J

THE

JOURNAL OF BOTANY

BRITISH AND FOREIGN

EDITED BT

JAMES BRITTEN, K. C. S. G., F. L. S.

late Senior Assistant, Department of Botany, British Museum.

VOL. LIS.

*5

LONDON TAYLOE AND PEANCIS

RED LION COURT, FLEET STREET 1921.

CONTRIBUTORS TO THIS VOLUME.

Agnes Arbee, D.Sc, F.L.S. Eleonoea Aemitage. Aethub Bennett, A.L.S. G. S. BotJLGEE, F.L.S. James Bettten, F.L.S. N. L. Beitton, M.D. Millee Cheistt, F.L.S. A. H. Chuech, D.Sc, F.R.S. H. A. Cummins. E. M. Cutting, M.A., F.L.S.

A. A. Dallman, F.C.S. H. N. Dixon, M.A.. F.L.S. S. T. Dunn, M.A.

J. F. Duthie, B.A.. F.L.S.

W. T. Elliott, D.D.S., F.L.S.

W. Fawoett, B.Sc.

AXTOXY GrEPP, M.A., F.L.S.

E. S. Gepp.

S. Geeyes, M.Sc.

W. B. Geoye, M.A.

James Geoyes, F.L.S.

M. J. Godfeey, F.L.S.

H. H. Haixes, F.L.S.

W. P. Hieex, M.A., F.R.S.

Ethelbeet Hoexe.

W. Hoetox-Smith.

B. D. Jacksox, Ph.D., F.L.S.

C. C. Lacaita, M.A., F.L.S.

L. V. Lestee-Gaelaxd, M.A., F.L.S.

G. Listee, F.L.S.

J. Cosmo Melyill, M.A., F.L.S.

S. Le M. Mooee, B.Sc, F.L.S.

W. E. Nicholson, F.L.S.

W. H. Peaesall, M.Sc, F.L.S.

R. Lloyd Peaegee.

Sir Dayid Peaix, C.M.G.,

F.R.S. H. W. Pugsley, B.A., F.L.S. J. Ramsbottom, M.A., F.L.S. Alfeed Rehdeb. A. B. Rexdle, D.Sc, F.R.S. H. J. Rlddelsdell, M.A. H. N. Rib-lev, C.M.G., F.R.S. C. E. Salmon, F.L.S. N. Y. Saxdwitii. 0. H. Sabgent. J. Small, D.Sc, F.L.S. T. A. Speague, B.A., F.L.S. T. Stephensox, D.D. T. E. Stephensox, M.Sc H. H. Thomas, M.A. H. S. Thompsox, F.L.S. W. Watsox, D.Sc, A.L.S. J. A. Wheldon. F. N. Williams, F.L.S. A. J. Wilmott, B.A., F.L.S.

A. H. WOLLEY-DOD.

THE

J 0 U R N A L 0 F 13 O T A N Y

BRITISH AND FOREIGN.

ORCHIS LATIFOLIA IX BRITAIN. Bi Rev. T. Stephenson, D.D., and T. A. Stephenson, M.Sc.

In discussing the problem of Orchis latifolia we cannot get any help from the specimen in Linnaeus' Herbarium. The plant has rather short, broad leaves, the lowest 11 cm. by 2o cm. It is widest half-way between the middle and the base. The bracts are long and broad, the flowers broad, with sidedobes of the lip hardly crenulate, and centre-lobe rounded. It cannot be seen whether the leaves were originally spotted or not.

In his description (Species Tlantarum, Edn. 1, p. 191) Linnaeus savs "folia parum maculata praesertim inferiora." In this connection " pa rum " should mean ''slightly." From the reference to "Bauh. Pin. p. SO " and the other works, it would seem that he might admit some unspotted forms.

In his article on the British Marsh Orchises in the Orel id Review (xxvi. p. 161, 1918) Mr. Rolfe says that the O. latifolia of Linnaeus never has spotted leaves. This is totally to ignore the description of Linnaeus himself. The Continental writers generally describe 0. latifolia as a plant with spotted leaves : Klinge says leaves always spotted; Camus, leaves spotted or not; Rouy, leaves usually spotted ; Schulze, leaves generally spotted ; Coste, leaves often spotted with brown. Most of the hgures in Reich. Icon. Fl. Germ, are of plants with spotted leaves, and so also are the figures of Schulze, Coste, and Camus. Some figures also have unspotted leaves — e.g. Flora Danica, t. 266, which is good " pratermissa" Smith, E. B. t. 2308, and Curtis, Fl. Londinensis, 250, which are O. incarnata.

Naturally it must be kept in mind that amongst all the spotted tvpes there are unspotted individuals. In this case reversion seems to be easv. Unspotted examples of O. ericetorum are often found, and perhaps more often of O. Fuchsii. The same is true of both forms of O. purpurella, and it would be expected to occur in the case of O. latifolia. Even then the plants with incurved lip, stronger looped pattern, and coarser texture of flower would be distinguishable from O. prcetermissa of the ordinary type, and by their usually lighter colour, less heavy pattern, and broader leaves from the darker or "northern" form of O. prcetermissa. In the case of hybrids, Journal of Botany. — Vol. 59. [Januaby/, 1921.] b

2 THE JOURNAL OF BOTANY

some have and some have not spots or rings. Even when hoth parents have spots, it is just possible that some individual arising from the cross might he unspotted.

It is necessary first to consider the connection of O. prcetermissa Druce with 0. latifolia L. If Rolfe is right, the former is nothing but "true" latifolia. It is interesting to note that the botanists of AVinchester College had discriminated this plant as a second tvpe of O. incamata. On this, see Bot. Soc. & Exch. Club Report, 1917, p. 174, and the whole article by Dr. Druce on the British Palmate Orchids. Another form with unspotted leaves has also to be con- sidered. This is abundant in Scotland, and, in our view, should be placed under O. prcstermissa, from which it differs in having a richer purple colour of the lip, with much heavier markings, and often slender narrow leaves. Although called by Druce " northern incar- nata" it must be distinguished from the plant which he has named O. incamata v. pulchella *.

AVe regard O. prcetermissa as certainly a distinct species, the identification of which is a great advance towards the complete elucidation of the puzzling forms of the Palmate Orchids. If this species, as well as the " northern incamata " be ruled out, we are left with a number of types of O. latifolia, of which most individuals have spotted leaves. The lip- types of O. prcetermissa and of its " northern " form are given in Journ. Bot. 1920, pt. 556, figs. 5-8.

In discussing the varieties of O. latifolia the question arises as to whether, the unspotted forms being eliminated, we have a true species left, and not a mere congeries of hybrids. As regards British forms, this opinion appears to be tending to dominate, and it may be venturesome to resist it. Yet we do not think the question by any means decided in that sense. Dr. Druce (B. E. C. Club Report for 1917, p. 109) says, " It is evident that much further field-work will have to be done before it is safe to reject latifolia as a British plant." Perhaps he would to-day grant even less than this. Dr. Heslop Harrison has found 0. latifolia in Durham, in profusion, far from other plants. AVinchester botanists, quoted by Druce (I. c. p. 175), consider that O. latifolia " is distinct from evident first-generation hybrids, and possibly exists as a constant species in places where one of the so-called parents does not any longer exist. It may possibly represent the result of hybrids self-fertilizing for some generations and so producing in time a type breeding true, but this poinl again must lie decided by experiment." This statement is open to some criticism from the Mendelian standpoint, since, if a type breeds true it is not due to any time-factor, but to the homozygous character of the uniting o-erm-cells. Another suggestion (/. c. p* 177) is " that the Marsh Orchids constitute either a recent group which has not vet settled down into a well-defined species, or else a species that has become variable and is breaking up into three types." Two quite distinct points have to be dealt with, one whether O. latifolia is ever found growing apart from other forms, in such numbers and of

* Dr. Druce has now published a description of this form in B. E. C. Keport, 1919, p. 577. He names it 0. pr&h rmissa v. pulchella, as it is closely similar to the v. pulchella of (j. incamata already so named by him (B. E C Ron 1117 p. 1G7). V

ORCHIS LATIFOLIA IX BRITAIN 3

such uniformity of type as to give fair presumption that it is pure, and the second, whether, when it is found growing with other forms, it is bound to be a hybrid.

On the first point, Col. Godfery (Journ. Bot. 1919, 138) quotes abundant evidence of the occurrence on the Continent of O. latifolia in plenty, of perfectly regular type, with spotted leaves, where no O. maculata could be found. If this species is thus certified for the Continent, it may very well be taken as probable that the very similar forms found in Britain are also true latifolia, even though they may be proved to be of slightly divergent types.

If of hybrid derivation, the prevalent forms would have to be referred to O. pratermissa and O. maculata for their parentage. This, however, would not hold good in the case of a fairly large batch of plants of O. latifolia which grows near the railway-line close to Borth. Here are many dark forms along with some lighter ones which may be hybrids, where O. ericetorum is present, but no trace of O. prtstermissa. At some distance, in the bog, O. iucar- nata grows, with pale leaves, usually very narrow ; but we should doubt its influence here. There is at least the possibility that O. latifolia is pure. Wherever else we have found it hitherto, O. prcetermissa has been alongside.

Taking up the second point, we think that our extended note on the bearing of the Mendelian theory on our problem has shown that a true species may arise from a cross, as well as from a mutation or series of mutations. Apart from experimental crossing, the pure forms could only be detected by such facts as their persistence through several generations, which would afford a presumption that they were breeding true, and their wide distribution, taken along with dis- tinctness and uniformity of type. As our own preference is decidedly for the mutation theory of their origin, we may say here that the bold and distinct type of lip-pattern of O. latifolia does not by any means suggest to us a cross. Mr. R. D. Laurie, M.A., of Aberyst- wyth College, has suggested that the species may be the result of a mutation in the direction of intensified pigmentation, the spots on the leaves as well as the heavier lip-pattern showing in a two-fold way the outcome of some germinal mutation — probably from O. pree- termissa. The fact that O. latifolia itself appears to cross very freely is also somewhat in favour of its being a true species. Of course, if it be granted that the forms are due to mutation, the specific rank follows ; for we need a name for the segregate. Even if the view be taken that there is extreme mutation, and that the forms are in a state of " polymorphic mixture," still the mixture is confined to a certain limited range of forms, and we need a name for the fluid aggregate, and we need to describe its chief types, if not for ourselves, then for those who come after us, who will have to note the ultimate outcome of the evolution of the forms. In his article in Journ. Bot., Dec. 1920, Col. Godfery deals with the matter from a different point of view, and perhaps better: see p. 289 (3).

As hitherto generally accepted, the characters by which O. lati- folia may be roughly distinguished from other forms are as follows : — The sepals are erect as in 0. incamata, purpurclla, and prcetermissu —

B 2

4 THE JOURNAL OF BOTANY

often quite erect, sometimes less so, but never weak and drooping, as in O. maculata. The spur, though variable, is stouter than in O. maculata, and usually, though by no means always, less so than in the others. The bracts are very variable, the stem more or less hollow. The leaves are broadest in the middle, often very broad and luxuriant, heavily keeled or nearly flat, nearly always spotted, lightly or heavily, or covered with handsome ringed spots. The most typical lip is much broader than long, with well-rounded side-lobes, which are hardly crenulate, the centre-lobe distinct, veiy variable in size, not much or at all exceeding the side-lobes, usually with a blunt and often rounded tip. The lip-pattern is usually a well-defined scheme of broken lines and spots, with spots outside the main lines or not. Sometimes the lines are very heavy and unbroken. The lip is slightly incurved, like a shallow saucer. The throat, where the spur opens into the lip, is much broader than in O. maculata. The colour varies considerably, from pale lilac, with darker markings, to very dark purple. Once we saw a fine, semi-transparent white specimen.

The forms known to us can easily be arranged in four main groups with two sub-groups.

Group A. Probably the most numerous and widely-distributed form has flowers with more or less pale lilac lips, regular, well-rounded side-lobes, a very small, rounded centre-lobe, the whole lip slightly incurved, saucer- wise, and the pattern of strongly-marked lines of darker purple, more or less broken, but often unbroken. The whole plant is about 2 "5 to 4 dm. in height. The leaves are generally broad and spotted, blotched or ringed. The spur is of variable thickness, often rather slender, though not so slender as that of O. maculata.

Sub-group a. In the same locality at Aberystwyth in which the main type of O. piirpurella is found, there is a group of very slender plants, 18 to 24 cm. in height, with flower-spikes 3 to 5-5 cm. long, whose flowers are pale and marked as in group A, and are fully as large, the lip being 10 to 12 mm. wide by 6 to 9 mm. long. The form is noticeable on account of its distinct and very slender habit. At a distance it might pass for a small specimen of O. ericetorum. It is growing in a field where there are many tall plants of 0. erice- torum, so that there is nothing in the situation to give rise to the slightness of the form. We have called it the " heath " form of O. latifolia. It is characterized by its stiff neat appearance, waxy ilowers, short bracts, and rather narrow leaves. At one time we supposed that this was the same as a plant referred to in the Winchester Keports as "down" latifolia: at Winchester it grows on the dry downs. Having now seen some complete specimens, by the kindness of Mr. McKechnie, we find that they belong in general habit to the broad-leaved group A, only being somewhat smaller owing to their dry habitat. On the downs at Winchester they grow with O. Fuchsii, which in general habit they there resemble, so much so that Mr. McKechnie suggests that O. latifolia is nothing but a luxuriant marsh form of O. Fucltsii. No such theory would be suggested by the plant-groups known to us, where O. Fuchsii, in damp situations, is tall and slender, quite unlike O. latifolia.

ORCHIS LATIFOLIA IX BRITAIN 5

which in general habit is much nearer to O. prteiermissa or to O. eri- cetorum, according to its type. The " heath " form at Aberystwyth appears to cross readily with other forms, including O. purpurella. lteichenbach (Icones Fl. Germ. t. 403) gives a form of O. latifolia which the author includes in " latifolise verse " : it suggests a very similar plant.

A single flower of this form is given in Journ. Bot. 1920, pi. 556, fig. 13. It may also stand as a representative of group A generally. A flower of the hybrid with O. purpurella is also given in fig. 11. The specimens were not specially selected, but a comparison of this flower with those of figs. 13 and 9 or 10 will show at a glance the intermediate type.

Group B. The chief difference between the plants of groups A and B is in the darker colour of the flowers and the less conspicuous pattern of the lip which is the necessary result. Apart from the spotted leaves, plants of this group often strongly resemble specimens of O. prcetermissa. The difference would be in the stronger-lined pattern of the lip, its thicker texture, and slightly incurved shape. Usually O. prcetermissa has a lip-pattern of fine spots and an almost flat lip of thin texture — besides the presence of the unspotted leaves. The plants are of about the same size as those of group A. The leaves are usually broad, but there is great variation, and occasionally they are very narrow. O. prcetermissa v. pulcTiella has heavy lip- markings as a rule, but not arranged in regular loops, as in O. hit (folia.

Sub-group b. We think it advisable to refer to a separate sub- group a set of plants growing at Borth, which have very dark purple flowers, heavy lip-pattern, and stout spurs. The leaves are dark green and rather rigid. As we have noted above, this set of plants is growing out of any connection with O. prcelermissa, unless it formerly occurred there and died out. O. ericetorum is found on the spot, and a mile away, across a wide tidal drain, O. incarnata. A lip of this t}Tpe is given in Journ. Bot. pi. 556, fig. 14, and it may also stand as a representative of group B generally.

Group C. In size and habit of flower this group resembles groups A and B, though the tendency is to a very large size. The colour is that of pale 0. ericetorum, and generally suggests one or other of the types of Spotted Orchis. The size of the lip in some cases is very large indeed, 13 mm. wTide by 8 or 9 mm. long. The lined lip- pattern is slighter, and there are more spots, often very small and almost covering the lip. The difference between these forms and O. ericetorum is in the evenly-rounded side-lobes, stouter spur, and more upright sepals. This form will be very near to O. Braunii (0. maculata superba~) — in fact, nothing but experiment could finally decide between them. We have usually found these forms growing in the neighbourhood of the darker ones, but in much smaller numbers. A lip of this type is given in Journ. Bot. pi. 556, fig. 15. In some cases even, the lip is crenulate, much as in O. ericetorum.

In this connection it may be useful to call attention to the fact that O. ericetorum (as well as O. Fuclisii) may grow to a very large size, and in this form may readily be mistaken for a Marsh

(I THE JOURXAL OF BOTANY

Orchis. It is probable that field-botanists pass these forms by without strict examination as O. la t ifolia or hybrids, the more so as they are usually found in marshy places. Obviously the determina- tion of possible hybrids will be much affected by the failure to detect a possible parent. Plants over 5 dm. high are common enough, usually with large and rather flat leaves, solid stems, and typical 0. ericetorum lips, the spur slender, sepals lax, and centre-lobe of the lip very small. Both forms of O. maculata appear to prefer situa- t i> ms neither very moist nor very dry ; but we have found large groups of both forms nourishing in wet sphagnum.

Group 1). We think a form of which we have specimens from Winchester and the Isle of Wight deserving of special mention. It was familiar to the late Mr. Hunnybun, who drew it, and was inclined to regard it as the best type of O. latifolia, owing to the great dis- tinctness of the lip, which did not suggest a hybrid origin. The lip is large and heart-shaped, the centre lobe being scarcely distinguishable. There is a beautifully regular pattern of fine lines, bounded by stronger ones, with a very few dots outside the lines. The spur is very stout, in one example curved a little. The Winchester plant had a few rather large spots on the leaves, the Isle of Wight plant unspotted leaves. These plants might be hybrids of 0. prcetermi ssa and O. latifolia, but it is just as likely that the form is a mutation. The Winchester plant is 38 cm. high, stem stout, with a small cavity, very leafy, the topmost leaf bract-like, the leaves seven in all, four upper ones reaching the basis of the spike, strongly keeled, the longest II cm. long, the widest 28 mm. wide, with a few spots and blotches here and there, rather grey-green, shining beneath. Bracts broad and long, exceeding the fiowers throughout the whole spike. The spike 6*5 cm. long, the flowers lilac, sepals erect, spur stout, rather short, the lip almost perfectly heart-shaped, the centre-lobe hardly apparent, and pattern of slight but well-marked lines and a few dots, the whole lip flat, side-lobes not crenulate. The lip of this type is shown in Journ. Bot. pi. 556, fig. 1G. It is of the Isle of Wight plant.

No doubt O. latifolia is intermediate in many characters between O. preetermissa and purpurella on the one side and O. maculata on the other. A study of the figures in the Plate referred to brings this out quite plainly. Itdoes not necessarily follow that all the forms are hybrids. In many groups of living organisms, species which are never questioned as valid arrange themselves in a regular series of gradations between extreme forms. In spite of arguments to the contrary, the heavy ringed spots of O. latifolia do not suggest to lis dilution, but rather intensification. They are very different from the faint stains that are often found in manifest hybrids. Nor does the characteristic line-pattern of the lip and its regular cup-like shape suggests combination of other types, but an independent variation. \1 any rate, we wish to urge these considerations pending the carry- ing out of experimental crossings, made and tested in accordance with known Mendel ian laws.

By way of summary, we may say that the view which our present knowledge suggests to us is that (J. latifolia is a true species,

OKCniS LATIFOLIA EN BRITAIN 7

and that a truly typical specimen of it cannot be confounded with any other species or with first-generation hybrids ; but that it is extremely variable — some of the forms resembling no other species, some verging more towards O. prcetermissa and some towards O. maculata, but at the same time not in any sense identical with these species.

P.S. — Since the above article was completed, Dr. Druce has published (B. E. C. Report, 1919, p. 90S) a long and interesting note, in which he combats, as we think quite successfully, Mr. Kolfe's contention that O. prcetermissa is the true O. latifolia L. Col. Godfery has also written in this Journal for December last strongly urging the recognition of O. latifolia L. as a valid species. We believe that on all the main points we are in accord with Col. Godfery's views with regard to the British plants. In this and his preceding article (Journ. B >t. 1919, 137-11-2) Godfery has produced ample evidence of the occurrence of a pure O. latifolia on the Continent. It is practically certain that British plants precisely similar to these belong to the same species.

THE LICHEN AS TRANSMIGRANT.

Br A. H. Church.

Oxce the probability of the direct progression of the Lichen from the sea has been put on a reasonable basis, from the analysis of its present somewhat secondary and recapitulatory organization, it becomes possible to make a fresh start at the right end of the story ; and to build up an account of the sequence of the progression, beginning at the marine inception of the problems concerned, instead of working backward from higher land-flora, as has been so often attempted in dealing with hypotheses of the origin of the earlier vegetation of the world-surface. All other vestigial relations of the Lichen now acquire a new and vital interest. It is to the reproductive phases and the stages of the life-cyle that one must look for further suggestions as to the older marine chapters in the history of the race, now seen to consist more probably of many polyphyletic and parallel lines of early algal organism, similarly faced with the problems of the transmigration at a common horizon. The intrusive and accessory algal ' gonidia ' may be largely omitted from consideration. It is on the fungus-component, in its own antecedent algal condition, that interest is specially concentrated ; and this is commonly of very normal Ascomycete habit, following again the presentation of what may be referred on general principles to the earliest phase of the Ascomycete progression. For example, consideration of the working- mechanism of the ascus suggests that : —

I. An open exposed hymenium, with asci discharging their con- tent of 8 ascospores in a volley by hydrostatic tension, is presumably the primitive method of spore-distribution in the group1; and that it was to this end that the ascus was elaborated in the first place from a unilocular sporangium emitting flagellated zoids in an

1 Buller (1909), Researches on Fungi, pp. 234, 210.

8 T1IK .TOUHNAL OF BOTANY

aquatic phase1. The apothecium with freely exposed hymenium may he so far regarded as older than the perithecium, in which protection of the developing asci becomes more important than the actual discharge of the spores in free air. On the other hand, many lichens present the perithecial condition ~, while the general parallelism of organization between a perithecium and a Flondean cystocarp in biological features of parental protection, as well as of nutrition and ostiolar mechanism of emission for immotile spores, suggests a similar origin in the sea, and Pyrenomycetous Lichens may be equally based on marine prototypes.

II. The production of numerous small ascocarps scattered over the general surface, margin, or tips of the soma, is undoubtedly more primitive than the restriction of the soma to one large cup in the manner of Peziza, or the enlarged convoluted hymenium of a Morch- ella ; and the former method of ascus-distribution — as imptying an indefinite number of parasitic sporophyte-stages, following indefinite production of oogonial ramuli — is again carried out much in the manner of the distribution of the cystocarps enclosing similar parasitic diploid generations among the Floridese. Thus it is evident that the apothecia are commonly associated with end-ramuli of the branching soma ; and when, in the limit, the entire shoot-system reduces to a single axis (as in monaxial Angiosperms, or the case of the Cycad from branched arboreal types), the Peziza-cw.]), as also the Agaric- model, present the highly specialized limiting case of a sequence of morphological reduction. In their retention of a primitive con- struction of small scattered apothecia, abundantly produced over a freely branching soma (Cetraria, TTsnea), the fruticose Lichen clearly antedates the more typical saprophytic Ascomycete ; though a multi- branched massive soma with countless perithecia still obtains in XyJaria.

III. Again, it is among the Lichen -forming Ascomycetes that one finds the most remarkable suggestions of vestigial sexual organs, as oogonial and antheridial ramuli, which with the exception of the Laboulbeniaceie, alone among Ascomycetes (and it may be said among all Euniycetes) present any definite suggestion as to the possibility of cross-fertilization, as opposed to decadent autogamy. Indications of a mechanism of undoubted spermatogamy, much in the manner of the Floridese — though very distinct in cytologieal details of the units — present recognizable examples of parallel progression in these widely divergent phyla; similarly expressing by the closeness of such convergence a condition of response to similar conditions of marine environment (as in reef-pools), and undoubtedly at the same algal horizon.

1 V. On the other hand, there is no need to labour the point that, though so curiously parallel, the early Ascomycetes and the Floridese present no direct connection in essential reproductive stages, any more than they do in their cellular somatic organization. The filamentous construction of the Floridean soma, with its mechanism of primary pit-connections, bears no relation to the mycelium of an Ascomycete.

1 Church (1919), Thalassiophyta, Bot. Mem. "3, p. 5G.

- A. Lorrain Smith (1911), British Lichens, vol. ii. p. 203. Pyrenodei.

THE LICHEX AS TKAXSMIGRAXT V

While the Floridean types still in the sea show the unilocular sporangium reduced to an output of 4 tetraspores — as the minimum number following an act of meiosis, and borne on a free-living individual, — the Ascomycetes, transmigrant to the land, still retain what may be fairly regarded as an older condition of 8 ascospores, beyond the meiotic division ; and the latter is borne on a parasitic sporo- phyte-phase. There is no indication that the limiting number 4, so constant among Florideae, when once established, should be increased to 8 with ecpial constancy : the claims of subaerial wastage, for example, may augment the output of asci, but are not likely to necessarily affect the established mechanism of spore-production.

While the algal precursors of the Lichen Fungi thus afford a glimpse of an older and now wholly lost race of marine algse, as well as expressing an older range of the Ascomycete type behind the decadent saprophytic forms — and even suggestions of a wider algal phylum beyond the horizon of the Florideae, — it becomes of interest to follow the course of what would probably have happened to such a race in passing through the vicissitudes of the transmigration epoch, to attain firstly a holosaprophytic habit, and subsequently a recovery of autotrophy in virtue of the helotism of intrusive algse, which has undoubtedly proved curiously successful under certain inferior con- ditions of subaerial environment, in which the plants have no modern competitors. In this respect special interest attaches to the condition described as involving ' intrusive organism ' in the older phase of the sea.

Starting from the lofty standpoint of the higher animal, one is apt to ignore the fact that every such individual organism may be preferably regarded as a special formation in the ecological sense, quite as much as an individual entity. Even in health we each carry an elaborate flora and fauna of our own. In the sea, conditions of life are even more complex, owing to the difficulty of finding suitable unoccupied substratum for the attachment of benthic organism as anchored hormon. In fact, the inception of free-swimming nekton at any point of the animal-series, may be largely regarded as one form of necessary response to this very restriction of available area in such stations. The same applies to the vast amount of epiphytic, epizoic, endophytic, and actively intrusive organism which charac- terizes the lower forms of plant-life in the sea. Germinating spoi-es, in absence of clean rock, must germinate, it at all, on other organism ; and more massive perennial growth-forms may commonly become wholly clothed and buried in a forest of epiphytes to the limit of their hapteron-capacity 1. Where the organism is of simple plasmatic nature, or is only invested by thin mucilaginous wall-substance, there is nothing to prevent intrusion by such spores — more particularly if possessed of flagellated and englenoid activity — and the consequent establishment of one organism within another; taking the chances

1 Cf. Tide-pool algfe smothered in Diatoms : Yendo (1914), Econ. Proc. Eoy. Dub. Soc. p. 105, on the commercial culture of Laminaria in Japan, by putting- clean rock in the sea at sporing-periods : Borgesen (1908) Botany of the Faeroes, p. 758 ; for list of 25 epiphytes on stems of Lamiiiaria Cloustoni (hyperborea), with zonation, p. 757.

10 Till: .TOUKNAL OF BOTANY

of shelter, and food-supply in the form of waste products of the ' host,' but also of diminished light-supply if within a plant, or of speedy digestion if inside an animal. Thus in the densely populated em ironment of the sea, it may he taken for granted that for benthie organism, anything may grow on anything else; and if the host be penetrable, anything may get inside — in every case with certain inevitable consequences. The general facts of intrusive organism thus afford a view of further possibilities of commensaiism or symbiosis, as so far the commonplace of the sea; though appearing more unusual, and hence a phenomenon attracting greater attention, in the case of higher plant-forms and the very specialized vegetation of the land; more particularly as we first become familiar with it in the impoverished versions of northern latitudes. Before emphasizing or exaggerating the unique nature of the Lichen-symbiosis, it may be well to consider the biology of the more general phenomena of ' symbiotic ' life in the sea, from which undoubtedly the Lichen Fungi have been at some time derived. Leaving on one side the case of suggested Bacterial symbiosis, as in forms presumably assisting nutrition by the decomposition of celluloses in the alimentary canal of even higher animals, and the effect of similar Bacteria in decomposing the humus-complex in soil, and so aiding the partial saprophytism of all higher land-vegetation — every grade of association is possible, from harmless and casual intrusion to facultative and obligate association, to symbiotic union closed by the digestion of the in- truder, to the state of complete helotism with loss of any somatic individuality in the intrusive units, or to a condition of complete dependence in the part of the host. Hence in marine biology the question of intrusive organism plays an important part, and soon attracted the attention of early land-biologists ; the possibility of intrusion being based primarily on the opportunity for penetration of the peripheral membranes or tissues of the ' host,' and all such phenomena again being commonly indicated as symbiosis in a loose sense. Recognition of green chloroplasts and algal units in many widely distributed and clearly animal forms was unavoidable, as seen conspicuously even in fresh-water organism as green amoeba?, green ciliate Protozoa as Sientor, the fresh-water sponge (Spongilla), the green Jlydra, as also in the green marine worm Convohota. The alg;e were isolated as Zoochlorella^ , and the photosynthetic formation of starch was demonstrated in several cases2; the vahre of such carbo- hydrate to the host being expressed by keeping green Spongilla exposed to light in filtered water for a month, and Hydra viridis for five weeks3. Similar phenomena are associated with the presence of Mi-own Flagellates, classed generally as Zoo&anthellce, also of widest distribution in the sea. Thus, Brandt (1883) 4 gives a long list of marine organism in which such intrusive 'yellow cells' may be observed; though these do not form starch as a photosynthetic

1 Beyerinok (1S90), Bot. Zeit. 48, p. 725.

2 Carter (1875) for Spongilla ; Ann. Mag. Nat. Hist. p. 187. Brandt (1883), Mitt. Zool. Stat. Naples, p. 229.

r> Brandt, loc. cit. p. 265. .

4 Brandt, loc. cit. p. 191. Famintzin (1889), Mem. Acad. St. Petersburg, xxx vi.

THE LICHEN AS TRANSMIGRANT 11

by-product or reserve, including Forarainifera (Orlitolites, Globi- i/rrina), Radiolaria, Flagellates, Ciliates (Vorticella), Sponges, Hydrozoa, Anthozoa, Ctenophora, Echinodermata, Polyzoa, Vermes (Convoluta, Mwtice), Tunicata and Mollusca, with reference only to intrusive flagellated or otherwise simple protoplasts of an order very comparable with the ' gonidia ' of a Lichen. The case of the Arthropod, with more or less chitinous or calcified exoskeleton, and that of the Vertebrate with similar exoskeletal armour, are alone omitted from the list. For example, among Kadiolaria, more than 100 species have been described as containing ' yellow cells ' or Zooxanthellae, of the nature of Brown Flagellates (whether of Cryptomonadine or Peridine nature), and such intrusive 'algal' zoids, invading the animal plasma, lead on to endophytic and parasitic modes of existence. The photo - synthetic intruder clearly gains shelter and a quiet medium ; and in the case of a holozoic host in which metabolic waste may be predominantly nitrogenous, a definite reason is established for such invasion by algal zoids, in a medium in which scarcity of ions of combined nitrogen is a limiting factor, as a ready solution of the nitrogen-problem. The mechanism of such intrusion may be readily ascribed to chemotactic action, as opposed to simple mechanical intrusion by germinating zoids with rhizoid processes, seeking attachment.

As examples of symbiotic association in which photosynthesis is apparently non-essential, may be instanced, even in the Plankton- phase of the sea, cases as that of the autotrophic Silieoflagellate Distepkanus speculum, occurring in the calymna or outer plasmatic layer of the Radiolarian Pheodaria which has no siliceous skeleton of its own. The Pheodaria apparently collects living Disteplianus and utilizes it instead of spicules. Another Kadiolarian, Airfocleptes, with radial spicules only, utilizes Diatom-frustrules for tangential needles. Free green algal units penetrate the mucilaginous sheath of Schizo- nema Diatoms, giving almost the effect of lichen-gonidia ; and just as Fungus parasites (Alycosphcerella) attack Pelvetia and AsccpriiyUum, so Naviculoid Diatoms may be endophytic in the mucilage of the receptacles of the latter. As amoeba? and filamentous Bacteria may live inside the sphere of Volvox (Molisch), so a Cryptomonad is described as living inside the sphere of PJueocystis ylobosa (Scherffel). Yellow cells or Zooxanthellre are described for Peridines as Pouchetia and Polykrikos ; just as in turn Gymnodines may invade Copepod eggs (Dogiel) and the alimentary canal of Copepods as Blastodinia (Chatton). A good example of such simple association is afforded by the Cryptomonad, C. Schaudinii, living and multiplying in great profusion in the Milioline Foraminifer PeneropUs pertnsus l of the Mediterranean. The Cryptomonad invader is of much the same size as the monokont gamete of the host, and the former are cleared out of the plasma before gametes are matured. Infection takes place at an early date, and the union suggests a simple case of parasitism. Although the Cryptomonad units each possess a brown lamelliform chloroplast, no special advantage to the host can be suggested when

1 Winter (1907), Archiv Protist, Kund. x. p. 16.

1L' Till- .lol'KNAL OF BOTANY

these arc itiside the calcified test, and Cryptomonads are not known to be doniinantly holophytie.

On the other hand, one of the most perfect examples of animal decadence and dependence on the intruder is afforded by the case of the green worm Convoluta roscoffetisis1, in which complete decadence of the metabolic organization of the host follows readily obtained food-supply by the photosynthesis of helot algse of the type of Carteria of the Chlamydomonads 3. Infection takes place in a larval stage, via the mouth, and is obligatory. With the alga growing and dividing, the host ceases to feed, but in absence of the al"'* it dwindles out; on the other hand, the algal units are reduced to complete helotism as residual chloroplasts.

1 n all animal-examples of such commensalism, symbiosis, or dependence, it may be noted once more that there is never any question of 'dual control,' and that there is never the slightest indi- cation of a special somatic factor being introduced as implying a new departure in the somatic organization as a 'consortium.' The expression 'intrusion' covers all the cases. Where there is no mechanical hindrance, anything may invade anything, in the chances of a moving medium ; such intrusion being but an extended phase of older processes of 'nutrition' by ingestion at the surface of freely- exposed cytoplasm. Suggestions of special adaptations in more success- ful plant and animal phyla to keep out such intrusions have been noted, as in the case of the abundant screen -formation of mucilage hairs in many seaweeds : the ' trichocysts ' of flagellates (Pouehetia, Polykrikos) have been freely interpreted from this standpoint. The general organization of higher animals with protective exoskeleton, and internal alimentary canal still exposed to ' infection ' at its digestive surface, follows the same principles ; and even in the plank- ton-phase the first inception of the cellulose or chitinous membrane has been regarded as owing its persistence in phylogeny to its value as a secondary utilization of the debris-heap of waste polysaccharides to this end.

Admitting the general facts and wide distribution of such pheno- mena of intrusion in early marine organism, botanical interest next centres in the manner in which such intrusive photosynthetic units may be utilized as a source of carbohydrate by the more dominant heterotrophic fungus of the Lichen-association ; since, though the algal cells may be intrusive in the fungus-soma, they are not as in the preceding animal-forms intrusive in the actual plasma of the host. Bat even here a broader view is required from the analogy of plant- life as known on the land. Heterotrophy is a phenomenon of signi- ficant importance in all massive plant-growths, even in the sea to the special factors of which it owes its inception; in that all tissues beyond the range of penetration of light must necessarily live at the expense of the surface-layers — over a range, that is to say, of rarely more than 100 /a. Light-penetration in the case of subaerial ve°-e- t at ion may be much more effective ; but the general fact that hetero-

1 Keeble and Gamble (1907), Q. J. M. S. li. 2. Keeble, Plant-Animals (1910), p. 123.

2 Doflein (191C). Protozoenkunde, p. 447.

THE LICHEN AS TRANSMIGRANT 13

trophic fungal tissue should derive soluble sugars (*. e. non-colloidal carbohydrate capable of passing polysaccharide membranes in the absence of direct haustorial perforation or plasmic continuity) \ from full autotrophic algal units, is really at bottom no more remarkable than the commonplace phenomenon observed in the nutrition of non-chlorophyll containing tissues of the more internal parts of higher plants, at the expense of the cortical layer with chlorophyll- content — all such tissues being equally heterotrophic. Heterotrophy is, in fact, one of the most general phenomena in all advanced benthic plant-organism ; the tissues need not be in direct plasmic continuity, and yet soluble materials in excess will be taken by tissues with less content— as shown again by the transfer of soluble substances across the junction of a graft with its stock. The ordinary metabolic mechanism of a massive plant is really run on the same general principles. The wonder, if any, is that the Lichen-habit should not be more general than it is ; and this opens up a wide problem as to the origin of the organization of the normal land-plant itself with its heterotrophic tissues. Given the opportunity, by a special set of biological factors, there is no reason why such intrusion should not work out a successful modus vivendi, and it remains to consider the inception of such conditions.

(To be continued.)

MYCOLOGICAL NOTES.— V. By W. B. Grove, M.A.

(Continued from Journ. Bot. 1920, 251.) Botdia insccjlrta (Oud.) Grove, comb. nov.

Sphceria insculpta Fr. ? Elench. ii. 95 (1828). Oud. Mat. Myc. ii, in Arch. Neerl. Sci. 1873, viii. 405, pi. 6, f. 9 !, and also in Nederl. Kruidk. Arch. ser. 2, i. 184, pi. 5, f. 9 (same figure).

Zignoella (?) insculpta ( Fr.) Sacc. Syll. ii. 225.

V'ialcea insculpta (Fr. ? Oud. ?) Sacc. Bull. Soc. Mvc. Fr. 1896, p. 67, pi. 5, f. 10.

Boyd i a remuliformis A. L. Smith in Trans. Brit. MycoL. Soc. 1919, vi. 151, f. 1.

Gf. also Duplicaria Empetri Fckl. Symb. Myc. p. 265, pi. 4, f. 22 (on Empetrum nigrum).

In August, 1919, I found a number of the trees in the Holly collection in Kew Gardens to be hadly infested with a fungus which had very remarkable and unusual spores. These spores have much the shape of two Indian clubs placed base to base, or rather of the sham Indian clubs used in schools, technically known as " sceptres." On investigation it was concluded that the fungus was possibly what Fries described (I. c.) as Sphceria insculpta. But further enquiry showed that it had also heen met with by other authors and had received various names, as given above.

The spores of all these fungi (except, of course, that of Fries)

1 Paulson and Hastings (1920). J. L. S. p. 497.

14

THE JOURNAL OF BOTANY

seem to be absolutely the same, although the external appearances are not all described alike. Since, however, it is difficult to believe that there should occur on twigs of Holly several distinct species having identical spores of the same unique character, the reasonable con- clusion seems to be that the external appearance is dependent upon outside conditions, while the spores remain essentially unchanged. The Kew Gardens specimens may be described as follows : — Perithecia scattered or aggregated, sometimes two or three being contiguous and immersed in a parenchymatous atro-olivaceous stroma, 300-500/4 diam., globose or lenticular, dull black with a paler centre, immersed in the soft bark only, not sunk at the base in the wood, covered by the shining epidermis and at length piercing it by a small pore. Asci often curved, fusoid-oblong, slightly tapering above, more so below, truncate at the apex, aparaphysate, about 120 x 16 //., soon deliquescing. Spores intertwined and obliquely distichous or

Fig. 1. Boydia insculpta. a, young ascus, X400; 6, immature spore ; c, older spore, X 600. (From Kew Gardens.)

Fig. 2. Puccinia •peueedani-'parisiensis. a, b, spores in air ; c, spore in water,

xooo.

(From near Whitstable.)

tristichous, elongated, doubly clavate-fusoid, i. e. attenuated at each end and narrowed to a filament in the middle, arcuate or flexuose, colourless, filled with guttules and oily protoplasm, 80-100 Xti-9 p, at length 1 -septate in the narrow part.

On twigs of Ilex Aqui folium var. Hendersonii, which it appears to kill. Kew Gardens, Aug., Sept. (Fig. 1).

'The perithecia have a very thick but soft wall, formed of dark greenish-olivaceous cells, either pseudo-parenchymatous or arranged more or less in parallel chains: the stroma, when present, is of the same character. The epidermis over them is for a long time un- broken and rises like a " boil," as if full of whitish pus, surrounded by a black border. At length the epidermis is pierced by a small pore at the summit of the boil : when several perithecia are contiguous,

MYCOLOlilCAL XOTES 15

the pore becomes a slit. The very oily spores may have many or only one large guttule in each loculus and, when mature, they readily break apart at the septum. The compound form seems to be more nearly allied to the Dothideacei than to the Valsei.

The Friesian species is described as growing on decorticated branches of Holly, and having the perithecium semi-immersed, with the upper part deciduous so as to leave the " nucleus " only behind, " nestling in the wood like a Stic/is." When old and exolete, there remains only a white pit surrounded by a black border. Of course, Fries does not describe (perhaps did not see) the spores.

Oudeinans' species seems to have been the same externally, since he does not describe that aspect and refers to Fries without hinting at any doubt, but he adds a description of the asci and spores which shows that they are exactly those of the British specimens. In 1918, Mr. D. A. Boyd discovered the same fungus in Ayrshire, and his specimens were briefly described by Miss A. L. Smith in the Trans- actions of the British Mycological Society (1. c.) as a unique species of a new genus, appropriately named Boydia after the discoverer, although she mistakenly supposes it to be a member of the Sphaerel- lace*. There cannot be the slightest doubt that both these latter are the same, but the Friesian species, which is dimidiate and sunk at the base in the wood might be different.

It is known, however, that the same fungus may occur, sunk in the wood or immersed only in the soft bark, according to circum- stances; e.g. this is true of Phomopsis Diospyri Grove and of Bhomopsis cinerascens Trav., as well as of other species. Often- times the latter case is found on the younger twigs, and the former on the older and more or less decorticated branches.

In 1896 Saccardo described a fungus with similar spores, but provided with a valsoid stroma, 2 mm. in diameter, although he also says that the perithecia are sometimes rather scattered. For this he formed a new genus Vialcea. It is contrary to the principles adopted in Saccardo's Sylloge to place such a species in the same genus as others which have discrete perithecia, but it seems to be now admitted that such a separation is not always accurate. For example, it is known with certainty that some species of Botryodiplodia (with clustered pycnidia, based upon or immersed in a stroma) are not distinct froni certain forms placed in Diplodia (with discrete pycnidia and no stroma). That is to say, the same fungus can develop according to its environment in one or the other form. If so, our fungus might well be placed as a form (f. sparsa) of Saccardo's. Until this is established without a doubt, however, the genus Boydia should be temporarily maintained, but according to the rules the specific name must be insculpta. Thus we should have to Avrite now Jlalcea insculpta and Boydia insculpta, although the latter may turn out to be merely V. insculpta f. sparsa.

Yet this is not quite the whole story. Fuckel, in his Symbols Mycologies (p. 265), described what he considered to be a phacidia- ceous fungus, on leaves of Empetrum nigrum, and invented for it a new genus JJuplicaria. The strange thing is that the spores of Duplicaria Mmpetri Fckl., according to the description and drawing

1 () THE JOURNAL OF BOTANY

(I. c), bear a very close resemblance to those of Boydia insculpta, though the asci are said to be accompanied by long filiform para- jihvses. AVhether these two species are not more closely allied than their present positions would indicate must be left to the future to decide.

PrjcciNiA Peucedani-parisiensis (DC.) Lindr.

In Massee's Mildews, Rusts, and Smuts (p. 122) this species is listed as British, but the specimens available to me for examination in 1913 did not show the true character, and consequently it was omitted from my British Bust Fungi. Towards the end of August last year, Mr. A. T. Eake, of Kew, collected Beucedanum officinale near'Whitstable, where it is abundant in a certain spot. On his showing me the plants, I pointed out that they bore a small number of sori of a Buccinia, and on examination this proved to be the species named above (fig. 2). Afterwards, on referring to Lindroth (Act. Soc. Faun. Flor. Fenn. 1902, xxii. 79), I found that he had received and records a specimen of the same fungus on B. officinale from Feversham (misspelt " Tewersham") in Kent.

This very rare parasite differs from its allies, B. bullata (Pers.) Wint. and B. Oreoselini (Str.) Fckl., in having its teleutospores striped with numerous narrow more or less parallel lines or ridges, which run lengthwise of the spore and occasionally branch or anas- tomose : rarely these lines or strise break up into rows of delicate warts. Those of B. bullata (which occurs on Beucedanum palustre *) are quite smooth ; those of B. Oreoselini are delicately verruculose, but not lineate.

To observe these characters well, a precaution may be needed. As is well known, faint markings of this kind become difficult to see in water — they must be examined in air. I have before me at the present moment a teleutospore of this Buccinia which happens to be divided longitudinally by an air-bubble ; on the half in air the stripes are discernible with ease, on the other half nothing is visible but faint granulations which might be put down to the granular proto- plasm within.

Phomopsis abietina Grove in Journ. Bot. 1918, p. 29.3.

Bhoma abietina Hartig, Diseases of Trees, 1888, p. 138, figs. 78-9 (Engl. ed.).

Bhomopsis pithya Lind, Dan. Fung. 1913, p. 421.

In this Journal (I.e.) I suggested that Bhomopsis pithy a Lind was not identical with Bhoma pithy a Sacc. = Sclerophoma pithya Died., as Lind asserted, but at that time it was impossible for me (not having seen any satisfactory specimens) to give a decided opinion. Since then I have received some excellent examples of the I'homopsis on dead Binus silvestris from Ayrshire, sent by Mr. D. A. Boyd, and also on Bseudotsuga Douglasii, kindly communicated l,v ]\lrs. Alcock of the Pathological Laboratory at Kew, from Perth- shire. These latter are parasitic on small branches of living Douglas

* The locality which is quoted (B)itish Rust Fungi, p. 193) for this fungus should be " Shapwick Bog, Somerset."

MTCOLOGICAL ETOTES 17

Fir, and are exactly what Lind intends : they also agree with what Hartig says and figures so closely that there can he no doubt it is the same fungus which he had before him. The description is as follows : —

Pycnidia densely scattered, convex, erumpent, black, 200-300 ll diam., usually more or less mouthless, sometimes pseudolocellate within ; upper part of wall composed of many thick and dark- brown layers. Spores oblong-fusoid, subacute below or at both ends, 5-6 x Lj-2 p, rarely 1-guttulate ; sporophoxes linear-subulate, 7-10 X 1/u, crowded, mostly straight, but unequal in length, rising from a thick pale olivaceous brown stratum.

There is a great similarity between this fungus and Fusicoccum abietinum Prill. & Del. (Sace. Svll. x. 241), which = Dothiorella pithya Prill. & Del. in Bull. Soc. Myc. Fr. 1890, p. 98, pi. 1-3, figs. 9-11, and is larger in all respects, the spores especially being 12-14 x 5—6 fx. But it is extremely likely that this is another instance of that dimorphism which is well known in relation to Phomopsis quercina v. Holm, and Fusicoccum quercinum Sace. In the latter case there can be found on twigs and branches of Oak every possible transitional state between the two, the spores of the Phomopsis state being 7-10 X l|-2^u, and of the extreme Fusicoccum state 15-16 x 3— 3| jx. In Phomopsis abietina there seems to be a still greater complexity, viz. a Phomopsis state, a Fusicoccum state, and a Dothiorella state ; and, after all, it is by no means improbable that the fungus called Sclerophoma pithya is nothing but a subsclerotioid state of the same species. Whatever the others may be, the Phomopsis state is a decided parasite, doing great harm to numerous species of Goniferae on the Continent, and may become equally dangerous in the Scottish forests.

NOTES ON JAMAICA PLANTS. By William Fawcktt. B.Sc.., and A. B. Rendle, F.K.S.

(Continued from Jorun. Bot. 1919, p. 314.)

EUPHOKBIACTLE.— III.

Phyllanthus caulifloru.s Griseb. A specimen sent by Dr. Brit- ton differs from the specimens from Swartz, which represent all that has hitherto been known of this species, in having inflorescences on the branch as well as on the trunk, thus combining the inflorescence characters of the two species, P. cauliJJorus Griseb. and P. axillaris Muell. Arg. Specimens of this group of species, P. caulijlorus, P. axillaris, and P. cladan+hus are much desired. (Fawc. & Rendle, Flor. Jam. iv. 258.)

RUTACE/E.

"Rhus? 1. Foliis pinnatis orato-acumiaatis subtus villosis, floribus racemosis tetrandris termi iudricibus.'n Tab. 8, fig. 3, Patrick- Browne, Nat. Hist. Jam. 186. There is a leaf of this plant from Browne in the Linnean Herbarium at the end of the 1th us cover, with Journal of Botany. — Vol. 59. [January, 1921.] c

IS THE JOURNAL OF BOTA.NT

a reference in Linnseus's hand to Browne's diagnosis (quoted above), but not otherwise determined by Linnaeus. The specimen includes no (lower or fruit, but from Browne's description and figure of the flower together with the leaf in the herbarium, wre have no hesita- tion in identifying it with Zanthoxylum flavum Vahl. (Fawc. & Bendle, Flor. Jam. iv. 173).

ANACARDIACEiE.

Comocladia troyentis, sp. nov. Arbor parva, trunco tenui. Folia 2-5 dm. 1., foliolis 6-8-jugis, membranaceis, petiolo et rhachi pilosis aut glabrescentibus, infimis ovato-ellipticis quam superiora tri])lo brevioribus, superioribus 5-10 cm. 1. ovato-oblongis aut ob- longis, apice acuto aut brevissime acuminato, basi obtusa insequi- laterali, margine undulato denticulato aut subintegro, nervis promi- nentibus ; petiolulo 1-3 mm. 1. Panicula parce puberula. Flores coccinei, subsessiles. Cah/cis lobi semiorbiculares, quam petala breviter ovata duplo breviores.

Sab. Troy, 2500 ft. Harris, 9349 !

Near C. jamaicensis Britton, but distinguished by the leaflets being in fewer pairs, and having an unequal base.

The following is a key to the species from Jamaica so far as known to us : —

Leaflets entire, not undulate, nor toothed. Plants glabrous, or more or less hairy. Leaflets in 6-8 pairs, upper usually more than 5 cm. 1. Leaflets with truncate or rounded base. 1. C. pinnatifolta.

Leaflets with cordate base 2. C. cordata.

Leaflets in 3-5 pairs, upper usually less

than 5 cm. 1 3. C. parvifoliola.

Panicle, twigs, and leaves brown- velvety ... 4. C. velutina. Leaflets toothed or undulate. Leaflets with large coarse teeth.

Leaflets in 8-10 pairs, apex acute to

shortly acuminate 5. C. Sollickii.

Leaflets in about 14 • pairs, apex ending

in a long slender acumen 6. C. grandidentata.

Leaflets with small shallow teeth.

Leaflets in 10-11 pairs, base subequal ... 7. G. jamaicensis. Leaflets in 6-8 pairs, base unequal 8. C. troyensis.

Note. — Under C. pinnatifolta we include provisionally C. pilosa Britton and O. pubescens Engler, specimens of which we have not seen. The considerable series which we have examined includes glabrous specimens and others with varying degrees of hairiness.

AQUIFOLIACE^.

Ilex florifera, sp. nov. Arbor usque 40 ped. alta, glabra, ramulis ciner< is. Folia 6-10 cm. 1., 4-5-8 cm. lat., rotundato-elliptica, Integra, margine subrevoluta, apice et basi rotundata, coriacea, costa

NOTES 0>* JAMAICA PLANTS 19

supra plana aut subcaniculata, subtus prominente, nervis utritique plus minus prominulis, venis obseuris, petiolis "7-1 cm.l. Injlores- centice (fern, tanturn cognita?) in foliorum axillis plurifasciculatae, 3-pluri-florse, corymbose, peduneulis c. "5 cm. in fl. usque 1 cm. in fr. Flores 4-meri. Calyx 1*2— 1*5 mm. 1., lobis tubo longioribus, trans- verse subellipticis. Petala oblongo-elliptica vel rotundato-elliptica, sublibera, 2*3-2*5 mm. 1., 1 '5-1*9 mm. lat. Ovarium 4-loculare, subovoideum, stigmate capitato. Drupa c. 3 mm. 1., subglobosa vel piriformis, pyrenis 4, 3-angulatis, levibus, linea dorsali instructis, c. 2'5 mm. 1.

Hah. In fl. & fr. Apr., May; Union Hill, near Moneague, St. Ann, Britton fy Rollick 2803! Albion Pen, St. Ann, 2000 ft., Harris, 12,012 !

Near I. dioica Maxim., but differs in the more slender twigs, the entire leaves, the many-flowered corymbose female inflorescence, and other marks.

Ilex uniflora, sp. nov. Frutex 12 ped. altus, ramulis validis, brunneo-cinereis. Folia 9-12 cm.l., 4-7 cm. lat., elliptica interdum ovato-elliptica, subinaequilateralia, apice obtusa, basi rotundata vel obtusa, integra, margine subrevoluta, coriacea, costa nervisque utrinque prominulis, venis obseuris, petiolis c. 1 cm. 1. Injl orescent ice (masc. ttantum cognitaj) in foliorum axillis plurifasciculatse, 1 -flora?, pedicellis glabris, 5-7 mm. 1. Flores 5-meri. Calyx 1 '2-1*5 mm. 1., lobis tubo brevi longioribus, ovato-triangularibus, apice obtusis. Petala elliptica, sublibera, c. 2"5 mm. 1., vix 2 mm. lat. Stamina 5, petalis c. duplo breviora. Pistillodium subplanum, stylo abortivo praeditum.

Hub. In fl. Nov. ; Holly Mount, 3000 ft, Harris, 12,201 !

Near /. nitida Maxim., but differs in the parts of the flowers being in 5 s, the male inflorescence 1-flowered, and flowers smaller.

CELASTRACE.E.

Maytenns microcarpa, sp. nov. Frutex 9-10 ped. alt. Hamuli grisei lenticellis creberrimis notati. Folia 5-10 cm. 1., ovato-elliptica brevissime et obtuse subacuminata, coriacea, margine integro sub- revoluta, Flores 7-8, in axillis foliorum fasciculati ; pedicelli tenues floriferi 2'5 mm. 1., fructiferi 5 mm. 1. C'afycis lobi transverse elliptic! 1 mm. 1. Petala rotundata L-5 mm. 1. Capsula ellipsoidea, subapiculata, brevissime stipitata, 7-9 mm. 1. ; valvar intus purpureaa 4-5 mm. lat. Semina plano-ellipsoidea, 5'5 mm. 1., 3*5 mm. lat., nigra.

Hah. In fl. Dec, Jan. ; in fr. Mav ; limestone rocks in woods, Peckham, Clarendon, 2500 ft. Harris, 11,054, 12,800 !.

Near M. jamaicensis Kr. & Urb., but distinguished by the smaller fruit.

Types in Herb. Mus. Brit. & Herb. Jam.

c 2

'J; I THE JOURNAL OF BOTANY

EPIPACTIS VIRIDIFLORA Reiehb.

By C. E. Salmon, F.L.S.

During the past summer I have had the opportunity of examin- ing, in the counties of East and West Gloucester (v.c. 33 & 34) and Monmouth (v.c. 35), a large number of living plants of an Epipactis which can only he placed under the ahove. In spite of much varia- tion in stature (18 to 75 cm.) and shape of leaf, yet the structure of the reproductive organs of the flower and the form of the epichile remained constant and pointed to Reichenbach's species.

It was evident at once that the plants, which invariably grew in woodland and were in their prime the latter half of July, could not be placed under either of the two forms, dunensis or vectensis, on account of their taller growth, moi'e robust habit, larger and broader leaves, longer bracts, and more numerous, larger, and more open flowers. Indeed, from Messrs. Wheldon & Travis's careful descrip- tion (Journ. Bot. 1913, 313) of the sand-dune plant and from an examination of dried material, it seems that this might rank at least as a good sound variety rather than a forma as suggested by Messrs. Stephenson. Col. Godfery goes further and says (Journ. Bot. 1919, 38) : " I am inclined to think that dunensis has gone far on the road towards differentiation as a species." On the other hand, it is quite possible that forma vectensis may prove to be a name covering small weak examples of type viridijlora itself.

Upon comparing my plants with Col. Godfery's excellent descrip- tion (I. c.) of his var. leptochila and with Surrey specimens. I was convinced that these West Country examples should be placed under his plant, though differing from, it in a few minor details. He care- fully distinguishes his variety from type viridijlora, and emphasizes the following points I had particularly noted in the living plants : — The tall robust stems, sometimes clustered (I saw five together in one spot), the ovate (sometimes almost orbicular) lower leaves, and the large open flowers (rivalling those of violacea) sometimes faintly tinged with purple. My plants possessed almost invariably a notice- ably long lower bract, leaf -like in form. G. F. Hoffmann (Deutsch. Fl. 182, 1801) described his Serapias viridijlora as having bracts longer than flowers, and Fl. Dan. v. t. 811 and Dietr. Fl. Boruss. viii. t. 509 show this feature, but not so pronounced as in my examples. In some specimens the sepals were much broader and less acuminate than one would expect in any form of viridijlora (c. 14 mm. long by 6 mm. broad) ; it is possible that these Avere hybrids with E. latifolia which grew witli them. The epichiles of the plants examined were fairly uniform in shape and match well f. 2 of t. 555 in this Journal, Sept. 1920; the hunches were two (with occasionally a small median one), more or less smooth in freshly-opened flowers, becoming more rugose as the blossoms aged.

As regards the reproductive organs, a sketch made on the spot of a side view exactly matches the drawing A 1 in Plate 553 (op. cit. Feb.) ; it was clear that the rostellum was, for all practical purposes, useless, as in no case could one extract the pollinia upon the point of a pencil as can be done so readily in the case of E. latifolia. Self- fertilisation was a eertaintv.

EPTPACTIS VIRIDIFLORA 21

Mr. T. A. Stephenson, who kindly examined one of the larger examples, wrote : — " Clearly a big form of viridiflora, much stouter and more lat /folia-like in its general louk than any we have had before. The pollen is veiy overhanging and friable, falling on to the stigma of itself, and the rostellum is visible as a rudiment in the bud only, having vanished in the open flower."

A possible arrangement of the British forms is suggested below : —

E. yiridiflora Reichb. [not British unless it includes]

forma vectensis T. & T. A. Stephenson in Journ. Bot. 1918, 1.

Isle of Wight (v.c. 10). var. dunensis T. & T. A. Stephenson [as forma']. Op. cit. 2. Anglesey (v.c. 52) ; S. Lancashire (v.c. 59) ; W. Lan- cashire (v.c. 60). var. leptochila Godfery in Journ. Bot. 1919, 38. Surrey (v.c. 17) ; E. 'Gloucester (v.c. 33) ; W. Gloucester (v.c. 34) ; Monmouth (v.c. 35); Shropshire (v.c. 40).

References should be made to the following valuable articles in tins Journal:— Messrs. Wheldon & Travis (1913, 343), Messrs. T. & T. A. Stephenson (1918, 1 ; 1920, 209), and Col. M. J. Godfery (1919, 37&80; 1920,33).

SOME BRISTOL PLANTS. Br Noel Y. Sandwith.

Ox September 22nd last my mother and 1 were botanizing with Rev. E. Ellman on Combe Down, near Bath, N. Somerset, when, on passing a field of potatoes, we were attracted by a quantity of a fine Fnmariti which we had never seen before. The plants were large, with many long diffuse branches and numerous rather lax and few- flowered racemes of large and very beautiful flowers. The sepals were as broad as the corolla tube, the fruit fairly small, with a very incon- spicuous neck, the fruiting-pedicels not recurved in any specimens we examined. The colour of the long corolla was a deep lustrous pink, that of the tips being blackish-purple. On submitting a specimen to Mr. C. Bucknall, he could bring it down to nothing but the very rare F. r>aradoxa, hitherto only known on the mainland of Britain in Cornwall, and first described by Mr. Pugsley in his monograph. A few days later we sent a small fresh example to Mr. Pugsley, and he replied that the plant certainly came under his F. paradoxa, which he now refers to the Continental F. Martinii of Clavaud. It is, of course, unquestionably a colonist in the Bath locality, but the extension of its range and its occurrence in Somerset is of some interest, apart from the facts that all the Fumitories, even the common one, are scarce or unknown about Bristol.

We were also fortunate enough to make two other rather im- portant discoveries for the Bristol district last summer. Galeopsis speciosa is very rare in this part of the country, and only a single specimen had been found, years ago, not far from Wells. Last

22 THE JOURNAL OF BOTANT

September we found plenty of beautiful plants in cultivated ground on the Somerset peat-moors between Ashcott Station and Glaston- bury. We have constantly seen the species in similar ground and on a similar soil in the black " warp " land of N. Notts, N. Lines, and S.E. Yorks, often associated with Erysimum cheiranthoides, just as it is in this Somerset habitat. If the plant be a colonist — though both it and the Erysimum have been claimed as natives of peaty land — it will rank with Galium Vaillantii and Ckenopodium Jicifolium, rare plants which are very firmly established on these peat-moors, the former occurring in extraordinary abundance in every piece of cultivated ground between Edington and Ashcott Stations, and the latter having been known there ever since the time of Thomas Clark.

Pinguicula vulgaris has once more turned up on the peat-moors after an interval of several years. One plant was observed by Mr. T. H. Green in July towards the eastern border of the moors, and we saw seven in the same enclosure in September. In another enclosure near by, Nitetta translucens grows in masses in several pools left by peat-diggers. It was first discovered for the Bristol district and for Somerset by Miss Honor Perrycoste when botanising with us on the peat-moors in June 1919.

The second new Bristol plant to be mentioned is Sparganiwm neqlectum, which; curiously enough, has never before been noted in our area, though recorded for both West Gloucester and N. Somerset. Search has occasionally been made, and Mr. White has always antici- pated its discovery. We were very interested, therefore, to see it in good condition on October 6th, growing in a pool (an old strontia digging) in a field near Yate Court, W. Glos. Mr. Salmon agrees to the naming, and we have suspicions that the plant also grows on the Somerset side within ten miles of the city. The characters of the fruit are very unlike those of S. ramosum and its variet}r microcarpum, but do not tally with those in Mr. Beeby's plate in eertain minor points.

Not far from Yate Court, we found last May a typical bush of Cradegvs o.ryacanthoides. This had not been known on the Gloucestershire side of the district until Miss Todd found it this year near llawkesbury, and Miss lioper has since reported another bush from Wiekwar.

BIBLIOGRAPHICAL NOTE.

LXXXIII. The Botany of the 'Herald.'

' The Botany of the Voyage of H.M.S. 'Herald' by Berthold Seemann was published during the years 1852-1857, as stated on the title-page, but in the absence of the original wrappers, no in- formation as to the dates of the various parts can be obtained from the book itself. As it contains numerous new species, the dates of publication are of some importance, and the following table has therefore been compiled from data furnished by contemporary periodicals : —

THE BOTANY OF THE ' HEKALD ' 23

Part 1, pp. 5-5Q, tt. 1-10, and map of YY\ Esquimaux-land

(1852).— See Gard. Chron. May 1, 1852, p. 278; Kew Journ.

Bot. iv. 156 (1852). Part 2, pp. 57-80, tt. 13-20, and map of Panama (1852). — See

Gard. Chron. Aug. 28, 1852, p. 517; Phytologist iv. 680

(1852) ; Amer. Journ. Sc. xv. 133 (Jan. 1853). Part 3, pp. 81-120, tt. 21-30 (1853).— See Gard. Chron. Dec. 10,

1853, p. 791 ; Kew Journ. Bot. v. 408 (1853). Part 4, pp. 121-160, tt. 31-40 (1853).— See Gard. Chron.

April 22, 1854, p. 255 ; Kew Journ. Bot. vi. 319 (1854) ; Phytol.

v. 201 (1854) ; Amer. Journ. Sc. xviii. 132 (July 1854). Part 5, pp. 161-200, tt. 41-50 (1854).— See Gard. Chron.

July 29, 1854, p. 487 ; Kew Journ. Bot. vi. 319 (1854) ; Amer.

Journ. Sc. xviii. 429 (Nov. 1854). Part 6, pp. 201-253, tt. 51-60 (1854).— See Trans. Linn. Soc.

xxi. 342 (1855) ; Amer. Journ. Sc. xix. 489 (May, 1855). Parts 7 and 8, pp. 255-320, tt. 61-80 (1856).— See Amer. Journ.

Sc. xxiii. 127 (Jan. 1857) ; Proc. Linn. Sue 1S56-7, p. Iv. Part 9, pp. 321-360, tt. 81-90 (1856).— See Amer. Journ. Sc.

xxv. 116 (Jan. 1858) ; Proc. Linn. Soc. ii. p. Iv ; Gard. Chron.

Feb. 14, 1857, p. 103. Part 10, pp. 261-483, tt. 91-100, and cancel-leaves 253-4,

279-80, 345-6 and four others (1857). — See Amer. Journ. Sc.

xxv. 116 (Jan. 1858); Proc. Linn. Soc. 1857-8, p. lxx ; Gard.

Chron. Nov. 7, 1857, p. 759.

The date of part 4 is a little uncertain : following Kew Journ. Bot. it is here given as 1853, but A. Gray gave 1854 as the date, and it was not reviewed in the Gardeners'' Chronicle until April 22, 1854. Another doubtful point is what pages were comprised in parts 4 and 5 respectively. We know that part 4 began at p. 121, and part 5 ended with p. 200, and it is here assumed that each part contained 40 pages, part 4 including pp. 121-160 and part 5 pp. 161-200. On the other hand, Asa Gray stated that part 4 included about half of the Composite, which suggests that it may have stopped at p. 152.

According to A. Gray and the Gardeners' Chronicle, 14 pages (7 leaves) were reprinted to correct errors and give additional infor- mation, and were issued with part 10 in 1857. These included pp. 253-4, 279-80, and 345-6 ; the four remaining leaves have not been traced. Cancel-page 254 is headed " Further Additions and Corrections, June 8, 1857 " ; cancel-page 280 includes a new species Tephrosia madrensis Seem., which appeared on the original page as Galactia marginalis Benth. ; and Mr. Britten points out that the leaf 345-6 in the Natural History Museum copy is fixed to the remains of one which previously existed. The writer is indebted to his colleague, Mr. S. A. Skan, for some important references, without which the preceding account could not have been completed.

T, A. Spragtje.

In his preface to the work, Seemann acknowledges the help of various botanists " whose labours will confer a lasting value upon it however small the merit that may he ultimately assigned to the

24 THE JOURNAL OE HOT ANT

parts worked up by myself." The names of the authors are attached to their various contributions, but it may be useful to bring them together in one list, with an indication of the regions to which they relate. The list is arranged under the authors' names : —

ChttRCHILL BABINGTON — Licheries : Eskimaux-Land, 47-49; Panama, 246-48; Mexico, 344-5; Hongkong, 432,

A. H. R. Giusebach — G-entianacece : Panama, 169, 170; Mexico, 318-19.

\V. H. Harvey — Alga: Eskimaux-Land, 49, 50.

F. Ivlotzsch — Euphorbiaceee : Panama, 99-106; Mexico, 276-8.

J. MlERS — Panama: Menispermaceie, 76-7<S ; Solanacece, 172-

176.

F. A. W. MlQUEL — Panama: Artocarpeee, 195-197; Piperacece, 197-200.

YV. Mitten — Hepaticae: Panama, 245-6.

W. Muneo — Hongkong: Cyperacece, 4t21— 3 ; Graminece, 423-4,

Neks ah Esknbeck — Panama: GyperaeeeB, 221-223 ; Graminece, 223 22",.

H. G. Reictienbach — Orckideeb : Panama, 214-5; Hongkong, 417-19.

F. ScHEER— Cactece : Mexico, 285-93.

C. H. Sckuetz-Bil'ONTinus — Composite : Mexico, 297-315.

John Smith — Filices : Eskimaux-Land, 44 ; Panama. 226-213; Mexico, 337-343; Hongkong, 425-431. Lycnpodiaccce: Panama, 213; Mexico, 343-4; Hongkong, 431. Marsileacece : Panama, 244 ; Mexico, 345. Equisetacece : Panama, 244.

J. Steetz — Gompusitce : Panama, 139-163 ; Hongkong, 384- 395.

W. Wilson — Musci : Eskimaux-Land, 44-46 ; Panama, 244-5 ; Mexico, 344; [Hongkong, 432].

In the LeguminoscB, Scrophular/acece, and Labiates of Panama, Seemann (106, 176, 187) acknowledges the assistance of George Bentham. Wilson's name is not printed in connexion with the Hongkong mosses, but Trimen has added it in MS. in the copy in the Department of Botany. Two species of Quercus from Panama (Q. Sfemannii and Q. Warcewiczii) are named and described (pp. 251-2) bv F. Liebmann.

The study-set of Seemann's plants is that in the National Her- barium : see Journ. Bot. LS89, 102-5. James Bhitten.

REVIEWS.

T!,r Cambridge British Flora. By C. E. Moss, D.Sc, M.A., F.L.S., F.R.G.S., Professor of Botany at the University College, Johannesburg, University of South Africa, assisted by Specialists in certain Genera, illustrated from Drawings by E. W. HuNNT- i!i n. Vol. III. Portulacacese to Furnariaceae. Folio; paper boards. Pp. xvi, 200, tt. 191. Cambridge University Press. 1920.

A series of complications, largely arising from the late war, is doubtless responsible lor the delay in producing this volume — the

THE CAMBRIDGE BRITISH FLORA 23

second in order of appearance — and for the greatly increased price — £'G los. net as against £2 5s. net — at which it is issued. These drawbacks may be regarded as inevitable, but they nevertheless militate strongly against the success and usefulness of the work. At this rate of progression, comparatively few of those who subscribed to the first volume can expect to see the last — some indeed have already passed away — and this consideration, coupled with the cost, will, we fear, deter many from embarking on what must prove an expensive enterprise.

The general features of the Flora were discussed in our review of the first volume (Journ. Eot. 1914, pp. 131-4), and there is no need to dwell upon these, as the new instalment naturally follows the plan of its predecessor. A fuller acquaintance with the work confirms the impression already conveyed that the help which can be rendered by typography is not adequately realised: for example, in the index (we note with pleasure that there is only one) all the names, whether retained or synonyms, are printed in roman type, although everyone knows the convenience of the usual differentiation by which the latter arc printed in italics.

The volume is almost entirely the work of Dr. Moss, who has however been helped in " Diaulhaeeee " — more usually known as Caryophyllaveft- — by Dr. Druce for Moenchia and Cerastium and by Mr. R. II. Compton for Li/c/t/u's and allied genera: Dr. Druce is also responsible for Jlonfia and Mr. Pugsley for Fumaria. The Syndies of the University Press acknowledge Mr. Wilmott's "valuable assistance both in correcting proofs and in dealing with matters which are usually settled by an editor, in the absence from England of Professor Moss." Mr. Wilmott also contributes a prefatory note on the late Mr. Hunnybun, to whose generous presentation of his drawings to the University the inception of the Flora is due : he pavs a high tribute to the artist's work, and justifies the " ultimate limitation of the portraiture to a single specimen " as resulting " in a corresponding gain of that permanent truth of observation which was to him the first requirement." As a rule, the drawings are suffi- ciently characteristic — in many instances they are excellent ; some- times, however, they are less satisfactory : the Chickweed, for example (t. 50), though doubtless an accurate portrait of the specimen Hunnybun had before him. certainly does not portray the plant as we commonly know it. The arrangement of the drawings on the plates is sometimes bad — e.g. Nuphar, Pulsatilla, and Montia fontana var. 1cm prosper ma ; and we do not understand the black shading in Stitchwort and some allied plants. The cost of the work might have been considerably reduced and its appearance improved if two or three species in genera such as Alsine, Sagina, and Arenaria had been brought together on one plate instead of very inadequately occupying one apiece : the appearance of the mysterious S. " boydi " on a folio page all to itself is comic, even though the " single tuft " be split into six fragments, as if by an explosion. The dissections as a whole are inadequate — an exception occurs in Fumaria, due, we understand, to the wise insistence of Mr. Pugsley, who is responsible for at least some of them ; it would have been better, we think

2G THE JOURNAL OF BOTANY

to have given these as blocks in the text, to which the excellent maps showing distribution form a useful addition.

An adequate criticism of the text could only be undertaken by one With a more thorough knowledge of the plants described than the present writer can claim. The elaborate division and subdivision of species has necessitated the creation of a number of names : it will take us long to become reconciled to the fact that when we read " Compton in Moss Brit. Fl. iii, p. 09," the reference is to the page actually under our eyes, and indicates that the name to which it is appended is here first published : there is presumably some reason for this departure from custom, but we have failed to find it stated. The notes throughout are often of much value and interest, and add greatly to our knowledge.

Turning over the pages, wherein it is pleasant to note the fre- quency with which this Journal is quoted, we observe that several ] .hints make their first appearance in a British flora: one indeed is altogether new — the small White Water-lily, usually regarded as a variety of Nymphcea alba, is raised to specific rank as N. occidental is. Jersey, it will be remembered, is included in the Flora ; we ■ have thence a remarkable form — "forma luxurious" — of Ficaria; Dian- thus gallicus Pers., of which "a single rather large patch" was found "on fixed dunes in St. Ouen's Bay"; and "Ranunculus aleae" — a name which looks somewhat less odd when spelt, in accordance with general custom, with a capital letter : it commemo- rates one Francis Alea, who was attached to the Madrid herbarium ami discovered it near the Escorial in 1843 : this was first found by Hunnybun in Jersey, where it is "quite a feature in some places in the dunes : the paler tint of the flowers and the more patulous habit enable the botanist to distinguish it at a glance from B. bul- bosus with which it grows." Several introductions receive the honour of plate and description, some of them we think, on somewhat insufficient grounds — e.g. Tunica Saxifraga, found "at the foot of a land-cliff on ground adjoining a public path near the railway station, Tenby : doubtless a garden escape originally and subsequently self-sown."

We note that Dr. Moss does not allow the claims of Aconitinn NapelluB to nativity, but regards it as " a recent escape from culti- vation " : " it is inconceivable," he says " that such a handsome and conspicuous plant could have been completely overlooked or ignored by all British botanists during the seventeenth and eighteenth centuries and the first part of the nineteenth." There is of course much to be said for this contention, but in speaking of it as a "recent" escape, Dr. Moss overstates his case. A reference to E. B. Suppl. 2781 shows that in 1819 it was found abundantly in Here- fordshire, and in 1820 very abundantly on the banks of a stream near Wiveliscombe in Somersetshire — in which county it has all the appearance of nativity — where it extended at intervals for three miles in these localities it must have existed for some considerable period before its discovery. The Steep Holm Peony (Pceonia mascula Miller, 1768; P. corallina Retz, 1783), which Marshall thought to be native, is regarded as " naturalised " : Smith (E. B. 1513), who by

THE CAMBRIDGE BRITISH ET.ORA 27

the way regarded Aeon it um as an introduction, thought it had " probably grown there from time immemorial," and mentions that its discoverer in 1805 met two fishermen who " could recollect having gathered its flowers sixty or seventy years ago " : it was however not noticed by Banks on his visit in 1773.

We cannot conclude this notice without referring to and endorsing a protest made by Dr. Rendle in his review of the book in Nature for Nov. 11, 1920. We had intended to pass the matter by with the remark that we did not accept as accurate the interpretation, worded in a needlessly offensive manner, placed by Dr. Moss on a notice in the Journal ; but Dr. Kendle has expressed on general grounds what might have appeared on our part the outcome of a personal grievance. His protest is as follows : —

" It is to be regretted that personal matters should have been introduced into a work of this kind. The Cambridge British Flora will, presumably, take rank as a standard work, a presentation of the knowledge and views of eminent British botanists at a period in the history of botany, and to perpetuate the differences of opinion winch have arisen on matters of very secondary importance detracts from the dignity which such a work should possess. The syndics of the Cambridge University Press would have been well advised if they had exercised a fatherly censorship on several paragraphs in the introduction to the present volume."

Conifers anil their Characteristics. By Charles Coltmax-Rogers. With Illustrations. Pp. xiii, 333. John Murra}r. Price 21s. net.

The genera of Coniferce, now about 40 in number, remnants of a very ancient and once more varied group, are not very difficult to discriminate — when, at least, both cone- and leaf-characters are avail- able ; nor when growing in a wild state do the small number of species inhabiting any one country present any very great difficulties. The great majority of the 380 species which the Order comprises are natives of temperate climates, so that more than 200 of them are cultivated side by side in our British pinetums. and then the diffi- culty of distinguishing between species nearly allied to one another, though from different regions, is considerable As Mr. Coltman- Kogers says in his " Prefatory," " it is hardly possible to carry into a friend's collection of growing trees the seven volumes of Ehves and Henry's Trees, Mr. Bean's two, Mr. Clinton Baker's, or even the one volume of Yeitch's Manual.'''' A pocketable key to the genera and species is, therefore, unquestionably a desideratum among the many who — whether as botanists, as growers, or otherwise — take an interest in conifers. It is, perhaps, hardly possible, even with the aid of a copious glossary, to render such a key practically useful to the non- botanical without the assistance of figures. The general habit of growth can, perhaps, be adequately described in words ; and all that is needed in the form of twig, leaf, and cone can be fully illustrated in black-and-white outline-drawings in the text, without that use of colour-printing and loaded paper which makes Graf Silva Tarouca's

L'S THE JOURNAL OF BOTANY

otherwise excellent Vhsere Freiland NacleUwlzer so much the reverse of portable.

Mr. Coltman-Rogers has essayed the formation of a key which,

he says, he has transcribed in pocket form. In the present volume the tables occupy pp. 263-305 ; and, apart from the drawback that they are in type far too small for the eyes of many people, they may serve fairly well for the discrimination of the species and varieties when the genus is known. There is, however, no key to the genera, and the characters of the subgenera only appear at the heads of the lists of species and not in subgeneric keys. The author has apparently so great a dislike for the word " genus " that the Sub- tribe Cal'litrince appears with the " Subdivisions or Species " Oallitris and Widdringtonia and the Tribe PodocarpecB with the "Sections" Podocarpi, Prumnopitys, Saxegothea, and Microeachrys. There is a good glossary in which, as elsewhere in the book, the author shows a considerable knowledge of Latin etymology ; but it is somewhat unfortunate that the two entries " Nut. A seed enclosed within a hard shell," and " Nux Banata. A nut enclosed in a pulpy covering, as a Yew berry " should appear in close juxtaposition ; and the defi- nition of "decussate" as "applied to leaves and branchlets arranged in pairs " hardly covers the case of the leaves of Juni penis, to which it is commonly applied. The " Identifying Tables " are preceded by 262 pages of gossip more or less concerned with confers. Two pages are devoted to violins and spruce-trees, twTo to Conium maculatum as a method of execution in ancient Greece, and a good deal of space to the short-comings of public school education, the relative knowledge of country life possessed by Virgil and Pliny, etc., etc. In a volume intended for the pocket one may resent the mere waste of space involved in such a discussion of the appropriateness or inappropriate- ness of a technical term as that which occupies most of p. 103, oi even a shorter circumlocution which can say of the stem of a Sequoia : —

"In shape it resembles, with its tapering stem, the familiar form of that Dairy Company's milk-can that we all know so well and hear so often, to the disquietude of our system and the disturbance of our nerves, rolled and jangled along the platforms of our island home railway stations."

A page or two might well have been saved by the omission of the entirely superfluous " the " before the names of species, as if they were Irish chieftains of exalted genealogy. When, however, the author writes of Abies sachalinensis that in Hokkaido " it is the sole representative of its species," he does not mean what his words strictly imply, that there is but one example on the island, as seems to be the case with Santalum fernandezianum on Juan Fernandez ; but simply does not know the difference between a species and a genus. It is clear that Mr. Coltman-Rogers loves his trees, and he may himself know them well ; but, while it is painful to find a tree- lover writing of Evelyn's " Sylvia,'''' it is a more serious defect when obscurity of style makes it difficult to gather the meaning of sen- tences intended to inform. The following sentence on the name of Picea Smithiana is a fair example : —

COXIFERS AND THEIR CHARACTERISTICS 29

" As far as the names go the vernacular aliases (e.g., Morinda) of this tree have — with all apologies to Sir E. Smith, first President of the Linnean Society, and after whom it was named — a far more pleasing ring than the name finally bestowed on it by priority of publication, but not given, we mark, by priority of name as bestowed upon it by the natives and European dwellers before the date 1812, when it was described and figured by Dr. Wallich."

The use of initial capitals to all specific names adds to the amateurish appearance of the book, as also do such abbreviations as " Ps. Ts. Doug. var. Colorado " in the alphabetical list of names which does duty as an Index.

A really useful handy guide to hardy conifers has yet to be written.

G. S. BOULGER.

BOOK-XOTES, NEWS, etc.

The journeys by which Kegixald Farrer enriched our gardens and greatly extended our knowledge of the flora of Tibet and China have been brought to a close by his death, which occurred during the last of them, on the frontier range between Burma and China, at the early age of 40. An article (accompanied by a portrait) in the (jtinlcnrrx Chronicle for Nov. 20, summarises his travels, which have been recorded at length by himself in the Journal of the Royal Horticultural Society, vol. xlii., and in the Chronicle for 1919-20. Farrer, though not strictly speaking a botanist, was in the first rank of collectors : he " sent home the best seed and obtained the best germinations " ; he never " succumbed to the fatal temptation to collect a plant simply because it was new : it was enough for him that it was beautiful and not yet in cultivation." Several hundreds of his plants have proved to be new, and many bear his name, though, so far as we are aware, it has not been bestowed upon a genus. On rock-plants and gardens, Farrer was a recognised authority — his book on The English Rock-Gar Jen, noticed at length in this Journal for 1919 (pp. 3-34-357), is a good example of his discursive style, which may perhaps be accounted for by the fact that he was also a novelist of some distinction.

Sir David Ernest Hutchins, whose death was reported from New Zealand at the beginning of December, was one of the ablest and most experienced forest officers in the Empire. Born in 1850, he was educated at Blundell's School, Tiverton, and entered the Indian Forest Service from the Nancy Forest School. After ten years' Indian service, he entered that of South Africa and devoted himself mainly for the next twenty-three years of his life to the study of extra-tropical forestry, reporting on the Transvaal in 1903, llhodesia in 1904, the slopes of Kenia in 1907 and 1908, British East Africa generally and Cyprus in 1909, West Australia in 1914-15, and New Zealand in 1916, and visiting Mexico to examine the pines suitable for introduction into Rhodesia. Hutchins's reports are alwa\'s marked by a careful discrimination of species with special reference to their ecology. Braclu/loena Hutchinsii Hutchinson, a Composite which

30 TTTE .TOriWAT, OF BOTAJsT

yields one of the most useful of East African timbers, was dedicated to him as practically its discoverer in 1910, and at the beginning of 1920 he was knighted for his services to the forestry of the Empire.

By the death of Odoakdo Beccari on the 25th October, at Florence (where he was born on Nov. 17, 1843) the Linnean Society has been deprived of its oldest Foreign Member, who was for inaiw years Director of the Botanic Garden in his native city. His name is associated with a large number of publications, the earliest of which — Malesia, extending from 1877 to 1890 — is devoted to the plants which he discovered during his travels in the Malay Archipelago and New Guinea in 1865-76. His chief work, however, was among the Balms, on which family he published numerous memoirs, the principal being a monograph of the Lepidocaryece.

The Transactions of the Lincolnshire Naturalists'1 Union for 1919 contains a paper by the Rev. E. A. Woodruffe-Peaccck on " Seed Dispersal " — a subject to which he has paid much attention. Special reference is made to the dispersal by birds, whose methods of transport are divided into three groups: "I. Internal Seed Carriage, in the crop, in the gizzard, in the alimentary canal : II. External Seed Carriage, in mud, in clay-balls, amid ruffled feathers, by mucosity : III. External Portion of Plant-Carriage, on the backs, round the necks, on the feet." The observations have been made during a long series of years on trees, bushes, and large plants, and contain much information of value and interest.

The Transactions of the Botanical Society of Edinburgh (vol. xxviii. pt. 1) contains an interesting paper by the Hon. W. R. Riduell on the pharmacopeia of The Vegetable Family Physician, a 12mo volume of 176 pages published at Boston, Mass., in 1836 : the author, Samuel B. Gunnons, was the editor of the Botanical Journal, a monthly magazine published at Boston; "he seems to have known considerable (sic) about the botany of his district, and most of his descriptions of plants are clear and easily recognisable." Col. H. II . Johnston contributes numerous notes, containing much of critical interest, on the Flora of Orkney, and Dr. Malcolm Wilson describes a new Phomopsis — P. Bseudotsugae — parasitic in the Douglas Fir. The editor has not yet realised that the tops of pages should be employed for conveying useful information, but is content to occupy them by a reiteration of the fact that they belong to the "Trans- actions of the Botanical Society of Edinburgh."

The fifty-second volume of the Transactions of the Devonshire Association contains a paper by Mr. G. T. Harris on the Fresh-water Algse of the county, with a survey of what has been done in their investigation since the first list appeared in the Flora Bevoniensis in 1S29. Special notes on some of the more interesting species are given, followed by a list of new records'for the county. The Twelfth Report of the Botany Committee in the same volume, edited by Miss C. E. Barter, contains numerous additions to the flora of the districts.

The appeal on behalf of the Watson Botanical Exchange Club, which was issued in October, has, we are glad to say, proved successful, and the Club, which since its formation in 1885 has done much

BOOK-NOTES, NEWS, ETC. 31

interesting and useful work, will continue, though it has become necessary to raise the annual subscription to 14.v. A change in the officers is announced : Mr. S. H. Bickham, who has acted as treasurer for thirteen years, is retiring, and will be succeeded by the present secretary, Mr. George Goode ; his place will be taken by Mr. H. Stuart Thompson, who occupied the same position in 1900-1905.

At the meeting of the Linnean Society on December 9, Mr. Lestei'-Garland exhibited a selection of the plants collected at Darfur by Captain Lynes and remarked on their geographical distribution : we hope to publish in an early issue a paper embodying Mr. Lester- Garland's observations. Dr. Daydon Jackson made an interesting communication on "'The Norsemen in Canada in a.d. 1000, with the Plants they collected." He explained that his remarks were limited to the introductory pai't of a lecture prepared four years previously, which had been postponed delivery. Starting from the paper read by Dr. Frithiof Nansen before the Royal Geographical Society on the Oth November, 1911, he emoted from recent papers by Daniel Bruun and H. P. Steensby in Jlrddelelser om Gronland, vols, xvi., xvii. in 1918, and a sliglvtsketch by Prof. H. 0.oJuel, in the current volume of the Svenska Linne-Sallsl-apets Arskrift, p. 01. The course followed by the Norsemen was narrated, from their colonies in Green- land across Davis Strait, to the North-east coast of Labrador, southward through Belle Isle Strait to the valley of the St. Lawrence, and the tract of country on its right bank, where vines were found growing, unsown corn, and a tree called ;' Masur," these being regarded as Vitis Labrusca L., Zizania aquatica L., and an Acer. The reasons why these voyages were not continued were explained as due to the weak colonies at that time in Greenland, the actual starting-point, and the opposition of the natives, termed " Skraellings," who prevented any attempts at settlements in li Vinland " — the "Wine- land of the sagas of Erik the Bed, and of Thorfinn Karlsefni, — the northern part of New Brunswick.

Mr. Martin Ni.thoff, of the Hague, has published a NaamJijst of the plants published in the Flora Batava, with reference to the volume in which each is described. It consists of two parts-— the first containing the Latin, the second the " Nederlandsche Namen." In the latter, the principle which often prevails in English books, pre- supposing that every species must have a vernacular name, is earned to an excess which seems to us absurd ; not only is each species so provided, but genera which have no equivalent in Dutch appear here under their Latin names with a Dutch rendering of the trivial — thus " Paxillus, zwarttluweelige."

The first number has been published of The Flowering Plants of South Africa, a new quarterly serial edited by Dr. Pole Evans, Director of the Botanical Survey of the Union of South Africa, published in England by Messrs. Lovell Reeve. Each part (15s.) will contain ten coloured plates from drawings by Miss K. A. Lans- dell ; the accompanying letterpress is by Dr. E. Percy Phillips, of the National Herbarium at Pretoria. The first part contains descrip- tions and figures of three new species — Arctotis Fostrri N. E. Br., Cyrtanthus contractus N. E. Br., and Leucadendron Stokoei Phillips.

32 THE JOURNAL OF BOTANY

The finding of Orchis hircina is, of course, interesting to British botanists, but one would hardly have expected to find the following in the Times (Nov. 13): — "Mr. Frank W. Stedman, a botanist, who was associated with the finding of the rare Lizard orchid at a time when it was believed to have become extinct in this country, died suddenly yesterday at Ash ford, Kent."

The Journal of the Linnean Society (Botany : vol. xlv. no. 302, Dec. 7) is chiefly occupied by " A Contribution to the Flora of Australia " by Mr. Spencer Moore, in which he discusses two species of Trihuhis described by Robert Brown, and describes new and rare species from older collections, mostly at the British Museum, and from recent collections by Dr. F. Howard and Mr. J. E. C. Maryon. Many new species and a new genus — Leptospermopsis, allied to Leptospermum — are described — this and two other novelties are figured, the plates being somewhat unsatisfactorily lettered " Aus- tralian Plants." The paper, considering the cost of printing, might have been more economically printed. The number also contains an account of a visit to Kunadiyaparawita Mountains, Ceylon, by Mr. Frederick Lewis, and on variation in the flower of Jasminum malabaricum, by Dr. H. H. Mann.

The Transactions of the British Mycological Society for 1919 (published 1 April, 1920) did not reach us for notice. It contains records of the "fungus foray " held at Baslow, Derbyshire, in 1919 : Miss Bayliss Elliott continues her " Studies in Discomycetes " and figures and describes two new species — Phoma conicola and Mollisia Populi ; Mr. Gr. O. Searle contributes an exhaustive study of Erysiphe Polyqoni in its relation to cultivated Brassicese, and there are numerous short papers by Miss Lorrain Smith and others.

The Department of Public Instruction of New South Wales is issuing a series of publications on The Australian Flora in Applied Art, the first instalment of which is devoted to the Waratah (Telopea speciosissima~). The handsome inflorescence lends itself readily to decorative treatment ; the volume contains numerous illustrations, showing its application to a great variety of subjects, ranging from architecture to umbrella-stands. " Its recognition as the leading flower in Applied Art," says Mr. 11. T. Baker in his preface, " is a pleasing connecting-link between the aesthetic taste of the autoch- thonous inhabitants of this Continent and the British race, for the former named it Waratah, signifying the finest in the Australian bush, to which decision one might add, if not the whole botanical world." Some of the designs for tiles and wall-paper are very effective. The so-called " Australian Waratah Legends " which appear as an Appendix under the heading "Literature" are wonderful examples of " English as she is wrote " in Australia.

Thanks to the generous support of our readers, the deficit which threatened to endanger the existence of the Journal has been cleared oft", and it will therefore continue, at any rate for the pre°ent year. A list of those who have contributed to ensure this result will be found on p. 3 of the wrapper.

33

EPIPACTIS LATIFOLIA IN BRITAIN. {Epipactis laiifolia All., including E. media Fries, as described by

Babington.)

Br Rev. T. Stephenson, D.D., and T. A. Stephenson, M.Sc.

In this Journal for September 1920 (pp. 209-212) Ave published a note on the British species of Epipactis. The present paper is an attempt to justify the position there taken up as to the forms hitherto generally assigned to E. latifolia or media. It seems advisable to start with Babington's diagnosis of the two species in his Manual of British Botany, and to discuss afterwards the literary questions involved, which; happily, are now resolved with almost complete finality.

The differences between E. latifolia and E. media, as given bv Babington, are that in latifolia the leaves are broadly ovate below and ovate-oblong above, while in media they are ovate-oblong below and narrow-lanceolate above, grading evenly into bracts : the lip of the flowers in latifolia is broader than long, with recurved tip and smooth lip-bosses ; in media it is longer than broad, the tip not recurved, and the lip-bosses plicate-rugose : in latifolia the lowest leaf-sheaths are appressed to the stem ; in media they are funnel-shaped. These are the chief differences, and if they really were constant, there would be no doubt of the existence of two distinct forms or species. But Dr. Bruce (B. E. C. Rep. 1913, ]). 337) long since pointed out that no reliance is to be placed in any of the characters thus distinguished, although in his edition of Havward's Botanist's Pocket-book, 1914, media still appears, with Linton's E. atroviridis as form (b), and Druce's platyphylla as form (c). No doubt Br. Druce would now delete media altogether.

In some years we found E. latifolia (media) very plentiful in W. Cardiganshire ; in fact, its abundance was simply in the inverse proportion to the amount of wood-cutting and roadside-cleaning that went on. We examined many scores of plants, and found that in the configuration of the leaves and flower-lips every possible cross- combination occurred. We found no viridiflora, and concluded that all the plants belonged to one very variable species. Some of these variations it may be worth while to describe.

One plant with a stout leafy stem of 27 dm. had broad ovate leaves below, very strongly veined, growing longer and more lanceolate above, and ending in an abruptly narrowed, bract-like leaf, 2-o cm. below the dense spike ; petals and sepals very broad, lips much broader than long, with recurved tip, and two pea-like side-bosses scarcely rugose at all. This suits fairly well the description of latifolia. Close to it was a taller plant of 6 dm., with similar flowers but leaves regularly graded into the bracts and nearly twice as long, the lowest funnel-shaped, the next ovate-lanceolate, then narrow-lanceo- late, grading into bracts. The lowest bract was actually larger than the topmost leaf, being 9 cm. long. Here is a media type of leaf, with a latifolia type of flower.

Another specimen wholly agreed with the description of media.

JoUHNAL OF BOTA>Y. VOL. 59. [FeBBLAKT, 1921.] 1)

31 THE .lul'KNAL OF BOTANY

All the leaves were lanceolate and evenly graded, the lip pointed- triangular with plicate-rugose side-bosses ; near it was a specimen with long leaves, very long-triangular lip, and nearly smooth side-bosses.

Many specimens agree with Linton's atroviridis : they have the narrower, graded leaf-type, and strongly-marked centre-Losses, as well as the side-bosses. These centre-bosses may be broad or narrow, short or long, scarcely exceeding the side-bosses or coming down the whole length of the lip, very distinct from the side-bosses or merged into them, and combined with either Jhe long or the broad lip-type.

Some forms agree with Druce's platyphylla. In one case the lower leaves were nearly circular, respectively 8 and 10 cm. in diameter ; they gradually merge into narrow, lanceolate leaves, approaching bracts in appearance. The flower has a roughly equilateral lip, with fairly large plicate-rugose lip-bosses and a central ridge. This lip is of a dull pink colour, the leaves very dark green, thick and glossy. Another specimen has the lowest leaf broader than long(5'o x5 cm.); in this case the leaves become ovate-lanceolate, merging into huge lower bracts (8-5 x 3 cm., 8-5 x 2-5 cm., and 75 x 2 cm.). As in the previous case, there is a very small central boss.

Quite frequently the latifolia type of leaf is combined with the media type of flower. There are before us four specimens with per- fectly ovate leaves throughout, with no graduation into bracts, except a single bract-like leaf in one case. In one of these plants a broadly ovate leaf 6 cm. long starts only T2 cm. from the base of the spike and reaches half-way up it. In another a similarly ovate leaf starts 8 cm. below the base of the spike. These all have flowers of a generally media type.

We do not find many cases of lip-bosses which Ave should call quite smooth, but in one of these the leaves are very narrowly lanceo- late, grading perfectly into narrow bracts ; in many cases the lip- bosses are almost smooth, and the leaves of media type.

In this connection we are rather at a loss to know jvjst what botanists mean by "smooth." "We rather think that the epithet is often applied to cases in which the bosses are not markedly knotted and wrinkled, but only a little uneven. We can hardly say that we hive seen any plants of any species of Epipactis whose lips Avere devoid of all wrinkles or furrows or convex mouldings of some sort; although some are much less moulded and rugose than others.

Besides single specimens from various localities, we have seen a fair nunber of plants from Grassington, almost all of which we should set down by their leaf-type to latifolia. The lip-bosses varied from nearly smooth to very prominent and lumpy : the centre-boss was absent in some, in others very small ; in one it was very distinct, in another there was a strong central ridge. An excellent note on the Grassington forms is given bv Dr. F. A. Lees in The Naturalist for March 1910 (p. 130).'

We have seen a large series of plants from Bath, sent by Mr. Bradley, none of which had very narrow or very evenly-graded leaves: in many the lowest leaf was orbicular, with a minute apex; those above were mostly lanceolate, of a somewhat intermediate type. Some flowers had very small centre-bosses or ridges. None had a

EPIl'ACTIS LATIFOMA LN JHUTAIX

31

lar<re or rugose centre-boss ; the side-bosses were uf medium size and rugosity, a few almost smooth.

It is thus evident that the amount of variation, when all the forms are reviewed, is very great. The variation of the leaves, both in size and shape, is considerable ; in one example the lowest leaf is ovate-lanceolate, 16 cm. long by 7 cm. wide, rapidly narrowing into narrow-lanceolate. In many, as has been said, the lowest leaf is orbicular. Some taper very abruptly, some very evenly. In some cases the leaves are very long in proportion to the stem, in others very short. They may graduate quite evenly into bracts, or the topmost leaf may be very large, and partly or wholly overtop the spike. In

Epipactis latifoUa, lip and flower types.

A. Broad. D. Longitudinal section through centre of lip.

B. Intermediate.

C. Narrow. E. Longitudinal section through centre of lip.

this case it may be ovate or long-acuminate. The lower bracts vary enormously. One has been noted 8-5 by 3 cm. They mav be very small, about a third longer than the flower. The topmost bracts are always very small, as far as we have seen. The colour of the leaves varies a good deal, being sometimes a deep, glossy green, and passing through various dull shades to yellowish green. The texture also varies, some leaves being tough and strongly ribbed, others much finer. Occasionally we have observed a variegated leaf, the green being broken by oblong yellowish patches, whose shape is determined by the venation.

The lips maybe roughly divided into three classes: (a) much broader than long, with tip recurved, sepals and petals broad ; (b) long- triangular, the point of the lip not (or not much) recurved, sepals and petals narrower; and (c) various intermediates. In the text- Hgure these types are indicated. Figs. A and D represent the broad

36 THE JOURNAL OF BOTANY

type; the details of the lip are shown in PL 555. 10; in this case there is a large centre-boss. Figs. Cand E represent the narrow type of lip ; PL 555. 11, gives the details of the same lip ; here the bosses merge into one another in a long-triangular moulding. Fig. B gives an intermediate type. Of this lip the details are shown in PL 555. 22 ; here there is a confluent triangular moulding of the lip, with a minute central boss besides. In all three cases the leaves were inclining to a narrow type ; but similar lips may be found with all types of leaf.

The character of the lip-bosses varies independently of the shape of the lip ; in all cases there are two bosses, and very often three. Sometimes they stand out from the surface of the lip abruptly, but more often the lip is moulded into a fairly smooth triangular swelling, out of which the bosses arise : in that case the central boss is some- times a mere vertical line or ridge upon the top of this elevation. Where there ax-e only two bosses there may be a deep depression between them ; they may stand out very little, with but slight folds, or may be very prominent and much knotted. The centre-boss may be a mere pin's head, or it may be as large or larger than the side- bosses, or it may be long and narrow, sometimes coming down to the point of the lip. And any of these characters of lip and boss may be combined with any of the characters of the leaves and bracts which have already been described. We have therefore every justification for coming to the conclusion that the distinctions set up in British floras between E. latifolia and E. media break down completely. All the forms must be assigned to the species latifolia, seeing that, as will be easily shown later on, E. media must be given up in any case. Purchas (Journ. Bot. 1885, p. 201) argues against E. media being counted as a British species, as also does Frevn (B. E. C. Pep. 1897).

At this point a few words may be written in reference to E. atro- viridis, described as a new species by W. R. Linton in his Flora of Derbyshire, p. 270 and plate (1903). The leaves are graduated on the stem, and neither very broad nor very long. The lip has a distinct centre-boss, but a recurved tip, and is broader than long ; apparently the character of the centre-boss more than anything else led to Linton's giving specific rank to the form. The Rev. E. F. Linton kindly sent a specimen which he thought was most probably E. atroviridis, but could not absolutely certify. The lip of this is given in Journ. Bot. 1920, PL 555. rig. 18. It has a long but not prominent centre -boss. The leaves are ovate-acuminate, of an inter- mediate type. In Cardiganshire a great many plants are to be found with well-marked centre-bosses and of many types of leaf, which might be brought in here ; but in view of what has been said about the intermixing of all these characters, it is vain to attempt to main- tain the distinctness of this species. Mr. Arthur Bennett in Journ. Bot. 1904, p. 24, questions the specific value of the form, and we entirely agree with his conclusions.

Dr. Druce has described a var. angustifolia (of latifolia) and var. 'platyphylla (of media). To what extent these forms represent constant variations we have -no means of deciding. We have found

E1MPACTIS LATIFOLIA IN URITAIX 37

broad and narrow types here and there, mixed with other forms. In B. E. C. Kep. iv. 503 (1916) the narrow-leaved form is reported as present in great quantity, with no other types present. We have not seen anything of the sort in localities known to us.

We may now study in more detail the specimens of lip-bosses figured in pi. 555. First, taking the whole aggregate of species, it will be noted that E. viridiforu presents the smoothest type of lip, otherwise there are considerable differences between the three types shown. Var. dunensis in fig. 1 has a broad type of lip, with three rather large, confluent bosses, whilst var. vectensis (tig. 2 i has a long- pointed lip and no centre-boss, and var. leptochila (fig. 3) with the same long-pointed form has a very different moulding, and a pair of ears, a feature paralleled in the case of E. latifolia in fig. 16. E. purpurata, tigs. 4 and 7, shows a somewhat heart-shaped lip, with more rugose side-bosses, ami a small central boss; the colour-scheme of this species is quite distinct (see op. cit. 11)20, p. 209).

E. atropurpurea (tig. 8) has by far the most rugose side-bosses ; we have never seen a specimen in which this was not so. In the lip figured, there is a strongly rugose centre-boss; in some cases, how- ever, this may be absent. The shape of the lip, in the example figured, is almost semicircular, and this is from a fairly representative plant. Fig. 5 shows a lip with highly rugose side-bosses, prominent centre-boss, and a pointed lip with sides incurved : this is from a Grassington plant which we take to be a hybrid of E. atropurpurea and E. latifolia. It is certainly not pure atropurpurea, and in that case, its intermediate character is verv evident from the drawing ; we have never seen another example of a thoroughly incurved lip.

Figs. 8-23 and 25 give examples of E. latifolia {media') : of these we may assign figs. 8, 9, 10, 13, 14, 17, 18, and 25 to the "broad" type, figs. 11, 12, and 16 to the "narrow" type, and tigs. 15, 19, 20, 21, 22, and 23, to the " intermediate " type, which has already been referred to. Of these the corresponding leaf-schemes were in part already indicated in our former paper.

In the "broad" set, tigs. 9 and 25 have broad leaves, fig. 14 narrow ones, and figs. 10, 13, and 18 leaves of intermediate type. The leaves of fig. 17 were not noted.

In the "narrow" set, figs. 11 and 12 had leaves of intermediate type, and fig. 16 very broad leaves. In the " intermediate " set, figs. 19 and 20 had broad leaves, whilst of the intermediates fig. 15 had narrow leaves, and fig. 23 leaves of broad type. Thus we have every sort of mixture of the three types of leaf and lip.

The varieties of the lip-bosses and moulding are very complicated. In figs. 10 and 13 there is a large, shallow centre-boss, in 8 and 21 a small, rugose centre-boss, in 12 and 16 the three bosses are confiuent, 16 being strongly rugose. In figs. 9, 17, and 22 there is a minute but strongly-marked centre-boss. In figs. 18, 19, and 20 there is a central ridge rather than a boss. In several other cases there is something of the sort. In figs. 15 and 23 there is rather a central trough than a ridge (cf. also fig. 2).

The general moulding of the lip has also to be considered. Thus in figs. 17 and 22 there is not only the distinct, minute centre-boss,

;{S THK JOURNAL OF BOTANY

but also a raised moulding in the centre, confluent with the side- bosses, giving a generally triangular effect. This is also seen in ficrS. H, 12, 15, and 16. In figs. 11, 15, and 22 we have what we suppose would be generally called smooth side-bosses ; figs. 8, 12, and 21 show the most rugose types.

It will be seen that the smoothest types are as smooth as E. viri- difiora, and the most rugose as rough as E. atropurpurea, though in the second case very differently wrinkled. In figs. 23 and 25 the side-bosses are only a little rugose, whilst in figs. 10, 13, and 21 they are almost pea-like. In fig. 14 they are the least elevated and figs. 13, 17, and 21 the most elevated.' In fig. 16 {cf. fig. 3) the lip runs up into ears.

By way of illustrating the distribution of the many types it may be noted that figs. 12, 15, and 17 were growing close together in a wood near Aberystwyth, and with them plants of still other types- one with a very large centre-boss, another with a very small one, and one with a narrow but rugose ridge. There were many differences in the leaf-type.

The colour of the flowers varies considerably. The sepals are usually of a dull green, giving to the plant a dingy appearance. From this they pass to various shades of dull, or sometimes bright red-purple. The figures given by Camus in his monograph of the Orchids of Europe (1908) have the flowers bright red, both for E. viridifiora and E. hi t (folia ; in Britain such bright colours seldom occur. The lip is of many shades of red-purple or lilac, sometimes white or greenish white : Mr. A. I). Webster speaks of a fine white variety of E. latifolia, found in North Wales, of which we have not seen specimens. The amount of light does not seem to make much difference to the tint; we have seen under the same bush, in flower at the same time, a pure green spike next to a bright purple one. We have found purple flowers in deep shade and dull green ones in profusion by the road-side, in quite sunny positions. At Southport the pale green viridiflora grows right out in the open, receiving every hour of sunshine.

We trust we have given enough evidence that the distinctions of E. latrfolia and E. media, as mostly given in British floras, entirely break down. All these forms, as also atroviridis, must without doubt be assigned to E. latifolia.

We have now to summarize the work recently done on the nomen- clature of the group, by which it has been conclusively shown that, whatever he the relations of the British forms, E. media, both of Fries and Babington, must be given up.

It would appear that Fries founded " media" on the characters of plicate-crrnate lip-bosses and tapering (not re volute) apex of the lip, whilst Babington emphasized more the narrower, evenly graduated leaves. We have seen that no such differences hold good for British forms.

Fiies {Mantissa altera, 1839, 55) gives three forms of his species media: "(ff) floribus albis. Serapias microphylla Bot. Dan. non Khrh. Serapias latifolia y aloe/is Wahl. Suec. p. 589. (b) floribus viridibus. Ser. riridi'/ora Reich, le. f. 1112 sec. Koch, (c) flori-

EPTPACTIS LATIFOLIA IX BRITAIN •U)

bus roseo-rubris .... E. atrgrubens Reich. Ic. f. 1111." Col. Godfery in this Journal for 1919 (p. 80), suggests that Fries's (b) viridiflora is really the green form of his (c), which is atro- rubens {atropurpurea), about whose identity there is no doubt. Fries rests on Koch for viridiflora, and Koch (he thinks) did not distinguish the green -flowered atropurpurea from viridiflora proper, and considered that they both belonged to latifolia. However, if we compare the plates of Reichenbach, whilst 1141 is undoubtedly atropurpurea, 1112 shows a plant with smooth ovaries and less rugged lip-bosses. This may be the same as Reich. Ic. Fl. Germ. 4b~, which is probably viridifiora proper.

As to {a) microplnjlla, Mr. Rolfe {Orchid Review, xxvii. pp. 78 f) thinks that the note " not of Ehrh." is an error, and that this form is E. microphylla Ehrh., figured in Fl. Dan. t. Nil, as Helleborine latifolia montana. It is not E. miorophylla Sw., as figured by Schulze, Reich. Ic. Fl. Germ. 484, &c: the figure in Fl. Dan. 811 is of a slender plant with broad leaves and a pointed lip, growing in highland woods. This does not look like any form of atropurpurea. Thus Fries's " media " proves to be a complex of two or three distinct species. In this connection we must plead guilty to a rather loose use of the phrase no men nudum in our previous paper (p. 210 i.

Turning now to Rabington's identification of Fries's media, it may be said that if, as Col. Godfery thinks, Fries's {b) is the green variety of atropurpurea, Rabington was wrong in his use of the name; for these plants really belong to his oralis. Rut if Fries's (b) was true viridiflora, then Babington was right in his identifica- tion, as we shall see immediately, but led us astray by giving a description which did not really apply to the forms. He was simply handing on the confused description of Fries.

The whole situation, as far as the Rritish forms are concerned, has been thoroughly cleared up by the skill and patience of Col. God- fery, in the article in this Journal referred to above. He shows that the plant named by Rabington as E. media was first diagnosed by him as E. viridiflora : specimens recently secured from his original Rritish locality turn out to be true viridiflora. Further confusion arose from the fact that the drawing of E. media in E. R. S. 2775, was drawn from a specimen of E. purpurata. but coloured like E. lati- folia. It is no wonder that viridiflora was submerged, and Col. God- fery is to be congratulated on his elucidation of the problem thus created. Now that we are able to apply the criterion of the different form of the reproductive organs, viridiflora can be diagnosed with certainty, that is, in the living state, and many forms hitherto assigned to media must be given to that species, and the rest to latifolia. Of course, when viridiflora has been identified in any district, the possibility of hybrids with other species cannot be quite excluded, in spite of the fact that the former is self -fertilizing ; but we have not hitherto heard of any such hybrids. It is certain that there are none amongst the very various forms the description of which has been the chief object of the present paper.

-Ill T!IK JOntXAL OF RD'I'.VW

THE LICHEN AS TRANSMIGRANT. Bv A. H. Church.

(Concluded from p. 13.)

On tracing back to the sea it begins to be clear that phenomena of intrusion at an improtected surface are the commonplace of marine biology, otherwise epiphytism is equally prevalent ; since in the abundant life of the sea, as in a tropical rain-forest, everything grows on top of or inside anything else available, in the increasing struggle for some sort of substratum. But the possibilities of penetration are limited by the constitution of the limiting membranes of the organism. This is again but a part of the wider problem which gives so many animal races an external membrane of chitin, becoming an impene- trable exoskeleton, while an internal permeable surface is retained as the alimentary chamber or canal for purposes of food-absorption. Though, in the case of plant-forms, the earliest soft mucilaginous membranes of polysaccharide waste may be readily penetrable (as in the case of intrusive green algal cells in Schizonema Diatoms), the more resistant polysaccharides of higher plants restrict further pene- tration, and thus acquire a ' protective ' value. Hence, since the provision of a firmer and more coherent and resistant polysaccharide, including a greater percentage of celluloses, follows the evolution of the higher alga in rougher strata of surface-water, so the possibilities of the intrusion of lower algse become less and less, until its excep- tional occurrence occasions remark ; just as land-plants under natural conditions are practically immune to bacterial attack, though many become susceptible to mass-infection under cultivation, and so appear as a novelty. Further it becomes evident that intrusion will be less and less a general phenomenon as cell-walls become more efficiently protective (though always open to special cases, as in the attack of definitely parasitic epiphytoid races and specialized endophytes), and will be less and less practicable in higher grades of marine vegetation ; again just as higher animals are attacked most readily from the unguarded alimentary canal or lungs. From this it follows that successful utilization of such a means of existence will not occur in the normally autotrophic races of the sea. That is to say, the Lichen-habit was never evolved in the open sea; but must be the product of subaerial and transmigrant conditions. That a few minor forms (cf. Li china) can endure partial submersion does not affect the argument; so can many Angiosperms. Zostera and Position in, for example, live permanently submerged, and in deep water ; but no one supposes that they were evolved there originally. The evolution of the Lichen in encouraging and utilizing phenomena of algal intru- sion, en mouse, and not by mere chance ' infection,1 is thus based on the possibility of penetration at the surface ; and this will become the more possible as the older plant-surface, originally presenting a frontage of photosynthetic units in closest contact, or in palisade- scries, begins to lose its essential function, and with it the special attributes given by active growth and division. A deteriorated cortical system prepares the way for successful intrusion. It remains

THE LICHEN AS TRANSMIGRANT 41

to consider the possibility of the deterioration of the photosynthetic surface-layers of a seaweed-soma of comparatively high grade, equivalent, that is to say, to a type of plant with a reproductive organization as efficient as that of the modern Floridea?.

Only a very slight acquaintance with the somatic mechanism of seaweeds is sufficient to indicate the manner in which this may be done. For example, in an ordinary Fucoid the photosynthetic units, which carry out the constructive metabolism of the plant in imme- diate association with the external food-solution, constitute but a relatively thin brown film, about 100 /j. deep, over the more massive mechanical tissues of the interior ; the peripheral more active units being 10-15 p only. Stripped of these outer layers, the whole plant reduces to a system of hyphal strands, as an interwoven mechanical tissue of ' descending by phse,' to all intents and purposes the mycelium of a Fungus-axis. The same effect is more readily produced in a Laminarian (cf. L.jlexicuulis), if kept in standing water for a day or two ; though in such a type the central tissues are less hyphal, and may even present a massive growth of mechanical tissue with annual increments as rinsr-effects (L. Cloustoni). A little consideration shows what happens. The actively metabolic surface-layers, accus- tomed to the free interchange of gases and food-salts with the external medium, are more sensitive to changes in the environment as they are the more superficial and the more actively synthetic. In standing water, soon deprived of all free oxygen, they die immediately in the dark owing to lack of oxygen for aerobic respiration ; and in the light they also soon fail for lack of further supply of the essential food-ions. On the other hand, the more internal tissues, previously wholly heterotrophic at the expense of the surface-laj'ers, and existing with a minimum oxygen-supply only available by diffusion from the periphery, are on a wholly different footing. Conditions that involve rapid death to the autotrophic surface-cells do not necessarily injure the internal mechanical and reserve tissues. This roughly visualizes the origin of the heterotrophic ' Fungus ' from residual vegetation of the sea, banked in standing wrater, or shaded under decaying masses of vegetation. A ' Fungus ' is to be regarded, not as a new land- growth, evolved de novo by the progressive elaboration of a mere weft of filamentous mycelium [however much some filamentous alga3 may have similarly given rise to filamentous phyla of Fungi (cf. Mucorini)~\ ; but a higher Fungus of the land is in short a ' skinned seawreed,' implying a more or less elaborated algal grow:tb-form, in which, on the death and decay of the older metabolic and autotrophic surface-layers, the exposed internal heterotrophic tissues continue their heterotrophic (and even conceivably anaerobiotic) existence at the expense of the soluble carbohydrates of the standing and non- aerated medium. The normal progression of heterotrophy in the transmigrant fungus-phyla, following the natural consequences of a residual marine environment of plant-rejectamenta, thus sufficiently accounts for the loss of the cortical and photosynthetic tissues, as also for the destruction of the protective ' cuticle ' and tissues pre- viously resisting intrusion. The two things go together, and need no longer excite surprise. The massive organization of a Fungus, with

|-_! THE JOURNAL OF BOTANY

morphological differentiation of centric or dorsiventral axes, and ramuli of different degree, is but the vestigia of an algal soma, adapted to the normal factors of marine biology, and following its seaweed phytogeny in the recapitulatory production of reproductive organs as ' trichome-derivatives,' whether in the form of a confluent ' hymenium' {Laminaria ; apothecia), or even as immersed and peritheciuni-like conceptacles (Jfuctts and Floridean cystocarps). Incidentally it now becomes clear why higher Fungi repeat the com- plex life-cycle and phases of sexual and asexual reproduction charac- teristic of the higher algal types of the modern sea (cf. Florideas): There can be no doubt that the heterotrophic Fungus-soma continues the heterotrophic region of the older algal somata, without being in any way a production de novo, designed to meet the special conditions of a saprophytic existence ; just as in an older Plankton-phase the heterotrophic ' animal ' arose by continuing the holozoic method of nutrition common to all naked flagellate life, but merely eliminated the autotrophic chloroplasts, or only their autotrophic function, since vestiges were retained as ' eves ' sensitive to the same octave of solar radiation.

Beyond such a stage of continued rather than incipient hetero- trophy, the Lichen covers a special biological case, also readily in- telligible to those who have seen the films and scums of low-grade green flagellate and coccoid forms which accumulate in any vessel of standing sea-water exposed to the light. While the main series of higher Fumycetes are clearly derivative from stripped seaweeds which emerged from the water to continue existence at the expense of stored carbohydrate of decaying land- vegetation, the Lichen repre- sents the case of similar simple or branched algal somata, remaining denuded of autotrophic tissue in standing pools, and hence soon smothered with a growth of green autotrophic flagellates, now ready to take advantage of the penetrable tissues of the enfeebled hosts, in their demand for a benthic substratum. The heterotrophic hosts, enclosed in a new mantle of photosynthetic units, merely require to continue their heterotrophic metabolism at the expense of the Avaste carbohydrate and photosynthetic oxygen, in such a medium onlv available in the vicinity of these autotrophic units. The synthesis is thus effected in terms of (1) a substratum for the flagellate on attaining the benthic habit, and (2) the only possible means of con- tinued life in the light, or even in the dark, for the more massive Fungus-' host.' The former point is of interest, since in the con- sideration of any possible advantage to the algal constituent of the Lichen, shelter is commonly postulated, or protection from desiccation ; the older vital problem of the struggle for substratum being naturally lost sight of in the consideration of subaerial migrants. It is in- teresting to note, again, that such a hypothesis precludes the possi- bility of the recovery of a cortical system by the host, which might be expected to follow in such a case as that of Funis, in which an apical cell, though itself largely heterotrophic, might conceivablv survive to regenerate a new somatic region by an apical meristem. lint apical meristems are only presented in special types among alga'; tlic primary ease being a condition of general intercalary growth and

THE LICHEN AS TRANSMIGRANT 43.

extension throughout the whole mass, or over localized areas [of. Enferomorpha (Chlorophyceae), Asperococcus (Phaeophyceae) ; DiJsea, Rhodymenia, Giyartina (Floridese)]. From this it may he said to follow that no Lichen is likely to have an apical meristem in the sense of a segmenting apical cell; and as a matter of fact the Lichen habit is largely confined to algal types of the lowest grade, in which intercalary growth and ramification of the most generalized character still obtain (cf. Mesogloia and Selminthora-tjjte with Usnea and ClaJonia). The more elaborate somata of the Lichen - series attain a phase with apical growth in the ultimate ramuli of the type of such Florideae as Puli/ides and Fur cellar ia, Chondrus and Giqartinq, in which the general presentation of Lichen-form is sufficiently striking in the living condition, to be even more empha- sized when subjected to desiccation on drying in the manner of modern iubaerial lichens.

Such a working-hypothesis covers much of the fundamental organization of a Lichen-fungus. One now begins to obtain a clear idea, not onlv of the locus of its transition to the heterotrophic condition, in common with associated saprophytic Ascomycete phyla, but of the mechanism of the original failure of the photosynthetic tissue, primarily due to the lack of oxygen for respiration over the night-period in stand- ing pools, as a phase of the same problem which was ultimately responsible for driving fish out of the water to become amphibians on the Land. From the standing pool above the tide-range, to existence wholly out of the water in damp air, or with the casual supply of the ' region of the splash ' and possibilities of atmospheric precipita- tions of 'freshwater.' is but a small departure. Familiar observation and recorded data as to the existence of even massive algal growths out of the sea, within the region of the splash, afford a general view of such translation from the water to air ; in which, again, further adaptations for increasingly drier conditions, to be ultimately ex- pressed by perennation in a state of even practically complete desicca- tion, follows a natural sequence of xerophytic progression. The existence of seaweeds out of the sea, exposed to damp air, is suffi- ciently familiar in the ordinary phenomena of the tide-range and the Fuci of the Salt-marsh ; the limiting factor being not so much desiccation in the case of subsaturated air, as the ultimate starvation implied in restricted capacity for proteid-synthesis on removal from the food-solution. Photosynthesis of carbohydrate may be increased, and so prove excessive, since this alone is useless for continued growth ; but a relatively very small amount of photosynthetic tissue, associated with sufficient chances of obtaining the necessary food-salts, may solve the problem, not only in subsaturated air, but also under drier conditions. Once it is clear that material for proteid-synthesis is of even greater consequence than water-supply, the way is prepared for xerophytic adaptation to economize the latter ; but no xeromorphic adaptations are of avail in absence of food-salts. Hence once removed from the sea-water, even the Lichen-association can be but a starved production, and cannot lead to any very great development of the soma. Desiccation may be faced, but not entire lack of food-salts, whether owing to fresh-water environment, or failure to establish

41 THE JOURNAL OF BOTANY

adequate absorption from the substratum. In this way the evolu- tionary history of the Lichen-association which suggests its possi- bilities, also equally defines its limitations, and at the same time covers the range of progression for all forms following a similar sequence of biological factors. Summarizing the stages briefly, it may be said that : —

(1) Failure of oxygen-supply in standing pools of sea-water was the most readily conceivable cause of the death of the autotrophic surface-tissues of the original sea-weed, thus leaving the thallus penetrable.

(2) Competition for substratum gave the mantle of green intru- sive ' gonidia ' of the type of Chlorella.

(3) The oxygen -problem may be also responsible for the greater success of the associated organisms on leaving the water for sub- saturated air ; though the cause of the plants being so exposed may be merely the expression of the pools drying up.

(4) The nitrogen-problem still keeps the plants impoverished; as

(5) The water-problem tends to keep them small and restricted to short seasonal periods, thus further delaying the rate of growth. The last again common to all land-vegetation of whatever grade, and theoretically a factor of less importance than is generally admitted in considerations of the progression of Land-Flora.

Ability to withstand the extreme desiccation of wind and hot sun to a condition of brittleness, often popularly regarded as distinctive of the vitality of Lichens, not only from the standpoint of the Fungus-constituent, as from that of the included 'protected' Algye, is by no means exceptional, or confined to members of this group. Even high-grade Basidionrycetes of the Polyporus and Agaricus model will similarly endure desiccation in the soma as a whole, as a part of their normal biological equipment ; rapidly recovering and even renewing the delicate mechanism of spore-discharge on being wetted. Thus Buller (1909) l describes Polysticttts hirsutus and Polyporus rigens renewing somatic growth afjter being kept air- dry for a year. Marasmius oreades was capable of recovery after 6 weeks, and renewed full activity of spore-discharge in a few hours after being wetted. Other species of Mierulius, Polystictus, Lenzites, and Dadalea, kept dry for 2 years and 6 months, only required about 4 hours' wetting to recover.3 For marine alga? Borgesen (1908) 2 gives Hildenbrandtia (Floridean and encrusting disc-type) grading into lichen-mixtures at 200 ft. above high-water mark on exposed Atlantic coasts. Enteromorpha intestinalis flourishes as a transmigrant in fresh-water streams, attaining a length of 6 ft. in running water; yet on exposed cliffs it covers large areas of rock-face bright green where sea-birds congregate (Rockall),3 and will endure desiccation until it may be powdered. Enteromorpha and Prasiola (Chloro- phycese) live abundantly exposed to fog, rain, and spray as the only source of water and food-salts, 40 ft. above the sea at Faeroe ; and

1 Buller (1909), Researches on Fungi, p. 106.

2 Lor. cit., p. Ill, fcr list of times and species.

3 Borgesen (1908), Botany of the Faeroes, p. 770.

THE LICHEN AS TKA>'SMIGRA>~T 4-3

Porphyra umbihcalis (Floridean) at 30 ft. dries so that it crackles when walked over. The desiccation of Pelvetia (Phseophycean), almost to hrittleness, is sufficiently familiar on sheltered British coasts. Hence endurance of extreme desiccation on a rocky sea-cliff may be presumed to follow normal processes of natural selection in any phylum of Algae or Fungi. The Lichen-soma is by no means unique in this respect, though it may be more familiar to the collector. Endurance of complete air-dry desiccation may be attained readily in the soma as a whole ; just as similar endurance of desiccation becomes the normal condition of the air-borne spore in every group of transmigrant plant-organism.

Such observations again suffice to indicate the possibilities of nutrition by drifted sea-salt which still characterize lichen-growths facing the sea ; and no one who has seen the shaggy covering of Lichen on the cliffs facing the Atlantic, can hesitate to believe that this is the primary transmigrant habit of the Lichen-series as a whole. They constituted one of the pioneer races of the older world, on the first rock-surfaces exposed above the retreating sea ; and they have held their station to the present time, since no other plant-organism can compete with them in endurance on such a feeble food-supply.

It will be noted that from the present standpoint, the precise nature of the intrusive algae becomes a matter of relatively very sub- sidiary importance, so long, that is to say, as it may be of discrete or Protococcoid habit ; while, in the case of transmigrant races, a change of helot intruders becomes not only conceivable but probable. Once the lichen-habit is established and becomes obligatory for these special fungus-migrants, it is possible that any readily available algal units may be utilized for the purpose, with little consequent effect on the morphology of the dominant fungus. These generalizations acquire additional interest in view of recent observations by Paulson (J. L. S. (1920), ]). 501 1 on the utilization of forms of Ohlorella in many common Lichens, which have long passed as symbiotic with Pleuro- coccus (Cystococcus), and what may be even in all cases the algal component of Physcia parietina, as described in the classical synthesis by Bonnier. The suggested evolution of a ' primitive ' Lichen in fresh- water ponds of the present day,1 or the synthesis of Physcia on the green dust of a tree-trunk in one's back garden, may serve to visualize some points of the process, but affords no adequate appre- ciation of the facts of the evolution of this remarkable race as a whole, as probably the oldest surviving phase of land-vegetation, — and also, in the more alga-like somata of fruticose forms, possibly the least changed of any subaerial types from a period beyond the range of stratigraphical geology. It is considerations as these which lend interest to what is otherwise one of the least considered groups in the range of botany. As the Brown Seaweeds, admitted often grudgingly and as a duty by old collectors to the repose of the Seaweed Album, afford the clue to the evolution of the parenchymatous land-plant, so the Lichen undoubtedly affords the best key to the development of all Fungi ; since while the latter pass on to the land and fresh-water.

1 Astern (1909), Annals of Botany, xxiii. p. 579. ' Botrydina, a primitive Lichen.'

1,6 THE .TOruXAL 01 BOTANY

the Lichens are nearest the sea from which they emerged, and still retain a more complete expression of the original algal soma, as well as indications of an older phase of sexual reproduction and the double life-cycle.

While the preceding generalizations may suffice to indicate the mechanism of the progression, so far as it concerns the somatic organization of the individual, through successive phases of reduced and heterotrophic saprophytism, to recovered autotrophy by the vicarious agency of intrusive green flagellates of the type of Zoo- chlorella, — -it remains to follow the simultaneous progression of the racial mechanism of reproduction, as presented in the process of fertilization and the provision of means of dispersal, as also the adjustment of the meiotic problem in the life-cycle, which may throw additional light on other phases of the problem, as well as confirming the general hypothesis. It should now be sufficiently evident that the phenomenon was eminently polyphyletic ; and many seaweed phyla, as the mixed flora of the littoral zone, residual in standing ponds, may have successfully solved the same problem. The fact remains that one Fungus-series, the Ascomycetes, has survived to show it better than any other; although the point is not invalidated that the latter systematic group probably constitutes a wide polyphyletic range of types, and one wholly empirical, since of unknown algal origin and now abruptly isolated.

Botanic Garden, Oxford. November 21, 1920.

SOME PLANTS FROM JEBEL MARK A, DARFUR. By L. V. Lester Garland, M.A., F.L.S.

In the early part of last year Captain H. Lynes, R.N., and a companion visited the province of Darfur in order to study the Natural History and collect specimens. Their stay was unfortunately cut short just as they reached the upper zone of the mountainous district which has very seldom been visited by Europeans and is sure to possess features of considerable interest. Captain Lynes started again for Darfur in November, and means to spend a year there. He will return, no doubt, with a large amount of detailed informa- tion ; but it is perhaps worth while to record at once the plants which he brought back from the high ground, and which he has been kind enough to present to the British Museum of Natural History. For the information contained in the following paragraphs I am entirely indebted to Captain Lynes.

Darfur is a vast plain about 20% larger than the combined British Isles, and with a mean altitude of about 2300 feet above sea- level. The highest part lies in the centre, whence it slopes gradually downwards and outwards, with local series of low, rolling hills. The volcanic massif of Jebel Marra, which forms the centre of the province, is in extent about fifty miles N. and S. by 20 miles E. and \\\. and its importance seems to have been much underrated up

SOME PLANTS 1'HOM JEBEL MARKA, DARFUB 47

to the present time. The greater part of it, according to Capt. Lynes's observations, has a mean altitude of 8000 feet, with peaks rising to 10,000 feet or more. No altitudes of anything like the same im- portance are found nearer than the Tibesti Mountains in the Sahara ((300 miles N.W.), the Abyssinian Plateau (700 miles E. by S.), the Great Equatorial Mountains (900 miles S.S.E.), and the Cameroons and Nigerian Mountains (900 miles W. by S.). Jebel Marra thus stands in a position of splendid isolation, and detailed information as to its fauna and flora cannot fail to be of very high interest.

The vegetation of the province as a whole " varies from pure desert in the North to rather rich park-land (Savannah) in the South." The greater part of the area is covered with " typical North Sudan bush and scrub." The low hills are either quite arid, or covered with the same vegetation as the plain. But Jebel Marra is " quite dif- ferent from anything else in the Province. These are well-marked vegetal zones, and the upper zone, from about 8000 feet upwards contains many temperate plants."

The following list includes only those species which were found in the mountainous district, but a considerable number of plants were also collected on the plains. A fortnight was spent at Kallokitting (ait. 4000 ft.), which Captain Lynes used as his base, and about 10 days on the mountain itself. The list, so far as it goes, indicates the presence of a composite flora in which North Temperate, Medi- terranean, Abyssinian, Sudanese, and widespread Tropical types are all represented : —

Adiantwm Gapillus-Veneris L. 7300 ft. (very widely spread). — Dn/opteris patens Desv. (Tropics). — Pteris longifolia L. 7200 ft. (all warm regions). P. aquilina L. (almost cosmopolitan).

Aloe commutata (Tod.) ? 7600 ft.

Ficus salicifolius Vahl. 6000 ft. (Abyssinia and Arabia). F. populifolia Vahl. 7200 ft. (Abyssinia and Arabia). F. pal mat a Forsk. 7100 ft. (Trop. Africa, Arabia, India).

Salix Safsaf Forsk. 72-30 ft. (Trop. and N. Africa).

Loranthus globuliferus A. Rich, (on Salix). 7250 ft. (Abys- sinia).

Polygonum serrulatum L. (widely spread).

Arenaria Schimperi Hochst. 9500 ft.

Ranunculus pinnatus Poir. 4000 ft. (Trop. Africa).

Acacia albida Del. 7300 ft. (N. Africa). — Indigofera arrecta Hochst. 7300 ft. (Abyssinia).— Desmodium Scalpe DC. 7300 ft, (Trop. Africa and India).

Monsonia senegalensis Guill. & Perr. (Trop. Africa).

Boswellia papyrifera A. Rich. 5000 ft. (Abyssinia).

Polygala ahyssinica K. Br. 9500 ft. (Trop. Africa and India).

Rhus villosa L. 60OO to 7000 ft. (Trop. and S. Africa).

Terminalia Browne/ Fres. 5200 ft. (Abyssinia). — Gombretum splendent Engl. Up to 6900 ft.

Epilobiuni hirmlum L. 7250 ft. (North Temperate).

Blceria spicata Hochst. 9300 ft. (Cameroons and Mts. of E. Africa).

48 THE JOURNAL OF BOTAXY

Mcesa lanceolata Porsk. 7250 ft. (Mts. of Trop. Africa).

Olea chrysophylla Lam.? 8000 ft. (Trop. Africa).

Cuscuta planifiora Ten. 9500 ft. (Mediterranean).

Mentha silrestris L. (Europe, N. and W. Asia, N. India, Temp. Africa). — Otosteqia scariosa Benth. 6000 to 7000 ft. (Abyssinia, Arabia).— Micromeria biflora Benth. 6000 to 7000 ft. (Trop. Africa, Arabia, India). — Lavandula coronopifolia Poir. 8000 ft. (Egypt, and the East).

Arnebia hispidissima Sieber. 9500 ft. (Trop. Africa, Arabia, N.W. India).

Witlmnia somnifera Dum. 8000 ft. (Tropics of Old World).

Linaria cegi/ptiaca B*um. forma. 9400 ft. (Egypt and Syria). — Veronica Anagallis L. (North Temperate).

Stereospermum Kunthianum Cham. 5200 ft. (Trop. Africa).

Hypoestes ForsJealei K. Br. (Abyssinia, Arabia).

Anthospermum pachyrrhizum Hiern. 8000 to 9400 ft. (Trop. Africa ).

Campanula Schimperi Vatke. 8000 to 9400 ft. (Abyssinia).

Lobelia sp., Sect. Hemipogon. 7300 ft.

Vernonia amygdalina Del. 7250 ft. (Trop. Africa). — Conyza

stricta Willd. forma. 8000 ft. (Trop. Africa to India). — Reli-

chrysum abyssinicum Sch. Bip. 8000 to 9000 ft. (Trop. Africa). —

G-eigeria alata Benth. Common (Ti'op. Africa). — Centaurea\sene-

galensis DC. (Trop. Africa).

HIERACIUM AMPLEXICAULE L. By J. Cosmo Melvill, D.Sc, E.L.S.

For a very long time this Hawkweed, of isolated position in the genus, has been known to occur on some ancient walls at Oxford, and, since the date of publication of the 3rd Edition of English Botany in 1866, where Carrie Barns, Clun, is given as Don's dis- covery, several localities in various counties have been named for it. Mr. S. T. Dunn, in his Alien Flora, mentions it as "thoroughly established in several places in old villages"; and in the last (9th ) edition of Babington's Flora Hawes and Cleish Castle, Kinross, are noted for its occurrence.

In July 1895, Mr. Henry Hyde, of Manchester, noticed a large growth of this plant on both sides of a bridge crossing theR. Mersey, between Stretford, Lancashire, and Sale, Cheshire, the two counties being divided by the river. Three years afterwards, on July 2nd, 1 898, 1 was conducted to the spot by Mr. Hyde, and found a very luxuriant growth, some specimens having even found a foothold on the masonry of the bridge itself; I at once recognized it as H. am- plexicaule, having only recently gathered it at Oxford, whence I sent specimens that year to the Botanical Exchange Club (see B. E. C. Report, i. 578, 1898). About ten years ago, I again visited the spot and noted that the plant was thriving and increasing exceedingly. From the parcel of plants collected in 1919, received from the Club, I was interested to find that Mr. B. S. Adamson, of Manchester

HIEEACIUM AMPUiXICAULE 49

University, had visited this locality and sent examples. These had heen queried as H. sciaphilum Ueehtr. var. amplifblium Ley, with which I do not consider our plant has any affinity.

While both Bos well (Syme) (E. Bot. ed. 3) and Babington (ed. ix) include our plant, Hooker in the Students Flora relegates it to obscurity, and Mr. F. N. Williams (Prodr. Fl. Brit.) and the Rev. W. R. Linton in The British Hieracia omit its mention altogether. There is therefore no consensus of opinion yet among botanists as to its proper status in our Flora.

To my mind it has considerable claim to fuller recognition. It has evidently been naturalized — especially at Oxford — for many years, and has come to stay, and, what is more, to appear from time to time in new localities. There being nothing particularly beautiful about it, save for exuberance of growth, I cannot think that anyone would pur- posely plant it, as others of the same genus are more attractive and brilliant, e.g. calendul/Jlorum or chrysantlu/m Backh., vlllosam L., or lanatum W. K.

I may add that it has long found a place, italicized, in the London Catalogue; Nyman (Conspect. Fl. Eur. p. 448) gives Spain, the Pyrenees, Jura, and the Apennines as its child' localities, making no mention of its appearance elsewhere adventitiously.

GLAMORG ANSI! I HE BRYOPH YTA.

Bv Elkanoha Akmitage.

During the botanical excursions in connection with the British Association Meeting at Cardiff last August, 1 was able to collect some Sphagna which have been kindly named for me by Mr. J. A. Wheldon, who tells me that only one Hog-moss {Sphagnum obesmm) had heen noticed in this county before. I was also able to add a few new records in mosses and hepatics which Mr. H. H. Knight has verified for me. Mr. Knight had worked at Bryophyta in Glamorgan previously, and has now handed me a list of his more recent records which have not appeared in the Census Catalogues, together with a few old ones from the Herbarium of the late Rev. A. Ley. In the following list the moss records are Mr. Knight's unless otherwise stated. The Sphagnum and hepatic records are the writer's ; vouchers of these gatherings for v.c. 41 have been deposited in the National Museum of Wales. The Sphagna were collected in two localities : on moorland at Mynydd-y-Glew at about 400 feet (M.), and on swampy moorland at Hirwain Common, 1450-1500 feet, below Craig-y-llyn (H.) — only two forms were common to both places. The other localities are detailed separately.

Sphagnum acutifolium var. versicolor f. venustum H. ; S. ph/mu- losum var. viride H.,var. versicolor f. teuellum M., var. ccerulescens M., and a form passing to ochraceum M. ; S. amblyphyllum var. mesophyllum f. sylvat/cum H. ; S. serratum var. serrulatum H. ; S. cuspidatum var. submersum f. crispatum H. ; S. mollusc// i/t var. vulgatwm f. co/npactum H. ; S. iuuiidatum var. ocalifoliuin Journal of Botany. — Vol. 59. [February, 1921.] e

50 THE JOURNAL OF BOTANY

f. subfalcatum M., f. laxifolium H., M. ; 8. auriculatum var. ovatum f. rufesceus M., subf. subsimplex M., var. variegatiim M., H. ; var. laxifolium M. ; 8. crassicladumw&Y. intermedium f. ovalifolium M. ; 8. cumbifo/ium var. glaucescens f. squarrulosum H., var. ^«Z- lescens M., var. Jlavescens M.

Oaiharinea crispa, Glyn Corwg, 1890, -£<?y'; Bolytrichum strictum, R. Perddyn, 1908 ; Rhabdoweisia fugax, Craig-y-llyn, Zcj/ ; Dicra- nella varia, Llantwit Major, 1908 ; Blindia acuta, R. Perddyn, 1909 ; Eissidens osmundoides, R. Perddyn, 1909; Rhacomitrium aciculare, R. Perddyn, 1909 ; i2. fascicular, Craig-y-llyn, 1920, E. A. ; iZ. canescens, sandhills, Pennard Castle, 1907, Merthyr Mawr "Warren, 1920, E. A. ; Tortula ambigua, Southerndown, 1908 ; T. intermedia, Porthcawl, 1908 ; Barbula Hornschucliia na , Porthcawl, 1908 ; Weisia rupestris, Resolven Waterfall, 1890, Ley, R. Perddyn, 1909 ; Zugodon Mougeotii, R. Perddyn, 1909; Bryum pendulum, Porth- cawl, 1909 ; B. fallens, Southerndown, 1908 ; Thuidium abietinum, sandhills, Porthcawl, 1908 ; Eurhynchium prcelongum, Cum Cathan, 1907; E. myurum, Gower, 1907; Hypnum clirysophullum, Southern- down, 1908.'

Alieularia scalaris var. procerior {fide W. Watson), Craig-y- llyn, 1920; Calypogeia Trichomanis and Scapania irrigua, Hirwain Common, 1920.

REVIEWS.

Henry Nicholas Ellacombe, Hon. Canon of Bristol, Vicar of Bit ton and Mural Dean : 1822-1916. A Memoir, edited by Arthur W. Hill. 8vo, cloth, pp. 318, 14 illustrations. 'Country Life' Offices, 20 Tavistock Street, London, W.C. : 1919. Price 10s. Qd. net.

" It has rarely been the fortune of one individual to exert on gardening in this country an influence so wide and beneficent as that wielded by the late Canon Ellacombe of Bitton " : thus begins the preface to the interesting and attractive volume which the Assistant l)irector of Kew Gardens has devoted to his memory.

The volume begins with a sketch of the Ellacombe family and an account of the Canon's father, Henry Thomas Ellacombe (1790- 1885), Avho preceded his son as Vicar of Bitton, near Bristol, and from whom the latter doubtless inherited his horticultural and botanical tastes. H. T. Ellacombe, as his letters preserved at Kew show, maintained an extensive correspondence with the leading botanists and horticulturists of his day : he is commemorated in Yucca Ellacombei, figured in Saunders's Refugium Botanicum (t. 317) in 1872, from a specimen in the Bitton garden.

Henry Nicholas Ellacombe, the subject of this memoir, was born at Bitton on Feb. 18, 1822, where he died on Feb. 7, 1916. Of his early life little is known, nor at any time could his career have been described as eventful. He went up to Oxford in 1840 and graduated in 1844 : at this period the " Oxford Movement " was at its height, and both father and son came under its influence. He was ordained in

HEXRY NICHOLAS ELL.VC'OMBE 51

1847, and in 1850 succeeded his father : after this his "life was centred almost wholly in Bitton parish, Bitton Church, and in his vicarage and garden." In 1852 Ellacombe married ; one of his sons, Dr. Gilbert Ellacombe, has sent from Rhodesia interesting plants to Kew, and is commemorated in Kalanchoe EUacombei. In 18G1 he became a member of the Bath Natural History and Antiquarian Field Club, of which in 1894 he was president, in succession to Leonard Blomefield : he contributed various papers to the Club ; the first (in 1869) was on " The Common English Names of Plants," a subject that much interested him and which he further developed in Tht Plant-Lore of Shakespeare, first published in 1878. This ran to three editions : the second (1884) contained a useful appendix of names from fifteenth and sixteenth century writers with which Shakespeare might have been familiar. The flowers of early writers alwa}rs attracted Ella- combe ; those of Gower, Chaucer, Spenser, and Milton formed the subject of papers in the Gardeners'1 Chronicle for 1915. The contents of his In a Gloucestershire Garden (1895) and In my Vicarage Garden (1902), notwithstanding their titles, are miscellaneous in character ; the best account of the garden is that by Mr. W. J. Bean in the volume before us, and in the delightful chapter by Miss Ellen Willmott, a frequent visitor to Bitton, headed " Canon Ellacombe and his Flowers." In this she mentions, among those which their owner regarded with especial affection, the little black pansy, which was " brought from Italy by his father in the early part of the nineteenth century ; it became a permanent occupant of the garden, and very few visitors left Bitton without a plant of it. The Rev. H. T. Ellacombe had identified it in Van der Glass's (sic) picture ' II Presepio,' circa 1450, now in the Pitti Gallery. The Canon paid a special visit to Florence to see the picture his father had mentioned, and he was greatly pleased to recognize unmistakably the little flower which is now so widely known as the Bitton black pansy." Those who have visited the Sala Hugo Van der Goes will not forget the charming representation of this little flower, which is strewn about at the foot of the vase of exquisitely-drawn flowers in the foreground of the picture.

The connection with Kew, which had been begun by his father, was renewed by the Canon in 1869, and continued throughout his life under three Directors, with each of whom he was on cordial terms. Sir Joseph Hooker dedicated to him the volume of the Botanical Magazine for 1881, and various species from the Bitton garden are figured in that periodical — it will be remembered that the editor of the book we are considering is Assistant-Director of the Royal Gardens, and to his personal reminiscences of Ellacombe, dating from 1912, much of its interest is due. Others have also contributed their quota : Miss Ellen Willmott, already cited, gives an attractive study of that admirable composer of madrigals, Robert Lucas de Pearsall (1795-1855), who was closely associated with both the Ellacombes — where, by the way, did he find in " the old Roman Catholic Requiem" the words "Da nobis pacem," which are quoted in his letter as from that source ? Reviews of Ellacomhe's books come from Mr. D. C. Lathburv and Sir Arthur Hort — the author of

52 THE JOUHXAI; 0? ISUl'AXV

the admirable translation of Theophrastus's 'Enquiry into Plants ; for this we are indebted to the Canon, who suggested that Sir Arthur should undertake it.

Although Ellacombe was not what is called a letter-writer, his notes, usually very short and connected with his hobby, give a pleasant idea of the man. One of the most interesting is that in which he announces the discovery of the two long-lost volumes of plant-drawings made for the Duchess of Beaufort at the beginning of the eighteenth century and preserved at Badminton ; of these very interesting collections a fuller description than that given bv Mr. Hill will be found in The Garden for August 28, 1920. There are pleasant accounts, mostly from his note-books, of his trips abroad, usually to Switzerland. The longest is that of Piora, above the St. Gothard Pass — a place which Ellacombe was the first to bring into notice ; this, with four other papers, including one on Roses, to which he was devoted, is printed at the end of the volume.

The book is attractively produced, and is embellished with numerous portraits and other illustrations : but it has one serious defect — not only is there no index, but the table of contents is. of the most meagre description. Such omissions in almost any book are reprehensible enough, but in a volume such as this, abounding as it does in references to plants, places, and persons of interest, it is little short of criminal.

British Plants: Their Biology and Ecology. By J. F. Bevis and H. T. Jeffebt. Second edition, revised and enlarged. Methuen, 1920. Pp. xii, 346. Price 7.?. 6d.

This useful' compendium of plant-biology has justified its existence and arrived at a second edition, in which "a rather large number of alterations and additions have been made in order to keep pace with the onward march of the subject." It makes a very read- able book for those who are anxious to know something more than the names of the plants they find, and serves as a good introduction to more serious botanical works. It is a multum in parvo, every biological aspect being touched upon : first, climate and its effect upon vegetation, then the ecological factors, water and its influence — xerophytes, water-plants, and tropophytes — light and heat, then soil and soil-biology. The first part appears to be essentially a conden- sation of Schimper's well-known work in which most of the important plant relations are touched upon. Some misleading statements need correction: e.g. (p. 27) transpiration signifies the evaporation of water from the shoot whether controlled or uncontrolled, epidermal or stomatal. The recent work appearing in the Jotirnal of Ecology casting doubt on the significance of the hydathode has been over- looked. What to the writer appears as a serious omission is the neglect of Blackmail's theory of limiting factors, which has now been applied to so many aspects of plant-biology, both in respect of water, light, CO;,, and heat, and also in Russell's book to soil-factors. The dependence of the activities of the plant on limitation by any of the numerous interactive factors should be made the starting-point of any modern consideration of plant-biology, and it is hoped that the

BUTTISK PLANTS 53

authors will in a future edition give the reader this most important clue to the plant's true relation to its surroundings.

In the second part the relation of varying morphology to the life of the individual plant is considered, the various terms in use being explained and the biological significance of the forms noted. The special biology of the climbing habit, of less usual methods of nutrition (parasites, symbiosis, etc.) and of food storage, each receive a chapter, and pollination and fruiting are dealt with at greater length.

The third part mostly considers the associations of plants and their relation to the habitats in these islands; it is preceded by short chapters on Evolution, the Origin of the British Flora, and the Classification of Plants. An attempt is made to remedy the sketchiness of these by three appendixes on Weissmann's law of Heredity, the Mendelian Theory, and Botanical Provinces, but readers interested in these aspects should supplement the information given. The work necessarily suffers a little from the difference between vastness of the area to be covered and the necessary limita- tions of publishing. Having written thus, it may seem contradictory to ask for more, but a little more attention given to cryptogams might stimulate the study of these groups by those who possess microscopes.

The indexing is good, and the print and paper very pleasing. Some misprints and mis-spellings occur : e. g., Sue da, which appears throughout. The book should certainly stimulate the reader for whom it was intended.

A. J. W.

Studies in Fossil Botany. By D. H. Scott, D.Sc, F.R.S., etc. Third Ed., vol. i. 8vo, pp. 434 ; 190 figures. 25s. net. London, 1920 : A. & 0. Black.

Each year that passes sees additions to our knowledge of the plants of the past. In reviewing these advances, we may sometimes notice that the pakeobotanist has discovered forms which are allied to living genera and which, as in the case of the fossil Osmundacea;, carry back the history of a familiar group over untold ages. In other cases the plants discovered are of strange types and apparently unrelated to living forms, or, again, the investigator may have un- earthed specimens which are synthetic and seem to unite some of the characters of one known group with the possession of features which are regarded as distinguishing some other different class of plant. As with all other types of research, these discoveries are published to the world in very many different journals in diverse countries and languages, and onlv the few can see them all and form a view of the knowledge which results. Dr. Scott's work gives a magnificent view in its true perspective of the edifice which has resulted from the study of fossil plants. It is an edifice built on a safe and substantial basis of fact ; speculations reared like pinnacles on an unsubstantial basis of problematical and badly-preserved specimens are eliminated; just so much detail is added to show the nature of the building and the texture of the work, while the main and usually uncontrovertible .

54 THE JOUENAL OF BOTANY

features of plant evolution emerge. As in previous editions, the illustrations form a noteworthy feature of the book, and many new and beautiful figures are included in this edition.

Twelve years have elapsed since the publication of the second edition, and during that period several noteworthy advances have been made in our knowledge of the Palaeozoic Pteridophyta. Some of them have helped to fill up gaps in the long continued researches on such familiar plants as Catamites and Sphenophyllum. In other cases work, such as that of Kidston and Grwynne-Vaughan, of Bertrand and of Gordon on the older Ferns, has not only increased the store of knowledge, but has cleared up uncertainties and provided a fund of authentic information, which enables us to reconsider some evolu- tionary theories maintained in former days without much positive evidence. To those who have not had the opportunity of studying the original papers, the brief sketch given here, with admirable illus" tration, of the past history of the Osmundaceae, must prove of considerable interest; as also of that remarkable group of Paleo- zoic ferns, the^ Zygoptendeae, with their curious petioles and two- or four-ranked pinna; which do not seem to have developed the flattened laminae now characteristic of leaves.

But it is the last and entirely new chapter in the volume to which all botanists will turn with eagerness. Many of us will begin the book by reading first this last chapter ; and we may be excused, for it contains an account of the earliest known Devonian land-plants. It is the first full and illustrated summary yet published by an independent expert, of Kidston and Lang's work on those 'most remarkable genera, Rhynia, Hornea, and Asteroxylon — work which must be regarded as one of the most important' contributions yet made towards our knowledge of plant evolution. In recent years some of us have almost despaired of finding any definite evidence of really early and generalized fossil plants. We seemed to have in the Lower Carboniferous and Upper Devonian rocks highly-evolved types of Perns, Lycopods, and Pteridosperms, nearly as complex as those varied types of Upper Carboniferous age. But now in the Psilo- phytales we have plants of a very simple type — leafless, rootless, with the simplest of vascular tissue, with the simplest sporangia, and much more like the primitive plants which we may have imagined. And moreover, these forms are not known from blurred impressions or from isolated fragments, but from petrified examples whose preservation is most perfect, whose stems give sections as fine as if cut from a recent plant, displaying such features as the stomata, rhizoids, and nvycorhiza. It is almost incredible that the simplicity of these forms should be due to reduction following the adoption of the mycorhizal habit, and we may have to revise our ideas upon the early evolution of plant- structures. Some theories of which little has been recently heard, such as Treub's ^protocorm" theory, may once more come into prominence, and Lignier's views on the morphology of stem and leaf will have to be given fresh consideration.

It is interesting to notice that Dr. Scott is favourable to the ideas which Dr. A. H. Church has recently put forward, and it may be that the discoveries of these most ancient plants will awaken a fresh interest in the study of the morphology of the marine al°-£e of the

STUDIES IS, FOSSIL BOTANY 55

present dav. Perhaps Church's theory provides a clue to the problem of why such generalized plants as the Psilophytales and such highly- specialized forms as the early Pteridosperms seem to be separated by a comparatively short space of Geological time. But whatever may be our opinion of current theories, the present work gives the botanist a ready means of becoming acquainted with the earlier plants which grew upon the earth.

H. H. T.

BOOK-NOTES, NEWS, etc.

An admirably illustrated monograph of The Leguminous Plants of Hawaii (Honolulu, July, 1920) has been provided by Prof. Joseph P. Rock, of the College of Hawaii. Published by the Experiment Station Committee of the Hawaiian Sugar Planters' Association, who are to be congratulated on their encouragement of science, special attention has naturally been paid to the economic species ; but from a botanical standpoint it is thoroughly satisfactory, the descriptions being full and scientific, with a clavis to the longer genera and a full synonymy: the notes on certain groups — e.g. the various forms of " Koa " — are of great interest and importance. The book is very well printed and has an excellent index ; its attractiveness is increased by the plates — 92 in number; many of them are taken from an earlier work by the same author on The Ornamental Trees of Hawaii, and show the whole tree.

The second part of the botanical section of the important work — Nova Caledonia — on the scientific researches in New Caledonia and the Loyalty Islands, edited by Fritz Sarasin and Jean lioux and published last year at Berlin and Wiesbaden, contains an enumeration of the Fungi by Miss E. M. Wakefield and numerous contributions on other groups : the phanerogams are undertaken by Messrs. Schinz and Gruillemin, who are editors of the section. The page-headings give no information beyond the number of the page ; it is strange that this opportunity for conveying useful information should be neglected in a work of this kind.

Science Progress — whose page-headings remain a standard example of neglected opportunities — contains a paper on " The Soya-Bean Problem," by Dorothy M. Atkins, B.Sc, in which the economic value of Glycine hispida is set forth and its cultivation described. " Experiments in some parts of the British Empire show that local conditions are favourable " to its cultivation, but no " serious attempt to start Soya-bean culture on an economic basis has been made." Nevertheless, Miss Atkins hopes (in italics) " that toe shall continue to encourage this promising crop, so that we may avoid repeating the history of our belated support of the sugar-heef'1 : presumably the " problem " is connected with the cultivation. Dr. Salisbury contributes the usual summary of recent advance in Botany, which would be more useful . if the month in which papers appear were named. There is still room for improvement in the proof-reading, and " Spencer le More " is an odd rendering of the name of our valued contributor.

The Imperial Bureau of Mycology, recently established at Kew, " is the outcome of a proposal adopted by the Imperial War Con-

30 THE JOURNAL OF BOTANY

ference in 1918 that a central organization should be established for the encouragement and co-ordination of work throughout the Empire on the diseases of plants, caused by fungi, in relation to agriculture," The committee of management consists of the leading biologists of the country, of whom a list is given in the Gardeners1 Chronicle for Jan. 1, from which the above information is taken. The Director is Dr. E. J. Butler, formerly Cryptogamic Botanist to the Indian Government, to which in May last year he became Agricultural Adviser.

At a time like the present, when there is so much to make up in the way of important publications which were delayed by the War, while the hindrances presented by the cost of printing show little prospect of removal, the fourth volume of the Recueil de V ' Instil ut Botanique Leo JSrrera issued by the University of Brussels seems to demand a protest. The volume contains nearly 700 pages and costs 50 francs ; it consists entirely of reprints of papers, most of them of no special importance and some already obsolete, contributed bv Errera and other authors to various journals at dates ranging from 1886 to 1899. No reason for its production is given, and in default of such explanation its publication seems unnecessary.

Mr. Oakes Ames has published a handsome volume, beautifully printed and admirably illustrated, a sixth fascicle of his Illustrations and Studies of the OrchidacecB (Boston, Merrymount Press). It contains an account of " the Orchids of Mount Kinabalu (Borneo)," by Charles Schweinfurth and Oakes Ames ; the latter also gives a continuation of his " Notes on Philippine Orchids." A new genus, PhiUppineea Schlecter and Ames, "the only endemic orchid genus of the Philippine Islands," is established for the plant previously described as Adenostylis Wenzelii.

In Notes from the Royal Botanic Gardens, Edinburgh (no. lxi ; dated Sept. 1920), Professor Bay ley Balfour continues to describe new species of Rhododendron, mostly from China, Avhence the supply of novelties seems inexhaustible ; seventy species are here named and described. Of Primula, which in the wa}r of novelties runs Rhodo- dendron close, Prof. Balfour describes from the same regions fifty- five novelties, with three of the allied genus Omphalogramma.

The Kew Bulletin issued in December (1920, no. 10) contains a paper on the Balsams of Chitral and the Kachin Hills by the late Major Toppin (1878-1917), arranged by Mr. S. T. Dunn from the MSS. and drawings which were bequeathed to Kew. Many new species are described and figured from drawings bj Major Toppin and Sir Joseph Hooker, whose notes on the collections are incorporated in the paper. The number also includes continuations of the " Dia- gnoses African* " and " Decades Kewenses."

We regret to record the death of the Rev. Canon H. W. Lett, which occurred at Aghaderg, Co. Down, of which place he had been rector since 1886, in December last, and of Mr. William Whitwell, of whom notices will appear later.

THE Journal of Genetics for October contains an interesting and important paper on the Bonavist Bean (Bolichos Lablab) by Dr. S. C. Harland.

Journ. Bot

Plate 557.

H. M. God/era, del.

]. X Xerapicamptii Forletii Godfery (Serapicamptis lingua x Anacamptis pyramidal is).— ■>. Single flower, sepals and petals flattened out (nat. size).— 3. Front (enlarged). — 4. Side view of column (enlarged). — 5. x Ophrys (0. arachnitiformis x 0. scolopax).

Cranbrookeana Godfery

0/

TWO NKW ORCHID HYBRIDS.

By Colonel M. J. Godfery, F.L.S.

(Plate 557.)

1. x serapica.mptis forbesii.

(Scrap/as Lingua L. X'Anacamptis pyramidalis Rich.)

Stem 22 cm. solid, cylindrical, glabrous. Leaves not gathered. Spike lax, 10-flowered. Ovary long (2.V cm.), slender. Bracts membranous, almost equal to ovary, narrow, lanceolate, acuminate, 3-nerved, tinged violet-red. Sepals free, spreading, lanceolate, acu- minate, reddish violet, paler than lip, midrib darker, a faint nerve on one side onlv. Petals slightly shorter, lanceolate, acuminate, darker red-violet. Lip flat, 3-lobed, side-lobes semi-orbicular, entire, mid- lobe narrow, lanceolate, acute, mucronate, concolorous, deep crimson, of a peculiarly brilliant almost velvety appearance, covered with microscopic short hyaline papillae. Column short, whitish, its walls produced on lip in two short raised slightly convergent sharp-edged ridges, concolorous with lip. Anther short, pear-shaped, apiculate, greenish. Fold of stigma clavate, red-violet. Rostellum violet. Viscid gland single, transversely oval. Pollinia small, dark brown. Caudicles ribbon-like, bright yellow. Stigma white.

The specimen described was found on May 5th, 1920, at Bordi- ghera, Italy, and sent to me by Colonel A. M. Forbes, who wrote : — " The orchids growing near it in quantities were pyramidalis and Serapias lingua, with one small group of »S'. longipetala about -40 yards away. We have never seen a specimen of O. laxiflora in that neighbourhood (though there are some at the mouth of the Xervia), nor of papilionacea, which we have never found yet. My wife thoroughly searched the locality where she found it, but could not find any other specimen, but there was one O. tridentata on a lower olive-terrace. "

The plant suggests so strongly the hybrids between Serapias and Orchis that it is evidently nearly allied to them. That the dominant parent is a Serapias there can be no doubt whatever, and it is almost equally certain that it is S. lingua. The size of the flower and of the lip suggest that species, and the shape of the lip is practically identical with that of lingua when flattened out — very different from the broad lip of cordigera, and equally so from the long one of longipetala. Hybrids between both the latter specie? and Orchis have a very much larger, broader, and more curved mid-lobe, much crisped and frilled at the edges, whilst in our plant the lip is flat in its natural position, with no fullness or frilling of any kind. More- over, in both & cordigera ai d S. longipetala the mid-lobe is densely covered with conspicuous hairs, whilst in S. lingua, though hairs exist, they are so much slenderer and fewer that the lip has often been described as glabrous. The lip in the present hybrid is glabrous except for microscopic papillae invisible to the naked eye. Finally, the guiding-plates or callosities at the base of the lip in 8. longipetala are distinctly divergent ; in our plant they are slightly convergent. Journal of Botany. — Vol. 59. [March, 1921.] f

."is THE JOU11NA.L OF BOTAST

It may fairly be concluded from these considerations, coupled with the fact that S. lingua was frequent where the hybrid was found, that that species was one of the parents.

To ascertain the other parent is not so easy. The abundance of Anacamptis pyramidalis in the immediate neighbourhood suggests that species, but it is rather shy of hybridization. The walls of the column in pyramidalis are prolonged in two erect guiding-plates on the base of the lip, analogous to the so-called callosities of Serapias, and the pollinia, as in Serapias, are affixed to a common gland. Anacamptis is therefore more closely allied to Serapias than is the srenus Orchis, and, since hybrids are well known to occur between Serapias and Orchis, there is no inherent improbability in the idea of a cross between Serapias and Anacamptis.

Well-authenticated hybrids have been recorded between S. lingua and Orchis papilionacea, and also between >S'. lingua and O. laxijlora. As the late Mr. Bicknell {Flowering Plants of the Riviera, sub t. lxiv.) savs that O. papilionacea has not been found at Bordighera, that species may be ruled out as a possible parent, especially since the published figures of S. lingua xO. papilionacea (Barla, Icon. Orch. pi. 22, figs. 4-8 and Camus, Mon. Orch. Eur. pi. 12. f. 337) at once put it out of court. O. laxijlora does not grow nearer to the place where our hybrid was found than the mouth of the Nervia, several kilometres away. Comparison with drawings of S. lingua X O. laxijlora found by my wife and myself in Italy has convinced us that O. laxi'lora is not one of the parents. We are therefore narrowed down to A. pyramidalis, which was plentiful in the neigh- bourhood, for though a single specimen of Orchis tridentata was found on a lower olive-terrace, there is nothing further to suggest the parentage of that species.

Of direct evidence of the influence of A. pyramidalis there is but little, but this is not surprising. The most salient features of that species are : —

(1) The long spur. No known hybrid between Serapias and Orchis shows any traces of a spur. As the influence of Serap>ias is so strong as to suppress this character entirely, Ave can expect no evidence in this direction.

(2) The guiding-plates on the lip. As the analogous callosity of S. lingua consists of one thickened cushion-like single mass, rounded at the apex and furrowed, its replacement in the hybrid by two quite distinct ridges may not unfairly be ascribed to the influence of Anacamptis.

(3) The saddle-shaped viscid disc. In the hybrid the rostellum is also transversely oval, so that in this respect Ave have a strong resemblance to Anacamptis. Moreover, in S. lingua x O. laxijlora the rostellum has "deux retinacles distincts " (Camus, Mon. Orch. Europe, p. 67). The fact that both pollinia are affixed to one single gland in S. Forbesii links it still more clearly to Anacamptis, and renders it more improbable that the second parent belonged to the genus Orchis.

The following points afford confirmatory evidence. The flowers are smaller than in S. lingua x O. laxijlora, the sepals and petals

TWO NEW ORCHID HTURIDS 59

more acuminate, the side-lobes of the lip not so broad, and the mid- lobe considerably narrower, in all which respects it approaches Anacamptis. The colour of the lip is a peculiarly bright crimson, almost impossible to reproduce, reminding one of the brilliancy of the most vividly-coloured specimens of A. pyramidal is on the Riviera. The ovary is long and slender, as in pyramidal is, which owes the shape of its spike to the long ovaries of the lowermost flowers standing out at right angles to the stem. The bracts closely resemble those of Anacamptis ; they are slightly shorter than the ovary, acuminate, membranous, and violet-red. In 8. ling 'a x O. laxiflora they much exceed the ovary, are broader, sub-obtuse, herbaceous, and green. None of these characters amount to positive proof, but taken together they form a chain of evidence, in conjunction with the presence of Anacamptis in the vicinity of the hybrid, as conclusive as can be expected in cases of this kind.

A difficulty — at first sight a serious one — may be raised on the ground that Serapias, with no spur, is adapted for visitation by sbort-tongued insects, such as Hymenoptera, and Anacamptis, with its long spur, for Lepidoptera, which have a long proboscis, and that it is very improbable that both plants should be visited by the same insect. The spur of Xiyritella, however, is only 2 mm. long, that of G yninadenia conopsea 15 mm. Yet hybrids between the two are, in some localities, by no means rare. Difference in the length of the spur is not therefore such a bar to cross-fertilization as might be supposed. In spite of its short spur, Midler observed that Nigritella was visited by no less than 4S different species of Lepidoptera. The difficulty, therefore, is more imaginary than real.

2. x Ophrys Cr.vxbrookeaxa.

(Oplirys arachnitiformis Gren. x O. scolopax, Cav. ) Stem 22 cm. high. Lower leaves resembling arachnitiformis, upper lanceolate, enfolding stem, acute. Bracts longer than ovary, erect, lanceolate, inrolled. Sepals greenish wbite, tinged with rose, with one green nerve, the lateral ovate rounded, with revolute edges, appearing triangular, the upper arched forward, oblong, truncate. Petals ligulate, narrow, pale yellowish green, ciliate, hispid in front with short erect hairs. Lip 3-lobed, side-lobes densely hairy, forming two projecting cones, mid-lobe semi-cylindrical, but not so much so as in scolopax, dark purple-brown, marked with two parallel irregular lines joined at the base by a collar. Appendix small, turned up in front, intermediate between the parents.

I gathered this plant on April 19, 1920, in a wood near Hyeres, France, under the impression that it was O. arachnitiformis, which in general appearance it much resembled, but the deeply-cut lip, with its forward-projecting conical side-lobes, so marked a character of O. scolopax, left no doubt as to the parentage of that species. It is not often that one finds a hybrid in which the evidence as to both parents is so clear. The two upper flowers had not }ret been visited by insects, but the ovary of the lower flower was. considerably developed. Both parents grew in the wood in some numbers.

I have named this plant after the Countess of Cranbrook, to whom

f 2

60 THE JOURNAL OF BOTANY

science is indebted for the discovery of the following hybrids at Hyeres: — Orchis Rainei Ilouy (O. OhampagneuxiixO. saceatd) of which I found a further specimen this year " Ophrys Rainei Alb. et Jahand." (O. arachnites x O. bomhylijlora) and O. olbiensis Godf. (O. arachnitiformis X O. Bertolonii). Last year she found, near Hyeres, a single specimen of Oplirys speculum Link, a North-African plant. This was first discovered in France by Moggridge. who found one specimen in 1865, and one the following year, near Menton. It was not seen again till it was found at Les Salins by Paine in 190S, at which place it has since been searched for in vain.

Explanation or Plate 557.

1. x Serapicamptis Forbesii Godfery (Serapias lingua x Ana- camptis pyramidal is i. 2. Single flower, sepals and petals flattened out (nat. size). 3. Front view of column (enlarged). -4. Side view of column (enlarged). 5. x Ophrys Cranhrookeana Godfery (0. arachnitiformis x O. scolopax).

ON HIERACIUM AUBANTIACUM L. By H. W. Pitosley, B.A., F.L.S.

Among my pleasant memories of childhood is a bed of orange hawkweed in my grandmothers garden which I suppose caught my fancy through the striking colour of its Mowers ; and for the sake of this association the plant, which of course is not uncommon in gardens, has always had a place in my own small plot. Being thus familiar with this hawkweed, I was rather surprised in the summer of 1916 to notice in the grounds of the manse at Aviemore, Inverness-shire, a profusion of a form which seemed to me distinct from my plant, and on my way home I met with this same strange form in cultivation at Pepper Arden, in North Yorkshire, where it was remarkably luxuriant.

This form is not characterized by a constant abundance of long leafy stolons by means whereof the plant rapidly spreads in all directions, but produces intermittently shorter and mostly under- ground stolons which effect a much slower increase. Its leaves are larger, of a duller green, with shorter, stiffer, and more abundant hairs, and in shape elliptical or obovate-lanceolate rather than Ungu- late or oblong. Its panicles, moreover, have fewer but larger heads of an orange-red or brick-red colour instead of brownish orange. These differences in habit, foliage, and flowers give the two forms the aspect of two distinct species.

On my return home in 1916 I saw by a reference to my herbarium that in 1910 I had collected this broad-leaved form, naturalized near Galashiels, without appreciating its distinctness ; and a large propor- tion of the British material in the National Herbarium at South Kensington, gathered mostly in the north, also belongs to this form.

In this collection is a specimen collected by G. Don in Banffshire,

ON UIERACIUM ArRAXTIACUM L. 61

which formed in part the basis of the introduction of H. aurantia- cum to the British Flora in English Botany, no. 1469 (1805), and from which the accompanying plate was probably drawn. The speci- men is a dwarf one, perhaps grown on poor soil, and the flowers are badly coloured in the plate, but it is reasonably certain that the plant belongs to the broad-leaved Aviemore form, with which the descrip- tion agrees. Backhouse, at p. 15 of his Monograph (1856), also refers to this form, and describes its rootstock as creeping, sub- stoloniferous, occasionally producing rooting stolons. Other British Floras describe the same plant ; Syme's English Botany and Hooker's Student's Flora remark '"stolons short or none"; and Babington's Manual "stoles often wanting." In Mr. Hanbury's Monograph, p. 7 (1889), which likewise mentions but one form, the stolons are said to be short or entirely absent, but the accompanying figure (no. 3) shows a distinct leafy stolon and narrow leaves, and I believe was not drawn from the broad-leaved form, although its flowers are sufficiently red.

It thus appears that the //. aurantiacum of British botanists generally is the broad-leaved form cultivated and naturalized in Scot- land and elsewhere, and that the narrower-leaved stoloniferous and more weedy plant commonly grown in present-day English gardens was not formerly known in this country and has not been botanically distinguished here — albeit it is now not only cultivated, but also naturalized in various English localities.

In most modern Continental works few variations of II. auran- tiacum have been noted other than those of the flower-colouring. According to Koch (Syn. Fl. Germ. ed. 1, p. 450 (1837)) the leaves may be oblong or obovate, and stolons present or absent — a description that would cover both of the British forms. Fries (Epicrisis, p. 24 (1862)) gives no varieties, but states that the plant, which is variable, is stoloniferous, with the leaves obovate or lanceolate. Rouy (Fl. de France, ix. p. 242 (1905)) describes a narrow-leaved form, which may be with or without leafy stolons, and adds one variety only. In DeCandolle's Brodromus (vii. p. 204 (1838)), however. Froelich diagnoses eight varieties. Unfortunately he does not well define the specific type, and the exsiccata cited ( Herb. Willd. no. 14659, kindly lent for inspection by Prof. Engler) are rather fragmentary ; but his first variety (majus) is clearly akin to our broad-leaved form, while the eighth {repens) recalls our narrow-leaved plant. This last will be further dealt with.

An elaborate treatment of the species is found in Naegeli and Peter's HieracienMittel-Europas-Biloselloiden (1885), where (p. 286) H. aurantiacum L. is divided into six groups, embracing nineteen subspecies. The group-characters are taken chiefly from the form of the phyllaries, the quantity of stellate hair on the involucres, and the colour of the leaves and flowers. The habit of growth and the other foliage-characters are considered under the separate subspecies, and rampant forms with narrow leaves may appear in the same group as others that are compact and broad-leaved. Geographical distribution is carefully shown throughout, though some of the subspecies are described from cultivated plants, and exsiccata are cited for some

02 THE JOURNAL OF BOTANY

forms, but not uniformly. Except in the case of plants with yellow or externally-striped florets, the colour of the ligules is termed " purple," which seems to indicate that the diagnoses were drawn up from dried material, the living flowers being certainly never purple.

Of the six groups the first — aurantiacum proper — is characterized by rather narrow phyllaries, short acladium, "purple" flowers, very pilose stem, and bright green or subglaucescent foliage. It contains eight subspecies, of which the third, H. aurantiacum L., sensu strieto, clearly coincides with the narrow-leaved English garden-plant. This is evident not only from the description, but from the exsiccata sent out by the authors, which show leafy stolons freely produced at the time of flowering. A number of varieties of this subspecies are dis- tinguished, of which our plant usually agrees best with var. setulosum, notable for the long hairs of its leaves. This subspecies is the most widely distributed form of the aggregate species, occurring throughout Central Europe and also in Scandinavia.

The broad-leaved British plant cannot be so surely identified with any of Naegeli and Peter's forms, but appears to be referable to the eighth and last subspecies (claropurpureum) of their first group aurantiacum. This subspecies, a native of Eastern Switzerland, is distinguishable by its partly underground stolons, its elliptical leaves with abundant stiff hairs above, its light-edged phyllaries, and its " clear purple " flowers. It seems to have been described from a cultivated plant, and no exsiccata for it are cited by the authors. Another subspecies of Eastern Switzerland, spanocJwetium, forming the collaborators' third group, is allied to this plant, and appears separable from subsp. claropurpureum only by its less hairy stem, more glaucous leaves, fewer but larger heads, and somewhat broader phyllaries. Flowering exsiccata of subsp. spanoclicetium in Herb. Mus. Brit, show no stolons and bear a considerable likeness to Don's Scotch specimen.

It may be concluded, therefore, that, of the two forms of H. aurantiacum known in Britain, one is referable to the Linnean type as defined by Naegeli and Peter, while the other is probably identical with their subspecies claropurpureum, though also resembling sub- species spanocheetium. Naegeli and Peter's nomenclature may con- sequently be followed for the two British plants if they are regarded as belonging to one species.

The differences of the two British forms, however, are so marked that, treating them alone, it does not seem possible to regard them as conspecific, especially when the average standard of Hieracium- species is remembered. Unfortunately I have never met with any wild forms of H. aurantiacum in Central Europe, and am thus the less able to judge whether a complete series of intermediate states between the different subspecies is likely to exist, so that no clear dividing-line is practicable. But the points of distinction between our two plants are so obvious, in spite of the dark hair-clothing of the inflorescence and the reddish flowers that are seen in both of them, that, after studying Naegeli and Peter's diagnoses, one seems warranted in doubting whether the basis of their grouping is suffi- ciently broad and whether sufficient importance is allowed to the

OX HIERACIUM AITBANTIACUM L. 63

differences in habit and foliage. The two British forms appear to me to show more essential differences than any of Naeeeli and Peter's primary groups, and I am therefore disposed to treat these two forms as two full species, under which many of the collaborators' subspecies may be placed. Taking this view, it becomes necessary to consider whether the Linnean type is correctly fixed hy Naegeli and Peter.

Hieracium auraniiacum is thus described hy Linnaeus in Species Plantarum, p. 801 (1753) :—

" Hieracium foliis integris, caule subnudo simplicissimo piloso corymbifero. Hort. Cliff. 388. Hort. Ups. 238.

" H. hortense noribus atropurpurascentibus. Bank. pin. 128. prodr. Go.

" H. alpinum non laciniatum, flore fusco. Bauh. pin. 12S. prodr. 65.

"H. germanium I. Col. ecphr. 2, p. 28, t. 30. " Pilosella polyclonos repens major syriaca, tlore amplo aurantiaco. Jforis. hist. 3, p. 78, t. 8. f. 7. " Habitat in Syria, Helvetia."

In the second edition (1703) the habitat becomes " In Syrise, Helvetia?, Au'strneque sylvis."

It will be observed that Linnaeus here furnishes no diagnosis, but relies entirely on citations from older works. The first of these is from his own Hort its Cliffortianus (1737), where the meagre description is of less importance than the citations, all of which, with one doubtful exception and excluding C. Bauhin's S. alpinum non laciniatum, distinctly relate to a broad-leaved garden-plant. The exception is Pilosella indica Cornuti, Canada, p. 209 (1035), a plant said to have been brought from the Indies, of which little is known, but which is represented in the Sloane Herbarium ( H. S. 310, f. 203) \>y what seems to be the broad-leaved garden form of S. aurantiacum. The final citation in Hort. Cliff, is Auricula maris liispanica from the posthumous Historia of John Bauhin, ii. p. 1010 (1051), which gives a full and accurate account of the plant, mentioning its underground stolons, its large, rough, distantly denti- culate leaves, some exceeding two inches in breadth, and the notable colour of its flowers. An interesting remark is that the author first saw this plant in 1008 in the royal garden at Stuttgart.

Linmeus's second reference, his Sort us Upsaliensis (1748), adds nothing material to the first one. The third citation, from Caspar Bauhin's Prodromus (1620) an&Pinax ( 1023), describes a garden-plant with leaves three inches broad ; and the fourth, from the same works, a smaller plant which the author suggests is the wild form of the larger one. It is remarkable that the native habitat of this wild form is given as Eastern Switzerland (Bhaetia), whence Naegeli and Peter record their broad-leaved subspecies claropurpureum and spano- chcetium. In the Pinax a reference is given under the garden-plant to a figure in Besler's Sortus Eystettensis (1013) which clearly recalls our broad-leaved Scottish form.

Linnaeus's fifth citation, from F. Columna's Ekplirasis (1616), concerns a cultivated plant obtained from the mountains of Germany,

04 THE JOURNAL OJT liOTANT

which is said to he of creeping habit and with large foliage. A good woodcut is given, representing a plant with oblanceolate, acute leaves, and a close panicle of many heads. The exact characters of this plant are uncertain, hut it is identified by C. Bauhin with his wild plant from Eastern Switzerland.

The sixth and final citation, from Morison's Historia (1699). is shown both by the description and the figure to refer to our broad- leaved garden-plant, and is identified by the author with the culti- vated plants of John and Caspar Bauhin already mentioned. The plant is apparently called "syriaca" in error, the name being taken from PiloseUa syriaca 0. B. Pin. (cpxoted among the synonyms), which was founded on P. maxima syriaca of Lobel and Tabernaj- montanus, a Syrian plant with square stem and opposite cauline leaves that is not a hawkweed.

It will be seen from Linnseus's citations that the hawkweed which he named H. aurantiacum was primarily a broad-leaved garden-plant known to many of the older botanists, whose descrip- tions are mostly free from ambiguity, and that a montane Swiss plant, with less defined characters but reputed to be its wild form, was included with it. Linnaeus apparently knew nothing of its native country, the habitats Syria and Switzerland being evidently copied from Morison and C. Bauhin. It is not mentioned in the Flora Sitecica, for Linnaeus was not aware of its occurrence in Scandinavia. A svnonym of this species earlier than any of those cited by Linnams is the Hieracium IX. of Clusius's Hist. Bar. Plant, i. lib. 5, p. cxlii (1601). Clusius mentions the stoloniferous habit, and comparing the umbelliform flowers with those of the common hawkweed, he writes " sed colore longe elegantiores, nempe impense fiavos et quasi aureos." The plant is stated to have been raised in Holland from seed obtained from Vienna.

The identity of the broad- leaved garden-form described by Linnaeus and earlier authors with the form grown in Scotland may be seen bv a reference to Parkinson, Gerard, and Bay.

Parkinson (Paradisus, p. 300 (1629)) furnishes a good descrip- tion of it as a British garden -plant under the name of Pilosella major, one of the synonyms in C. Bauhin's Pinax. The description contains an allusion to its broad leaves, which are well shown on the accompanving plate, and an English name "Golden Mouse-Eare " is introduced.

The phut is not mentioned by Gerard himself, but in the second edition of the Herball, lib. 2. p. 305 (1633), T. Johnson accurately diagnoses it as Hieracium hortense latifolium, and furnishes a specially good figure which unmistakably represents our Scottish broad-leaved form. Johnson also identifies it with the cultivated plants of C. Bauhin and E. Columna cited by Limueus, and with the plate in Hortus Eystettensis. He mentions that it is a rare garden- plant, and quaintly hrin^s in a second English name, saving: "The stalkes and cups of the fioures are all set thicke with a blackish downe or hairinesse as it were the dust of coles ; whence the women, who keep it in gardens for noveltie sake, have named it Grim the Colliar." This is an allusion to a humorous comedy, popular in

ON HIEKACIITM AUBANTIACITM L. U.">

Queen Elizabeth's reign, called "Grim the Collier of Croydon." Parkinson curiously objected to this name as "both idle and foolish."' Ray (Historic, i. lib. 5, p. 243 (ltJSti)) adopts J. Bauhin's name for our plant, adding those of Parkinson, Johnson's Gerard, and C. Bauhin as synonyms, and drawing- attention to its probable identity with Hieracium IX. of Clusius.

It is thus reasonably certain that the plant intended by all these authors is the broad-leaved form known in Scotland, naturalized and in cultivation, and that this form was at one time, if not at present, extensively "Town on the Continent.

A reference to the Linnean Herbarium, where the specimen of H. aurantiacum, though fragmentary, can be seen to belong to this form, confirms this view; moreover, the ten specimens of Auricula muris hispanica in the Sloane Herbarium (indexed in the copy of Kay's Historia in Herb. Mus. Brit.) can likewise be seen to be all this same form, showing that it was the plant in general cultiva- tion and commonly known to the botanists of that period. It may he added that the old name "Auriculae muris," or "mouse-ear," also points to this plant.

It therefore seems established that this hawk weed, which was introduced into gardens in various European countries about the end of the sixteenth or beginning of the seventeenth century, became widely known owing to the unique colour of its flowers, and, as a garden-plant, formed the basis of Linmeus's account of H. auran- tiacum. Linmeus's name must therefore remain with this form if it is separated as a species from our rampant, narrow-leaved plant that Naegeli and Peter take as the specific type of H. aurantiacum.

A wild and a garden variety of this species, both with broad leaves, are given by Haller (Hist. Hclvet. p. 21 (1708)), who distinguishes them by the colour of the flowers being orange and red respectively. Jacquin (Flora Austriaca, v. t. 410 (1778)), depicts in a very fine coloured plate a subalpine form of the broad-leaved plant, with underground stolons, foliage somewhat more lanceolate, acute and toothed than in our garden-form, and rather large, deep red flowers.

The earliest undoubted reference to a narrow-leaved II. auranti- acum seems to be that in Allioni's Flora Pedemontana, i. p. 213, and iii. tab. 14. f. 1 (1785). The plate here pourtrays a plant, gathered on Mont Cenis, with narrow, oblong leaves, and. as Allioni states in the text, sulphur-coloured flowers with fimbriate-laeiniate ligvdes. This figure, representing a monstrosity, is strangely cited by Naegeli and Peter for their subspecies claropurpureum var. Occi- dent ale.

Another narrow-leaved form is described and figured by Vahl in Flora Danica, vii. p. 5 and tab. 1112 (1799). The plant drawn shows a long, leafy stolon, oblong-spathulate leaves, and orange-coloured flowers; and it is clearly akin to Naegeli and Peter's typical sub- species aurantiacum and the narrow-leaved British garden-form. This plate, however, like Jacquin's, is not cited by Naegeli and Peter. This is the first record of the occurrence of II. aurantiacum in Scandinavia.

GO THE JOURNAL OF BOTANY

In 1800 a. distinct species of this group was described by Will- denow in Spec. Plant, ed. 5, p. 1564, as H. repens, with a diagnosis " H. scapo ramoso multitioro nudo piloso, foliis oblongis obtusiusculis integerrimis, stolonibus repentibus," the Mowers being stated to be " saturate tlavi." This plant is reduced to a variety repens of H. aurantiacum by Froelich in DeCandolle's Prodromus, vii. p. 1204 (1838), but is made a synonym ? of H. prussicum (H. collinumx Pilosella) by Naegeli and Peter (I. c. p. 373). Through the kind- ness of Prof. Engler I have been able to see the type-specimen in Willdenow's herbarium, no. 14663. This has the facies of a narrow- leaved H. aurantiacum, but when carefully examined it is found to lack both the red colouring of the ligules and the black involucral hairs characteristic of that species, and I think it is probably a hybrid such as Naegeli and Peter suggest, although the long pilose hairs of the stem seem to show the influence of H. aurantiacum. Willdenow's name cannot therefore be adopted to represent Naegeli and Peter's narrow-leaved type.

In works subsequent to Be Candolle's Prodromus I have been unable to find this narrow-leaved rampant plant distinguished either as a species or a variety, except by Naegeli and Peter, by whom it is treated as the true H. aurantiacum L., sensn stricto. It has been shown at length that this name really belongs to the broad-leaved garden-form, and as it is intended, for reasons already explained, to deal with the two plants as distinct species, a new name for tins narrow-leaved form is required. It is therefore proposed to rename it H. brunneo-croceum in allusion to the colour of its flowers.

In thus dividing this group of hawkweeds into two species, I am fully aware that while this treatment should serve readily to dis- tinguish the two plants known in Britain, it may not be of equal use in elucidating the relationships of the various wild Continental forms, with which I am unacquainted in nature, and of which I have examined only the limited collections in Herb. 'Mus. Brit, and (partially) at Kew. But the forms described by Naegeli and Peter seem largely divisible into two groups, whereof one may be dis- tinguished by a predominance of underground stolons and a tendency to broad, elliptic foliage, while the other set has more pronounced, leafy stolons and narrower, oblong leaves ; and I think that a large proportion of these forms can thus be naturally associated with the two species recognized in this paper.

I may add that I have not traced the origin of the narrow-leaved form as an inhabitant of British gardens. I have reason to believe that the particular plants I knew as a child were brought with Gen- t a urea montana from Western Switzerland in the early seventies, but of this I am not certain. The form being widely distributed, however, it is not unlikely that it was introduced at different times from various habitats and then disseminated bv gardeners owinc: to its very easy propagation. Examples from different localities some- times show minor differences that may be derived from the wild stocks.

The two species may be diagnosed thus : —

HlERACIUM AURANTIACUM L. Sp. PL SOI (1753); Smith, Eng. Bot. no. 1469 (1805); Syme, Eng. Bot. ed. 3, v. 166 (1866);

OX HTETCACTrH ArKANTIACTM L. G7

IT. aurantiacum subsp. claropurpureum ? Naegeli & Peter, Hier. Mittel-Europ. Pilosell. 291 (1885).

Pilosella major Park. Parad. 300, cum icone (1629) ; Hieracium hortense latifolium Gerard, Herb. ed. 2. lib. 2, 305, cum icone bona (1036); Auricula muris hispanica Pay, Hist. i. lib. 5, 243 (1680). -Icones. Jaequin, Fl. Austr. t. 410 (forma foliis angustioribus lanceolatis dentatis) ; Eng. Bot. no. 1469.

Exsiec. G. Bon, Banffshire, in Hb. Mus. Brit. ! Baker, Wilton Woods, Cleveland, Yorks. in Hb. Mus. Brit. ! Trimen, Stanmore Heath, Middlesex, 1866, in Hb. Mus. Brit.!

Stolons rather short, mostly underground and scaly, producing rosettes of leaves round the parent plant. Stem rising from a radical rosette of leaves, 20-60 cm. high, pilose with spreading hairs 4-5 mm. Lone; hairs, except near the base of the stem, dark or black-based, mixed with dark, glandular and stellate hairs above, with 1-4 leaves in the lower half, decreasing upwards. Panicle corymbose, often only 2-4-headed when naturalized, but up to 20-headed in cultiva- tion. Acladium and peduncles rloccose and black-glandular, sparingly pilose with dark hairs. Bracts grey-floccose and pilose externally. Leaves elliptical, or, especialby when cultivated, elongate below and becoming obovate-lanceolate, commonly 10-20 cm. long (but reach- ing 30 cm.), and 2-25-6 cm. (rarely 7 cm.) broad, obtuse or mucronate, with the inner and eauline acute, subentire or distantly denticulate (very rarely subdentate), deep dull green or glaucescent, pilose, on the upper side abundantly, with stiff, light-coloured hairs 2-4 mm. long ; marginal hairs shorter (1*5—2 mm.). Heads about 25 mm. in diameter, deep orange-red or brick-red (sometimes purplish when dried), with involucres 8-9 mm. long, nearly truncate below. Phyllaries rather broad, obtuse, dull green, or when dried, blackish preen, with rather broad, pale margins (inner red-tipped), slightly floeculose, pilose with dark hairs mostly 2-3 mm. long, and with scattered dark glandular hairs. Styles light livid but appearing yellowish in contrast to the red ligules.

Native in Eastern Switzerland and probably in the Tyrol and

Austria.

Cultivated for over three centuries on the Continent and in Britain ; now grown here chiefly in the north of England and in Scotland.

Naturalized in many Scotch localities, as Banffshire (Bon i, Forfarshire ( Gardiner), Perthshire (between Killin and Kenmore, R. Brown, 1793), Kinross-shire (B. Stuart), Lanarkshire (Bothwell AVoods, Glasgow. J. P. 1870), and Selkirk (Galashiels, 1910 !) ; and in England at Berwick-on-Tweed (Hb. Hume), Yorkshire (Baker), Middlesex (Trimen), and Isle of Wight (St, Lawrence, B. Cooper).

Hieracium brunneo-cioceum, sp. nov.

H. aurantiacum auct, mult, non L. ; H. aurantiacum subsp. aurantiacum Naegeli & Peter, I. c. 288.

Icon. Fl. Danica, t. 1112 (ut H. aurantiacum).

Exsiec. Hier. Naegel. nos. 17!, 80!, 122! E. S. Marshall, no. 4190! (omnia ut H. aurantiacum).

Stolones, saltern in cultis, longi, vulgo supra terrain foliosi sed

08 THE JOURNAL OF BOTANY

nonnumquam infra squambsi, tandem folia rosulata e quibus caules Horiferi crescunt gerentes, per sestatem totam producti ut planta late serpat. Caulis gracilescens, 10-(30 cm. altus, basin versus folio supremo minimo 1— 1-foliatus, oninino pilis fuscis 3-6 mm. longis patentibus pilosissimus, superne parce floccosus et pilis obscuri- oribus basi nigrescentibus glandulisque riumerosis nigris brevioribus vestitu*. Ramiticatio corymbosa, pauea-25-cepbala, florifera densa, fructifera laxior. Acladium pedunoulique floccosi, parce pilosi, dense nigro-glandulosi. Bracteaj virides, externe pilosse. Folia angusta, oblongo-lanceolata lingulatave aut interdum oblanceolata basi at- tenuate, 6-10 cm. (raro 16 cm.) longa, P25-2 cm. (raro 3 cm.) lata, obtusa vel mucronata (interiora caulinaque acutiora), subintegra vel obscure sinuato-denticulata, dilute viridia, superne setis plus minusve longis (2-5 mm.) ssepe numerosis inferne marginibusque pilis brevi- oribus rarioribus vestita (in formis umbrosis minus breviusque pilosa). Calatbia circa 20 mm. in diametro, brunneo-crocea sed interdum in medio pallidiora (in sicco obscuriora). Periclinium 7-8 mm. longum, inferne rotundatum vel tandem truncatum. Squama3 angusta?, obtusse, olivacese (in sicco nigrescentes), marginibus obscure vel obso- lete pallidis, interiores apice rubro-tinctse, omnes plus minusve flocculosse, pilis fuscis vel nigris ad -1 mm. longis et basin versus etiam glanduliferis nigris brevioribus pilosissimae. Styli pallide lividi, per ligulas croceas, ut videtur, lutescentes.

Stolons, at least wben cultivated, long, mostly above ground and leafy, but sometimes underground and scaly, terminating in rosettes wbich develop flowering-stems, continuously produced and enabling the plant to spread rapidly. Stem rather slender, 10-60 cm. high. very pilose throughout with spreading dusky hairs 3-6 mm. long, sparingly floccose and with darker or black-based pilose hairs mixed with numerous dark glandular ones above ; with 1-4 leaves towards the base, decreasing upwards. Panicle corymbose, few-25-headed, dense in flower, but laxer in fruit. Acladium and peduncles floccose, sparingly pilose, and densely black-glandular. Bracts green, ex- ternally pilose. Leaves narrow, oblong-lanceolate or Ungulate, or sometimes oblanceolate and attenuate below, 6-10 cm. (rarely 16 cm.) long and P25-2 cm. (rarely 3 cm.) broad, obtuse or mucro- nate (inner and cauline more acute), subentire or obscurely sinuate- denticulate, pale green, with more or less abundant pilose hairs of variable length (2-5 mm.) above, and more sparingly and shortly pilose below and on the margins (in shade-forms less and more shortly pilose). Heads rather small, about 20 mm. in diameter, brownish-orange (darker when dried), but sometimes paler in the centre, with involucres 7-8 mm. long, rounded or Anally truncate below. Phyllaries narrow but obtuse, dark green (blackish when dried) with obscure or obsolete paler margins (inner tipped with red), more or less flocculose, abundantly pilose with dark hairs up to 4 mm. lono-, and with black glandular hairs intermixed towards the base. Styles light livid, appearing yellowish in contrast with the ligules.

Native in the hilly districts of Central Europe from Savoy and Piedmont to North Germany, Galicia, Transsylvania, and the Banat ; also in Sweden (ap. Naegeli and Peter).

021 HIEEACHJM AUKAKTIACUM I,. 69

Cultivated in Britain for over fifty years and now frequent in English gardens.

Naturalized in Worcestershire (Selly Oak, Thompson, 1902) ; Leicestershire (Bellgrave, Bell, 1909); Oxfordshire (Great Tew, L876); Carmarthenshire (Glynnir, Ley, 1876) ; W. Somerset (Cul- bone, Sayne, 1867); X. Devon (Trentishoe, Carruthers, 1883, and Barnstaple) ; W. Cornwall (Newlyn East, Beid, 1903) ; and probably in other counties.

THOMAS WALTER (1740 ?-88) AND HIS GRASS.

Br James Britten, F.L.S.

The herbarium of Thomas Walter, author of the Flora Caro- liniana, is one of the most interesting collections in the Department of Botany. Its history is given in a note, probably in the hand of the younger John Fraser (fi. 1799-1852), on the first page of the folio volume containing it : " The Herbarium of Thomas Walter, Esq., of South Carolina, author of the 'Flora Caroliniana,' pubd 1788. Presented May 23, 1819, to the Linnean Society of London, by John Fraser, son of John Fraser, the indefatigable North American Botanical Collector from the years 1786 to 1811, who died in London the latter year." It was purchased by the Department at the sale of the Society's surplus collections in 1863, for the small sum of fifteen shillings.

Each specimen is mounted, through a slit, on a small piece of paper of irregular shape, which was apparently cut from another volume in which the collection was originally contained ; this was cut up and rearranged alphabetically — probably b}r Fraser : on each of these slips the name (sometimes incorrectly applied) is in most cases written. The names are mostly in two hands : sometimes both appear on the same ticket. Mr. S. F. Blake and I, when he was over here in 1915, spent some time in investigating these, and the conclu- sions we arrived at are given in his paper on Walter's plants in Bhodora, xvii. 130. I am not, however, quite satisfied : for example, on the label of Spircea trifoliata, "which Mr. Blake accepts as Walter's the word stamens is spelt " stemnys," which one can hardly suppose Walter, who was evidently a man of considerable education, would have written : I do not, however, think this is in the same hand as the other labels identified by Mr. Blake as Walter's. On the other hand, in the labels attributed to Fraser " nova genera," as a singular, is of frequent occurrence, and this appears also in Eraser's printed list. Most of the labels bear a number ; I have not found out to what this refers. On the first page is inscribed " Walter's Herbarium 1786-1788. J. F." (i.e. John Fraser), and it is on their resemblance to this that the identification of the labels is based. A few of the specimens were loose in the volume when it was received ; these were fastened down when it was re-bound. The plants described in the Flora are not all represented in the her- barium; the specimens are usually small, but mostly sufficient. Many of them bear names in Smith's hand, added when the collection was at the Linnean Societv.

70 THE JOURNAL OF BOTANY

The importance of the collection was early recognized by American botanists. Pursh (Fl. Amer. Sept., pref. xvii : 1814) consulted it when it was in the possession of the sons of John Fraser, and quotes some of the names in it in his book. It seems to have attracted but little attention from English botanists, as when Asa Gray made inquiries for it on his first visit to England in 1839, Brown and Bentham knew nothing of it ; Lindley, however, discovered it in the possession of John Fraser, who sent it to the Horticultural Society's rooms for Gray's inspection. Gray * gives an account of it in the Journal which he wrote for John Torrey : he found " the examination very tedious, as the specimens are very often not labelled, except with the genus in his [Walter's] ' Flora,' so that I have first to make out his own species, and then what they are of succeeding authors. The specimens are mostly mere bits, pasted down in a huge folio volume. I suspect this was done by Fraser, and the labels have sometimes been exchanged, so that it requires no little patience. Some of the things I most wished to see are not in the collection, and there are several in the collection which are not mentioned in the ' Flora.' You would laugh to see what some of the things are that have puzzled us : thus, for instance, his ' CucubaJus polypetalns ' is Haponaria officinalis'., his ' Dianthus carolinianus ' is Frasera ! in fruit." Gray