. 3 : i { si 3 il { i F 4 3 i d (S EDS Ee Mol 1: No. 2 1912 APRIL The Scottish *" o« Botanical Revs A QUARTERLY MAGAZINE INCLUDING THE Transactions of the Botanical Society of Edinburgh Edited by M‘Taccart Cowan, Jr. EDITORIAL COMMITTEE WILLIAM Barciay. CARLETON Rea, B.C.L., M.A. ARTHUR BENNETT, A.L.S. A. Lorrain SmitTH, F.L.S. A. W. Bortuwick, D.Sc. W. G. Smiru, B.Sc., Ph.D. H. N. Dixon, M.A., F.L.S. James Stirton, M.D., F.L.S. Rosert H, MELpRum. G. S. West, M.A., F.L.S., A.R.C.S. EDINBURGH NELCE SAO, PD. BELLEVUE Price 2s.6d. Annual Subscription, payable in advance, 7s. 6d., post free in U.K. in amalgamation with a7 R. S. RICHARDSON & CO. CUTLERS, Etc. 52 Georae St., Edinburab ‘Phone 2489. Ay, Adie Wedderburn Optical Repantiadnn Spectacles and Eyeglasses Lorgnettes, Reading Glasses Field and Opera Glasses Instrument Department Barometers, Barographs Thermometers, Rain Gauges Surveyors’ Instruments Photographic Department Cameras and Kodaks Plates, Films, Papers Negatives Developed and Printed R. S. RICMARDSON & CO. = = Cutlers and Cycle Agents = = 52 GEORGE STREET, EDINBURGH Established 1828 (Adjoining the Music Hall) Telephone 2489 ee _____ Superior SHEFFIELD CUTLERY of every description DRESSING CASES - - MANICURE and SHAVING REQUISITES PRINCIPAL DEPOT for the Renowned GILLETTE SAFETY RAZOR The Razor which really Shaves without Stropping or Honing AUTO-STROP, DURHAM-DUPLEX, CLEMAK AND OTHER SAFETY RAZORS ‘‘EASY-OPEN” POCKET KNIVES of many kinds Our “RING” KNIFE has attained Great Popularity CERMAN-SILVER IVORY, STAC, PEARL 1/; 2/6, 3/6 from 3/6 each also in also in STERLING SILVER COLD (HALL-MARKED) TTT TT i : ih. Practical Simplicity, Quality of Material, and Careful Workmanship have resulted — i -__ ae in thousands now being in use SPARBROOK and other CYCLES ALL ACCESSORIES, including Many Novelties REPAIRS IN ALL DEPARTMENTS UNDER PERSONAL SUPERVISION SUBSCRIPTIONS FOR i912 ARE NOW DUE. WHEN CHEQUES ARE USED EXCHANGE SHOULD BE ADDED SUBSCRIPTIONS Subscriptions are payable in advance by cheque or postal order to M‘'TAGGART COWAN, Jr., 40 Great King Street, Edinburgh. Single copies from Messrs. Neill & Co. Ltd., Edinburgh. CONTRIBUTIONS Contributions on all subjects connected with botanical science and teaching will be fully considered. A preference will be given to contributions relating particularly to Scotland. All manuscript must be clearly written, and when possible typed. Illustrations should if possible be adapted for reproduction in the text—that is, pen-and-ink drawings or photographs. Where special plates are desired, contributors should first communicate with the Editor, as only a limited number of these can be published. Authors will receive 25 copies of their communications free. Additional copies, in covers, may be had from the Printers at the undermentioned rates, provided such orders accompany the manuscript. 25 copies. 50 copies. 100 copies. 8 pages . > 3s, Od, hs, Ody 8s. od. (a % ge ooAOels 8s. Od. 13s. Od: Co re 7) gS Od: 13s. Od. 20s. 6d. All communications on Editorial matters, Advertisements, and Books Jor Notice should be addressed to M‘TAGGART COWAN, Jr., 40 Great King Street, Edinburgh. ESTABLISHED ABOVE A CENTURY. ORTH BRITISH & MERCANTILE INSURANCE COMPANY IN WHICH ARE VESTED THE SHARES OF THE RAILWAY PASSENGERS ASSURANCE COMPANY, and the OCEAN MARINE INSURANCE COMPANY, LTD. FUNDS - - £21,000,000 FIRE—LIFE—ANNUITY. ACCIDENTS OF ALL KINDS|THIRD PARTY LIABILITY | PLATE GLASS ACCIDENTS AND ILLNESS | LIFT ACCIDENTS FIDELITY GUARANTEE EMPLOYERS’ LIABILITY BURGLARY AND THEFT MARINE MOTOR CAR ACCIDENTS LOSS OF PROFITS TRANSIT OF SECURITIES. The Company now issues Combined Policies specially designed for the benefit and convenience of householders, affording Complete Indemnity against Fire, Burglary, and Accidents to Domestic Servants. LIFE DEPARTMENT. INCREASED BONUS FOR THE QUINQUENNIUM. Compound Bonus, £1, 10s. per cent. per annum, that is, on Sums Assured, and previous Bonuses—an increase of 2s. per cent. per annum on the rate for the previous Quinquennium. Representing a Bonus addition for the Quinquennium of £7, tos. per cent. on sums assured on five-year policies, and £18, 5s. per cent. on sums assured on oldest Policies. CHIEF OFFICES: EDINBURGH: 64 PRINCES ST. LONDON: 61 THREADNEEDLE ST., E.C. il The Scottish Botanical Review OY [| APRIL The Geological Relations of Stable and Migratory Plant Formations. By C. B. Crampton, M.B., C.M., of H.M. Geological Survey. PAGE Part III. (continued from ~. 16), THE RELATION OF SOILS TO CLIMATE AND PHYSIOGRAPHY . ; ‘ 57 Part IV. THE STABLE TYPES OF VEGETATION IN BRITAIN (to be continued) : : ‘ : 5 . 68 PART III. THE RELATION OF SOILS TO CLIMATE AND PHYSIOGRAPHY (continued). THE chemical effect of lime on plant-association is not merely a result of the outcrop of limestone rock at the surface. The physiography or climate may, indeed, be such that it is inappreciable, even where limestone forms the solid rock and sub-soil ; and, conversely, the kind of plant-association may indicate the presence of lime in the surface waters in places remote from limestone or highly calcareous rocks. A few species have been shown to suffer in their metabolism as a direct consequence of considerable quantities of lime in the soil; thus plants of the acid moorland and heath rarely gain a footing except where lime is wanting or has been leached from the surface (2). The presence of lime in the soil is, however, usually taken as a gauge of fertility, and this is particularly the case in cold- temperate humid regions where the surface-leaching of lime so rapidly leads to soil-acidity and humus accumulation, with VOL, I. 5 58 THE SCOTTISH BOTANICAL REVIEW increasing sterility and loss of available nutriment in the soil. In tropical rain-forest, in spite of the greater rainfall and deep leaching of the sub-soil, excessive humus accumulation and surface-acidity are checked by the rapidity of plant decay that follows a greater degree of warmth(1). This not only results in a more rapid oxidation of decaying plant-tissues and removal from the surface of organic acids, but probably also in a more complete return of mineral food to the soil. Agricultural experts have proved that lime benefits the soil in various ways, mechanical and chemical, but the latter appear of more consequence from the ecological standpoint. The importance of lime, as a chemical base, has been demonstrated as an agent of neutralisation in the acidity that follows the biological processes of humification and nitrification, and also for setting free potash from inert, zeolitic combinations in the soil. It is further considered essential for the working of the most widely distributed, known form of free-nitrogen-fixation (22). Plant physiologists believe that calcium performs important duties in certain metabolic processes in the higher plants, but the great differences in the amounts of lime in the soil, leading to the choice of habitat in various groups of calciphile plants, would negative far-reaching relations in this direction. In the case of heath and bog plants, there are several inde- pendent experimental results proving that calcium carbonate is not in itself prejudicial to their growth (23). Further, that a number of plants should be distinctly calciphile in one district, though growing elsewhere where lime is deficient, suggests that the presence of lime is not the only factor involved. We should therefore look for relations between calcium carbonate and other substances in the soil for a full under- standing of the conditions underlying the segregation of calciphile and calciphobe species, and though further ex- perimental work is needed before we can arrive at a solution of the problem, it may be well to define those physiographical factors which seem to have bearings on the subject. In cold-temperate, humid regions the so-called calciphobe species are confined to habitats deprived of calcium carbonate through leaching of the surface or stagnation of the soil waters. STABLE AND MIGRATORY PLANT FORMATIONS 59 These habitats accumulate acid humus, or may fail to do so on account of soil-porosity or the physiographic relations of drainage and exposure. In the latter cases plant-growth is discontinuous and retarded, and the removal of dead vegetation by mechanical or fungal agencies is more than usually effective. The calciphobe species thus inhabit sterile soils, where any humus that accumulates is acid in character, inhibiting bacterial action, and the plants have therefore to be of such a nature that the nitrogen, necessary for their growth and metabolism, can be obtained from minute quantities of inorganic nitrogen-salts or from organic compounds, either directly or by heterotrophic, fungal, or parasitic association (24) (13). Under the calciphobe associations we would therefore include on the one hand those of the heathland, etc., where the surface has been long subjected to leaching ; and on the other hand, those of the moorland and others, where stagnation of thesoil-waters leads toacidity and acid-humus accumulation. In the former case lime is removed from the soil by leaching ; in the latter, lime may in some cases accumulate, but is locked up in the form of humates faster than the supply is capable of neutralising acidity. It is probable that variations in the ratio between the rate of development of acidity, and that of neutralisation by calcium carbonate or other bases, lead to the differentiation in the moorland of such associations as grass- moor, calcareous bog, and various types of moorland “flushes,” and also between sour marsh and marsh-meadow, all of which depend greatly on physiographic factors for the differentiation of their habitats. The calciphile plants, from the point of view of their physiographic relations, may be roughly sub-divided as follows :— 1. Aquatics or semi-aquatics, usually found in waters holding excessive quantities of lime in solution, and including a number of algze and some mosses that become encrusted with calcium carbonate, and probably abstract carbonic acid from bicarbonate of lime in solution. These should perhaps be classed with the sinter-organisms, saline plants, and others that have evolved special protective powers against conditions 60 THE SCOTTISH BOTANICAL REVIEW unfavourable to most plant species and are now almost unable to live elsewhere where competition is involved. 2. Certain species of lichens confined to limestone surfaces. A number of these lichens eat deeply into the surface of the rock, which indicates that their metabolic processes are closely connected with their choice of habitat. In their relations to lime they are probably to be classed with the preceding group. Lichens usually avoid surfaces whose nature or exposure subjects them to exfoliation or destruction, and the colonisation of soluble limestone surfaces by lichens may require the presence of special adaptations to prevent extermination. Some of these lichens “sow their seed,” as it were, in holes in the rock, and the thallus is crustaceous, gelatinous, or evanescent, rarely, if ever, foliaceous or fruticose. 3. Certain aquatics, semi-aquatics, reedswamp, fen, and marsh plants, which flourish where abundance of lime and other salts are supplied by the waters, owing to the physio- graphic relations of the habitat to the river system. The abundance or poverty of lime and alkaline salts has been generally considered to be an important differentiating factor between fenland and acid-moorland, and this is a direct result of the different physiographic relations of their water-supply to the drainage system (3) (26). Fen associa- tions chiefly consist of species from reedswamp and marsh. The rapid growth of fen vegetation indicates considerable quantities of easily available nitrogen compounds, since in the majority of the plants there is no reason for thinking they can obtain their nitrogen by heterotrophic means. The nitrogen is therefore probably obtained autotrophically, and may be directly derived from ammonium salts or nitrates correlated with the abundance of lime and other salts in the waters. . Graebner has pointed out the rapid exhaustion of mineral nutriment that takes place in the waters of moorland pools and small lakes, and how the pools then fill with floating forms of Sphagnum which gradually supplies sufficient humus for other moorland plants to take root and spread upon the surface (2). This poverty of inorganic, nutrient salts in waters immediately derived from acid-moorland surroundings STABLE AND MIGRATORY PLANT FORMATIONS 61 should be contrasted with the fenland conditions, where such salts are constantly replenished by the river water, and induce a rapid growth of rooted plants through the greater part of the period of subaqueous humus accumulation. The humus accumulation in fen is primarily of physiographic origin, from the subaqueous nature of the habitat, and from the rapid growth and periodical extensive death that characterises the vegetation. Fen-humus is protected from erosion or much admixture with inorganic silts by the reed-belt and the physiographic relations of the habitat. An evolutionary succession in plant association accompanies the change from reedswamp to fen, with the gradual accumulation of humus up to flood water- level. The spongy nature of the humus, and its relation to the water-level, ensure a fairly constant supply of alkaline waters at various depths, and vigorous species of the reed- swamp persist in close association with an increasing number of marshland species, which invade the surface of the humus as the upper limit of the water is approached (3) (25). Fen- land differs markedly from moorland floristically and ecologi- cally, but its relations to reedswamp and marsh are very close. In any defined area, the aquatics, reedswamp, fen, and marsh are related by a similarity in their supply of inorganic food but are segregated by the physiographic conditions which lead to physical differences in the habitat. In comparison with moorland peat the humus substratum of fen appears to be chemically inert, though some of the fen plants are prob- ably saprophytes. The ammonium and potassium salts brought down in the river-waters would be locked up as inert compounds with the humus derivatives of the fenland peat, but where calcium carbonate is present in excess the ammonium and potassium would be set free as carbonates, the calcium taking their place ; or where sufficient inorganic bases were present, the original ammonium and potassium salts would be protected from combining with the humus derivatives (22). Nitrates would probably be more abundant than ammonium salts in the river-waters, but might undergo reduction in the presence of much organic matter, and nitrification of ammonium salts would also be prevented, so it seems likely that fen plants 62 THE SCOTTISH BOTANICAL REVIEW chiefly obtain their nitrogen from inorganic ammonium salts. The importance of lime, in determining fen vegetation, is possibly due to its powers of retarding the transformation of inorganic ammonium and potassium salts into less assimil- able combinations with humus derivatives, such as occur where lime is wanting and acid humus accumulates. Where the latter begins, inorganic compounds of nitrogen are largely superseded by combinations of nitrogen and humus, and moorland plants take possession. This may ensue from changes in the physiography cutting off the supply of calcareous waters and inducing stagnancy, or from a lowering of the water-level followed by leaching of the surface layers of the humus (3) (26) (13). Deeper rooted fenland species may then continue to live for a time side by side with shallower rooted phanerogams, mosses, and lichens from the moorland. The scarcity of inorganic nitrogen compounds in acid bog- waters suggests that the numerous species of algze, and some mosses, é.¢. the Sphagnacez, may be capable of using some of the more oxidised, soluble compounds of humus with ammonia, such as ammonium crenate, for the purposes of obtaining nitrogen, while the densely tufted species of mosses, with a highly developed tomentum of rhizoids, may be partially saprophytic for similar purposes. If these mosses had far to search for their inorganic food, one would not expect the closely interwoven and limited extension of the rhizoids that characterises the species that live in humus, The tomentum of Polytrichum strictum has so close a re- semblance to a mycorhiza that it may well be believed to function in a similarsmanner. Micorhize have been detected in several species of hepatics, and are probably also present in mosses (35). 4. Mesophytes that flourish in moist, well-aerated soils adapted to the formation of nitrates. In this climate porous soils need to be sufficiently deep, and well provided with humus, to retain moisture and supply nitrogen compounds, and with sufficient carbonate of lime to prevent acidity. Habitats having these soil-characters will only occur where there is an abundant supply of lime in the soil-waters, or STABLE AND MIGRATORY PLANT FORMATIONS 63 where the soil consists largely of limestone fragments. Plants with these requirements should therefore be more abundant in limestone districts than elsewhere. Damp crevices in limestone, rain-wash, screes, and the neutral humus of well- lighted woods on limestone slopes, and deep, sandy alluvia flanking calcareous waters are the most likely habitats to meet these conditions. Shallow-rooted plant associations, indicating the presence of lime in the surface waters, are frequently met with in calcareous ‘‘ flushes” where, owing to the slope of the ground, waters derived from calcareous springs, or from the surface run-off from calcareous rocks, periodically flush areas which otherwise would be sterile, through atmospheric leaching or an acid condition of the soil (26). The soil of these calcareous flushes must often be well aerated, owing to the periodic nature of the flow of waters, and their high oxygen content. Many of them are only flushed during wet weather, and the soil at other times being well aerated and warm (dry, calcareous flushes), may Ets favour the conditions for the production of nitrates. Where deep residual soils overlie limestone, or where deep soils, overlying other rocks, contain abundant lime in solu- tion brought from higher levels, the vegetation may well be luxuriant, as noted by Hilgard, since the lime promotes rapid humification and nitrification in the soil. This luxuri- ance of the vegetation may be lasting in character, provided the climate ensures a continuous supply of humus on the one hand, and alkaline ground-waters on the other. The super- ficial layers of the soil in such cases may contain but little lime, and the shallow-rooted ground associations would then be no index to its presence. 5. Plants of a distinctly xerophytic habit, and often dwarf in stature, including a number of phanerogams and mosses that are chiefly confined to dry limestone habitats or chalk, where the soil is excessively calcareous, dry, and poor in humus. On exposed limestone surfaces subject to drought, or on chalk, humus fails to accumulate under our existing climate unless a leached soil is formed (3). Such limestone surfaces contrast with the leached surfaces of other well-drained or porous soils where acid humus accumulates. The physical 64 THE SCOTTISH BOTANICAL REVIEW drought that has to be endured by shallow-rooted vegetation is increased by this poverty in humus. It is on such surfaces that the most typical group of the so-called calcicole flora is developed. But even at its closest contact the influence of limestone on plant-association is probably very complex, and may be different in different species at very short distances. Certain species that are common in the clefts of limestone pavement, eg., Oxalis Acetosella, are also abundant on the most acid type of humus, provided they are shaded from the sun. Others, as 7eucrium Scorodonia, are abundant both on limestone and on leached rocks which accumulate humus, provided they are well drained. Evica cinerea, Calluna vulgaris and Arctostaphylos Uva-urst grow on bare limestone in some places, and on leached rocks and acid humus else- where (27). The species considered most distinctive of lime- stone do not, however, often grow in situations developing acid humus. They appear to be chiefly confined to dry limestone soils, where the accumulation of humus is very scanty. The reason for this preference for limestone is unknown, but their restriction to dry limestone soils, where humus fails to accumulate, and where they may sometimes be found associating with some of the plants characteristic of very dry, heathy, or acid soils, suggests that there may be something in common between such apparently diverse habitats. What does the Bearberry find in common in such habitats as dry limestone slopes in the Alps, and the most acid type of peat bogs in the north of Scotland, other than physical and physiological drought? In the one habitat it finds abundant lime and little humus, in the other abundant acid-humus and practically no lime. ,In this connection certain statements by Schr6ter in regard to the habitats of Pzzaus montana are of peculiar interest. He writes: “How fundamentally diverse are the habitats of the mountain pine, which can grow on the dry, loose, calcareous talus of a hot, southern slope, and on high moors, which are mainly composed of bog-mosses and are subalpine bogs dripping with water but poor in mineral matter. The former soil is poor in humus, rich in mineral matter, and dry ; the latter is a substratum rich in humus, poor in mineral matter, and always saturated with water. STABLE AND MIGRATORY PLANT FORMATIONS 65 Common to both is only one character—poverty in assimil- able nitrogen” (28). Again, Schréter and Kirchner say : ‘‘One character especially must be brought into prominence which all forms of Pzzws montana have in common—namely, the poverty of assimilable nitrogen in the soil ; limestone crags, limestone screes, and dune sands all show very little nitrogen content, and in the humus of the ‘ Hochmoor’ it is present in a form available only with difficulty. Pznus montana is, however, specially well equipped against nitrogen poverty in the soil (according to P. Muller’s recent investi- gations), through its capacity to assimilate nitrogen with the aid of endotrophic mycorhiza ” (20). Nitrogen is the necessary element of plant food which for its existence in an assimilable form depends most on the local relations of the soil to plant life. The supply, and form of occurrence of nitrogen in the soil, would therefore appear to be of great importance in defining the type of vegetation that can persist over any particular area of the land surface. Apart from aquatic habitats and those liable to flooding, the relation between the humus content and the nitrogen content in natural soils is probably a close one. Where the soil is strongly acid the humus and nitrogen content may be high, but the latter in a condition physically unavailable for plant life. In soils where acidity is prevented by abundance of calcium carbonate the humus content might appear to be an index to the available nitrogen ; but since the nitrogen chiefly becomes available through the destruction of humus, and may perhaps further result directly from plant decay without the intervention of humus formation, this is by no means necessarily true. In the case of very shallow, dry limestone soils plant decay is rapid, apparently owing to the frequent aeration of the soil and the prevention of acidity, and the accumulation of humus is minimised. The available nitrogen in such a soil would therefore depend on the balance from the loss of ammonia that occurs during the processes of decay. In cultivated soils excessive liming is believed to cause a wastage of nitrogen, both as ammonia, and through too rapid nitrification and leaching depleting the stock of humus (22). In natural, shallow, dry limestone soils the loss of ammonia 66 THE SCOTTISH BOTANICAL REVIEW during decay therefore becomes a question of some import- ance, since the pabulum necessary for organisms promoting free-nitrogen-fixation would seem to be very limited. The vegetation of these dry limestone habitats is usually stunted or dwarfed, doubtless due in part to drought, strong illumination, or the effects of grazing by animals; but it seems to be a matter worth investigating whether this dwarfing be due to these factors alone, or to further factors amongst which poverty in available nitrogen may be of importance. It has also to be shown to what extent the xerophilous characters of these typically calciphile plants may be a result of the concentration of mineral solutions in the soil waters. If little loss of ammonia accompanies the rapid decay on dry limestone soils, the latter may form a habitat for plants peculiar in being fairly well supplied with ammonium salts, but subject to a physical or physiological drought preventing the rank growth usually found in habitats where there is a liberal supply of nitrogen compounds. The peculiarities of such plant societies would be, that they had plenty of avail- able nitrogen, but avoided competition with the rank growth found on all fertile soils with a good supply of water. If, on the other hand, there is a considerable loss of ammonia on plant decay, the habitat would be peculiar in being subject to drought and nitrogen starvation, but retaining a neutral or alkaline reaction of the soil. It is possible that both conditions occur on dry limestone soils from differences depending on such factors as the depth and nature of the soil, and especially its degree of alkalinity, its clay and humus content, and its local temperature or degree of illumination, in their total effects on the retention of ammonia and nitrification. Whether nitrogen-demanding or subject to nitrogen- starvation, the plants of these dry limestone soils occupy a habitat where the relations of calcium carbonate to fertility are on an entirely different footing from those of the fertile, moist, calcareous soils with abundant humus, and the so-called calcicolous plants need critical analysis from this standpoint. A similar scarcity of humus occasionally occurs on bosses of dolerite, where the soil covering the rock is very thin. STABLE AND MIGRATORY PLANT FORMATIONS 67 The grass covering such bosses is then very green and dwarf, but many of the dolerite bosses in Scotland still have an acid humus covering, an inheritance from past climatic conditions. Dolerites contain a large proportion of alumina, lime, iron, and magnesia, and a high percentage of phosphoric acid, but the alkali content is low. The weathered surface is con- tinually leached of its lime, but the oxides of iron and silicate of alumina form a highly ferruginous soil of an easily deflated, powdery nature when dry. In these cases, in addition to the effects of the drought and lime in promoting rapid decay, the poverty in humus may be in part due to the oxidising effects of the iron oxides on decaying organic matter. The grassland of these dolerite bosses in Scotland is described by Robert Smith in his “ Botanical Survey of the Edinburgh District ” (30). The association resembles in some respects that of the chalk downs, but is much poorer in species. In very dry positions, soils consisting largely of lime and iron would appear therefore to favour rapid eremacausis rather than humification, but these matters await the result of experiment, and a short recapitulation may be given to enforce the point, that the relations of lime to vegetation vary widely with the physiography and climate. 1. In arid regions lime accumulates in the soil apart from the nature of the district rocks. 2. In humid climates lime tends to be leached from porous surfaces and elevated soils and to gravitate in solution to lower levels. 3. In sufficiently warm, moist, aerated soils with neutral humus, lime in solution induces fertility through its powers of liberating plant food and its favourable influence on bacterial growth. 4. In cold, saturated soils its main action may be one of neutralisation and precipitation of soluble humus substances, giving freedom of action to inorganic salts of ammonia and potash. 5. In humid climates, warm or cold, hard-jointed lime- stone tends to develop an underground system of drainage which may gradually lead to drought and eventually to the barren condition know as limestone pavement. 68 THE SCOTTISH BOTANICAL REVIEW 6. In dry, shallow soils, on limestone, chalk, or dolerite, humus fails to accumulate, and this possibly leads in certain cases to a scarcity of nitrogen. 7. The relations of lime to humus must vary greatly from point to point, according to the physiography in a limestone country, since areas subject to drought alternate with areas where lime and humus occur together in a deep, moist, or marshy soil, where the degree of fertility would depend on warmth and aeration, or stagnancy. In alpine regions in particular these relations must be very complex, and plants which appear to be calcicole may have markedly diverse reasons for their choice of habitat. Some may require easily available, inorganic nitrogen-compounds but be capable of living in a soil subject to physical drought, a combination limiting their distribution to dry, neutral, or faintly alkaline soils where their dwarf habit would not influence them adversely through competition. Others, having similar requirements as to mineral food, but incapable of withstand- ing drought, would need to compete with rank-growing vegetation, or occupy humus-filled crevices in rocks with a calcareous drainage; while others again, with a small demand for nitrogen but incapable of withstanding soil- acidity, or of competition with a heath flora on an acid soil, would be restricted to the driest limestone-habitats where plant decay is accompanied by loss of nitrogen. PART IV. THE STABLE TYPES OF VEGETATION IN BRITAIN. There are four chief stable types of vegetation in this country : Moorland, Woodland, Heathland, and Grassland. Moorland covers extensive areas in the north and west of the British Isles. It is there often the dominant type of vegetation to sea-level, but further south and east only covers wide stretches on mountain plateaux, or on low-lying peat flats, where it has succeeded fenland through changes leading to leaching of the surface or stagnation. The formation of moorland peat depends chiefly on perennial great atmospheric humidity, and rather low temperatures. In the north-west, where these conditions have been pre- STABLE AND MIGRATORY PLANT FORMATIONS 69 eminently fulfilled in the past, peat has taken possession of the ground almost irrespective of the nature of the under- lying rocks, and, in extreme cases, even of the physiography. Moorland peat overlies clays and sands, sandstones and limestones, as well as all kinds of metamorphic and igneous rocks. To the south and east, where the past climate has been less frequently favourable for peat growth, the moor- land is more and more restricted by physiographic and edaphic factors. The greater part of the moorland is at present in a retrogressive phase owing to a late change in climate (31) (26). This retrogression has been accelerated locally by physiographic factors, and the peat has been entirely or partially removed from considerable tracts, includ- ing the better-drained slopes and wind-swept cols and summits, especially in areas which have developed a system of under- ground drainage, as porous debris and limestone pavement. The moorland associations, over large areas, owe their present persistence to an edaphic inheritance of thick peat, formed under more favourable conditions for peat accumula- tion in the past. This cloak of peat, and a highly acid drainage arising from its degradation, act as a barrier to the invasion of the moorland region by other plant-associations to which the present climatic conditions are not entirely unfavourable. Pine-forest is capable of colonising the retrogressive bogland, but the spread of the pine during the late retrogressive phase was perhaps prevented by early human interference. A large part of the rest of the country was once covered by woodland. It possessed the stations best protected from the wind, and where the summer temperatures were highest, z.e. the southern and eastern lowlands, and the northern and western valleys. In the north and west woodland becomes more and more restricted to the warm, sheltered valleys and steep slopes where peat has found no lodgment. The dominance of tree-species in the woodlands has depended on their specific capabilities of resisting unfavour- able climatic and edaphic conditions on the one hand, and competition on the other. The past climate and the method of dissemination of the seed have probably been the chief regulators of their periods of immigration, while competition, 7O THE SCOTTISH BOTANICAL REVIEW physiography, and the soil have largely controlled their present distribution. Birchwoods are widely distributed in the moorland region, since they are capable of withstanding the acid condition of the soil associated with atmospheric leaching and moorland drainage. Humid, cool summers are less unfavourable to their growth than to the other tree-species with which they are unable to compete in close canopy on account of their great light requirements. The early post-glacial immigration and spread of the birch was, comparatively speaking, unchecked by the severity of the climate and sterile condition of the soil, but its dominance further south was rapidly limited by competition with other trees. Natural pinewoods are at present almost restricted to the Central Highlands of Scotland, but extensive pine-forests formerly covered areas of the moorland peat. The pine appears to have greater capabilities than the birch of with- standing drought and soil-acidity, but requires somewhat higher summer temperatures for its full development, and its former widespread colonisation of moorland peat, probably coincided with a stage of moorland retrogression when the humidity was less, and the summer temperatures higher, than those favourable for rapid moorland growth. The oakwoods are subdivided by Moss, Tansley, and others as the chief associations of three separate plant- formations (3). Thus they distinguish: (1) damp oakwoods of Quercus pedunculata as the chief association of a plant- formation of clays and loams ; (2) dry oakwoods of mixed QO. pedunculata and Q. sessiliflora with a plant-formation of sand soil ; and (3) oakwoods of Q. sesszliflora alone with a plant-formation of siliceous soils. It must not, however, be forgotten that clays, loams, and sands normally form the lower ground in Britain, while the so-called siliceous soils, derived from the more resistant sedimentary, igneous, and metamorphic rocks, form the steeper-sloping and more mountainous country. It should also be noted that deep, fine-grained soils and coarse, shallow, mechanically formed soils bear parallel relations to the physiography of the softer and harder rock masses. The slopes of the hills have been recently glaciated and the lower ground has been the STABLE AND MIGRATORY PLANT FORMATIONS 71 receiver of the finely comminuted drift. Moreover, the hilly ground of the so-called siliceous soils is in its distribution in this country chiefly northern and western, where the atmo- spheric humidity is comparatively great and the summer and winter temperatures approach one another most closely. The Q. sesst/iflora woods have their chief distribution on the Carboniferous shales of the Pennine region and the harder, Silurian shales of Wales, rocks which form hilly country, from their structural relations or metamorphism, but on prolonged weathering yield clays almost indistinguishable from the clays of the lowland oakwoods. In such cases any difference in the soil would be chiefly due to the effects of physiography on the maturity and depth of soil, drainage and climate, and the oaks as a whole may be said to show a preference for clay-soils, soils retentive of moisture and with large reserves of potash. Quercus sesstliflora is reported to have a less Continental and more Atlantic distribution than Q. pedunculata, and to rise to greater height on the European mountains. The dis- tribution of the two types of oak in this country seems therefore to have been guided as much by the physiography, and their past and present relations to climate, as by the specific nature of the soil. Q. pedunculata is the lowland type requiring deeper soils, greater stability in the amount and nature of the soil-waters, and higher summer temperatures for full development, while the Q. sessz/zflora woods form a montane belt intermediate between those of QO. pedunculata and the birch, where the physical conditions are more exacting but the competition less. It is probable that the heat and light requirements, the ability to stand wind-exposure, the kind of root-system, and the capabilities of the different species of flourishing under stronger or weaker solutions of salts and acids in the soil- waters, are the chief factors guiding the distribution of the British forest-trees. Ashwoods and beechwoods are considered chief associations of a plant-formation of calcareous soils by Moss, Rankin, and Tansley (3). They are also known on the Continent to show a preference for limestone. The ash is very deep-rooting, and is considered to flourish best under stronger mineral solutions Fa THE SCOTTISH BOTANICAL REVIEW in the soil-waters than most other British trees. It is light- demanding, and therefore unfitted for competition in close canopy, but reaches perfection with greater humidity and cooler summer temperatures than Q. pedunculata. In oak- woods it is frequently confined to wet flushes in sloping hollows, where often a very black humus is formed from forest litter apparently under the influence of alkaline drainage. The oaks avoid these damp spots perhaps for similar reasons that they avoid most limey soils, but whether this be from excess of salts in solution or for other reasons remains to be shown. Many acid-humus and cold-enduring plants appear to succumb to very slight excess of mineral food in the soil-waters, and it is probable that the different forest-trees have very different optima in respect to the strength of mineral soil-solutions about their roots. Ash-trees probably find somewhat similar conditions to those of the flushes, in the soil in the crevices of limestone and amongst debris on steep limestone slopes in which they deeply root. A habitat where it can root deeply, obtaining a relatively large supply of mineral food-solutions and yet avoid competition with the oak or beech, appears to be the natural one for ashwoods. The influence of physiography in causing contrasting habitats in limestone areas has already been referred to, and the beech has very different capabilities of rooting to those of the ash. A deep, well-drained, uniformly good soil may perhaps induce deep-rooting in the beech; but with its dislike for undrained soils and with its great spread of strong, com- paratively superficial roots, it appears often to have specially selected a slightly raised, rocky mound when growing in oak- woods. In the Lowlands of Scotland raised mounds, due to small exposures of whinstone or other rock, are frequently the site of small but strong beech-clumps. These of course are plantations, but may be the successful trees out of mixed plantings. Beech-trees growing on poor, loose soil are frequently uprooted by storms, but those growing upon the rock escape. Their success on the bare chalk may then perhaps be accounted for by their power of strong superficial anchorage, and the deep shade and abundant leaf-fall with which they protect the surface from evaporation. STABLE AND MIGRATORY PLANT FORMATIONS 73 The peculiarly abundant litter of the beech is stated to contain more nitrogen and potash than most forest litter, and the poverty of undergrowth allows it to quickly come into close contact with the soil. Rocks containing large quantities of lime or iron, such as the limestones and_basic-igneous rocks, may retard the formation of acid humus in beechwoods owing to the action of lime and iron onhumus. The probable effects of abundant lime and iron in retarding the formation of humus on chalk downs and dolerite bosses have been dis- cussed above. In beechwoods, where the surface is damp and shaded, the result would be rather to promote rapid humification through the retardation of acidity. Forest litter is considered by some to afford little return of nitrogen to the soil, but in the case of beech litter on chalk or decomposing dolerite, with the diminished acidity, a considerable return of nutriment may perhaps occur. The capability of the beech of growing on poor soil may be explained by its shallow roots, the known relation of the roots to mycorhiza, and the abundant litter and shade afforded by its leaves. In respect of the various habitats chosen by British forest- trees the relations between recurrent violent gales and the long life of trees must have full consideration. Oaks are rarely uprooted by storms from sites they have chosen on deep soils, though their branches are frequently broken. The poorly grown oaks planted on dolerite or sandstone exposures in Scotland are, however, often overturned, and an examina- tion of such trees shows a poor development and frequently a rotten condition of the descending roots. Beeches grown on loose soils are easily uprooted, and a hard substratum giving a firm hold to their spreading roots appears necessary for stability. Nor is a deep or shallow-rooting of importance only in respect of the soft or hard nature of the ground and the effects of wind, since a shallow soil is rapidly leached, unless the rock contains sufficient bases to neutralise the acidity incurred by the accumulation of forest litter. The beech can flourish and withstand storms on a firm surface with a shallow soil, and a substratum containing large quantities of lime or other bases would tend to the neutralisa- tion of surface-acidity. The oak stands best on a deep soil where its deep roots can obtain a firm anchorage. Exposure VOL. I. 74 THE SCOTTISH BOTANICAL REVIEW of the surface and superficial leaching are of less importance provided the deeper soil-waters retain their alkalinity. As shown by Graebner (2), heathland has developed within the present sphere of the climatic influence of the woodland province through leaching of the surface layers of the soil where these are more than usually porous.‘ Within this province moorland has a physiographic or topographic distribution, while within the moorland region heaths closely resembling those of the heathland proper may arise through physiographic causes preventing humus accumulation, or forwarding its aeration and disintegration as great porosity of the surface, steepness of slope, and exposure to insolation or wind. With an expansion of the moorland province a greater area of the heathland would doubtless progress to moorland, and the former might therefore be looked upon as a delayed or abortive stage of moorland. The heathland in the south-east of this country is evidently an extension of the heaths of the N. German Plain, as pointed out in the ‘‘Types of British Vegetation.” Other heathland areas in the north and west bound and intersect the moorland, and are evidently abortive or retrogressive phases of the latter. The heathlands of the N. German Plain and S.E. Eng- land cover a district which appears to have been peculi- arly subject to marked changes of vegetation in post-glacial times. During stages of Continental extension and summer drought, the moorland province contracted towards the Atlantic border, and wide areas of peat-covered ground were invaded by pine-forest. At the same time the Russian steppes probably advanced far into the N. German 1 T have used the terms ‘‘ moorland province” and ‘‘ woodland province” to denote the areas in which the climate tends to the dominance or decadence of moorland or woodland as, the case may be; the terms ‘‘moorland region” and *“ woodland region” to comprise the areas where either moorland or woodland is the dominant natural vegetation at the present time ; and the term ‘‘ district ”’ for separate areas occupied by stable plant-formations within such pro- vinces or regions. The province shifts with secular changes of climate, but, owing to physiographic or edaphic factors, plant-formations, which stabilised within a certain climatic province, may persist in a retrogressive condition, for shorter or longer periods, beyond its shifted boundaries ; and owing to similar factors plant-formations differing widely from the dominant vegetation of a climatic province may arise and hold their own within such a province. Stable plant-formations might thus be classified as (1) dominant or climatic, where the prevailing factors are existing climatic factors; (2) retrogressive or relic, where the prevailing factors are past climatic factors ; (3) topographic, where the pre- vailing factors are physiographic and edaphic (19). STABLE AND MIGRATORY PLANT FORMATIONS 75 Plain, and the woodlands were forced to retreat towards the Atlantic border and on to the slopes of the mountains. Marr has pointed out the significance of the presence of fossil steppe mammalia, and of living steppe plants, in the sandy tracts of the S.E. of England. He suggests that these plants may be relics of a former steppe period in which the steppe had a western extension as far as the S.E. of England (3). The heathland tract has perhaps resulted from the altera- tion of markedly different climatic influences within this district. During the Ice Age, the N. German Plain was widely covered by sandy drift, which was afterwards subjected to the sifting action of wind. Sand covers much of the northern tract, while the fine dust or loess has accumulated along its southern borders In Late Glacial times an advance of the steppe covered wide areas in loess and entombed many steppe mammalia, as described by Nehring (32). With later advances the area would be further subjected to the sifting influence of wind. With the late return of a moor- land insular climatic phase, the more fertile sands and loess would be covered by forest, but certain sandy areas would be rapidly leached of their soil-nutriment and only heath or moorland would obtain a footing. With prolonged leaching of the surface of the sandy tracts, the heathland would advance at the expense of the forest, as shown by Graebner. If the steppe plants in the heaths of the S.E. of England are relics of a former steppe phase, as Marr suggests, they have probably managed to persist in the sand-dune areas of the S.E. coasts, and have reinvaded the heath, when the moorland climatic phase passed away. It is, however, possible that these steppe types were reintroduced into the S.E. counties from the Baltic coast in early historic times. All the stable types of vegetation have their homologues in migratory formations. The moorland is represented by fenland. Migratory woodlands are represented by alder and sallow swamps, and a kind of open heath is frequent, especially within the moorland region, on leached sandy and gravelly alluvia. It is, however, with the grasslands that we find the predominance is on the side of the migratory formations. Perhaps the only truly stable grassland in this country is that of the chalk downs. 76 THE SCOTTISH BOTANICAL REVIEW Apart from grass-moor, or grass-heath, the various acid types of grassland can hardly be considered to have more than a temporary significance, except where their habitat is governed by migratory factors of surface change. Nardus stricta is normal to the flood-rims of streams and acid ‘‘flushes” in the moorland, but often rapidly invades any northern or high-lying area, where peat is undergoing denuda- tion or has but lately been removed. The distribution of Nardus grassland in this country appears to be mainly governed by the extent of acid drainage from moorland peat and the hard, leached ground from which peat has been denuded. Peat subject to snow lie is also usually covered by Nardus. Nardus has no value as pasture, and the trampling and manuring of cattle or rabbits tends rapidly to its disappearance. Azra flexuosa, on the other hand, gener- ally avoids flushed ground and affects soils which have been long subjected to atmospheric leaching. Its normal habitat is grass-heath and the acid-enduring woodlands of the moor- land region. It rarely forms true grasslands, either by itself or in conjunction with other species, except where the migratory factors have lately been in operation.’ Thus while Nardus affects frequently flushed moorland alluvia, Aira is abundant on the older, deserted leached terraces. The so-called neutral and alkaline grasslands (3) (33) of Festuca, Agrostis, etc., usually form pasturelands. Their per- manence depends either on constant grazing, treading, and manuring by cattle, which is favourable to their establish- ment, or on physiographic or migratory factors setting limits to the occupation of the ground by woodland, heath, or moorland. Where they form pastureland, one or another type may be artificially established on almost any class of soil, after a short tillage ; but where they occur naturally, physiographic and edaphic factors are always prominent. The climate of this country favours pastureland in places where the growth of woodland is retarded by the shallowness of the soil-cap, wind exposure, periodic flushing or flooding, or by the situation forming a favourite resort of animals (3) ; but for the persistence of pasture on the same spot, it is 1 Dr. W. G. Smith points out that Azva flexuosa is often a tussock-plant on new soils exposed to leaching, e.g, waste heaps. a STABLE AND MIGRATORY PLANT FORMATIONS TL. necessary that heath, moor, and marsh should be excluded by edaphic factors, ze. that souring or acidity of the soil should be prevented. Good drainage is therefore essential on the one hand, and sufficient bases to neutralise the acidity induced by constant leaching, on the other. The former is attained through soil-porosity or slope, the latter either by periodic flushing of the surface with waters capable of pre- venting acidity, or through the soil-layer being very thin and resting on rock, which by weathering directly liberates sufficient bases to neutralise acidity. The chalk and certain limestones and basic-igneous rocks are the chief rocks which may perform this function in this country, and owing to the surface action of solvents tend to have smooth outlines, and thin very finely divided residual soils on exposed elevations and slopes. Exposed hills and bosses of these rocks are colonised with difficulty by trees owing to the shallowness of the soil, but afford all the conditions necessary for permanent grassland. The chalk grassland occupying the smooth, rounded contours of the chalk downs in the south of England, is believed to have existed from prehistoric times. Parts of the tops of the downs, where a leached soil has accumulated, show heathy patches (3), but the chalk is so pure that a soil of this nature is generally absent, even on the gentler slopes. Under present climatic conditions, humus fails to accumulate under the grassland of the chalk downs. As in all pastures, this may be partly due to constant grazing, but it certainly is accentuated by edaphic conditions. During the colder, wetter periods in the past, when the formation of moorland peat was at its height, acid humus may have accumulated from the absence of drought and the inactivity of the bacteria of decay, and the plant-associations of the chalk downs may well have been very different from the present ones ; but with the change of climate to a drier one with warmer summers, the xerophytic grassland has found conditions peculiarly favourable for its establishment. Under the present climate neither acid nor mild humus accumulate, and the dwarfed chalk-grassland flora has evolved under conditions promoting recurrent drought and comparative soil-sterility, but without the soil-acidity which defines the flora in similar sterile areas on other rocks. 78 THE SCOTTISH BOTANICAL REVIEW Another type of grassland found in this country, the ‘Calpine pasture” of Robert Smith (34), may closely approach the stable types, and perhaps in some cases might be classed along with them. In its usual development, how- ever, it shows clearly its origin in periodical flushing of the surface of the hill-slopes with water, and there is indeed every stage between alpine grassland and the ‘‘ moorland flushes ” described in the ‘‘ Vegetation of Caithness.” The centre of distribution of the grassland of these moorland flushes lies in the migratory grassland of the higher sandy alluvia of the moorland streams. The typical alpine grass- land is best developed on the slopes which rise above the moorland, and consequently uninfluenced by acid drainage from above. Its chief development apparently occurs on the slopes receiving the most heat and the least atmospheric precipitation, ze. the slopes where the upper limits of the moorland follow the lowest contours. In its typical develop- ment Vardus stricta is scarce or absent and the prominent grasses are Festuca vivipara, Anthoxanthum odoratum, and alpine forms of Azra cwspitosa with Alchemilla alpina and other plants. As noted by W. G. Smith, it always shows the greatest development where the slopes are formed of the finest rain-wash derived from mica-schists or other rocks which suffer deep and complete mechanical division through weathering, and especially where these rocks contain more than the usual small amount of soluble alkaline bases (3). Where sparsely developed, the grassland is restricted to the slopes immediately beneath the crags from which it receives the surface run-off of water and rain-wash, and wide stretches of alpine grassland are always surmounted by extensive areas of crag or rock-debris. Many of the smaller grassland stretches migrate rapidly with the changing spread of surface waters, and all owe their persistence to the delay in leaching by atmospheric precipitation, induced by flushing of the surface and the finely divided nature of the soil. The following scheme of classification shows some of the possible relations of the stable types of vegetation in Britain :— 1. Arctic-alpine Region—the sanctuary of plant-formations from a former more widely extended arctic-alpine and tundra province. The province is very limited at present in this STABLE AND MIGRATORY PLANT FORMATIONS 79 country and the plant-formations chiefly or entirely migratory. 2. Moorland Region—the stronghold of plant-formations isolated through the influence of an acid-humus barrier, due to the maximum combined effects of the late glaciation and the range of the migrations of the moorland climatic province. 3. Mesophytic Forest Region—occupying the area where the above barrier is ineffective or wanting within the meso- phytic forest province. 4. Steppe Province—absent in Britain. We may tentatively exemplify topographic stable plant- formations in the following :— 1. The S.E. Heath Formation, which is invasional in the Mesophytic Forest Region through leaching of the surface. 2. The N.W. Heath Formation (possibly including the extinct pine-forest in the peat), which is invasional in the Moorland Region through climatic and topographic retro- gressive changes in the moorland, 3. The Chalk-grassland Formation — invasional in the Mesophytic Forest Region through the physiographic and edaphic nature of the substratum, perhaps comparable with the “Sunny Pontic Hill” formation of Graebner (2) and chersophyte formations of Warming (13). It is probable that both the Moorland and Mesophytic Forest will eventually be split up into plant-formations based chiefly upon differences in the physiography of the habitat. (To be continued.) REFERENCES. (22) Hatt, A. D.—The Soil. 2nd ed., 1909. (23) Weber, C. A.—Jahresber. der Manner vom Morgenstern, 1goo. Heimatbund an Elbund Wesermundung. GRAEBNER, P. — “Studien iiber die norddeutsche Heide.” Englers Jahrb., 1895. TRANSEAU, E. N.—‘‘ The Bogs and Bog Flora of the Huron River Valley.” Bot. Gazette, 1905-6. (24) Frtu, J., and ScHrROTER, C.—Die Moore der Schweiz. 1904. WarmMiING, E.—(Ecology of Plants. Eng. trans. Oxford, rgo9. (25) Yapp, R. H.—‘“‘Sketches of Vegetation at Home and Abroad: Wicken Fen.” The New Phytologist, 1908. 80 ' THE SCOTTISH BOTANICAL REVIEW (26) Crampton, C. B.—The Vegetation of Caithness considered in relation to the Geology. 1g1t. (27) SmiTH, W. G., and Rankin, W. M.—“ Geographical Distribu- tion of Vegetation in Yorkshire: Harrogate and Skipton District.” Geograph. Journ., 1903. PRAEGER, R. L].—A Tourist Flora of the West of Ireland. 1909. (28) SCHROTER, C.—Das Pflanzenleben die Alpen, 1908, p. 88. (Quoted in Warming’s CEcology of Plants.) (29) SCHROTER, C., and KircHNER, O. — Lebengeschichte der Bliitenpflanzen Mitteleuropas, vol. i., 1905. (30) SmitH, R.—‘‘ Botanical Survey of Scotland: Edinburgh Dis- trict.” Scot. Geograph. Mag., 1goo. (31) GEIKIE, J.—“ On the Buried Forests and Peat Mosses of Scot- land and the Changes of Climate which they indicate.” Trans. Roy. Soc. Edin., xxiv., 1867. Lewis, F. J.—‘‘The History of the Scottish Peat Mosses and their relations to the Glacial Period.” Scot. Geograph. Mag., 1906. CaJANDER, A. K.—Beitrage zur Kenntniss der Entwickelung der europaischen Moore. Fennia, xxil., 1905. (Quoted in Warming’s CEcology of Plants.) (32) NEHRING, A.—‘‘ The Fauna of Central Europe during the Period of the Loess.” Geol. Mag., 1883. (33) ScartH, G. W.—‘‘The Grassland of Orkney: an C&cological Analysis.” Trans. of Bot. Soc. of Edin., rg1t. (34) SmitH, R.—‘ Botanical Survey of Scotland: North Perthshire.” Scot. Geograph. Mag., 1900. (35) Cavers, F.—‘‘On Saprophytism and Mycorhiza in Hepatice.” New Phytologist, ii. 2, 1903. (36) FLAHAULT, CuH., and ScHROTER, C.—‘“ Phytogeographical Nomenclature.” III° Congrés International de Botanique. Bruxelles, Mai, 1910. (37) FLICHE, P., et GRANDEAU, L.—Papers in Annales de chimie et de physique. 1873-1879. ANTHELIA: AN ARCTIC-ALPINE PLANT ASSOCIATION 81 Anthelia: An Arctic-Alpine Plant Association. By AW ..Gesmuth,oBise~Ph.D IN the course of the ascent of Ben Lawers by the Inter- national Phytogeographical Excursion last August, Professor C. Schréter and Dr. E. Riibel, two experienced Swiss botanists, pointed out a _ plant association extremely characteristic of the higher Alps. This association presents many points of interest as one in which several Hepatice and Mosses play the part of pioneers in colonising a sub- stratum which owes its origin in the first place to topography and in the second place to the action of running water. While the term ‘‘arctic-alpine zone” in Scotland is a con- venient term for general use, most botanists will appreciate that the zone is by no means uniform in its development. Just as the vegetation of the lowlands presents itself as woods, moors, grassland, and other types, each with sub- types, so in the arctic-alpine zone there are many sub- divisions (1) (2). The plant association now under con- sideration is one of these subdivisions, and it is proposed to bring together here some information which may direct attention to it and may stimulate the study of others. Last August on Ben Lawers, after ascending the morainic valley of the Tuim Bruic or Carie Burn, and traversing in succession the zones of NMardus-Juncus squarrosus grassland and the Adchemilla alpina pasture, the lower levels of the south-west corries were reached. Shortly after leaving the clear springs which emerge from the rocks about 3000 feet at the highest limit of a definite stream channel on the Lawers side of the valley, the Swiss botanists drew attention to a long, dark, crusted tract descending from near the base of the “Gentian Cliff,” a very conspicuous tract in the rock-strewn green sward of this part. Other examples were seen in ascending the slope towards the low neck between Ben Lawers and B. Ghlas, frequently as dull dark stretches following a series of shallow troughs. Towards the summit the grassy turf becomes more limited, and Adlchemilla alpina is ' Read before the Botanical Society of Edinburgh, 11th April 1912. 82 THE SCOTTISH BOTANICAL REVIEW more and more restricted to sheltered places beside blocks. Here the most conspicuous plant association is Rhacomitrium lanuginosum with Carex rigzda and other mat-forming plants, but at frequent intervals the darker patches occur. The summit ridge shows many tracts of this black mossy crust, and in the ‘“Cernua Corrie” there is one large patch at the rough wall near the ruins of the Ordnance hut. One has also recollec- tions of other summits where this dark-crusted humous surface is a feature. The Swiss botanists recognised these patches and tracts as ‘*Schneetalchen,” a term introduced by Oswald Heer in 1836. In L. Schroter’s “ Taschenflora ” (3) the word is trans- lated as ‘‘snow-valley,” but as English equivalent ‘‘snow- gutter” or ‘‘snow-flush”” more nearly expresses the kind of little runnels suggested by the original term. The vegetation of these snow-flushes has been described by authors like Kerner, Christ, Stebler, and Schréter. The more recent observations of Brockmann-Jerosch (4) indicate that, before dealing with the vegetation, attention should be directed to the topographical and physical factors that bring about its evolu- tion. Anyone who has seen the snow melting on the higher hills can recall the early emergence of rocks or snow-bleached green slopes and knolls. In Switzerland and the Tyrol these are bedecked with flowers while the snow still lies a few yards away. Day by day the snow-patches decrease, becom- ing more limited to the lower depressions or sunless slopes. The snow-water soaks through the turf seeking the lower de- pressions till it flows from below the snow and streams away to still lower levels. Thus in troughs and depressions of undulating ground and along the foot of slopes or escarp- ments there is a system of temporary water-courses which, so long as the snow is melting, are more or less under water. The summer rain-water will tend to follow the same course, but with this difference that the larger supply of ice-cold water is replaced by more occasional trickles of warmer surface-water. On steep slopes these streamlets descend with some force and carve out little stream-beds (snow-water channels), but on gentle slopes or flats or in depressions the force of the flow is not sufficient to erode ; the water wanders slowly through the turf and deposits accumulated suspended ANTHELIA: AN ARCTIC-ALPINE PLANT ASSOCIATION 83 matter as a sediment.! The snow-water carries the dust which gathers on lying snow, fine particles of mineral matter and fragments of plants, and it also collects other materials as it trickles over the surface. Snow-dust, according to Ratzel, may contain 50 per cent. of organic matter, and as this with the finely divided mineral matter is laid down amongst remains of last year’s vegetation, a rich soil is built up. The soil is dark and finely fibrous and when lifted adheres together as clods. On Lawers this turf contains numerous particles of glittering mica, while other samples from Suther- land show particles of gneiss. Brockmann-Jerosch points out that this substratum owes its origin solely to the action of snow-water and rain, and that the snow-flush vegetation occurs where the snow-water collects, not necessarily where the snow-patches lie longest. The substratum results, there- fore, from slow sedimentation combined with growth of vegetation which is not swept away but remains perennial, rising gradually each year on the new sediment. This building-up may be seen where a boulder or piece of rock is present and becomes gradually embedded. On these grounds we regard this habitat as migratory and comparable to the flushes described by C. B. Crampton (2); the snow-flush vegetation will, therefore, come under that author’s group of migratory plant formations. During spring the snow-flushes are soaking and _ their moisture is retained far into the summer, hence in a moist season they remain slimy and slippery underfoot. In a dry summer (as in August I9I1) they become cracked and crusted. Although it may be surrounded by a green sward of Alchemilla alpina with arctic-alpine grasses (Festuca, Poa, Deschampsia, Nardus), the snow-flush is mainly dark in colour except where tufts of Guaphalium supinum, the dark-green mats of Salix herbacea, or moss-tufts of Polytrichum have become established. The vegetation of the snow-flush begins with cryptogams, 1 These characteristics suggest the term ‘‘snow-flush.” The original definition of ‘‘flush” by C. B. Crampton (2) is: ‘‘Sloping ground gives rise to springs. Where the springs are of small volume, or where they are of a temporary nature flowing only in wet weather, the growth of plants prevents marked erosion of the surface and shallow gutters floored with vegetation result. These may be called ‘flushes.’ The vegetation and plant association of these flushes depends on the nature and source of the flow of water ” (p. 62). 84 THE SCOTTISH BOTANICAL REVIEW and these frequently remain as the dominant vegetation. The Swiss botanists give the place of pioneer to one of the Hepatice, Anthelia Juratzkana. This plant lies close to the surface, and in the fresh, moist condition forms a bluish- green carpet. In summer it is often seen in the dry con- dition as a dark-brown or black mat, which with the lens appears as a tangle of shoots, about 1 mm. thick, closely beset with minute leaves, the whole recalling in miniature a mat of low-growing matted Cad/una. As pointed out by Professor Schréter, the mat (in August) was dark but with a greyish coating which in his book (5) is ascribed to a thin covering of filaments of fungi. This is also recorded by Heeg (8), who states that the roots are permeated by fungal filaments and may rank as mycorhiza such as have been described for other Hepaticz (9). Specimens taken from a typical snow-flush on Ben Lawers in August last, along with others from similar habitats in Sutherland, supplied by Dr. Crampton, were submitted to Dr. Symers M. Macvicar, who reports that they belong to the genus Anthelia, but as the specimens are sterile it is unsafe to say whether the species is A. /uratzkana, or A. julacea, or a mixture.t Dr. Macvicar (10) states that A. Juratzkana ascends to the summits of the highest hills in Scotland (4300 feet on Ben Nevis), and that it rarely descends below 1900 feet. In a recent letter he also suggests that the snow-flush association is probably what he has named the Marsupella association [(10) p. 7], and he points out that Arnell and Jensen have observed and described the occurrence of Azzhelia and associated Hepatic in Scandi- navia. C. Schréter (5) cites a letter in which W. Arnell says that Anthelia Juratzkana flourishes best on soils periodi- cally flooded by snow-water, and it also occurs on the banks of streamlets, and more sparingly on bare soil; also that the species becomes more abundant in Scandinavia the higher one goes, and that it is abundant in Spitzbergen. Riibel (7) follows the Scandinavian botanists in distinguishing an ‘‘Anthelietum,” which he regards as the basis of the 1 Anthelia Juratzkana (Limpr.), Spruce, in C. Schroter (5); A. ju/acea, Dum., var. clavuligera, Nees., in Brockmann-Jerosch (4): 4. /zratzkana(Limpr.), Trevis (Jung. nivalis, Sw.), in S. M. Macvicar (10). A. 7zlacea (L.), Dum., is the only other Scottish species. ANTHELIA: AN ARCTIC-ALPINE PLANT ASSOCIATION 85 snow-flush types of vegetation to be indicated later. It is thus clear that experienced observers in Northern Europe and the Alps recognise in A. /uratzkana one of the pioneers of the snow-flush. Another cryptogamic element in the snow-flush vegetation is the genus Polytrichum, according to the Swiss accounts generally P. sexangulare, Flork., and P. alpinum, L., two species recorded as characteristic of the summit region of Scottish mountains. On Ben Lawers we saw cushions of Polytrichunz (sp. not identified) in the snow-flushes pointed out. This genus of mosses has several species which are adapted for life in moorland flushes, as indicated by C. B. Crampton [ (2) p. 62)]. They can withstand periodical submergence and soon grow through the shallow deposits of sediment laid down, so that they aid in binding these deposits into a humous turf; the close, compact growth also enables the tufts to withstand periods of drought. The snow-flush vegetation includes a limited number of flowering plants, but according to Swiss accounts some of these are very characteristic. The following is a list of species recorded in August IQII (ze, after a dry summer) on the western slopes of Ben Lawers (about 3500 feet) in two snow-flushes where A zz¢helza was a conspicuous element :— Polytrichum. Guaphalium supinum. Rhacomitrium lanugtnosum. Salix herbacea. Solorina crocea (orange lichen). Szbbaldia procumbens. Carex pilulifera. Euphrasia (? scotica). C. rigida. Festuca ovina (vtvipara). A short list was recorded (August 13, 1898) by Robert Smith on Ben-y-Ghloe at about 3000 feet on patches where snow had recently melted:—Salix herbacea, Guaphalium supinum, Alchemilla alpina, Galium saxatile. The Swiss “ Schneetalchen ” is thus described (3) :—‘“ Poly- trichum septentrionale usually appears as a pioneer and covers the ground with a dense, dark-green carpet. Soon afterwards Avenarza biflora begins to penetrate this carpet ; its slender stems and small leaves are half hidden in the moss, so that the stellular flowers seem to be scattered over the carpet as if by accident. Next the creeping, radiating 86 THE SCOTTISH BOTANICAL REVIEW mats of Cerastium trigynum associate themselves with these two plants. Later on Guaphalium supinum steps on the scene, or the long red creeping shoots of the five-leaved lady’s mantle (Alchemilla pentaphyllea) become interlaced, so as to form connected masses.” Brockmann-Jerosch (4) gives the results of examinations of nineteen examples of snow-flushes, recording the number of times each species was noted. Ribel (7) also gives results from forty-eight stations. Both authors worked in the Rhetian Alps, an extensive area of high altitude. The following table gives—for Scottish species only—the num- ber of times of occurrence recorded from the above two memoirs :— | Rubel. Brockmann. | Gunaphalium supinum : : ; x4 44 15 Salix herbacea . ; ; si 43 16 Polytrichum, spec. div. ; : : 31 frequent Poa alpina . i ‘ : j 2 5 Taraxacum officinale (alpinum) , ; z 9 Veronica alpina . : : : Sp 20 | 7 Alchemilla pentaphyllea 16 8 Anthelia Juratzkana . : fl I5 frequent Sibbaldia procumbens : : : ail 14 7 Cerastium cerastioides yi 13 II | Carex Lachenalit : : : 9 | Polygonum viviparum | 8 | _Euphrasia minima . : ; : ‘ | 5 ar Deschampsia cespitosa Gi The species recorded for snow-flushes by Oettli (6) for the Churfurst and Sentis areas are also included in the above table. These lists are characteristic for stations on crystalline rocks. The catalogue from calcareous rocks is much longer, with few Scottish species, and has Salzx retusa as a character- istic plant. The Scottish lists, taken along with other notes, suggest that the same association is represented as on the crystalline rocks of Switzerland. It is probable that in Scotland Cerastizum cerastiotdes and Veronica alpina also occur on snow-flushes, while Poa alpina, Deschampsia ANTHELIA: AN ARCTIC-ALPINE PLANT ASSOCIATION 87 ceéspitosa (arctic-alpine form), and Polygonum viviparum are widely distributed over other arctic-alpine associations. The various authors point out that the conspicuous flowering species of the snow-flush vary from place to place. Here Polytrichum may form a carpet, there Salix herbacea a mat, or Gnaphalium in white tufts. Riibel gives prominence as tone-imparting species to Polytrichum, Salix herbacea, Sibbaldia, Alchemilla, Gnaphalium, and Ligusticum Mutellina, and he demonstrates that the association presents sub-types according to the dominant plant ; thus Avenxaria biflora is given as much more abundant when Polytrichum is the conspicuous plant ; on the other hand, Guaphalium is fairly constant in all the sub-types of snow-flush. This patchy occurrence of various plants in separate stations or within the same snow-flush leads Brockmann to the conclu- sion that accident or chance plays an important part in the constitution of the plant-covering. In other words, the snow-flush is an open association into which species from neighbouring plant communities migrate. The existence of comparatively large pure patches of a _ single species, frequently observed in the snow-flush, is attributed to the capacity of the plants for vegetative propagation. It seems to be essential for existence in the snow-flush habitat that the species should form low matted tufts sufficiently com- pact to resist periodic flooding, and along with this the power to extend laterally by short horizontal shoots. Guaphalium supinum forms mats from which the short flowering stems arise, and short horizontal shoots extend along the ground to give off new leafy shoots which flower in some future year. Cerastzum, Sibbaldia, Alchemilla, and Veronica alpina extend by lateral branches rooting in the underlying carpet of Hepatice or Moss. Salix herbacea forms a loose mat extending comparatively rapidly by underground horizontal branches which root and may form new plants. That this habit of growth leads to consider- able stability may be experienced when one collects specimens ; either one brings away much soil attached, or the shoots break off short, leaving most of the plant behind. Another interesting adaptation is noted by Brockmann- 88 THE SCOTTISH BOTANICAL REVIEW Jerosch, namely, the presence of a felted coating of hairs which entangles air when the plant becomes submerged by occasional water, and so prevents flooding of the leaf. He recalls the silky white sheen on such leaves as Guaphalium, Cerastium, Stbhaldia, and Alchemilla alpina when seen under a shallow layer of flood-water. The evolution of the snow-flush vegetation is indicated in E. Riibel’s account. Azthela, perhaps preceded by still lower organisms,! forms a humous turf, fairly stable and liable to invasion by other species. Polytrichum follows later, and more or less takes the place of Azthelza. Later still Salx herbacea or Alchemilla assumes chief place. Riibel has observed where the different sub-types occur beside each other that Sa/zr takes the higher and drier situations, which in Switzerland adjoin the extensive plant association of Carex curvula. ach of the stages of vegetation probably indicate stages in the evolution of the habitat, since the later vegetation will tend to give it increased stability. In time the accumulation of sediment, humus, and vegetation may be such that the snow-water is diverted to new situations, where the sequence will begin over again. During the various phases other species secure a footing and flourish well or ill according as the habitat suits them. Riubel suggests that Taraxacum and Cerastium cerastiotdes find in the snow-flush that abundance of organic matter which they require; in Switzerland both are characteristic species of the “lair-flora,” that is, places manured by sheep, goats, or other animals, where grasses like Poa annua flourish. Again, Carex Lachenalit he has observed invading the snow-flush from neighbouring marshes. All these observations point in one direction: that the snow-flush is a series of migratory associa- tions. These were recently defined in these pages (11) (p. 8): “ Migratory formations are of comparatively short persistence on the same habitat, which sooner or later undergoes change or destruction, with renewal elsewhere. Their associations tend to rapid degeneration from plant invasion. All stages of progressive successions of associations are encountered.” It seems to us that along these lines the somewhat complex distribution of our arctic-alpine vegetation must be studied. 1 Dr. Macvicar (zz /z¢t,) informs me that Aztheléa is preceded by an alga. ANTHELIA: AN ARCTIC-ALPINE PLANT ASSOCIATION 89 LITERATURE. (1) Types oF BriTIsH VEGETATION.—Edited by A. G. Tansley. Cambridge, rg1t. (2) Crampton, C, B.—Vegetation of Caithness. 1911. (3) ScHROTER, L.—Taschenflora des Alpen-Wanderers (with notes by C. Schroter). Zurich, 1903. (4) BROCKMANN-JERoScH, H.—Die Flora des Puschlay. Leipzig, 1907. (5) ScHROTER, C.—Das Pflanzenleben der Alpen. Zurich, 1908. (6) OxttTLi, M.—Beitrage zur Oekologie der Felsflora. Zurich, 1905. (7) RtsBet, E.—Pflanzengeographische Monographie des Bernina- gebietes. Zurich, 1gtt. (8) Herc, -——.—Lebermoosen Niederdsterreichs. (9) Cavers, F.—“ On Saprophytism and Mycorhiza in Hepatic.” New Phytologist, 11., 1903. (10) Macvicar, S. M.—‘ Distribution of Hepaticee in Scotland.” Trans, Edin. Botan. Soc., xxv., 1910. (11) Crampton, C. B.—‘‘Stable and Migratory Plant Formations.” Scot. Bot. Review, i., 1912. Mosses from the Western Highlands. By jjames stiton, MVM Deer LS. LEUCOBRYUM PUMIL_LUM (Michx.) has been found at last near Gairloch, Ross-shire, on the 20th of September 1911. Bryo- logists have searched during a long course of years for this moss throughout Europe and Great Britain, but until lately quite in vain. I have a distinct recollection of hearing Pro- fessor Schimper of Strasbourg, author of the “Bry. Eur.,” while on a visit to this country in 1865, urge those interested in mosses to institute an organised search for a Leucobryum having leaves with deeply cucullate apices. Several in Glasgow directed their attention for twenty years thereafter towards the discovery of such a moss, but without success. In 1882 a Mr. Piffard discovered, in the New Forest, England, L. minus (Hampe), then reckoned a variety of the common L. glaucum (1.), but now named L. albtdum (Lindeb.), for what reason I know not. At this stage the search gradually abated for nearly thirty WOES Me 7 gO THE SCOTTISH BOTANICAL REVIEW years longer until, in 1911, L. pumz2lum (Michx.), the moss to which attention had been drawn fifty years previously by Professor Schimper, was found near Gairloch, growing on the debris of the Torridon Rock. I have carefully compared genuine specimens of this moss from Dr. Braithwaite, who got them from Mrs. Britton, keeper of the Moss Herbarium in Bronx Park, New York, and the Scottish and American specimens agree in every particular. Whilst working amongst the Torridon Rock what struck me as very peculiar is the fact that, in spite of its hardness, it is easily disintegrated by the rootlets of slender mosses. Many such, chiefly species of the genus Grimmia, are found growing on large exposed masses of this rock and mostly in saucer-like depressions. On detaching these tufts they appear as if they had been reared in loose sand, whilst a corresponding cavity in the rock is revealed. I may return to this subject on another occasion. Another peculiarity presented is the deep dark-green colour assumed by mosses srowing on it, while the same mosses growing on other rocks usually exhibit a greyish appearance more or less deep. That this prevailing deep green is not of the same constitu- tion throughout may be inferred from the fact, that in some instances it changes, in the course of two or three weeks, to a deep coppery hue throughout the entire plant, in others it slowly turns to a dingy yellowish-green, while in the majority of cases the colour remains nearly unchanged for many months. An instance of the first of these changes in colour is the following which, besides, is remarkable otherwise, owing to the peculiar shape of a proportion of the upper cells of the leaf, viz. barrel-shaped, with lateral walls, convex and rugose, while the ends are usually narrowed and nearly straight. Grimmia rubescens, sp. nov.—In small, dense, convex tufts of a dark green above, rapidly changing throughout to a dark copper colour, quite unlike the dingy grey usually assumed by others of the same genus; stems about an inch long or less, simple or dichotomously divided, much less fre- quently fastigiately branched; leaves closely arranged around stem, spreading slightly and straight when moistened, narrowly ovate lanceolate, ending in a broad apex, latit. MOSSES FROM THE WESTERN HIGHLANDS QI ‘o7—"I mm., surmounted by a broadish hyaline hair tapering to a point, sparsely spinulose, variable, from a third to nearly as long as the leaf proper ; nerve narrow below (‘04 mm.), widening somewhat upwards, then narrowing to summit, pale then copper-coloured; margin entire, recurved in lower half, plane afterwards, the other margin quite plane throughout, the upper half thickened by two couples of transverse cells and towards apex interruptedly bistratose throughout ; cells in four to six short rows at inner base narrow, cylindrical, separate (especially laterally), °og—07 by 004-6 mm., slightly granular, ultimately pellucid, outwards becoming (rather abruptly) nearly elliptical, dark, opaque, and granular outwards to recurved margin; on the plane margin two to five short rows /yaline, sharply oblong, close, "013-02 by ‘007-9 mm., up from base cells very irregular in size and shape, barrel-shaped or only irregular in outline, and slightly narrower than those at base but of the same length, or cells merely round but more of such nearer apex and nerve ; only four capsules in a young state were found ; seta pale, short, straight or slightly bent, about a third of an inch long ; capsule elliptical, rugose ; lid and teeth red, short, pointed; acumen short, blunt, red, not more than a fourth the length of capsule. There is another moss rather closely allied to the preced- ing, but as it presents one strange peculiarity such as I cannot recall having ever seen previously, I consider it right to describe it here. Grimmia undulata, sp. nov.—In rather dense convex tufts of a dark-green colour above, becoming darker in the her- barium, ultimately assuming a dingy brown in the lower half ; stems slender, about an inch long, simple or slightly branched ; leaves laxly arranged around stem, nearly appressed, with the longer hairs everted ; when moistened spreading somewhat, straight and slightly undulating, re- maining attached to the stem nearly to base, long, slender, ovate lanceolate acuminate, terminating in a slender hair very variable in length, from a mere point below to about half the length of leaf above, slightly spinulose ; cells at central base long, cylindrical, close, often attached, ‘06-09 by “006-9 mm., slightly granular, ultimately hyaline, outwards g2 THE SCOTTISH BOTANICAL REVIEW shorter, bluntly oblong, with two to four short rows of cells next margin smaller still, sharply oblong, ‘016 by ‘006 mm., but all basal cells becoming hyaline ; upwards cells become graduaily shorter and merge nearly transversely into the upper dense, opaque, chlorophyllose, round or roundly quadrate cells, length variable from ‘oog to ‘O16 mm. in longer diameter, in upper third somewhat smaller, ‘007-9 mm. ; nerve soon turning brown, firm, solid, breadth near base "05-065 mm., a little broader upwards then lessening to acute apex, of dense structure within; the posterior wall shows a single row of chlorophyllose cells as in the pagina, but nearer the front a row of three largish pellucid cells ; margin of leaf recurved a little more than the lower half, entire; a thin transverse section of leaf shows the margins thickened in a remarkable manner, by a round cluster of cells grouped together without order, in number from seven to fifteen so as to form a thick round club three times the thickness of the rest of pagina ; this thickening is often seen a little below the middle of leaf but smaller. Barren. On the red rock in several places near Gairloch. This moss is rather closely allied to Gr. trichophylla (Grev.). Another interesting moss deserves to be recorded, viz. a Bryum, so closely resembling in size and appearance the greenish forms of B. axgenteum that I passed it on more than one occasion without inspection. Bryum elegantulum, sp. nov.—Tufts very dense, slightly convex, green above, pale but tending towards a reddish tint below ; stems very slender, about half an inch or less in height, simple or sparsely divided, red and red-radiculose in lower half, to which also minute acutish leaves are attached ; upper leaves rather suddenly enlarged, but still minute, scarcely a millimetre in length, closely arranged and even imbricated so as to form, at and near apex,a rather elongated comal group; such upper leaves are oblongo-ovate, concave, narrowing convexly above to a point, from which arises a-long, slender acumen about one-third the length of leaf proper, or from ‘2 to 35 mm. long, terminating acutely and having the same construction as the rest of the pagina, but the cells composing it a little longer; nerve slender and thin, breadth near base about ‘03 mm., tapering and ending in the acumen; margin MOSSES FROM THE WESTERN HIGHLANDS 93 plane, entire, having two rows of cells less than half the breadth of those of the rest of the pagina but longer ; cells at base in two to four transverse rows (across base), oblong or quadrate, with thickish, opaque walls, ‘025-032 by ‘or2- ‘022 mm., those above for the rest of leaf acutely rhomboid, large, close, ‘04-06 by ‘o14—02 mm. _ Leaves almost destitute of chlorophyll. This moss is certainly distinct from &. argenteum in the narrow border-cells of the leaf, in the nerve which reaches the apex and, from the character of the cells in the lower part of the acumen, probably extends a little into it, as well as from the length and constitution of the acumen itself which, in 4. argenteum, is only represented by a bluntish knob; lastly, from the differences in the shapes of the leaves as well as from differences in the paginal cells in the two mosses. In woods in Lovedale, near Gairloch. Only one long, strong seta was observed in a tuft, but as it dropped off I could not found anything on it. At Onich, on Loch Linnhe, a tuft of Plagiothectum Miilleri was picked up in fine fruit. The station is unusual, not many feet above sea-level. This is the only specimen which has been found in fruit in Great Britain. In 1909 at Onich, on Loch Linnhe, a moss was discovered to which a proper place could not be assigned, but as the same moss was secured in I9II at Gairloch, I feel constrained to give an outline of its main characteristics. Barbula incavata, sp. nov.—Tufts large, very dense, green above, changing in the herbarium to a dingy yellowish green, brown below; stems simple or slightly branching, slender, about half an inch long ; leaves incurved when dry, spreading a little and straight when moist, 4ol/ow throughout, elliptical or narrowly so, length 1°5 mm. by *35 mm. in breadth near middle, blunt and round at apex; margin plane, slightly incurved towards the summit, faintly crenulated or nearly entire ; nerve strong, pale, then fulvous, vanishing a little below apex ; cells at inner base oblong or somewhat rhomboid, detached, ultimately pellucid, variable, ‘022-04 by ‘oo8—o1I mm., outwards smaller, thinner, ultimately hyaline, latit. ‘oo6- ‘008 mm. across, upwards changing transversely into others, 94 THE SCOTTISH BOTANICAL REVIEW very peculiar, giving at first the impression that they are very minute, viz. ‘004-7 mm. across, but such are ultimately seen to be the dimensions of particles of chlorophyll, whereas the cells proper are bluntly quadrate, close, with very thin transparent walls, containing four particles of chlorophyll in each, and ‘oog—014 mm. across, ultimately dark and opaque; such are found throughout the rest of the leaf, the larger next the nerve. Leaves are minutely papillose, more especially in their upper third, less frequently on the margin ; height about ‘0025 mm. At Onich near seashore in two places ; in one place similarly situated at Gairloch, IgII. Scottish Forms of Sparganium. By Arthur Bennett; A: FOLLOWING up the notes in the January number, I here give the forms of this genus, which I have from Scotland. The specimens have all been seen by the late Mr. Beeby and by Dr. Rothert. The names are given in the sense and value of the describers, although opinions may differ as to their being varieties, states, etc. Sparganium ramosum, Curtis (sensu Beeby).! j- microcarpum, Neuman in Hartm., “Sk. FL,” 112, 1899.” —Crianlarich, Perth, E. S. Marshall sp., 1893. Dalmally, Argyll, E. S. Marshall sp., 1893. Baldernock, Stirling, R. M‘Kay sp., 1887. S. affine, Schniz. a zostertfoltum, Neuman (/.c.).—Unst, Shetland, W. H. Beeby sp., 1887. B deminutum, Neuman (Zc.).—Loch na Criche, Moidart, v.c. 97, S. Macvicar sp., 1895. y microcephalum, Neuman (é.c.).—Loch between Rackwick and Orgill, Hoy, Orkney, E. S. Marshall sp. S. minimum, Fries, var. flacctdum (Meinh. sp.).—* Mél. 1S. -erectum (a), Linn, “Sp; Pl,’ ed. 1, 1754,078e 2 First part, all published. SCOTTISH FORMS OF SPARGANIUM 95 Biol.,” xviii., 3, 393, 1893. Isle of Gigha, v.c. 101, A. Somer- ville sp. Colonsay, v.c. 102, M. M‘Neill sp. var. rostrata (Larsson sp.), “Fl. Vermland och Dal,” ed. 1, 1859. Islay, v.c. 102, A. Somerville sp. The var. mzcrocarpum of ramosum is the one most nearly approaching in habit, etc., to S. eglectum, Beeby. The only Scottish specimens seen of S. neglectum, Beeby, are from the Isle of Lismore, v.c. 98, S. Macvicar.' Mr. Beeby remarked on these specimens:—‘‘ From a care- ful examination of your Sparganza, they certainly appear to me to be S. weglectum. But hitherto the northern limit has been v.c. 62, N.E. York; while on the west side there is nothing north of Cheshire!. Hence I should have been glad to have seen a ripe fruit. The extension of range is remark- able ; however, it occurs in Denmark (v. sp.), in almost the same latitude as Berwick, c. 55° 50, while Lismore is c. 56° 30. There are records for the mainland of Sweden on good authority up to about 50° 10%. Another at 59° will decidedly require confirmation. I cannot do other than name your plant S. neglectum.”—Beeby, 8/10/1898. Contrasting these against Mr. Burkhill’s from Scarborough, York (1896), the fruit is decidedly smaller, and the Scar- borough specimens would perhaps come nearer to the var. oocdrpum, Celak, in “ Oest. Bot. Zeit.,” 425, 1896. S. affine zostertifolium is taken as the type by Neuman; whether this is so with reference to Schnizlein’s specimens | am unable to say, as I have not seen a type specimen. Ascherson and Graebner? under affine cite “S. alpinum, Don ex G. Don, in ‘ Loud. Hort. Brit.,’ 375, 1830, name only.” But in Headrick’s “Survey of Forfar” (1813) Don uses the name S. zatans* as found in the Lake of Forfar. So far as | have seen there seems to be little variation in the other species, z.e. S. s¢mplex, Huds., in Scotland. The following are additional records to “Topl. Botany” and Supplement :— S. ramosum (Ebudes), 102, “S. Ann.,” 1906. S. neglectum, 74, Wigton, “ Ann. Scot. Nat. Hist.,” 101, 1910. ’ 1 “Ann. Scot. Nat. Hist.,” p. 39, 1899. 2 « |e Oe ‘ - oa >» —) RB ‘Vater / ' ae hy I aTAI sna ama oe re satemalinat 1 ‘gil slayo head ‘Sat a ae r aA A¢ AS fm aad i ni ELL - inh Pm Oa Laas | - “tes i oe ap EY Vi Woda om eo Ris ; Vi am 4 : rf ré A ay Fi mE “vy om q i © a co \ q ba ia ad | AW Fig. PLATE I. 1. Median section of a young carpel of Ag/aonema commutatum. x about 25. ons! . Young ovule of A. commutatum, showing the young embryo-sac (¢.5.) — and the two integuments. x90, . Young ovule of 4. modestum. x90. . Nucellus of an ovule of 4. modestum of about the same age as that shown in fig. 3. 250. ¢.s., embryo-sac. . Nucellus of the same species, showing a division of the primary sporo- genous cell. . Young ovule of 4. simplex. x50. The embryo-sac (é.s.) contains two... | nuclei. . The nucellus of the ovule shown in fig. 5. 250. . Cross-section of an embryo-sac of A. s¢mplex, containing two vimies placed near the middle of the sac. x 250. . Two longitudinal sections of an embryo-sac of 4. commutatum, with two nuclei, placed at opposite poles of the sac. x 250. Ve SGOT. BOT. REV. = CT rc ‘a ai —“* es ee) mnt Li? -D.H Camppeli, Dei. PLATE | CAMPBELL,- Embrys-sac of Aglaonema i wie ae aed ae ; een Pal! lS ae | eae Ot Ren ; uu MOUNLUY LS d ' PMP « Vol ee H) o ban #) j H \ a . , y ? 2s ) ~ ‘ \ } 4 , J ia [ j cB A re 54 } } j r' a" 4 ; ay . 4 - sy ~~ A ZI \ . F 4 f ; Lf ll aTagd ‘ ts ital > ow? 7 te eskauna er onk-ovsdns me to aoitasz wiring 1 .3ff ee A sO2b%-” .dee orl? Yo bas rowol sdb. yoo 8 ‘oe one mo a8 ditw vsvqywr-.K ag ol ir ee } ee af, -islona gsr o roa al gt ni a rhea jaa ed} to alisioM .pt 21 we ; A ied Dy, fofun soit ow! ort bas ( as wd one-one ms to moitosé pt 3iT, ‘panied oo ad od, SUT jr bHt ti wore pi daidw Yo 200 ylmo \ ozs — .sbivib ot } - dato wot diw , mma to asz-oyrdats 5 to etioltose owl OT (21 2g _ -orsinn 9f} to sao (RI .o82 oni lo 2sloq sdi is etisq i beoslq ya fs me x .ausloun Iasslatlo s 21 ; isfoua 1alyq ‘W to pae-oyrdms as Yo motges: salyqgoriM- A 3 y OF 21 ci beg tadd es eye omnse ; AN to > oan-orndin ae 10 KOgB ott lo moitose-2201) 81 . gil ; 7 \,0gS* — .zute1eqqs yye a Site ,Re siee of to moiges leiiase oi iguoilt aoites2 .o1 sit | aah x detaia ee i th i nude Fig. Fig. Figs. Fig. Figs. Fig. Fig. Fig. PLATE II. 10. Oblique section of an embryo-sac of A. sémplex, with two nuclei, at the lower end of the sac. x 250. 11. Ovule of 4. szmplex with an embryo-sac containing an egg apparatus and two free nuclei. x 40. 12, 13. Details of the embryo-sac shown in fig. 11. 12 shows the egg (0) and the two free nuclei ; 13, the synergide. x 250. 14. Section of an embryo-sac of 4. s¢mplex in which were two nuclei, only one of which is shown in the figure. The nuclei were beginning to divide. x 250. 15, 16. Two sections of an embryo-sac of A. commutatum, with four nuclei, placed in pairs at the poles of the sac. 15 shows one of the micro- pylar nuclei ; 16, a chalazal nucleus. x 250. 17. Micropylar region of an embryo-sac of 4. commutatum of about the same age as that figured in 15, 16. 18. Cross-section of the apex of an embryo-sac of A. s¢mplex, showing the egg apparatus. x 250. 19. Section through the central region of the same sac, showing one of the two free nuclei. x 250. ae SCOT BOT REV D.H. Campbell, Del. PLATE, I. CAMPBELL.—Embryo-sac of Aglaonema. PLATE III. Figs. 20, 21. Four sections of an embryo-sac of A. sémplex containing an egg apparatus and two free nuclei which were in process of division. a, 4, x 120; c,d, x 250: ¢, the egg ; d, one of the synergids. Figs. 22, 23. The dividing free nuclei. 650. starch granule (?) at one pole Fig. Fig. Fig. Fig. of the nuclear spindle. 24, a-c. Three sections of an embryo-sac of A. simplex in which the egg apparatus was lateral in position; d, outline of the egg apparatus. The endosperm (e7.) is beginning to form. x90. 25. Young endosperm-cells from the sac shown in fig. 24. x 250. 26. Transverse section of the egg apparatus of A. simplex; 0, egg; sy., synergids. x 250. 27. Basal part of the same, showing two of the four large free nuclei. x 250. Figs. 28, 29. Young nuceijlus and embryo-sac of A. modestum. The embryo-sac (é.s.) has entirely destroyed the apical tissue of the nucellus. In 28 there is apparently a second embryo-sac (e.s.) (?). x90. SCOT BOT REV D.H. Campbell, Del. PLATE I CAMPBELL, -Embryo sac of Aglaonema J VOL be | ‘ «A rs | j ew SY. 7 7 ve a yt == i Py ” ger Nitrate: > i Wihkck s. er - ra of ' ¢ tf f 43 a Tala es aay, ; BRE Ve = VI bina 7 el } hi a } dali Gh sali Auto Bn (see ats; 2noijase owT Of pt 9a soit seed oft ta enw (9) eutersqqs ygo odT cone od? bila i ow? y elisa sift to otto sti ingze1q S19 ,p2ul of uods YIinoraqqe ,tsloun dl a a | 00 .» mi mwode 518 92913 Io mm ‘a wage * ope ortine ort to [loo rsdjoas mort isloua Mame ssxd? to quov) .1g¢ .pill \ OER esiiunm «A To semye bits (.405) oyrdms hellao-owT .s¢ pil / aeqéohus gavoy oi gaiwode pse-oyrdens omse oft lo eq iwwol .p¢ “tere ga hee hellso-on0 « guisizinos pwlowe Ao one-odenT pe ’ / ex — islam gerd} to qi s sbeot ausloua-nolen agial .2¢ .pi cova he eae ag bat a ‘A To on2-oyidand ag ait ) {.2) isloua 9aii Yo bas issaled) xe wit | OB % " Aw ! bollc.ows fr wip As to one-oiirdons whio ak ee a8 vy Ost x { 36 V bie x: allao-wnaqzobas mow ialoua coien'l Oh Of .20iT! AT {.x9) Sead gave hi oe ait Fig. 30. Fig. 31. Fig. 32. Fig. 33- Fig. 34. Fig. 35. Fig. 36. Fig. 37. Fig. 38. Figs. 39 Fig. 41. PLATE IV. Two sections of an embryo-sac of A. szmplex in which the endosperm filled the sac. The egg apparatus (0) was at the base. Three large nuclei, apparently about to fuse, were present in one of the cells; two of these are shown in @ x90. Group of three small nuclei from another cell of the same sac. x 250, Two-celled embryo (em.) and synergide of A. s¢mplex. x 200. Lower part of the same embryo-sac, showing the young endosperm. Embryo-sac of A. stmplex, containing a one-celled embryo (em.). x 120. Large fusion-nucleus made up of three nuclei. x about 500. Embryo-sac of 4. modestum, with egg apparatus at the base, and two free nuclei(ez.). x 120. Chalazal end of an older embryo-sac of the same species; 0, the egg (or perhaps one-celled embryo). x 120. An older embryo-sac of A. modestum with a two-celled embryo (em.), x 120, , 40. Fusion nuclei from endosperm-ceils. x 250. Embryo-sac of 4. modestum, with a young embryo (em.). The primary endosperm in the concave side of the sac is conspicuous. 120. — PCOT, BOT. REV D.H Campbell, Dei PLATE IV CAMPBELL,~ Embryo-sac of Aglaonema 5 a - ; ‘ 3 iS » ‘ vie ¥ co es. er 7 . ‘ 3 i I 64 ~s 3e a ‘ } THE EMBRYO-SAC OF AGLAONEMA I15 (4) CAMPBELL, D. H.—‘ Studies on the Aracez.” Annals of Botany, XlV. I-25, 1900. (5) CamMpBELL, D. H.—“ Studies.on the Araceze. The Embryo-sac and the Embryo of Aglaonema and Spathicarpa.” d/d., xvii. 665-687, 1903. (6) CaMpBELL, D. H.—‘“‘Studies on the Aracez, III.” Jbid., xix. 329-349, 1905. (7) Gow, J. E.—‘‘ Morphology of Spathyema fetida.” Bot. Gazette, xlill, 131-136, 1907. Note on Victoria regia, Lindl. By John Arcangeli, Professor of Botany at Pisa, Italy) THIS plant, the most beautiful of all the species in the Nymphezacee or water-lily family, has been recently culti- vated in the Botanic Garden of Pisa (Italy) with excellent results and very little expense, using chiefly solar heat. The germination of the seeds was carried out in a small tank of zinc, with water, gently warmed from below by a petroleum lamp. The sowing was made in the month of March or April in small pots immersed in the water of the tank. When the seedlings were sufficiently grown, with leaves 3-4 inches in diameter, one or two seedlings were planted in the bottom of a basin which was placed in a greenhouse without any apparatus for artificial warming but with glazing turned towards the south, and which was, during a large part of the day, in direct sunshine. In these conditions the seedlings, having been kept from the month of June at a temperature from 25 to 40° C., grew quite well and continued to vegetate vigorously during the months of July, August, September, and October. In the process of vegetation the plant developed a dozen and more very fine flowers, of nearly 30 cm. in diameter, which expanded successively at intervals of from three to four days. The blade of the leaf is circular in outline, in the seedlings 5-12 cm. wide and reaching at maturity I-1'70 m. in diameter. The lifetime of each flower 1 Read before the Botanical Society of Edinburgh. 116 THE SCOTTISH BOTANICAL REVIEW was several days, and during the last few days they exhibited marked movements of nutation, rising gradually to water- level in the morning and sinking in the afternoon. The flower presented also the striking phenomenon of two succes- sive expansions in two different days, the first in the penulti- mate and the second in the last day, in each case during the afternoon, the one with white, the second with red corolla, and graceful changes of colour with beautiful gradation. The plant ripened fruits with perfectly formed seeds, although without cross fertilisation. By these experiments, accomplished in the years 1907, 1909, and I910, we may conclude that, if Vuctorta regia may grow on pools in Sicily on open ground (“ Bull. Soc. Tote. d’Orticultura,” 1907, p. 114), probably also in southern Italy, it may be grown in the plains of central and northern Italy with an appropriate greenhouse and practically using only solar heat. The trials made to cultivate this plant on open ground in our botanic garden till now have all completely failed, because by night the culture-tank being without cover is cooled by radiation ; nevertheless that is not to say that under special conditions and in very hot summers the cultivation may not be successful. Meanwhile, lately, during 1911, in the same greenhouse employed for the experiments above quoted, some seeds left in the bottom of the basin from the preceding culture germinated in the month of May, without artificial heating, and two of the seedlings so obtained grew and reached maturity, producing leaves of more than I m. in diameter, and flourished perfectly, bearing several flowers. I may say also that in the seeds of this plant I have been able to observe germinal asynchronism, viz. that the seeds germinate at different times, and then they may do so in different successive years, as is the case with Huryale and many other plants, an arrangement which is very profitable for the conservation of the species. SHORT NOTES 117 Short Notes. [Zt ts hoped that all will combine to make this section as complete as possible by the prompt recording of all *‘ new records,” etc. | Philonotis rigida, Brid.—Mr. James M‘Andrew made an interest- ing addition to the Perthshire list of mosses last May, when he collected a specimen of /#ilonotts rigida, Brid., at Aberfoyle (v.c. 87). The determination was confirmed by Mr. H. N. Dixon. This species is not entered in the census catalogue for any of the Scottish counties, although it has been reported from Orkney at least, if not from elsewhere in Scotland. As, however, it is mainly a southern species it was thought safer, in the absence of specimens, to omit the record. R. H. MELDRUM. Cornus suecica, Linn., in Peeblesshire (v.c. 78).—I am glad to be able to record this species from Peeblesshire. On roth June rgrt my brother, Mr. W. T. Blackwood, drew my attention to a small patch on the Dollar Law containing about a dozen flowering heads. The area covered was very limited, but later in the day another small patch was found. So far as I am aware this plant is at present recorded from Scottish vice-counties 112, 108, 107, 106, 105, 98, 97, 96, 94, 92, go, 8g, and 88. South of Perth and Forfar it has not been noted except from 62 York n. east, and 68 Cheviotland. In “ Topographical Botany ” Watson, however, gives, ‘‘83 Edinburgh by a trick?” but in the appendix to both the first and second editions of Lightfoot’s ‘ Flora Scotica” it is recorded from the Pentlands on the authority of Dr. Hope. It would be more satisfactory if this old record were confirmed, as it probably can be. The Peeblesshire record is interesting as being a connecting link between the English and the other Scottish stations. G, G. BLAcCKwoop. Zostera nana, Roth., in Aberlady Bay, Haddingtonshire (v.c. 82), etc.—With reference to the extract from the Report of the Botanical Exchange Club for 1gro, given in the first number of this magazine (p. 53), I have to point out that I recorded Zostera nana from Aberlady Bay so long ago as 1889 in the ‘‘ Transactions of the Botanical Society of Edinburgh” (vol. xvii. p. 415). In my note, which is entitled, ‘On the Occurrence of Zostera nana, Roth., in the Firth of Forth,” I also recorded the plant from the mud flats west of Cramond (v.c. 84), and near Torryburn (v.c. 85). These records apparently escaped Professor Trail’s notice when he drew up his ‘“Topographical Botany of the River-Basins of Forth and Tweed” (Trans. Bot. Soc. Edin., 1903). WILLIAM Evans. 118 THE SCOTTISH BOTANICAL REVIEW Petasites albus, Gaertn., in Fife.—A specimen of this plant was brought to me in February ; it was found growing near a burn in the Chapel district, and was first noticed in flower at the beginning of February. This species seems to be spreading in Scotland, and is well established in several places in the counties of Edinburgh and Linlithgow. Iam not aware of it having been recorded before for this side of the Forth. N. MILLER JOHNSON. Notes from Current Literature. ‘“Notes from the Royal Botanic Garden, Edinburgh” (No. xxil., November 1g1t), deals with Scheuchzeria palustris, L.—This was discovered in fair quantity on Rannoch Moor in July 1910 by G. W. Scarth, growing in a very wet peaty marsh associated with Carex limosa. In these notes Mr. Scarth gives a summary of the British localities where the plant has been observed. Four of these are in or near the Vale of York, four in Shropshire, and one in Northampton. The only known Scottish station was the White Myre of Methven, but Perthshire botanists have failed to find it in recent years, probably as a result of the change of vegetation due to the nesting of a colony of black-headed gulls. It is therefore of interest to have the record of the occurrence in Scotland of this rare plant. In the December number of the ‘‘ New Phytologist ” is an interest- ing account of the “floristic results” of the International Phytogeo- graphical excursion in the British Isles which was so successfully undertaken last summer. A considerable number of additions to the British Flora were noted, chiefly owing to the wider knowledge of our Continental friends, and of these a considerable number are from Scotland. This should encourage our Scottish workers to redouble their energies, as there certainly is a vast amount of valu- able floristic work still to be done; in the lowlands of Scotland in particular. The novelties include Castalia alba, Wood, var. candida (Presl.), froma loch near Dunkeld and from Galway. Mr. Druce says: “The distinguishing characters are the absence of stamens from the upper part of the ovary, leaving the neck bare ; the pollen grains are smooth (not tuberculate, as in a/va) and somewhat larger. The lowest pair of leaf-veins are curved, and, if produced, would cross, enclosing an oval area.” Sagina nodosa, Fenzl., var. montlifera, Lange, Southport dunes, etc. A Sagina from Ben Lawers, which has long been a cause of perplexity to botanists, was the cause of much discussion, but pending further examination it has been assigned to Sagina glabra, Koch. Mr. Druce says: “ From S. saginoides it may be known by its NOTES FROM CURRENT LITERATURE 119 large flowers, though doubtless in herbaria it will be often found to represent that species. From .S. swbu/ata its more creeping habit and more woody root-stock will distinguish it. Owing to the difference in the capsule it may be worth while distinguishing the Scottish form from the Continental species as var. scotica.” Another interesting plant from Ben Lawers, A/chemil/a vulgaris, L., var. acutidens (Buser), pointed out by Dr. Ostenfeld, who is very familiar with this plant in the Faroes. Professor Graebner also pointed out Calluna vulgaris, Hull., var. Erike, Ascherson and Graebner, which is a prostrate rooting plant with descending branches, the flowers being turned downward ; it is said to retain its character under cultivation. The claims of Juncus ranarius, Fries., to be considered a species are discussed, Professor Graebner being of opinion that it is worthy of full specific rank. Mr. Druce records it from many places, including Scottish records for vice-counties 85, 90, 105, 107. The Report includes descriptions of several new varieties and a considerable number of vice-county records, while there are also valuable notes upon many of our more critical species and varieties. An interesting memoir on the vegetation of Natal has just been published by Professor J. W. Bews of Natal University College, Pietermaritzburg (formerly of the University and Royal Botanic Garden, Edinburgh), as a reprint from “Annals of the Natal Museum,” ii. part 3. Natal may be said to present a coastal belt, then a series of terraces at about 1000, 2000, and 3000 feet respec- tively, with a mountainous region above 5000 feet. Into this system the chief rivers cut back deeply so that in the higher region rugged gorges are formed, while in the lowlands the valleys are broad and flat. Climatic details are given, and in discussing them the author throws useful light on the occurrence of veld and forest. The rain-clouds from the Indian Ocean deposit first on the coastal belt, and a dense bush is the natural vegetation. Towards the mountains the rains fall mainly on the rising edge of each terrace, while the intervening terrace-plateaux are much drier; the result is a forest-zone on each terrace margin and a treeless veld on the plateau. The valleys are much drier than the highlands, especially in winter, the dry season, when they are also colder. The vegeta- tion of the valleys resolves itself into “rocky stream thickets” in the narrower higher valleys, and a dry “thorn veld” of trees and grasses in the broader drier valleys. In view of the economic development of Natal as regards forestry, stock-grazing, and agriculture, the author has done great service in this memoir through his study of natural vegetation in relation to climate and physiography. A series of excellent photographs of landscapes illustrate the physiognomy of the country, and the numerous floristic lists will be valuable for the systematist. The memoir as a whole conveys the impression that field experience in the varied topography of Scotland is an excellent preparation for the study of the vegetation of other lands. 120 THE SCOTTISH BOTANICAL REVIEW “The Summit-Flora of the Breadalbane Range,” by Peter Ewing, F.L.S. (‘Glasgow Naturalist,” vol. iv. No, 2, February 1912). “‘ Mycological Notes,” by D. A. Boyd (‘Glasgow Naturalist,” vol. iv. No. 1, November rgr1), contains a number of new records for the Clyde area. “* Additions to the List of Mosses of Dumbartonshire,” by John R. Lee (“Glasgow Naturalist,” vol. iv., No. 1, November rg1t). ‘“‘The Weeds of Arable Land in relation to the Soils on which they grow, II.,” by Winefred E. Boenchley, D.Sc., F.L.S. (‘‘ Annals of Botany, January 1912). “Perthshire Roses,” by Wm. Barclay (‘Transactions Perthshire Socy. of Nat. Science,” vol. v. p. 11). ‘‘The Alge of the Bruce Peninsula,” by A. B. Klugh (“Ottawa Nat.,” xxv. pp. 94-98, I9II). “Phycological Studies: V. Some Marine Algze of Lower California, Mexico,” by Marshall Avery Howe (“Bulletin Torrey Botanical Club,” November 1911).—Several new species are described and figured. “ Aposporie et sexualité chez les Mousses,” by El. et Em. Marchal (Academie Royale de Belgique, “ Bulletin de la Classe des Sciences,” 1911, Nos. 9-10.) “Mousses du Sahara,” by M. A. Coppey (‘‘ Bull. Soc. Bot. de France,” t. xi. 7, October 1911).—Several new species described and figured. “Sur la présence du Plagiothecium curvifolium Schliep. dans les Vosges et la Jura, et sur le valeur specifique de cette Mousse,” by M. A. Coppey (‘ Bull. Soc. Bot. de France,” t. xi. 7, October 1911). “‘Pestalozzia hartigii Tubeuf. En ny. fiende i vara plantskolor,” by Torsten Lagerberg (‘‘Meddelanden fran Statens Skogsférsdk- sanstalt,” 8, 1911). “Om tvakénade blommor hos Salix caprea” (‘‘ Uber zweigesch- lechtliche Bluten von Salix caprea”), by N. P. Herman Persson (‘Svensk Botanisk Tidskrift,” 1911, Haft 3). ‘‘Om utbredningen af Melampyrum pratense L. f. aureum Norm. i sddra Norrland” (‘‘Uber die Verbreitung von Melampyrum pratense L. f. aureum Norm. in stidlichen Norrland”), by N. P. Herman Persson (“Svensk Botanisk Tidskrift,” 1911, Haft 3). ‘Die Polygalaceen der Rheinprovinz,” by W. Freiberg (‘‘ Verhand- lungen naturhistorischen Vereins der preussischen Rheinlande und Westfalens”). Contains valuable notes on many species and varieties. NOTES FROM CURRENT LITERATURE I21 ‘ Florule oltensis Additamenta, ou Nouvelles Annotations a la flore du département du Lot,” by M. Ern. Malinvaud, X.? (‘ Bull. Soc. Bot. de France,” t. xi. 7, October 1911), contains interesting notes on several species. “Sur quelques espéces japonaises et chinoises du genre Scrofu- laria,’ by M. G. Bonati (‘‘Buill. Soc. Bot. de France,” t. xi. 7, October 1911). ‘Plantes nouvelles, rares ou critiques,” by MM. les Abbés Coste et Soulié (Bull. Soc. Bot. de France,” t. xi. 7, October 1911).— Geranium bohemicum, VL. ; Vicia stcula, Guss. ;x Geum cebennense (G. svlvaticum x urbanum), Coste et Soulié; Artemisia chamemelifolia ; x Thymus aveyronensis (T. vulgaris x Serpyllum), Coste et Soulie ; x Thymus vivariensis, Coste et Revol., etc. “The Relation of Climax Vegetation to Islands and Peninsulas,” by Roland M. Harper (‘‘ Bulletin Torrey Botanical Club,” November IQII). “An Ecological Study of Whitewater Gorge,” by L. C. Petry and M.S. Markle (‘Proceedings of the Indiana Academy of Science,” 1910). “Induced and Occasional Parasitism,” by D. T. MacDougall (‘‘ Bulletin of the Torrey Botanical Club,” October 1911). ‘Om. ljungbranning fér skogskultur,” by Edward Wibeck (‘‘ Med- delanden fran Statens Skogsférsdksanstalt,” 8, 1911). “En margborrsharjning i 6fre Dalarna,” by Torsten Lagerberg (“Meddelanden fran Statens Skogsférsdksanstalt,” 8, 191r). “ Origin and History of our Garden Vegetables and their Dietetic Values,” by Professor G. Henslow, M.A., F.L.S., V.M.H. (‘Journal of the Royal Horticultural Society,” December 1911). ‘On Tumour and Canker in Potato,” by A. S. Horne, B.Sc., F.G.S. (‘Journal of the Royal Horticultural Society,” December 1911). Reviews, Book Notices, etc. THE TWENTY-SEVENTH ANNUAL REPORT OF THE WATSON BOTANICAL EXCHANGE CLUB, IgIO—IQII. UNpER the enthusiastic management of its Hon. Secretary and Editor the Report of this Club continues to increase in size and usefulness. As usual, the Report is confined to notes upon the specimens sent in for distribution by members, and these contain a VOL. 1. 9 INA) 2. THE SCOTTISH BOTANICAL REVIEW large number of critical species, upon which the remarks of the various specialists are most helpful to all studying the British flora. There is also a figure of /umaria major, Badarro, reproduced from the drawings made by Mr. E. W. Hunnybun for Ze Cambridge British Flora. The Report contains a photograph of the late Rev. Augustin Lea, and a sketch of his life by the Rev. E. F. Linton. Mr. Lea had been for many years a referee of the Club and its largest contributor. We have selected and reproduce here some of the more valuable notes :— ‘“‘ #. major, Badarro. The plant from Gilly Tresamble, Perran-ar- worthal, which I distributed in 1904 (see Report, 1904-5, p. 7), has been identified by Professors Schinz, Ascherson, and Graebner, as well as by Dr. Fedde of Berlin, who is working out the genus fumaria for Engler’s ‘ Pflanzenreich,’ as /. major, Badarro. Every year since its discovery, on Oct. 8, 1904, I have seen thousands of plants of it among potato, turnip, mangel and cabbage crops in the parishes of Perran-ar-worthal and Gwennap. Rouy et Foucaud (‘Flore de France,’ vol. i. p. 176) place this plant with / sfectadilis, Bischoff, under / agraria, Lag., and they describe it as follows :— ‘Bractées, lancéolées, égalant ou dépassant les peédicelles ; sépales ovales, courts, égalant environ le quart de la longueur de la corolle, et a peine plus étroits qu’elle, pfofondément dentés, 4 nervure médiane, peu ou point carénée; silicule globuleuse a mucron cylindracé, mince ; feuilles courtes, 4 lobes courts, peu écartés.’ None of the hundreds of Cornish specimens which I have examined have had the bracts more than one-half as long as the fruiting pedicel, and the small oval sepals are only rarely slightly dentate at the base. From all other species of /wmaria occurring in Britain / major may be distinguished by its long, ultimately lax raceme of 20-25 large rosy-pink flowers, which often are much recurved. On the Continent it is said to flower from April to June; in Cornwall its flowering season extends from the early part of September to late October.— F, Hamilton Davey. ** Anthyllis Vulneraria, L., var. coccinea, L. Dry banks, Polzeath, E. Cornwall, v.c. 2, August 1910.—H. E. Fox. A. Vudlneraria, L.., var. coccinea, L.., has flowers red, concolorous. Our plant named A. Dz/lenit, Schultes, is different ; it has cream-coloured flowers, tipped with red. I have seen var. coccinea from S. Devon, Cornwall, and (strange to say) Ben Lawers, Perthshire, whence Mr. C.-P. Hurst sent me fresh specimens a few years ago. A. Dillenii, which is always a small plant—I have gathered var. coccinea nearly a foot high, with larger heads—seems to be strictly littoral, ranging from E. Sussex! to W. Sutherland !—E, S. M. “ Vicia gracilis, Lois. Field of wheat, Coton, Cambs., v.c. 29, August 23, 1910.—A. J. Crosfield. Yes, good gracilis, coming under the a /eiocarpa, Gren. and Godr., as it has glabrous pods. I do not possess any examples of the hairy-podded form (8 eriocarpa, G. & G.), and do not know if it occurs in Britain. As mentioned in ‘Journ. Bot,’ 1908, p. 264, V. gracilis may always be separated REVIEWS, BOOK NOTICES, ETC. 123 from any forms of ¢e¢vasperma by the hilum and funiculus characters, and I noticed last year, when gathering both near Billingshurst, Sussex, that in the former the standard and wings are clear lilac in colour, not striate or very faintly so, whilst in ¢etrasperma both are strongly veined with purple. I noted, too, that the upper calyx-teeth of graci/is are lanceolate and those of ¢e/rasferma triangular-acute. —-C, E.S. ‘* Asperula taurina, L. Wood near Abercorn, Linlithgowsh., v.c. 84, May 28, 1910. This has been well established here for many years.—M‘T. Cowan, jun. Correct.—S. T. D. Add the county to the labels. Not on record, so far as I know, for Linlithgow.—A. B. “ Taraxacum ? Loose sand dunes, Hunstanton, W. Norfolk, v.c. 28, June 6, 1910. This does not appear to agree with the description of any of our recognised dandelions.—C. E. Moss. T. erythrospermum Andrz. ‘The same form as occurs on the Haddingtonshire coast, with very pale achenes —M‘T. C. This has the finely cut foliage of Z. exrythrospermum Andrz., but the achenes are paler in colour than usual.—A. B. J. Nearest to var. devigatum of our forms.—A.*B. Handel-Mazzetti, in his Monograph of Zaraxacum, has changed the names—I think wrongly. Dr. Moss’s plant clearly comes under what we call Z. erythrospermum, Andrz., DC. ‘The fruit is faze pinkish brown, not brick-red. Mr. W. H. Beeby informed me that we had probably one or two subspecies of that in Britain, besides the type.—E. S. M. ““Gentiana precox, Towns. Chalk Downs, Freshwater, Isle of Wight, v.c. 10, June r910.—H. E. Fox. This is G. Zngulata, C. A. Agardh, var. precox, ‘Towns.,’ Murbeck. I studied it carefully in Wiltshire, where it is usually associated with G. Amaredla, L., and came to the decided conclusion that they were specifically distinct. Probably all the alleged south-country inland stations given for G. campestris, L., belong to this plant, which usually sheds its seeds before G. Amareda is in flower.—E. S. M. “ Veronica arvensis, L., var. Cavenham Heath, W. Suffolk, v.c. 26, May 16, 1910. I find this variety, or perhaps form, is frequently gathered in mistake for V. verna.—C. E. Moss. Apparently this is the 8 glandulosa, Legr., in ‘ Bull. Soc. bot. France,’ 30, p. 70. Rouy (‘ Fl. de France,’ vol. xi. p. 50) says of it: ‘ Plante tres velue, glandu- leuse.’ I have gathered it in a more extreme form on the sandhills between Deal and Sandwich, E. Kent; and it is probably not uncommon.—E. S. M. “ M. rubra, Sm. f. Old quarry near Ross, Herefordsh., v.c. 36, Sept. 29, 1910.—A. Ley. This is the plant recognised by Malinvaud as MW. rubra, Sm.,and gathered by him as a subspontaneous weed in the neighbourhood of houses in France. It should be noted that M. rubra, Huds., is probably quite a different thing, which he describes (‘Flora Anglica,’ 1798, pp. 252-3) as ‘ floribus verticillatis ; caulibus diffusis ; foliis subsessilibus, ovatolanceolatis, serratis, acutis, subnudis’; while his J/. sativa is credited with ‘ floribus verticillatis ; caulibus erectis; foliis petiolatis, ovatis, serratis, acutis, villosis.’ 124 THE SCOTTISH BOTANICAL REVIEW Most British authors look upon J/. rubra as a species or variety differing from JZ. sativa in its longer leaf-stalks, while Hudson regarded his JZ. rubra as differing from that species by having shorter petioles. Hudson should not therefore be cited as the authority for the plant with longish petioles usually referred to as AZ. rubra in British floras—S. T. D. This is near to JZ. ocymiodora, Opiz, in ‘ Naturalientausch,’ No. 10, p. 22 (1823), but differs in the exserted stamens, and the stem base nearly glabrous.— A. B. I agree to this as MW. rubra, Sm., and, according to my herbarium, it is the usual British form, though I am aware that it is not quite like the figure in ‘English Botany,’ ed. iil. JZ. rubra, as I know it, has short roundish-ovate bracts, not so ovate or ovate-oblong as they are figured. —E. F. L. . “WM. gentilis, L. Roadside ditch, near Malvern, Worcs., v.c. 37, Sept. 9, 1910.—S. H. Bickham. I think that this is not typical gentilis, which has calyx-teeth much more hairy, but rather var. Wirtgeniana, F. Schultz, as it has the long-petioled leaves, stalked whorls, etc., of that form. I see in Druce’s ‘ List of British Plants’ that this variety is placed under JZ. rubra, but I thmk that L’Abbé Ch. A. Strail is right in regarding it as a form of genti/is.—C. E. S. M. gentilis, L., approaching var. Wirtgentana in the subglabrous stem and leaves and calyx thinly hairy, but failing in the points L’Abbé Strail (‘B.E.C. Rept.’ 1887, p. 187) emphasised, eg. ‘the floral whorls are all stalked, the lower ones with very long stalks, in the variety.’ This plant has the whorls mostly sessile, a few only in the middle of the spike being shortly stalked. This is borne out by Dr. Wirtgen’s specimens of AZ Waortgeniana, F. Schultz, No. 4, ‘Herb. Menthar. Rhenan.,’ ed. ii., which are also nearly glabrous in foliage.—E. F. L. The Abbé Strail, in ‘ Essai de classification et descriptions des Menthes qu’on rencontre en Belgique’ (‘ Bull. Soc. roy. de botanique de Belgique,’ xxvi. (1887), pp. 63-168) gives ‘ (1) Calice tubuleux, a dents longuement subulees, etc. (AZ. Wirtgeniana, Schultz) ; (2) Calice campanulé et a. dents plus ou moins courtes (AZ. gentilis, Smith).’ He gives ‘JZ. Wirtgentana= M. rubra, Lej. et Court., ‘Comp. fl. Belg. et Sm.”’ On the whole it seems to me, judging by Strail’s detailed descriptions, that this belongs to gem/ézlis. —A. B. “ Populus |monilifera|, 9. Plantation at Glen Parva, Leics., v.c. 55, June 1t910.—A. R. Horwood. Not Populus monilifera, Ait. (1789) =P. deltoidea, Marsh. (1785), which is very rare in Britain, even as a cultivated tree; nor is it P. monztlifera, Mich. fil. (=P. montlifera, Loud.), which is the common ‘ Black Italian poplar’ of cultivation ; but it seems to be P. virginiana, Fougeroux (‘ Mem. Soc. Agric. Par.,’ 1787). I have not seen this description, but I think this is the plant intended by Continental writers (e.g. Ascherson and Graebner, ‘ Flo. Mitteleur.’) by their P. wxginiana. Some examples of it in Kew Gardens are named ?. marylandica ; but the description of ‘ P. marylandica, Bosc.,’ in Lamarck’s ‘ Encycl. suppl. iv.’ does not fit the plant. This poplar does not appear in any of the British REVIEWS, BOOK NOTICES, ETC. 125 floras or lists. However, it is subspontaneous in several fenny places in Suffolk, and this, I suppose, must count as its first British record. It is sometimes planted, as on the roadside in West Suffolk between Barton Mills and Icklingham, also in grounds and gardens, as in Cambridge. The following poplars belonging to this group are usually confused by British botanists:—(1) P. nigra, Linn. (indi- genous in southern and eastern England) ; (2) P. de/totdea, Marsh. (very rarely cultivated in Britain; indigenous in N. America); (3) P. canadensis, Moench (the ‘ Black Italian poplar’; commonly cultivated ; origin unknown); and (4) /. virginiana, Foug. (culti- vated ; origin unknown). /. virginiana is usually (? always) a pistillate tree; P. canadensis is usually (? always) a staminate tree. The above three introduced poplars have o, 1, or 2 glands at the base of each lamina: these glands are absent in P. xigra. P. deltoidea is slightly ciliate at the margin of the lamina. P. canadensis (the male tree) has terminal leaves which are decidedly less acumi- nate than those of P. virginiana (the female tree).—C. E. M. “Poa Chaixit, Vill. In a wood, Eddleston, Peeblessh., v.c. 78, July 1909.—Ida M. Hayward. This grass seems to be spreading in Britain, or has it been overlooked ?—A. B.” BRITISH PLANTS: THEIR BIOLOGY AND EcoLocy. By J. F. Brvis, B.A., B.Sc., and H. J. Jerrery, A.R.C:Sc.; F-L:S. Pp. xit+ 334. London: Alston Rivers, tg11. Price 4s. 6d. net. In this we have at last an attempt to form a text-book which will be in accordance with recent advances in botanical science and at the same time in such a form as can be readily understood by elementary students. Here we have the old facts of morphology stated in a new and much more readable way; form is discussed in relation to function, and plant life considered in relation to its environment, while the reciprocal relationships between plants and the various external factors affecting them are discussed one by one, examples being chosen to illustrate what is meant from well-known British or garden plants, thus fixing the knowledge imparted upon the memory. It is impossible to avoid the use of hard terms in many cases, but the paralysing effect of these upon the elementary botanist has to a great extent been minimised by the derivation of each term being clearly stated. The book is divided into three parts :— Part I. is devoted to environment and its influence upon vegeta- tion, and such fundamental factors as water, light, heat, discussed, and the various effects upon plant forms, due to excess or paucity of these essentials, shown. Part II. is devoted to plant biology, and here we find plants divided into biological divisions with chapters upon the “ Division of Plants according to their Longevity” (annuals, biennials, perennials, etc.) and the different sections of perennials shown (evergreen perennials, deciduous plants, etc.). There are also chapters entitled 126 THE SCOTTISH BOTANICAL REVIEW “Classification of Plants according to their Mode of Growth” (terrestine plants, epiphytic plants, and the various forms of these) and ‘Classification of Plants according to their Mode of Nutrition” (green plants, saprophytes, parasites, etc.). This part also contains much information upon the different modes of vegetative reproduction and the storage of food-reserves, and many illustrations are given in the text. Part III. is perhaps of most interest to the general reader as, starting with a discussion of evolution, mutation, etc., it goes on to discuss the origin of the British Flora, and the various well- marked plant associations recognised in this country. This part of the book, consisting of some 97 pages, is particularly well done and eminently readable. The book on the whole is a very good résumé of the works of Schimper, Warming, and the British ecologists in a form easy of assimilation, and after having gone through it the student should be able to read the original works of the various authors with intelligence. Of course some points which are set forth as facts are open to question, but it must be remembered that in writing a text-book it is necessary to give the reader something definite to hold on to; if discussions are entered into and side issues noticed, the results as regards the student are apt to be extremely undesirable. The book is no doubt capable of being misused if looked upon merely as a cramming medium, and is perhaps more suitable to the teacher than to the scholar ; still we cannot help feeling that it may well do something to mitigate the irritation felt by those who, not being interested in botanical science, yet have to acquire some slight knowledge in order to pass the necessary examinations, and show them that botanical science is something more than a collection of hard names and dry facts. TEU "WOODPECKER” Umbrell is the latest invention of the kind. It can be had at all prices from 1Q/6, for Lady or Gentleman. The ‘““WOODPECKER” Sunshade in any colour. TO BE HAD ONLY FROM THE PATENTEES JAS. ROBERTSON & SON, Umbrella Makers, 43 Queensferry Street, laa eee 30 & 32 Leith Street, | George Prescott & Go. (J. H. MURRAY), Specialists 52 QUEEN STREET, GLASGOW; | DUKE STREET, DUBLIN. Spectacle & oS ee Eyeglass ‘i ) Specialists. y ie (i f Grane ty, The New ee Invisible Frameless Spectacles & Eyeglasses A Specialty. 100 Lothian Road, Edinburgh TELEPHONE—1478 Central. Glasgow—8630 City. Dublin—2279. iil rT it Telegraphic Address: *‘LARCH,” EDINBURGH. Telephone No. 2034. PND ( pee Awarded 4 Gold and 4 Silver Medals at recent Flower Shows held in Edinburgh. “a An DAVID W. THOMSON Nurseryman and Seedsman 113 GEORGE STREET — EDINBURGH. —— Seed Warehouse—l18 GEORGE STREET. Selected Vegetable and Flower Seeds, Bulbs and Forcing Plants, Retarded Bulbs and Plants, Garden Tools, Manures, &c. Catalogues Post Free on Application. NURSERIES—Granton Rd. and Boswall Rd. AN EXTENSIVE AND WELL-GROWN STOCK OF Forest Trees of all kinds, Ornamental Trees and Shrubs, Rhododendrons and Flowering Shrubs, Game Covert Plants, Fruit Trees, Roses and Climbing Plants, Herbaceous and Aipine Plants, ALL IN SPLENDID CONDITION FOR REMOVAL. INSPECTION INVITED. Catalogues Post Free on Application. iv ig Contents PAGE THE GEOLOGICAL RELATIONS OF STABLE AND MIGRATORY PLant Formations. C. B. Crampton, M.B., C.M. (Zo be continued) . 5 : : ; : 57 ANTHELIA: AN ArcTIC-ALPINE PLaNnt Association. W. G. Smith, B.Sc., Ph.D. ; : . : : 81 MossES FROM THE WESTERN HIGHLANDS. James Stirton, MD EES. : ; : ran ; ; 89 ScoTTisH FoRMS OF SPARGANIUM. Arthur Bennett, A.L.S. 94 NEW OR IMPERFECTLY DESCRIBED SPECIES OF ACACIA FROM WESTERN AUSTRALIA’ Alex. Morrison . 96 THE Empryo-Sac OF AGLAONEMA. ere Houghton Campbell ; 100 Norte oN VICTORIA REGIA, LINDL. Tatts pene ct eS SHortT Notes :— Philonotis rigida Brid., in Perthshire. R. H. Meldrum 117 Cornus suecica, Linn., in Peeblesshire. G.G. Blackwood 117 Zostera nana, Roth., in gag William ‘ Evans : : ‘ eee $b Petasites albus, Gaertn., in Fife. N, Miller Johnson. 118 NOTES FROM CURRENT LITERATURE j ‘ ee i: E Reviews, Book Notices, ETC. :— The Twenty-Seventh-Annual Report of the Watson Botanical Exchange Club, 1910-1911 ; 121 British Plants: their Biology and ei se F. Bens ang it..J.opehery : ic eats AATE: SCOTTISH WIDOWS’ FUND (ESTABLISHED 1815) The Greatest Institution for Mutual Life Assurance in the United Kingdom FUNDS ANNUAL REVENUE £21,000,000 £2,250,000 MANY VALUABLE FEATURES attach to the Society’s Policies—e.g. Large Compound Bonuses, which are added from the issue of the Policy. Liberal Surrender Values, Loan Values and Paid up Policies. All Options are available after payment of one year’s premium. Prospectus and Quotations sent free on application SPECIAL PROSPECTUS FOR LADIES AND CHILDREN HEAD OFFICE—9 ST ANDREW SQUARE, EDINBURGH GLASGOW OFFICE—114 WEST GEORGE STREET LONDON OFFICES—28 CORNHILL, E.C., & 5 WATERLOO PLACE, S.W. WILLIAM BRYCE “centric ca.s80 54, = EDINBURGH KERNER AND OLIVER’S The Natural History of Plants. - 2000 [ilustrations. 2vols. 21S. FLOWERING PLANTS, GRASSES, SEDGES, AND FERNS OF GREAT BRITAIN. By Anne Pratt, edited bySter. Coloured Plates. 4 vols. for 328. THE a FLOWER GARDEN AND HOME GROUNDS. By W. Rosinson. 15S. for 11s. 3d. BENTHAM AND HOOKER’S British Flora. With Illustrations, 2vols. Lach 9s. : WILD FLOWERS AS THEY GROW. By CorKE AND NUTTALL, Coloured Plates. 2vols. Each 5S. SATURDAY IN MY GARDEN. A Practical Guide to the Making and Management of Small Gardens. Vet 3s. 6d. WILD FLOWERS MONTH BY MONTH. By Step. Coloured Plates. 2 vols. Morocco Binding. 18s. f JOHNS’ FLOWERS OF THE FIELD. Revised by BouLGER. Coloured Plates. New. 53. Sd. BEAUTIFUL FLOWERS AND HOW TO GROW THEM. By Wricut. 100 Coloured Plates. 2 vols. For 12s. 6d. ; BRITISH WILD FLOWERS IN THEIR NATURAL COLOURS AND FORM. By Prof. HEnsLtow. 8S. /or 6s. CATALOGUE OF BOOKS IN NATURAL SCIENCE IN PREPARATION. POST FREE ON APPLICATION