandy 12 loam developed. The poorly drained phases occupy the only other large areas, the Egmont clay loam corresponding to the O'Leary, the Armedale to the Charlotte- town and the sand loams. SOILS - CULLODEN SANDY LOAM - DUNSTAFFNAGE SANDY LOAM - KILDARE SANDY LOAM - ARMEDALE SANDY LOAM »-E+ - EGMONT CLAY LOAM Generalized soil map of Prince Edward Island (Based on Whiteside, 1950). Drainage, controlling as it does both the type of soil and the types of plant cover developed at each site, is of first importance as an ecological factor. The Island has low summits of under 500 feet above sea level, but short slopes to the sea or rivers give most areas adequate drainage. Thus we may with reason speak of a Central Upland although its summits south of Fredericton lie at 480 feet at most. From this center the highest divides lie in a line almost parallel to the Prince-Queens county line from Park Corner to Bonshaw. These are the well- drained hardwood hills. As the long slopes, consequently, lie to the east and west, it is on the lower broader divides running in these directions, that the nearest small swamps or bogs occur. East of the Covehead Road the upland shades off altogether at heights of 100 to 125 feet. South of Montague River and Orwell Bay, hilly country rises to an upland with summits at 400 feet near Iona and 425 at Caledonia; there are short slopes to the south and west, longer ones toward the north and east. A fringe of lowland surrounds the hills; extensive swamps parallel Murray River, both to north and south. East of the Hillsborough River the low sandy lands are cut off from the sea by the Pownal escarpment which, with heights of 150 to 300 feet, diverts drainage to the long slopes toward the river. The lakes of this area are scarcely 13 drained at all, their waters filtering through swamps toward the headwaters of creeks tributary to the Hillsborough. From Fort Augustus and Pisguid eastward, the hills form a belt subdivided by the valleys of northward-draining rivers, while from Souris to Boughton River or even Montague River, the only well- drained land is on the seaward ends of the southern peninsulas separated by southeastward-draining valleys. Thus, within these fringes of hills, there are extensive swamps and large poorly-drained areas. The way in which the southeastern-directed valleys are the site of the Dunstaffnage kames and eskers suggests that drainage toward this side may have been impeded by glacial deposits. The large northeastern peninsula resembles the smaller southeastward ones in having better drained areas nearer the tip and large swamps on the gradual slopes of its broader base. In western Prince (i.e., west of Summerside and the Miscouche isthmus) two NE-SW trending divides are the main feature; one, the backbone of West Prince (i.e., west of Portage), lies near and parallel to the northwest coast and reaches heights of 175 feet; the other, reaching only 125 feet, lies transversely across Central Prince from Abrams Village to Bideford. Central Prince north of this ridge includes the largest area of poorly drained Egmont soils, the second large area of the Egmont lying south of it toward the Miscouche isthmus. In western Prince the lower summits make for poor drainage except where the seaward ends of valleys have cut ravines. Lesser streams have often cut down fairly easily into the postglacial clays found below 75 feet, becoming wide, slow, and bordered by an alluvial flat suitable for alder thicket. Eastward, stonier alluvium is found along the larger north-flowing streams (Hunter River, Winter River, Indian River). Coastlines in the Island appear widely different, the North Shore with its wide crescentic sweep and great bays all barred by sand islands built by the Gulf currents, the South Shore with its three great southwest-facing bays of muddy shores. The east coast with its many estuaries tending to be barred by sand spits is an open Gulf coast like the North, though more sheltered. Any appreciable depth of sand along the shore tends to become windblown into dunes, usually by onshore winds and so parallel to the coast, but also by longshore winds that cross-pile them. Vegetational Features The vegetational history of the Island is, at best, hypothetical until analyses may be made of peat and lake-sediment samples for fossil pollen. However, the changing distribution of land and sea areas and the fluctuations of the climate of eastern North America have been the most significant factors, and the waning of the fourth (Wisconsin) ice sheet the point of departure. In Gaspe and western Newfoundland, mountains may have stood above the ice at its maximum extent; but on the Island, with its greatest relief only 500 feet, nunataks cannot have existed. The possibility of plants having survived the glaciation on the Island depends on whether the Central Upland really was 14 unglaciated. On the Island occur none of those species whose characteristic distribution suggests localized survival. Secondly, Hulten propounded a hypo- thesis that the continental shelf, bare during glaciation, has served as a refugium during the Ice Age. If so, the Laurentian (Gulf) endemics of coastal habitat may have reached the Island thence as the ice melted away and the sea advanced. As the ice accumulated, it depressed the continent beneath it. As it shrank through melting, the continent returned to its normal level strip by strip, each released along what is called a 'hinge-line.' One hinge lay across Prince Edward Island: consequently as the ice melted back toward New Brunswick, the sea temporarily inundated the depressed portions of Prince County to as much as 75 feet above present sea level. At the same time, by reason of a contemporaneous hinge, the sea covered southern New Brunswick and Maine. But with so much of the world's water in the form of ice, the continental shelf was extensive above sea level, and a migration route along the Atlantic Coastal Plain brought southern plants to Nova Scotia. Here too, the effect on the flora of the Island has been very slight, but a few of the bog plants must have used this route. The relatively fresh waters of this postglacial sea gave its coasts as a migration route to estuarine plants: only one (Samolus) occurs of those discussed in this connection by Fassett. However, a few of the coastal plants which entered the Great Lakes along postglacial shores must have arrived at the Island about this time. Interest- ingly enough, two of the salt marsh plants of the Bay of Fundy reach their northern limits in Bedeque and Hillsborough Bays, suggesting migration by such a continuous inner coastal route. The forests of Prince Edward Island, part of the 'Acadian Forest' of the present, seem to have arrived in two waves of invasion. The Boreal element is general and well represented: predominantly of conifers, white and black spruce, jack pine, larch, and balsam fir, with white birch and mountain ash, it is the only forest of the North Shore and the bogs. It may have had 7000 years to cross the waters of the postglacial sea and its shrunken remnants, or have entered by land from Nova Scotia. At least its representation in the Island is rather complete. The northern deciduous forest element, represented by the "northern hardwoods" (beech, sugar maple, yellow birch) and by white pine, hemlock and red spruce, which forms the climax forest of the uplands, shows a full quota of tree species but very little of the herbaceous "spring flora" so characteristic of the mainland hardwoods. This reached its greatest extent during the 'thermal maximum' about 3000 years ago, almost certainly after the Island had become isolated by Nor- thumberland Strait. VEGETATIONAL ASSOCIATIONS Previous Knowledge of Vegetation Our source of information regarding the nature of the vegetation before 1800 is confined mainly to the kind of notes made by travelers. Since 1820, visitors to Prince Edward Island have seldom felt that their observations could 15 add anything to our knowledge of its plant cover. The early French records are presented by Harvey (1926). Jacques Cartier, the discoverer, in 1534 found the "treeless lands'7 of the North Shore covered by "pease. . . and wild oats like rye, one would say sown and tilled," the Lathyrus japonic us and Elymus mollis of sandy beaches, by "white [?J and red gooseberry bushes, strawberries, raspberries," Ribes hirtellum, Fragaria >vir- giniana and Rubus idaeus ssp. strigosus indicating an early stage in succession on cleared or burnt land. The forest was composed of "cedars, yew trees [balsam fir?], pines, white elms, ash trees, willows, others unknown to us [Europeans]. " This association of Thuja occidentalis, Abies balsamea (?), Pinus strobusy Ulmus americana, Fraxinus americana, Salix spp., marks his landing as in western Prince. Nicolas Denys in the middle seventeenth century regarded "firs" as dominant, as conifers are on most of the lowlands, and remarked on the presence of beech and birch. These latter, Fagus grandifolia and probably its associate Betula lutea, are part of the characteristic upland vegetation. Gotteville, founder in 1720 of Fort LaJoie near Rocky Point, noted the local association of oak," cherry, beech and pines of mast size" (Quercus rubra var. borealis, Prunus sp. , Fagus, Pinus strobus), which may be found even now in the woods on the grounds at Riverside near Charlottetown. Quercus rubra was known from the North Shore as large groves near Tracddie and mixed with Pinus strobus at Malpeque. Gaudet (1956) summarizes some later observations beginning with that of Rochon (1759). Rochon noted a large grove of cedar between Malpeque and Cascumpeque, made up of "two kinds, red and white" — a valid distinction among spruce and pine, common though invalid in hemlock, but locally unheard of in cedar. Probably he referred to the spruce, since he maintained that the "incense" (gum) of the red was chewed among the Acadians. Lord Selkirk in 1792 distinguished the hardwood and the pine communities, too: "The most common species of timber are beech and maple, . . . frequently intermixed [with] birch of different kinds, spruce, firs and other species of the pine tribe," but "in some places, the pines entirely predominate, this . . . indicating a soil of an inferior quality." Johnstone (1820) remarked that Prince Edward Island was "one entire forest of wood," listed the prevalent hardwoods (with elm included, oddly enough) and softwoods, and noted that "promiscuous" mixtures of hard and softwood were to be found in places while in other parts clumps of a particular kind were found by themselves. Major Pollard (1898) spoke of virgin forest of evergreen "firs" mixed with oak, birch, maple, ash and poplar, and of majestic pine and hemlock growing amid an undergrowth of hazel, alder, aspen, juniper (i.e. larch), cedar, and tangled brambles. Stewart (1806), classifying land by its vegetation type as to degree of suitability for settlement, generalized the ecology of the central upland and its eastern borders: 16 €t (« (( on the best land, maple [Acer saccharum. A. rubrum], beech iFagus], black and yellow birch one species, [Bet ula luteal, mixed with firs [Picea rubens, Abies balsameaj and pine IPinus strobus], with an undergrowth of yew [Taxus canadensis]99 . on the next best land, no evergreens mixed with hardwoods, and more yew". on the third grade, poor land, with a thin upper stratum and the sub-soil cold and hard . . . the hardwoods are not mixed with evergreens and there is no yew". "the worst land is covered with spruce [.Picea glauca, P. mariana], small white birch [Betula populifolial and scrubby pines". "swamps are covered with black spruce [Picea mariana]99. From relics of the original vegetation, it would appear that the best land lay on the gentler slopes of the central upland, the next best on the steeper slopes, north slopes and ravines, and the third undoubtedly on the dry hilltops. The poor land comprised the sandy parts of the valleys and lowlands near the shore, the morainic lands of Tracadie, and perhaps much of the east-central lowland (e.g., the pine groves of Mermaid). Associations From the vestiges of forest in the woodlots, swamps and ravines, the virgin forest sketchily described in the records gathered by Harvey and Gaudet and in Stewart's land classification may be reconstructed. (A more uniform picture is thus arrived at than today's patchwork remnants permit.) Except for dunes, salt marshes and bogs, the land was entirely forested. 7. Forest. On the uplands the dominant forest, usually of mixed forest aspect, was of the "northern hardwoods" (beech, sugar maple and yellow birch) and their coniferous associates (Braun, 1950), white pine and hemlock, their pro- portions varying with the nature of the site. Red spruce, considered by Halliday (1937) the unique species of the Acadian Forest Region which covered the Maritimes, made a poor third to the other upland conifers. Halliday' s general- ization (after Macoun) by the words "in spite of the flat topography and low elevation, maple occurs generally throughout, although in other parts of the [Central Acadian] section [including southern New Brunswick and the Gulf and Fundy slopes of Nova Scotia] it is confined to higher and better-drained positions" fails to appreciate Island topography. The differentiation of the upland forest by site, already indicated roughly by Stewart's first three land types, corresponds closely to that found by Long (1952) in the western portion of the Central Acadian Section, near Fredericton, N.B. Pure stands of beech (Stewart's third type) dominated the dry hilltops of the central and southeastern uplands; sugar maple with some yellow birch pre- dominated on the flatter summit areas of the north-eastern peninsula and the low uplands of western Prince, occurring also in the central uplands on the more gradual slopes. On gentle north-facing slopes or in shady ravines, red 17 89470-2 maple and hemlock predominated, particularly in the central and southeastern uplands; red maple alone or with yellow birch grew on the almost imperceptible O'Leary clay slopes. On the Island red spruce is less characteristic of low sites, probably because it here approaches its northern summer-temperature limit. Typically, the ground cover of such sites is sparse and a shrub layer almost lacking. The striped maple (Acer pensylvanicum) and mountain maple (A. spicatum), small and straggly but essentially trees, form a subsidiary element in the stands dominated by sugar maple. The herbs are mostly spring-flora perennials: Smilacina racemosa, Ranunculus abortivus, Aralia nudicaulis, Trillium undulatum, T. cernuum, Medeola virginiana, Trientalis borealis, and such localized species as Claytonia caroliniana, Viola pensylvanica, Osmorhiza claytonii, Panax trifolius, semi-saprophytes like Listera convallarioides , Pyrola elliptica, sapro- phytes like Monotropa uni flora and Corallorhiza maculata and parasites like Epifagus virginiana. Where hemlock was abundant a few saprophytes alone grew; under red spruce woods, the mossy floor would harbor Oxalis montana, Viola incognita, and various Pyroleae (Chimaphila, Orthilia). On the lowlands a wider range of sites made for a more heterogeneous forest. Deviating least from the hardwood forest is that of the stream-valley clays in western Prince. Dominated by red maple, now mainly in alders and always with the occasional elm, and interstitial species, elder (Sambucus canadensis) and less invariably, Cle*natis virginiana and Viburnum trilobum, with Salix lucida around open wet s puces, it corresponds closely to the Acereto-ulmetum (maple-elm community) of the upper clay plain of postglacial terraces in Kamouraska County, Que., described by Hamel (1955). Along streams cedar may form pure stands, usually with a limited undergrowth of herbs: Viola spp. most typical. White and black ash are also scattered along streams here, where the tall herbs Eupatorium maculatum, Aster puniceus and A. umbellatus, Urtica spp. and Geum laciniatum grow. Without ash, the same association is found along the larger rivers of the north slope, far to the east of Prince. The herbaceous plants of the red maple forest vary with the wetness of the site: with poor drainage, the ferns Onoclea sensibilis and Osmunda cinnamomea with scattered Rubus pubescens are most common; with better but still damp sites, Aralia nudicaulis, Thelypteris noveboracensis, Dryopteris spinulosa, Aster acuminatus and Carex de bills var. rudgei are more typical. Red maple forest without elm is characteristic of the depressions of the sandy east-central lowland. The abundance of glacial and marine sands in the lowlands makes for a contrasting forest, probably then dominated by white pine or white pine and oak, associated with red pine and wire birch in the most gravelly sites. Where these sands are little above the water table, the most extensive cedar woods of west Prince occur. The undergrowth of the pine woods consists of widely scattered Pyroleae, but under wire birch (Betula populifolia), the shrub Comptonia peregrina, the goldenrods Solidago puberula and S. nemoralis, and the shrubby Ericaceae 18 are admitted. Black spruce (Picea mariana), an occasional element in these low dry sites, dominated the true swamps of the lowlands almost exclusively, save for larches (Larix laricina) where they bordered on bogs. Along the North Shore, exposed sites and older fixed dunes allowed white spruce (P. glauca) forest to hold its own. Forest changes. The major change in the forest since settlement began has been the wholesale clearing of land for crops and pasture. As well-drained fine sandy loam soils were most favorable to agriculture, the uplands were cleared more completely than the lowlands; some 72% of the central upland is cleared at present In the west, too, the most extensive clearing has been in these areas, though the large areas of poorly drained soil reduce the average cleared area to 50%. A somewhat similar situation exists in the southeast, where the fringe of lowlands is less completely cleared but the upland soil is less favorable. Here again some 50% is cleared and in use. In the northeast, because of the large swamps the highest proportion of woodland remains; some 40% is cleared. By the combined average, 60% of the Island is devoted to agriculture, an excep- tionally high proportion for Canada east of Quebec City. The remaining 40% is by no means all in woods, for some 8% of the total is in unimproved waste land, marsh or barren. The result has been the reduction of hardwood-pine-hemlock forest to a greater extent than black spruce and red maple, which form the most extensive forest areas. Economic demand has made for selective cutting of much of the woodland that remains. In the 18th century all visitors noted that contemporary /7defence resource1', the mast timber of white pines. This was cut and exported to England; later white pine and oak were used in shipbuilding, and the former in woodworking. White pine stumps rotted from the back pastures within the last 70 years in many areas. The great size of the hemlock made it useful for boards, often only for barn siding because of its relative weakness, and its bark was sought for tanning. Sugar maple, used till fifty years ago as a source of sugar, was often maintained in a grove near the farmhouse, especially in the west; most of the birch from these groves was undoubtedly removed for fuel, though the paper birch has often replaced it. During the past twenty years, the demand for pit props and pulp wood has led to the cutting of much second-growth white spruce in Kings County. One of the earliest forces released by the settler was fire. About 1738, when the North Shore was only a base for drying fish, fire burnt from Tracadie eastward to East Point, devastating the entire north-eastern peninsula. It must have spread easily in the dry spruce and pine forest. One result of this burning was the encouragement of the heath shrubs; the resulting blueberry barrens of Tracadie have been perpetuated by accidental and deliberate use of fire ever since. The fire of 1840 in central Prince produced a great blueberry barren at Conway which persisted for over thirty years. (This the biggest fire of Prince County history burned from Campbellton to Bidef ord. ) The opening up of the hardwood stands by selective cutting has brought 19 89470- 2Vi disease to several forest trees. Yellow birch and white birch (as a member of the forest) have been devastated by dieback. The beech of hilltops has often suffered from the recently (1934) introduced aphid-borne canker (Nectria coccinea) so badly that only low thickets of stump-sprout trees persist; however, there are many more undamaged stands than in Nova Scotia. Very little white pine remained to suffer from or spread the blister rust when it struck early in this century. The larch of bogs, a species apt to grow in extensive pure stands, was also reduced by an introduced insect, the sawfly, in the 1890's. However, the role of native diseases in the case of successional species such as the larch, with their large pure stands, is probably less permanently significant; destruction and recovery may take place cyclically, as it seems to with the marine1 eel grass Zostera attacked by wasting disease. Forest succession. New plant communities (modifications of old ones) have been brought about by human intervention in the forest. Fire was not an unknown thing in 1534; even the discoverer Cartier found the community of raspberries and gooseberries in "treeless lands", suggesting fires set by lightning or Indians. And there were plants adapted to survival of fire: as in Nova Scotia (Martin, 1955) the perennials with deep rhizomes, the bracken and sheep laurel, sprout following a fire in the same summer. Other heaths such as the blueberries are close behind; the fireweed (Epilobium angustifolium) with its cork-protected rhizome flourishes in this community. The seeds which drift into the bare spaces are usually grasses such as Agrostis scabra and Danthonia spicata, or fleshy fruits carried by birds: raspberries (Rubus idaeus) and gooseberries or currants (Ribes spp.). The jack pine (Pinus banksiana) whose slowly opening cones are burst open by the heat of fire, formerly a rare element of bogs and Kildare beaches, has become abundant near Tignish and East Bideford, following fire. The hardwood forest opened up by cutting and disease takes over a century to regenerate. Consequently, much of the upland forest at present is in various stages of succession. Openings often permit a dense growth of fern, Dennstaedtia punctilobula with clumps of Dryopteris spinulosa, to spring up; such partial openings are colonized by plants with berry fruits: Actaea rubra and the red elder Sambucus pubens are characteristic. Red spruce and red maple are apt to follow, or papei birch and fir. More complete cutting brings forth an earlier stage characterized by sedges (Carex emmonsii, C. deflexa) and red maple stump sprouts, sometimes with aspens and wire birch. However, the maples become the shade trees soon, and the species of partial openings appear, with the new sedges Carex deweyana, C. arctata and C. communis. With deeper shade, the herbs Clintonia borealis, Maianthemum canadense, Smilacina racemosa and Cornus appear. On the poorly drained soils, cutting has been less extensive. However, fire has not been less frequent. In either case the barrens that result persist longer than in other sites. The typical heaths are there, but rhodora (Azalea canadensis), wintergreen (Gaultheria procumbens) and leatherleaf (Chamaedaphne calyculata) are the most important, with sheep laurel next to them, and Spiraea 20 tomentosa accompanying them. Such sites are apt to become peaty, with the tall shrubs Nemopanthus mucronata or Vibrunum cassinoides in clumps and the lichen-coated larch and black spruce growing up very slowly. The orchids Spiranthes lacera, S. cernua and the club moss Lycopodium tristachyum,. all typical of exposed sites, find a place on the hummocks in such barrens. On the heavier lowland soils, partial red maple cutting has given room to aspens and fir with some spruce. Osmunda cinnamomea replaces bracken as the common fern. As a mossy ground cover is established in increasing shade, the sedges Carex brunnescens, C. debilis var. rudgei in damp, C. leptalea and C, disperma in wetter spots become typical; Cornus canadensis and Coptis groenlandica are very characteristic herbs. Along the wooded margins of swales or rivers these sedges may be accompanied by Athyrium filix'femina, Streptopus amplexifolius and Ribes lacustre. The cutting of red or white spruce woods on damp soils usually produces vegetation first in the damp pockets with peat and bog plants such as asters and Epilobium spp. and sedges; then fireweed, rasp- berries, elder, among the slash; then, red maple with balsam fir and paper birch following closely. Cutting in the pine woods produces a heath barren much like the habitat some ten years after a fire: same goldenrods, bracken, blueberries, sheep laurel and Pyrola rotundi folia, under wire birch and aspen with scant regeneration of pine — except in central Prince, where good examples may be seen near Freeland and Perceval River. Reversion from pasture follows a different course, comparable to the suc- cession on dunes. The grass is colonized directly by sun-tolerant young white spruce, with some alder and bayberry rapidly (15 years) shaded out. Until natural »: :#* if? ■Kn , *kjS§ f i .1 "■ White spruce quickly invades neglected pastures. Springton. 21 thinning occurs, only saprophytes (fungi) appear beneath; then moss (Pleurozium) and Maianthemum with Cornus canadensis. On the steeper slopes, Corylus cornuta and Diervilla lonicera appear at the sunlit edges of the spruce woods. In the southeastern uplands, the pastures may be colonized by wire birch and red maple rather than white spruce if these are close at hand. On damp or heavy soil the regeneration of forest from pasture is conducted by marsh fern {Thelypteris palustris) and larch, rather than white spruce; in flat sandy loam soil, the pasture succession parallels that of the dunes most closely, goldenrods (Solidago canadensis, S. graminifolia) giving way to bayberry, and that in turn to white spruce. Thus, forest succession on the lowland soils has previewed the later stages of succession in the habitats adverse to forest: the bogs, dunes and salt marshes, the lakes and ponds. 2. Fresh-water communities. Spring and rain-fed lakes form a very different plant community from that of stream-fed lakes. They are low in mineral nutrients and at first sandy-bottomed for lack of sediments. Thus, with the growth of plants they have a tendency to become acid from the accumulation of only partially decayed plant debris; consequently, they have a tendency to become bog lakes. The larger, Keefe's Lake, Glenfinnan Lake and Lake Verde, though partly Keefe's Lake with zone of Juncus militaris In the two to three-foot depths. 00 bordered by peat, show little tendency this way; the smaller Mermaid Lake more; the ponds at Village Green and Murray River and the Lot 10 Lake at Portage are definitely bog-lakes. The larger lakes have now acquired in their middle depths one or two feet of muck, in which the water lilies are rooted. Nuphar variegatum is found in all the lakes, boggy ones as well, in this medium; the rare Nymphaea odorata only in Lake Verde and one of the Murray River ponds. The scarcely encumbered sands of the shallower water are the medium of Eriocaulon septangulare and Lobelia dortmanna with emergent flowers, and of the completely submerged Isoetes riparia, Myriophyllum tenellum, Eleocharis acicularis and Elatine minima. The smaller Mermaid and Verde Lakes have a border of floating-leaved Sparganium multipedunculatum and S. angustifolium (species which reappear in the sandy barrier-beach ponds of the North Shore with larger Myriophylla and Potamogeton natans). The three larger lakes have an inshore border of various rushes, all with Eleocharis palustris, in Glenfinnan and Keefe's Lakes overshadowed by tall Juncus militaris. Mermaid Lake has Scirpus americanus and S. subterminalis in its sandy shallows. The peaty bank bordering open water at Mermaid and Glenfinnan \ Lakes is dominated by S. cyperinus; sandy borders are usually Shallows of the sandier shore of peat-bordered Mermaid Lake, showing Nuphar variegatum lily pads among the Scirpus subterminalis rushes. 23 dominated by Myrica gale and clumps of Ilex verticillata, species which likewise persist to the small pond or even swamp stage. With the transition to bog-pond, a floating mat of peat and heath roots encroaches on the water in an encircling ring. Its outer edge is the characteristic site of Woodwardia virginica, the dis- tinctive characteristic heaths include Gaylussacia dumosa, Kalmia polifolia and even G. baccata (also found on barrens;) the wet peat is the site of Habenaria blephari glottis, Sarracenia purpurea, Drosera rotundi folia, Eriophorum virginicum, E. tenellum and the other more tolerant cotton grasses. To the rear, the heaths of peaty barrens become more common, and the succession follows the same course. Most characteristic of the northern and eastern coasts is the formation of barrier-beach ponds. These show a succession on the whole similar to that of stream-fed lakes, which indeed they are. The artificial millponds, mostly a century old at least, likewise belong to this type. The distinctive appearance of these lakes is given by the far greater abundance of 'bulrushes/ Typha latifolia, which extend well out into the much greater (more quickly accumulated) organic debris of the shallows. One upland lake, Pisquid Pond, also stream-fed, answers equally to this description. The cattail {Typha) border of all these ponds is filled in by Alisma triviale, Polygonum amphibium, Lysimachia thyrsi- flora, P otentilla palustris, fringed by a separate border of Sparganium eurycarpum with scattered Bidens cernua, Sagittaria latifolia, Polygonum hydropiper and Bidens frondosa. The influx of mineral nutrient (often fertilizer from the hill farms) stimulates the formation of a water bloom of Lemna minor and no doubt of the other Lemnaceae. Pisquid Pond and Fernald's "pond at Grand Tracadie" are characterized by the dense belt of Ceratophyllum demersum and floating Spirodela polyrhiza outside the Typha zone. These ponds are far richer in pond- weeds than the lakes: Potamogeton foliosus and P. friesii are the most charac- teristic. Where manure reaches the ponds, Acorus calamus typically occurs. These pond shores show some diversity: M.W. Smith (1946) notes that on firm clay margins Scirpus validus is more abundant than Typha, and that Hippuris vulgaris occurs only around the influx of cold streams, However, the 5. validus persists into the foul-smelling quaking-marsh stage that follows the filling of such ponds by vegetation. This stage often shows large quantities of Equisetum fluviatile, a minor zone-former on open pond shores, and much Glyceria grandis. The first shrub to follow is hoary alder (Alnus rugosa) with J uncus effusus and /. balticus filling in. As the alder thickets become more dense, Chrysosplenium americanum and Hydrocotyle americana occupy the mucky pools; asters, Cala~ magrostis canadensis, Rosa virginiana, with Habenaria psycodes and Polygonum sagittatum cover the ground. Black spruce would commonly follow. Such an association (without the pools) is also found on damp headlands in western Prince. Differences between upland and lowland streams are apparent in their vegetation, perhaps largely due to the difference in oxygen content between swift-running streams and slow. The sandy-bottomed upland streams and rivers 24 usually have Ranunculus trichophyllus as the only submerged species, joined by Myriophyllum exalbescens in the southeastern upland. In slower water long- trailing beds of Potamogeton praelongus become common, and nearer the lowlands this species is replaced by P. perfoliatus var. bupleuroides, or more rarely P. alpinus. In the Dunk, Zannichellia palustris (often halophytic) penetrates for two miles above salt water. 3. Bog communities. Although peat (usually of Sphagnum subsecundum) is found in pockets in the lowland woods of red maple, and bog sphagna may take over a poorly drained fire barren, the bog community is determined by peat and 'f^lS^'f A boggy depression with sedges and cotton grass in the center, heath plants (here mainly Azalea canadensis) surrounding, and black spruce in the back* ground. Near Mermaid Pond. is essentially a later stage in lake succession. The vegetation of floating mats described previously is that of bogs, with the exception of species of the water's edge like W oodwardia virginica. All the other members of that community occur in the great bog of Black Banks, oddly enough the only one of the great bogs whose lake-basin origin might be questioned. Like many smaller pond-basin bogs, the Black Marsh at North Point is a sea of leatherleaf {C ham ae daphne calyculata) and cotton grass {Eriophorum an gusti folium) interspersed with islands of stunted larch and black spruce, each with its ' 'beach" of Scirpus cespitosus 25 The heath of ChamaedaDhne calyculata at Black Marsh with islands of black spruce. var. callosus. These clumps include Nemopanthus mucronata, Aronia prunifolia, and Amelanchier ? fernaldii. In the hummocks of peat, Rubus chamaemorus and Vaccinium oxycoccos are characteristic. Pools in the peat are surrounded by Sarracenia, Drosera rotundifolia and occasionally Utricular ia cornuta. Damp depressions harbor the magenta-flowered orchids Calopogon pulchellus, Pogonia ophioglossoides and rarely Arethusa bulbosa. The great bogs of Portage and Miscouche are largely wooded, as the red maple and wire birch and black spruce have encroached on the open peat. Smaller bogs are essentially similar but lack the full variety of species. 4. Maritime associations. Various communities, fresh as well as salt, range along the coast usually in series of narrow bands in the treeless zone occasioned by the immediate proximity of the sea. Eroding cliffs, windswept sand of beaches and dunes, waterlogged hollows, estuarine flats, brackish lagoons, salty tidal marshes, and exposed headlands present a great diversity of habitats with corresponding diversity of floras that can be sketched only briefly. The yielding shale and slumping sandstone of the Island's capes erode too rapidly to support a permanent or even a consistent community of plants. Weeds are the typical opportunists: Senecio sylvaticus, Sonchus asper, and those halophytes of weedy genera, A triplex patula and Plantago juncoides. The wetter and more permanent grassy clay cliffs may support plants such as Triglochin 26 palustris. (Inland cliffs are equally limited: the only ones worth the name are those of the Dunk which support the Island's only Polypodium virginianum and Sphenopholis intermedia.) The sandy beaches and wet stretches of sandy soil favor Elymus mollis rather than the marram grass Ammophila breviligulata, but its associates are the beach pea (Lathyrus japonicus), Cakile edentula and the silverweed Potentilla anserina, often with Arenaria peploides and Glaux maritima, or, on the south shore, Xanthium echinatum. Deeper sand is found windblown into ridges of dunes, usually parallel to the coast and thus sheltering alternating hollows, the dune slacks. The dunes are stabilized by the rhizomatous Ammophila, the dune grass par excellence. Spaces among the grass are filled by Lathyrus japonicus Sand dunes tend to become stabilized by dune grass (Ammophila breviligulata) and later colonized by white spruce (Picea glauca)* (itself rhizomatous), by Carex silicea, Iris hookeri or Artemisia stelleriana or by small annuals such as Trifolium procumbens or Euphorbia polygoni folia. On the lower dunes Sonchus arvensis is usual. At Bothwell two species of tludsonia cover the dunes, but even there as on all North Shore dunes, the prevalent species is//, tomentosa. Small dicranoid mosses, filling space, become the site of Sagina nodosa colonies. Blowouts in the dunes are first colonized by the Cakile of sandy beaches. The first woody species to arrive is the bay- 27 A lingering patch of marram grass in the white spruce forest on the inner dunes* Brackley Beach. berry among which Anaphalis margaritacea and Solidago graminifolia are the perennial weeds; the second is the white spruce which, except for rare replace- ment by cedar in west Prince, becomes the forest. However, while the spruce is scattered and the sand loose, the shrubs Empetrum atropurpureum, Juniperus horizontalis, or Toxicodendron radicans (poison ivy) may appear, and the per- ennial herb Smilacina stellata. The slacks may be wet and carpeted with the moss Aulacomnium palustre t in which case the characteristic vascular plants are Epilobium palustre, Cicuta bulbifera, Juncus pelocarpus and Vaccinium macrocarpon. Solidago sempervirens never extends to slacks within the first ridge of dunes, S. graminifolia and Aster novi'belgii replacing it. Drier slacks are colonized by Juncus balticus and /. alpinust often with the orchid Liparis loeselii filling gaps. On dry or wet slacks, the asters, goldenrods and everlasting only slightly precede the arrival of bayberry. Sometimes the sand may blow inland for a short space, favoring the growth of the warmth-loving grass Poa compressa, or into barrier-beach ponds creating sand flats colonized by Limosella subulata9 Rumex maritimus, Scirpus americanus, S. maritimus and S. paludosus. The very local species Aster 28 laurentianus belongs to this habitat. Salt spray blown over the cliffs on to sandy ground may create the association of a sandy beach at a higher level; this is particularly true on the Kildare sands of western Prince. Muddy shores, in this predominantly sandstone land, are mainly located around estuaries, the largest salt marshes being those of the Dunk and Hills- borough Rivers. So few species colonize these saline habitats that their charac- teristic coloration marks the zones, bright-green Spartina alterniflora on mud submerged at every tide, reddish-green S. patens above it and blue-green Puccinellia pumila replacing the latter where hay-cutting or tramping have reduced it. Zones from lowest to uppermost are dotted with Plantago juncoides and bushy Limonium nashu, and occasional clumps of Triglochin elata. Bare patches of mud on the marsh are colonized by Salicornia europaea and Suaeda maritima, or in upper zones by Spergularia marina or S. canadensis. A weedy belt of A triplex patula follows the high-water mark with the biggest roll of eelgrass wrack. The upper limit of salt-water influence is a weedy tension-zone of which Solidago sempervirens and Spartina pectinata are the most constant members, and Ligusticum scothicum one of the less. Pools on the salt marsh are often occupied by submerged Ruppia maritima. The muddy banks of estuarine creeks are tne characteristic habitat of Car ex paleacea and C. salina, while a carpet of Eleocharis parvula may here grow even below the Spartina alterniflora belt. The sea water affords a home to only one vascular plant, the rhizomatous eelgrass Zostera marina, which forms great beds off sandy shores in sheltered bays or estuaries. Besides the dune succession which leads to white spruce forest, the latter may appear on exposed land above the beaches of headlands. These headlands may even be treeless as in the case of extreme North Point, covered by Festuca rubra, Plantago juncoides var. laurentiana and Trifolium arvense, or East Point covered by hummocky Empetrum nigrum with sandy patches of Plantago juncoides var. decipiens. At Stanhope, for instance, where the dune sand lends force to the effects of the west wind, there are large areas of Juniperus communis, J. horizontalis and Empetrum nigrum with scattered Lechea intermedia, Vaccinium vitis-idaea and introduced Thymus serpyllum. The borders of taller spruce woods in such sites are characterized by Deschampsia flexuosa, while pasture edges grow abundant Euphrasia americana. 5. Cultivated vegetation. The types of use made of the farmland (87% of the Island) is shown by this separation (1951): Cultivated hay 203,783 acres Improved pasture 197,937 acres Grains 177,636 acres Potatoes & turnips 37,156 acres Farm wood lots 346,191 acres Unimproved 103,318 acres Miscellaneous crops 39,283 acres Total farmland 1,095,304 acres 29 North Point, an eroding cape exposed to the full force of the Gulf storms, is essentially barren of coastal vegetation* Unimproved land may include scrub and heath used as pasture together with swamp or bogland. All this land, in hay, oats, other grains, potatoes and turnips, is of interest botanically as a source of annual weeds, which grow in the disturbed ground as space fillers among the crop plants. Here one finds in particular the mustards and stitchworts, knotgrasses and pigweeds. But cul- tivation varies with the crop and has different effects. Potatoes, grown in the usual rotation, are relatively weed-free except for a little lamb's guarters {Chenopodium album) or knotgrass (Polygonum aviculare). The ridging of rows in the cultivation of root crops helps 'clean' the soil, and oats, the usual grain, grows too closely for cultivation and supports an abundant crop of devil's paintbrush (Hieracium floribundum, H. pratense and //. aurantiacum) and mustards, especially wild radish [Raphanus raphani strum). The practice of chemical weed- killing will alter the proportions greatly, while not extinguishing any species. Just so in the 1900's, the clean grain seed enforced by inspection eliminated the corn cockle [Agrostemma githago) but has reduced the abundance of only a few of the other species affected. Mustard seed lives so long in the soil that, once established, even crop rotation proved no way to counteract it. The planted hay O crop, following in the common rotation, is a timothy-clover mixture, either red clover or alsike being usual. The practice of making silage has not yet become widespread but, especially in the central upland, one sees sweet clover {kelilotus officinalis) or alfalfa {Medicago sativa) abundant in the grass planted for this purpose and as pasturage. In the second year of the hay crop the clovers have largely died out, but have served their purpose in nitrifying the soil; this year it is a timothy hay crop and the spaces for weeds are still small. The fifth year, it is usual to pasture the field. The rotation may vary from this (the usual for mixed farming) according to the specialty of the farm. Farms with more land in potatoes {Solarium tuberosum) rotate fields every three years through potatoes, grain and clover hay; cattle farms, to assure more land in fodder crops, use a grain, clover hay, timothy, and pasture rotation. Because of the rotation the establishment of perennial weeds is unusual, but a biennial such as the daisy {Chrysanthemum leucanthemum) , establishing itself in the oatfields as the Perhaps nowhere else in the world do the daisies (Chrysanthemum leucauthemum) grow in as great profusion to brighten the landscape in early July; hay fields are solid white with them. Hunter River* rosette, survives the first mowing and flowers next June so abundantly as to whiten the hayfields of the Central Uplands. While rotation alters the tillage and the vegetation of cropland every year, 'permanent pasture1 supports a different vegetation. Although the original pasture grasses must have been introduced from Europe, those which now dominate the pastures are essentially naturalized and seldom planted. Among these the bent, called browntop or P.E.I, bent {Agrostis tenuis) is most abundant in the Island as in other comparably cool humid temperate regions. It is replaced behind the dunes or on moraines by the Canada bluegrass (Poa compressa), another European, 31 favored by warm well-aerated soils here where, generally, the climate tends to be too cool for it; in damp or clayey pastures by the creeping bent (Agrostis palustris), or where these are too much trodden for perennials to survive, by the annual bluegrass (Poa annua). The weeds of pastures are, like the grasses they must compete with, perennials: black-eyed Susans {Rudbechia serotina) and native goldenrods {Solidago canadensis, S. gr ami ni folia) where the grass is heavy; poisonous ragwort (Senecio jacobaea) and prickly thistles (Cirsium arvense) avoided by the stock even where the grass is overgrazed. Where pasture is neglected, forest succession is initiated. On the sandy Culloden soils, as on the fixed dunes, goldenrod and hawkweed (Hieracium pilosella) are replaced by a dense shrubbery of bayberry (Myrica pensylvanica) which is soon overtopped by white spruce or wire birch and white birch. In wet places, it is the soft rush {J uncus effusus) and marsh fern (Thelypteris palustris) which thrive on over- grazed lands, and the larch which replaces the white spruce as the sun-tolerant weed tree. On the upland, fir and red maple succeed the white spruce in whose shade they germinate, and thus regenerate the climax forest. Areas in ;/nonagricultural human use" have, besides various weeds not found on agricultural land, also the deliberately planted shade trees and garden shrubs and herbs. Residential Charlottetown streets are lined with European lindens {Tilia europaea) and Scotch elms (Ulmus glabra) to the exclusion of other species. The former are planted, usually for avenues, at many of the older estates in the country; the latter are not. The English oak {Quercus robur), never seen in towns, is grown at a few old farms near navigable estuaries. In the western villages where shade trees are few, balm of Gilead (Populus gileadensis) or the native balsam poplar (P. balsamifera) or the "English" (Lombardy) poplar (P. nigra var. italica) are most often planted; in Wellington and Alberton box elder [Acer negundo) gives the streets their bright dusty green. Lilac (Syringa vulgaris) is the favorite ornamental shrub, followed perhaps by old double roses (Rosa cinnamomea). The use of dwarf elm (Ulmus pumila) or Siberian pea (Caragana arborescens) for hedges is a sure sign of recent archi- tecture and of Experimental Farm leadership. Along the railroads wild barley (Hordeum jubatum) and barnyard daisy (Chrysanthemum leucanthemum) and various annuals such as small toadflax (Chaenorrhinum), small yellow clovers (Trifolium procumbens), crab grasses (Digitaria ischaemum) and foxtail grasses (Setaria glauca, S. viridis), and, more recently, prostrate pigweeds (Amaranthus albus) and spurges (Euphorbia supina), used to be common. But by the use of weed killer these are being greatly reduced and in some cases may be prevented from establishing themselves along rights- of-way. Used along roadsides to save the effort of brush-cutting, weed killer spray has killed a band of alder, elder, wire birch and escaped box elder, as well as the weeds, leaving only grass and dead sticks. The use of western grain as feed for stock or poultry has led to the arrival of western weeds at sidings and farmyards: pineapple weed (Matricaria matricarioides, already naturalized), occasional wild oats (Avena fatua), prickly lettuce (Lactuca 32 scariola), great ragweed {Ambrosia trifida, whose establishment could ruin much of the tourist industry), cockleburs (Xanthium italicum), sunflowers {Helianthus spp.) and flax (Linum usitatissimum) signify such locations. The waste land around houses has long supported its vile crop of couch grass (Agropyron repens) and burdocks (Arctium minus), or catnip (Nepeta cataria) and motherwort (Leonurus cardiaca) where hens have scratched in the shade of hedges. PHYTCGEOGRAFHY Absence and Presence In a discussion of species absent from Prince Edward Island, it is necessary to take into account present and past factors in plant distribution. The simplest hypothesis to explain the presence of the present native flora is by natural diffusion from eastern New Brunswick and northern Nova Scotia across the waters of Northumberland Strait, varying in bredth from 9 to 40 miles. To test this hypothesis, a list was drawn up of the indigenous flora of the adjacent mainland within the drainage area of Northumberland Strait. Comparison of this Island list with Roland's Flora of Nova Scotia shows that 98% of the Island flora is found in Nova Scotia, in some part or other; only one or two species in the Island are not represented in either Nova Scotia or New Brunswick (Rumex persicarioides, Bidens heterodoxa). Of the flora of northern Nova Scotia, some 650 indigenous species, a full 80%, cross to the Island. The absences lie mainly in one of three ecological communities: (1) the aquatics, probably because the lakes of the Island are so few and small, (2) the intervale plants, for lack of river-alluvium areas in the Island, (3) the geophytes of the spring flora under hardwood forest. In the face of abundant hardwood forest in the Island, histori- cal factors may serve to explain this last absence. That diffusion across the waters of the Strait has taken place, there need be no doubt. Prince County, it seems, has never been linked by land to New Brunswick since the retreat of the ice, yet there are over 30 species there which are absent from the rest of Prince Edward Island and present in New Brunswick. Secondly, there are 15 species in the Island that do not occur in Nova Scotia at all, but which do occur in New Brunswick. And, if one postulates that it was the saline waters of the Strait which barred most of the "spring flora geophytes" from the Island, how can one account for the presence of a few of them (Viola pensylvanica, Claytonia caroliniana, Panax trifolius) in localized areas without assuming that each colony represents diffusion across the Straits, either from the west or south? It may be argued that ecological necessities have confined some of these local species to western Prince, but their migration via a Nova Scotian land bridge (see Geology Chapter 2) is out of the question. Of the species that may have migrated by a land bridge formerly connecting the Island to Nova Scotia, the greater part can give no evidence. The intervale 33 89470-3 and hardwood flora of Nova Scotia by its distribution pattern in Nova Scotia (from the New Brunswick border to "the hub", thence southwest to Kings County and northeast to Cape Breton), let alone by its fossil history in the United States, shows that it has entered Nova Scotia via the Chignecto Isthmus and therefore since the recession of the postglacial sea that submerged the the Isthmus and the presumably earlier submergence of any land bridge to Prince Edward Island. The species that have entered Nova Scotia from the southwest rather than from New Brunswick are referred to as the ' 'Coastal Plain Element" (not ;/Coastal Plain flora/' for even in Nova Scotia they are off the Atlantic Coastal Plain). However, none of these reach the Island. But, other species of southwestern affinities, occurring both in Nova Scotia and in southern and eastern New Brunswick, appear in Prince Edward Island. These are either plants of sandy lands, such as the Panica, Hudsonia ericoides and Viola fimbriatula, or of pond and bog, such as Woodwardia virginica, Habenaria blephariglottis and Gaylussacia dumosa. Possibly these species colonized sandy deposits, bog and swampy pond hollows in the new land left by the receding postglacial sea, arriving from nearby Nova Scotia rather than from New England to the southwest. Several such species stop at Northumberland Strait, but of the dozen or so which Nova Scotia and the Island have in common, many (e.g., Panicum depauperatum, Lactuca hirsuta, Viola fimbriatula) seem absent from northern Nova Scotia, their ranges broken between the Annapolis Valley and the Island. Thus, there is little or no positive evidence of the significance of the land bridge in the migration of the present flora to the Island. One can only hope that pollen analysis will resolve some of the past history now obscured by wider diffusion. Another group whose distribution may show evidence of the past continuity of Prince Edward Island with the mainland are those which follow the North Shore. The simply psammophilous or halophilous species follow round from the Gulf coasts of the Island to the Northumberland Strait coast. But other species confined to the Gulf shore show ranges extending no farther south than the Gulf shore of New Brunswick (e.g., J uncus alpinus, the Gulf endemic Aster laurentianus ) or that coast and northern Cape Breton {Parnassia parviflora, Empetrum atropurpureum, Eleocharis pauciflora) or also with outposts on the cold coasts of the Bay of Fundy (Juniperus horizontalis and Iris hookeri) or the Bay and the outer coasts of Nova Scotia (Euphrasia randii). While ecological requirements (cooler summers, exposed sites) may have kept them from the Strait shores, their disrupted ranges at these their southern limits on the coast suggest a continuity dating back to a colder postglacial stage. An interesting range pattern is shown by some eight species which occur in New Brunswick, Prince Edward Island and Cape Breton but not on the Nova Scotian mainland. Most of them (viz., Aster junciformis, Betula pumila, Caltha palustrisf Galium labradoricum, Vaccinium uliginosum), are near their southern limit but others such as Comandra r ic hards iana, Anemone canadensis and Hypericum ntajus, while surely reaching their farthest east in the Gulf area, are 34 not boreal types at all. If these migrated overland by way of northern Nova Scotia, those near their southern limit may have lost mainland stations through warmer summers; if so, the three latter must have migrated via Prince Edward Island and probably after the Gulf assumed its present form, or else they would have mainland stations. And the Prince Edward Island summer except along the Gulf shores is as warm as that of the mainland. Local Distribution. Within the Island the flora tends to be fairly uniform. Some 250 of the 625 native species, at least, may be seen from the distribution maps to be wide- ranging in the province. Some of these seem to be rare in Kings County. However, as that area was the least completely investigated during the survey, many apparent gaps there may eventually be filled. Others show the distribution: western Prince and east-central lowlands, their absence from the central uplands and the southeast being due to the absence of suitable habitats. The same might, of course, be said of some 40 common halophytes-psammophytes restricted to the coastline. Another 30-odd species (in Carex, section Stellulatae, in Amelanchier, Crataegus and Rubus particularly) were too imperfectly understood for their ranges as mapped to be reliably complete. However, in spite of the exclusion of at least 320 of the native flora from consideration, 23% showing very restricted ranges in the Island may be of phytogeographic interest. Some 16 show restriction to the eastern part of the Island, some 44 to Prince west of Summerside, another 23 so widely scattered and scarce east of that isthmus as to suggest a late spread from it, some 14 a restriction to the central uplands, some 25 to the North Shore, 22 to the east-central lowlands, altogether 144 of narrow range. Of these groups none can be quite regarded as homogeneous. Some of the North Shore species are of relict status (e.g., Aster laurentianus) , possibly related to the unstable shore habitat. Many of these are species (e.g., Iris hookeri, Euphrasia randii) restricted, possibly by the cooler summers, to the Gulf area and the coasts of the adjacent Maritime Provinces and New England. One is an Atlantic element, of the Coastal Plain type (Corema conradii). The group restricted to western Prince includes many species abundant in that area {Carex gracillima, Salix rigida, Populus balsamifera, Thuja occidentalism Pinus banksianaf Geum aleppicum, Fraxinus americana, Veronica scutellata, Eupatorium perfoliatum). The equal failure of Thuja to pass the Chignecto Isthmus into Nova Scotia, almost due south of the Miscouche Isthmus, suggests some historic barrier. Among the seven or more native species of the Island that occur also in New Brunswick but not in Nova Scotia most are of this coastal endemic relict status. Pilea pumila, however, stands out as one that reaches as far east as the upper Hillsborough River but does not enter Nova Scotia though its range presents a hardwood forest pattern. The fact that some 23 (e.g., Viburnum trilobum, J uncus dudleyi) which are more abundant in western Prince have stations east of the central uplands suggests that species of more efficient dispersal may cross the barrier while others must very gradually 35 89470-3Vi make their way round it by way of the coastal streams. Furthermore, many of the western Prince group are confined west of the isthmus at Portage (e.g., Angelica atropurpurea, Solidago gigantea) or to theTianish area (e.g. , Ranunculus pensylvanicus, Senecio aureus), which reduces the significance of the Miscouche isthmus to that of one of several bottlenecks to the slower migrants. But the most striking thing about the species of restricted range, on the whole, is how few their stations are. This is particularly true of these species in the central upland, for which I have advanced the argument of chance arrival after the formation of Northumberland Strait accompanied by ineffective dispersal. Thus, we may look upon all species local in Prince Edward Island, except for the few relict survivors of the North Shore, as late arrivals that have been slow to fill their scattered ecological niches. The introduced flora (316 species) includes many diverse elements, which may be separated historically. The most important is the European element. Of 226 introduced species mapped in this list, 200 are of European origin, only 26 of American origin. The reasons for this are twofold: the source of the cul- tivated plants of Prince Edward Island (including the South American potato) was western Europe, hence the weeds brought in with them are mostly European; secondly, most of the American species which would be considered as weeds (e.g., the goldenrods) were native to the area and are classified as indigenous. Of the European species, at least 28 were introduced as crop plants (the degree of intent is difficult to assess with pasture grasses), another 30 as ornamentals and shade trees, while 142 would be classed, at least in the present age, as weeds. (Many of the mustards, for instance, have lost their status as cultivated plants). Historically, the introduction of European plants belongs to the period of settlement from the British Isles, with attendant clearing and cultivation. After Confederation, the building of railways, trade with mainland Canada, and the establishment of a train-ferry link with New Brunswick all helped to introduce weeds from North American sources. The railways have become the route of Chaenorrhinum minus, Tragopogon pratensis, Lactuca scariola, Euphorbia supina, Collomia linearis, and many other weeds. Though the source of the first three is originally European, all have entered the Island from American sources. Since the period 1890 to 1910, this has been true, for the modern centralized overseas trade does not come to Island ports. Apart from Blythe Hurst's sporadic and continual introductions, the only recent European weed to arrive by sea was Arnoseris minima. Weed seeds have been brought in with feed grain and these have given rise, particularly at railway sidings, to an element from western Canada, often species originally Russian. One question that has often concerned phytogeographers is the degree to which plants find themselves at the limit of their range within an area. The insular position of Prince Edward Island makes such decisions rather arbitrary, since the limit imposed by the Gulf is much farther south on its North Shore than in Gaspe or Cape Breton, which have many more plants near their southern ii 36 limits. Tentatively, then, one may divide the native species of the flora into two groups: those which cross the Gulf — about 35 species (or some 22 percent), and those that are found on either side beyond it — some 490. Of the former which come in from the southwest, practically all are at the northern margin of their range, only a few (e.g., Thuja occidentalism Pilea pumila, discussed above) are at the eastern margin and do not spread on into Nova Scotia. Those that are at a southern limit, some 30 species, mostly of the North Shore only, make up a mere 5%; those at their western limit another 5%. The latter are the 'Atlantic' species which reach eastern New Brunswick and also Newfoundland, but seem restricted to a relatively coastal migration pathway. Half of the Atlantic group of the Maritimes do not extend as far inland as Prince Edward Island. Thus, the great majority of those 68% of the Island's species that also range across the Gulf are species which in the Maritimes stop at nothing but the Atlantic, the ubiquitous species of this area. That many of these are far from common in the Island shows the significance of the other factors (e.g. cooler summer temperature, late time of arrival, slow rate of dispersal, the few lakes, the wide separation of marshes) in limiting the distribution of species there. Summarizing Statistics Although no final summary of the flora can be expected, as additional introductions and adventives find their way into the flora and occasional native species continue to be discovered, a tabulation of the species now listed will give some idea of the composition of the flora Taxonomic rank Native Introd. Total Species 624 316 940 Varieties 69 13 82 Hybrids 12 2 14 When compared with that of Nova Scotia it is a small flora: Nova Scotia (Roland, 1947, modified by later discoveries) has 1012 native species and hybrids and 243 varieties, but the province has an area ten times as great as Prince Edward Island and has a greater variety of habitats. The significance of agriculture and settlement in the flora of Prince Edward Island may be shown by comparison of the introduced flora with that of Nova Scotia; the larger province has 392 introduced species and hybrids, and 30 introduced varieties, only a little more than the Island. For the Nova Scotian flora, Roland has not distinguished the adventive (recently or occasionally arriving species which as yet do not per- petuate themselves in the flora) from the fully naturalized species, but at least 101 "introduced species" of his list are not naturalized inNova Scotia. Prince Edward Island has a naturalized flora of 199 to Nova Scotia's less than 291 species, for 117 Island species must be regarded as merely adventive. 37 A system often used in comparing floras as to the structure of the vege- tation they make up is the "Raunkiaer spectrum." Area £ Ch H Cr_ T_ Connecticut 15.0% 1.9% 49.4% 21.7% 11.7% Prince Edward Island 15.2% 4.0% 47.1% 24.4% 9.0% Gaspe 12.1% 9.0% 49.0% 22.1% 7.5% These figures for the native flora show a decided similarity between all these temperate forested regions of the northeast. However, the trees and shrubs (F - phanerophytes), form a smaller proportion of the flora in Gaspe where there are alpine areas. For the chamaephytes (Ch) or generally woody plants with low horizontal or decumbent stems, the proportion rises from Connecticut north to Gaspe, which with its alpine areas has many more in the flora. For the hemi- cryptophytes (H), herbaceous plants with stolons or rosettes or tillers near ground level, the proportion shows no significant trend; likewise for the crypto- phytes (Cr), rhizomatous and/or aquatic plants (no bulb plants in P.E.I.). The proportion of annuals (T - therophytes) falls off to the northward, showing the Island flora once again in the logical intermediate position. Addition oi the naturalized and adventive plants reduces the proportion of all classes relative to the therophytes, which rise to 13.5% in Connecticut, 14.1% in Gaspe, and overwhelmingly in the Island to 18%, probably because the native flora is small relative to Connecticut and the cultivated area large relative to Gaspe. The significance of the recent survey can be seen from the fact that, although over 80 species have been excluded from the list of Hurst and the one of MacSwain and Bain, the 716-species list of Hurst in 1940 has been expanded to a 939-species list (or, excluding adventives often listed by Hurst an 822- species list). REFERENCES This list includes both major reference works and articles dealing specifi- cally with Prince Edward Island or its plants. For species mentioned only incidentally the appropriate reference is included in the body of the list. The important records in contributions of a serial nature, such as J.M. Macoun's "Contributions to Canadian Botany" and Groh's "Canadian Weed Survey," have been treated in this way to facilitate reference. Adams, John. 1937. Some additions to the flora of Prince Edward Island. Can. Field Nat. 51: 105-107. Auer, Vaino. 1930. Peat bogs in southeastern Canada. Canada Dept. of Mines, Mem. 162. Bain, Francis. 1890. The Natural History of Prince Edward Island. Charlottetown, Haszard. Bain, Francis. 1892. Additions to "Plants of Prince Edward Island". 3 p. Charlottetown, Prince Edward Island Nat. Hist. Soc. 38 Broun, E. Lucy. 1950. Deciduous forests of eastern North America. Philadelphia, Blakiston. Britton, N.L. and Addison Brown. 1913. Illustrated Flora of the Northern United States and Canada. 2nd ed. New York, Scribner, 3 vols. Burgess, W.H.L. 1890. Pteridophyta. In Macoun, John. Catalogue of Canadian plants. Montreal. Campbell, J. Ewen. 1952. A New Flora of Prince Edward Island, rev. ed. Charlottetown, Laboratory of Plant Pathology, mimeographed. Chalmers, Robert. 1895. Report on the surface geology of eastern New Brunswick. north-western Nova Scotia and a portion of Prince Edward Island. Geol. Surv. Can. Ann. Rep. 7(M): 149 p. Churchill, J.R. 1902. Some plants from Prince Edward Island. Rhodora 4: 31-36. Clarke, J. A. 1906. Weeds and their eradication. In Prince Edward Island Dept. of Agriculture, Annual rept. ...1906. Charlottetown, p. XXVIII. Copeland, E.B. 1947. Genera Filicum. Waltham, Chronica Botanica. Dore, W.G. Phytogeography of the Maritime provinces (unpublished notes). Dore, W.G. and A.E. Roland. 1942. The grasses of Nova Scotia. Proc. N. S. Inst. Sci. 20: 177-288. Ennis, B. 1928. The life forms of Connecticut plants and their significance in relation to climate. Conn. Geol. Nat. Hist. Bui. 43: 1-100. Fassett, N.C. 1928. The vegetation of the estuaries of northeastern North America Proc. Boston Nat. Hist. Soc. 39: 73-130. Fernald, M.L. 1925. Persistence of plants in unglaciated areas of boreal America Am. Acad. Arts Sci. Mem. 15: 239-342 Fernald, M.L. 1950. Gray's manual of botany. 8th ed. New York, American Book Co. Gaudet, J.F. 1956. Forestry past and present on Prince Edward Island. Maritime A.I.C. Conference, Charlottetown, 3 p. (mimeographed). Gleason, H.A. 1952. The new Britton and Brown illustrated flora of the northeastern United States and adjacent Canada. New York, New York Botanical Garden. 3 vol. Gorham, R.P. 1944. The known distribution of the buckthorns in the Maritime provinces. Acadian Nat. 1: 118-124. Groh, H. 1927. A weed survey of Prince Edward Island. Sci. Agr. 7: 388-395. Halliday, W.E.D. 1937. A forest classification for Canada. For. Serv. Bui. 83. Hamel, Aubert. 1955. Esquisse ecologique des comtes de 1 'Islet et de Kamouraska. Can. J. Bot. 33: 223-250. Harvey, D.C. 1926. The French regime in Prince Edward Island. New Haven, Yale University Press. Hulten, Erik. 1937. Outline of the history of arctic and boreal biota during the Quaternary period. Stockholm. Hurst, Blythe (Sr.). 1933. Flowering plants and ferns of Prince Edward Island. Trans. Roy. Can. Inst. 19: 251-273. Hurst, Blythe (Sr.). 1940. A new flora of Prince Edward Island. Charlottetown, Guardian Printing Office. Hurst, Blythe (Sr.). 1941. Addenda to 1941. In later reprinting of Hurst (1940). 39 Lochhead, William. 1902. The weeds of Prince Edward Island. In Prince Edward Dept. of Agriculture, Annual rept. ...1906. Charlottetown, p. I - XXVII. Long, K.D. 1952. Forest types and sites of the Acadia Forest Experimental Station. Fredericton, Acadia For. Exp. Sta. (M 226), mimeographed. MacLeod, J.W. 1947. The forests of Prince Edward Island. Forestry Chron. 23: 190-193. Macoun, John. 1883-1903. Catalogue of Canadian plants. Montreal and Ottawa, 7 parts. (Except for two specimens cited in earlier volumes, all Island vascular plant records are published in Part 5 (1890), the last volume on vascular plants.) MacSwain, John. 1907. The flora of Prince Edward Island. In: Spotton, H.B. The Elements of Structural Botany; P.E.I, ed. Toronto, Gage, p. 1-104. MacSwain, J.F., and F. Bain. 1891. List of Prince Edward Island plants. Charlottetown, P.E.I. Nat. Hist. Soc, 8 pp. Marie-Victorin, Frere. 1935. Flore laurientienne. Montreal, Imprimerie de la Salle. Martin, Lynton J. 1955. Observations on the origin and early development ot a plant community following a forest fire. Forestry Chron. 31: 154-161. Owen, E.B. 1949. Pleistocene deposits of O'Leary map area, Prince County, Prince Edward Island. Geol. Surv. Can., Paper 49-6. map. Robinson, B.L. and M.L. Fernald. 1907. Gray's new manual of botany. 7th rev. ed. New York, American Book Co. Roland, A.E. 1947. The flora of Nova Scotia. Truro, Truro Printing * Black crowberry (Empetrum nigrum) mats the dune slacks along the North Shore; a clump of bayberry in the center foreground* [Rhus glabra L. does not extend northeast of central Maine. Reported by MacSwain and Bain, MacSwain, Hurst.] Toxicodendron radicans (L.) Ktze. Map 535. (Barkley, Ann. Mo. Bot. Gard. 24: 417. 1937) POISON IVY Local; usually near coast, most common in prince. Damp, sunny sites: hollows among inner dunes, tension line around salt marsh, open swamp thickets: "dry rich woods" (Bain) inapprop- riate for Island occurrences. The irritating skin rash caused by contact with the oil of its stems and leaves seems like an allergic reaction, immunity often being lost only after several contacts with it. The oil may be removed by scrubbing with laundry soap if the rash has not set in, otherwise the only precaution to be taken is bandaging to prevent spreading the irritation; baking soda may be used as a palliative, perhaps because it dissolves the oil. Island specimens belong to the northern low variety, var. rydbergii (Small) n. comb., based on Rhus rydbergii Small, Mem. N.Y. Bot. Gard. 1: 268. 1900. {Rhus Toxicodendron, Bain, MacSwain and Bain, Hurst; R. radicans ]_., Campbell). AQUIFOLIACEAE (ILICINEAE, MacSwain) Ilex verticil lata (L.) Gray Map 536. Lowlands, throughout. Swampy streams, borders of damp woods, in shade the leaves thinner and with translucent spots, the "var. tenuifolia." {Prinos verticillatus L., MacSwain and Bain); MacSwain, Hurst. 190 535 Nemopanthus mucronata Acer spicotum 537 538 Acer pensylvonicum Acer sacchorophorum 540 191 Nemopanthus mucronata (L.) Trel. Map 537. Scattered throughout. Swamps and thickets; in open bogs more compactly branching and gnarled. (A', canadensis DC, MacSwain and Bain; A', fascicularis Raf., MacSwain); Hurst. ACERACEAE (SAPINDACEAE, MacSwain) Acer spicotum I am. Map 538. General in Prince and Queens; in the northeastern hills of Kings. Openings or understory of hardwood forest, on well- drained sites: knolls, upland or ravines. Bain, MacSwain and Bain, MacSwain, Hurst. Acer pensylvanicum L. STRIPED MAPLE Map 539. Scattered throughout; common throughout and at borders of upland. Ravines, gullies and slopes; understory in mixed or hardwood forest. Possibly the "white maple (A. Negundo)" of Stewart? Bain, MacSwain and Bain, MacSwain, Hurst. [Acer sac char ophorum K. Koch (Rousseau, Contr. Inst. Bot. Univ. Montr. 35: 1-66. 1940) ROCK MAPLE, SUGAR MAPLE Map 540. Throughout. With Beech, forming the climax forest of well-drained sites, the uplands and knolls. The sweet sap, gathered between March 25 and April 10 (Stewart) is boiled for maple sugar, eaten as a candy and used as maple flavoring now that imported sugar is more economical. In 1890, Bain estimated the annual production as 25,000 pounds. No commercial production now. The curled grain of the wood was used in ornamental woodwork as "bird's-eye maple" (Stewart). Extensively planted as a shade tree. (^4. saccharinum, Stewart, Bain, MacSwain and Bain, MacSwain; A. saccharum, Hurst, Campbell).] [Acer saccharinum L. , is occasionally planted but is not native east of the Saint John. (A. dasycarpum Ehrh., MacSwain); Hurst.] Acer rubrum L. RED MAPLE Map 541. Common throughout. Hardwood forest (well-drained sites), often pure on uplands after clearing, due to stump-sprouting; in damp sites, very abundant, with spruce forming mixed woods. Often planted as a shade tree, though more brittle than A. saccharum. Its red fall color is our best. Stewart, Bain, MacSwain and Bain, MacSwain, Hurst. The forma tomentosum (Desf . ) Dansereau present (Bonshaw). Acer negundo L. MANITOBA MAPLE, BOX ELDER Map 542. Introduced from the Prairies as a shade tree, though very brittle; naturalized along Indian River, freguent before 2,4-D spray (1952) along roadsides near Alberton. (Negundo aceroides Moench, MacSwain); Hurst. Forma sanguineum L. Martin Montague. BALSAMINACEAE (GERANIACEAE, MacSwain and Bain, MacSwain, both in part) [Impatiens pallida Nutt. Not seen, but to be sought. MacSwain, HurstJ 192 541 542 Impatiens capensis Impatiens parviflora 544 Rhamnus cathartica 546 193 89470-13 Impatiens capensis Meerb. Map 543. Common throughout. Damp thickets along streams. (/. fulva Nutt., Bain, MacSwain and Bain, MacSwain; /. bi flora Walt., Hurst); Campbell. Impatiens parviflora DC. Map 544- Charlottetown and environs: naturalized from Europe (originally central Asiatic) before 1912. Covering dampish ground in shrubbery; ditches, thickets. 1 Lot (Groh). Hurst. RHAMNACEAE Rhamnus alnifolia L'Her. Map 545. Local, western and eastern lowlands. Swamps and wet woods, in shade rarely fruiting. Harbors the crown rust of oats, Puccinia coronata, in aecial stage. Gorham 1944, Campbell. [Rhamnus cathartica L. BUCKTHORN Map 546. Introduced from Europe for hedges; occasionally escaped (e.g., Summerside). As alternate host of crown rust of oats {Puccinia coronata), its other stations plotted by Gorham (1944): Fredericton, Cymbria, Charlottetown, Georgetown; also in Souris, at all these stations planted. 3 Lots (Groh). Hurst.] [Rhamnus davurica Pall. "Introduced in Prince Edward Island about twenty- five years ago [at the Experimental Farm] and spreading in a limited fashion on farm [the same] near Charlottetown,'' as R. cathartica var. davuria, Gorham.] *Rhamnus frangula L. European; introduced for hedges on the Experimental Farm, Charlottetown, and naturalized there in the hedgerows and wood edges. (Frangula alnus Mill.). VITACEAE harthenocissus quinquefolia (L.) Planch. Map 547. VIRGINIA CREEPER Eastern North American, introduced as a creeper for houses; persistent, and at Souris escaped to roadsides. [Parthenocissus inserta (Knerr) K. Fritsch, as to Island records probably refers to the former species. Fernald (1950).] TILIACEAE *Tilia europaea L. Dunstaffnage, edge of woods behind school. European, introduced for avenues, and escaping. MALVACEAE [Ma/va neg/ecfcr Walk. Map 548. Around Savage Harbour since 1930's (Messervy); Albany (1950). Naturalized from Europe. Waste places (M. rotundifolia, Hurst 1940; Campbell).] [Malva verticillata L. var. crispa L. Indian River mill, planted under lilacs; two plants escaped to roadside. European, introduced.] 194 Parthenocissus quinquefolia 547 548 Hypericum perforatum 549 550 Hypericum boreole 551 552 195 89470- 13Vi Malva moschata L. Map 549. Scattered throughout; planted as ornamental around houses, cemeteries; naturalized commonly along roadsides. Pink or white petals. Ma cS wain and Bain, MacSwain, Hurst. [*Abutilon theophrasti Medic. Char lot tetown, railway yards, adventive from central Canada; originally from India.] HYPERICACEAE (GUTTIFERAE, Fernald 1950) : Hypericum perforatum L. Map 550. Becoming common along road- sides and in sandy pastures, central and eastern regions; Richmond its western limit (1953). European, naturalized. Poisonous to stock. MacSwain and Bain, MacSwain, Hurst. Hypericum boreale (Britton) Bickn. Map 551. Prince County and east-central lowlands. Muddy shores and ditches, often in boggy places. (//. mutilum, MacSwain and Bain, MacSwain, Hurst. ^Hypericum majus (Gray) Britton Map 552. Local: McNeills Mills corner (P); Dalvay House. Damp sand: roadside pit and dune slacks. At both stations, as in Cape Breton, mixed with //. canadense. Hypericum canadense L. Map 553. East-central and southeastern region and western Prince County. Damp open sites; muddy fields, shores, bogs. MacSwain, Hurst. Hypericum virginicum L. var. fraseri (Spach) Fern. Map 554. Scattered throughout lowlands. Damp sites: pond shores, dune slacks, creek banks. (Elodes virginica (L.) Nutt., MacSwain and Bain; E. campanulata Pursh, MacSwain); Hurst. ELATINACEAE Elatine minima (Nutt.) Fisch. & Mey. Map 555. Glenfinnan Lake, Lake Verde; submerged (in six to eighteen inches of water) on sandy shores, though gathering fine clay. CISTACEAE Hudsonia ericoides L. Map 556. Bothwell, sand dunes. Forma leucantha Fern, originally described from here (Fernald, Rhodora 19: 76. 1917). X Hudsonia intermedia (Peck) n. comb. Bothwell, sand dunes. Always associated with //. ericoides clumps amid the more abundant H. tomentosa. (//. tomentosa var. intermedia Peck, Rep. N.Y. State Mus. 45: 86. 1893). Fernald (1950). Hudsonia tonentosa Nutt. Map 557. North Shore, locally from Bothwell to Malpeque. Sand dunes. MacSwain and Bain, MacSwain, Hurst. 196 553 Hypericum conadense 554 Hypericum virginicum var. fraseri 555 556 Hudsonia tomentosa Lechea intermedia 557 558 197 Lechea intermedia Leggett Map 558. Mermaid (0); Murray River; near North Shore, Morell to East Bideford. Sandy open sites: weedy on sandy roadsides, spruce or pine woods and exposure barren. 2 Lots (Groh). (L. minor, MacSwain, doubted by Hurst, Campbell); Groh, Hurst. VIOLACEAE Viola cucullata Ait. BLUE VIOLET Map 559. Common throughout. Damp sites, shady or open, often weedy in damp fields, stream banks. Pale- flowered forms were referred to V. Watsoni by Greene. Bain, MacSwain and Bain; (V. palmata var. cucullata (Ait.) Gray, Watson in MacSwain; doubted by Hurst). X Viola n.elissaefolia Greene. Brackley Point Road (Fernald et aL), is V. cucullata X septentrionalis, Viola nephrophylla Greene Map 560. West Prince, very charac- teristic of cedar swamps; also in ravine woods. Viola fimbriatula Sm. Map 561. Vicinity of Charlottetown, open sandy woods. Rocky Point, specimens referred by Fletcher to "V. fimbriatula X nesiotica". (V. dentata, J.M. Macoun, Ott. Nat. 13: 159. 1899). Viola septentrionalis Greene Map 562. Wellington Centre, Summerside, Hunter River. Well-drained sites: dry fields and banks. Hybrid with V. cucullata from Brackley Point Road. Robinson & Fernald 1908, Hurst. Viola pollens (Banks) Brainerd WHITE VIOLET Common throughout. Damp sites in woods, along brooks, in marshes. (V. blanda, Bain, MacSwain and Bain, Watson in MacSwain, Hurst; V. Macloskeyi ssp. pallens (Banks) M. Baker). Viola incognita Brainerd Map 564. Scattered throughout. Well- drained (rarely damp) sites in coniferous woods; characteristic under second- growth spruce. (V. alsophila, Watson in MacSwain; not referable by Hurst). Viola renifolia Gray Map 565. Common in West Prince, scattered eastward. Woods: ravine slopes, stream banks or cedar swamp. Viola pensylvanica Michx. var. leiocarpa (Fern. & Wieg.) Fern. Map 566. YELLOW VIOLET Rare: near Alberton (Watson in MacSwain); scattered in the central hilly region. Open hardwoods. (V. pubescens, Watson in MacSwain, Hurst). Viola adunca Sm. var. minor (Hook.) Fern. Original station of plants cited by Brainerd in Rhodora 15: 109. 1913 as received from Watson, not recorded. (V. Labradorica Schrank, Watson in MacSwain, Hurst; V. adunca var. glabra Brainerd); Campbell. 198 Viola cuculloto 559 560 Viola nephrophylla Viola septentrionalis 561 562 Viola pollens Viola incognita 563 564 Viola renifolia Viola pensylvanica 565 566 199 Viola arvensis Murr. Map 567. Most common in Queens County, records since 1919 (at Kinkora). Weed of fallowed grain fields, rarely roadsides; flowering early. Naturalized from Europe. Hurst 1940; (V. tricolor "not well naturalized, " from 1 Lot, Groh, Hurst). THYMELAEACEAE Daphne mezereum L. DAPHNE Map 568. New Annan, 1953 {Poole); Hunter River; Dunstaffnage (Messervy). Flourishing roadside colonies were apparently wiped out in the winter of 1944-5 (Hurst in Guardian, 21 June 1945), but it has escaped to woods at Hunter River. European, introduced ornamental. Hurst. LYTHRACEAE Lythrum salicaria L. Map 569. Scattered roadside stations, in damp ditches. Formerly (till 1955) most abundant in the marsh (now drained) at the Experimental Farm, Charlottetown, the parent colony. Introduced- from the St. Lawrence?; European. 1 Lot (Groh). Hurst. ONAGRACEAE (OENOTHERACEAE, Bain) Epilobium angustifolium L. FIREWEED Map 570. Common through- out. Open, usually dryish land; characteristic of the clearings and burns of coniferous woods. 49 Lots (Groh). (E. spicatum Lam., MacSwain and Bain); MacSwain, Hurst. Epilobium strictum Muhl. Map 571. Local: East Point, Bristol, Mount Stewart, Bunbury, Grand Tracadie, Summerside, Bloomfield. Peat bogs, marshes. Bain 1892, J.M. Macoun, Can. Record Sci., Apr. 1894, 82. Epilobium leptopbyllum Raf. Map 572. Common in western Prince County; scattered in lowland Queens; Murray River. Marshy spots, including dune slacks, thickets, pond borders, bogs. (E. rosmarini folium Pursh (not Haenke), MacSwain and Bain; E. densum, Hurst; E. strictum, Campbell). Epilobium palustre L. Map 573. Western Prince, east-central Queens. Swamp or boggy sites; dune slacks. MacSwain and Bain, MacSwain, Hurst. Var. palustre Tignish (Macoun). Var. oliganthum (Michx.) Fern. Usual form. (var. monticola, Roland, not Haussk.). Epilobium adenocaulon Haussk., vars. Map 574. Rather common, throughout. Damp, open spots; thickets; roadside ditches. 8 Lots (Groh). (E. coloratum, MacSwain and Bain, MacSwain, Hurst); Groh, Hurst. Var. adenocaulon. Common. White-petalled forms occur: York. (E. glandulosum var. adenocaulon (Haussk.) Fern., Campbell). Var. occidentale Trel. Tracadie. (E. glandulosum var. occidentale (Trel.) Fern., Roland: map). 200 Viola arvensis 567 j 569 &v Lythrum salicaria 0»rv Y » ^, f — V\\ Epilobium anqustifolium 570 Epilobium strictum Epilobium leptophyllum Epilobium odenocoulon 573 574 201 Epilobium ciliatum Raf. Brackley Point, "salt marsh" (probably dune slacks), {Macoun). Annotated by Trelease as "perhaps E. ciliatum Raf., treated by me as a dwarf crisp-pubescent form of E. adenocaulon." J.M. Macoun, Can. Record Sci., Apr. 1894, 83. Oenothera biennis L. (Oe. novae scotiae Gates) Map 575. Common weedy species, particularly in the central uplands. Dry sandy places: roadsides, barren pastures, and waste places. (Very variable in stem and leaf color, in width and shape of leaves and length of internode, in presence or absence of red papillae as bases of hairs, in abundance of strigose hairs on capsule, contrasting forms often occurring together). 34 Lots (Groh). (Oe. muricata, Groh, Hurst in part; Oe. parvi flora, Campbell in part); Bain, MacSwain and Bain, MacSwain, Hurst. Var. strigosa (Rydb.) Piper, Fl. State Washington, 407, 1906. Tignish (var. hirsutissima, Fernald, 1950 not Gray). Oenothera cruciata Nutt. Map 576. Char lottet own, Brackley Beach, Fortune Bridge. Dry sandy roadsides. On the railway siding below the Charlotte- town Experimental Farm, its dark green, glossy, lanceolate leaves, in size only gradually reduced to bracts of the inflorescence, and the red cortex of its roots, contrasted with the light green, oval-oblong, more denticulate leaves abruptly decreasing into the bracts of the inflorescence, and the brown root-cortex of Oe. biennis. But there were also plants with linear petals of various widths, whose vegetative characters were those of Oe. biennis, though no intermediates with Oe. biennis floral characteristics were found. Gleason (1952), on the other hand, reduces Oe. cruciata to Oe. parviflora L. Hurst. Oenothera parviflora L. {Oe. ammo phi lo ides Gates & Catcheside) Map 577. Common coastwise, rare inland. Beach bars, dune sand. (Oe. muricata, Groh, Hurst in part); Campbell in part. [Oenothera fruticosa L. Report in 1891 may have intended Oe. tetragona, but neither occurs east of southern Nova Scotia. MacSwain and Bain.] Oenothera perennis L., vars. Map 578. Common in the uplands, scattered, throughout. Weed of old fields, sandy stream margins, rarely of gardens. 4 Lots (Groh). (Oe. pumila L., MacSwain and Bain, MacSwain, Groh, Hurst); Campbell. Var. perennis Throughout. Var. rectipilis Blake Concentrated between Brackley Beach and Mount Stewart. Circaea alpina L. Map 579. Scattered throughout. Damp shady sites: streambank woods, swamps. MacSwain and Bain, MacSwain, Hurst. 202 Oenothera parviflora 577 578 Myriophyllum exalbescens 203 HALORAGIDACEAE Myriophyllum exalbescens Fern. Map 580. North Shore ponds and streams; also near Montague. Fernald cited specimens from Tignish and Black Pond (Rhodora 21: 121. 1919). Cool waters; well-aerated (usually on sandy bottom); on the North Shore, common in the cooler ponds, elsewhere only at influx of streams (M.W. Smith 1946). (M. spicatum, MacSwain, Hurst), Campbell. fayriophyllum verticil latum L. var. intermedium Koch Brackish millpond, Tignish (Fernald et al.). (Fernald 1950). HIPPURIDACEAE Hippuris vulgaris L. Map 581. Local: frequent near Tignish; Wellington; frequent in ponds in National Park; Cambridge (K). Muddy shores in shallow water". . .only in cold spring water"; agreeing withFassett's observation that Hippuris is usually a species of cold waters (M.W. Smith 1946). Bain 1892, MacSwain, Hurst. ARALIACEAE Aralia racemosa L. Map 582. Rare, central upland: Dunk River, Hunter River. Deciduous or mixed woods, sandstone cliffs or knolls. Bain, MacSwain and Bain, MacSwain, Hurst. Aralia hispida Vent. SARSAPARILLA Map 583. Scattered through- out, very common in Kings. Open, sandy woods, dry or damp; weedy in clearings and slash or burn. "Sarsaparilla" noted by Stewart as more abundant here than anywhere on the mainland north of Boston. The drupes collected for a tonic "wine." (Aralia sp., Stewart); Bain, MacSwain and Bain, MacSwain, Hurst. Aralia nudicaulis L. Map 584. Common throughout. Well-drained shady sites, in hardwoods or coniferous. "Ginseng in great plenty in forest, in large timber and good soil"— Stewart. (Not ginseng, however). (Panax trifolium, Stewart); Bain, MacSwain and Bain, MacSwain, Hurst. Hanax trifolius L. Map 585. Northeastern uplands only: Mount Stewart (Roland map); Fortune Bridge to Glencorradale. Well-drained hardwood sites with rich leaf mold, and spring sunshine; plants wilting and dying back about July 7th. (A dwarf ginseng). Hurst 1940. UMBELLIFERAE Hydrocotyle americana L. Map 586. Local: Prince and Queens, east to Watervale, Tracadie. Damp ground, partial shade: thickets, stream banks. Hurst 1940. Sanicula marilandica L. Map 587. Very local: West Prince; Hunter River; Dundee. Open damp woods and stream-bank thickets. 204 Hippuris vulgaris Panax trifolius Hydrocotyle americana 585 Osmorhiza cloytonii 58B 205 Osmorhiza claytonii (Michx.) C.B. Clarke Map 588. Common in northeastern upland; very local westward: Bonshaw, Portage, Pleasant View. Slopes, hollows or runs in hardwood forest. (0. brevistylis DC, J.M. Macoun Can. Record Sci., Oct. 1895, MacSwain); Hurst. Cicuta bulbifera L. Map 589. Scattered throughout, usually near the coast in lowland: swamps or marshes. Poisonous. 1 Lot (Groh). MacSwain, Hurst. Cicuta maculata L. Map 590. Common in Prince and east-central lowlands; Murray Harbour. Marshes: often in roadside or railway ditches. Poisonous. 9 Lots (Groh). Bain, MacSwain and Bain, MacSwain, Hurst. Carum carvi L. CARAWAY Map 591. Common in Prince County and around old settlements eastward. European; introduced for the fruits ("seeds") used as a flavoring (in bread, cake, etc.), now naturalized. Waste places, fields. Bain, MacSwain and Bain, MacSwain, Hurst. [Aegopodium podagraria L., collected at Brackley Beach {Hurst) , is the var. variegation Bailey, introduced as garden plant, persistent but not naturalized. Groh & Frankton. Can. Weed Surv. 5: 44. 1948.] Sium suave Walt. Map 592. Common throughout, except in uplands. Wet places: river banks, pond margins, marshy ditches. Poisonous. 14 Lots (Groh). (S. latifolium, MacSwain and Bain; S. linear e Michx., MacSwain and Bain; 5. cicutae folium (Schrank) J.F. Gmel., MacSwain, Hurst); Campbell. Ligusticum scothicum L. Map 593. Coastal: freguent in Kings, rarer westward at least to Ellerslie. Tension line at head of beach or salt marsh. 2 Lots (Groh). (Coelopleurum Gmelini, Macoun); MacSwain, Hurst. *Conioselinum chinense (L.) BSP. Map 594. Pleasant View (P), along shady banks of creek gullied into clay beds. *Coelopleurum lucidum (L.) Fern. Wood Islands, turf along top of low cliff. Macoun's specimens called C. Gmelini from Brackley Point belong to Ligusticum scothicum. Angelica atropurpurea L. Map 595. West Prince, marshy thickets or open swamps in woods. MacSwain, Hurst. Fastinaca sativa L. PARSNIP Map 596. Scattered, in central Queens, around Summerside and West Prince. European, introduced and cul- tivated for the edible roots; naturalized and weedy in waste places. 7 Lots (Groh). MacSwain and Bain, MacSwain, Hurst. Heracleum lanatum Michx. (Rickett, Rhodora 46: 389. 1944) Map 597. Local: Western Prince County; Hunter River to Rustico Island; northeastern region; and Belle River (K). Damp, open places: grassy banks along streams or above shores. (//. maximum, Campbell); MacSwain and Bain, MacSwain, Hurst. 206 ! Cicuto Uilbifera Cicuta rr.aculata 589 590 591 Ligusticum scotliicum 593 594 Conioselinum chinense Angelica atropurpurea Pastinaca sativa 595 596 207 Daucus carota L. WILD CARROT Map 598. Sparsely scattered throughout. European; naturalized. Open, dryish sites: fields, roadsides and waste places. MacSwain and Bain, MacSwain, Hurst. Var. sativa is the cultivated carrot. CORNACEAE Cornus alternifolia L. f. Map 599. Local in northeastern and central uplands and on Prince County knolls. Well-drained, hardwood sites, flourishing at the open edges, and in hardwood forest often failing to flower. MacSwain and Bain, MacSwain, Hurst. [Cornus rugosa Lam., recorded by Bain (1890), is not present, perhaps due tc absence of rocky sites (C. circinata L'Her., Bain, doubted by Hurst, Campbell.] Cornus stolon ifera Michx. Map 600. Common in Prince County, scattered eastward along the rivers, and on the northeastern uplands. Damp, open groudn; often forming thickets. (C. sericea, MacSwain and Bain); MacSwain and Bain, MacSwain, Hurst. [Cornus florida L. Northeastern limit in Maine; included by wishful thinking. MacSwain and Bain.] Cornus canadensis L. PIGEONBERRY Map 601. Common through- out. Woodlands, particularly characteristic of the floor of coniferous woods but also weedy in brush and along roadsides. The pair of leaves below the "whorl" may or may not develop. The bright scarlet berries," mawkish sweet," fatten fowl fast —Stewart. Bain, MacSwain and Bain, MacSwain, Hurst. ERICACEAE (including PYROLACEAE, Fernald (1950) and VACCINIACEAE, Bain) Chimaphila umbel lata (L. ) Barton var. cisatlantica Blake Map 602. Local: uplands of West Prince, center and southeast. Dry mixed or coniferous woods, rarely in dry old fields. MacSwain and Bain, Hurst. Grthilia secunda (L.) House, vars. Map 603. Western Prince; central and northeastern regions. (Pyrola secunda L., MacSwain and Bain, MacSwain, Hurst, Campbell). Var. secunda Dry mixed or coniferous woods, frequent. Var. obtusata (Turcz.) House Bloomfield, Dundee, in larch swamps; Harmony, mossy spruce forest. Moneses uni flora (L.) Gray Map 604. Scattered, in upland regions, West Prince, central, southeastern and northeastern. Mossy coniferous woods, in ravines or around swamps. (M. grandiflora Salisb., MacSwain and Bain, MacSwain); Hurst. 208 Heracleum lanatum 597 598 Cornus alternifolia Cornus stolonifera 599 600 Cornus canadensis Chimaphila umbellato var. cisatlantica 601 602 Moneses uni flora 603 604 209 89470-14 hyrola minor L. Map 605. Rare: Hunter River (Macoun); Harmony (K) (Fernald & St. John). Damp fir woods. hyrola chlorantha Sw. Map 606. West Prince. Coniferous woods. MacSwain, Hurst; (P. virens Schweigg., Campbell). Forma pauci folia (Fern.) Camp One of the original collections is from Alberton. hyrola elliptica Nutt. Map 607. Throughout, the most common Pyrola. Dry woods, on uplands, knolls or sandy soil. MacSwain and Bain, MacSwain, Hurst. f-yrola rotundifolia L. var. americana (Sweet) Fern. Map 608. Scattered; Prince County, Charlottetown area and the southeast. Open dry or sliqhtly damp sites: barrens due to burning or clearing; among the Betula populifolia^ accinium angustifolium, etc. (P. rotundifolia, MacSwain and Bain, MacSwain; P. americana Sweet, Hurst); Campbell. hyrola asari folia Michx., vars. Map 609. Rare: west and east, woods and swamps. Var. asari folio Pleasant View (P), rich thicket alona stream in clay beds. Var. purpurea (Bunge) hem. Dundee, larch swamp (Fernald et aL); report from Mermaid Lake (Hurst), needs confirmation, as acid peat is an unfavourable habitat for it. (P. uliginosa Torr., Hurst); Campbell. Fernald (Rhodora 50: 212-3. 1948) remarked that Roland had erroneously mapped var. asarifolia as present in the Island and var. incarnata (i.e., purpurea) from Nova Scotia, while the text indicated the contrary; Fernald (1950) made the same mistake, recording var. asarifolia from P.E.I, and var. purpurea from Nova Scotia. Monotropa uniflora L. INDIAN PIPE Map 610. Scattered throughout. Dry woods, more commonly under conifers (as a saprophyte tolerating continual shade). MacSwain and Bain, MacSwain, Hurst. Monotropa hypopithys L. var. americana (DC), Domin Farwell, Am. Mid. Nat. 10: 39. 1926. Map 611. Bedeque {Fernald & St. John); Brackley Point {Macoun); Kilkenny Road (Hurst); Harmony {Fernald & St. John). Woods, often under conifers., var. rubra (Torr.) Farwell, Roland. {Hypopithys monotropa Crantz, Copeland); MacSwain, Hurst. hterospora andromedea Nutt. Prospect Creek, pine woods. July 1888 (Macoun) J.M. Macoun, Can. Record Sci., 1897, 280; MacSwain and Bain, Robinson & Fernald, Hurst. Ledum groenlandicum Oeder Map 612. Common in the lowlands. Acid swamps, bogs and barrens due to fire or cutting. (L. lati folium Ait. , Bain, MacSwain and Bain, MacSwain); Hurst. 210 Pyrola chlorantha 605 606 Pyrola rotundifolia 607 Monotropa uniflora 609 Monotropo hypopitys Ledum groenlondicum 612 211 89470- 14VS Azalea canadensis (L.) Ktze. RHODORA Map 613. Common except in the uplands. Acid barrens, bogs, swamps. Flowering May to early June in Queens; till the end of June at North Point. (Rhodora canadensis L.« MacSwain and Bain; Rhododendron Rhodora G. Don, MacSwain; Rh. canadense (L.) Ton., Hurst, Campbell). [Azalea viscosa L., not known east of Maine. MacSwain and Bain.] Kaln.ia angustifolia L. DWARF LAUREL Map 614. Common through- out. Bogs, open or cleared swampy and acid woodland (barrens), on fire barrens in pine stumpland (Stewart). Poisonous. 44 Lots (Groh). MacSwain and Bain, MacSwain, Hurst. Kalmia polifolia Wang. Map 615. Scattered: West Prince, central Queens and eastern regions. Peat bogs. (K. glauca Ait., MacSwain and Bain, MacSwain); Hurst. *Andronieda glaucophylla Link Map 616. Western Prince: Waterford, East Bideford peat bogs. Abundant at edge of floating mat on bog-ponds, later persisting amid peat. {A. polifolia, MacSwain and Bain ?). Chaniaedaphne calyculata (L.) Moench var. angustifolia (Ait.) Rehder Map 617. Common in lowlands. Open, acid sites: bogs, or "barrens" due to burning or clearing of woods. (Cassandra calyculata (L.) D. Don, MacSwain and Bain, MacSwain); Hurst. [Lyonia ligustrina (]_,.) DC. extends northeast only to southern New England. (Andromeda ligustrina L. , Bain), 1892.] Epigaea repens L. var. glabri folia Fern. MAYFLOWER Map 618. Scattered: western Prince, central Queens and southeastern uplands. Acid, bushy barrens or mossy floor of the succeeding second-growth mixed woods. MacSwain and Bain, MacSwain, Hurst. Gaultheria procumbens L. CHECKERBERRY, BOXBERRY Map 619. Common east of Charlottetown, local in West Prince. Open, dry acid sites: dry peat or bare patches in open woods, barrens and old fields. Fruting calyx edible, wintergreen flavored. MacSwain and Bain, MacSwain, Hurst. Chiogenes hispidula (L.) T. & G. Map 620. Scattered: West Prince; central and eastern regions. Covering mossy hummocks in coniferous woods at edge of swamp or bog. Fruit edible, wintergreen-flavored. (Ch. serpyllifolia Salisb. , MacSwain and Bain, MacSwain) ; Hurst ; (Gaultheria hispidula (L.) Bigel., Campbell). Arctostaphylos uva-ursi (L.) Spreng. var. coactilis Fern. ens_ var. qlabrifolig 618 Gaultheria procumbens Chiogenes hispidula 213 [Calluna vulgaris (L.) Hull, the Scotch Heather, possibly introduced but not seen. See also Limonium. MacSwain, Hurst.] Gaylussacia dumosa (Andr.) T. 6c G. var. bigeloviana Fern. Map 622. Rare: Black Banks; Village Green; near Murray River. Peat bogs. Fruit edible but insipid compared to the next. Hurst 1940. Gaylussacia baccata (Wang.) K. Koch Map 623. Scattered, locally abundant, in Prince County, central and southeastern regions. Sandy barrens, dry old fields, bogs. The f. glaucocarpa (Rob.) Mackenz. reported by Messervy (Hurst). {Vaccinium resinosum Ait., MacSwain and Bain; G. resinosa (Ait.) T. & G., MacSwain); Hurst. *Vaccinium uliginosum L. var. alpinum Bigel. Map 624. Waterford (West Prince), peat bog hummocks. Sea-level records are rare south of the Gulf. Vaccinium myrtilloides Michx. BLUEBERRY Map 625. Scattered throughout. Common in West Prince and the eastern half of the Island. Dry acid sites: "barrens'' from clearing or fire; more shade-tolerant than V. angustifolium, persistent in open second-growth woods. Fruit less sweet than those of V. angustifolium, but abundant and edible. (V. canadense Kalm ex Torr. , MacSwain and Bain, MacSwain, Hurst); Campbell. [Vaccinium tenellum Ait., of the southern States, is an erroneous record based on V . tenellum sensu Pursh, presumably V. angustifolium var. nigrum. MacSwain and Bain.] Vaccinium angustifolium Ait., vars. BLUEBERRY Map 626. Common throughout. Dry acid sites: clearings, barrens, old fields. Following fire, extensive even-aged stands reach maximum fruit production the second year (as in Lot 12 after the fire of 1840, the bushes still persisting in 1870, or on the famous fire barrens of Tracadie, extending toward St. Peters). The most valuable native small fruit. "So abundant locally as to afford swine feed for weeks. A gallon of gin can be made from a bushel" (Stewart). Var. laevifolium House Common throughout, variable in leaf form, twig color and fruit size. (V . corymbosum, Stewart; V. pensilvanicum Lam., not Mill., MacSwain and Bain, MacSwain, Hurst); Campbell. Var. nigrum (Wood) Dole Map 627. Scattered in West Prince, Kings; colonies with leaf form of var. laevifolium. (V. tenellum, MacSwain and Bain; V . Brittonii Porter; V. pensilvanicum var. nigrum Wood, Hurst); Campbell. Vaccinium vitis-idaea L. var. minus Lodd. Map 628. North and east coasts; rarely inland: Keefe's Lake. Exposed sites: inner sand dunes and sandblown barrens; open sandy woods, old fields and banks near sea (V. Oxycoccos, MacSwain and Bain?); Churchill 1902, Adams, Hurst 1940. 214 Arctostaphylos uva»ursi var. coactilis Gaylussocia dumoso 622 Goylussocio boccata Voccinium uliginosum var. alpinum 623 624 Vaccinium myrtilloides Vaccinium angustifolium 625 Vaccinium angustifolium var. nigrum Vaccinium vitis-idaea 627 628 215 Vaccinium oxycoccos L. CRANBERRY Map 629. Locally, through- out, in peat bogs. Smaller fruit then the following. Ours are referable to, or transitional to var. microphyllus (Lange) Rousseau & Raymond (in Nat. Can. 79:(2): 82 1952). {Oxycoccos vulgaris Bong., MacSwain and Bain; 0. palustris Pers.); MacSwain, Hurst. Vaccinium macrocarpon Ait. CRANBERRY Map 630. Common, central and eastern regions, largely coastal. "Marshes adjoining upland; low wet poor sandy land" (Stewart), dune slacks and barrier-beach pond bogs, occasionally inland peat bogs, then particularly in floating mats. Fru^it large, high in benzoic acid, used for sauce and flavoring milder fruits. "Exported, and used for sauce, the overwintered berries less acid" (Stewart). (V. Oxycoccos, Stewart; Oxycoccos macrocarpus (Ait.) Pers., MacSwain and Bain); MacSwain, Hurst. On the peaty sand borders of Mermaid Lake, cranberries (Vaccinium macrocarpon) form mats among the Scirpus cyperinus tussocks, scattered Eriophorum virginicum and Rhynchospora alba. Where the peat has been trampled (see cart track in background) Juncus spp. form small clumps. PRIMULACEAE [Lysimachia vulgaris L. Charlottetown, rubbish heap {Fernald & St. John). European, garden plant, not established.] [Lysimachia punctata L. var. verticil lata (Bieb. ) Boiss. Map 631. Murray River, dump by river, cultivated in village. European, garden plant.] Lysimachia terrestris (L.) BSP. Map 632. Scattered, chiefly coastal but inland around ponds and lakes. Muddy or marshy pond shores, stream banks and ditches. Hybridizing with L. thyrsiflora. (L. stricta Ait., Bain 1892); Hurst. 216 Vaccinium oxycoccus var, tnicrophyllus Voccinium mocrocarpon 630 Lysimachia punctoto var, verticillata Lysimachia terrestris 632 Lysimachia nummularia Lysimachia thyrsiflora 633 634 217 Lysimachia nummularia L. Map 633. Rare: Montrose (P); Charlotte- town; North River (?), 1905 {Warren). European, naturalized. Ditches and stream banks. Campbell. Lysimachia thyrsiflora L. Map 634. Scattered, coastal and lowland. Damp, open ground: shores, marshes, pond margins. Hybridizing withL. terrestris. MacSwain, Hurst. X Lysimachia commixta Fern, in Rhodora 52: 199. 1950 {L. terrestris X thyrsiflora). North Lake (K) {Fernald et aL). [*Lysimachia ciliata L.Wellington, a few plants among grass in churchyard (1952). Adventive (or introduced ?) from Nova Scotia.] Trientaiis boreal is Raf. Map 635. Well-drained, woodland or mixed wood sites, where competition reduced by shade. Forma ramosa Vict. Upton, near Charlottetown. (T. americana (Pers.) Pursh, Bain, MacSwain and Bain, MacSwain, Hurst); Campbell. [Anagallis arvensis L. Map 636. North River, marine sand beaches, 1891 {Bain). European, adventive. MacSwain, Hurst.] [Centunculus minimus L. Island records, including Fernald (1950), based on Macoun's misidentification of Crassula aquatica.] Glaux maritima L. var. obtusifolia Fern. Map 637. Coastal, frequent. Sandy beach heads and salt marshes, and salt-sprayed meadow above cliffs, where plants depressed (by trampling ?) and much branched. {G. maritima, MacSwain and Bain, MacSwain, Hurst, Campbell). Samolus parviflorus Raf. Map 638. Selkirk, shore of brackish pond {Fernald & St. John). {S. Valerandi, MacSwain and Bain; S. Valerandi var. Americanus Gray, MacSwain; 5. floribundus HBK., Fassett 1928, Hurst); Campbell. [Hottonia inflata L., to be excluded. MacSwain and Bain.] PLUMBAGINACEAE (included in PRIMULACEAE, Bain) Limonium nashii Small "SEA HEATHER" Map 639. Coastal, common. Salt marsh mud., (var. trichogonum Blake, Roland; Static e Limonium, Bain, MacSwain and Bain, MacSwain; L. carolinianum, Hurst, Campbell). OLEACEAE (ORNACEAE, Bain) Fraxinus americana L. WHITE ASH Map 640. West Prince; occasionally planted elsewhere as a shade tree. ''In good land but in no great quantity" (Stewart). Thickets on clay stream banks. {F. excelsior, Stewart); MacSwain, Hurst. 218 Anagollis arvensis 635 636 Gloux maritima Samolus porviflorus 637 638 Fraxinus omericana 639 640 219 Fraxinus nigra Marsh. BLACK ASH Map 641. Scattered in western Prince; Brackley Point area. Swamps and clay thickets. Wood of no value; twigs used for baskets and brooms (Stewart). (F. americana, Stewart; F. sambucifolia Lam., MacSwain); Hurst. Syringa vulgaris L. LILAC European, ornamental, persisting around old house sites. 12 Lots (Groh). Hurst. MENYANTHACEAE Menyanthes trifoliata L. Map 642. Very local in Prince: Nail Pond; Stanhope; locally abundant in east-central lowlands. Mucky wet thickets and pond-borders. (Including var. minor Michx.), Hurst. APOCYNACEAE Apocynum androsaemifolium L. Map 643. Common in West Prince, around central uplands and at Murray River. Weedy in open soil: clearings, railway banks, and roadsides. 12 Lots (Groh). MacSwain, Hurst. ASCLEPIADACEAE *Asclepias incamata L. Avondale (Q), alder swamps. Sets seed. Asclepias syriaca L. MILKWEED Map 644. Georgetown, abundant, fruiting, and spreading along railway (located by Messervy); Brackley Point, introduced from New Brunswick (1931) and spreading in Hurst's lawn, but not setting seed. Campbell. CONVOLVULACEAE [Convolvulus sepium L. Map 645. Throughout, common in Prince and Queens. Open places, twining and climbing on tall herbs: tension line of seashore, weedy on damp roadsides and railway banks. (Represented by the pink-flowered forma coloratus Lange), (var. americanus Sims, Roland). 18 Lots (Groh). MacSwain, Hurst.] Convolvulus arvensis L. "Near New Glasgow, July". Adventive, European, (Messervy, ca. 1930). Hurst. Cuscuta gronovii Willd. Map 646. Plat River (P), Pleasant View (P), parasitic on herbs in clay-bottom alder thicket. POLEMONIACEAE [Collomia linearis Nutt. Map 647. Milton, 1950 {Bassett)] adventive by C.N.R. from region of concentration in Bay of Chaleur region. Campbell.] BORAGINACEAE Symphytum asperum Lepechin Map 648. Charlottetown (1912); Springvale, near Milton (1952). Naturalized along streams by roadside; introduced garden plant from the Caucasus. Macbride, Rhodora 18: 24. 1916. 220 Menyanthes trifoliota 641 642 Apocynum androsaemi folium Asclepias syrioca Convolvulus sepium 64S Symphytum asperum 221 [Symphytum officinale L.# introduced to gardens from Europe, weedy in 2 Lots (Groh). Without specimens, may refer to the preceding.] Ly cops is arvensis L. Map 649. Sporadic weed, naturalized from Europe: Ellerslie (1948), Rustico (before 1884, 1952), Commercial Cross (1954), Lower Montague (1909); roadsides, hayfields, gardens on sandy soil. Macoun 1884, Adams, Hurst, Groh in Can. Weed Surv. 2: 32. 1944. [Echium vulgare L. Map 650. Iona (Q), 1915. European, weedy. Groh, Can. Weed Surv. 3: 42. 1946.] Myosof/'s scorpioides L. FORGET-ME-NOT Map 651. Southern rivers of the central upland from Dunk River to DeSable and Kelly's Cross. European, naturalized along stream beds. Campbell. Myosof/s laxa Lehm. FORGET-ME-NOT Map 652. Very local, western Prince and eastern regions. Marshy, perennial streams; often in thickets. MacSwain, Hurst. Myosof/s arvensis (L.) Hill Map 653. Charlottetown, 1914 to present; Souris. European, naturalized. Weedy in lawns and waste sandy places. Groh, Hurst. Lappula myosotis Moench Map 654. Becoming frequent, West Prince (1912), Borden (1926), central Queens (1950). European, naturalized. Weed of railways, grain fields and chicken runs, probably introduced in grain. 2 Lots (Groh). Johnston Contr. Gray Herb. 70: 47. 1924. (L. echinata Gilib., Groh, Hurst, Campbell). LABIATAE [Ajuga reptans L. Garden plant introduced from Europe. Not naturalized! Hurst.] Scutellaria lateriflora L. Map 655. Common in Prince, scattered in east-central lowlands. Damp shady sites: thickets, swampy woods. MacSwain and Bain, MacSwain, Hurst. Scutellaria galericulata L. var. pubescens Benth. Map 656. Western Prince, lowland Queens and much of Kings, probably throughout. Damp, more open sites than the former: thickets, swamps, abundant at pond margins. (5. galericulata, MacSwain and Bain, MacSwain, Hurst; S. epilobiifolia A. Hamilton, Campbell). Nepeta cataria L. CATNIP Map 657. Infrequent: Bedeque, Brackley Beach, Wood Islands. European, formerly planted around houses and usually remaining near them. Very attractive to cats. MacSwain, Hurst. 222 is arvensis 650 Myosotis scorpioides 651 652 Myosotis arvensis 654 Scutellaria lateriflora Scutellaria qolericulata 656 223 Glechoma hederacea L., vars. Map 658. Rather common, at least in Prince and western Queens. European. Weed of human occupation: dooryards, gardens, hedges and roadside banks. 6 Lots (Groh). MacSwain and Bain; (Nepeta Glechoma Benin., MacSwain; /V. hederacea (L.) Trev., Hurst); Campbell. Var. hederacea Charlottetown (1914), {Fernald & St. John). Var. micrantha Moricand The usual form. {Nepita hederacea var. parviflora (Beuth.) Druce, Roland). Prunella vulgaris L. Map 659. Common, throughout. Open places: weedy in pastures, fields, roadsides, thickets. 30 Lots (Groh). Apparently the native var. lanceolata (Barton) Fern. Forma rhodantha Fern., Bear River and doubtless elsewhere. MacSwain and Bain, MacSwain, Hurst. Leonurus cardiaca L. Map 660. Formerly planted around houses*, persistent or escaping to hedgerows, as at Brackley Beach. European. 2 Lots (Groh). MacSwain, Hurst. Galeopsis tetrahit L., vars. Map 661. Common, throughout. European. Weed of waste places, roadsides, and cattle-trodden thickets. 27 Lots (Groh). MacSwain and Bain, MacSwain, Hurst. Var. tetrahit. Rare: Kilmuir {Bruce). Var. bifida (Boenn.) Lej. & Court. Common. [Lamium hybridum Vill. Map 662. Charlottetown 1912 {Fernald et a/.); Summerside, 1935 {Hurst). Adventive, European. {L. purpureum, Hurst 1940, Groh in Can. Weed Surv. 2: 33. 1944).] [Lamium purpureum L. Map 663. Summerside, gardens and dumps (1951), probably not established. European. Hurst's records based on L. hybridum specimen.] [Stachys arvensis L. At or near Campbell's Mills, New Glasgow, 1938 (S. Campbell). No specimen to vouch for this record. European, adventive. Hurst 1940.] Stachys palustris L. Map 664. Wellington, Summerside, near DeSable, North River to Charlottetown. Naturalized, European. Forming colonies in ditches by roads. 3 Lots (Groh). MacSwain and Bain, MacSwain, Hurst. [Stachys aspera Michx., record to be excluded. MacSwain and Bain.] [Stachys officinalis (L.) Trev. Map 665. European, introduced in Blythe Hurst's garden and escaped to fence line. Campbell.] Satureja acinos (L.) Scheele Map 666. Georgetown 1941 (Hurst); Portage, 1953; St. Eleanors, 1954. On the railway, adventive from central Canada. European. Hurst 1941. *Like Nepeta Cataria, never in gardens but always in the bare shady ground around houses, these species regarded by Victorin (1935) as "commensals" of man, not as introductions,, 224 658 Prunella vulgaris 659 Galeopsis tetrahit Lamium hybridum 661 662 Stachys palustris 663 664 225 89470-15 Origanum vulgare L. MARJORAM Map 667. Abundant, along road between New Glasgow and Rusticoville, colonies spreading vegetatively only, discovered 1941 by W.L. Holman. Naturalized, European. Hurst 1941. Thymus serpyllum L. ssp. chamaedrys (Fries) Vollman (Hegi, 111. Fl. Mit. Eur. 5: 4323. 1928) THYME Map 668. Established in West Prince, near DeSable, along St. Peters road northeast of Charlottetown, and in the Park. 1 Lot (Groh). Naturalized European. Light sandy soil, forming clumps. White-flowered form, Dalvay (Bassett). (T. pulegioides L.; T. Serpyllum, Groh, Hurst). Lycopus uniflorus Michx. Map 669. Scattered throughout. Damp open or partially shady sites: fields, pond margins, thickets. (L. virginicus, MacSwain and Bain, MacSwain, doubted by Hurst); Hurst. Lycopus americanus Muhl. Map 670. Scattered throughout in low- lands. Damp, open fields, swamps and stream banks. (L. sinuatus Ell., MacSwain); Hurst. [Mentha rotundifolia (L. ) Huds., not seen. Bain 1892.] Mentha spicata L. SPEARMINT Map 671. Roadsides; Richmond, Charlottetown, Stanhope Road (Macoun). European, introduced and naturalized. Mentha piperita L. PEPPERMINT Map 672. West Prince; Charlotte- town, Peakes Station, Bothwell. European. Naturalized along roadside ditches. Mentha gentilis L. Map 673. Royalty Junction, 1901; near wharves, Charlottetown, 1912 (Fernald & St. John); "frequent about Tracadie" (Churchill). {M. rubra Huds., Churchill, Hurst, doubted by Campbell); Groh, Hurst, Campbell. Mentha arvensis L. var. villosa (Benth.) S.R. Stewart. Map 674. Rather common, throughout. Native. Damp open sites: stream banks, ditches, pond margins. (M. canadensis L., MacSwain and Bain, MacSwain; M. arvensis var. canadensis (L.) Briq., Hurst); Campbell. Var. arvensis, often cultivated in gardens for flavoring; introduced from Europe. 226 Stachys officinalis 665 Origanum vulgare Thymus serpyllum 667 Lycopus uniflorus Lycopus americanus 671 672 227 89470- 15Vi SOLANACEAE Solanum dulcamara L. Map 675. Very local; Ellerslie, Charlottetown near Hillsborough Bridge since 1912, Murray River. Naturalized, European. Damp waste ground, often along streams. Poisonous. Campbell. [Solanum tuberosum L. POTATO Most important crop of the Island. South American. Springing up on dumps, never established.] [Solanum nigrum L., frequently recorded: "Too common" (Stewart), "common nightshade" (Bain); Charlottetown, 1943 (R.R. Hurst), cultivated or casual in garden. Not seen. European. MacSwain, Hurst.] [Solanum rostratum Dunal "Casual", in Charlottetown. Hurst 1940.] [Nicandra physalodes (L.) Pers. Map 676. Charlottetown, 1935. South American, introduced garden plant, casual escape. MacSwain, Hurst.] [*Hyoscyamus niger L. Map 677. Miminegash, roadside, 1919 {Perry). Adventive; naturalized from Europe around Bay of Chaleur.] [Datura stramonium L. Map 678. Occasional in Charlottetown; New Haven (1946). Asiatic, adventive from southward. Wharves, railways, roadsides.] SCROPHULARIACEAE Verbascum thapsus L. MULLEIN Map 679. Scattered throughout, more common eastward. European, naturalized. Dry or stony open sites; abundant in old fields, especially overgrazed pastures. MacSwain and Bain, MacSwain, Hurst. Linaria vulgaris Hill Map 680. Throughout, very common in central region. European, naturalized. Weed of dry waste land along roads and railways, rarely in pastures. 17 Lots (Groh). Forma leucantha Fern. Emerald Junction (Bassett). MacSwain, Hurst. [Linaria maroccana Hook. f. Brackley Beach (1927), casual introduction in Hurst's garden. African. (L. reticulata, Groh & Frankton, Can. Weed Surv. 7: 100. 1949).] 228 Mentha gentilis 673 674 Nicandra physolodes 676 Datura stramonium 678 Verbascum thapsus Linaria vulgaris 679 610 229 Chaenorrhinum minus (L.) Lange Map 681. C.N.R., now throughout. European, naturalized. Weed of railway cinders; 2 Lots, 1926 (Groh). (Linaria minor (L.) Desf., Groh, Hurst), Campbell. Chelone glabra L., vars. Map 682. Fairly common throughout in lowlands and along streams. Damp ground: marshy fields, thickets. MacSwain and Bain, MacSwain, Hurst. Var. glabra Throughout; the only variety from southeast of Charlottetown. Var. dilatata Fern. & Wieg. Prince County and North Shore: Bloomfield, Mount Stewart (Fernald etal.)t Stanhope {Bassett). [Pentstemon hirsutus (L.) Willd. Not known from the Maritimes. (P. pubescens Soland., Bain 1892, MacSwain); Hurst.] fAimulus ringens L. Map 683. Local in Prince and lowlands of Queens; rare eastward. Damp muddy stream banks, pond margins. MacSwain, Hurst. toimulus moschatus Douglas Map 684. Locally in West Prince; abundant around Bonshaw; Dundas. Pacific North American; naturalized. Springs, stream borders. Hurst. Limosella subulata Ives Map 685. North and eastern coasts. Estuaries and barrier ponds, brackish muddy sand. (L. aquatica var. tenuifolia, Macoun, Hurst); Campbell. [Veronica longi folia L. Map 686. Summerside, 1912 (Fernald et a/.). European, introduced to gardens, casual escape. Referred by Pennell, Rhodora 23: 11. 1921, to his V. maritima forma E., by the collectors to var. media (Schrad.) Hartm.; not the common or garden V. maritima forma A. of Pennell.] Veronica serpyllifolia L. Map 687. Scattered. Prince and Queens; throughout? European, naturalized. Damp grassy places along upland streams, wood roads, in old fields. Bain, MacSwain and Bain, MacSwain, Groh, Hurst, (var. nummularioides Lee. & Lem. ). \ 230 Chaenorrhinum minus Chelone glabra 681 Mimulus ringens 683 684 Veronica longifolia 686 Veronica serpyllifolia 687 688 231 Veronica officinalis L. Map 688. Frequent in Queens and Kings, mainly in uplands. European, naturalized (not native). All material seems referable to var. tournefortii (Vill.) Reichenb. MacSwain and Bain, MacSwain, Hurst. Veronica chaniaedrys L. Map 689. Sherwood cemetery; Brackley Point Road, mainly in uplands. European, naturalized before 1912. Hurst 1940. Veronica scutellata L. Map 690. Local in Prince County west of Summerside. Mucky pond shores and pools in swamps. (V. Anagallis, MacSwain and Bain, MacSwain; V. Anagallis-aquatica, Hurst). Veronica americana (Raf.) Schwein. Map 691. Common in West Prince and the northeast; scattered in center and southeast. Around shallow,, still water: streams, ditches, pools. MacSwain and Bain, MacSwain, Hurst. [Veronica Anagallis-aquatica L. records probably refer to V. scutellata.] [Veronica arvensis L. was probably listed on no authority as Hurst only mistaken- ly attributed it to MacSwain.] [*Veronica peregrina L. Bunbury, weed in the Cotton Memorial Nursery (1956), adventive from central Canada.] Veronica persica Poir. Map 692. Charlottetown, 1912; 1 Lot, 1926 (Groh). Garden weed, from the Middle East. (V. Buxbaumii Ten., MacSwain; V. Tournefortii, Groh, Hurst); Campbell. [Veronica agrestis L. Dundas, 1952 (G.C. Warren). Garden weed, European. Record possibly referable to the preceding. MacSwain, Groh, Hurst.] Me/ompyrum lineare Desr., vars. Map 693. Common, central and eastern regions. Dry, open sites (hemiparasitic on grass roots): barrens, old fields and succeeding wire-birch woodland. (M. pratense, MacSwain and Bain; M. Americanum Michx., MacSwain); Hurst. Var. lineare Common. (M. lineare Desr., MacSwain and Bain). Var. americanum (Michx.) Beauverd Mount Stewart, Tracadie, Charlottetown, Wood Islands (growing among var. lineare). (M. Americanum Michx., MacSwain and Bain). Euphrasia randii Rob., vars. Map 694. North Shore: in thin turf, exposed sands or headlands along coast. Specimens cited by Fernald and Wiegand, Rhodora 17: 187-9. 1915. Var. randii Swampy shore of Campbell's pond, Sea View (Fernald & St. John). Var. reeksii Fern. Dry brackish sand, Cape Aylesbury {Fernald et a/.). Var. farlowii Rob. Grass, East Point (Macoun). [Euphrasia rigidula Jordan Cozen's Pond near Malpeque (Fernald et al. 8027), cited by Fernald & Wiegand (Rhodora 17: 197. 1915) as E. stricta HBK.. non Host (this species). E.O. Callen considers this material incorrectly determined, and referable to the following.] 232 Veronica chamaedrys 689 690 691 692 Melampyrum lineare Euphrasia randii 693 694 233 Euphrasia canadensis Townsend Map 695. Malpegue (Fletcher), cites as of Fowler by Fernald & Wiegand (Rhodora 17: 195. 1915), Cousins Pond (det. E.O. Callen); certainly at other stations along the North Shore and perhaps elsewhere. X E. aequo I is Callen (E. "strictiformis"?), a hybrid of E. americana and E. canadensis, occurs in. disturbed habitats where both species are present. Mount Royal near O'Leary, roadsides (det. Callen). E. americana Wettst. Map 696. Common, widespread and variable; dry open fields and exposed turfy situations. (E. officinalis, MacSwain and Bain, MacSwain, Hurst). Odontites serotina (Lam.) Dum. Map 697. Central and southwestern regions, west to Summerside. European, naturalized at least by 1900. Weedy, in turf of fields, on dry roadsides or sandy places; hemiparasite on grass roots. (Bartsia Odontites Huds., MacSwain; 0. rubra Gilib., Groh, Hurst, Groh in Can. Weed Surv. 4: 38. 1947); Campbell. Rhinanthus crista-galli L., vars. Map 698. Scattered near North Shore from Covehead eastward and elsewhere in Kings (?); near Tignish. Dry fields; headlands, tension line of salt marsh. 5 Lots (Groh). MacSwain and Bain, MacSwain, Hurst. Var. crista-galli The form lacking anthocyanins, occasional near Tignish. 'Var. fallax (W. & G.) Druce Only from eastward. OROBANCHACEAE Epifagus virginiana (L.) Barton Map 699. Recorded from West Prince and vicinity of Charlottetown; perhaps co-extensive with Fagus. Parasite on roots of beech. Orobanche uniflora L. Map 700. Langley Beach near Southport, solitary specimen, June 1928 (Messervy). Root-parasitic on herbs, in damp meadows. LENTIBULARIACEAE Utricularia minor L. Map 701. Watervale (Q), marshy creek; Mount Stewart, pools in marsh. Macoun, Hurst. Utricularia cornuta Michx. Map 702. Rare: East Bideford; Mermaid Lake. Mucky spots in acid peat bogs. 234 Euphrasia canadensis Euphrasia americano 695 Odontites serotina Rhinonthus cristo-golli Epifagus virginiana 700 Utricularia minor 701 235 PLANTAGINACEAE Plantago major L., vars. Map 703. Common throughout. Bain, MacSwain and Bain, MacSwain, Hurst. 37 Lots (Groh). Var. major Naturalized from Europe. Weed of waste ground (yards, roadsides) and lawns. Forma intermedia (Gilib. ) Pilger, the glabrous plants, Charlottetown; forma paniculata Domin, the plants with lateral spikes from lower axils of main spike, Pownal (Malte). Var. scopulorum Fries & Broberg Brackley Point, Grand Tracadie, Wood Islands; maritime sands. Similar but more elongated plants from the muddy bottom of the former millpond at Wellington had the purple petiole bases of P. rugelii. [Plantago rugelii Dene. Campbellton (Watson specimen determined by Macoun, not preserved). Should be sought.] Plantago j unco ides Lam. var. decipiens (Barneoud) n. comb., based Plantago decipiens Barneoud, Mon. Plantag. 16 (1845), (not seen). Map 704. Scattered, maritime. Sandy headlands, beach heads and salt marshes. (P. maritima, Bain; P. maritima var. juncoides, MacSwain and Bain; P. marina, MacSwain and Bain; P. decipiens Barneoud, MacSwain, Hurst; P. juncoides var. decipiens (Barneoud) Fern., Campbell, all in part only). Var. decipiens Common form. F. pygmaea (Lange) Rousseau (Nat. Canad. 59: 234. 1943), East Point, exposed headland, in sand and sandy cart track, while f. decipiens grew among the grasses. Var. laurentiana Fern. Wood Islands (Fernald, Rhodora 27: 102. 1925), Souris; North Point; transitional form at Campbellton. Headlands. Plantago oliganthos R. & S. Map 705. Coastal: Murray River, Souris, Charlottetown, Brackley Beach, Ellerslie. Salt marshes. (Wisely not distinguished from P. juncoides by local botanists). Plantago lanceolata L. Map 706. Queens and Kings Counties, scattered. European, naturalized. Open dry or waste ground, weed in pastures, fallow fields and roadsides. 3 Lots (Groh). MacSwain and Bain, MacSwain, Hurst. RUBIACEAE Galium triflorum Michx. Map 707. Throughout, in Alberry-Charlotte- town soils. Well-drained sites on floor of mixed or hardwood forest. MacSwain and Bain, MacSwain, Hurst. ^Galium mollugo L. var. erectum (Huds.) Briq. Southport; Charlottetown Experi- mental Farm, weed in hay and grain fields. European, naturalized. (G. erectum Huds.). Galium palustre L. Map 708- Common throughout. Open marshy places; along streams and ditches. 10 Lots (Groh). Hurst. 236 Plantago juncoides var. decipiens var. laurentiana 703 704 705 Plantago oliganthos 706 Plantago lanceolata Galium palustre 707 708 237 Galium trifidum L. Map 709 Scattered around the coast. Wet open places: marshy pond, dune slacks, springs. MacSwain and Bain, MacSwain, Hurst. Var. halophilum Fern. & Wieg. Tracadie Beach {Churchill). Intermediate material from Bothwell. Galium tinctorium L. Map 710. Scattered, Prince County and east- central lowland. Damp muddy spots: pond margins and thickets. (G. Claytoni Michx.); Hurst 1940 (?). Galium labradoricum Wieg. Map 711. In dune slacks at Nail Pond (West Prince); bogs, Mount Stewart to St. Peters. Fernald 1950, Campbell. Galium asprellum Michx. Map 712. Fairly common in Prince County and east-central lowland. Damp alder thickets; marshy ground along streams. MacSwain and Bain, MacSwain, Hurst. *Mitchella repens L. Portage, mossy hummock in damp mixed woods. [Cephalanthus occidentalis L. Not known east of Queens Co., N.S. MacSwain and Bain.] CAPRIFOLIACEAE Diervilla lonicera Mill. Map 713. Most frequent in central regions; open, dry or upland soils at border of thickets. (D. trifida Moench, MacSwain); Hurst. Lonicera villosa (Michx.) R. & S., vars. Map 714. Local in Prince County and east-central lowlands. Swamps and bogs. (L. caerulea, Hurst); Camp- bell. Var. so/on/'s (Eaton) Fern. Usual type in the east-central region. Var. calvescens (Fern. & Wieg.) Fern. Miscouche bog, associated with the last. Lonicera canadensis Bartr. Map 715. Scattered in Prince (especially in West Prince) and northeastern upland. Stream banks and ravine slopes in mixed woods. (L. ciliata Muhl., MacSwain); Hurst. Symphoricarpos albus (L.) Blake Introduced for facing house-walls and for hedges, persisting but seldom escaping to roadsides. 4 Lots (Groh). Ours all var. laevigatus (Fern.) Blake. (S. rivularis Suksd., Roland). (5. racemosus Michx., Groh, Hurst); Campbell. Linnaea boreal is L. var. americana (Forbes) Render Map 716. Scattered throughout, lowlands and ravines. "Mossy-iloored fir woods'7, often covering hummocks. (L. borealis, MacSwain and Bain, MacSwain); Hurst. 238 Galium trifidum Galium tinctorium 709 710 711 712 713 714 Linnaea borealis 715 7U 239 Viburnum alnifolium Marsh. Map 717. Very rare and confined to southeastern upland: Surrey, mixed woods in ravine (Fernald & St. John); Lewes (Profitt). Viburnum cassinoides L. Map 718. Common throughout, except central upland. Swamps, damp acid barrens and clearings. (V. nudum, Bain, MacSwain and Bain); MacSwain, Hurst. Viburnum trilobum Marsh. HIGH-BUSH CRANBERRY Map 719. Frequent in Prince County; very local eastward: Rustico Island (Hurst), Montague (Hurst), Bangor (MacDougall). Sometimes planted as ornamental, and escaped: road to Charlottetown airport (R.R. Hurst). Damp thickets and open swampy woods. Cranberries "smaller less pleasant on very ornamental bushes" (Stewart). (V. acerifolium, Bain, MacSwain and Bain, Hurst, Campbell; V. Opulus, Bain 1892; V. Opulus var. americanum Ait., Hurst); Campbell. Sambucus canadensis L. ELDER Map 720. Common in Prince and Kings; local in Queens. Moist, open places along streams, swamps. 12 Lots (Groh). Fruit the basis of elderberry wine. Foliage poisonous. MacSwain and Bain (with doubt), Groh, Hurst. Sambucus pubens Michx. Map 721. Common throughout. Well-drained sites, clearings or open hardwoods. Red berries unpleasant. Bain, MacSwain and Bain; (S. racemosa, MacSwain, Hurst); {R.R. Hurst). CUCURBITACEAE Echinocystis lobata (Michx.) T.