944 R44 C.2 Cflp^y *. C*>t~*j, ■»^t-^<-^»v*-^< /U+J. Gf>U 3/v$ THE ORIGIN OF THE BRITISH FLORA. THE ORIGIN OF THE BRITISH FLORA BY CLEMENT REID, F.R.S., F.L.S., F.G.S., OF THE GEOLOGICAL SURVEY OF ENGLAND AND WALES. LIBRARY NEW YORK BOTANICAL GARDEN f LONDON : DULAU & CO., 37 SOHO SQUARE, W. 1899. C. i_ LOKDOX : Prin'ed by Steaxqkwavs and Sons, Tower St., Cambridge Circus, W.C. PREFACE. While embodying in this book the results which I have accumulated during the past twenty years, I should like to take the opportunity of thanking the many friends who have assisted me. The first to do so were Mr. Carruthers and Professor A. G. Nathorst, whose work, in fact, led me to undertake these studies. In the troublesome work of determining the plants I have been greatly aided by the constant courtesy and assistance of the officers of the Botanical Department of the British Museum, especially of my friends Mr. E. G. Baker and Mr. A. B. Rendle. At Kew also I have received every facility for the work, and to Mr. J. G. Baker, the late keeper of the Herbarium, I owe much. Messrs. G. and H. Groves have also assisted me at various times with specimens of recent plants which I was unable to obtain for myself, and others have been received from friends whose names are too numerous to mention. With regard to the geological material that I have obtained from others, specimens have been vi Preface, received from so many sources that I must leave the reference at the head of each locality to speak for itself, only acknowledging the special aid that has been given by Mr. James Bennie, in collecting the plants of the ancient silted-up lakes of the Scottish Lowlands. For the constant encourage- ment of Sir Archibald Geikie, Director-General of the Geological Survey, I am also very grateful. CONTENTS Chap. Page. I. — Introduction . i II. — The Present Flora of Britain . . 10 III. — Means of Dispersal 20 IV.— Changes in Geography and Climate . 33 V.— Deposits Containing Fossil Plants. . 48 VI.— Former Distribution of British Plants. 97 Appendix.— Table Showing the Range in Time of the British Flora 171 THE ORIGIN OF THE BRITISH FLORA. CHAPTER I. Introduction. In the year 1876, happening to be engaged on the Geological Survey of East Norfolk, I was led to commence observation on the plants of the Preglacial ' Cromer Forest-bed.' At first I confined my efforts to collecting the animals and plants, some of the latter being afterwards determined by Mr. Carruthers. But it soon became obvious that, in order to obtain any satisfactory knowledge of the subject, it was necessary to collect and study the ripe seeds and fruits of our British plants, and to devote much of my leisure to the work of comparison ; fossil seeds had seldom been collected in this country, and recent plants with perfectly ripe seeds were seldom to be found in our herbaria. From a study of the plants of the Cromer Forest-bed, the work gradually expanded into an examination of any Newer Tertiary plants that could be found in Britain, and as during the past twenty years my employment on the Geological Survey of England has necessitated a close scrutiny of our Newer Tertiary deposits, especially in the south and east of England, I have been brought continu- ally face to face with the problems of the origin of our S: B ! 2 Origin of the British Flora. fauna and flora, and the relations these bear to the climatic changes through which this country has passed. Moreover, this life spent principally in field, and moor, and forest has forced me to observe how each changing season is marked by corresponding adaptations in the animals and plants, such as enable the species to preserve themselves, to multiply, and to spread ; or, if adaptation fails at any point, through some climatic irregularity, how sweeping and rapid may be the extermination of all except some few accidentally favoured individuals. While col- lecting seeds and fruits for comparison with the fossils I was compelled particularly to observe their many adaptations for dispersal, and also their times of ripening, and the abundance or scarcity of ripe seeds. It was impossible under such circumstances to avoid seeing the close connexion which must exist between the present geographical distribution of plants and animals and bygone changes in climate and in physical geography. Edward Forbes' * essay was read and read again ; but it soon became apparent that his brilliant generalisations, though far in advance of the date when they were written, were only partially true. Much of his reasoning was fallacious. To explain the presence of Arctic and of Iberian plants in Britain, he showed that outliers of the Arctic flora stranded on our mountain peaks could be accounted for by an appeal to the climatic conditions of former days, when a similar flora covered the whole of our Islands, and was not confined to isolated mountains. He did not see, apparently, that the use of this reasoning precluded the use of the * ' On the Connexion between the Distribution of the existing Fauna and Flora of the British Isles, and the Geological Changes which have affected their area, especially during the epoch of the Northern Drift.' — Mem. Geo/. Survey, Vol. I., pp. 336-432 (1846). Introduction. 3 converse hypothesis of a warm climate continuous from Preglacial times to account for the Iberian plants in the west of Ireland and in Cornwall. Either might be true, but scarcely both ; for the Irish and Cornish plants are not such as could survive a colder climate like that postulated by Forbes to explain the migration of the Arctic species. We have obtained direct evidence, since Forbes wrote, that all Ireland was at one time strongly glaciated, and also that Arctic plants once occupied the lowlands of Devonshire. This problem of the origin of our flora is one which can be solved, I think, by the historical method, and that seems to be the proper mode of attacking it. No doubt the imperfection of the geological record is so great as to make the task an exceedingly difficult one ; for nowhere have we yet discovered a continuous sequence of deposits, all fossiliferous, such as would give a connected history of our recent animals and plants from their first appearance in Britain to the present day. The exact order of succession of the deposits, of the physical changes, of the climatic alternations, and of the waves of migration, is still uncer- tain ; though a definite historical record is gradually being built up by the comparison and correlation of numerous overlapping chronicles, each recording at most some three or four of the subordinate stages or periods. This work of correlation, as already mentioned, has been greatly facilitated by a detailed examination of extensive areas, and a close study of the geology of the more recent deposits. In this way I have been enabled to trace the connexion between the strata, and often to speak with confidence as to the date of groups of fossils which otherwise would have had to remain as isolated finds. My own researches have been largely aided and supplemented by the examination of material obtained from friends working in districts which I 4 Origin of the British Flora. have had but slight opportunity of studying. This has especially been the case with regard to the lacustrine deposits of the Scottish Lowlands, so minutely examined by Mr. James Bennie. The results of these investigations will be found summarised in Chapters IV. and V. of this work. In the examination of our recent flora I have looked at the plants mainly from the point of view of the field- naturalist. Their climatic and geographical distribution ; the periods of ripening, and the means of dispersal of their seeds ; their competition with other plants ; and their dependence on, or destruction by animals, were the circum- stances especially noted — more so than critical distinctions of varieties and sub-species. Not that these distinctions are considered unimportant, but mainly because of the difficulty of studying them without a complete herbarium, too heavy to transport during constant changes of station. Moreover, botanists have almost ignored the essential distinction between a varietal form due to local conditions, and a true sub-species or race ; for many of our named sub-species have evidently no more claim to such rank than have luxuriant garden specimens. Forms, for instance, of the water-crowfoot (Ranunculus aquatilis) or of the lesser spearwort (R. Flammula) growing in a well-manured horse- pond or ditch have no claim to rank as sub-species, unless they can be found also under more natural conditions, and come true from seed. Again, the prostrate maritime form of broom found in Cornwall (Cytisus scoparius, var. pro- stratus) has similarly no claim to varietal rank, for Mr. Mitten tells me that seeds gathered by him grew in his garden into the common erect form of broom. A botanical visit to the Dingle Promontory, in Kerry, in company with Mr. Edmund Baker, produced several instances of this sort. We examined Saxifraga innbrosa and its allied Introduction. 5 forms, of which we found several, each occupying well- defined small areas, and apparently possessing definite characters. But, as more and more of the patches were examined, these distinctions were found to melt away ; for each fresh patch yielded a slightly different form, so that finally we were able to obtain a nearly complete series of intermediates seeming to connect the extreme .S". umbrosa with the extreme 5. Geum, all of them living within a small area under similar conditions. Pinguicula vulgaris and P. grandiflora, on the other hand, we found growing together in abundance, and quite distinct except at one spot where, below a rock on which both grew, we found a number of hybrids. In this case the allied forms, some- times only ranked as sub-species, are both good species, and have different geographical distributions, though they over- lap at more than one point. Botanical books are full of similar anomalies, often due to a natural desire to announce the discovery of a form new to Britain ; but for the student of geographical distribution varietal names founded on such material are worse than useless. For they tend to confound sub-species, which, if found in isolated areas show, in all probability, a transportation of the seeds from one to another, with varieties or forms, which will reappear wherever the parent species is subject to particular con- ditions. A flora like that of the British Islands may be studied in so many different ways, that it will be well to define at once the standpoints from which it is viewed in the following pages. I do not propose, nor do I feel competent, to touch on the questions of the evolution of the species, or of their relationship to each other ; what will be attempted in Chapter II. is, to give a sketch of the existing flora as a whole, to note its composition, and the distribution of the species. Chapter III. will deal with the means of 6 Origin of the British Flora. dispersal of the various species which constitute our flora, with special reference to the present and past distribution of the plants. Finally, I propose to give an historical account of each species as far back as geological evidence will yet allow it to be traced. It may be considered presumptuous to attempt such a task ; but, though the following Chapters are most imper- fect, yet they may do good by directing attention to lines and methods of research which are as yet little appreciated. The section on the geological history of our flora, being a record of the actual distribution in space and time of our plants from direct observation, will perhaps be the one to which botanists will most readily turn. It may be sug- gested, however, that the section on means of dispersal is equally important, and that the connexion between the different Chapters is so close that it is impossible properly to appreciate the relationship of the living plants to their fossil representatives without a study of the subject from various points. Though the present volume is professedly occupied with a discussion of the origin of the British flora, it should not be forgotten that in questions of geographical distribution it is impossible to separate animals from plants, for many plants are directly dependent on certain animals for means of dispersal. Moreover, certain animals are dispersed by the same means as flowering plants, have the same difficul- ties to contend with, are no less dependent on climatic conditions, and are almost equally tied to a single spot during the lifetime of the individual. The land mollusca in particular are in these respects so like the more sedentary species of flowering plants that I have not hesitated to speak of them where they help to illustrate the subject under consideration. Beetles, I believe, would also be of use ; but of this order I have unfortunately no knowledge, Introduction. 7 and at present few of the numerous fossil species occurring in our Pleistocene deposits have been determined. Fresh- water mollusca, freshwater fish, and amphibia seem to obey the same laws of geographical distribution as aquatic plants : the species are usually of wide range, provided the barriers are not excessively broad or high, and the climatic conditions are suitable. The geological sketch has been greatly condensed ; for it is obviously impossible to deal with so complicated a subject in a limited space, and all that can be done is to give some indication of the climatic conditions, local peculiarities, and character of the flora at each spot where plant-bearing deposits are found. The thorny subject of bygone alternations of climate is perforce discussed, for it lies at the root of our inquiry. I have also been obliged to deal with another equally vexed question, the submer- gence or elevation of the land in Pleistocene times; for this obviously has a most important bearing on the possible survival of plants within our Islands. In discussing the past climatic changes, while giving the preference to the evidence derived from remains of plants belonging to existing species, I have not hesitated to supplement this by an appeal to other groups of organisms, or to inorganic geology ; for an assemblage of Arctic mammals, a group of Arctic or desert mollusca, a morainic deposit, or erratics brought by floating ice in an Arctic sea, are as good evidence of climate as a group of plants, and are often discoverable in strata in which no plants are preserved. Perhaps it will be asked why, if the British flora is to be treated from standpoints which involve a consideration of climatic and geographic changes such as cannot be merely local, a still wider view is not taken, and this flora dealt with as a mere outlier of the Pala^arctic one ? To this I may reply, firstly, that the fossil plants of the periods 8 Origin of the British Flora. dealt with are at present almost unknown outside Britain, Sweden, and North Germany, and speculation would have to take the place of an appeal to direct evidence. Secondly, that Britain is not by any means simply an outlier of the continent of Europe. Its flora is an insular one of peculiar character, unlike that of any part of Europe, and unlike that of an oceanic Island. Few, if any, of the species are confined to Britain ; but the Islands contain a selection of the continental species best adapted for dispersal, and best able to hold their own in a changing climate. Britain, within the lifetime of existing species, has been subjected to many fluctuations of climate, which have left their mark on the flora. On these fluctuations was superimposed a series of orographic changes, such as must have tended greatly to modify local conditions, and must sometimes have aided, sometimes have hindered, the dispersal of the seeds. The following pages deal, therefore, with an insular flora of exceptional type ; in the building up of which selection and sweeping extermination have played so vigorous a part, that the flora now consists largely of an assemblage of the more readily dispersed of the Palsearctic species. Time has not permitted any large amount of variation or formation of sub-species in these Islands ; and in this our flora is totally different from the more ancient floras of oceanic islands, which were beyond the reach of such violent climatic fluctuations as have affected Britain. There is one point which needs explanation before we proceed further. I have been obliged in the following pages to go back to the popular and original use of the term ' seed.' Of the two senses the popular one seems to be by far the most useful scientifically, for it refers to the thing that is sown, not to an embryo with or without Introduction. 9 certain appendages and coverings, which in function may be quite indistinguishable from others belonging to the fruit. A seed, therefore, for our present purposes is the one-seeded unit of dispersal. All our British fruits, with the single exception of that of the Cornel, divide into such one-seeded portions, which tend to be dispersed separately, so that the young plants do not interfere with each other. These units may be seeds in the strict botanical sense, or they may be complete one-seeded fruits ; sometimes they are stones or carpels, one-seeded, or at any rate with only one of the seeds properly developed ; in other cases they include the dried calyx, or other parts of the flower or receptacle. Constant explanation would be needed if an attempt were made to define botanically what part of the fruit is referred to in each case — it is more convenient to accept the perfectly understood popular usage. lO CHAPTER II. The Present Flora of Britain. WHEN the British Flora is carefully studied, it is found to be composed of numerous elements, and can be divided into several well-marked groups. The grouping of the species, however, varies according to the point from which they are viewed. Disregarding purely botanical affinities, which are not under consideration in this volume, the assemblages necessarily differ according as the flora is looked at from the standpoint of relationship of the plants to climatic conditions ; or from the standpoint of habitat, including variations in soil, and shelter ; or again, from that of local distribution. No one of these methods will enable the plants to be grouped into 'provinces' satisfactory for all purposes. Each set of conditions overlies and modifies the distribution which either of the others alone would tend to bring about. If we begin with the broadest classification, that based on climatic conditions, we find at once that this is not merely a question of average, or of extreme temperature. It is temperature plus amount of moisture, modified in various ways by the season at which the rain falls, the amount of sunshine, and the season at which the sun is felt. A flowering plant has varying needs at different seasons ; and the satisfying of these is so essential to the existence of the species — not necessarily, I would remark, the same thing as essential to the existence of the in- dividual— that, if the conditions are unfavourable for any The Present Flora of Britain. 1 1 one of them, the plant cannot maintain itself. The seed must have the right temperature, soil, and amount of moisture to enable it to germinate and grow. The young plant must have sufficient vigour to defend itself against parasites or aggressors — not like the wheat which cannot grow among our ordinary weeds, and depends on human protection. The climatic conditions at the time of flowering must be favourable, or the ovule may not be fertilised. For the ripening of the seed a certain critical temperature must be reached, and maintained for a sufficient time. The cold or wet in the winter must not be such as to destroy the seed before it has germinated. All these conditions must be favourable or the plant can- not establish itself. An annual plant must seed every year, and go through the whole round safely, or it will be destroyed. A perennial plant need seed and grow from seedlings only once in a generation. As instances of what is meant by these remarks I will take a few common plants. The horse-chestnut grows well even as far north as Bergen in Norway, and in Britain it produces abundance of ripe seeds every year ; but even in the south of England, as far as I am aware, it never succeeds in establishing itself from self-sown seeds. The common elm (Ulmns campestris), on the other hand, in England only produces perfect seed about once in forty years. Forty years is far less than the lifetime of an elm, and if the tree seeds once in a lifetime, and the seed germinates, the species may establish itself. Perfect seeds have not come under my observation, and I cannot there- fore say whether this elm does grow from seedlings. It is generally said only to occur where planted. The butcher's broom (Ruscus aculeatus) is an instance of a plant which just manages to hold its own. After watching its fruiting for twelve years in succession, I find that as a 12 Origin of the British Ftora, rule only about one plant in fifty produces any fruit, and these are not only few in number, but, as they ripen in November, an early winter may prevent them ripening at all. The plant being perennial and hardy can survive, but it has evidently reached its northern limit in Britain.* The sycamore, maritime pine, and common rhododendron (R. ponticiwi) are instances of plants undoubtedly intro- duced, which seed and grow freely from seedings in the South of England. That they were not till lately members of our flora is evidently due to geographic, not to climatic conditions. We cannot point to any British annuals which do not seed freely in some part of the Islands, for the sufficient reason that an annual which cannot seed well may be entirely exterminated by a single exceptional season. This points to a probable explanation of the curious tendency noticed in the floras of small oceanic islands, for genera ordinarily annual and herbaceous to be repre- sented by perennial species. This may be explained in the following way. In many annual plants a few in- dividuals become biennial ; these in an island devastated by an exceptional gale at flowering time, by a swarm of locusts, or other adverse conditions, would be the only ones to survive, and natural selection would thus tend to perpetuate the biennial or perennial forms which so characterise these islands. This change of annual into perennial forms, however, in all probability has had little effect on the British plants ; for the Islands, besides being too large, are sufficiently close to the Continent to receive occasional seeds or pollen of the same species, which by intercrossing would tend to keep the species true. * The exceptionally warm and dry summer and autumn of 1898, however, caused Ruscus to fruit so freely in Hampshire that I counted upwards of forty ripe berries on each of several plants. The Present Flora of Britain. 13 Climatic conditions cause two very distinct floras to be represented in Britain. The lowland flora is in the main the temperate flora of the neighbouring lowlands of Belgium and France. The upland flora, on the other hand, consists of numerous more or less isolated outliers of the flora which overspreads the lowlands of the Arctic Regions and occupies the mountains of Scandinavia. This latter assemblage is found at higher and higher elevations as it is traced southward, and is confined to hills sufficiently high to have an average temperature approaching that met with at the sea-level within the Arctic Circle. As the fall of temperature is about iQ Fahr. for every 300 feet of elevation, a sub-arctic climate is found over a considerable area in Scotland, and on a certain number of isolated hills in England, Wales, and Ireland. The seeds of the British Alpine plants are invariably small and usually very minute, a peculiarity that will be again alluded to. Local conditions govern the distribution of large groups of species. First, there are the sea-coast plants, which are all confined to a narrow belt near the sea. This flora is very uniform throughout Britain, though some of the species are found only on the south coast and a few only on the east. The seeds of maritime plants are of various descriptions, and often of large size. Many of them are scattered far and wide by the sea, though the plants only establish themselves where a suitable habitat occurs. Thus the sea-coast flora includes a good many plants like the sea- kale {Crambe maritima) , which tend to appear sporadically wherever the habitat is suitable and to disappear again after a few years — as though dispersal were easy, and the range of the species was limited by climatic rather than by other considerations. Many of the sand-dune or shingle-beach 14 Origin of the British Flora. species are more properly desert plants, and are only confined to the coast because in Britain we have no other suitable regions. The aquatic flora consists largely of species of wide range, which have a remarkable power of reaching isolated rivers, lakes, or ponds. Though some of these species are confined to limited areas, most of them tend to re-appear wherever the local conditions are favourable. They are apparently more limited in their northerly range by un- favourable climate than by difficulty of crossing barriers. Several of the aquatic plants of limited range are almost confined to the East Anglian broads and rivers ; but this limitation is evidently due to the more extensive and connected waterways of that district, rather than to other conditions. Not one of our aquatic plants is a member of the Alpine flora, or belongs to the Lusitanian group found in Cornwall and in the West of Ireland. Among the marsh and peat-moss plants are many of which the local distribution is evidently governed by climate and geographical position, and is not dependent on soil or amount of rainfall. A large group of these plants consists of upland forms, such as the Arctic willows and sedges. Anothersetis confined to the Eastern Counties; though these are few in number, notwithstanding the large area of swampy ground there found. A third group is confined to the South-west of England, or to the West of Ireland. The anomalies in the distribution of our peat-moss and marsh plants are very striking, especially as this flora probably has been less affected by human agency than any other, except the Alpine. Man may have drained a certain number of swamps, and thus exterminated some species, principally in the Fenland ; but it is not probable that he has had much to do with the introduction of new The Present Flora of Britain. i 5 species, or the transfer to other widely separated localities of species already in Britain. Marsh plants, of all the groups, are the least likely to be introduced accidentally or on purpose by man. Many of the heath or barren-land plants might be classed equally well as marsh species, for gravelly or sandy areas tend to become peaty and waterlogged in our climate. The most marked characteristic of this flora is the occurrence in it of certain gregarious plants, which occupy definite areas in enormous profusion, though entirely absent from others equally suitable. Several of our heaths, for instance, are very local, though all of them occur abundantly where found at all. The British plants which have a marked western geographical distribution within the Islands nearly all belong to the marsh and heath groups. Of the other open-land groups, that belonging to good soil and clayey meadows is surprisingly restricted, and many of the species are probably late introductions. It is not difficult to see the reason why we have so few species characteristic of our wide areas of clayey pasture. These, till recent times, were woodland, not open prairie, and since the destruction of the woods they have been under cultivation or closely grazed. We have therefore nothing equivalent to the prairie vegetation of North America or other drier climates. Several plants confined to the eastern counties belong, however, to this group ; for there the dry cutting winds of winter probably always prevented the forest growth from extending to the sea, even where the soil was richest. The other meadow species have generally a wide range throughout Britain, wherever the climate is suitable. Our woodland plants are extremely difficult to deal with, partly on account of the wholesale destruction of the 1 6 Origin of the British Flora. ancient forests, partly because of the extensive planting, which has introduced trees belonging to other districts and has profoundly modified our woodland flora. To take one or two instances, the Hornbeam is one of the principal ancient trees of Essex and other south-eastern counties ; but in the New Forest it only occurs sporadically, near houses and villages, and such would seem to be its ordinary mode of occurrence in most parts of Britain. We cannot, however, say positively that it can only be reckoned as indigenous over a certain limited area, though the evidence points in that direction. The Scotch Pine is equally doubtful, for it was abundant throughout Britain when our existing peat-mosses began to form ; it afterwards disappeared throughout the south of England ; but now that it has been re-introduced it seeds freely and is fast spreading, especially in Hampshire and Dorset. It is probable that as far back as Roman times trees were planted round the villas for shade and beauty, and Roman officers would probably have given preference to southern forms which reminded them of their native lands. Thus such trees as the Horse Chestnut, Spanish Chestnut, Sycamore, Lime, and probably the Vine and Fig-tree, would be introduced. Some of the trees died out, others established themselves from seedlings and still remain ; but except through the negative evidence of the geological record there seems to be no satisfactory way of telling which of our rarer trees were thus introduced. Besides the forest-trees, we have a large number of plants which are confined to woods; we have also several species of land-snails, which are similarly restricted to ancient forest and are not found in modern plantations. The moisture and shelter of our woods make the general character of the undergrowth fairly uniform throughout Britain ; though we possess a large number of woodland The Present Flora of Britain. i j plants which are confined to a few widely separated localities. Some of the LiliacecB and Boraginece, for instance, though abundant where they occur, are curiously local, most of them being absent from extensive areas apparently as well suited for their growth as those in which they are found. In the altered state of our woods these anomalies are particularly difficult to understand, for the plants usually do not appear to group themselves into assemblages confined to special districts, and the distribution of each species has to be studied separately. Not one of our woodland mollusca or plants, unless the Arbutus be reckoned as a forest species, falls into the special groups confined to the eastern counties, to Cornwall, or to the West of Ireland. It is a question whether the absence of Lusitanian woodland species may not be due merely to the destruction of forests in Cornwall and in the West of Ireland; but this cannot be determined till the sub-fossil plants of the forests buried under the recent peat in these districts have been collected and examined. It is possible that some of the difficulties may be cleared up when we have studied each patch of ancient woodland, however small ; for by searching small isolated patches of old forest we can often find outliers of the sedentary wood- land mollusca and plants, such as probably once extended over wide areas now bare or under cultivation. A certain number of our plants are confined to lime- stone rocks or to calcareous soils ; but it will be sufficient here to remark that none of them is characteristically eastern or western, and that scarcely anything is yet known of any of them in the fossil state. In addition to these classifications according to climate or habitat, there is yet another, certain species being eastern and others western. Though we have a con- C 1 8 Origin of the British Flora. siderable number of plants which are confined to the Eastern Counties, they, or at any rate the majority of them, have not a correspondingly eastern distribution on the Continent, and so many of them occur throughout the greater part of Europe, that the present local distribution in Britain may be, after all, climatic rather than geo- graphical. The Eastern Counties are considerably drier and more sunny than the others, in this agreeing more nearly with the mainland of the Continent. Our western plants, on the other hand, are yery peculiar, for we find in Cornwall and Devon, and also in the West of Ireland, groups of plants characteristic of the Pyrenean region. These plants occur usually not as rarities but in profusion, so that in parts of the West of Ireland the common species which carpet the hill-sides are Iberian forms unknown elsewhere in Britain. There is also another peculiarity which must be taken into account when we discuss the origin of these outliers — though Pyrenean plants occur both in the south-west of England and in the West of Ireland, the species found in the two districts are not the same. Thus Cornwall pos- sesses two of the Pyrenean heath-plants, Erica ciliaris (another outlier of which occurs in Dorset) and Erica vagans; while the four found in the West of Ireland, Erica Mackayi, Erica mediterranea, Dabeocia polifolia, and Arbutus Unedo, are all different from the Cornish ones. The only western plants common to the two regions are three spurges, two of which are sea-coast species. Nearly all the Pyrenean plants found in the British Islands, including the only tree belonging to this group, have minute seeds, the numerous large-seeded trees and plants which are associated with them in Spain not extending into Britain. Three American plants also occur in Ireland, but the The Present Flora of Britain. 19 distribution of these is too peculiar to permit of any attempt at explanation in the present state of our know- ledge as to the former range of these species. Spiranthes Romanzoviana occurs in Cork, and in North America and Kamtschatka ; Sisyrinchium angustifoliuni is found in bogs in Gal way and Kerry, and also in Arctic and Temperate North America ; Eriocaulon septangitlare is an aquatic plant occurring in Skye and the West of Ireland, and also in North America. From the above notes it will be seen that Britain shows signs of a geographical distribution of plants largely in- dependent of that due to climate ; or, perhaps we should say, not governed by existing climatic conditions. The cause of these peculiarities will be best discussed when we have examined into the means of dispersal possessed by dif- ferent plants ; but it will be as well at once to say that the subject is beset with difficulties, and at every turn we meet with instances of anomalous distribution, such as make a botanist inclined to suggest ' accidental introduc- tion by man ' were it not that many of the species are marsh or woodland forms, long established and most un- likely to be brought by human agency in any form. Per- haps future research may show that many of the outliers were once less isolated, and that the present distribution is not so unaccountable as it seems. Such has already been shown to be the case with many mammals and mollusca, which geology proves had once a much wider distribution; but the flora of our Later Tertiary deposits has not yet been collected and studied so thoroughly as has the fauna. 20 CHAPTER III. Means of Dispersal. WHEN the adaptation of plants for dispersal is spoken of, one thinks of winged seeds, or of clinging burrs, of floating nuts, of succulent fruits which tempt birds, or of other obvious adaptations. These, however, form only a few of the contrivances made use of by nature to aid plants to hold their own and to extend their range. On considering what is necessary to the existence of a species, it soon becomes evident that modes of dispersal that seem to be merely accidental really depend on some modifi- cation of the seed or plant. They are often alternative methods without which the very life of the species would be in danger. No plant of the Temperate Regions — I do not speak of Tropical species — would be likely to hold its own for long periods if it were confined to a single station. The sweeping climatic waves which time and again have passed over our latitudes within the life-time of the exist- ing species must have compelled every one now found in Britain to move. When deep snow and ice smothered our uplands, the Alpine flora had to descend to the lowlands ; when a warmer climate returned, the Arctic plants had to leave the low ground and again climb the heights The lowland plants, on the other hand, with few exceptions, had to leave the country when the Reindeer, Arctic Fox, Means of Dispersal. 2 1 and Lemming inhabited Salisbury Plain, and the Arctic Birch and Bearberry grew in the lowlands of South Devon. The Temperate flora has returned again ; but the fact that the whole, or nearly the whole, of our plants have been compelled at least twice, probably many more times, to migrate long distances, shows that the British flora as it now exists must be a flora highly specialised for dispersal. In this respect it is probably more specialised than any tropical flora, which has been developed in an unvarying climate, but under a struggle for existence more violent to the individual. We should expect to find, therefore, that the British flora consists of a selection of the more mobile plants of Europe, without the accompanying sedentary forms. As the best illustration of what is meant, we may take the proportions of plants with minute seeds and of plants with large seeds to the total number, in orders represented both in the flora of Britain and in that of Europe ; the numbers not including plants that have seeds, either large or small, modified in special ways for dispersal over long distances. The approximate percentages are as follows : — Percentage in Britain. Percentage in Europe. 24-5 31'3 17*6 12-4 Large seeds Small seeds The composites, which at first sight appear to form an order particularly adapted for dispersal, constitute, how- ever, a much smaller proportion of the British than of the continental plants. This, I believe, is due to the general deficiency in our flora of prairie vegetation — the majority of the composites are prairie species, and until the last thousand years Britain, while possessing a temper- ate climate, was mainly woodland, so that there 22 Origin of the British Flora. was only comparatively small area suited to their needs. Before studying more minutely the means of dispersal available, it may be well to ask, in this connexion, what are the requirements that are usually essential to the life of the species. In the first place, it is necessary that the seed should be sown beyond the limit of the patch of soil exhausted by the parent plant. For this a very slight mobility is requisite. Secondly, in the case of British plants, some method is ordinarily needed by which they are enabled to cross barriers, such as rivers or straits, or tracts of desert in which the plant cannot flourish. I use the term ' desert ' as implying areas unfavourable to any particular species. A desert from the human standpoint is a sandy waste without water, which is unsuitable for the plants and animals useful to man. Such an area may be gay with flowers, and is no desert to the Gorse or Horned Poppy — the desert to them is the luxuriant meadow or forest, which they cannot overpass unless their seeds are carried by some rapid messenger. To a water-plant the dry land is a desert ; to a mountain plant the lowlands are desert ; to the lowland plants the mountain is a desert ; and to go further, to certain plants everything but limestone rock is a desert. Consequently the British Isles consist not only of an Archipelago with numerous islands, but from the points of view of different plants the area forms quite different Archipelagos, of low- lands with scattered mountain tops, of non-calcareous country with isolated limestone, or of dry land with scat- tered lakes. In gregarious plants, such as heaths and rushes, the necessity for scattering the seeds beyond the shadow of, and beyond the soil exhausted by the parent species, may Means of Dispersal. 23 mean that only the outer individuals of each cluster, presumably on the average those that have already been selected by the dispersing agency, have much chance of propagating themselves. In the case of small-seeded gre- garious plants like the heaths, without highly specialised means of dispersal, this difficulty probably tends to keep the seeds small and chaffy, so as easily to be scattered by the wind. The berry-bearing heath-plants on the other hand, though equally gregarious, have seeds fewer, larger, heavier, and with thicker walls. These latter have been modified for dispersal by birds. The small-seeded heaths without special adaptation for dispersal are often singularly local ; though occurring in profusion, they tend to occupy widely separated areas, and are absent from other districts equally favourable. The berry-bear- ing species are of more general occurrence in suitable localities, though individually they may not be so abundant. * Other species have special methods of throwing the seeds beyond the shadow of the parent plant. The Gorse, Wood-sorrel, Geranium, and Spurge forcibly eject their seeds from the ripe pod or capsule. The acorn is attached lightly for some time after it is ripe, and grows at the end of a thin branch which, lashed by the October gales, flings the acorn as boys throw clay-pellets from the end of a switch. Many umbelliferous plants have a similar mode of scattering their seed ; for when ripe the carpophore splits and the seeds hang loosely by their upper ends to the two whip-like filaments. At the same period the withered plant hardens and becomes very elastic, so that any passing animal causes it to spring back and throw off the seeds, which unless thus scattered, tend to hang on till they decay. This process one can study in a patch of these withered umbellifers, part of which is accessible to 24 Origin of the British Flora. animals, and part of which is cut off by a fence so that it has remained undisturbed. Umbelliferous plants which possess burrs, however, behave quite differently. They are less tall and springy, and, like other plants with burrs, are so arranged as to scrape the burrs against any passing animal, but usually not to fling them. Many plants have capsules so arranged as to scatter the seeds when forcibly disturbed, but not otherwise to drop them. The Poppies, Wild Hyacinth, Henbane, and various caryophyllaceous plants, have capsules erect in fruit and opening above, and the stems become stiff and elastic when the seeds are ripe. In some plants such as Erodhim, the seed can actually crawl away from the parent. Certain trees, such as the Ash, Maple, Hornbeam, and Pine possess winged fruits which when detached by a breeze tend to be carried short distances, clear of the shadow of the parent, though the seed itself is of considerable weight. They com- bine in this way the advantages of a large embryo, which gives the young plant a copious store of nutriment to draw from while it is competing with the short herbage, with a seed sufficiently mobile to reach places where it can obtain sunshine and new soil. The majority of our plants, as already remarked, have other means of dispersal, which will enable the species occasionally to overleap barriers — a faculty very different and probably far more important than the slow spreading over short stages that has just been spoken of. Here it may be pointed out that this conquest of the land foot by foot or yard by yard is insufficient to account for the present distribution of our flora. It cannot surmount barriers, and will not account for the mode of occurrence of such a plant as Erica ciliaris, which occupies in profusion two compact areas, one in Cornwall and one in Dorset, and has every appearance of spreading in each case from a Means of Dispersal. 25 single seed accidentally transported from some distant region. The British flora is full of anomalies of this sort. I may also point out as a geologist that sufficient time cannot be allowed for this method of spreading, even on the unwarrantable supposition that our plants could find a continuous belt of suitable country all the way from Central Europe, or whatever country they were obliged to take refuge in during the Glacial Epoch, to the furthest point they have now reached. Though the Postglacial period counts its thousands of years, it was not indefinitely long, and few plants that merely scatter their seed could advance more than a yard in a year ; for, though the seed might be thrown further, it would be several seasons before an oak, for instance, would be sufficiently grown to form a fresh starting point. The oak, to gain its present most northerly position in North Britain after being driven out by the cold, probably had to travel fully six hundred miles, and this without external aid would take something like a million years. I doubt whether anything like this time has elapsed since the Arctic flora occupied the lowlands of the south of England and the reindeer inhabited Central France. Most of our plants have special adaptations for dispersal over long distances, and, as the different modes of trans- portation must necessarily lead to different geographical distributions in different orders, a classification of plants and animals founded solely on method of migration ought to throw much light on some obscure problem in geo- graphical distribution. I am afraid, however, that at present we have not sufficient direct evidence and can only speak in a general way of these facilities; though new observations are made from day to day, and Darwin collected a large body of evidence on this subject.* The * Origin of Species, 6th edition, pp. 323-330. 26 Origin of the British Flora. main directions in which British plants are specially adapted for dispersal are the following : — Modification. Abundance of minute seeds (Heaths, Rushes, Saxi- frages, Caryophyllacese, &c). Abundance of large edible seeds (Oak, Pine, Horn- beam, Ivy, &c). Edible fruits with hard stones (Blackberry, Haw- thorn, Holly, Arbutus, &c). Winged seeds (many Com- posites, Willows, &c). Winged seeds with lax hairs ( Willow - herbs, Willows, Bulrush, &c). Burrs and hooked seeds. Floating seeds. Cut- leaved submerged water- plants ( Water - crowfoot, Water-milfoil, &c). Mode of Dispersal. Readily moved by accidents of all sorts. Eaten or dropped by birds ; most are destroyed, but some are transported un- injured. Eaten by birds and mam- mals ; seeds passed unin- jured. Transported by wind. Cling to feathers or fur. Transported by water. Collapse and cling when re- moved from the water ; stems fragile, and broken pieces grow. Carried on legs of mammals or of wading birds. The first group, the minute-seeded plants, is a very large one, and it will readily be understood that the plants belonging to it include nearly all the British species which show strikingly anomalous distribution. Nearly all of our Alpine plants, of the Lusitanian species found in Ireland Means of Dispersal. 27 and Cornwall, and of the peculiar eastern-county plants belong to this group, the larger seeded species found associated with them on the Continent being absent. These plants seem therefore to possess in a pre-eminent degree the power of crossing seas like that which separates Ireland from the Pyrenees. They are probably trans- ported freely by migrating birds, either on their feet or in their feathers ; but the moist-soil species must also have been carried in profusion in the cakes of mud which adhere to the flanks of oxen that have rested in a moist meadow till the earth has dried on them. Before fences were made, the migrating horses, oxen, and bisons, in this way must have carried such seeds for long distances, and any adhering to the head of an animal would be carried across an arm of the sea uninjured. It must be remembered, however, that the autumn migration of mammals, which is the migration when nearly all the seeds are ripe, would have been southward in Britain, and consequently could only carry plants in that direction. The northward migration taking place in spring, few seeds would be carried, except such as had become entangled in the fur and were shed with it next summer. Wading and swimming birds, on the other hand, commonly come to Britain from the north and east in autumn, leaving the colder districts at a time when the seeds are ripe, thus bringing the smaller ones to this country. This is probably the reason why so large a pro- portion of the minute-seeded Arctic plants are found in Britain, though many of the species only occur in small numbers and at various scattered localities. The next group, that containing the plants with large edible unprotected seeds, is a small one in this country ; but it is of especial importance on account of the difficulty the species present when we try to account for their pre- sence in these Islands, except on the hypothesis of a former 28 Origin of the British Flora. greater continuity of the land. The difficulty is so real that I have devoted particular attention to the attempt to discover in what manner large soft seeds, which cannot be carried in fur or feathers, and are killed by digestion, can be transported across deserts. It will be shown in Chapter IV. that since suitable climatic conditions came into existence there has been no sufficient change of land or sea to give a continuous land passage from the Continent for these plants — yet, here they are and their presence must be explained. The British plants to which these remarks particularly apply are the following : — the Oak, Beech, Ash, Maple, Privet, Spindle Tree, Ivy, Flags, Convolvulus, various Mallows, White and Yellow Waterlilies, and Apple. In each of these, except sometimes in the Waterlilies and Apple, the fruit is eaten for the nutriment contained in the seed itself, which is therefore generally destroyed. No doubt in many of these plants the seeds are occasionally dispersed by rivers ; but this will only scatter them along the lower part of the same river-basin or at most some distance along shore ; it will not carry Waterlilies to isolated lakes or to other river basins, nor can dry-soil plants be carried thus to scattered islands. The largest edible seed we have is the acorn ; if it can be transported freely for considerable distances uninjured, the difficulty in the other cases must be more apparent than real. In peat-mosses, on open chalk downs, and in ploughed fields, often a mile or more from the nearest mature tree, one constantly finds seedling Oaks, which last a few months or, perhaps, a couple of years, and then die, the conditions being unfavourable. I have for several years noted the position of these seedling oaks, finding them in places where no mammal would take the acorns. For instance, they are common in any of the New Forest Means of Dispersal. 29 peat-bogs that are within a mile of an Oak-tree. They are common also in some places on the top of the escarp- ment of the South Downs, half a mile from Oaks, and 300 or 400 feet above them. They are always associated with empty acorn-husks, stabbed and torn in a peculiar way. In October and November rooks feed in the Oak-trees, and I have long felt convinced that they were mainly responsible for the dispersal of acorns. On October 29th of 1895, in the middle of an extensive field, bordered by an oak-copse and scattered trees, I saw a flock of rooks feeding and passing singly backwards and forwards to the Oaks. On driving the birds away, and walking to the middle of the field, I found hundreds of empty acorn-husks, and a number of half-eaten pecked acorns. It was noticeable that many of them were not shed acorns, but were accom- panied by acorn-cups, the stalks of which had been bitten to tear them off the tree. The reason for the selection of acorns in cups is probably that they are easier to carry —a shed acorn must be an awkwardly large and slippery thing for a rook's beak, one with a stalk will be more convenient. Several uninjured acorns were found, one, almost uninjured, had been driven by a single peck deep into the soft soil of a mole-hill. In this way oak-woods must spread rapidly ; but we still want observations as to the extreme distance to which acorns are thus carried. I have seen seedling Oaks at a distance of a mile from the nearest tree (not necessarily the tree from which the acorn came) and have found the characteristically torn husks somewhat further away.* Mr. J. J. Armistead, moreover, recordsf that he once found a young Oak in a sheltered ravine among sea-cliffs on the northern coast of Hoy, Orkney. The tree was * Nature, No. 1358, vol. liii., p. 6 (1895). t Zoologist, p. 19 ( 1 891). 3