Bibliotheque botanique EMILE BURNAT Cataloyue %" Provient de . Li\i'('s proNt'iuiiil (Ic l;i l)il)li()tli(''(|ii(' hol.-iiiifun^ (PEmile BurnaK l8"2S-l<»-20), insrivs en oclohiv ll»20 dans la l)il)liotht'(|n{' {\\\ Conservatoire bolaniqne de (ien i>iii' n^ \1 o xj- .^ise, K^"? THE JOURNAL OF BOTANY BRITISH AND FOREIGN. EDITED BY JAMES BRITTEN, K.S.G., F.L.S •otanicav VOL. XLIX ILLUSTRATED WITH PLATES AND CUTS LONDON : WEST, NEWMAN & CO., 54, HATTON GAEDEN. 19 11. XT I/, M-f LONDON: PRINTED BY WEST, NEWMAN AND CO. HATTON GARDEN, E.G. CONTEIBUTOES TO THK PRESENT VOLUME. Eleonora Armitage. E. G. Baker, F.L.S. James Britten, F.L.S. C. E. Britton. Cedric Bucknall. W. H. Burrell, F.L.S. E. M. Cardew. W. G. Clarke. E. H. COMPTON. A. D. Cotton, F.L.S. A. A. Dallman, F.C.S. F. H. Davey, F.L.S. H. N. Dixon, M.A., F.L.S. E. Drabble, D.Sc, F.L.S. H. Drabble. G. C. Druce, M.A., F.L.S. S. T. Dunn, B.A., F.L.S. J. E. Drummond, B.A., F.L.S. J. W. Ellis. P. EwiNG, F.L.S. Antony Gepp, M.A., F.L.S. E. S. Gepp. L. S. Gibbs, F.L.S. John Gerard, S.J., F.L.S. W. B. Grove, M.A. W. Hodge. A. E. HoRwooD. E. Jackett. A. Bruce Jackson. B. Daydon Jackson, Ph.D., Sec. L.S. J. H. A. Jenner. B. Kaalaas. H. M. L. Kensit, B.A. C. E. Larter. H. W. Lett. E. F. Linton, M.A. G. Lister, F.L.S. E. S. Marshall, M.A., F.L.S. Spencer le M. Moore, F.L.S. C. E. Moss, D.Sc. W. H. Pearson, A.L.S. E. P. Phillips, M.A. Henry Pierson. David Prain, F.E.S. H. W. PUGSLEY, B.A. E. F. Eand, M.D., F.L.S. H. P. Eeader, M.A., O.P. A. B. Eendle, D.Sc, F.E.S. E. A. ElCHARDS. H. J. ElDDELSDELL, M.A. W. MOYLE EOGERS, F.L.S. Ida M. Eoper, F.L.S. C. E. Salmon, F.L.S. Annie L. Smith, F.L.S. F. Stratton, F.L.S. E. W. Swanton. H. Takeda. H. H. Thomas. H. S. Thompson, F.L.S. K. M. Toms. E. F. Towndrow. Carl Traaen. W. G. Travis. W. J. TUTCHER, F.L.S. H. F. Wernham, B.A. G. S. West, D.Sc, F.L.S. William West, F.L.S. J. A. Wheldon, F.L.S. E. M. Williams. F. N. Williams, F.L.S. A. J. WiLMOTT, B.A., F.L.S. W. Wilson. A. H. WOLLEY-DOD. E. A. Woodruffe-Peacock, F.L.S. Directions to Binder. Tab. 509 to face page 1 „ 510 41 „ 511 73 „ 512 105 „ 513 137 „ 514 177 Portrait OF AuGusTiN Ley . 201 The Supplements (i. ' Notes on the Flora of Denbighshire ' and ii. 'A List of British Roses') should be, placed separately at the end of the volume. Joupn. Bot. Tab. 509. \ n / '., 2b 2c' 2f 3d H. N. Dixon del. TENERIFFE MOSSES. West, Newman proc. THE JOURNAL OF BOTANY BRITISH AND FOREIGN. TENERIFFE MOSSES. By H. N. DixoxN, M.A., F.L.S. LIRRART PEW YORK BtlTANiCAl. UAKDHM (Plate 509.) The Mosses dealt with in this article were collected by Dr. J. H. Salter, and sent to me for determination. Comprising about sixty packets, they included a comparatively large number of plants of note, three being new species, one of them of an extremely interesting nature, and several others being additions to this group of islands. For purposes of reference it will be well here to enumerate the new discoveries, an asterisk being prefixed to those new to the Atlantic Islands, and a dagger to those new to the Canaries but recorded previously for one or more of the other groups : — ■ jWeisia crispata Jur. *Aulacomniiovi androrjijnnm ■''Encahj2)ta vulgaris Hedw. Schwaeg. ■''Grimmia flaccida Lindb. ■■'Brachythecium Salteri, sp.nov. ■''G. tergestina Tomm. ■'■'Homalothecium harhcUoides, ■'■Orthotrichinn rupestre Schleich. sp. nov. ■•'•Bryum validicostatum, sp. nov. Dr. Salter has written the succeeding notes as to the localities, and the conditions under which the collections were made : — • " The Mosses enumerated in the following list were collected coin the island of Teneriffe during a stay of fourteen months (Dec. ^ 1908 to Fel). 1910). The greater part of this time was spent at ■-- Guimar, a small town and health resort situated at an elevation of about 1200 ft. upon the southern side of the island. The aspect is south-east, and, being sheltered by the central ridge of the island, the climate is peculiarly dry and sunny. No rain is expected during the five summer months. The rainfall in 1909 was ten inches, which may be taken as about the annual mean. The substratum is everywhere volcanic, largely consisting of old lava-flows. It follows that all Mosses growing in the open are exposed to extreme desiccation. Last season even the winter rains failed, so that there was severe drought in January and February of the present year. The succession of the zones of vegetation in the island is well known. Above the cultivated Journal of Botany.— Vol. 49. [Jan. 1911.] b A THE JOURNAL OF BOTANY coast lands is found, at a heiglit of from 2000-4000 ft., the ' Monte Verde,' or belt of evergreen woods, consisting largely of LaurmecB and Myrica Faya, with species of Ilex, Erica, Arbutus, Viburnuvi, and Cistus. The deep, rocky ravines called ' bar- rancos,' which everywhere furrow these wooded slopes, contain the springs which furnish most of the scanty water-supply of the island, and here, in dampness and shade. Mosses and Hepatics flourish luxuriantly. In the three magnificent ' barrancos ' of Eio, Badajos, and Ahavingo, all easily accessible from Guimar, most of the Mosses in the following list were collected. To the Monte Verde succeed the scant remains of the pine-forests, still lying within the influence of the cloud-belt. Above the upper limit of the Pinar sterility reigns, vegetation being almost restricted to certain endemic Leguminosce. There is here intense dryness of the atmosphere, strong insolation by day, a great difference between day and night temperature. Upon the Peak, at the shelter-hut (10,700 ft.), no Mosses w^ere noted, and w^ith difficulty a very few Lichens were found. At the Rambleta (11,700 ft.) the only moss met with was Grimmia tricliopkylla, maintained by the moisture due to condensation of steam escaping from a volcanic vent. Short visits were paid to some specially favourable locali- ties, as to the laurel-woods of Teno at the w^estern extremity of the island, the little patch of forest at Agua Garcia, and the luxuriant evergreen woods of Las Mercedes and Taganana, which support a cryptogamic flora unequalled elsewhere in Teneriffe. All these localities are situated upon the older formations, which evidently retain much more moisture than do the later volcanic deposits which now cover the greater part of the island." For assistance in drawing up the following list I am greatly indebted to Mons. J. Cardot. Ceratodon 'purjmreus Brid. Pinar, above Ladera, Guimar (no. 3). Cynoclontium Bruntoni B. & S. Pinar, above Ladera, Guimar (no. 12), c. fr. The capsule is more symmetrical, smoother, and with less distinct neck than in our usual European form, and the leaf-cells more pellucid. Leiicobryum glaucuin Schp. Taganana (no. 41). Fissidens serrulatus Brid. Agua Garcia (no. 42), c. setis. An unusual form, having the cells practically smooth. — F. adiantoides Hedw. Agua Garcia (no. 40). Gymnostomum calcareum Bry. germ. Monte Yzana, Guimar (no. 7). — Var. muticum Boul. Barranco del Eio, Guimar (no. 15). Weisia crispata (Bry. germ.) Jur. Barranco del Rio, Guimar (no. 17), c. fr. New to the Canaries. The peristome in several of the capsules is more highly developed than usual, the teeth red and quite conspicuously protruding beyond the capsule mouth ; in other capsules it is paler and much shorter. The leaves are stoutish, with wide points and strong nerve, often cucullate, and wath the margins widely enrolled. Timmiella anomala Limpr. Barranco, above Arafo (no. 8), c. fr. Barranco del Rio, Guimar (no. 23), c. fr. TENERIFFE MOSSES O Tortilla atrovirens (Sm.) Linclb. Old lava-flow, Guimar (no. 22), c. fr. This is probably the var. edentula (B. & S.) Par., having the peristome-teeth very short and fragmentary, and arising from a somewhat broad, papillose basal membrane. The degree of de- velopment of the teeth varies, however, a good deal, and the_ variety would seem to be of dubious value. — T. muralis Hedw. Above Arafo, about 4000 ft. alt. (no. 36), c. fr, A small form. Encalypta vulgaris Hedw. Cumbre, above Candelaria, at about 6000 ft. (no. 30), c. fr. The form with calyptra smooth or nearly so. New to the Atlantic Islands. Grimmia flaccida Lindb. (G. sphcerica Schp., Schistidinm jyul- vinatum Brid.). Monte Yzana, above Guimar, at about 7000 ft. (no. 4), c. fr. New to the Atlantic Islands. — G. tergestina Tomm. Near Monte Yzana, above Guimar (no. 20). Sterile, but deter- minable by the leaves. New to the Atlantic Islands. — G. campes- tris Burch. Eisco de Fiomarcial (no. 11), c. fr. Old lava-flows, Guimar (no. 16), c. fr. — ? G. Doniana Sm. Near Monte Yzana (no. 13). A sterile plant, with the habit of G. Doniana or nearly that of G. arenaria, but with usually rather wider leaves and shorter basal cells, seems perhaps to belong here ; but it appears to be dioicous. It may probably be nothing but a dioicous form of G. Doniana, but in the absence of fruit it is scarcely safe to record it as such. G. Doniana has not been recorded from the Canaries. — G. trichophylla Grev. Rambleta, Pico do Teyde, 11,700 ft. (no. 27), S ■ — <-?• commutata Hiibn. Eisco de Fiomarcial (no. 9), c. fr. Eecorded by Pitard and Proust from two localities on Teneriffe, but only in the sterile state. Ancectangium compactum Schwaeg. Barranco del Eio, Guimar (no, 24). — A. angustifolium Mitt, Ladera de Guimar (no. 34), c. fr. Orthotrichnm Sturmii Hornsch. Pinar, Cumbre above Cande- laria, about 6000 ft. (no. 26), c. fr. Pinar de Guimar (no. 35), c. fr. I have recorded this plant under the above name in deference to Schiftner and others, as it is in all probabihty the same plant as is already recorded from several localities, of one of which Schiff- ner (Hedivigia, xli. p. 287) writes : " Die zweischichtigen Blatter lassen keinen Zweifel iiber die Eichtigkeit der Bestimmung." Dr. Salter's plants have the leaves in the whole of the upper part bi-stratose, but the capsule, especially in no. 26, tapers at the base into the seta entirely as in 0. rupestre. Unless 0. Sturmii is to be separated from 0. rupestre on the ground of the bi-stratose lamina alone (and intermediate stages even of this character are frequent), it appears to me quite impossible to define any constant concourse of characters to distinguish it from 0. rupestre. — 0. rupestre Schleicli. On old escohon stem {Cytisus prolifer), Cumbre de Candelaria (no. 37), c. fr., teste Cardot. k tall, arboreal form, which is quite distinct from the plant above mentioned, and can- not be separated from 0. rupestre. This species (apart from the above 0. Sturmii) does not appear to have been recorded from the Canaries, or, indeed, from the Atlantic Islands. Funaria hygrometrica Sibth. Barranco above Guimar (no. 10), c. fr. B 2 i THE JOURNAL OF BOTANY Hajjlodontium Notarisii (Mitt.)' Broth. Barraiico above Arafo (no. 5), c. fr. Ejn-pterygium Tozeri (Grev.) Lindb. Fuente la Kosa (no. 29). Bryum argenteum var. lanatum B. & S. Santa Ci'uz Eoad, Guimar (no. 25). — B. Donianum Grev. Guimar (no. 32), c. fr. Bryum (Eubeyum § Trichophoea) validicostatum, Card. & Dixon, sp. nov. (Tab. 509, fig. 1.) Caespites densi, humiles, circa 1 cm. alti, pallide virides, subnitidi, getate rufescentes. Caulis sat crassus, in sectione transversa subpentagona, fasciculum cen- tralem magnum, rete laxmia tenue, cellulas externas 1-2 seriebus minores, incrassatas, ptilcherrime rubras exhibens. Folia caulina valde conferta, humida regulariter imhricata suberecta, sicca ad- pressa, imrum mutata, ])aullo contracta, nullo modo torquata, unde caules crassiuscule clavati teretes ; comalia densissime conferta, 2-2-5 mm. longa, 1 mm. lata, late oblongo-ovata, subol)tusa, nullo modo acuminata, perconcava, marginibus per totam fere longi- tudinem fortiter anguste recurvis, apicem versus denticulatis. Eete laxum, e cellulis elongate hexagonis, 60-80 /x longis, 20-25 /x latis, parietibus firmis, baud incrassatis instructum, marginalibus in 2-4 seriebus perangustis, limbum angustum unistratosum lute- scentem distinctum efformantibus ; cellulae apud basin baud de- currentem laxiores, rectangulares, saepe rubellse. Costa validis- sima, OBtate purpurea, ad basin 150-180 [x lata, inde cito angus- tata, apud mediam partem folii 60-80 /a lata, apice in cuspidem lutescentem subintegrum exiens ; in sectione transversa plano- convexa, dorso maxime prominens, prope basin 2-4 cellulis ven- tralibus magnis, ducibus 3-4 multo mii:ioribus, fasciculo comitum parvo, stereidearum permagno, cellulis dorsalibus circa 8, trans- versa ellipticis, majusculis. Innovationes graciliores, foliis mi- noribus, minus confertis, vix comatis, aliquando paullo lenissime torquatis. Dioicum. Seta et theca illis B. Doniani et B. i^latylomatis similes ; operculum magnum, alte acute conicum. Spori 10-13 ji. Peristomium pallide aurantiacum, dentes lamellis circa 30 dorso brevissime acutissime prominentibus, endostomii meinbrana pal- lida, 'prcealta, processus breviores, cilia nodosa, vix appendiculata. Hab. Barranco del Eio, Feb. 1909 (no. 19). Eisco de Fio- marcial, Feb. 1909 (no. 6). Mons. Cardot informs me that he also possesses the same plant from the Azores, leg. Carreiro ; and that the plant recorded as B. Teneriffce by Schiffner [Bornmiiller, no. 1673, and by implication also, no. 1680, from Monte (Tafera), Gran Canaria] , also belongs here (cf. Schiffner, " Neue Materialen .... der Atlantischen Inseln," Hedwigia, xli., 288). Probably some other records for B. TeneriffcB may also properly belong to this species. The affinities of this plant are with B. Tcneriffa Hampe, B. platyloma Schwaeg., and B. Doniamivi Grev. ; but it appears to be quite distinct. B. Donianum differs at once in the thickened border, and the leaves usually twisted when dry, B. ylatyloma in the wdde border and spirally twisted leaves. It is, I think, even more distinct from B. Teneriffce, which appears to have been some- TENERIFFE MOSSES 0 what misunderstood. I have examined the type of B. Tenerijfm'va. Hampe's lierbarium at the British Museum, and find it a plant of quite a different habit, with long, lax innovations bearing flaccid, flexuose leaves, as described by Hampe, whose note, " B. capillari proxima, sed foliis lato-limbatis superne dentibus aciculiformibus seu ciliformibus nonnullis longioribus vel brevioribus primo mo- mento distinctum," quite removes it from any very close affinity with the present plant. The stout nerve, purple at the base, and the teretely imbricated branches with the leaves erect and ap- pressed when dry, give the plant a very different appearance from any form of B. cainUare or the allied species. Bryum clava- tulum Card. & Dixon from the Azores has somewhat the same habit, but the leaf structure is quite different. From B. obconicum it differs in the stout nerve, wider, more obtuse, densely imbri- cated leaves, &c. The peristome characters also may have some value. The capsules were nearly all operculate, but mostly fairly well ripened. In the two cases where the lid had already become partially separated from the capsule, I found the annulus entirely adherent to the rim of the hd, and showing no signs of spHtting up or separation under pressure or after heating in water ; but I can scarcely think this to be a normal character. Mnium undulcUum L. Taganana (no. 38). Aulacomnium androgynum Schwaeg. Eock-boring made in search of water, Eisco del Frayle, above Vilaflor (no. 31). The usual gemmiferous form. Genus and species new to the Atlantic Islands. Bartratnia stricta Brid. Barrancos above Guimar (no. 2), c.fr. Philonotis fontanel Brid. By watercourse, Vilaflor (no. 33). Poyonatum aloides (Hedw.) P. B. Evergreen woods, Teno Mountains (no. 39), c. fr. Polytyidiiim juniperiniim Willd. Agua Mansa (no. 1), c. fr. Hedioigia ciliata var. leucoplum B. & S. Foot of Eisco de Fiomarcial (no. 45), c. fr. Leucodon sciuroides var. Teneriffcd Een. & Card. Evergreen woods, Teno Mountains (no. 49). Pterogonium gracile Sw. Barranco del Eio (no. 46) ; Ladera de Guimar (no. 54). Leptodon longisetus var. flagellifer Mont. Evergreen woods, Teno Mountains (no. 59 a). This variety is totally different in its habit and size from the type of L. longisetus (which differs little in its vegetative characters from L. Smithh), and it is difficult to realize that it belongs to that species ; the stems are four to five inches long, with long, often tiagelliform branches, resembling slender forms of Neckem turgida or N. mediterranea in general appearance. Neckcra intermedia Brid. Taganana (no. G2) ; evergreen woods, Teno Mountains (no. 64).— A'", cephalonica Jur. Evergreen woods, Teno Mountains (no. 59), c. fr. Homalia lusitanica Schp. Taganana (no. 58).— J/. Webbiana Mont. Taganana (no. 57). A robust form, with rigid stems, 6 THE JOURNAL OF BOTANY three inches long. Mons. Cardot points out that the nerve is nearly always double, while very short, while in H. Webbiana it is described as " nervo vix conspicuo unico " ; there appears, however, to be some variation in the character of the nerve, which is often almost obsolete, as well as in the habit and size ; and specimens in the British Museum collection agree well with our plant. Thamnium alopecurum B. & S. Evergreen woods, Teno Moun- tains (no. 55) ; Agua Garcia (no. 60). Homalothecium barbelloides Card, k Dixon, sp. nov. (Tab. 509, fig. 2.) Caules elongati, 6-10 cm. longi, infra arete intertexti, minime radiculosi, plus minusve distanter irregulariter ramosi, niticli, sericei, laete virides, inferne straminei, gracilUmi, mollissinii, flexuosi, filiformes, subcompressi ; habitu omnino Barbellarum gracillimarum, praecipue B. pendulce (Sull.) Fleisch. Caulis tenerrimus, in sectione transversa subrotundus, fasci- culo centrali vix ullo, reti laxiusculo pertenui, cellulis externis in 2-3 seriebus minoribus, valde incrassatis, extremis substereideis. Folia sat conferta, sericea, madore erecto-patentia, ssepius subcom- planatula, ramea raro paullo secunda, sicca parum mutata, i^arva, e basi valde concava lanceolata, stricta, sensim in acumen acutis- simum tenuem angustata, pluries plicata, 1-5-1-75 mm. longa, marginibus planis, integris vel apicem versus subdenticulatis, tenuicostata, costa longe sub apice evanida, circa | longitudinem folii attingente. Cellulse laminae angustissime lineares (circa 2- 2-5 /x latas), basin versus parum mutate, ad angulos tantum nonnullae breviores, latiores, subincrassatae, irregulares, subisodia- metricae, pellucidai, alas yarvas, male definitas nee autem obscuras effingentes. Folia ramea minora, angustiora, superiora saepius in subuiam tenuissimam flexuosam angustatae, tenuissime costatae, vix plicatae. Cetera ignota. Hab. Evergreen woods, Teno Mountains, Dec. 1909 (no. 52). A very pretty and delicate plant, the position of which in the absence of fruit must remain somewhat doubtful; th« structure of the leaves, however, agrees very closely with that of Homalo- thecium, and it will probably prove eventually to be rightly placed here. In habit it is totally distinct from any of the species of Homalotliecium, or indeed of the allied genera, and is an almost exact counterpart of Barbella pendula (Sull.) Fleisch. It is an exceedingly interesting addition to the flora of these islands. The acumen of the leaf varies considerably at different parts of the stem, sometimes being straight and comparatively short, though nearly always exceedingly attenuated ; at others, being longly subulate and often fiexuose or undulate. The branches are often so attenuate as to be almost flagelliform. The leaves are somewhat amplexicaul at base, deeply concave with the margins broadly platter-edged, so that when in the natural position they appear lanceolate from a very narrow basal line ; on pressure, however, they can be flattened out, and then show a broad line of insertion and an almost triangular outline, as in other species of Homalothecium. TENERIFFE MOSSES / Brachythecium Salter! Card. & Dixon, sp. nov. (Tab. 509, fig. 3). Caespites densissimi magni tumicli, sat robusti, flavo-virides, hand nitidi ; caules 3-i cm. longi, procumbentes, irregulariter vel subpinnate ramosi ; rami ascendentes, suberecti, valde conferti, inaequales, crassiusculi, obtusi. Folia caulina imbricata, arcuato- decurva ; ramea arete imbricata, e basi cordato-ovata sensim vel abrupte acuminata, acumine plus minusve attenuato, ssepius semi- torto, dense serrulato (caulinorum acumen tenuius minus serra- tum); ovanid, concavo-carinata, pluriesprofimdej^Ucati, marginibus ad basin late, superne angustissime distincte reflexis. Costa ad basin valida, f— | folii longitudinem gequans, apice sajpe in 1-3 sjiiculos dor so folii yrominentcs exiens. Areolatio sat densa, e cellulis lineari-vermiculatis, 40-50 jx longis, ad infimam basin et ad angulos paucis latioribus subquadratis instructa. Autoica. Fl. masculi numerosi, tumidi, inter femineos mixti. Folia pericbaetialia erecta, convoluta, apice subulato vel loriformi flexuoso vel subsecundo. Seta 1-1-5 cm. longa, rubra, scaberrima. Theca parva, breviter turgide ovata, cernua ; exothecii cellulae parvaB, incrassatse, collenchymatae, rectangulares angulis rotun- datis. Operculum breviter conicum. Peristomium aurantiacum ; dentes extern! sat breves, lamellis circa 20 intus prominentibus, endostomii membrana circa ^ longitudinem dentium gequante, flavida, ciliis 2-3 praelongis, tenuissimis, minima tamen fragilibus, brevissime appendiculatis. Spori 12-16 //. Hab. Fuente la Kosa, April, 1909 (no. 56). A very distinct species, most nearly allied to B. trachypodium, but entirely different in its habit, the absence of the glossiness that is a marked character of that plant, the turgid subjulaceous branches, the markedly recurved leaf-margins, less sharply toothed leaves, and small capsules, which are abundantly produced. The apex of the nerve usually ends in one to three serrulations, showing as short spinulose projections from the back of the leaf, as is frequently the case in B. trachypodium and B. Starkei. Bracliytliecium rutabulitm B. & S. Ladera de Guimar (no. 47). — B. purum (L.) Dixon. Taganana (no. 63). — B. illeccbnuii (Schwaeg.) De Not. Barranco del Rio, Guimar (no. 44 a.), c. fr. Eurhynchium Teesdalei (Sm.) Schp. Barranco Badajos (no. 14). It is on record for La Palma, but I have not seen any record from Teneriffe. — E. circinatum (Brid.) B. & S. Ladera de Guimar (no. 48). — E. meridionale (Schp.) De Not., forma. Barranco del Rio (no. 44), c. fr. A robust plant with strongly striate, rigid leaves, more like a small form of E. striatum. I take it to be the E. canariense (Hampe & C. M.) referred to by Ren. & Card, in Bull, de l'H6rb. Boissier, ii. 438 (1902). Bhaphidostegium substrumulosum (Hampe) Card. {B. Wel- witschii Schp., Sematophyllum auricomum Mitt.). Agua Garcia (no. 28), c. fr. Amblystegium riparium (L.) B. & S. Barranco del Infieruo, Adeje (no. 43). Fuente Madre, Vilaflor. O THE JOURNAL OF BOTANY Explanation ch^ Plate. Fig. 1. Brijum validicostatum. a, plant, natural size, b, stem (dry), x 3. c, leaf viewed from front, x 20. c', do., viewed from back, x 20. d, leaf apex, X 50. c, nerve section at base, x 80. Fig. 2. Homalothecium barbcUoidcs. a, plant, natural size, b, part of stem, X 2. c, leaf in natural condition, x 20. c', do., under pressure, x 20. d, leaf apex, x 200 (the cells are drawn slightly tco wide), e, upper cells, X 200. /, alar cells, x 200. Fig. 3. Brachijthecitiin Salteri. o, plant, natural size, t, &', stem-leaves, X 20. c, upper cells, x 200. d, alar cells, x 200. e, apex of branch leaf, X 40. f, capsule, x 3. FURTHEE NOTES ON THE FLORA OF FLINTSHIRE. By a. a. Dallman, F.C.S. During the past year I have been largely occupied in the investigation of the flora of Denbighshire in its various aspects, and a vast amount of information has now accumulated : of this I hope shortly to give some account. Owing to this large ex- tension of my original undertaking, it has not been practicable to devote very much time to personal field work in Flintshire during this period, although a little has been accomplished. This, how- ever, has been largely compensated for by the kindness of various friends and correspondents, and, as a result, a considerable amount of new material has been acquired, and it may perhaps be as well to place some of this on record. The present contribution is in the nature of a supplementary county list, in the sense that the records quoted represent extensions of distribution, or in some instances additions to the county flora. In preparing this list I have only utilized a fraction of the material to hand, and so it must not be regarded as exhaustive, but rather as supplementary to papers on the suljject whicli I have contributed to this Journal (see Journ. Bot. 1907, 138; 1908, 187, 222; 1910, 40, 53, 90). Consequently the present contribution is entirely and designedly of a systematic character. Work dealing with local ecology, Welsh plant names and plant lore, floral biology, and other aspects of the subject, has occupied much time, but such is purposely excluded. These phases of the undertaking are reserved for detailed treatment in the Flora of Flint and Denbigh, which is in preparation. The present affords another pleasant opportunity of more formally expressing my indebtedness to several friends and con- tributors. To Miss F. M. Thomas and Dr. E. J. Haynes Thomas, of Chester, I am under special obligations for much valued help and co-operation. Mr. R. H, Day, as a resident observer, has rendered valuable aid in connection with the vegetation of the Cwm area, and in various other ways. Mrs. New (Backford) and Miss Cummings (Caerwys) are jointly responsible for an acceptable series of records of plants observed in the vicinity of Caerwys. Miss M. Jones, of Tan-yr-ywen, Llanfynydd, has also gone to some trouble to keep me supplied with observations and material FURTHER NOTES ON THE FLORA OP FLINTSHIRE 9 dealing with the flora of her district; Miss H. M.WiUiains (Aston) has supphed some useful notes chiefly on plants occurring in the neighbourhood of Shotton and Aston. Mr. W. Hodge, of North- wich, has contributed some very welcome observations, including notices of several species hitherto unrecorded for Flint. I have also to thank Mrs. Macdonald (Cwm), the Eev. J. Evans Jones and Mrs. Evans Jones (Dyserth), the Rev. W. Wright Mason (Bootle), Dr. W. B. Russell (Colwyn Bay), Mr. W. Whitwell, F.L.S., and Dr. J. W. Ellis, of Liverpool. Plants which are additions to the county hst are indicated by an asterisk. With few exceptions, the plants in the following Hst were observed in 1910. Clematis Vitalba L. Four Crosses, Meliden Road, near Dy- serth, Mrs. Evans Jones. Lane from Gronant, up hillside. Day. Cerrig Heilyn Lane, Dyserth, 1910. Hedge between Dyserth Castle and the Meliden Road. ■■'■Banunculus Baudotii Godr. Plashes, near the lighthouse, east side. Point of Aire, Hodge. — Var. '■■'confusns Godr. In a ditch by the road leading from Rhyl, west end, across the marsh to Rhuddlan, Hodge. — B. Lenormandi F. Schultz. Pond near Gwern-to, above Llanfynydd, Jones (sp.). Near Cwm, Day.— B. sardous Crantz. Seen in two stations in the neighbourhood of Rhyl, Hodge. — "'B. arvensis L. Several plants as a casual on the Cop near Saltney, Thomas (sp.). Hellehorus fcetidus L. Plentiful in Maes Mynan Woods, May, 1908, G. Loftus £ H. S. Marsh (photograph). Nant-y-Ffrith, only one plant seen, Thomas. Aquilcgia vulgaris L. A single plant seen near Cymmau, Ffrith, Thomas. Hedge-bank by road south of Newmarket, Day. Berber is vulgaris L. Plentiful about the ruins of Old Hawarden Castle, Thomas. Near Truly Farm, Babell, near Caerwys, Neio. ■'■NymiJiaa lutea L. In the pond belonging to Higher Kinner- ton Vicarage, Thomas. ■''Papaver Argemone L. Seen in two stations in the vicinity of Ehyl, Hodge. Corydalis claviculata DC. In a little wood on the edge of the meadows by Tan-yr-ywen, Llanfynydd, Jones (sp.). Alyssum maritimum L. The station-yard, Rhyl, evidently an outcast or escape, Hodge. ■■'Erophila prcBCOx DC. Steep limestone rocks in Nant-y-Firith in almost inaccessible situations. Only a few plants seen, Tliomas. Sisymbrium officinale Scop. var. ■•'leiocarpum DC. Several plants on the Cop at Saltney, in suspicious company, 1909. Helianthemum canum Baumg. Plentiful on the western side of the Gop, above Gop Farm, Newmarket, Day. A frequent and typical plant of the carboniferous limestone of the littoral portions of Flint and Denbigh. Lychnis Githago Scop. Corntield between Saughall and Queen's Ferry, Williams. 10 THE JOURNAL OP BOTANY Stellaria aquatica Scop. By- the Clwyd near Ehyllon, below St. Asaph. — S. nemorum L. Marshy meadow near Caerwys, Neiv. Sagtna apetala L. By the drive, Hawarden Park, Ellis. SiJergularia salina Presl. Eiver-bank between Shotton and Queen's Ferry, Williams. — S. margincUa Kittel. Dee shore between Sandycroft and Queen's Ferry. Hypericum montanum L. A solitary plant seen in lane between Marian mawr and Marian Mills, Day. Malva roUindifolia L. Koadside by the Common, Caerwys, Neio. In several stations near Ehyl, Hodge. Plentiful about Newmarket, Day. Geraniwn pratense L. In the quarry by the railway station, Caerwys, Hodge. Field between the railway and the Alyn, a third of a mile or so south of Khydymwyn Station. A few plants just behind the west platform of Khydymwyn Station. Erodium maritimum L'Herit. At the foot of the sandhills on the land side at Gronant, Hodge. ■'Oxalis corniculata L. Occurs in quantity in a semi- wild state as a garden weed in the grounds of Dyserth Vicarage. ■•'Impatiens hiflora Walt, Western end of the "H" Bridge, Bhyl, Hodge. Euonymus europxeus L. On the left-hand side of the road, about a hundred yards after passing the keeper's cottage, on the way to Tryddyn from Tan-yr-ywen, Jones (sp.). Cerrig Heilyn Lane, Dyserth, Day. Melilotus alba Desr. On the Cop near Saltney, Thomas. Trifolium medium L. Near Hawarden. — T. hyhridum L. The Cop, Thomas (sp.). — "T. filiforme L. Occasionally on the Ehyl Golf Links, Hodge. — T. striatum L. On the Common, Caerwys, New. Ornithopus perpusillus L. Hawarden Park, Williams. Vicia gemella Crantz. Plentiful on a bank by Ty-coch Farm below Cwm. — V. hirsuta Gray. Banks by the Aston Hall Colliery line, W"illia7ns. Between Buckley Junction Station and Kinnerton Station, Thomas. Eailway embankment near Caerwys, Hodge. Geum rivale L. Meadow by Tan-yr-ywen near Llanfynydd, Jones (sp.). On the right-hand side of the road from Tryddyn to Nerquis, immediately after passing Carreg Llech, Jones. Amongst alders in some swampy ground by a streamlet south of the railway, above Aston Hall Farm, Williams. Poterium officinale Hook. fil. Meadow between the Gop and Gwaenysgor, 1894, Herb. Day. Saxifraga granulata L. About twelve plants noticed by chil- dren from Llanfynydd School while looking for " nature study " specimens in a wood near Tryddyn village in May, 1909, Jones (sp.). A very rare plant in Flint and Denbigh. Chrysosplenium alternifolium L. Nant-y-Ffrith Woods, Thomas (sp.). ' On the Flintshire side of the Eiver Cegidog between Ffrith and Cefn-y-Bedd, Thomas. Cotyledon Umbilicus L. On the right-hand side of the little hill after passing the Old Smithy near Gwern-to, Jones (sp.). Wall PURTHEE NOTES ON THE FLORA OF FLINTSHIRE 11 in lane near Cefn-y-Bedd Station, TJiomas. Eocks in quarry by Ehuallt. '■•-Sedtmi Telephiuin L. Limestone quarry near Caerwys Station, Hodge. — S. album L. Disused quarry at the base of the carboni- ferous limestone range, on the side of the south-east road from Prestatyn, Miss Olive Franklin (sp.). — '''S. anglicum Huds. On a dry bank at the Ehyl end of the road leading from Prestatyn, out of Bastion Eoad, near the site of a burnt-down stackyard, Hodge. I sent several clumps to Mr. Day, of Cwm, from the neighbour- hood of Dolgelly, where this plant is very plentiful. At that time this species had neither been seen in nor recorded for Flintshire, although carefully sought for. With a view to naturalizing it in the county, these clumps were planted on an old wall on Mynydd Cwm (Cwm Mountain) immediately above and south of Bryn Goleu. These plants appeared to be thriving when I visited this locality in September. — S. reflexum L. Eoof and walls by Marian Mills, near Newmarket, Day. — S. rupestre L. Limestone rocks in Nant-y-Ffrith, Thomas. Occurs here at an elevation of about 900 feet. Myriophyllum aUerniflorum DC. In the plashes near the Point of Aire lighthouse, at the shore end of the road leading from Talacre Station, Hodge. Epilohium parvifloruvi Schreb. Seen in two stations near Ehyl, Hodge. By a little pond near the footpath between Plassau and Newmarket, Day. Bryonia dioica Jacq. Lane leading to the Dee Bridge at Shotton, Williams. Smyrnium Olusatrum L. Plentiful on roadside about a third of a mile from Meliden towards Prestatyn. '■'Carmn Petroselinum Benth. & Hook. f. Sandy bank on the Ehyl Golf Links, Hodge. MyrrJiis Odorata Scop. Plentiful in a field adjoining the inn at Cymmau, and extending some distance down the road towards Cefn-y-Bedd, Thomas. Scandix Pecten-Veneris L. Near Aston, one plant only, Williams. The Cop, near Saltney, Thomas. Cornfield near Ehyl, Hodge. Fmniculum vulgare Mill. Seen in three places in the vicinity of Ehyl, Hodge. CEnanthe fistulosa L. Very common on the marsh between the Golf Links, Ehyl, and the railway, Hodge. Pond in field by roadside about a mile east of Ehuddlan, towards Cwm. Silaus flavescens Bernh. In three places in fields on the farm called Ehydorddwy-wen, near Ehyl, Hodge. "'Cancalis daucoides L. Several plants as casual on the Cop at Saltney, Thomas (sp.). Valeriana dioica L. Near Llanfynydd, Jones (sp.). Swampy ground by a streamlet south of the railway, about Aston Hall Farm, Williams. Dipsacus sylvestris Huds. On Eliydorddwy-wen Farm, near Ehyl, Hodge. Quan-y behind the post-office, Ffrith, Hodge. On 12 THE JOURNAL OF BOTANY the embankment of a bridge which crosses the railway below Gronant. Antennaria margaritacea Benth. & Hook. fil. An odd plant near Buckley Station, Thomas (sp.). Inula Helenium L. In fair quantity in a field near Dyserth, Bev. J. Evans Jones (photograph). Bidens tripartita L. In a farm j)ond on the roadside, Town Ditch, Thomas.- — ''B. cermia L. Marshy place about half a mile from the shore at Gronant, Hodge. Anthemis Cotula L. and ''''A. arvensis L. In a field near the electrical generating station, Ehyl, Hodge. ■■'Matricaria suaveolens Buch. Very plentiful by roadside. Town Ditch, near Hope, and also opposite the Schools, Higher Kinnerton, Thomas. Waste land behind the Palace Hotel, Rhyl, west end, in Lake Avenue, Hodge. Plentiful in lane south of Warren Dingle and about three quarters of a mile north-east of Hope Exchange Station, Thomas. Near Buckley Station, Thomas. '^'Petasites fragrans Presl. Entrance to Dyserth Church ; a small patch to the left of the bridge over the stream. Senecio sylvaticus L. var. ■■'•aiiriculatus Smith. Hedgebank at the eastern end of the Rhyl Golf Course, Hodge. — S. viscosus L. In the quarry near the station at Caerwys, Hodge. Several plants amongst cinders on the south platform of Caerwys Station. — S. erucifoliiis L. Beyond the Grange Laundry near Rhyl, Hodge. '^'Arctium nemorositm Lej. Associated with Smyrnium Olusa- trum, near the front of Nant Hall Hotel, Prestatyn, Hodge. Carduus pycnocephalus L. Newmarket, Day. In disused quarry by roadside between Cwm and Rhuallt. — G. nutans L. The Gop, Day. — G. crispus L. Near Newmarket, Day. — Var. ■••acanthoidcs L. In the quarry near Caerwys Station, Hodge ; between the station and Caerwys, close to Cement Works, New. Serratula tinctoria L. Near Llyn Helyg, Day ; hedgebanks below Plass-yn-Cwm, extending for some distance towards Cwm, Day. Leontodon hirtus L. Pastures by the side of the road and close to the cottages at the Rhyl end of the marsh road to Ehuddlan, after crossing the railway behind the Marine Park, Rhyl, Hodge. Lactuca virosa L. A single plant growing on difficultly accessible limestone rocks near Dyserth. Gampamda latifoUa L. Very fine and plentiful in the quarry, Afon wen, near Caerwys Station, Hodge. Vaccinium Vitis-Idcea L. Pen-llun-y-gwr, just below the summit, Thomas. Lysimachia vulgaris L. By the edge of a pit close to a small wood at the southern end of Rhydorddwy-wen Farm, near Rhyl, Hodge. ■•'Anagallis fosmina Mill. In a meadow about half a mile east of Caerwys, Netv. Ligustrum vidgare L. Steep limestone rocks in Nant-y-Ffrith, Thomas. Clearly native here. FURTHER NOTES ON THE FLORA OF FLINTSHIRE 13 ■■'•Polemoniiim co&rideum L. Probably an escape, by the side of the stream at the ford, Cilcain, Hodge. Cynocjhssum officinale L. Near Carreg-y-ty, on the way to Cymmau, going up the lane past Llanfynydd Church, Jones. Be- tween Newmarket and the Gop, Day. Echium vulgare L. Eailway embankment, Llannerch-y-mor, Mason. Borago officinalis L. Near Prestatyn, Hodge. Litliospermum officinale L. Between Caerwys and the Station, Neio. Linaria viscida Moench. Plentiful on the railway embank- ment about fifty yards to the west of Caerwys Station, Hodge. Veronica hybrida L. I am glad to find that this still occurs in the county, as I was afraid it had long been extinct. First observed by the Eev. W. Bingley in 1801, it does not appear to have been seen in Flintshire by any subsequent observer until this year. It was found by the Eev. J. Evans Jones, in some quantity, on carboniferous limestone, and I was pleased to discover it in a second Flint station last August.- — V. AnagaUis-aqnatica L. var. -'anagallidiformis Bor. With the type in the marsh ditches between the Ehyl Golf Links and the railway, Hodge. '^'Euphrasia gracilis Fr. Ehyl Golf Course, Hodge. — E. cnrta Wettst. Craig Fawr. Bartsia Odontites Huds. var. '■'■serotina Dum. Ehyl neigh- bourhood, Hodge. Between Cwm and Ehuddlan. Verbena officinalis L. The Common, Caerwys, New. Mentha Pulegiwn L. Between Cwm and Ehuddlan, Herb. Day. '■'■'• Calamintha iMrviflora Lam. There are Flintshire examples of date 1842 in Miss Potts's herbarium at Chester. Nepeta Cataria L. Near Ehyl, Hodge. Galeopsis speciosa Mill. Near Ehyl, and also in the vicinity of Caerwys, Hodge. Leonurus Cardiaca L. In one place on the coast road between Ehyl and Prestatyn, Hodge. Lamium hybridum Vill. Eoadside between Aston and Queen's Ferry, Williams (sp.). Eoadside between Cymmau and Ffrith, Thomas. Two or three plants near Cwm, Day. — L. album L. Wall near Tryddyn, Jones (sp.). Plantago lanceolata L. var. spharostachya Eoelh. The Gop, Day. Polygonum Bistorta L. Entrance to Hafod Abley Wood near Llanfynydd, and also in a field on the south side of the wood, Jones (sp.). Euphorbia exigua L. The Leet, Williams. In a corn-field about a quarter of a mile south of Ehyl Cemetery, on the farm called Ehydorddwy-wen, Hodge. Eailway track at Dyserth Station, Day. Humulus Lupulus L. By the stream south of Newmarket, Day. By a little pond near the footpath lietween Plassau and Pen-y-cefn-isaf, between Cwm and Newmarket, Day. Hedge close to Aston Hall Farm. Taxus baccata L. In fair amount and clearly native on lime- stone rocks and elsewhere in Nant-y-Ffrith, Thomas. A farm not 14 THE JOURNAL OP BOTANY far away is known as Tan-yr-ywen (=: Under the yew tree). The yew also finds mention in another Fhnt place-name in Moel Plas- Yw ( = The bald hill of the hall of the yews), an elevation in the Clwydian range. Hellehorine latifoUa Druce. In the wood at Caerwys, Hodge. — Var, media E. S. Marshall. In the woodland at the commence- ment of the Leet, near Coed-du, Thomas. Orchis pyramidalis L. Near Caerwys, Neio. Ophrys apifera Huds. In the neighbourhood of Dyserth, Bev. J. Evans Jones. Habcnaria bifolia E. Br. In a wood near Caerwys, New. Koadside near Cae-glas near Llanfynydd, Jones (sp.). The Leet, Thomas (sp.). By the road between Cwm and Llyn Helyg, Day. — H. albida Br. Near Pant Asaph, Neio. AsjMvagus officinalis L. a maritimus Mill. Near the shore at Prestatyn, Bev. W. W. Mason (sp.). Between Gronant and Point of Aire, Hodge. *ConvaUaria majalis L. Seen wild in one place in May, 1910, by Miss Cummings and Mrs. New. Juncus compressus Jacq. Prestatyn, growing amongst J. Ger- ardi, Maso7i (sp.). Lemna trisulca L. Pond in field near Plas-yn-Cwm, Day. Alisma ranunculoides L. Common in the ditches on the marsh land lying between the Rhyl Golf Links and the railway, Hodge. ■■'•Potamogeton pralongus Wulf. In the Clwyd in quantity, north-east of St. Asaph, Day. — P. crispus L. Ditches on the marsh between the Rhyl Golf Links and the railway, Hodge. Zannichelliapedunculata Reichb. In a ditch by the side of the path leading across the marsh to Rhuddlan from Rhyl, by the side of the Clwyd, about a quarter of a mile from the railway crossing at the back of the Marine Park, Rhyl, Hodge. Marsh below Shotwick and Puddington. ■'■Eriophorum vaginatum L. Gwern Mountain, Thomas. Carex divisa Huds. The Cop, Thomas (sp.). — C vulpina L. Lane between Newmarket Road and Marian Mills, Day. — C. remota L. Near Cwm, July, 1897, Herb. Day. — G. sylvatica Huds. In the lane leading from Caerwys, station past Afon wen, up to the village of Caerwys, Hodge. — C. acutiformis Ehrh. By the side of a ditch on the eastern side of -Rhydorddwy-wen Farm near Rhyl, Hodge. — G. axillaris Good. In damp meadows near the " Sheepskin Factory," Caerwys, Nero. Avena fatua L. var. ■■'pilosissima Gray. On the waste land on the north side of the Marine Park, Rhyl, Hodge. Phragmites communis Trin. '•'fi nigricans Gren. & Godr. Inland of Rhyl and in ditches near Prestatyn, Hodge. Glyceria plicata Tr. In a damp place by the side of the lane (about half way up) leading from Caerwys Station to the village, Hodge. Morfa Rhuddlan. — G. distans Wahl. Prestatyn, Mason. Bromus unioloides H. B. K. The Cop, Thomas (sp.). Hordeum murinum L. Along the Ochr-y-Foel Road, Dyserth, just above the station. 15 A NEW SECTION OF LONCHOCARPUS. By S. T. Dunn, B A., F.L.S. Among the type examples of Millettia Thonningii (Baker in Oliver, Fl. Trop. Afr. ii. 128) we find specimens of two African trees with few-paired pinnate leaves and racemes of handsome flowers, one, however, having its racemes ebracteate, pedunculate, short and loose, wliile in the second they are long, compact, and floriferous to their densely bracteate base. The names under which they were originally described, vi/. (1) liohinia TJwnningii Schum. & Thonn., and (2) Millettia Griffoniana Baill., are in- cluded among the synonyms {loc. cit.). Baillon's tree is easily recognizable by its long crowded tail- like racemes, and is represented by a remarkably complete series of specimens collected all down the coast of tropical West x^frica, from Sierra Leone to Angola, and there is no doubt that it is not only specifically but generically distinct from the other. Even over this large area, indeed, it maintains so close a uniformity of characters that it cannot conveniently be divided into varieties, though it shows a tendency in the southern portion of its range to produce flowers with slenderer pedicels and smaller bracts upon a thinner rachis. To this form belongs the Kamerun tree, which Dr. Harms has described as Derris leptorJiacliis. There is no doubt whatever that this and Baillon's tree belong to the same species, and that with it must be associated a second species dis- covered by Kitson in S. Nigeria, but to what genus they should be referred was an open question until fruit was found ; Harms was as well justified in referring his tree to Derris as Baillon his to Millettia. Fruit has, however, now been collected in the Kamerun forests, and, being indehiscent and wingless, necessitates the removal of these species from Derris and Millettia to Lonclio- carj^us, with which their turbinate calyx also agrees. They differ, however, so markedly from any species of that genus known to me that it appears convenient to keep them together under the sectional name of Gaudaria, a designation suggested by the tail- like appearance of their racemes. Lonchocarpus § Caudaeia sect. nov. Flores in rachidibus longis pendulis sessilibus nodoso-racemosi. 1. L. Griffonianus Dunn, nom. nov. Millettia Griffoniana Baill. Adans. vi. (1866) 222 ; M. Thonnincjii Baker in Oliver, Fl. Trop. Afr. ii. (1871) 128 ; Derris leptorhachis Harms in Engl. Jahrb. xxvi. (1899) -302. West Trop. Africa. Sierra Leone, Barter, n. 1623, 3265 ; Gold Coast, W. H. Johnson, n. 456 ; Lagos, Moloney, n. 21, Millen, n. 129 ; Nigeria, Old Calabar, Mann, n. 2282, Chevalier, n. 5012 ; Kamerun, Preuss, n. 1155 ; Batanga, Bates, n. 318 ; Galloon, Soyaux, n. 101, Bates, n. 487 ; Congo, Hens, n. 335, Christian Smith ; Angola, Wehvitsch, n. 1860-2. 16 THE JOURNAL OP BOTANY 2. L. multifolius Dunn, sp. n. Arbor magna, ramis exigue tomentosis. Folia 12-15 cm. longa, 7-juga, stipellata ; foliola oblonga, acuminata, basi obliqua, rotunclata, 4-5 cm. longa, char- tacea, supra glabra opaca, subtus prsecipue in costa tomentosa. Racemi axillares ad 25 cm. longi, fere ad basin fioriferi ; floras 8-10 mm. longi, nodoso-racemosi, nodis ad 5 mm. longis, pedi- cellis 2-3 mm. longis, ut rachidibus tenuiter tomentosis, apice bracteolis 2 subulatis ; calyx campanulatus, 3-4 mm. longus sub- glaber, dentibus 4, tubi quartam partem aequantibus, ovatis ; petala glabra ; vexillum sine ungui rotundatum, basi breviter cordatum ; stamen vexillare medio connatum ; ovarium tenuiter sericeum multi-ovulatum. West Trop. Africa. S. Nigeria, Kitson, March 21, 1909 (Brit. Mus. Herb.). This species is easily distinguished from L. Grijfoniamis by its more numerous leaflets. SPIB^A ULMARIA L. var. DENUDATA Boenn. By a. R. Horwood. Mr. Druce does me an honour in singling out my six-year-old note on this plant." / also " hold no brief for (or against) clenn- data," but when one finds that one and the same plant is made up of typical Spircea Ulmaria, the so-called variety denudata, and a leaf of an intermediate type, it is reasonable to say that denudata has no title to rank as a variety, and to assume that it is merely a state, unstable in the same specimen, and of no more value than lupus in man, and like that due to pathological causes. Of what real value, to take a similar case, is the variety Pryorii of Pajjavei- Bhoeas ? In the herbarium the red colour of the hairs of the peduncle, its chief distinction, is lost. Varieties based on whether some portion of the plant is hairy or glabrous throughout are far from satisfactory, e. g. Sisymbrium officinale var. leiocarpum, Acer campestre var. leiocarpon, and the like. In the nomenclature of Lichens far more names rank as " subspecies " and " forms " than as varieties, and this seems much more sensible than adhering to the higher rank, in the last case, of variety. For evolution exhibits three modes of progress. We have heredity, which tends to preserve species in their original form, and to perpetuate the type. The active agent of evolution is variation, which tends to multiply its stock by the production of variations. These last, as subdivisions of a former species, heredity tries to maintain without further alteration, whilst the antagonistic factor of variation tends to multiply them afresh. Then Mendel's Law gives us a tendency in some to go back to their original form, while a simultaneous effort is made to produce the same or fresh forms. In this Scylla and Charybdis of conserva- tion and progress a state of instability may be created ; some * See .Journ. Bot. 1910, 281. MARINE ALG.E FROM THE KERMADECS 17 forms are evolved or produced which, hy their purely physiological origin, are evolved alone by physiological stimulus, and are not perpetuated, or their origin may be pathological. Whilst a variety may be defined as a variation of some important morphological or structural significance, which persists alongside of a type from which it was manifestly derived, the status of a " form," "state," &c., must be regarded as inferior to those differences in individual species which are called varieties. This is because permanent variations are useful adaptations, whilst forms or states are merely ephemeral reactions to stimuli called forth by a dis- organized state, an abnormal environment, or the effect of ad- jacent species. The ultimate test of species, variety, or form is found only by cultivation for several generations, and this in the present instance does not appear to have been done. The greater plasticity of plants (compared with animals) renders them more liable to become abnormal, and open to the attack of other plants. This is well illustrated in the case of microscopic fungi, e. g. rusts, smuts, where the condition they induce is closely analogous to the white coloration of the under surface of Spiraa Ulmaria. It may be due perhaps to the plant's inability to manufacture its food-substances, and to answer fully to the stimulus of chlorophyllization, which is stimulated exter- nally and produces an inner derangement. To sum up, species are the biological units once differentiated by variation, which goes on complicating original differences, so that permanent varieties occur, and at the same time plants are open to detrimental effects of environment and resulting reaction, often impermanent, as here. These must not be confused with permanent intermediate forms obtained by hybridization and crossing, which remain constant, and are produced by definite interbreeding. When species, variety (so-called), and an intermediate state are commingled in the same unit, it is reasonable to suggest that the two last are of no permanence, and in this case that the denudata form of Spircsa Ulmaria is a malformation. MAEINE ALG^ FEOM THE KEEMADECS. By a. & E. S. Gepp. The following is a list of marine algge collected by Mr. Eeginald B. Oliver, of H.M. Customs, Christchurch, New Zealand, during his visit to the Kermadec Islands in 1908. An account of the vegetation of the islands is published by him in Trans. New Zealand Institute, vol. xhi. 1909, pp. 118-175, in which he gives a history of the past botanical investigation of the Kermadecs, a description of tlieir Geology, Climate, Introduced Animals and Plants, Plant Formations, Geographical Distribution, kc. The following facts are largely derived from his paper. The Kermadec Islands are four — Sunday Island or Eaoul Journal of Botany. — Vol. 49. [Jan. 1911.] c 18 THE JOURNAL OF BOTANY Island, Macauley Island, Citrtis Island, and French Keck — and are situated half-way along the submarine ridge which connects New Zealand with the Tonga group. They are volcanic and composed of tufas. Sunday Island is the largest of the group, and its coast is rocky save for the gravelly beach in Denham Bay on the south-west side, and some sand on the north shore. These islands lie far away from any land from which they could have derived their flora and fauna ; and hence their great interest to students of geographical distribution. The preponderance of the flora is of the New Zealand type, bringing the islands into the New Zealand biological region. Dr. Cockayne has made the Kermadecs a separate province of that region ; but Mr. Oliver endeavours to show that they unite with Lord Howe Island and Norfolk Island to form a natural division, for which he proposes the name " subtropical islands province." The marine algae collected by Mr. Oliver were sent to Mr. E. M. Laing, who very kindly invited us to identify them. He intends to publish a list of them himself in New Zealand, but suggests that in the meantime we should give the names in this country. As regards the geographical distribution of the species, it will be seen that none of them is endemic unless it be the Galaxaura, but that the majority are found in New Zealand. Mr. Oliver {loc. cit. p. 156), writing on "Dispersal," says: "The nature of the material cast up on the shores of Sunday Island points to the south or south-west as the direction from which the strongest and most frequent [ocean currents] reach the group. Several kauri (Agathis aiistraUs Salisb.) logs, some bearing brands of New Zealand firms, are lying on the shores of Sunday Island, and on the north coast is a balk of Oregon pine, supposed to have formed part of the cargo of the ' Elingamite,' wrecked on the Three Kings Islands in 1903. Again, during the months of July to October, 1908, when strong westerly winds prevailed, a large amount of seaweed was cast up in Denham Bay. ... It is evident that a strong ocean current flows from New Zealand in a north-easterly direction ; and this in my opinion is sufficient to account for the preponderance of New Zealand forms in the flora of the Kermadec Islands." A study of the charted ocean currents of the South Pacific confirms this view, and shows at once that the Kermadecs are in a very interesting position. A. branch of the great South Equa- torial Current sweeps down to the south-west past Samoa, Fiji, and the Friendly Islands. It runs on past Norfolk Island and Lord Howe Island, and skirts down the coast of the Continent of Australia (East Australian Current). Then colliding with the cold subantarctic water of the strong West Wind Drift, it is deflected, and now runs up north-east past the islands of New Zealand, and strikes across to the Kermadecs. Here it joins another branch of the South Equatorial Current, which has swept down south from the Friendly Islands. The conjoined currents pass away to the south-east. Thus the Kermadec group is washed MARINE ALGiE FROM THE KERMADECS 19 by oceanic currents from the north and from the south-west, the latter being the stronger of the two. The result of this upon the algae is shown by the fact that the majority of the species received belong to the New Zealand flora, and at least one {Peyssonnellia) to the flora of the Friendly Islands. The Kermadecs were visited by H.M.S. 'Herald' in 1854 under Captain H. M. Denham, whose name is perpetuated in the chart of Sunday Island. On board were the naturalists Milne and MacGillivray, who made a small collection of plants, which were described by Sir Joseph Hooker (Journ. Linn. Soc. Bot. i. p. 125 (1857)). The few algse included in the collection are in Kew Herbarium, but apparently were never published. No other algae appear to have been gathered on the shores of the Kermadecs until Mr. Oliver made his long stay on Sunday Island in 1908. However, Dickie, in his report on the algae of the ' Challenger ' Expedition, records Cystophora eloncjata as found floating in the deep sea off Raoul (or Sunday) Island. We make further refer- ence to this under Caryophyllum elongatum. Mr. Cotton, of Kew Herbarium, has also received lately some marine algae from the Kermadecs ; and it is with profit that we have compared the two sets and exchanged views with him. Ulva Lactuca Linn., nos. 1334, 1340. " On rocks between tidal marks, Meyer Island." Geogr. Distr. General. Ulva latevirens Aresch., no. 1341. Cast up on Denham Bay Beach. Geogr. Distr. South Australia. Tasmania. Enteromorpha compressa Grev., nos. 1332, 1333. " On rocks between tidal marks, Meyer Island." Geogr. Distr. N. and S. Atlantic. N. and S. Pacific. Cladop}iora fusca Martens, nos. 1318, 1348. " In rock pools, Meyer Island." We have little hesitation in referring our plants to C. fusca, though we have had no opportunity of seeing an authoritative example of it. Martens' description is a little puzzHng in respect of the branches, which are stated to be twice as thick above as at base — a character which does not appear in the figure. Our plants agree with Martens' figures, but attain larger dimensions, our biggest specimen being three inches high. In Kew Herbarium are other specimens from the Kermadec Islands, whicli were collected during the visit of H.M.S. ' Herald ' in 1854. They bear the name CladopJiora prolifera, but whether they were ever published we do not know. The range of the species extends to Ceylon, where it was collected by Ferguson. His specimens (no. 86) are represented in tlie British Museum Herbarium. The type-specimens were collected at Mampawa, on tlie west coast of Borneo, and at Palal)uan. Geogr. Distr. Borneo. Vauchcria sp., no. 1308. Growing epiphytically. Meyer Island. c 2 20 THE JOURNAL OP BOTANY Ganlerim raceviosa var. uvifera Weber v. Bosse, forma inter- viedia W. v. B. Meyer Island. Geogr. Distr. Indian and Pacific Oceans. West Indies. Codium tonientostim Stackhouse, no. 1313. Meyer Island. Geogr. Distr. General. DurviUcea sp. Cast up on Denham Bay Beach. (Not sent.) Hormosira Banhsii Decaisne, no. 1344. Cast up on Denham Bay Beach, Sunday Island. (Not sent.) Geogr. Distr. Australia. New Zealand. Tasmania. Carpophyllum maschalocarpum Grev., no. 1349. "Probably not growing locally; in di'ift-weed only." (Not sent to us.) Geogr. Distr. New Zealand. Carpophyllum elongatum (comb. nov.). Syn. Cystophora elongata Dickie in Journ. Linn. Soc. xv. (1876), p. 241. CariJophylluvi angustifolium J. G. Agardh, Lund Univ. Acta, xiv. Math. No. iv. p. 8 (1877). Though this species was not gathered by Mr. Oliver, it was found by the 'Challenger' Expedition "floating freely on the surface in small bunches, inhabited by oceanic animals, thirty miles distant from Raoul Island, Kermadecks, July 14, 1874. Vesicles large, inch to 1|- inch long, elliptical subacute, and with copious efflorescence of mannite " [Dickie MS. in herb.]. It is represented in both the British Museum and Kew Herbarium, and it appears to be identical with the CmyojyJiyllum angustifolium collected by S. Berggren in New Zealand and described by J. G. Agardh. We have not seen Agardh's type, but we find Dickie's specimens to be exactly like that figured by Mr. R. M. Laing in Trans. Proc. New Zealand Institute, xxxii. p. 67, pi. v. fig. 1 (1899), which was found by him in the Bay of Islands, " apparently from the same tidal pools from which Berggren obtained it." Mr. Laing redescribes the species {loc. cit.), and what he says of his plant applies to the 'Challenger' specimens. These latter may be presumed to have drifted across to the Kermadec group from the north end of New Zealand. Geogr. Distr. New Zealand. Carpophyllum Phyllanthtis Hook, et Harv., no. 1350. Drift, Denham Bay Beach, Sunday Island. Geogr. Distr. New Zealand. Carpoijlujllum plumosum J. Ag., no. 1359. Cast up on Denham Bay Beach, Sunday Island. Mr. Oliver says : " This species, as well as C. maschalocarjmm, DurviUcea sp., and Hormosira Banksii, do not, I think, live in Sunday Island waters, but are brought by ocean currents. They are found washed up on the beaches, and usually have barnacles attached to them." Geogr. Distr. New Zealand. Taonia australasica J. Ag., no. 1317. Meyer Island. Syn. Spatoglossum cuneatum J. Ag. ex J. B. Wilson in Proc. Roy. Soc. Victoria, iv. p. 162 (1892). MARINE ALG^ FROM THE KERMADECS 21 Mr. J. Bracebridge Wilson collected this species on Jan. 23, 1886, at Port Phillip Heads, as the specimen in the British Museum shows. In his catalogue of algae collected at or near Port Phillip Heads and Western Port, he gave it J. G. Agardh's MS. name, Spatoglossum ciineatum. Agardh described the plant in his Ana- lecta Algologica Gont. i. p. 30 (1903), under the other name. Mr. J. B. Wilson's rich herbarium of Victorian algas is in the British Museum ; and it may be added that an interesting biographical notice of him and a portrait were published by Dr. J. H. Maiden iu the Victorian Naturalist, xxv. pp. 116, 117 (1908). Geogr. Distr. Victoria. Gymnosonts nigrescens J. Ag., nos. 1362, 1365, 1374. Cast up on Denham Bay Beach. Geogr. Distr. North, South, and West Australia. Dictyota p-olificans A. & E. S. Gepp, nos. 1314, 1323. Meyer Island. Geogr. Distr. New South Wales and Queensland. Sargassum fissifolium J. Ag., no. 1367, on rocks near low-water mark, Meyer Island ; no. 1368, on rocks between tides, Fleetwood Bluff, Sunday Island; no. 1369, cast up on Denham Bay Beach, Sunday Island. Mr. Oliver says : " This is the most abundant seaweed found growing on the rocks of Sunday and Meyer Islands. It does not grow above the level of low-water mark, except on rocks in exposed places, where they are continually washed by the waves, and in rock-pools near the south side of a cliff". Its distri- bution is thus determined directly by the amount of heat received from the sun." Geogr. Distr. Queensland. Chantransia sp., no. 1361. Without fruit. Epiphytic on Gynmosorus nigrescens, Denham Bay Beach. Galaxaura sp., no. 1351a. Kocks near low-water mark, Meyer Island. The same species was collected during the visit of H.M.S. 'Herald' in 1854, and is placed under G. lapidescens in Kew Herbarium. To what modern species these plants should be referred we do not attempt to decide until we can make an adequate study of the entire genus. The large number of new species created by Kjellman in his monograph of Galaxaura have vastly increased the difficulty of identifying specimens. Zanardinia marginata J. Ag. ( = Brachycladia marginata Schm.), no. 1355. Sunday Island. Geogr. Distr. North and South Atlantic. North and South Pacific. Indian Ocean. Gelidium longij^es J. Ag. " Densely tufted. On rocks between tidal marks, Nov. 1908." This specimen agrees in habit and structure with a specimen from New Zealand in the British Museum Herbarium collected by Berggren, and named by J. G. Agardh Gelidium longipes. The Kermadec plants differ only in being shorter and in not being so freely branclied, nor somewhat compressed at the apices, as in Berggren 's plant. However, the 22 THE JOURNAL OF BOTANY Kermadec specimens appear to be young and still growing. We find in one a tetrasporic branch, but as a rule the thallus has the appearance of being not yet mature. Geogr. Distr. New Zealand. Pterocladia capillacea Born., nos. 1310, 1352. Growing epi- phytically, Meyer Island. " Called turtle moss by the inhabitants, because eaten by the turtles." Sunday Island. Geogr. Distr. Atlantic, Mediterranean, Cape of Good Hope, Indian Ocean, China, Japan, Australasia. Gracilaria confervoides J. Ag., no. 1358 (pro parte). Cast up on Denham Bay Beach, Sunday Island. Geogr. Distr. Atlantic, Pacific, and Indian Oceans. Laurencia Forsteri J. Ag. ? no. 1358 (pro parte). Cast up on Denham Bay Beach, Sunday Island. Geogr. Distr. South and West Australia. Plocamium hrachiocarpimi Kiitz., nos. 1335, 1336, 1337, 1339. " Cast up on Denham Bay Beach, Sunday Island, Oct. 1908." Geogr. Distr. New Zealand. Martensia elegants Her., no. 1327. On rocks near low- water mark, Coral Bay, Sunday Island. Geogr. Distr. South Africa. West Australia. New South Wales. Nitophyllmn dec2imbens J. Ag., no. 1363. On Carpophyllum plumosum, cast up on Denham Bay Beach. Geogr. Distr. New Zealand. Delisea pulchra Mont., nos. 1338, 1353. Cast up on Denham Bay Beach, Sunday Island. "Intestines of a turtle, twenty yards in length, full of this plant in all stages of digestion. A little of the same species was found growing on the outside of the neck of the turtle." Geogr. Distr. Eastern Australia. Tasmania. New Zealand. Asparagopsis Sanfordiana Harv., no. 1330. Meyer Island. The specimens are very rotten, but appear to be altogether refer- able to this species. Hitherto it has only been recorded from Western Australia, but in Kew Herbarium is a specimen from Keppel Bay, Queensland. Geogr. Distr. Western Australia and Queensland. Euzoniella incisa Falk. {^= Polyzonia incisa J. Ag.), nos. 1311, 1319. Meyer Island. Geogr. Distr. South-west Australia. Tasmania. New Zealand. Spongoclonium Broivnianum De Toni {CalUthamnion Browni- anum Harv.), no. 1326. On rocks near low water. Boat Cove, Sunday Island. Geogr. Distr. West Australia. Peyssonnellia rubra J. Ag., no. 1354. Eock-pools, Sunday Island. This is the same as no. 39 of Harvey's Friendly Islands Algce. Geogr. Distr. Adriatic. Polynesia. Melobesia sp., no. 1322. Meyer Island. PLANTS OF THE AZORES 23 Amphiroa anceps Decne., no. 1346. On rocks below low-water mark, Meyer Island, May 19, 1908. Geogr. Distr. Norfolk Island. Cheilosporitm elegans Aresch., no. 1347. On rocks below low- water mark, Meyer Island. Geogr. Distr. New Zealand. New South Wales. Gorallina officinalis L., nos. 1320, 1357. Rock-pools, Meyer Island ; also Sunday Island. " Forms a dense coating up to an inch high on rocks from low tide to about half-tide mark, and in rock-pools." Geogr. Distr. General. Gorallina sp., no. 1329. Sunday Island. Gorallina Giwieri Lamx., nos. 1342, 1343. Cast up on Denham Bay Beach, Sunday Island. Geogr. Distr. South Australia. Tasmania. PLANTS OF THE AZORES. By G. Claridge Druce, M.A., F.L.S. In the early March of 1909 I made a short stay in San Miguel, the chief island of the Azores, which I reached after a tempestuous passage from Madeira. The islands are about 700 miles from Portugal, 1150 from Britain, and 1700 from Newfoundland. The country is well cultivated, maize being the chief corn crop, but potatoes, sweet-potatoes, and other vegetables are grown, and there is a considerable extent of pasturage on which many cows are reared. Sheep, too, are raised not only for their flesh and wool, but are also used as draught animals. The people are honest, industrious, and kindly. The climate, although some- what damp and relaxing, is very equable, frost being unknown at the sea-level. The rainfall is only about twenty-nine inches, but there are boisterous winds, which sweep the place with great violence. Tropical, or perhaps more correctly subtropical, vegetation flourishes, and some of the Quintas abound with a most varied collection of trees and flowering shrubs brought from almost all parts of the world. Many adventitious plants abound, and indeed give a key-note to the flora. Pineapple culture has now almost replaced the growth of the St. Michael oranges. The scenery is somewhat marred by the high lava-walls, built to shelter the gardens, and they stretch so far from the town of Ponta Delgada that it is difficult to obtain a view, or to get free from them into open country. The Azores are practically volcanic cones, San Miguel being entirely volcanic, so that calcareous species are absent. In the Crater of Fundas, with its extremely interesting geysers, there is very beautiful scenery, the high walls of the crater being covered with vegetation, including not only native species, but the planted Japanese Cryptomeria and the Australian wattle, besides the 24 THE JOURNAL OF BOTANY naturalized Indian Hedychium. From a lake a small stream of clear water, bordered by white Japanese azaleas, flows by the hot springs, where its muddy meadows are full of the tara [Colocasia antiquorum) ; the hedgerows leading to the Fundas and in other parts of the island being formed of Hydrangea, with its magnificent show of pink or bluish blossoms. The Sete Cidades is also a strikingly romantic place, and the summit of the hill commands aii unusually interesting and charm- ing view, not only of the beautiful coast-line and the great crater with its sapphire and its emerald lake, but of the perfectly formed subsidiary crater, now a wealth of vegetation. The botany of the island has been w^ell explored by Dr. Carreiro, to whose kindness I am much indebted, both during my visit and subsequently for sending to me a hundred of the typical island species. The Azores are interesting to British botanists also from the fact that H. C. Watson visited them and prepared a valuable and interesting list of the plants which was published in F. du Cane Godman's Natural History of the Azores, in 1870. The most recent work on the island flora is that by Prof. Trelease, which appeared in the Eighth Annual Beport of the Missouri Botanical Garden (1897). To the fact of my visit being made so early in the year is pro- bably due the discovery of five new species of mosses, one being new to science. It was pleasant to meet with the very local endemic species Campanula Vidalii on the main island of San Miguel, as well as to add two or three other species to the flora. I am indebted to Prof. Hackel, Mr. H. N. Dixon, and to Mr. S. Macvicar for naming the grasses, mosses, and hepatics respectively. Papaver Bhceas L. var. Pryorii Druce. Ponta Delgada ; possibly introduced. I saw it also at Funchal, in Madeira. Cardamine caldeiranum Guth. var. amplifolitmi Trelease. This interesting endemic plant, which, although a distinct species, is somewhat intermediate between G. amara and G. sylvatica, was seen at the Fundas. Watson found that it maintained itself "self- sown and quasi-spontaneously " in his Surrey garden. Sisymbrium officinale Scop. var. leiocarpon (Jord.). This is, as Trelease observes, the prevailing form ; noticed also in Madeira. Gerastium viscoswu L. The common form has flowers with distinctly larger petals than our British plant. Spergula arvensis L. The form with papillose seed alone noticed, i. e. S. vulgaris Boenn. Hypericum humifusum L. Near Lomba da Cruz. — H. foliosum Alton. On the ascent to Sete Cidades. Lowe (Fl. Mad.) says this is distinct from the Madeii'an grandiflorum, therefore it is probably endemic. Oxalis purpurea Jacq. Well naturalized at Ponta Delgada, &c. Vicia atropurpurea Desf. A form of this handsome species was seen westward of Ponta Delgada. Trifolium angustifolium L. In two or three places on lava- walls about Ponta Delgada. Not given for San Miguel by Trelease. PLANTS OF THE AZORES 25 Lupinus tenuis Forsk. North of Ponta Delgacla, doubtless as a relic of fodder culture. Not given by Trelease. Ornithojnts compressus Druce. In the roadway in the Furnas. Buhus rusticanus Merc. Appears to be the prevailing bramble throughout the island. Fracjaria indica L. Naturalized north of Ponta Delgada ; I also saw it by the roadside near Turin, in Italy. — F. vcsca L. Just the British plant at Kibiera Grande. PotentiUa erecta Hampe var. ins igni s (Domin, in lit.). P.Tor- mentilla var. insignis Domin in Fedde, Nov. Sp. v. 66, 1908: — " Caules robusti elato-erecti, superne tantum pauci-ramosi 25- 35 cm. alti, dense adpresso-pilosi ; folia magna foliohs sessihbus coriaceis (intermedia 2-3 cm. longo) obovato-cuneatis, utrinque 4-6-dentibus lanceolatis acutis instructis, adultis quoque supra et subtus ad nervos valde prominulos longe adpresso-pilosis ; stipulge oblique obovato-rotundatge 2 cm. et ultra long® et inter- dum latiores, inciso serrate, flores ca. 15 mm. late tetrameri ; calycis adpresso-pilosi, sepala externa internis paulo breviora ; petala diam_etro transversiali latiora." — K. Domin, in lit. San Miguel, F. D. Godman in Herb. Kew. 1865. Wolf, Mon. Gatt. PotentiUa, p. 647, 1908. I saw this interesting plant in several places in the crater of the Fundas ; it has a robust bushy growth, quite erect, and has four to five petals. I should think it is a distinct species, bearing the same relation to our own P. erecta d^ndi procumhens as the Azorean Gardamine does to our C. aniara and sylvatica ; it might appropriately bear the name P. insignis (Domin) suggested by that botanist. Tillaa muscosa L. Trelease observes that it had not been recently collected, and that excellent local botanist my friend Dr. Carreiro also told me he had not seen it. I found it on damp roadsides in the Fundas, on the lava-ridges by the track to the Sete Cidades, and in the Quinta of Dom Jose do Canto at Ponta Delgada. Lythrum Hyssopifolia L. Fundas. Fuchsia macrostemma Euiz & Pavon. Naturalized on walls in several places, as Trelease observes. — F. Bicartoni Hort. is naturalized to the west of Ponta Delgada. Hedera canariensis Willd. var. azorica. Between Kibiera Grande and Fundas, and on the ascent to Sete Cidades. Sherardia arvensis L. var. maritima Griseb. Ponta Delgada ; the type only is mentioned by Trelease. Viburnum Tinus L. var. subcordatum Trelease. In some quantity near Lomba da Cruz. Watson recognized the difference of the leaf-characters from those of V. Tinus. LeycesteriaformosaWnil. Completely naturalized in the Fundas. Bubia angustifolia L. Sete Cidades, &c. ; quite distinct, I think, from our B. peregrina, an opinion stated also by H. C. Watson. Kentranthus Sibthorpii Heldr. & Sart. Not mentioned by Watson. A narrow-leaved plant, identified as above by Trelease, naturalized at Ponta Delgada, &c. Erigcron canadense L. This ul)iquitous alien species has found 26 THE JOURNAL OF BOTANY its way into San Miguel. — E. mucronatus DC. is also naturalized in many places, as at the Furnas. — E. linifolius Willd. Ponta Delgada. Gnaphalium luteoalbumLi. and G-imrpureiim L. Ponta Delgada. Bidens inlosa L. Naturalized at Ponta Delgada, and likely, as in Madeira, to become a pest. Petasites frag rails Presl. Near Ponta Delgada. Senecio malvcefolius DC. Rare ; on cliffs at Capellas. — S. viikanioides Otto. This African species I saw on the site of an old garden west of Ponta Delgada. It is quite naturalized in Madeira. Urospermum picroides Schmidt. On waste ground at Ponta Delgada ; Trelease says " not recently collected." Campanula Vidalii Watson. This very interesting and local species was discovered by Captain Vidal, who picked it on an in- sulated rock off the coast of Flores. Watson was unable to find it on the mainland. Hunt, according to Watson, was afterwards more successful in finding it locally on the coasts of Santa Maria and San Miguel, whence its introduction to English gardens is said to be due. Trelease, however, states that it " occurs on cliffs and detritus by the sea-shore and on outlying rocks round the entire island of Flores," and goes on to say that, as regards Santa Maria and San Miguel, " the impression there and in Terceira is that it occurs in cultivation only, and was originally derived from Flores." It is therefore pleasing to be able to add a station where it is indubitably native on San Miguel. This is at Capellas, a fishing-town of about three thousand people, nearly nine miles from Ponta Delgada. In the narrow bay is a considerable whale fishery, which gives the town its prosperity. The bay is surrounded on three sides by very steep cliffs, especially on the southern side. In one a path has been cut in the solid lava, by which access is given from the top to the sea-level. A wall protecting from the sea is thus formed ; on this I climbed to see what vegetation grew on the seaward side, where the cliff is vertical. - A plant, which I took to be a Euplwrhia, grew temptingly out of reach, but I managed to reach a few of the leaves and a flower-stalk of the preceding year, and then recognized this extremely local species. Dr. Carreiro has since seen the plant there, and I believe supports my view of its being indigenous in a station where it is not likely to be eradicated. The plant is full of an extremely viscid milky juice, which can be drawn out in long threads. It grows well from seed in England. Vaccmium cylindraceum Sm. Lameiro, &c. An endemic species. Erica azorica Hochst. With other brushwood this species is being rapidly destroyed, owing to the culture of pineapples. The brushwood is cut into small pieces and spread about nine inches thick in the glass-houses, then covered with some soil : into this compost the pots of pineapples are placed, and the plants ripen without other artificial heat than that induced by the fermentation of this vegetable substance. PLANTS OF THE AZORES 27 Lysimachia azorica Hornem. Very closely allied to our own L. nemorum, with which indeed it is merged in Index Keicensis. Watson considered it distinct ; an analogous case to that of Potentilla insignis. I saw it sparingly in the Sete Cidades. Myrsine africana L. var. retusa DC. Near the Fundas, Lagoa, &c. ; an interesting species with copious berries covered with bloom, looking temptingly eatable, but when eaten producing un- pleasant eifects. Vinca difformis Pour. This beautiful species was a great adornment to the delightful woods of the Fundas. Centaurium lyulchellum Druce. In the Fundas, but not quite typical, being laxer, and with slightly larger and paler flowers ; perhaps worth separating as var. azoricmn. New to the Azores. Tecoma radicans Juss. This North American species is natu- ralized to the west of Ponta Delgada. Linaria Cymhalaria Mill. Ponta Delgada ; Fundas. Antirrhinum Orontium L. Frequent, but, as in Madeira, with paler flowers than our British plant. Vero7iica mregrina L. This North American species is natura- lized in garden ground at Ponta Delgada and by the roadway in the Fundas. — V, officinalis L. Ponta Delgada. Linaria Cymhalaria Mill. San Miguel, new to that isle. Mentha Pulegium L. Near Povoacao. Marrubium vulgare L. Near Ponta Delgada. Ballota nigra L. " Not recently collected," Trelease. I saw it in Ponta Delgada. Plantago lanceolata L. var. eriophylla B.-Webb & Berth. Ponta Delgada ; Lomba da Cruz. Alternanthera Achyranthes Br. Ponta Delgada; Villa Franco da Campo. Polygonum hydropiperoides Michx. Fundas. Che7iopodium miirale L. Ponta Delgada. Euphorbia azorica Hochst. Ponta Delgada. Buxus sempervirens L. A narrow-leaved plant in the Fundas ; probably planted. Salix cinerea L. Probably planted, but not given in Tre- lease's list. Persea azorica Seubert. Sete Cidades. Mercurialis ambigua L. f. Ponta Delgada. Ficus repens Kottl. This Indian species was naturalized in the Fundas, by the stream-side near the park of Viscount Porto Formoso. Hedychium Gardnerianum Rose. This Indian species is com- pletely naturalized in the Fundas, forming in many places the chief undergrowth in the woods. Iris foetidissima L. Near Villa Franca. Amaryllis Belladonna L. Naturalized in a iew places. Canna indica L. Naturalized west of Ponta Delgada. Arum italicum Mill. Fundas ; Ponta Delgada. Richardia africana Kunth. Now abundantly and completely naturalized by stream-sides in several places. 28 THE JOURNAL OF BOTANY Colocasia antiquorum Schott. Naturalized in the Funclas, where it is extensively cultivated in moist ground by the streams near the geysers. From the tuberous roots, "Tara," a mucila- ginous fecula, is obtained, and largely used as an article of diet. Garex macrostyla Lapeyr. var. 'peregrina L. H. Bailey. Furnas. — G. Paircei F. Schultz. Dr. Carreiro had kindly sent this from St. George Island under the name G. muricata. New to the Azores. Festuca Myurus L. New to San Miguel. I saw it on' some wall-tops in Ponta Delgada ; it is not quite typical, and, as Pro- fessor Hackel says, shows an approach to F. Broteri. Dicksonia GulcUaLi'Iler. Furnas; common and very beautiful. Adiantum Gapilliis-veneris L. About the hot springs of the Furnas. Asplenium Hemio7iitis Li. Sete Cidades. — A. monantJinm Li. Furnas. — A. lanceolatum Huds. Ponta Delgada, &c. — A. marinum L. Capellas. Wooclwardia radicans Sm. Fundas ; abundant. Blechnum Spicant Eoth. Fundas. Dryopteris aristata Druce. Very luxuriant in the Fundas ; Sete Cidades. — D.paleacea Druce. Sete Cidades. — D. mollis (Sw.). Sete Cidades. Gymnogramme Tetta Schl. Sete Cidades. Hymenophyllitm unilaterale Bory. Fundas. — H. tunhrigensc Sm. Sete Cidades. Osmunda regalis L. Fundas. Selaginella Kraussiana Br. Fundas ; Sete Cidades ; north of Ponta Delgada. Lycopodium cernuum L. Fundas ; a most beautiful species. Sometimes growing on concretionary rock deposited by the hot springs. — L. Selago L. var. suberectum Baker. Fundas. Equisetuvi ramosissimum Desf. In the Fundas. I gathered at Fundas Targiona liypophylla L. and Gousinia marcliantioides Raddi, which appear to be additions to the Azores. The few mosses I collected included one new to science, Bryum clavcUulitm Cardot &, Dixon, besides five additional to the Azores, i. e. Tivnniella Barhula Limpr., Barbula gracilis var. viridis B. & S., Tortilla muralis Hedw., Barbula vinealis Brid., and Bryum mitrale Wils. ; these, with other species gathered by me, have already been recorded by Mr. H. N. Dixon in Journ.-Bot. 1909, 372. THE GENUS DIASTELLA. By E. p. Phillips, M.A. The genus Diastella was founded in 1809 in Knight's ProteecB, p. 61, and defined as follows: — "Diastella Salisb. Flores in Capitulo plus minus truncate terminali. Bractea 1 inter singulos, praeter Involucrum gemmacearum, membranacese. Petala vix irregularia, post anthesin apice erecta, basi tantum cohtierentia. Pericarpium ut in Leucadendro. Stylus barbatus, tandem de- THE GENUS DIASTELLA 29 ciduus. Frutices, saepius procumbentes. Folia simplicia, in pluribus viridia, rarissime ad apicem 3-dentata." Seven species were described, viz. D. bryiflora, D. serpyllifolia, D. vacciniifolia, D. imrilis, D. myrtifolia, D. ericafolia, D. humifusa. In 1810 (Trans. Linn. Soc. x. pp. 15-226) appeared Robert Brown's famous essay on " The Proteaceae of Jussieu," but no mention is made of the genus Diastella, the species composing it being regarded by- Brown as belonging to Mivietes — not, however, to the original Mimetes of Salisbury (Parad. Lond. sub n. 67), but to Mimetes in the wider sense which Brown himself gave to it. If we now turn to Brown's account of Mimetes, we find that he subdivided the genus into two sections : (1) Caiyitula axillaria, (2) Capitula terminalia, Mimetes spuria, and I have found by a comparison of Salisbury's specimens (at Kew) with Brown's types at the British Museum that these two sections correspond respec- tively to the genexa, Mimetes Salisb. and Diastella Knight, thus : — Mimetes E. Br. Sect. i. Capitula axillaria = Mimetes Salisb. Sect. ii. Capitula terminalia. Mimetes spuria = Diastella Knight. Sprengel (Syst. Yeg. i. 464) is the first author who mentions the genus Diastella, but he merely states that Diastella bryiflora Knight = Leucospermum puberum R. Br. Roemer & Schultes (Syst, Veg. iii. 266) and Steudel (Nomencl. Bot.) make the same statement, which by a comparison of the two herbaria proves to be incorrect ; what we really find is that D. bryiflora Knight = Mimetes thymelaoides R. Br. The next important work bearing on our subject is Endlicher's Genera Plantarum, where the author (Suppl. iv. 2, p. 78) makes Diastella Knight a section of Leucospermum R. Br., but strangely enough keeps the plants to which Knight applied the name Diastella under the genus Mimetes R. Br. In his account of Mimetes, Endlicher subdivides the genus into two sections, Eumimetes and Pseud oviimetes, which correspond exactly to Brown's sections. On comparison with the result already arrived at from Brown's account of Mimetes, the following table becomes clear : — Mimetes R. Br. Sect. i. Eumimetes Endl. = capitula axillaria R. Br. — Mimetes Salisb. Sect. ii. Pseudomimetes Endl. = capitula terminalia R. Br. = Diastella Knight. How Endlicher came to apply the name Diastella to a section of Leucospermum I do not know. Meisner (DC. Prod. xiv. pp. 259, 262, 264) follows Endlicher, and adopts the same names for the subdivisions of Leucospermum and Mimetes, as also does Engler (Pflanzenfam. iii. i. 135, 137) ; Bentham & Hooker (Gen. PL), while keeping Diastella as a synonym of Leucospermum R. Br., state that some species of the former genus belong to Mimetes R. Br. sect. Pseudomimetes Endl., thus : — 30 THE JOURNAL OF BOTANY Leucospermum R. Br. = Uiastella Knight (partly). Mimetes R. Br. Sect. i. Eumimetes Encll. = Mimetes Salisb. Sect. ii. Pseudomimetes Endl. = Diastella Knight (partly). The Index Kewensis takes us no further than Bentham and Hooker, as the following reductions given there will show : — DIASTELLA Knight = Leucospeemum R. Br. D. bryifora Kn. = L. puberum R. Br. D. ericmfolia Kn. = M. purpurea R. Br. D. Jmmifusa Kn. D. myrtifolia Kn. = M. myrtifolia R. Br. D. parilis Kn. = L. parile R. Br. D. serpyllifolia Kn. D. vacciniifolia Kn. = M. vacciniifolia (no authority). After a careful comparison of Knight's types of Diastella with Brown's Mimetes I arrive at the following results : — Diastella Kn. = Mimetes, sect. ii. R. Br. D. bryiflora Kn. = M. thymelcsoides R. Br. D. ericaefolia Kn. = M. purpurea R. Br. D. humifusa Kn. (no specimen seen). D. myrtifolia Kn. = M. myrtifolia R. Br., var. /?. D. parilis Kn. = M. myrtifolia R. Br., var. a. D, serpyllifolia Kn. = M. divaricata R. Br. D. vacciniifolia Kn. = M. divaricata R. Br. All the authors dealt with above agree in subdividing Mimetes into two very defined sections (Mimetes Salisb. and DiasteZia Knight), but these sections differ so much in habit, structure, &c., that I have split Brown's Mimetes into two genera, and have kept up the names published by Knight. GENERIC differences. Mimetes Salisb. Habit Erect plants, rarely de- cumbent. Inflorescence ... Capitula3-13-flowered, partly hidden in the axils of the upper leaves, aggregated at the end of the bran- ches ; capitula oblong. Involucre-bracts Membranous or cori- aceous, usually showy and coloured. Style H-^i in. long, glab- rous. Stigma f-3 lin. long, cylindric, linear, nodulose, or sinuate at the junction with the style (rarely gradually tapering in- to the style), some- times swollen above. Diastella Knight. Prostrate shrubs, rarely erect. Capitula many-flow- ered, terminal, soli- tary, rarely 3-nate ; capitula subglobose, often truncate. Somewhat leathery, rarely conspicuous. 2f -4f lin. long, hairy below. |— J lin. long, subcla- vate, gradually ta- pering into the style. LEICESTERSHIRE PLANTS 31 The principal references for the genera are as follow : — DiASTELLA Salisb. apud Knight, Prot. 61 (1809). Mimetes sect. ii. E. Br. in Trans. Linn. Soc. x. 109 (1810). Mivietes sect. Pseudomimetes Endl. Gen. PI. Suppl. iv. 2, 78 ; Meisn. in DC. Prodr. xiv. 261 ; Engler, Pflan- zenfam. iii. i. 135. non Leucospermum sect. Diastella Endl. 1. c. ; Meisn. op. cit. 269. nee Leucospermum (partim) Benth. & Hook. Gen. PI. iii. 170; Engl. op. cit. 137. Mimetes Salisb. Parad. sub n. 67 (1807). Mimetes sect. i. R. Br. 1. c. Mivietes sect. Eumiinetes Endl. I.e.; Meisn. op. cit. 262; Benth. & Hook. op. cit. 171 ; Engl. op. cit. 135. LEICESTERSHIRE PLANTS (1905-1910). By a. R. Horwood. Some six years have elapsed ■■'• since any additional notes upon the distribution of flowering plants in Leicestershire have ap- peared ; and as the Elora of 1886 is now very out-of-date, and during this interval much of interest has accumulated, I publish these records as a further contribution to the county flora. The results here set forth are the joint work of several botanists and others whose names appear below ; but especial mention should be made of the share taken by the Rev. H. P. Reader in this recent work. He has been the constant companion (as well as botanical guide) of the writer on excursions made with the object of survey- ing the county anew botanically, and the following notes are, except where the initials of other workers are cited, our joint work. An asterisk denotes a new record ; a dagger stands for plants of alien origin. In one or two works that have been overlooked in compiling the Flora, some records of interest occur that we have incorporated with the recent notes. In one of these by the Rev. Irwin Eller,f the author states that a number of plants were " planted " in the Belvoir district, e. g. Acorns Calamus, Muston ; Silene nutans, Stathern; Aster Tripolium,NQ\e of Belvoir; Inula Pulicaria, Mus- ton ; Lepidium latifolium, Muston ; Thlaspi arvense, Stathern ; Geranium lucidum, Stathern ; Leonurus Cardiaca, Stathern ; Trifolium subterraneum, Muston ; Medicago maculata, Muston ; Myrrhis Odorata, Belvoir. Some of these are found, it is true, as casuals. Others are not native, and we should be quite at a loss to account for their occurrence in these stations had no such state- ment been published. As the book is rare we have cited the cases alluded to here, so that others may be on their guard. • See Journ. Bot. 1904, pp. 337-349 ; 1906, pp. 261-266. t History of Belvoir Castle, 1841, Appendix. 32 THE JOURNAL OF BOTANY Of a similar kind is a nuisance which may be of a more wide- spread natm-e. In this part of the country, at any rate, it is the custom of herbalists to travel (they are often railway guards) far afield in search of plants used medicinally, and to root them up and bring them home, and plant them in or near their houses, e. g. on railway banks. This has a twofold artificial effect : (a) plants are redistributed by man ; (b) laid open to extermination in their original habitats. I have not seen this practice described before, and it may serve to explain some present anomalies of distribution. Owing to one cause or another the following plants, moreover, have disappeared from v.-c. 55 since 1886 : — Arabis Jdrsuta, Dianthus deltoides, Sagina nodosa, Hypericum elodes, Trifolmm glomeratum, Lathyrus palustris, Drosera rotundifolia, D. anglica, Siitm latifoUum, CEnanthe silaifoUa, Antennaria dioica, Inula Puli- caria, Anthemis nobilis, Senecio campestris, Arnoseris pusilla, Crepis palitdosa, HypochcBvis glabra, Vaccinium Vitis-idaa, AnagalUs tenella, Gentiana Pneumonanthe, Limosella aquatica, Orobanche major, 0. purpurea, Pinguicula vulgaris, Mentha piperita, M. gen- tilis, Calammtha Nepeta, Littorella juncea, Polygonum minus, Bumex p)ulcher, Aristolochia Clematitis, Empetrum nigrum, Sjn- ranthes autumnalis, Orchis ustulata (? an error), Habenaria bifolia, Buscus aculeatus, Tofieldia palustris, Luzula Forsteri, Scirpus pauciflorus, S. ccBspitosus, Bynchospora alba, Schoenus 7iigricans, Carex dioica, G. filiformis, Asplenium Adiantum-nigrumr A. Tri- cJiomanes,'-'' A. viride, Cystopteris fragilis, Lastraa Thelypteris, Osmunda regalis, Lycopodium Selago, L. inundatum, L. clavatum, Pilularia globulifera. These are mentioned here in case any of them are still known to occur in localities in Leicestershire unknown to us, and as a supplement to the list of Cryptogams cited as extinct here since 1886.t W. B. = W. Bell.— L. S. B. = L. S. Biggs (Leicester Museum). — J. B. = J. Bradshaw.— B. E. C. R. = Botanical Exchange Club Reports. — E. F. C. = E. F. Cooper. — I. E. = The Eev. Irwin Eller.— T. R. G. = T. R. Goddard (Leicester Museum).— A. R. H. = Author's initials.— C. B. H. = 0. B. Headley.— J. E. J. = J. E. Jordan. — F. L. F. K. = Mrs. F. L. Foord-Kelcey. — E. E. L. = E. E. Lowe (Leicester Museum).— G. M. = G. Mercer.— H. P. R. = Rev. H. P. Reader. — T. E. R. = T. E. Routh. — N. H. T. = Natural History of Tutbury, Sir Oswald Mosley, 1863, Appendix by E. Brown.— W. A. V. = W. A. Vice.— W. B. E. C. = Watson Botanical Exchange Club Reports. The order followed is that of the tenth edition of the London Catalogue. Thanks are due to the Rev. W. Moyle Rogers, E. F. Linton, Prof. E. Hackel, Messrs. J. E. Bagnall, Arthur Bennett, H. and J. Groves, and J. W. White for naming several of the more critical species. * Found where probably they have been planted recently, but not in truly native habitats. t See Journ. Bot. August, 1907. LEICESTERSHIRE PLANTS 33 Clematis Vitalha L. Martinshaw Wood, C. B. H. (W. B. E. C. 1906). Certainly not indigenous here, nor apparently anywhere in the county. Anemone nemorosa L. Cotesbach. ■ — \A. ranunculoides L. Piper's Hole, Croxton Kerrial, /. E. Doubtless an escape. ■■-\ Adonis annua L. (= autumnalis L.). Moira (N. H. T.). Eving- ton, /. B. Abbey Park, F. Bussell. Banunculus circinatus Sibth. Moira (N. H. T.). — B. tricho- jjhyllus Chaix. Scraptoft. — B. Drouetii F. Schultz. RatcUffe, Scraptoft, Stonton Wyville. — B. heteroijhyllus Weber. Brackle- bridge (W. B. E. C. 1907). — i?. peltatus Schrank. Six Hills, Cropston Eeservoir, terrestrial form, Lowesby (W.B. E. C. 1906). Hallaton. — Var. ■Hruncatus (Hiern). East side of Groby Pool, A. B. H. — Var. florihundus (Bab.). Stoney Stanton, W. B. (W. B. E. C. 1906). — Var. ijenicillatus (Hiern). Blaby Brook, W. A. V. (W. B. E. C. 1905).— i?. Flammula L. Form approaching b. radicans Nolte. Cropston Eeservoir. — B. acris L. c. Borceanus Jord. Leicester, C. B. H. (W. B. E. C. 1906). Barkby, W. B. (W. B. E. C. 1908). —i?. sardous Crantz. Aylestone, L. S. B.— B. Ficaria L. var. incumhens F. Schultz. Ratclift'e. Also a stoloni- ferous form with sinuous leaf margin. A dwarf form of the type was found down by the Great Stretton Brook. Galtlia iKilustris L. b. Guerangerii (Bor.). Botcheston, L. P. S. Aylestone, Gracedieu, H. P. B. ■''\Erantliis hyemalis Salisb. South Croxton, Long Spinney, Scraptoft. Berberis vulgaris L. Moira (N. H. T.). Blaston. Castalia alba Wood. Market Harborough. Papaver Bhoeas L. var. Pryorii Druce. Lowesby. — P. Lecoqii Lamotte. Aylestone, L. S. B. Stony Stanton, Medbourne. iGlattcimn j^lioiniceum Crantz. Belgrave, G. M. '^j-Neckera bulbosa. Knighton, W. B. Wanlip. iCorydalis lutea DC. Walton. Fumaria officinalis L. Form with retuse fruit. Knighton, W. B. (W. B. E. C. 1907).— -li^. densiflora DC. Rothley Plain, established W. B. '■'Badicula Nasturtium-aquaiicum R. & B. var. siifolia R. & B. Scraptoft, A.B.H. Arabis glabra Bernh. Woodville (N. H. T.). Cardamine amara L. Botcheston Bog. Gracedieu, W. B. — C. flexuosa With. Newtown Unthank, Launde, Gaddesby, Halla- ton, Medbourne. Great Glenn, T. B. G. Cropstone, Glenfield, L. S. B. Hoby, W. B. (W. B. E. C. 1910). Erophila verna E. Meyer. Croft, W. B. (W. B. E. C. 1907). Ratclift'e. — b. maj use ula (Jord.). — c. s/cnocar^ja (Jord.). Croft, W. B. (W. B. E. C. 1907).—^. pracox DC. Blaby, W. A. V. (W. B. E. C. 1905), with unusual leaf characters. Croft Quarry, W. B. (W. B. E. C. 1907). [Cochlearia Armoracia L. Medbourne, Market Harborough, on the Leicester road. Birstall, G. M. Hesperis matronalis L. Blaston. Journal of Botany. — Vol. 49. [Jan. 1911. j d 34 THE JOURNAL OF BOTANY Sisymbrium Thalianwn Gay. Buddon Wood, Bagworth, Wan- lip, Kirby Muxloe. — -S. officinale Scop. var. leiocarpum DC. New Humberstone. — ■■-\S. Irio L. New Burton and Ashby Eoad (N. H. T.). Erysimum cheiranthoides L. Thurmaston, Aylestone, L.S.B. Ingarsby, Spinney Hill Park, N"^w Humberstone. ■■HEucUdium syriacum. BU .y Mill, W. A. V. (W. B. E. C. 1905). Introduced in foreign corn. Camelina sativa Crantz. Elmesthorpe. — b. fcetida (Fr.). Aylestone. Brassica Napus L. Tilton, Medbourne, Stony Stanton, Old Humberstone, Branston, Hallaton. — B. rapa L. Aylestone, L. S. B. — B. nigra Koch. Stony Stanton. — B. alba Boiss. Desford Station. Diplotaxis muralis DC. b. Babingtonii Syme. Market Har- borough. Coronopus procumbens Gilib. Overseal (N. H. T.). New Hum- berstone, Scraptoft. This is one of the most long-lived of plants when placed in water in a vase. A specimen exhibited on the flower-table of the Leicester Museum lasted three months, water being changed weekly. Lepidium ruderale L. West Bridge, Thurnby, Aylestone, New Humberstone, Foston. — '''flv. sativum L. Woodhouse Eaves, H. P. B. Aylestone, L. S. B.—\L. Draba L. New Humberstone, West Bridge, Desford. Thlaspi arvense L. Crown Hills, /. E. J. Great Easton. BapJianus Baphanistrum L. b. sulphureum. Billesdon Coplow. Aylestone. Beseda lutea L. Between Saltby and Sproxton, H. P. B. Bel- grave Birstall, G. M. Groby, L. S. B. Saxby, B. Barnes. — B. Luteola L. West Bridge, Birstall, G. M. Saxby, B. Barnes. Viola odorata L. x V. hirta L. (= F. sepincola Jord.). Wymond- ham, G. B. H.— V. hirta L. Breedon Cloud Wood (N.^H. T.). Katcliffe, Tilton Hill, Baggrave Blaston, Hallaton, Wymondham, G. B. H.^V. hirta x odorata (= V.permixta Jord.). — ■•'¥. calcarea Greg. Near Ingarsby Tunnel, 1906. — V. sylvestris Kit. Burbage Wood, Thurlaston.— F. canina L. Bardon Hill, W. B. (W. B. E. C. 1907). Not typical. — V. arvensis Murr. a. agrestis Jord. Stoughton, South Knighton, W. B. (B. E. C. K. 1908). — f. segetalis (Jord.). Syston, A. B. H. Lutterworth, W. B. (B. E. C. E. 1908).— b. ob- tusifolia (Jord.). Forest East (W. B. E. C. 1910). — c. Llotjdii (Jord.). Abbey Park (B. E. C. E. 1908). Blackbird Lane, W. B. (B. E. C. E. 1908). Of the first Dr. Drabble says " it may be a large-flowered form, due to rich soil." Of the second {ibid.) he says "not typical." — e. ■•'subtilis (Jord.). Branston, Saltby, A.B.H. —arvatica (Jord.). Knighton, W. B. (W. B. E. C. 1910).— TimSaZi (Jord.). Oadby, W. B. (W. B. E. C. 1907). Dr. Drabble calls this a small and delicate state. — Kitaibeliana Eoem. & Schultz. Oadby, W. B. Polygala serpyllacea Weihe. Baggrave, Burbage Common. Eoecliffe, W. B. (W. B. E. C. 1906). \Sa2)onaria officinalis L. Aylestone. Birstall, G. M. LEICESTERSHIRE PLANTS 35 Silene latifolia E. & B. . Tilton, Aylestone, L. S. B. Birstail, G. M. Branston, Sixlthy.—h.jmbemla (Jord.). Waltham, F. L. F. K. (B. B. C. R. 1907).— 5'. noctiflora L. Moira (N. H. T.). Branston, Saltby, Arnesby, Goadby Marwood. — \S.dicliotoma Ehrh. Eving- ton, /. B. Lychnis dioica L. fl,. albo. T'lton, L. S. B. King's Norton, Marefield. In l)otanical district -. fJ of the county (as defined in tire Flora of 1886), or between Rugby and Lutterworth, this species does not occur, a fact which must be correlated doubtless with the absence also of Mercurialis lyerennis and the scarcity of Arum macidatum. — L. Githago Scop. Kirby Muxloe, Desford, L. S. B. Evington, /. B. South Croxton, Ragdale. Cerastium semidecandrum L. Breedon (N. H. T.). On old walls and dry places about Stathern, I. E. — C. arvense L. Swith- land, Lowesby. Stellaria apetala JJcrisb. Kirby Muxloe. — S. neglecta Weihe. Narborough Bog, L. S. B. — b. nmbrosa (Opiz). Botcheston, between Newtown Unthank and Kirby Muxloe. — c. decipiens E. S. Marsh. Newtown Unthank. Cropstone Reservoir, L. S. B. Nar- borough, C. B. H. The tubercles rather rounded, and in this respect, according to Mr. C. E. Salmon, agreeing with a plant from S. Croxton referred to S. neglecta. — S. uUginosa Murr. Seal Wood (N. H. T.). Desford, Baggrave, Aylestone, L. S. B. -'-\Spergula sativa Boenn. Ashby Parva, 1910, H. P. B. Spergularia rubra Pers. Windmill Hill, Woodhouse Eaves. Hypericum maculatum Crantz. Whitwick (N. H. T.). Ulvers- croft. — H. humifusuni L. Billesdon Coplow, Ashby Parva. — H. pulchrum L. Newtown Linford. Malva moschata L. Harby, Belvoir, /. E. Sanvey Castle, Lowesby Hill, Aylestone, New Humberstone. Cropstone, T. B. G. Mountsorrel, E. E. L. This plant is apparently especially charac- teristic of the marlstone formation and sandy soils generally. — M. rotundifolia L. Walton. Packington (N. H. T.j.— fill, pusilla Sm. Ratby. Tilia cordata Mill. Swithland Wood, W. B. (B. E. C. R. 1907). \Linum usitatissimuvi L. Amongst corn, Eastwell, I. E. Ayle- stone, New Humberstone, Old Humberstone, Glen Parva. ^Geraniuvi plicBum L. Aylestone, L. S. B. — G. pyrenaicum Burm. fil. R. Soar, Thurmaston, G. M. Railway banks, near Glen Parva, W. A. V.~G. pusilhim L. Overseal (N. H. T.). Wal- ton, Medbourne, Branston. — G. lucidum L. Wanlip. Castle Donington, L. (& P. S. — G. Bobertianum L. fl. alba. Stoughton, Tilton, Burbage. Erodium cicutariwn L'Herit. Branston. Birstail, G. M. Oxalis Acetosella L. Croxton Park, I. E. Launde, Blaston. Scalford. Bhanmus Frangula L. Woolsthorpe, I. E. Acer campestre L. b. leiocarpon Wallr. Lowesby. Knighton Road, W. B. (W. B. E. C. 1907). Genista tinctoria L. Croft, Burbage, Stathern, Saltby. Cytisus scoparius Link. Seal Wood (N. H. T.). 36 THE JOURNAL OF BOTANY Ononis spinosa L. Keyham. A white-flowered form occurs here. — 0. repens L. Billesdon Coplow, Salthy. -\MecUcago sativa L. New Humberstone, Marefield, Syston, Laughton. Mountsorrel, E. E. L. Between Edmonthorpe and Wymondham, B. Barnes. — M. dcnticulata Willd. Birstall, G. M. — M. arabica Huds. East side of Cliarnwood Forest (N. H. T.). Belgrave, G. M. Wanhp, G. M. Melilotus altissima Thuill. West Bridge, Belgrave Pumping Station, Aylestone, Glen Parva, Church Langton. Smeeton, T.B. G. — M. officinalis Lam. (= M. arvensis Wallr.). Evington, J. B. Belgrave, G. M. New Humberstone. Trifoliitm medium L. Moira (N. H. T.). Burbage, Keyham, Stathern, Branstone. Aylestone, L. S. B. — ■■'jT. ochroleucon Huds. Mountsorrel, E. E. L. — T. arvense L. Birstall, G. M. Mountsorrel, E. E. L. — T. striatum L. Tilton Hill, Oropstone, Sproxton. — \T. hybridum L. South Croxton, Tilton, Branstone. — T.fragiferum L. Houghton, King's Norton, E. Langton, Arnesby, Sproxton. AnthylUs Vulneraria L. Laughton Hill, Wigston Station. Mountsorrel, E. E. L. Lotus tenuis Waldst. & Kit. Woodville (N. H. T.). Near Birstall. Onobrychis vicicefolia Scop. Mountsorrel, E. E. L. Between Edmonthorpe and Wymondham, B. Barnes. (To be continued.) SHOBT NOTES. New County Kecords in Bryophyta. — The following Mosses and Hepatics are new for the counties of Westmorland and Ayr respectively. I have been kindly helped by Messrs. Wheldon and Ingham in naming them. In June, 1910, I found, in Mardale, v.-c. 69, Meesia trichoides Spruce, in fruit, and BracJiythecium 2)lumosii7it, B. & S. var. nomomallum Schimp. In Eiggindale, Bazzania tricrenata (Wahl.), Trevis. In August, 1910, I found, at Craig Lure, Ayrshire, v.-c. 75, Sphagnum compactum De Cand. var. subsquarrosum W., S. latifoUa. It is, however, a good species, as the old plants show always thin roots, smaller, not very rough leaves, few, not very patent branches, longer, redder flowers and narrower, brown seeds. S. latifoUa has, as is well-known, broad, almost black seeds, leaves one foot long and very rough, and ' fathom-long ' (fadenlange) 1-3 inch thick roots. The leaves of both species are always dry and never exude salt like the leaves of S. tomentella, to which, when they dry, white salt frequently adheres." Becker evidently considered his plant as related to S. latifoUa rather than S. tomentella, but after a careful examination of a number of authentic specimens of S. sareptana, named by Becker himself, and a comparison with numerous plants of tomentella (some verified by Becker) and latifoUa, I have come to the con- clusion that we must either consider it distinct from both or class it as a variety of L. tomentellum. Nyman (Conspec. Fl. Europ. 609, 1881) adopts the latter course, calling it, however, a sub- species, and Gandoger (Nov. Conspec. Fl. Europ. 396, 1910) follows this arrangement. In the examination above referred to I was much struck by the amount of variation as regards hairiness of scape, leaves and bracts. In some examples the bracts were practically glabrous, in others distinctly hairy, and even on the same individual, variations occurred. This hairiness of bract surely indicates an affinity with L. tomentellum, as the bracts of S. latifoUa are quite glabrous. For the present, I prefer to regard it as a variety of L. tomen- NOTES ON LIMONIUM 75 tellum, distinguished from that, usually at a glance, by the more- branched scape with numerous sterile branches and laxer spikes, and, at a closer inspection, by being much less tomentose on the scape, by the almost glabrous bracts (the outer one more herba- ceous also), and by the acute calyx-lobes. As far as I have been able to judge from dried specimens, the character mentioned by Becker regarding the presence of salt on the leaves holds good, and is yet another distinction. From S. latifolia the plant is easily separated by its larger calyx (with often intermediate teeth), more glabrous scape, less rough leaves, more herbaceous outer bract, &c. It has more than once crossed my mind that S. sareptana might be the result of L. tomentellwn, with its decidedly hairy bracts and calyx and rounded calyx-lobes, crossing with S. lati- folia, which has glabrous bracts, less hairy calyx and acute calyx- lobes. A careful examination of the living plants in their native surroundings would probably soon decide this point, which, if proved, would explain in a great measure the very varied facies which the plant presents. I have been unable to find any figures of L. tomentellum that can be called satisfactory, hence the plate now given. Trautvetter (Z. c. 357) asserts that the S. Gmelini of Reichb. Icon. Crit. 237 is his S. tomentella f. trachycauUs, but as that figure indicates a plant with a very lax spike and acute calyx-teeth, it would seem to fit var. sareptanum much better and to this I should assign it. The S. scoimria Pall, of Reichb. Icon. Grit. 236 is indicated by Trautvetter as his glabella form of tomentella, but I think, in spite of the rather large liowers shown on the plate, it really represents L. Gmelinii as noted in Journ. Bot. 1909, 287, and has nothing to do with any form of L. tomentellum. The synonymy, description and distribution of Limonium tomentellum and its variety are as follows : — Limonium tomentellum 0. Kuntze, Rev. Gen. PL ii. 396 (1891). Statice tomentella Boiss. ! in DC. Prodr. xii. 645 (1848) ; Ledeb. Fl. Ross. iii. pars 1, 461 (1847-9). S. Gmelini Bieb. Fl. Taur. Gauc. i. 250 (1808); Roem. & Schult. Syst. Veg. vi. 778 (1820) ; non Willd. S. Gmelini W. S tomentella Trautv. f. trachycauUs Trautv. in Melang. Biol. Bull. Acad. St. Petersb. 357 (1856). S elata in hb. DG. ! non Fisch. Exsicc. — F. Schultz, herb. norm. nov. ser. Gent. 9, 896 ! Becker, PI. Wolg. infer. 44 ! Planta plus minusve hisjndula ; folia plus minusve hispidula, tuberculis albidis incomposite adspersa ; scapus saepius supra medium ramosus ; rami inferiores steriles nulli aut raw pauci ; spicae breves et densijioro}, raro laxifiorae ; bractea exterior fere omnino mevibranacea, plus minusve hisjndula, raro fere glabra ; bractea interior quam bractea exterior vix duplo longior, plus minusve hispidula, raro fere glabra; calyx crassus, copiosc hirsutus aut velutinus, raro fere glaber ; calycis lobi brevissiuii triangularo- rotundati aut acutiusculi, dentibus intermediis interpositis. g2 76 THE JOURNAL OF BOTANY Plant 25-55 cm. high, more or less finely hairy. Leaves more or less finely hairy (with occasionally stellate hairs) on both sides, rarely almost glabrous, pinnately veined, broadly-or-oblong-obovate or ovate, blunt or more acute, often emarginate, usually mucro- nate, normally rather shortly petioled, showing white tubercles when dry. Scape more or less finely hairy (with usually some stellate hairs), rarely almost glabrous, erect, slightly flexuose, branched usually from above the middle, sterile branches absent or very few. Branches and hranclilets ascending or ascending- patent and recurved. Scales triangular-acuminate, usually finely hairy, never foliaceous. Spikes patent usually scorpioid, short and rather dense-flowered, rarely longer and lax-flowered. Spike- lets 1-2 flowered, sometimes with an additional abortive one. Outer bract l|-lf mm. long, orbicular-or-triangular-ovate, apex more or less acute, keeled with the keel projecting as an apiculus, usually wholly membranous except for slight herbaceousness near base, more or less finely hairy (rarely almost glabrous). Middle bract lf-2 mm. long, irregularly oblong-ovate, apex truncate or rounded, hyaline with veins, glabrous. Inner bract 2^-3J mm. long, orbicular-ovate, with broad membranous margin which is often, together with herbaceous portion, emarginate or jagged at apex, more or less finely hairy (often velvety) over herbaceous portion (rarely almost glabrous), usually a little less than twice as long as outer bract. Bracteole 1 or absent, 2|-2f mm. long, irregularly ovate, apex rounded, pointed or jagged, hyaline with veins, glabrous (rarely with a few hairs). Calyx 3f-4J mm. long, stout, densely hairy (or velvety) on veins and between same from base to about 1 mm. from tip of lobes, or less hairy and then chiefly on veins (rarely almost glabrous) ; calyx-lobes very short (^ mm. or less) rounded-triangular or (rarely) more acute, short sub-lobes usually present ; veins of calyx strong, not reaching to base of calyx-lobes. Corolla with lobes emarginate. Differs from all forms of L. vulgare and L. Gmelinii by the presence of minute hairs on scape, leaves, and bracts. Distribution. — European Eussia, S. E. (Gouvs. Saratoff ! and Astrakhan !) Sarepta ! Siberia ! Russia, S. & S. W. (Nyman). L. TOMENTELLUM O. Kuntze var. sareptanum. Statice Gmelini Reichb. ! Icon. Grit. Gent. 3, 37 (1825) ; non al. S. sareptana Becker! in Bull. Soc. Nat. Mosc. xxi. I, 12 (1858). S. tomentella Boiss. subsp. S. sareptana B£ck. Nyman Gonspec. Fl. Europ. 609 (1881); Gandoger, Nov. Gonspec. Fl. Europ. 396 (1910). S. intermedia Gzern. {fide Gandoger, I. c). Ico?i.— Reichb. Icon. Grit. 237 ! Exsicc.—F. Schultz, Herb. Norm. Nov. Ser. Gent. 13, 1211 ! Becker, PL Wolg. infer. 42 ! (The description given below is contrasted with that of the type ; characters common to both are omitted) : — Planta parum hispidula aut glabriuscula; folia tuberculis albidis carentia, saepius emarginata; scapus saepissime infra THE GENUS CANEPHORA 77 medium ramosus; rami inferiores steriles; spicae longiores et laxiflorce spiculis saepe non contiguis; bractea exterior herbacea margine meiiibranacea, /erg glabra; bractea interior quam bractea exterior ixiullo i^lus duplo longior, fere glabra; calyx tenuior, multo minus hirsutus; calycis lobi breves, triangularo-ac?6^i. Plant very sparingly finely hairy, rarely almost glabrous. Leaves usually slightly emarginate, not showing white tubercles when dry. Scape often more or less glabrous in upper part, branched from below the middle, numerous sterile branches usually present. Spikes rather long and usually quite lax-flowered, often with spikelets not contiguous (as in L. humile Mill.). Spike- lets usually 1-flowered. Outer bract herbaceous with a broad membranous margin, apiculate or not, glabrous or almost so. Middle bract 1|— 2J mm. long. Inner bract 2f-3J mm. long, glabrous or almost so, a little more than twice as long as outer bract. Bracteole 1 (rarely 2), 2|— 3^ mm. long, glabrous. Calyx 4:-4| mm. long, more slender than that of tomentellum, irregu- larly hairy in lower half (not usually any higher) on veins and sparingly between same, some veins wholly glabrous ; calyx- lobes short, but usually slightly longer than those of tomentellum (J mm. or more), triangular-acute. Distribution. — European Russia, E. & S. E. (Gouvs. Saratoff ! and Orenburg !) Sarepta ! It will be noted that all the localities for L. tomentellum and variety are inland. Description of Plate. 1. Limonium tomentellum, two-thirds natural size. 2. Portion of spike of ditto, natural size. 3. a, b, c, d, c, outer bract, middle bract, inner bract, bracteole, and calyx of L. tomentellum, all enlarged four times. 4. a, h, c, d, e, outer bract, middle bract, inner bract, bracteole, and calyx of L. tomentellum var. sareptanum, all enlarged four times. THE GENUS CANEPHOEA. By H. F. Wernham, B.Sc. I. The Genus. This genus of Bubiacece is, so far as is known, confined to Madagascar; three species only have been described hitherto. An examination of the material available in the British Museum and Kew herbaria has suggested the necessity for some enquiry into the definition and affinities of the species. Canephora was described originally by Jussieu [Genera Plan- tarum, p. 208, 1789) as follows : — " Canephora. Flores 3-6 intra calicem communem sub-5-fidum, sessiles squamulis distincti. Singulis calix 5-6-fidus. Corolla parva campanulata 5-6-loba. Antherae 5-6 oblongue sessiles. Stigma 2-fidum. Fructus .... pisiformis coronatus 2-spermus. Frutex ; pcdunculi axillares soli- tarii, a basi ad apicem incrassati, desinentes in calicem communem 78 THE JOURNAL OF BOTANY calathiformem. Caracter ex Cqmmersoni sjyecimine Madagascari- ensis innominato. . . . Confer cum Faramea.'" Gmelin (Syst. 372, 1791) gives a summary of the generic characters based upon Jussieu's account; he adds a specific name, C. madagascariensis, but without description. Lamarck (Encycl. Meth. Illustr. 377, t. 151, fig. 1) in the same year gives a figure of the genus, to w^hich we shall return presently. Willdenow (Sp. PL 976, 1797) gives brief descriptions of two species, C. axillaris and C. cajntata, the latter of which has since been recognized as synonymous with BurchelUa capensis E. Br. The former is described as " C. foliis subrotundo-ovatis, floribus solitariis axil- laribus." Poiret (in Encycl. Suppl. ii. 57, 1811) gives the generic characters at length as follows : — " Le caractere essentiel de ce genre est d'avoir : un calice commun, tubule, a plusieurs fieurs; un calice propre, a cinq ou six decoupures ; une corolle campanulee, a cinq ou six divisions ; un fruit couronne, a deux semences. Chaque fleur offre : — 1*^. Un calice commun, d'une seule piece, tubule, dent6, a plusieurs fleurs separ^es par des ecailles ; un calice propre, presque campanul6, a cinq ou six decoupures. 2o. Une corolle monopetale, petite, campanulee, a cinq ou six lobes ovales, aigus. 3". Cinq ou six etamines ; les filamens presque nuls ; les antheres oblongues, point saillantes. 4^. Un ovaire inferieur, un peu arrondi, surmonte d'un style filiforme, plus long que la corolle, termine par un stig- mate bifide. Le fruit ... en forme de pois, couronne par les decoupures du calice, a deux loges, a deux semences." Poiret retains Willdenow's two species : C. axillaris is thus described : — " C. floribus sessilibus, solitariis, axillaribus ; foliis subrotundo- ovatis, breviter petiolatis. Arbrisseau dont les tiges se divisent en rameaux glabres, cylindriques, noueux, d un blanc-cendre, garnis des feuilles opposees, mediocrement petiolees, ovales, un peu arrondies, fermes, coriaces, luisantes, glabres a leurs deux faces, tres entieres, longues d'un pouce et plus, un peu aigues, a nervures fines, distantes, a peine rameuses, presqu' opposees ; les fleurs sessiles, axillaires, solitaires, plus courtes que les feuilles." Poiret also points out that the plant under description was collected by Commerson in Madagascar, so that C. axillaris Lam. is almost certainly identical with the plant on which Jussieu based his generic characters, and therefore the type,' also, of the C mada- gascariensis of Gmelin. Poiret refers the figure (Encycl. Meth. t. 151, fig. 1) to C. axil- laris ; but both figure and description are curiously inaccurate in regard to the inflorescence. The organ, which is in reality a common peduncle bearing a few flowers (see Jussieu's description supra), is represented as a long funnel-shaped calycine structure with several teeth at the margin, and at least a dozen small flowers appearing in the mouth. A. Richard, however, in his generic description of Ganepliora (Mem. Soc. Hist. Nat. Par. v. 261, 1829), refers to the organ in question not as a common calyx but as a flattened branch : "flores THE GENUS CANEPHORA 79 in apice ramuli compressi, sensim dilatati et apice cupulgeformi " ; and, to judge from the specimens in the National and Kew her- baria, this estimate of its nature is correct beyond question. Eichard makes the important addition that the ovary is multi- ovulate ; he does not refer to the seeds. De Candolle (Prodr. iv. 617, 1830) includes the genus under the head " Genera rubiacea non satis nota aut loco suo omissa," and his description is based mainly upon Richard's, with the addition, already pointed out by the earlier authorities, that the fruit is two-seeded. The affinity " inter Gardeniaceas Sarcocephaleas post Lucingeam inserendum " is suggested. Bentham & Hooker (Gen. PI. ii. 74, 1873-6) base their de- scription upon De Candolle's account, but suggest the inclusion of Canephora in the Mussandecd, a tribe characterized by valvate aestivation. This, however, is incorrect; for, as Baillon points out in Bull. Soc. Linn. Paris, i. p. 199, 1879, the aestivation is con- torted, a fact which I have been able to verify in all the plants examined. It would appear, therefore, that this genus should have a place in the Gardcniece, as suggested by De Candolle ; and K. Schumann has adopted this course in his account of the BubiacecB (Nat. Pflanzenf. iv. 4, p. 80). The general appearance and structure of the flower is certainly suggestive of Garclenieoi rather than of MusscBudecB. One point in the generic description which needs amendment concerns the number of the floral parts, which has hitherto been given as 5-6 or 5 ; as will be seen below, four of the five species are 4-merous. I have had no opportunity of examining fruits or seeds, and so must leave unsettled the rather important question as to whether the presence of one seed only in each loculus of the fruit is to be regarded as a constant generic character ; the earlier descriptions seem to point to this conclusion. The Musscendem have, typically, many-seeded fruits, whereas many of the Gar- deniece, e. g. Tricalysia (with which Canepliora is closely asso- ciated by K. Schumann), have few-seeded fruits. The existence of the latter character in Canepliora may thus acquire a certain phyletic importance. In any case, the number of ovules in the immature ovary is small — 5-7, according to K. Schumann. The critical generic characters of Canephora may, then, be summarized as follows : — Shrubs or trees, usually glabrous ; flower campanulate, with contorted aestivation of the corolla ; inflore- scence consisting of a few flowers, borne in the hollowed apex of a phylloclade, and surrounded by a common involucre ; ovary bilocular, with few ovules. II. The Species. Until 1897 the genus CanepJiora was known by the species C. axillaris W. alone. In that vear an accurate figure of this species was published in the Hist. PI. Madag., by Drake del Cas- tillo (t. 444), which forms part of Grandidier's " Histoiro" of the 80 THE JOURNAL OP BOTANY island, together with the figure of a second species, C. Humblotii, wliich is distinguished by its long oblong leaves and 4-merous flowers ; unfortunately, however, no description of this second species is given. A third species — also 4-merous — was described by Augustine De Candolle from material collected by Mocquerys (no. 142) in the Maroa district. This is C. maroana, of which the author states that " Le C. Humhlotii doit etre tres voisin du notre. II en differe cependant par la forme du limbe." This being so, the additional species now described appear to be quite distinct. The five species fall conveniently into the clavis exhibited below. The leaf-shape, characteristic as it is, seems to afford the best preliminary guide towards identification : — - Flowers 5-merous, leaves broadly elliptical, mostly 5-6 cm. long 1. G. madagascariensis. Flowers 4-merous, leaves nearly always more than 7 cm. long. Leaves rounded at the base, subsessile . . 2. C maroana. Leaves acute at the base. Leaves narrow-lanceolate, leathery . 8. C. angustifolia. Leaves broadly elliptical, submembranous 4. G. Gouclotii. Leaves long-oblong . . . . 5. C Humhlotii. Descriptions of these five species are given below ; that of the first, upon which the genus was founded, has been revised. 1. C. MADAGASCARIENSIS Gmelin Syst. 372 (1791), nomen. G. axillaris Willd. Sp. PI. 976 (1797). Frutex omnino glaber, ramulis subteretibus nodosis, foliis coriaceis nitidis ellipticis apice obtusis basin versus in petiolum brevem plus minus subito angustatis, stipuhs triangularibus cuspi- datis, inflorescentiis ssepissime 3-4-floris, floribus subsessilibus, calyce campanulato 5-dentato dentibus late triangularibus, corolla campanulata lobis 5 oblongis apice spathulatis quam tubus brevi- oribus utrinque glaberrima, staniinibus 5 subsessilibus antheris linearibus obtusiusculis basi lobatis, stylo filiform! stigmate in duobus lobis subaequalibus diviso, bacca. ... Hab. Fort Dauphin, Madagascar, Scott Elliot, 2945 ! (Herb. Mus. Brit, and Kew) ; J. Gloisel, 263 ! (in Herb. Mus. Brit.). Folia 5-6 (3-5-7-5) cm. long, x 2-7-2-9 (1-5-3-3) lat. Cost® secundariae angulo 70 -"-SO^ ad venam centralem insert®, utrinque 6-8 distantes. Stipulae 2 mm. long. Inflorescentias pedicella sub anthesin ca. 1'6 mm. long., ad apicem fere 6 mm. lat. Calyx totus 3-5 mm. ; dentes ca. 1 mm. long. Corollse lobi 3 mm., tubus 4 mm. long. Antherge, dimidio exsertse, ad 2-1 mm. long. This, the species which has been the longest known, is the only one so far known which has pentamerous flowers. 2. C. MAROANA Aug. DC. BuU. Herb. Boiss. ser. ii. 1, 1901, p. 584. I have not seen this plant, but the rounded leaf-base is a remarkable feature. The original description is appended: — ■ Frutex ramulosus glaber, ramis teretibus, foliis subsessilibus oblongis (8-11 cm. longis, 3^-4 cm. latis) basi rotundatis apice acuminatis subcoriaceis nervis lateralibus 8-10 adscendentibus THE GENUS CANEPHORA 81 inter se arcuatim connexis, stipulis basi inter se unitis longiuscule cuspidatis. Inflorescentia pauciflora — in specimine uni- vel bi- flora — pedunculis phyllodineis oblongo-linearibus 4 mm. latis basin versus attenuatis. Flos subsessilis 10-12 mm. longus. Calyx minimus 4-dentatus dentibus triangularibus subconcavis. Corolla infundibuliformis, 4-lobata lobis oblongis 4-5 mm. longis obtusis utrinque glaberrimis. Antherae lineares ad faucem sessiles apice subulato exsertae. Ovarium 2-loculare ovulis placentas axillari insertis. Stylus filiformis stigmate bifido. Discus epigynus annu- laris niger. Maroa, en foret (Mocquerys, 142). Arbuste de 2 a 3 m., a fleurs blanches. 3. C. angustifolia, n. sp. Frutex omnino glaber desuper rufescens, ramulis praecipue junioribus angulatis, foliis petiolatis anguste lanceolatis utrinque acutis coriaceis margine refiexo supra nitidis in siccitate nigrescentibus subtus levioribus, stipulis brevi- bus latis cuspidatis connatis persistentibus, infiorescentiis pauci- (saepissime 1-2-) floris, floribus manifeste pedicellatis, calyce infun- dibuliformi 4-dentato dentibus triangularibus, corollae glaberrimae late campanulatae lobis 4 oblongis obtusis quam tubus vix brevi- oribus, staminibus 4 antheris linearibus acutis basi bilobis fila- mentis brevibus sed manifestis, stylo filiformi striate in lobis duobus complanatis diviso, bacca. . . . Hab. Madagascar, J. V. Thompson in Herb. Mus. Brit. ! Folia 6-8 cm. x 1-1'8 cm. ; petiolus 2-4 mm. Vena centralis utrinque eminens, saepe rubens ; costae secundarige utrinque 6-8 patentes nee valde conspicuse. Stipulas 4 mm. long., basi 4 mm. lat. Inflorescentiae pedicella sub anthesin 2 cm. long., ad apicem 5 mm. lat. Calyx totus 3-5 mm. long., dentes ca. 1'5 mm. long. CoroUse lobi 6 mm. long., tubus 7 mm. long. Antherae 5'5 mm. long. Stilus 9 mm. long. Apart from the very characteristic leaf-form, this species is remarkable for the relatively large size of the floral parts, especially the anthers, and for the presence of distinct filaments. The following, again, although nearly allied apparently to Aug. De CandoUe's C. maroana, seems to be undoubtedly distinct therefrom, in virtue of its broader, elliptic, scarcely acuminate, petiolate leaves, acute at the base, its smaller flowers, and the larger size of tlie shrub as a whole : — 4. C. Goudotii, n. sp. Frutex glaber 4-5-pedalis, ramulis obscure angulatis, foliis pergamaceis petiolatis ellipticis apice sub- ol)tusis vix acuminatis basi in petiolum brevem angustatis mar- gine reflexo, stipulis triangularibus cuspidatis in l)revem vaginam cohaerentibus tardiuscule deciduis, inflorescentiis paucifloris, flori- Inis subsessilibus, calyce infundibuliformi 4-dentato dentibus tri- angularibus, corolla anguste campanulata lobis 4 ovato-lanceolatis suljacutis tubum vix aequantibus utrinque glaberrima, staminil)us 4 subsessilibus antheris linearibus acutis basi integris, stylo filiformi stigmate in duobus lobis inaequalibus diviso, bacca. . . . Hab. " Les montagnes qui bordent les rives du Saramdon." Madagascar, Goudot in Herb. Mus. Brit. ! 82 THE JOURNAL OP BOTANY Folia 9-13 cm. x 4-2-5-5 cm. ; petiolus 4-7 mm. long. Vena centralis utrinque eminens, costseque secundariEe angulo ca. 70° ad earn insertae, vix arcuatge, subtus nee valde conspicuis, supra utrinque ca. 10 eminentes, necnon alige minus conspicuse et irre- gulariter dispositse. Stipules totse 5 mm. long. Inflorescentiae pedicella sub anthesin 3-5 cm. long., ad apicem 4 mm. lat. Calyx totus 3 mm., dentes vix 1 mm. long. Corollae lobi 3-5 mm., tubus 3'8 mm. Antherge 2-2 mm. long. Lastly must be added some account of the species figured without description in Grandidier's Histoire. The following is based upon the figure in question : — 5. C. HuMBLOTii Drake, Hist. PL Madag. 1897, t. 444. Frutex glaber, ramuHs subteretibus, foliis petiolatis coriaceis (?) longi- uscule oblongis breviterque acuminatis basi angustatis, stipulis triangularibus mucronatis, inflorescentiis 3-4-iloris, floribus sessi- libus, calyce subtubuloso 4-dentato dentibus triangularibus obtusis, corolla campanulata lobis 4 plus minus anguste ovatis apice rotun- datis quam tubus brevioribus, staminibus 4 subsessilibus antheris lineari-fusiformibus apice subulatis obtusis. Compared with the figure of C. axillaris, the leaves are much longer and relatively narrower ; tiie pedicel of the inflorescence is longer and narrower at the top ; the corolla is more deeply lobed ; and the anthers are longer and not so stout. ALGOLOGICAL NOTES. By G. S. West, M.A., D.Sc, F.L.S. x\mong the Algae I am constantly collecting and those which are submitted to me for examination are many species of con- siderable interest and about which little is known. New facts concerning the structure and life-histories of some of these Algae are always coming to light, and our knowledge of their geogra- phical distribution is being gradually extended. It seems desirable that these numerous facts and records should be published in a collective form rather than as isolated notes which are liable to be overlooked, and I therefore propose to issue periodically a series of " Algological Notes" in which they will be embodied. Critical remarks upon little-known species, discussions upon the validity of certain species, and the systematic position of others will also be included ; and, in addition, short reports on small collections of Algas from various parts of the world will appear from time to time. I. — ALGiE FROM NEAR ElVADEO, NORTH-WEST SpAIN. A few small tubes of Algae were collected from running water near the town of Kivadeo by Mr. W. Fawcett in November, 1909, and were forwarded to me by Dr. A. B. Eendle. The material contained thirty-three species of Algifi, among which were a few Desmids and a number of rather interesting Diatoms. ALGOLOGICAL NOTES 83 Chlorophyce^. — Netrium Digitus (Ehrenb.) Itzig. & Eothe, Penium Navicula Br6b., Tetmemorus granulatus (Breb.) Ealfs, Euastrum oblong iiin (Grev.) Ealfs. Bacillarie^. — Eunotia Arciis Ehrenb., E. hinaris (Ehrenb.) Grun., E. pectinalis (Kutz.) Eabenh., Achnanthes linearis W. Sm., Navicula a232Jendiculata Kiitz., N. atomoides Grun., N. horealis Ehrenb.', N. Brehissonii Kiitz. var. diminuta V. Heurck, N. con- tenta Grun. var. biceps V. Heurck, N. cryptocephala Kiitz., N. falaisensis Breb., N. major Kiitz., N. jjeref/rma Kiitz. var. Menisculus Schum., A', rliynchocephala Kiitz., N. s'tauroptera Grun., N. tuscula Ehrenb., N. viridis Kiitz., Stauroneis anceps Ehrenb., S. gracilis Ehrenb., Vanheurchia vulgaris (Thw.) V. Heurck, Gomphonema parvulum Kiitz., Cocconeina turgidum (Greg.) nob. [= Cymbella turgida Greg.] , Nitzschia Denticula Grun., N. obtusa W. Sm., N. obtusa var. brevissima Grun., A''. Palea (Kiitz.) W. Sm. var. debilis V. Heurck, Hantzsclna aviphioxys (Ehrenb.) Grun. Myxophyce^. — Phormidium tenue (Menegh.) Gom., Oscilla- tor ia animalis Ag., 0. tenuis Ag. II. — A Diatomaceous Earth from Lewis, Outer Hebrides. Some three years ago Mr. F. J. Lewis, of the University of Liverpool, forwarded me a sample of a Diatomaceous Earth which he had collected to the south of Monach, near Loch Keiter, in Lewis, and recently he has very kindly supplied me with the particulars of its occurrence. The deposit was about 6 or 7 in. in thickness, and was overlaid (1) by a layer of peat from 2J to 3 ft. in thickness, containing trunks and stools of Betula alba, and (2) by some 7 to 9 ft. of peat, derived from Scirpus lacustris and Sphagnum. The deposit, which rested upon a few inches of fine sand on rock, is soft and friable, and of a rather dark grey colour. It is a fairly pure Diatomaceous Earth of freshwater origin, consisting of a number of species, two of which stand out conspicuously from the remainder. These two species occurred in great quantity and about equal proportions, and not only are they much the largest species present in the deposit, but are among the biggest of fresh- water Diatoms. They are Surirella robusta Ehrenb. and Navicula nobilis Ehrenb. var. Dactylus (Ehrenb.) V. Heurck. x\mong the large Diatoms were numerous acicular sponge- spicules. The matrix of smaller Diatoms consisted of the following species :—Tetracyclus lacustris Ealfs var. emarginatus (Ehrenb.) W. Sm., Tabellaria fenestrata (Lyngb.) Kiitz., Fragilaria capucina Desmaz., F. construens (Ehrenb.) Grun. var. binodis Grun., Eunotia pectinalis (Kiitz.) Eabenh. var. undulata Ealfs, E. prarupta Ehrenb. var. bidens (Ehrenb.) Grun., E. robusta Ealfs var. tetraodon (Ehrenb.) V. Heurck, E. Veneris Kiitz., Navicula major Kiitz., N. viridis Kiitz., Stauroneis Phanicenteron Ehrenb., Gomphonema acuminatum Ehrenb., G. constrictum Ehrenb., G. intricatum Ehrenb., Cocconema delicatulum{Ki\t'A.) nob. [= Cymbella delicatula 84 THE JOURNAL OF BOTANY Kiitz.] , Epithemia turgida (Ehrenb.) Klitz., Surirella ovalis Breb. var. angusta (Kiitz.) V. Heurck. - Fragments of two species of Synedra were also present, viz. S. Acus (Kiitz.) Grun. and S. Ulna (Nitzsch) Ehrenb. III. — New and Eare British Alg^. 1. Stichococcus scopulinus Hazen, "Ulotrich. and Chaetoph. of tlie United States," Memoirs Torr. Bot. Club, xi. no. 2, 1902, p. 161, t. 22, f. 4-6. This species was collected by Mr. W. B. Grove on wet stones at Studley, Warwickshire, in December, 1909. It seems to be a very characteristic species, having cells 3-3-5 /x in diameter, and up to 10 (or even 12) times longer than broad. It has not previously been recorded for the British Islands. 2. Closterium tumidum Johnson in Bull. Torr. Bot. Club, xxii. n. 7, 1895, p. 291, t. 239, f. 4 ; W. & G. S. West, Monogr. Brit. Desm. i. 1904, p. 156, t. 19, f. 15-18. The characters of this species were drawn up to include, amongst other forms, the Desmid originally described and figured by Ealfs as " Closterium Cornu Ehrenb. var. /5" (vide Ralfs, Brit. Desm. 1848, p. 176, t. 30, f. 6a-e). Of this form Ralfs figured the zygospores, and these are generally accepted as representing the zygospores of C. tumidum. They are subrectangular, with retuse sides and produced angles. Quite recently Mr. W. B. Grove collected in Warwickshire numbers of rather small forms of G. tumidum, agree- ing exactly with that figured by W. & G. S. West, I. c. pi. 19, f. 15. Many of these had conjugated and the zygo- _ spores were subglobose. It seems de- Fig. l.-Clo.<.n«»Umn/*m f^^ble, therefore, to regard this small Johns, var. ^ili(tro>^pora, var. n. form as a distmct variety. A, vegetative cell; B, zygospore, Var. SPH^ROSPORA, var. n. (Fig. 1). X 500. Var. cellulis parvis et brevibus ; zygo- sporis subglobosis vel ellipsoideo-glo- bosis. Long. cell. 48-66 /a, lat. 8-8-5 /x ; lat. apic. 3-4 /x ; diam. zygosp. 23-5-26 /x. Hah. Earlswood, Warwickshire, March, 1910. 3. Pleurococcus rufescens (Kiitz.) Breb. This species of Pleurococcus is exceedingly rare in the British Islands. In October, 1907, Mr. W. B. Grove collected very fine specimens attached to wood in a rain-water tank at Bewdley, in Worcestershire. Diam. cell. 11-5-17 /x. 4. Richteriella botryoides (Schmidle) Lemm. in Hedwigia, xxxvii. 1898, p. 306, t. 10, f. 1-6. This limnetic Alga occurred in small quantity associated with Actinastrum Hantzschii in the surface waters of the canal at Lifford, Worcestershire, in October, 1910. It is quite possible that this and other allied Algae are ALGOLOGICAL NOTES 85 widely distributed in the British Islands, but up to the present this is only the third record of the occurrence of Bichteriella hotryoides. 5. Lagerheimia wratislawiensis Schroder in Bar. Deutsch. hot. Ges. 1897, xv. p. 373, t. 17, f. 7. This Alga occurred sparingly among numerous members of the Protococcales in a small pool in the grounds of Studley Castle, Warwickshire, in October, 1906. It is, perhaps, the most distinctive species of the genus, and this is the first British record. In no single instance did the parietal chloroplast possess a pyrenoid, and, therefore, after comparison with Schroder's original description and figure, this must be regarded as additional evidence that the presence or absence of pyrenoids in the chloroplasts of the Protococcales is largely a question of nutrition. Long. cell. 12 /x, lat. cell. 9 /x ; long. set. 30-31 fx. (Fig. 2, G.) 6. Lagerheimia genevense Chodat in Nuova Notarisia, 1895, p. 87, f. 1-12. This minute Alga of the subfamily PhytJieliecB occurred in the helioplankton of Blackroot Pool, Sutton Park, Warwickshire, in August, 1907. The cells were cyhndrical, but Fig. 2. — A, Lagerheimia (jenevense Chodat. B-F, L. genevense var. suh- glohom (\ lenww.) GhoA. \ C-F, stages in formation of autospores. G, L.xcratis- laiviensis Schroder. A and B, x 1000 ; C-G, x 500. proportionately a little shorter than those figured by Chodat, and the bristles were rather short. Long. cell. 8 ju, lat. 4-5 /x ; long, set. 7-8 IX. (Fig. 2, A.) Var. suBGLOBOSA (Lemm.) Chodat, Algues vertes de la Suisse, Berne, 1902, p. 188. Lagerheimia subglobosa Lemm. in Hedwigia, xxxvii. 1898, p. 309, t. 10, f. 9. I agree with Chodat that this form can scai'cely l)e regarded as a species, as it differs only from L. genevenne in its ellipsoid cells. It occurred in the helioplankton of i3racebridge Pool, Sutton Park, Warwickshire, in April, 1910 ; and also in quantity in a small pool in the grounds of Studley Castle, Warwicksliire, in October, 1906. From the last-named locality many of the examples showed the formation of autospores, 86 THE JOURNAL OF BOTANY four of which ai'ose in each mother-cell. Long. cell. 5-6'5 //, lat. 4-4-6 /x; long. set. 10-13-5 /x. .(Fig. 2, B-F.) 7. Ankistrodesmus Spirotaenia, sp. n. (Fig. 3, A and B.) Cellulis solitariis, lihere natantibus, inter algas limneticas repertis, angustissiniis et multe elongatis ; parte mecliana angustissime fusiformi, ad polum unumquemque in spinam longissimam tenuissimam et rectissimam sub- gradatim producta ; chromatophora pallide viridi, taeniformi, spiraliter contorta, anfracti- bus circiter 3^-4, pyrenoidibus nuUis. Long. cell. 171-185 ^/x; lat. max. 2-2-1 /x. This curious species occurred sparingly among other limnetic Algge in the surface waters of the canal at Lifford, "Worcestershire, in October, 1910. Its exceedingly narrow cells, drawn out at the apices into fine, straight, spine-like processes of great length, and the nature of the chloroplast, are characters which serve to distinguish it. The chloroplast is rather remarkable, consisting of a rather narrow, pale green band, wrapped spirally round the interior of the widest part of the cell, and having about four complete turns. No pyrenoids were observed. It is of a length approaching the Closteri- opsis-type, but differs from all the elongated species of Ankistroclesvms in the spirally twisted chloroplast. I would here draw attention to a Diatom, Nitzschia Tcenia W. Sm., from which Anhistro- B I desmus Spirotania differs in being a member of the Protococcales, and therefore in the / possession of a chloroplast, in the structure and nature of its cell- wall, in the equal dia- ^ meter of its cells when viewed from all sides, and many other characters. 'I consider this comparison rather irrelevant, but I institute it purposely after reading certain statements Fig. 3. — A and B, made by Mr. G. I. Playfair. In a footnote to Ankistrodesmus Sjnro- ^^^^ ^^ ^.j^^ most amusing papers which has teHt«, sp.n. : A, outline -\ ,^ t\ -i- i.i • j_i of cell • B median por- ^ver appeared on the Desmiaiacea, this author,- tion w'ith' chloroplast. replying to the fact that I had twitted him C, A. falcatus var. with having described a common attenuated ackularis{A Br.)G S species of Nitzschia (N. acicularis) as a Closte- West, an elongated -v i ^i j. t j-i ii -j. foi-m A X 600-'''""'^' I'eniarks that 1 was apparently "quite B, X 800; C, X 500. ' unconscious" of having "described and figured the same Diatom, Nitzschia reversa, as Anki- strodesmus nitzschioides, sp. n." In the first place, I must remark * G. I. Playfair, " Polymorphism and Life-History in the Desniidiacece," Proc. Linn. Soc. New South Wales, 1910, xxxv. jDart 2, p. 459. ALGOLOGICAL NOTES 87 that Mr. Playfair's description and figure (as far as they go, and they are very imperfect) of " Closterium naviculoideiim " not only agree very well with Nitzschia acicularis, but that author actually states that there was some doubt in his mind as to whether the plant was not the Diatom Nitzschia acicularis Smith. Under such circumstances Mr. Playfair had no reason for foisting another synonym upon an already overburdened nomenclature. As regards Ankistrodesmns nitzschioides, I may say, first, that I am not " quite unconscious" of any Alga I have described ; and, secondly, that my remarks under the diagnosis of the species clearly show that I had carefully considered its resemblance to the attenuated species of Nitzschia, all of which I know quite well. The attenuated spine-like apices of this Alga, like those of A. S])irot<2nia and others, are of quite a different nature from the attenuated frustules of any species of Nitzschia, and, moreover, the cells are isodiametric. It might also be mentioned that " chromatophora pallide viridi, pyrenoidibus nullis " is a statement of fact. 8. Ankisteodesmus falcatus (Corda) Ealfs var. acicularis (A. Br.) G. S. West. Some very long forms of this variety occurred abundantly in the reservoir at Barnt Green, Worcestershire, in May, 1910. Long. cell. 146-165 /x ; lat. 3-3-4 fx. (Fig. 3, C.) 9. Actinasteum Hantzschii Lagerh. in Ofvers. af K. Vet. Akad. Forhandl. 1882, n. 2, p. 70, t. 3, f. 25-26 ; Schroder in Ber. Deutsch. botan. Ges. xv. 1897, t. 25, f. 3. This rare Alga has few British records. It occurred in small numbers in the surface- waters of the canal at Lifford, Worcestershire, in October, 1910. Long. cell. 16-21 jx ; lat. cell. 3-3-5 fx. It is known from the plankton of the Kiver Thames. 10. Hydeueus fcetidus (Vill.) Kirchn. This member of the SyngeneticcB was collected in the swiftly running Eiver Plym at Shaugh Bridge, in Devonshire, in April, 1910. The collectors were Miss Olive E. Hood and Mr. K. H. Whitehouse, who informed me that the current was very strong and the water very cold, due to melting snows on the moorlands from which the river flows. It is an interesting record for one of the southern counties, wliere one would scarcely expect to find such a denizen of alpine regions. In the British Islands it has been previously recorded from Scot- land and North Yorkshire. 11. Synedea actinasteoides Lemm. in Ber. Deutsch. botan. Ges. 1900, xviii. p. 30. This interesting Diatom occurred in some tow-nettings made in the canal near Lift'ord, in Worcester- sliire, in October, 1910. It has not previously been recorded for the British Islands. It belongs to that section of the genus Synedra named by Lemmermann " Belonastnun," in which the cells are associated to form free-floating colonies, each colony consisting of a cluster of Diatoms radiating from a centiul point. Long. valv. 54-59 /x ; lat. 2'6-3 /x. About 8 cells were present in each colony, but the number, although variable, was well within the limits given by Lemmermann of 4-16. The dimensions were 88 THE JOURNAL OF BOTANY slightly greater than those of the typical form so frequent in the rivers and lakes of Germany, but were not so great as the measure- ments of var. lata Lemm. (Z. c. p. 30). 12. Ehizosolenia morsa W. & G. S. West in Trans. Koy. Irish Acad, xxxiii. sect. B, 1906, p. 109, t. 11, f. 5-7 ; G. S. West in Journ. Linn. Soc. (Bot.) xxxix. 1909, p. 77, t. 3, f. 15-16 ; W. & G. S. West in The Naturalist, 1909, p. 292, and fig. 2 on p. 187. B. eriensis H. L. Smith var. morsa W. & G. S. West, 1905. This species was found in association with Synedra actinastroides .in tlie surface-waters of the canal at Lifford, in Worcestershire. It is a characteristic constituent of the plankton of all the British lake-areas, and its occurrence in the Midlands of England is both noteworthy and interesting. Lat. valv. 4-8'5 /x. IV. — DiPLOCH^TE Collins and Polychj^tophora W. &G. S.West. The publication of a second species of Polychcetophora, namely. P. simplex G. S. West,* has induced Collins f to place both P. simplex and P. lamellosa j under his genus Diplochcete. He states, with some reason, that if P. simplex is to be included in the same genus as P. lamellosa, then both must be trans- ferred to the genus DiplochcBte, of which a solitary species was described in 1901. § It seems, however, impossible to regard Diplochcete soUtaria and Polychcetophora lamellosa as species of the same genus, and my own personal view is that for several reasons they should remain as species of different genera. There now remains for discussion the systematic position of Polychcetopliora siinplex. When I described this plant I had grave doubts whether it should be placed in the genus Polychceto- phora or a new genus created for its reception. The suggestion made by Collins that it should be placed as a species of Diplochcete seems to me almost a more erroneous view than its inclusion in the genus Polychcetophora. The American '^l^n'i— Diplochcete solitaria — possesses flattened cells with thick walls, each cell being furnished with two bristles, usually at opposite poles, and arising from the lower half of the cell. The figure given by Collins shows these bristles to be rather short, stiff' and thick, and very different in character from the slender, flexuose bristles of the British plants. Moreover, Dipilochcete solitaria is a marine Alga epiphytic on Laurencia ohtusa, a fact which must not be lost sight of. Few genera of Algae are common to both fresh watei" and the sea, and it is highly probable that these two plants, Diplochcete solitaria and Polychcetophora simplex, have originated independently from quite different sources among the lower green forms. Under these * G. S. West, " Some Critical Green Algffi," Journ. Linn. Soc. bot. xxxviii. 1908, p. 279, t. 20, f. 1-G. t F. S. Collins, "The Green Algae of N. America," Tufts College Studies, Mass. vol. ii. no. 3, 1909, p. 278. \ W. & G. S. West, "Notes Alg. III.," Journ. Bot. March, 1903, t. 448, f. 1-4. § F. S. Collins, " The Algas of Jamaica," Proc. Araer. Acad, xxxvii. 1901, p. 242. THE FLORA OF THE WORMS HEAD 89 circumstances, and considering that the characters of each are sufficiently distinctive, I would suggest that the most reasonable, and probably the most correct, way out of the difficulty is to establish a third genus for the reception of the Alga described as " Polychatopliora simplex." I would suggest OligochcetojyJiora as the name of this genus, and the following as a synopsis of the three genera discussed above : — PoLYCHJETOPHORA W. & G. S. Wost, 1903. Freshwater. Not known to be epiphytic. Cells subglobose, ellipsoid, or ovoid, solitary or more usitally 2 to 6 aggregated to form a suhfilamentous thallus ; cell-tvall very thick and strongly lamellose. Each cell furnished with 8-12 long, flexuose, simple seta of a delicate character and without a sheath, mostly arising from the sides of the cells, hut also from the dorsal surface. Chloroplast single, parietal, and often indistinct in its limitations. 1. P. lamellosa W. & G. S. West. Diam. cell. 19-35 fx ; crass. cell, membr. 2-8-10-5 /jl ; long. set. 86-183 /x. DiPLOCH^TE CoUins, 1901. Marine. Epiphytic. Cells ellip- soid and somewhat flattened, solitary ; cell-ivall thick and homo- geneous. Each cell furnished with 2 rather short, stiff, simple setcs, arising from the lower half of the cell, usually at opposite poles. 1. D. solitariaCoWm^. Diam. cell. 25-30 ju ; crass, cell, membr. 5-8 /x; long. set. 55-67 /x ; crass, has. set. 5-6 /x. Oligochaetophora, gen. n. Freshwater. Epiphytic. Cells subglobose or ovoid, loosely aggregated to form very small colonies ; cell-wall very thin and homogeneous. Each cell furnished wnth 2-4 long, flexuose, simple seta, of a delicate character and without a sheath, arising only from the dorsal surface of the cell. Chloro- plast single, parietal, possessing 2 or 3 small granules of starch. 1. 0. simplex nob. [= Polychcetophora simplex G. S. West, 1908] . Diam. cell. 15-20 /x ; long. set. 50-210 (plerumque 120) /x; crass, has. set. 0-6 fx. THE FLORA OF THE WORMS HEAD, AND THE NATIVITY OF CERTAIN DISPUTED SPECIES. By the Rev. H. J. Riddelsdell. Last June I spent three weeks at Rhosili, a village close to the Worms Head, and at the western extremity of the peninsula of Gower. The peninsula forms the western part of Glamorgan- shire, and extends some eighteen miles west from Swansea. The coast consists of lofty limestone cliffs, rising to more than 200 ft. at Rhosili. North of the village the limestone is replaced by a considerable hill of Old Red Sandstone, which forms a rocky boggy heathland, producing Sedum anglicum, Drosera, Osmunda, and the like. The hill ends seawards in low cliffs, fronted by a good bay of smooth sand. The limestone cliffs extend from Rhosili south to tlie Worms Journal of Botany. — Vol. 49. [March, 1911.] h 90 THE JOURNAL OF BOTANY Head, and thence (with a few breaks) east to the Mumbles, near Swansea. In the neighbourhood of the Head there is Httle sand ; high tide covers the foot of the chlTs. The Head itself is a very desolate stretch of land, shaped like a long peninsula or island, which in turn comprises three parts or islands connecting by short necks of rough sea-blackened rocks ; the whole is joined to the mainland by a very rough and tiresome passage-way of boulders and rocks, exposed only at half tide. Some parts of the Head are precipitous ; the largest " island," which is also nearest to the mainland, is precipitous on one side, the other forming a gentler slope. The outermost "island" rises steeply to its extremity, and then drops sheer into the sea. All over the Head the limestone is too near the surface for cultivation to be possible : it is a great breeding-place for gulls. Sheep are pastured during part of the year. I examined the flora of the Head and neighbouring cliffs, not in order to get a complete list of the native plants ; the list given below does not pretend to that. I wished rather to test the question as to whether some of our well-known species are native or not. The conclusion that is forced on my mind, more and more in proportion as I study the plant associations of such out- of-the-way spots as this, is that we have full right to treat, e. g. Malva sylvestris and Anagallis arvensis, as real natives of Britain. Some parts of the land about Ehosili and the villages near are cultivated. Ploughed land comes down in places to within two hundred or three hundred yards of the sea. But there is none, and I believe there can never have been any, within about a mile of the Head, or within two miles of its extreme point. Of course, as birds frequent the Head, and men and animals find their way on to it, the chance of plant introduction is by no means absent ; a remark which applies equally well to any part of England and Wales. The Worms Head is set forth here only as an excellent ex- ample, more free from complicating influences than others in the neighbourhood, of the occurrence of some species' of disputed status. Many localities from the cliffs close by, or from cliffs in other parts of the county, or from wet hollows in sand dunes, &c., could be cited for some of these species. The evidence drawn from the data under consideration is, I believe, strongly in favour of their being native. It is not indisputable ; nothing is. Species were massed variously on different parts of the Head. The outermost island was almost entirely given up to Silene mari- tima and Statice maritima. The portion of the middle "island" nearest the mainland, stood out red, to the distant observer, with masses of sorrel. The largest " island " had quantities of gorse on the more level parts, the higher ground being occupied largely with Helianthemum, Silene, &c. Scilla verna was scattered over most of the surface, sometimes so thickly as almost to exclude other vegetation. Ranunculus acris, B. Ficaria, Cochlearia officinalis ; C. danica, in large quantity. Helianthemum canum, which likes the rockier THE FLORA OF THE WORMS HEAD 91 and more exposed ground more than does H. Chanicecistus. Poly- gala vulgaris ; Silene maritima becomes gradually more frequent seawards until it occupies nearly the whole ground. Cerastium tetrandrum, viscosum, vulgatum, Stellaria apetala, Arenaria ser- pyllifolia ; Sagina maritima, in many forms and sizes, not assign- able to any named in the London Catalogue, I believe, though var. densa occurs on Burry Holm, four miles away. his mind, nor that Batters for reasons of priority employed a different nomenclature. With regard to the latter there is direct evidence to the contrary, as in the list of changes of nomenclature given by Batters at the end of the Catalogue neither species is referred to. There would appear, therefore, to be no doubt that the two species were unintentionally united. The distribution of the species according to Foslie is as under: — L. Eacemus Fosl. North Atlantic : Mediterranean, Adriatic, English Channel (Falmouth, only dead specimens known), Bahamas, Eed Sea. Indian Ocean : Mauritius, Eodriguez. * At the same time Foslie records L. fasciculatum var. eunana from Larne Lough, on the east of Ireland. The writer has seen the Larne Lough specimens, and they are sterile, and he cannot help thinking that Foslie's earlier deter- mination of the material is the more correct, namely that it is a form of Litliothamniou calcamon and not a form of L. fasciculatum. CH^ROPHYLLUM AUREUM IN BRITAIN 117 L. FAscicuLATUM Fosl. Co. Galway I Roundstone Bay (McCalla, Hanna, Foslie), Fahy Bay (Hanna in Herb. Batters). Co. Cork: Schull (ex Johnson, in list of Irish Corallinacese). Co. Waterford : Ballynacourty (ex Johnson I.e.). Exsicc. — Hohiies, Algce BritanniccB Bariores no. 262. CHMBOPHYLLUM AUBEUM L. IN BRITAIN. By G. Claridge Druce, M.A., F.L.S. In English Botany, under t. 2103, published in January, 1810, there is a description of the above-named Umbellifer, which Smith says was found by G. Don "in the borders of fields, between Arbraath [Arbroath] and Montrose, and at Corstorphine, near Edinburgh." Subsequently Don (Herb. Brit. fasc. ix. ? 1812, No. 207) repeated the above record, adding, " I observed what I believed to be an intermediate plant between this and the C. syl- vestre, near the village of Kirkliston, about ten miles west from Edinburgh." The specimen in his Fasciculns is correctly named, but no one seems to have found the plant again in either of the above localities, and Hooker [Student's Flora, p. 531) says, "Scot- land, G. Don, not confirmed." It was therefore a great pleasure to hear from Mr. James Eraser, of Leith, in 1909, that he had found it at Callander. Last year I had perforce to break my journey at this popular centre, and after dinner strolled out ; to my surprise, within five minutes' walk of the ' Dreadnought ' Hotel, I met with G. aureuvi, growing in great quantities by the Teith, up the banks of which I traced it north-westwards for more than a mile, growing in im- mense quantities to the exclusion of C. temulum. It was not only on the banks of the river, but also in some meadows and fields and under walls in the vicinity ; doubtless its very abundance having led the numerous botanists who must have visited this place to think it was a form of Anthriseus sylvestris Hoffm., wliich it somewhat suggests. A closer examination, however, shows that this plant is a perennial, and it has not the furrowed stem of A. sylvestris. From C. temuUim, which its somewhat spotted stems with swollen nodes might suggest, its perennial growth, its leaves of a brighter green and firmer texture and witli acute segments, and its umbels with denser and more showy flowers, clearly distinguish it. It was pleasant to see, so firmly established, another of Don's reputed discoveries. A word or two may be said as to the status of the plant. Its Continental distribution is rather against its being Scottish, and Nyman says it occurs in South Germany, Switzerland, Italy, &c., but not in France (except the Pyrenees), Holland, Belgium, or Northern Germany, while it is adventitious only in South Norway. It is a plant which would scarcely tempt the horticulturist to intro- duce it to his garden but, like the Chervil, it has ax'omatic pro- lis THE JOURNAL OP BOTANY perties, and it may have been at one time used as a condiment, since Mr. Clement Eeid informs me he has found a single fruit in Eoman Silchester. At present we can only regard it as a well- established alien. Now that attention has been called to it, the plant may be found elsewhere in Scotland. The specific name is somewhat misleading, referring, as it does, not to the flower, which is of a purer white, I think, than either temulum or sylvestris, but to the fruits, which as they ripen have a greenish golden hue. DOUBLE FLOWERS. By a. R. Horwood. In an interesting communication Miss Helen Saunders (p. 62) describes her recent discovery of wild double daffodils, and she would explain them as due to overcrossing, or excessive inter- breeding as we may call it ; but she states that gardeners do not procure them by any uniform plan, nor universally by crossing. It is certainly the case that they are commonest under cultivation. And yet in spite of this and the known neglect by horticulturists of artificial crossing to account for them, she proposes this method as responsible for those found double in a wild state, when a reason lies to hand which is constant under all conditions, viz., the irritation caused by insects, which is common also to many other types of monstrosities or malformations. This agency Miss Saunders dismisses as not reasonable ; but surely a fact may be known without a reason for it '? And so it is that the double flower has been found by repu- table botanists such as Molliard (" C6cidies Florales," Ann. des Sci. Nat. s6r. viii. 1895, p. 67) to be due to mites which set up an irritation, and cause thereby teratological structures. The work of Masters amply illustrates the variety of these monstrosities. Peyritsch has shown also that by their means petals become sepals and stamens petals. Is it then unreasonable to say that a double flower with additional petals is due to insect irritation ?, If so, we should have to dismiss then as unhkely the supposition that the fastigiate galls upon trees — with which double flowers are strictly comparable — are due to insects as equally unreasonable. But it is almost a truism to remind anyone that galls are actually produced by certain GecidomyidcB, allied to those that cause double flowers. And just as these galls, as Darwin has remarked [Animals and Plants under Domestication, 1868, p. 283), come remarkably true to character each time, though so different amongst themselves and produced, moreover, by different insects so closely allied, so double flowers initially produced by gall-insects, once brought into existence, come true generation after generation. If one compares a double daffodil with the fastigiate flower-like gall or " helix" of the "rose" willow, where is the difference in the nature of the work done ? True, there is a difference in the organs or parts — in the one case, petals, in the other, foliar appendages — but every A NEW BRITISH FLORA 119 botanist knows in his own experience many cases of such pheno- mena— that the budding of one organ out of another prohferation (as this is called) is due also to insect agency or soil conditions, and what is possible in the one case may assuredly be held to be so in another. Vivipary and prolepsis are equally intelligible by appeal- ing to soil conditions in the one case, insect agency in the other. Moreover, the existence of imperfect stamens in these plants, although the flowers are said to differ from the common double dafi'odil, indicates not so much a case of crossing, but one in which the production of seed has been prevented by some abnormal cause. This non-production of seed is well known in the violet, which lacks petals in the seed-producing state, but is here due to another reason, doubtless the early season of flowering and want of cover for the seeds or means of burying them or ripening them. In the case of the double narcissus, insects disturb the regular working of the floral structure, and activity is centred in petal- formation instead of seed-formation, just as petals in the violet are dispensed with when the embryo is destined to mature in the ovary. If further suggestion were needed as to the probable cause of the production of double daffodils at Chittlehampton, it is to be sought in the well-established fact that similar teratological abnormalities can be induced by a mere pin-prick, so that mecha- nical stimulus may also be the inducing agent, as well as insect intrusion. Turning, lastly, to analogy in the animal kingdom, would it be more unreasonable to make the suggestion we have made, or to draw attention to ascertained facts as to the origin of double flowers, than to suggest to a Cingalese that pearls were due to the irritant action of sand upon the shell-wall of an oyster '? With this case for reflection and comparison, I would venture to suggest that Miss Saunders may have rather too readily thrown aside the knowledge of experienced botanists and horticulturists. A NEW BEITISH FLORA. From time to time notices have appeared in this Journal announcing the preparation of a new British Flora. The neces- sity for such a work on modern lines has during the last few years become increasingly obvious, and Prof. Trail in his presidential address at the British Association last year (reprinted in this Jour- nal, 1910, pp. 241-250) emphasized this necessity and suggested preliminary steps towards carrying out an adequate scheme. There is of course ample provision for the field botanist in the way of handbooks which enable him to determine the various forms he may meet with, and througli the medium of this Journal and the Exchange Clubs he is kept fairly ait courant with critical investigation ; but something moi'e than this is required, notably in the direction of illustration. It is therefore with much satis- 120 THE JOURNAL OF BOTANY faction that we learn that the Cambridge University Press have made arrangements for the issue of such a Flora, under the capable editorship of Dr. C. E. Moss, the Curator of the University Herbarium. Dr. Moss, although a comparatively recent addition to the students of the Bi'itish Flora, has already, as our pages have shown, established his reputation as a careful investigator of the life-history of plants and of the works of the earlier writers and observers ; he is, moreover, young, and will thus, in the natural course of events, be able to carry out a scheme which will require time to bring to completion : no better choice could, there- fore, have been made. It is expected that the Flora will be completed in ten volumes, which will be issued annually. The cost will be about two guineas a volume. The high price is necessitated by the illustrations. Each species, in addition to many varieties, will be illustrated by Mr. E. W. Hunnybun's beautiful pen-and-ink drawings, about 1750 of which have already been completed. Each of Mr. Hunnybun's drawings has been made from living specimens ; each plant has been drawn natural size, and, in addition to a general drawing of each plant, there will be enlargements of critical organs ; and, in the case of critical species and varieties, the name of each specimen drawn from has been vouched for by some competent authority, whose letter of identification has been preserved. In addition to these drawings, there will be numerous photographs of plants in their natural habitats, and maps in- dicating the distribution of certain genera and species will be freely supplied. Engler's system of classification will, generally speaking, be followed. This system is becoming very generally adopted, and already there are German, Swiss, and American floras based on this system, which without doubt represents a great advance in several respects on that of Bentham and Hooker. The first volume to appear (which will be vol. ii. of the Flora) will deal with the earlier Dicotyledonous families, from the Salieacece to the ChenopodiacecR. The Monocotyledonous families will appear last. Gymnosperms and Pteridophytes will also be included, and per- haps the Charophytes. In addition to the systematic descrip- tions— which will be in English and not in Latin — of the various groups, there will be short statements of the affinities of the various orders and families, and the geographical distribution of the various groups will be given. So far as is possible, recent accepted work on morphology and ecology will be incorporated in the descriptions. Dr. Moss will utilize the services of speciaHsts in several of the more critical genera. Already many well-known botanists have promised their aid, e.g. Mr. G. C. Druce {Cerastium, &c.), Mr. A. B. Jackson (Thymus), Eev. E. F. Linton {Salix), Eev. E. S. Marshall {Cochlearia, Epilobium), Mr. Bennett (Potamogcton), and Mr. H. W. Pugsley {Fumaria). The first volume will deal with such difficult genera as Populus, Quercus, Ulmus, and Sali- cornia, on all of which the author has been engaged for some years. HENRY CHICHESTER HART SAMUEL ALEXANDER STEWART EDWARD PERCEVAL WRIGHT WILLIAM HUNT PAINTER SOME IRISH BOTANISTS 121 It will be necessary to keep a uniform standard of the extent and limits of the species, and with this end in view it may be desirable for the author to act in an editorial capacity when dealing with the genera written by specialists ; but, so far as is possible, each specialist will be given a free hand in the treatment of his particular genus. Special attention will be paid to nomenclatorial matters, and the International Eules will, in general, be followed. Vol. i., vv'hich will be about the fourth or fifth volume to appear, will contain the Gymnosperms and Pteridophytes, and the intro- ductory chapters. The latter will deal with topographical distri- bution, and with such questions as the origin of the British Flora. SOME lEISH BOTANISTS. By the Editor, (with portraits.) For many years this Journal has given prominence to bio- logical notices of botanists of these kingdoms who have been removed by death from the fields in which they have won more or less distinction, or have at least put on record in these pages some evidence of their attachment to botanical investigation, repre- senting different grades of society but one in their common interest in botany. In the case of the three Irishmen of whom I now propose to give a somew^hat belated notice, ample accounts have appeared in the pages of the Irish Naturalist, to whom I am indebted for many of the following facts, and also for permission to reproduce the excellent portraits which accompany the memoirs. Althougli late, it seems right that some notice of them should appear in these pages, to which two at least were frequent contributors. Henry Chichester Hart (1817-1908) was born at Raheny, Co. Dublin, on July 29, 1847, at which time his father. Sir Andrew S. Hart, was Vice-Provost of Trinity College, Dublin, where he graduated B.A. in 1869. He came of a Donegal family, and in his early youth began the investigation of the county which, after several papers in this Journal from 1879 to 1896 and reports in the Proceedings of the Royal Irish Academy, culminated in his Flora of Donegal, published in 1898. Numerous other papers dealing with the botany of various parts of Ireland will be found in the two publications named, in whicli his name appears as early as 1873, in connection with the discovery of Alchemilla alpina in Wicklow ; those in this Journal were often written in narrative form and, though thus less convenient for comparison than a mere list, make far more interesting reading, containing as they do many observations on the people and customs of the different districts, and on other points connected Journal of Botany. — Vol. 49. [April, 1911.] k 122 THE JOURNAL OF BOTANY with his excursions. His papers related mostly to phanerogams, but in 1886 he published a list of localities for the hepatics and mosses which he had collected at various times in Ireland. Some of his later notes on Irish plants were printed in the Irish Naturalist. His many observations on the plants of the Irish mountain ranges are summed up in a paper contributed to the Proceedings of the Eoyal Irish Academy in 1891. The tribute paid to Hart's work in the second edition, founded on the papers of his life-long friend A. G. More, of the Cybele Hihernica, by its editors, Messrs. Colgan and Scully, is evidently well deserved : he " has done more," they say, " to further our knowledge of Irish plant distribution than any other explorer of recent times." His last note in this Journal records his discovery in Skye of Arabis alpina, of which Mr. Druce recently exhibited Skye specimens at the Linnean Society. He died at his residence, Carrablagh, on Lough Swilly, Donegal, on Aug. 7, 1908. But Hart's botanical work was by no means limited to these islands. In 1876 he was appointed naturalist to the British Polar Expedition, and his exceedingly interesting account of the localities visited and of the plants collected by himself and his colleagues extends throughout this Journal for 1880, its earlier production having been deferred through ill-health. In 1883 he took part in an expedition organized by the Palestine Exploration Fund ; his report on the botany of Sinai and South Palestine is published in the Transactions of the Eoyal Irish Academy, xxviii. (1885), and that on the botany of the Jordan Valley and Western Palestine in the Journal of the Exploration Committee for 1885 : the former contained descriptions and figures of three new species, and added about seventy plants to the Palestine flora. The observations detailed in these two papers, with the zoological results of the expedition — for Hart was also an accomplished zoologist, although his work in that direction does not come within the scope of this notice — are brought together in his volume entitled Sovie Account of the Fauna and Flora of Sinai, Petra, and Wddy 'Arabah, published by the Exploration Com- mittee in 1891. The specimens collected both on the Polar and Palestine expeditions are in the National Herbarium and at Kew. Hart became a Fellow of the Linnean Society in 1887, but his name does not appear in recent lists. For an account of Hart's striking personality and various in- terests, and a general estimate of his work, reference must be made to the biography by his intimate friend and companion on many expeditions at home and abroad, Mr. E. M. Barrington, published, with an excellent portrait, in the Irish Naturalist for December, 1908. This contains a bibliography completing that published in Irish Topographical Botany, p. cxxii. Samuel Alexander Stewart (1826-1910) As is pointed out by the Eev. C. H. Waddell in his interesting sketch in the Irish Natiiralist, Stewart stands alone among Irish- SOME IRISH BOTANISTS 123 men as an example of the " working-man naturalist " of whom England and Scotland have furnished numerous instances. His grandfather went to America about 1798 and settled in Phila- delphia ; here Samuel was born on Feb. 5, 1826. At the age of twelve he returned with his family to Belfast, wdiere in spite of many difficulties he succeeded, at a time when facilities for the purpose were few, in obtaining a good general education. The classes conducted in natural history by Ralph Tate in 1861-64 under the Science and Art Department enabled Samuel to develop his bent in that direction ; he distinguished himself in every class, obtaining first class certificates and often silver or bronze medals in geology, mineralogy, and in systematic and physiological botany and zoology. To these classes may be traced the establishment in Belfast of a taste for natural history ; Tate early established field excursions for his more promising pupils, of whom Stewart was one, and in 1863 founded the Belfast Field Club, of which Stewart was always an enthusiastic supporter, and to whose Pro- ceedings he contributed numerous papers on botany and geology. In 1871 Stewart became a Fellow of the Botanical Society of Edinburgh ; in 1890 he was appointed assistant curator of the Belfast Museum, succeeding to the curatorship in 1891. In 1904 he was elected an Associate of the Linnean Society — an honour which was celebrated by his friends at a public meeting in the Museum, when he was presented with an address and a purse of £120. In 1907 he retired from the Museum, but still continued to take an interest in his favourite studies ; his death on June 15 last year was the result of a street accident. He was buried in the City Cemetery, where a suitable monument will be erected to his memory by the members of the Field Club. Mr. Praeger, who adds to Mr. Waddell's account of Stewart's life an appreciation of his work, appends to it a very full biblio- graphy. His contributions to this Journal range from 1884 to 1899, in which latter year (p. 396) he defended the Cybele Hiber- nica from certain criticisms it had received. Most of his papers naturally appeared in the Irish Naturalist and in other Irish pub- lications. Stewart was generous in the distribution of specimens and was a corresponding member of the Botanical Exchange Club. Both his biographers bear witness to his modesty, energy, and accuracy, as well as to his willingness to help all who asked his assistance. Edward Perceval Wright (1834-1910) He was born in Dublin on Dec. 27, 1834, and began life as a clerk in a commercial company. In 1853 he entered Trinity College, where he graduated in 1857 and was in the same year appointed director of the College Museum. He studied natural science under G. J. Allman, then the University Professor of Botany, and in 1854 began to publish the Natural History Eevietv. In 1857 Wright was appointed Director of the Dublin Natural History Museum ; in 1862 he took the degree of M.D. and prac- k2 124 THE JOURNAL OF BOTANY tised in Dublin as an oculist. His appointment as locum tenens for Harvey, then Professor of Botany, in 1865-66 turned his thoughts again in the direction of natural science ; and in 1869 he became Professor of Botany, in succession to Alexander Dickson who had been appointed to Glasgow, and Keeper of the Herbarium in Trinity College in 1870 — the former post he resigned in 1904, owing to failing health, retaining the latter for some time longer. Before his definite appointment, Wright occupied himself with botanical travel in various directions. His expedition to the Aran Islands in 1865 resulted in a paper published in the Proceedings of the Dublin Natural History Society for 1866-7, which was his principal contribution to Irish botany ; his last published note, however (Notes from Bot. School 237 (1904)), relates to the occur- rence of Euphrasia occidentalis in Ireland. In 1867 he visited the Seychelles, where he stayed six months ; of his visit he gives an account in a privately printed letter addressed to the President of the University Board, who had obtained for him an extended leave of absence : the botanical results of the visit are published in the Transactions of the Royal Irish Academy xxiv. (1871). Wright spent the spring of 1868 in Sicily, and published the zoological results. He was at least as competent a zoologist as a botanist : the list of his papers in the Royal Society's Catalogue shows that his scientific interests extended over a very wide field. Perhaps his most important botanical work was connected with Algae, on whose structure and development he published a series of memoirs. Wright took great interest in the Herbarium of which he was Keeper, and in the first number of the Notes from the Botanical School of the College — a journal published under his control and at his expense from 1896 to 1905 — he gives a useful account of its history and contents, in the course of which it becomes apparent that, but for his activity, there would have been no herbarium to keep, as it had been untouched from 1866 to 1869. He was extremely generous in lending the important South African collec- tions made by Harvey in connection with the Flora Gapensis, and apparently not always quite careful in noting to whom they were sent ; I remember being fortunate enough to put him on the track of the Asclepiadacece which had been lent to a botanist who at one time proposed to work at the order, and had for a long period been lost sight of. Dr. H. H. Dixon's account in the Irish Natm-alist, to which I am indebted, makes no reference to Wright's genial personahty. He was a thorough Irishman — it is understood that his Home Rule proclivities diminished his popularity among his colleagues ; and on his rare visits to the National Herbarium he brought into it a breeziness characteristic of his race. After his retirement he spent much of the winters at Poggio Gherardo, the classical villa near Florence where Mrs. Janet Ross has for many years extended charming hospitality to literary men. Here I last met him in 1908 ; he was as genial and amusing as ever, but there was a feel- ing that the visit would be his last, as indeed it proved to be. william hunt painter 125 Three Seventeenth Century Botanists. The following, which I had intended to print as a separate note, may fitly appear as an appendix to the foregoing. In the Correspondence of John Bay, printed for the Ray Society in 1848, occur the names of three Irish botanists of whom nothing further appears to be known. Their names occur in a letter written to Ray on Aug. 26, 1696, by Dr. Francis Vaughan, " a learned physician of Ireland, living at Clonmel, in the county of Tipperary " — so Ray describes him in a letter to Sloane (p. 313) — in which Vaughan says : " Dr. Wood, Dr. Mitchell, and I have resolved to be as curious as our leisures will permit in making a collection of what plants this kingdom affords. We have begun this summer at Wexford, where we casually meet to drink the medicinal waters, and in the month we stayed there we made up a catalogue of about 280 plants, to which we have and design to augment as opportunity offers to any of us " (p. 304). Vaughan sent Ray an account of the poisoning of " eight young lads " in Co. Tipperary by eating the roots of CEnantJie crocata (p. 314), which was sent by Ray to Sloane on March 16, 1697, and is printed also in Phil. Trans, xx. 84 (1698). Later (April 24, 1697) Vaughan sent Ray an account of a case of poisoning by Eupliorhia hyherna which occurred eight miles from Clonmel—" some of the Irish use this root boiled in milk as a cathartic"; and speaks of the local employment of Hypericum elodes — " the native Irish call it Birin Yarragh, which signifies Herha dysenterica, and use it in that distemper with good success," as Vaughan himself did (pp. 319, 320). Vaughan's account of Q^^nanthe is referred to by Threlkeld (Syn. Stirp. Hibern. (not paged) 1727) and that of the EupJiorhia in Molyneux's appendix to that work (p. 22), but neither Threlkeld nor K'Eogh (Botan. Hibern., 1735) mention him or the others among their authorities or subscribers. Wood, who signs himself N. W., writes to Ray on Aug. 31, 1696, from Kilkenny as to the use of "dullysk" in Kerry (p, 305), and from the same place (where he doubtless resided) on April 28, 1697, about the depositing of eggs in rushes by insects (p. 320). There is a letter from Mitchell written from Dublin to Sloane (Sloane MS. 4075 f. 203), dated 6th Oct. 1724, of purely medical interest. Mr. Praeger can tell me nothing of Vaughan, Wood, or Mitchell, but suggests that some reader of the Journal may throw light upon them. WILLIAM HUNT PAINTER. (1835-1910.) William Hunt Painter was born in Birmingham on July 16, 1835. Of his earlier days I know nothing, but in 1861 he entered tlie ministry of the Church of England. He had intended to engage in missionary work in connection with the C.M.S. and for this purpose was trained at the Islington Missionary College, but owing to obstacles he took up home work, his first curacy being 126 THE JOURNAL OP BOTANY at Barbon, Westmoreland. Here he made the acquaintance of the Eev. Eobert Wood, by whose advice he began the study of botany. In 1865 he came as curate to High Wycombe, where I made his acquaintance, and liad his company on many botanical rambles. After occupying a similar position at Edgbaston, Derby, and Bristol, he was appointed rector of Stirchley, Shropshire, in 1894, remaining there until 1909, when, in consequence of failing health, he went to live in Shrewsbury, where he died on Oct. 12 of last year : he was buried at Stirchley. Although it cannot be said that Painter occupied a leading position among British botanists, he did a good deal of useful work, most of which has appeared in the pages of this Journal. Among phanerogams his principal work was connected with the flora of Derbyshire, on which he published a long paper in 1881 and supplementary notes in 1889 ; these formed the basis of his volume Contributions to the Flora of Derbyshire, published in the latter year, reviewed somewhat critically by Mr. Bagnall (Journ. Bot. 1899, 318) ; a supplement, which appeared first in the Natura- list for 1899-1902, and was subsequently issued as a small volume, showed that the author himself was fully conscious of the imper- fections of his book. During his stay at Falmouth in the spring of 1898, Painter took up the study of mosses, which from that time forward formed his chief botanical interest, and occupied his attention during his holidays; papers on the mosses of Falmouth, Derbyshire, Brecon, and Cardiganshire, will be found in this Journal for 1900, 1902, 1904, and 1906 respectively. Painter was for many years a member of the Botanical Ex- change Club, and was a considerable collector, although his speci- mens were not of the first order. Many of them are in the British collection of the National Herbarium ; but his own herbarium with his geological and other collections was presented to Uni- versity College, Aberystwyth, before he left Stirchley. The English Churchman of Oct. 20, which gives some account of Painter's not remarkable ecclesiastical career, says that " in his death the Church of England has lost a faithful and devoted minister who was ever jealous for the maintenance of its Protestant principles " ; he belonged, indeed, to the straitest sect of Evangeli- calism, a school which has been steadily declining in influence since the days of the Oxford Movement. The portrait is from a photograph lent by his widow. James Britten. SHORT NOTES. Introductions at Paisley. — The following is a portion of a paper published in the Selborne Magazine for March : — " During the summer now ended the strange flora of a waste piece of land in the south of Paisley, Eenfrewshire, N.B., excited much interest and discussion among the members of the local Naturalists' Society. In addition to a large number SHORT NOTES 127 of the commoner wild flowers of the district, forty plants, either casuals for the district, or, if indigenous, of very rare occurrence in the shire, were observed. This waste land, which is about one-eighth of an acre in area, occupies part of the site of an old sandstone quarry, and has been slowly formed by the dumping of much heterogeneous material into the excavation caused by former quarrying. Road-scrapings, sand, soot, lime, broken bricks, dung, garden refuse, and the sweepings of the kilns of a local brewery have all contributed their quota to the levelling up of the ground. The resultant soil, as may be imagined, is of a very varied nature. The only reasonable explanation that can be offered of the appearance of many of the plants in this part of the kingdom is that they have been introduced to this particular waste land as seeds, through the medium of the contributions from the brewery. A collection of the forty plants was made, and after their identification had been verified by P. Ewing, Esq., F.L.S., of Uddingston, they were exhibited at a meeting of the Natural History Society of Glasgow on Oct. 25th, and subse- quently at a meeting of the Paisley Naturahsts' Society on Nov. 4th. The following plants were included in the collection : — \Banunculus arvensis L., \Delphinium Ajacis L., \Glaucium corni- culatum Curt., Sisymhrmm Sojjhia L., \Erysimum repanduvi L., E. clieirantlioides L., Gonringia orientalis Dum., Gamelina sativa Crantz., \C. sativa Crantz var. fceticla (Fr.), \Leinclium ruderale L., Thlasin arvense L., Beseda lutea L., Saponaria Vaccana L., \Silene noctiflora L., Malva rotundifolia L., fiU. parviflora L., ■\Linum angustifolmm Huds., Medicago hispida Gaertn. var. den- ticulata Willd., Melilotus officinalis Lam., M.albaDe'&r., Trifolium arvense L., \Vicia lutea L., V. bithynica L., Lathyrus Apliaca L., iCaiccalis daucoides L., ]G. latifolialu., \Biipleiiriim rotundifoUum L., Antheniis Cotula L., \Cirsium arvense Scop. var. setosum G. A. Mey. f. argenteum (Buch.- White), dehor ium Intybus L., \ Am- brosia trifida L., Lysimachia vulgaris L., jLappula echinata Gilib., Convolvulus arvensis L., \ Datura Stramonium L., \Cala- mintha Acinos Clairv., \Amaranthus retroflexus L., i Setaria viridis Beauv., Phalaris canariensis L., \Avena fatua L." ToRTULA CANESCENs Mont. — Miss C. E. Larter in her paper on Devon Mosses and Hepatics {Trans. Devonshire Assoc, for the Advancement of Science, (&c. xl. 1908) records this rare species from Devonshire, specimens in the Torquay Nat. Hist. Museum, collected in 1869 by Mrs. Griffiths, having been identified by Mr. E. M. Holmes. Last year Miss Larter had the pleasure of re- finding it in its original locality, Meadfoot Cliffs, where it was also gathered, under her direction, by Mr. G. Wrey. In February of this year among some mosses sent to me by Miss Larter for determination I detected this interesting species in young condi- tion, sent as a species of Pottia. It was with some ditticulty that I was able to convince myself that my determination was correct, for the hair-points were, comparatively speaking, short, and the Plants marked (t) have not been previously recorded for Renfrewshire. 128 THE JOUENAL OP BOTANY calyptra was quite distinctly papillose ! However, specimens sent a few weeks later with normally developed hair-points and good mature capsules set any doubts at rest ; and I think it is worth while putting on record this strange form. The calyptra varied considerably, but was, I believe, fairly constant within each tuft ; in some quite smooth, in others slightly scaberulous, while in a few tufts it was quite as rough as in, e.g., Pottia Wilsoni. I find the calyptra markedly scabrous also in specimens gathered at Stanner Eock, Eadnor, by Eev. C. H. Binstead. The fact is the more interesting because a smooth calyptra is usually given as a generic character of Tortula, as contrasted with Pottia, in which it is not unfrequently rough. The rather peculiar structure of the nerve in this species may be here referred to, as I have not seen it noticed elsewhere. Viewed from in front the nerve appears slightly thickened and spongy in the upper half of the leaf; and not unfrequently richly chlorophyllose cells in two or three rows give it a dotted appearance somewhat as in certain exotic species of Gatharinea. The explanation is seen on cutting a section. The ventral cells of the surface of the nerve are large, elongate (two or three times as high as wide), frequently appearing (in section) subcylindric and often detached from one another except at the base, and crowned with a beautiful fringe or crest of papillae. It thus shows a distinct approach to the structure of the nerve in T. atrovirens (Sm.) Lindb. (c/. Limpricht, Laubm. I. fig. 183), but the ventral cells are less numerous (in 3-4 rows), and less elongate in section ; while, on the other hand, when almost free from one another they are even more striking, and give an almost lamel- liferous appearance, as noted above, to the nerve. — H. N. Dixon. JuNcus Leersii Mars, in Berks. — Eecently a herbarium which was made by Mr. W. H. Holliday, a friend of Henry Boswell, of Oxford, came into my possession. Holliday was in the fifties a very enthusiastic bryologist, as his herbarium testifies, while his collection of local plants consisted of most carefully selected, well dried, and accurately named specimens. These were chiefly obtained from the neighbourhood of Oxford and of Great Marlow in Bucks, many of his specimens being the earliest evidence of their occurrence in the latter county. These include Lythrwn Hyssopifolia L., which Gotobed recorded from near Windsor — a locality which may be in Berkshire. The Oxford plants include the earliest voucher for Agrimonia odorata, and," from Berkshire, the Eush named above, which is a hybrid of /. conglovieratus witli /. inflexus. This he gathered near South Hinksey, in 1860, in the vicinity of /. diffusus, itself a hybrid (/. effusus x inflexus). — G. Claeidge Deuce. Deaba muealis L. in Berks.— Last September, Mr. Herbert Napier, who has been diligently studying the mosses near Oxford, brought me a specimen of the above plant which he had found in Wytham Woods, in Berkshire. I visited the spot and found that it was plentifully scattered over a small planting of young ash trees which had been recently cut, but with the exception of FOSSIL PLANTS 129 Solanum nigrum there was no other ahen present. But as the plant was quite restricted to this area, which is a nursery for young trees, we can scarcely claim the plant as a native of the county. My own belief is that the Draba was brought with ash saplings (probably from Derbyshire) to the nursery near the Abbey ; that place being found unsuitable, they were transferred to this hilly woodland (on coralline oolite) and that the plant was thus introduced. — G. Claridge Deuce. Malaxis paludosa Sw. — This was discovered in 1908 or 1909 in a small mountain bog near Brecon by Miss de Winton. She could only find it in one of many bogs of the kind in the neigh- bourhood, and there was very little of it. I have a specimen. The plant is given in Top. Bot. for no other Welsh county than Carnarvon. — H. J. Riddelsdell. REVIEWS. Fossil Plants. A Text-Book for Students of Botany and Geology. By Prof. A. C. Seward,"F.R.S. Vol. ii. 8vo, pp. 624, with 265 illustrations. Cambridge University Press. Price 15s. net. One of the most important contributions to botanical literature made during the past year was furnished by the publication of the second volume of Professor Seward's Fossil Plants. The first volume of this work appeared in 1900, and, though we have had to wait so long for tlie second, its excellence amply atones for its delay. In the present volume the study of the fossil members of the Pteridophyta has been completed, the consideration of the Seed-plants being reserved for a third part. Thoroughly up to date, it deals fully witli fossils of the Mesozoic and Tertiary age, as well as with the Carboniferous remains, and gives a large number of references to original papers, thus enabling any subject to be followed out in detail. Its full descriptions of the important plants should prove very useful to students who are beginning the study of the fossil types. At the same time the scope of the book is a wide one, all the better-known forms being mentioned, and it will therefore appeal to all who study the Pteridophyta ancient or modern. The consideration of the Sphenophy Hales, which was begun in the first volume, is concluded by an account of the recently dis- covered fertile shoots belonging to that group. The Psilotales are considered in a separate chapter ; the author prefers to retain the PsilotacecB as a division of the Pteridophyta, including only Psilotum, Tmesipteris, and a few doubtful fossil forms, and does not unite them with the Splienophyllales, as several writers have recently done. Pages 30-279 are devoted to the Lycopodiales. As in the case of the other groups, the account of the fossil members of the class is preceded by a l)rief account of its recent members, in this case the genera Lycopodium, Selaginella, and Isoetes. By means of these and similar admirably selected and written sections, the 130 THE JOURNAL OP BOTANY consideration of the plants of the past is brought into close touch with the study of their modern" descendants, and a much greater interest is imparted to the book than it would have possessed had it dealt entirely with fossils. The Lycopods are treated very fully. The anatomy of the typical species, Lepidodenclron vasculare Binney, is described in detail, and useful summaries of the anatomy of nine other species are given. XJlodendroid and Halonial forms are discussed, and the principal types of Lepidodendroid cones are described. The Bothrodendrece. are given a separate chapter, while the results of much modern work is incorporated in the chapters on Sigillaria and Stigmaria. The second half of the book deals with the Ferns. The intro- ductory chapter on modern ferns is a most interesting piece of work, particular attention being given in it to the orders of ferns most closely related to some of the fossil types. In the sub- sequent chapters the past history of the OsvumdacecB, Matoninea, and other families is described and illustrated in a lucid manner. Several of the groups of Mesozoic Leptosporangiate forms are nearly related to recent genera, but, as we go back to Carboni- ferous times, forms with sporangia of the Marattiaceous type appear to become dominant. Summaries are given of a number of the genera which have been instituted for some of these types. A large number of the Palaeozoic ferns had, however, little similarity to any of the modern forms, but exhibit many features which may be regarded as primitive. A considerable amount is known of the Botryopterea, and Zygojyterecs, and they have been placed in a group to which various names have been given. Prof. Seward points out that not one of these names is altogether satis- factory, and proposes the new designation of Ccenopteridece, for this group. He thinks that the name Primofilices, which Arber in- troduced for them, implies that they were absolutely rather than relatively primitive. The concluding chapter is on the various genera of fern-like plants of uncertain affinities, probably either ferns or Pterido- sperms. While this chapter may present little to interest the general student of botany, it will be invaluable to those who wish to study the floras of the past. It gives a short critical account of those genera of fossil plants — such, for example, as Glossopteris — which entered so largely into the composition of ancient floras. These genera are founded almost entirely on- structureless leaf- impressions, and are therefore often neglected by botanists, but we urgently need more information about them. As in vol. i., we have at the end of the book a large biblio- graphy of the works mentioned in the text. It contains about six hundred references to articles from very numerous sources, and cannot fail to be most useful to students and investigators. The book is profusely illustrated with several hundreds of drawings and photographs. Most of them bring out very clearly the points of interest, but we could wish that some of them had not been reduced so much. There are few important subjects in the text which are not amply illustrated by figures. A TEXT BOOK OF BOTANY 131 The author is heartily to be congratulated on the way in which he has reviewed and summarized the enormous amount of material with which he has had to deal. Little that is really essential has been omitted, and compression has never been carried too far. It is also not a book written to support a theory, though many theories which the author holds are incidentally presented to the reader. -o- tt m A Text Book of Botanv. By J. M. Lowson. Seventh Impression. (Fifth Edition.) Pp. viii., 607, 354 figs, in text. London: Clive, University Tutorial Press. 1910. Price 6s. This book has been enlarged considerably since its first edition, but the general method and plan remain unchanged. The student is advised in the preface to read Part i. merely " in a general way" — a somewhat uninviting beginning. In Part ii. the Angiosperm is treated in detail, including morphology, physiology, and the characters of a few natural orders. Part iii. deals with Vascular Cryptogams and Gymnosperms in a series of types, together with some account of the homologies and relationships between these and flowering plants. This account is fairly clear, but its brevity, necessitated by the scope of the book, makes it somewhat misleadingly incomplete. This third part concludes with a chapter on Ecology. Part iv. comprises accounts of lower cryptogamous types, and Part v. deals with " Additional Natural Orders," concluding with " Test Questions " and a few practical hints. The type employed in the body of the book is tolerably good, but the small type of the numerous additional paragraphs is somewhat trying. The details of elementary morphology are treated throughout in a clear and succinct manner, and the reader is assisted materi- ally by the lucid and boldly-lettered diagrams. An industrious student with a good mechanical memory might readily master these details from this text-book, unaided by a tutor and without previous knowledge of the subject : the author has, without doubt, the knack of presenting facts intelligibly. There is, however, a lack of that organic continuity which we should welcome in a modern text-book ; fact is too seldom, and at best too barely, associated with function ; the subservience of structure and habit of the individual parts to the plant considered as a whole and as a living organism is not emphasized ; the student's mind is centred wholly upon each detail as it is presented in its turn ; he forgets the life to which the detail contributes its quota of service. In this connection the isolation of physiology and its relega- tion to a couple of chapters is ominous. Even here the vital principles — notably those of energy and respiration, the keynote of every living organism— take a minor place and are veiled in a tangle of detail and an array of formidable experiments, the expression of the outward and visible signs, merely, of those principles. The examinee who is asked for an account of respira- tion will think more of Sachs' apparatus than of energy and metabolism. The vast importance of the water-current^ — of its 132 THE JOURNAL OF BOTANY continual motion, its entrances and its exits, and its consequent influence on the habit of plants — here suffers in a similar way. So, again, with photosynthesis ; the immense part played by the green plant as the entrance-door of the sun's radiant energy into the organic world is unrevealed. The cliapter on Ecology has been added, doubtless to meet the requirements of the examinee. In this difficult field, where broad, vital, scientific principles are essential to the appreciation of the barest rudiments of the subject, we look vainly in chapter xviii. for any boldness in the display of those principles. Test questions 129 and 130, beginning respectively " Enumerate the . . . plants . . ." and " Write a list . . ." are significant: we should regret to find such in a public examination. The reader gains either a very misty idea, or none at all, of the plant association, that funda- mental consideration which lies at the very root of truly scientific ecology ; plant societies are named, but left undefined ; the formation, the expression of the organic succession of associations within a uniform habitat, is not even mentioned. In this section the tutor, with whom we found that we could dispense in the study of morphology, will be needed sorely. The systematic botany is dealt with in two portions isolated from each other and from the account of floral morphology ; this is in keeping with the tendency to mechanical treatment which is to be deplored throughout the book. The description of the "horizontal" floral diagram is clearly portrayed; but the all- important longitudinal section of the flower, showing the relation of floral parts to the receptacle, that great factor in classification, is left practically unnoticed. This text-book will teach the student many facts, but it will not make him think : if public examinations are to be memory- tests, the author will have rendered great service to a large and growing public — the examinee. HEW Contrihiizioni Diatoniologiche, ix-xi. Achille Forti. Venezia, 1910. Atti del Eeale Istituto Veneto di Scienze, Lettere ed Arti. Anno acad. 1909-10, tom. Ixix., parte seconda. This publication is a continuation of the excellent series of contributions on Diatoms which have emanated from Dr. Achille Eorti during the past few years. No. ix. is. a most exhaustive account of the genus Cerataulus, in which the author summarizes, with copious critical notes, all that is known concerning the systematics and geographical distribution of this rare genus. No. X. is an account of a number of recent and subfossil Diatoms from Abyssinia. No. xi. is an extensive list of Diatoms found in certain calcareous tertiary deposits at Bergonzano and Marmorito, in Italy. The three contributions are illustrated by nine photo- graphic plates of very great merit, which add materially to the value of the work. All scientific students of Diatoms will welcome further contributions from the same author. ^ q Wparp 133 BOOK-NOTES, NEWS, dc. We learn with very great pleasure that the Eev. E. S. Marshall has been elected an Honorary Fellow of the Botanical Society of Edinburgh, " in recognition of his great services to British Botany." How great these services have been none know better than the I'eaders of this Journal, to which Mr. Marshall has been a valued and constant contributor for more than twenty-five years, and we are sure that not only they, but the many others who have benefited by Mr. Marshall's knowledge, always placed freely at the disposal of fellow-workers, will join us in congratulating him in the distinction — for the number of British Honorary Fellows is limited to six — which has been conferred on him. "When it is remembered that Mr. Marshall during all these years has been engaged in active clerical work, some notion may be formed of the energy and perseverance which are salient features of his character. We hope it may be long before it is necessary to give some account of his botanical work ; those who are fortunate enough to possess plants from his herbarium know that in addition to his critical and local knowledge Mr. Marshall is conspicuous among British botanists for the admirable care with which his specimens are selected and dried. Ad multos annos ! At the meeting of the Linnean Society on January 19th, Mr. C. H. Wright read a paper on the Flora of the Falkland Islands, based upon a collection made by Mrs. Eleanor Vallentin — whose father (Mr. W. Wickham Bertrand) was one of the earliest settlers in the West Falklands — but expanded to enumerate all the plants recorded from the Falklands. An endeavour has been made to define the distribution of plants in the islands and to show what changes have taken place in the flora since the publication of the Flora Antarctica in 1847. The plants are chiefly of dwarf habit, often with aromatic leaves, and conspicuous, often scented, flowers, which are produced chiefly between November and January. The earliest to appear is Draha funiculosa Hook, f., in September. The extermination of the fox [Canis antarcticus) has rendered possible the keeping of sheep, with the result that plants previ- ously common have now become rare ; amongst these are the Tussac Grass [Poa flahellata Hook, f.). Cinnamon Grass {Hiero- chloe redolejis R. Br.), and Blue Grass (Agrojjyron rejyens Beauv.). Primula farinosa var. magellanica Hook, f., while still abundant, is much dwarfed in those islets where sheep have been introduced. Veronica elliptica Forst. f. attains a height of 7 feet, and is the tallest plant on the islands, the next being Chiliotrichum amelloi- deum Cass, (the Fachima plant). Azorella ccBSpitosa Cav. (tlie Balsam-bog) forms hard masses up to 10 ft. long and 4 ft. higli, which rapidly decay on being wounded. The flora shows a great affinity with that of Magellan and Chile. At the meeting of the same Society on February 2nd, Dr. Jackson showed a series of lantern-slides, explaining the genesis of the portrait of Carl von Linn6, painted by Alexander Roslin, 134 THE .70URNAL OF BOTANY and the various copies, including the original three-quarter length portrait now at Versailles, though in a somewhat poor condition ; and showing that the Lapland drum in the Hoffman portrait, and on the lap of the foreground figure in the engraved title-page of the Flora La])pojiica, is a magic drum, and not a hotanic press. Mr. S. T. Dunn read a paper entitled "A Eevision of the Genus Actinidia Lindl." The genus comprises twenty-four climbing shrubs with a peculiar floral structure. The name, which is derived from the Greek word axrn;, a star, refers to the remarkable aiTangement of the numerous styles, which radiate, like spokes of a wheel, from the summit of the ovary. The appearance is striking, and is only paralleled by the gyncecium of Dillenia. The species inhabit the whole of the far east of Asia, from Japan to the Malay Peninsula, and extend to Sumatra and Java. The genus has presented some difficulty in the separation and grouping of its species. This difficulty is to a great extent removed when the facts of the dis- tribution of the various species are collected and compared, for the genus is found to be naturally divisible into sections, which present a well-marked gradation both in floral and vegetative characters, as well as in geographical range. The question of systematic position has from the beginning been an open one — Bentham, in Bentham and Hooker's Genera Plantarum, placing it in Ternstrcemiacece ; while Gilg, in Engler and Prantl's Pflan- zenfamilien, makes a separate section for it in Dilleniacece. On a comparison of the actual characters by which it differs from each, it is found that in its versatile anthers, its numerous seeds, and moderate (not minute) embryo it differs from all known Dillen- iacece, ; while the only character by which it differs from all TernstroemiacecB is the presence of raphide-bearing cells. Its close connection with Saurauja and Glematoclethra, genera still more closely allied with Ternstramiacece, taken in connection with the above evidence of its leaning towards that Natural Order, seems to support Bentham's view. At the meeting of the Linnean Society on March 2nd, 1911, Mr. E. M. Holmes showed a specimen of Griffithsia globifera J. Ag. from Milford Haven; Mr. Cotton remarked on the spread of some of these alien algae in our waters. Mr. H. W. Monckton thereafter showed a series of lantern-slides from photographs taken during his visit last autumn to Sweden as a delegate on behalf of the Society to the International Congress of Geologists, especially those taken at Upsala, some of which showed places connected with Carl von Linn6, including a front view of his house in the old Botanic Garden. Dr. B. D. Jackson then showed a supplementary series of lantern-slides, chiefly from old prints, concerning the history of the old Botanic Garden. He stated that when Linne and Rosen had exchanged Chairs in January, 1742, and the former had thereby become prefect of the garden, he took immediate steps to rearrange the garden, provide glass-houses, and rebuild the house attached, which belonged to the prefect. The last slide showed the old poplar close to the entrance, the only specimen which can be regarded as coeval with Linne, inasmuch BOOK-NOTES, NEWS, ETC. 135 as the laurels and a few other veterans had heen transported to the new botanic garden early in the nineteenth century. The twelfth volume of the Proceedings of the Washington Acadeimj of Sciences is mainly occupied by a monograph of the Lichen Flora of the Santa Cruz Peninsula, California, by Albert W. T. C. Herre, in which the following new species are described : Verrucaria melas, V. Stanfordi, Cyphelium occidentalis, C. Ander- soni, Bacidia ioessa, Acarospora Hassei, A. arenosa, Placynthium duhium, Lecania Dudleyi, Lecidea j^f^t-cifica ; and a new genus, Zahlhrucknera (Z. calcarea). Another botanical paper is that on the PolytrichacecB of Western North America, by T. C. Fryes, with numerous illustrations. Annual and Biennial Garden Plants, by A. E. Speer (Murray, 7s. 6fZ. net), dealing with " their value and uses, with full in- structions for their cultivation," is the most recent addition to the flood of garden books which has been pouring out for several years and is apparently still unexhausted. It is a widely spaced, well-printed volume, alphabetically arranged, with numerous, often very poor and scrappy, illustrations by the author ; some of these, indeed, — e. g. Argemone grandiflora, Godetia — give little idea of the plant, and others, such as Foxglove, are taken from very inadequate specimens. Many of the plants included are of course so common that no description is necessary, but of others the descriptions seem very inadequate. A good deal of space is occupied by what are supposed to be English names, but does anyone ever call Schizanthus retusus "Notched Fringe-flower " or Oenothera biennis "Large Kampion "? and can " Cotgrave " be a name for the Sweet William ? Some attempt — not always successful, e. g. " Gri'ffithi" "La'blab" — is made to give the proper pronunciation of the Latin names ; misprints are numerous — " semplerfiorens," " Barbery." The cultural directions seem full and useful. We are glad to see that Floioers of the Field, as enlarged by Mr. Boulger (S.P.C.K., 7s. Qd.), has gone into a new edition — the thirty-third of the work — to which has been added a biography and portrait of the original author, the Eev. C. A. Johns. First published in 1853, the book has been to many botanists the introduction to a knowledge of British plants ; in its present form it has been once more carefully revised, and includes the latest additions to the British list, the nomenclature adopted being in accordance with the rules of the Vienna Botanical Congress, as confirmed last year at Brussels. We still regret that the craze for colour which at present dominates popular taste has necessi- tated the introduction of a number of coloured plates, which add materially to the weight of the book though but little to its use- fulness ; poor as some of these are, they are however preferable to those which appear in another publisher's issue of the work, to which the S.P.C.K. edition is in every respect superior. The small (8th) part which concludes the first volume of Mr. Williams's Prodromus Flora Britannicce contains the completion of the Uhodoracece and the Pyrolacece, with an introductory note 136 THE JOURNAL OF BOTANY and a conspectus of the orders and families of British plants. In plan and treatment it differs in no way from the preceding portions, which have been noticed at length in these pages. There is the same originality and the same wealth of out-of-the- way and interesting information, as well as the same disregard for accepted rules and conventions, e.g., in the substitution of Uva-ursi for Arctostaphylos, as to which see this Journal for 1910, pp. 183, 206, on the latter of which Dr. Jackson shows that the name, rendered as above by Mr. Williams, was originally written as two words and is therefore "excluded by analogy." The author acknowledges a grant from the Eoyal Society in aid of the publication of the work, the next part of which will be "published as soon as circumstances permit." Mr. James Kynoch (8 College Road, Brighton) has published (price Qd.) a fourth and enlarged edition of Wild Floioers of Bar- mouth and its Neighbourhood, in which has been incorporated the list published by the Eev. T. Salwey in 1863. This indeed con- stitutes more than half the book and its principal value ; it is more extensive than the title of the book implies, as it includes lists of ferns, mosses, hepatics and algae. Although it can hardly claim to be GvitiGal—Erythrcea latifolia is given as a Barmouth plant — and is erratic in its spelling, the list is interesting and sug- gestive ; it has long been out of print, and Mr. Kynoch has done well to render it accessible. Messrs. Routledge announce the publication of a volume on The Alpine Plants of Europe, by Mr. H. Stuart Thompson, which we may anticipate will be a useful addition to the numerous works already devoted to the same subject. Mr. Thompson possesses two qualifications often lacking to compilers of such books — botanical knowledge and an intimate acquaintance with the plants in their natural habitats — and these qualifications cannot fail to give a special value to his work. The last part (Feb.-July 1910, pubhshed Feb. ,1911) of The Essex Naturalist contains a Report on the Lichens of Epping Forest by Mr. R. Paulson, some notes on the coast-flora of the county, by Mr. J. C. Shenstone, and an interesting and exhaus- tive biography of Dr. Benjamin Allen of Braintree (1663-1738) who, although not himself a botanist, was a correspondent of Ray and Dale. The Selborne Magazine for March contains', besides the paper on introduced plants at Paisley, from which we give an extract on p. 126, a useful article on Botanical Photography, by Mr. P. F. Visick. By direction of the London County Council, a tablet has been affixed to No. 32, Soho-square (the National Hospital for Diseases of the Heart), where for many years Sir Joseph Banks lived. We regret to announce the death of Mr. W. A. Clarke, of Oxford, and of Colonel Beddome, whose name was for many years intimately associated with Indian botany. Notices of both will appear in an early issue. Ic 4 2b' 2b I 'f \ \ rA ^ i ■ \ '' ' 3 2c 2d ! t .^^'• -^^ 2f / ! 2f ■ "^^^ -"^ 3 b 3b' 6 b 2a 5 b / '' \h \ \ 5d \ . '^ .' '\' ^ 1 1 1 ! I ^i, 4 b 6d H?:3a^ H. N. Dixon del. HYOPHILOPSIS. &e. West, Newman proe 137 HYOPHILOPSIS, A NEW GENUS OF POTTIACE^, With Further Contributions to the Bryology of India. By H. N. Dixon, M.A., F.L.S. (Plate 513.) Since my last article on Indian mosses in this Journal (Dec. 1910), I have received for study several small collections. These include a further collection by Mr. Sedgwick, mostly about Fort Purandhar, in the Poona District of the Western Ghats ; a small number collected by Mr. C. E. C. Fischer, Deputy Conservator of Forests in Coimbatore, on the Attapadi Hills, Mysore District, in 1910, and sent to me by Rev. E. Blatter, S.J. While from Northern India Mr. L. B. Hall, F.L.S., has sent me a few mosses collected by Miss Craig at an altitude of about 7000 ft., near Ghoom, in the Eastern Himalayas. And lastly, Mr. W. R. Sherrin has sent me the packing of skins in the British Museum of (a) a flying squirrel from the Naga Hills, Assam, and (b) a squirrel from Sikkim, each containing some interesting species. These, as well as those mentioned in my last article, are dealt with here. Hyophilopsis Card. & Dixon, gen. nov. Pottiacearum. Stirps parva, gregaria, habitu inter Pottias et Hyophilas ludens. Caulis perbrevis, subsimplex, fasciculo central! latiusculo, male definito. Folia oblongo-subspathulata ; areolatio peiiaxa, e cellulis magnis subhexagono-rectangularibus pellucidis parce papulosis instructa, basilaribus laxissimis, marginalibus in 1-3 stratis angustissime linearibus, limbum distinctum angustum effor- mantibus. Theca et seta facie omnino Hyophilae. Annulus latus, persistans. Peristomium rudimentarium, e membrana annulari lutea papillosa perangusta intra annulum recondita, dentibus bre- vissimis, fragmentariis, papillosis aurantiacis saepe obsoletis, vix supra orificium exsertis instructum. H, entosthodontacea Card. & Dixon, sp. nov. (Tab. 513, fig. 1.) Dense gregaria, viridis. Caulis perbrevis, simplex vel subsimplex ; in sectione transversa rotunda, fasciculo centrali majusculo, reti reliquo homogeneo fere, laxo, parietibus firmis subincrassatis, rubris, cellulis externis omnino similibus. Folia inferiora parva, tenuissima, superiora rosulata, multo longiora, 2-3 mm. longa, latitudine perinaequalia, ssepe anguste oblongo- sublingulata, nunc elongate obovata vel subspathulata, acuta vel subobtusa nee acuminata, concava, Integra vel sub apice obscure obtuse subdenticulata, flacciclissima, tennia, pellucida vel chloro- phyllosa, marginibus planis ; costa tenuis, in summo apice soluta, in sectione duces 2, fasciculum comitum, fasciculum cellularum stereidearum vel substereidearum dorsalem, cellulas magnas inanes ventrales exhibens. Areolatio perlaxa, cellulae superiores sub- hexagono-rectangulares, irregulares, circa 30 fj. longae (apud f folii longitud.) et 8-12 /x latae, apicem versus breviores, interne sensim elongatge, basilares {dimicliam j^ytem folii occupantes) laxissima, Journal op Botany. — Vol. 49. [May, 1911.] l 138 THE JOURNAL OF BOTANY elongate rectangulares, perpellucidoe ; superiores pagina dorsali papillis majusculis 2-4, seriatim dis-positis, pellucidis plei'umque inconspicuis praeditae, parietibus omnibus tenuissimis ; cellula) ad marginem 2-3 seriebus angustissime lineares, limbum i^&vc^'ngustum optime (lefiniUim flavescentem per totmn folii amhitnm efi'ormantes. Seta ad 1 cm. longa, inferne rubra, superne flavida ; theca breviter cylindrica vel angustissime elliptica, castanea, collo perbrevi dis- tincto, opei'culo rostrato, tertiam partem tbecae aequante. Calyptra cucullata, saepius scaberula. Exothecii cellulse regulares, rectan- gulares, in seriebus longitudinalibus dispositae, orificium versus nonullae aliquot breviores, vix quadratae ; stomata sat numerosa, magna. Annulus latus, bene evolutus, ut videtur longe persistens. Peristomium generis. Sporae 8-10 ft, laeves, virides. Dioicum, vel potius revera rhizantoicum ; fl. masc. terminales, in caulibus cum femineis intermixtis, aetate propriis, imprimis verisimiliter e cauli- bus fertilibus ortis sitae. Hab. On walls and mortar, very common, Purandhar, Poena District, Oct. 1910; leg. Sedgwick (no. 119) c. fr. A verj^ distinct plant, and one somewhat difficult to place in Pot- tiacece, the fruiting characters being very nearly those of HyoiMla, while the areolation is rather Pottioid. The capsule, however, is not gymnostomous, as the basal part of a peristome is quite distinctly present, though reduced to very small dimensions. The leaves, bluntly pointed, extremely thin and pellucid, with lax thin- walled cells and a narrow, distinct, not thickened border, are, at first sight, suggestive of EntostJiodon, whence the specific name. The papillae, of course, at once distinguish them, but they are frequently inconspicuous, and often (except in profile) masked by the chlorophyll ; in their well-developed condition and on empty cells, on the other hand, they are very distinct, and their arrange- ment a very unusual one in this family ; the only close resem- blance that I remember seeing being in the lower cells of one or two New Zealand species of Tortula (e. g. T. Petricei Broth, and T. rubra Mitt.). Ancectangium Steacheyanum Mitt. Earth-banks, Fort Puran- dhar, Poena District, May, 1910 ; leg. Sedgwick (nos. 81, c. fr. ; 89, a dense, yellow, sterile form, with closely curled leaves when dry ; 106, c. fr., a vivid green form ; 107, a rather robust, sterile form). Near Binsar, Almora, 1910 ; leg. Miss E. Shepheard (no. 14), c. fr. A. Walkeei Broth. On stones, Panchgani", W. Ghats, Feb. 1909 ; leg. Sedgwick (no. 44), st. ; teste Brotherus. The cells are very short quite to base, only a very few at extreme base next the nerve shortly rectangular. A. Thomsoni Mitt. Binsar, Almora, 1910 ; leg. Miss E. Shepheard (no. 12), c. fr. A robust plant, closely resembling in habit Trichostomum mutabile Bruch. It should be noted that Wilson's name {A. crispulum) cannot be maintained, being only a nomen nudum. A. BicoLOR Ren. k Card. " Taulea, India, 12-13,000 ft., HYOPHILOPSIS, A NEW GENUS OF POTTIACE.E 139 Aug. 7, 1883; leg. Duthie, no. 4099" in Herb. Mitt., c. fr. An unnamed specimen with the above labelHng, sent from Mitten's herbarium unnamed, agrees quite well with an original specimen of A. hicolor (Sikkim ; leg. G. A. Miller, 1901), kindly sent me by Mons. Cardot ; Duthie's plant being only a little more slender and laxer in foliation. A. DuTHiEi Broth. {Molendoa Duthiei Broth, olim). Among some unnamed Indian specimens of Ancectangium put into my hands for study from Mitten's herbarium, were three labelled as follows: — "6342a, Ancectangium versicolor, Pahang Gadh, India, July 19, 1886; leg. J. Duthie " ; " 63426, Ancectangium ohtusatum, India; leg. J. Duthie"; "6242c" (probably a slip for 6342c), " Ancectangium, ln(\.\?i,; leg. J. Duthie." The two specific names are no doubt MS. names of Mitten's. The numbering, together with the absence of particulars in the labelling of the last two, would seem to imply that they were collected all at one time. At first sight there is a considerable difference between {a) and (b), the latter being of denser growth, and the leaves being, as the name implies, for the most part obtuse. Closer examination, however, shows that too much dependency must not be placed on this character, for the leaves in {h) are often acute and are very variable, while in [a) they are equally variable, and if more fre- quently acute are also at times quite obtuse ; and the structure and form of capsule, &c., being identical, they must be held to belong to one species. The plant marked (c) is certainly closely allied, and, I think, without doubt identical, although the areolation is a little larger and much more pellucid. The capsule agrees, and is marked in all three specimens by a tendency to an asymmetrical form, especially in the immature condition. The lid, which is extremely long-beaked in (a), is usually shorter in (h) and (c), but varies, and apparently the shortness is, in some cases at least, due to accidental causes. A further plant of Mitten's labelled " India ; leg. Duthie," is also a still more compact and obtuse-leaved form of the same thing. I have compared these with Ancectangium Duthiei Broth. (Alampi-la, coll. Duthie, no. 12,717) in the British Museum, original specimens of this species, and I have little doubt that they are all referable to that species. The leaf structure agrees exactly, but the Alampi-la plant has shorter, denser leaves than the laxer forms. The basal cells, the basal margin regularly re- curved (almost always on one side only), the somewhat asymme- trical capsule, all agree, and there is really no difference between A. Duthiei and any of the above plants, except in the habit, which in Mitten's (c) is indeed almost if not quite identical ; and in the pointing of the leaves, which, as I have stated, is quite a variable feature. The fruiting characters would seem to support the view that it is an Ancectangium rather than a Molendoa. This extends the distribution of A. Duthiei, as I suppose, some distance eastward, and also shows it to be a variable plant. In L 2 140 THE JOURNAL OF BOTANY this connection it is necessary to compare with it A. laxuvi C. M., Schensi, China, which I have examined in the British Museum (leg. Giraldi, det. C. Miiller, no. 2078). At first sight this differs markedly from A. Duthiei in the excurrent nerve and cells more ohscure, very papillose, and the back of the nerve papillose. But the lower leaves agree exactly with the Indian plant, the cells vary in size and distinctness, as they do in A. Duthiei. The back of the nerve also varies, and in Mitten's plant it varies from smooth to distinctly papillose {A. Duthiei is described as nervo . . . dorso Isevi), so that I do not think this can be held to have any importance. The leaf margin in the Chinese plant, while normally plane, is very occasionally slightly reflexed below. Tlie basal cells are shorter, but otherwise agree ; as also does the capsule (not at all, as described by C. Miiller, " angustissime eylindrica "), so far as the condition permits of comparison. The plants are certainly very nearly allied, but in view of the excurrent nerve, almost plane margins, and shorter basal cells of the Chinese plant, I should hesitate to unite them, unless further investigation should result in the discovery of intermediate links. Orthotrichum (Calyptoporus) Griffithii Mitt. MS., sp. nov. (Tab. 513, fig. 2.) Corticola ; dense caespitulosum, immiliivi, vix •75 cm. altum. Folia densa, erecto-patentia, j^lerumque ohtusa, rarius snbobtusa, late oblonga, ad 3 mm. longa, ad -75 mm. lata, marginibus ad basin breviter anguste recurvis, su])erne ]}lanis, sai)e omnino fere i^lanis ; costa tenuis, profunde canaliculata, dorso valde prominens. Cellulae incrassatas, subrotundo-ellipticee, 8-10 /x latae, humillime papillosae vel omnino fere laeves, basilares paullo elongatse, anguste lineares. Autoicum. Vaginula pilosa vel sub- nuda ; calyptra baud inflata, parce pilosa vel subnuda. Seta per- brevis, vix -5 mm. alta, e summa ochrea parum excedens. Theca 2)arva, immersa vel suhemergens, ad 2 mm. longa, madefacta elliptica, collo hrevissiino, circa quartam partem thecae aequante, apud basin nullo modo in setavi decurrente, sed abnvpte rotundo ; siccitate apud orificium angustata, superne leniter plicata, ve- tustate (vacua) vix sub ore constricta, cum collo paullo an- gustato subcylindrica. Striae 8, debiles, e 3-4 tantu7ii seriebus celkilarum parietibus longitudinalibus incrassatis instructae, vix ad medium thecae productae, luteo-aurantiacae ; intervallorum cellulae in seriebus phiribus (8-12) dispositce, pallide luteae, rectan- gulares parietibus tenuioribus ; stomata immersa, numerosa, per totam ^-partem inferiorem theca sparscB ; cellulae circumsistentes valde protuber antes, stomata tamen plermnque minime obtegentes. Operculum rostellatum. Peristomium fuscum, dentes in paria 8 connati, vetustate medio fissi, dense papillosi, papillis nunc irre- gulariter nunc in striolis verticalibus vel obliquis dispositis ; cilia 8, fusco-aurantiaca, filiformia, solidiuscula, nonnunquam papillosa. Sporge 16-24 /x, aurantiacae, sublaeves. Hab. " On trees, Oongar, 5000 " ; Mitten in sched. Allied to 0. Schimperi Hamm., 0. pumilum Sw., and 0. micro- carpum De Not. In the form of the capsule base it is very distinct, and only comparable to 0. Schimperi, from which it HYOPHILOPSIS, A NEW GENUS OF POTTIACE^ 141 differs, as from the other two species, in the sHghtly larger spores, and especially in the bands of the capsule, which, in the allied species, are broad, wider than the intercostal spaces, and composed of 4-0 longitudinal series of cells ; in 0. Griffithii they are very faint and narrow, often composed (except at the extreme orifice of the capsule) of one to two rows of indistinctly marked cells only, and the intervals of thin- walled cells very much wider. The leaves also are very different in form and texture, the margins but little revolute, and the cells incrassate. 0. 2^uinilum differs in the form and recurved margins of the leaves, and especially in the more tapering base of the capsule. This latter character also separates it from 0. microcarinim, which also has the overlapping cells of the stomata scarcely protuberant. Orthotrichum sp. A specimen from Mitten's herbarium, labelled " 209, Nubra, 11,000, 1 Sept. 1848 " in Mitten's hand, is, I have very little doubt, an undescribed species. It is closely allied to 0. speciosuvi, but the only matured capsule shows a marked departure from that species in the structure of the capsule. The exothecium cells in 0. speciosum are very regularly arranged in longitudinal rows, regularly rectangular-oblong in form ; in Mitten's plant these cells are irregular, both in form and arrange- ment, being of various shapes, from oblong to triangular and rounded. The calyptra is also quite naked, the capsule scarcely striated. In view, however, of the scanty material in poor condi- tion, I do not venture to describe it as new, but it is a species for which a look-out should be kept among Orthotricha from the higher altitudes in the Himalayas. 0. CRENULATUM Mitt. An unnamed specimen in Mitten's her- barium, labelled " Keris Shayuk Valley, Tibet, Dr. Thomson," is certainly this species, mixed with Drummondia Thomsoni Mitt. ; it is probably part of the original gathering. The ribs of the capsule consist each of 6-8 rows of cells ; the stomata are im- mersed, but the overlapping cells, instead of being very protuberant and greatly hiding the guard-cells, as in 0. Simicei, are scarcely prominent, and leave the guard-cells almost uncovered. The leaf- cells also are more distinctly papillose. It is no doubt a good species, differing in the denser shorter habit, the less emergent capsules, and the narrower, more concave leaves, as pointed out by Mitten, as well as in the above characters. A further specimen of Orthotrichum from Mitten's herbarium, from the Nilghiri Mts., and probably collected by Strachey, is named 0. revolutum Mitt. MS. It is very closely aUied to 0. anomaluin ; in fact, except that the stomata are almost entirely covered by the overlapping cells, and are scattered about the middle of the capsule, I can find no difference. But the material is scanty and poor, the peristome much destroyed, and it is per- haps best left undetermined. Still another plant, labelled (in a hand I do not recognize) " Mungalin/'- Sept. 1887, alt. 1000' ; only a fragment on a schist * ? Mungaliri ; the writing is indistinct. I have not been able to trace this locality. 142 THE JOURNAL OF BOTANY block," contains two species, one of them similar to 0. rupestre, but showing a very remarkable development of peristome, the teeth being almost covered to "half their length with irregular lumps of preperistome. The material is, however, too poor to found a species upon ; but it may be hoped that the plant will be rediscovered. AuLACOMiTRiUM. A scrap was sent for study, labelled in Mitten's hand " Nilgiri, Beddome, 195 pt. c. Thyridiuvi Indicmn." 1 have compared this carefully with original specimens of Aula- comitrium humillimum Mitt. (Japan ; leg. Maingay, det. Mitten, Aug. 1891) at Kew, and the two plants are certainly identical. I asked Mr. E. S. Williams to be kind enough to examine Mitten's specimens of Thyridium indicum, to see if any trace of an Aula- comitrium occurred there, but he replied : — " Under 195, Beddome, I find a sterile plant named Thyridium indicum Mitt. MS., of which there are three separate specimens, all alike, and no mixture of any sort in the packet. It is very easy to believe, how^ever, as you suggest, that so small a bit of moss, as is the specimen of Aulacomitrium, might have been accidentally mixed with the Thyridium, and really belong to a very different region." More- over, the scrap so labelled appeared in every particular of condi- tion identical with the type-specimens of A. humillimum, and I think the conclusion is inevitable that it is a fragment of the Japa- nese plant which had casually become placed with the Thyridium. Macromitrium Perrottetii C. M. Binsar, Almora, 1910 ; leg. Miss E. Shepheard (no. 18), c. fr. Macr. squarrulosuvi C. M. would appear to be nearly allied to M. Perrottetii, but I have not been able to see an authentic specimen of the former. Mons. Cardot writes that a specimen from the Nilghiris (leg. Perrottet), so named by Bescherelle, does not agree with C. Mliller's descrip- tion, in that the leaves are not spreading in the dry state, and does not therefore appear to be correctly named ; it agrees with Miss Shepheard's plant, and both only differ from M. Perrottetii in having the leaves a little smaller, and the seta shorter. I have examined M. Perrottetii at Kew, where it is well represented, and shows considerable variation in size and length of seta, several specimens being quite identical in these respects with the Binsar plant, e. g. " No. 221, Herb. Ind. Or. Hook. & Thorns., Chongtoom, ISikkim, 9000 ped., J. D. H." No. 239 in the sanie set (Nunklow, Khasia, 5000 ped., H. & T.), it may be mentioned, is something quite different, and incorrectly named. It is also perhaps worth noting that the specimen of " M. squarrulosum MillL, Perrottet, Neilgherries," in Herb. W. P. Schimper, at Kew, is certainly in- correctly determined, being a species of quite a different affinity, with leaves curled when dry. The distribution of M. Perrottetii is from the Sikkim Himalaya to Ceylon. M. ANGULOSUM Tliw. & Mitt. Nuwara-Eliya, near Galle, Ceylon, 1898; leg. J. H. Darrell (no. 135), c. fr. M. SULCATUM (Hook. & Grev.) Brid. On stones, Panchgani, Feb. 1909 (no. 34), st. ; and on trees, Mahableshwar, Jan. 1909 (no. 58), c. fr. ; leg. Sedgwick. HYOPHILOPSIS, A NEW GENUS OF POTTIACE^ 143 PoHLiA EiGEscENS Brotli. {Webeixi rigescens Mitt.). Ghoom, Himalayas, alt. 7000 ft., 1910 ; leg. Miss Craig (no. 6), c. fr. One or two stems of this apparently rare species. Brachymenium (Dicranobryum) Fischeri, Card. & Dixon, sp. nov. (Tab. 513, fig. 6.) Pusillicm, dense csespitosum, sub- nitidum, viride, setate pallidum. Caulis perbrevis, innovationibus julaceis, foliis nunc densissime nunc laxiiis imhricatis, vix 5 mm. lougis, tenellis. Folia minuta, vix -75 mm. longa, -3 mm. lata, corclato- ovata, acuta nee acuminata, concava, marginibus integerrimis, omnino planis vel ad infimam basin angustissime obscure recurvis ; costa validiuscula, in cuspidem longiusculam subintegram patulam excurrens. Eete chlorophyllosum, e cellulis subrhomboideis, 25- 35 /x longis, 8-10 /x latis, parietibus pertenuibus instructum, mar- ginalibus angustioribus, elongate oblongis, basilaribus laxe rect- angularibus. ? Dioicum. Seta tenuis, pallide rubra, 1-1-5 cm. alta. Theca erecta, symmetrica, parva, vix 2 mm. longa, collo brevi, in setam subahrwpte defluente, superne latior, sicca sub orificium breviter distincte angustata. Calyptra parva, vix sub orificium producta. Exothecii cellulae irregulares, parietibus crassiusculis, orificium versus multo angustiores. Peristomii dentes inferne saturate aurantiaci, superne pallidi, lanceolati, intus altiuscule lamellati, dorso minutissime papillosi, suturis horizontalibus aque ac linea media validis ; endostomium liberum, pallide aurantiacum, corona basilaris mediam partem dentium vel pauUo altius attingens, processus rudimentarii. Sporae minutae, 10-12 /x. Operculum conicum, siccitate ore latins. Hab. 'Closely intermixed with Philonotis imbricatula Mitt, on rocks, Muttikulam Gap, Attapadi Hills, Coimbatore, alt. 3100 ft., Sept. 1910; leg. C. E. C. Fischer (no. 4), c.fr. Closely allied to B. Borgenianum Hampe, of Madagascar and Mauritius, but having the leaves wider above and not or scarcely acuminate, the cells slightly narrower, the nerve more longiy excurrent, the capsule smaller and shorter, and in proportion wider. B. squarrosulum Card., a recently described American species, is also near it, but has the capsules inclined and some- what asymmetric, and considerably larger. B. Walkeri Broth, has the seta longer, and the capsule much longer, and with a long narrow tapering neck. The capsule is noticeable in the dry state from having the orifice much contracted, being narrower than the base of the operculum, and so forming a sort of neck between this and the expanded part of the sporangium. B. TURGiDUM Broth. Mahableshwar, 1909 (nos. 64, 65) ; Panchgani, 1909 (no. 66) ; on Euphorbia, Fort Purandhar, 1910 (no. 84) ; all c. fr. ; leg. Sedgwick. No. 64 has tho capsules a little less turgid and more erect than in previous gatherings, and the seta longer ; (65) has both forms of capsule, with the longer seta ; (84) has the leaves somewliat longer and slightly falcate. B. Walkeri Broth. Stones, Panchgani, Feb. 1909; leg. Sedgwick (no. 52), c. fr. Det. Brotherus. 144 THE JOURNAL OF BOTANY Anomobryum cymbifolium (Lindb.) Broth. Mahablesliwar, Jan. 1909; leg. Sedgwick (no. 37), st. Bryum (Areodictyon) sahyadrense Card. & Dixon, sp. nov. (Tab. 513, fig. 5.) Dense caespitosum, vix 1 cm. altum, rufo- variegatum, nitidum. Caulis parce ramosus, inferne laxiuscule, superne magis conferte subcomose foliosus. Folia elongate ob- longo-ovata suhspathulata, acuta, nee acuminata, perpellucida, marginibus omnino planis, integerrimis, costa valida, imlchre ruberrima, in cuspidem longam strictam excurrens. Areolatio anguste rbomboidea, cellulse 75-100 [i longae, 10-12 [x latse, mar- ginem versus vix angustatse ; rete basin versus elongatum, cellulis rectangularibus, infimis in seriebus pluribus saepe brevioribus, subquadratis ; omnibus echlorophyllosis perpellucidis, parietibus pertenuibus. ? Dioicum (fl. masc. baud inventi). Flos femineus nunc lateralis, nunc terminalis inter ramulos dichotomos ; foliis comatis, majoribus, magis acuminatis. Seta l'5-2 cm. alta, rubra, superne flexuoso-curvata, theca inclinata vel horizontalis, majus- cula, 3-4 mm. longa, 1-1-25 mm. lata, turgide piyfiformis collo longiusculo in setam sensim defluente, submicrostoma, operculo breviter conico. Annulus latus. Exothecii cellulse irregulares, parietibus firmis subincrassatis collencliymaticis. Peristomii dentes pertenues, pellucidi nee colorati, late lanceolati, circa 1- 1-25 mm. longi, ad basin •35--4 mm. lati, subobtusi, marginibus valde irregularibus, per totam longitudinem papillis quasi ciliolatis, facie dorsali ubique dense minute papillosi, juxta basin transverse vel oblique striolati ; sutura media angulata, seque ac sutursB transversse, pertenues, segre discernendse ; facie ventrali superne conferte inferne parce grossiuscule papillosi, suturis horizontali- bus circa 16tenuibus bene distinctis humilibusnotati vixlamellati, unaquaque sutura serie horizontali papillarum ornata. Endo- stomium aurantiacum, membrana basilaris circa quartam partem dentium aequans, parce papillosa ; processus dentes plus minusve superantes, angustissime lineares solicUusculi pernodosi, aliquando hie illic irregulariter longius appendiculati, saepe papillosi, ciliis interjectis 2-3 rudimentariis. Sporse 20-27 /x, punctulati. Hab. Eocks, Purandhar Fort, Poena District, Oct. 1910 ; leg. Sedgwick (nos. 114, 111), c. fr. Derivation from Sahyadri, the vernacular name for the Western Ghats. A distinct and very interesting species, in its vegetative cha- racters somewhat resembling B. ghatense Broth. & Dixon (v. infra), but marked by the longly excurrent vinous red nerve and larger leaves, while the peristome structure is entirely different, and, in- deed, very striking. (Fig. 5d is drawn on too small a scale to bring out the characters properly.) The outer teeth are pale and un- coloured, very delicate in texture, highly irregular and prettily fringed with papillae at the margins ; the papillae of the dorsal surface to- wards the base of the teeth arranged in lines which are usually hori- zontal, but show a pecuhar variety in direction. The processes of the inner peristome, which is free or lightly adherent to the outer, are longer — at times considerably longer — than the outer teeth, and while filiform are not delicate nor fragile, but rather robust. HYOPHILOPSIS, A NEW GENUS OF POTTIACEiE 145 They show no median Hne or perforation, and as a matter of fact each one represents only one lateral half of the complete pro- cess, the other half remaining undeveloped, or having been ab- sorbed; occasionally fragments of the missing half remain attached here and here to the side of a process, and show that if complete it would be of an extremely narrow linear-lanceolate outline, pierced by narrow slits. The striolation of the outer teeth varies considerably ; in some capsules the lines are practically always horizontal, in others they take various directions, as referred to in the quotation belovv^ from Mons. Cardot's letter. The somewhat doubtful position of the fertile flower, together with the peculiar peristome, suggest a relationship with Mielich- hoferia, under which at one time Mons. Cardot was inclined to place the plant. The development of the outer peristome seems to necessitate its reference to Areodictyon, but it raises the question of the relation of the Section Areodictyon to the re- mainder of the genus Bryum and the allied genera of Mielich- hoferia and Brachymeniuni. Dr. Hagen has already expressed the opinion that there are good reasons for elevating Areodictyon to generic rank {cf. Eev. Bry. 1907, p. 85). I quote the following from a very interesting letter of Mons. Cardot on this question : — " J'ai pense un moment a placer cette Mousse dans le genre Mielichhoferia, en la rapprochant des Mielichhoferiopsis, mais je crois qu'il serait en somme preferable de la reunir aux Bryum de la section Areodictyon. Elle se rapproche des especes de ce groupe par I'ornementation des plaques dorsales des dents de I'exo- stome, mais en differe par les bords irreguliers de ces memes dents, par les segments de I'endostome plus etroits, non perfores, appendicules au sommet, et par la nervure longuement excurrente. Je vous propose done de la placer en Bryum (Areodictyon), mais je me demande, avec MM. Hagen et Nicholson (Rev. bryol. 1907, pp. 84-85), s'il ne conviendrait pas de considerer ce groupe com me un genre particulier, qui, a mon avis, se rapprocherait plus, par la structure du peristome, des Mielichhoferia que des Bryum ou des Brachymenium. La distinction entre les Mielichhoferiees et les Bryees, basee sur la seule position des fleurs, me parait bien artificielle, car on se trouve souvent en presence de cas douteux, sur des especes appartiennant aussi bien a I'une qu'a I'autre de ces deux tribus. A mon avis, il serait bien preferable d'etablir la separation des deux tribus en prenant comme caractere distinctif la structure du peristome, et en attribuant aux Mielichhoferiees les genres a endostome pourvus d'un emembrane basilaire courte et de segments tres etroits (y compris les Stableria, les Ortho- dontium et les Areodictyon), et en ne laissant dans les Bryees que les genres a membrane basilaire 6levee et a segments plus larges. Je serai heureux de connaitre votre avis sur cette question. J'ajoute incidemment que la Mousse de Gr^ce qui a 6t6 publico recemment sous le nom de Mielichhoferia Coppeyi Card, n'est autre que le Bryum (Areodictyon) splachnoides (Harv.) C. M. de I'Himalaya, que M. Nicholson a decouvcrt aussi en Cr^te. En decrivant le peristome de cette esp6ce, M. Coppey compare tres 146 THE JOURNAL OP BOTANY justement le dessin forme par les stries cles plaques dorsales du peristome a celui que Ton observe a la face interne du pouce humain, et cette comparaison peut s'appliquer egalement a notre espece." I am strongly inclined to agree with the view here presented, but I have hesitated to make so great a change as it would involve within the narrow limits of this article, and content myself with suggesting it as probably a desirable rearrangement of these groups. Bryum (Areodictyon) ghatense Broth. & Dixon, sp. nov. (Tab. 513, fig. 3.) Humile, tenue, habitu variabile, nunc laxe nunc dense csespitosum, interdum compactum brevicaule interdum laxe elongate tenuicaule, rufescens vel pallide rubellum. Caulis brevis, innovationibus ssepe elongatis, ad 1 cm. altis. Folia caulina superiora dense comata, ovata acuta vel ovato-lanceolata acumi- nata, innovationum maxime variabilia, ovata, lanceolata, denique anguste lanceolata tenui-acuminata ; tenerrivia, perijellucida, seepe versus basin ruberrima, margine piano, elimbato, integerrimo ; costa valida, per totam longitudinem subaequa, ruber, j^eycurrens vel sapius in mispidem brevem integram excurrens. Areolatio teniiissima; cellulte magnitudine multo variabiles, cum elongatione folii pariter elongatte, illaa foliorum breviorum breviter hexagono- rhomboideae, foliorum longiorum anguste rhomboideo-lineares, ad marginem parum angustiores ; alares rectangulares. Seta tenuis, flexuosa, rubra, 2|-4 cm. alta. Theca erecta vel sub- inclinata, cum collo subsequilongo in setam sensim attenuata aut pyriformis aut clavata, micivstoma, castaneo-rufa, operculo breviter mamillato. Exothecii cellulse valde irregulares, parietibus firmis, curvatis. Annulus latus. Peristomium imrvum, luteum, maxime variabile, dentes externi angusti, inter se remoti, ad basin baud conjuncti, intus lamellis circa 15 trabeculati, dorso superne Iseves inferne dense papillosi, papillis nunc irregidaribus nunc in seriebus transversalibus obliqiiisve dispositis. Endostomium nunc rudi- mentarium, nunc e corona basilari tantum instructum ; bene vero evolutum processus tenues e membrana basilari hyalina dimidiam fere partem dentiitm aquante parvos lincarcs angustissime rimosos fugaces dentibus adha3rentes exhibens ; ciliis nullis. Sporie 20- 25 /x, sublaeves. ? ? Autoicum. Flores terminales (masculi ?) im- maturi vel abortivi in ramulis propriis solum inventi. Hab. On stones of a wall, Mahableshwar, Jan. 1909 (no. 63) ; on a stone pillar in a sunny place, Mahableshwar, Jan. 1909 (no. 61) ; on stonework of a small bridge, Lonavla, Sept. 1908 (no. 5 ; a dense compact form, with short stems, oval, shortly pointed leaves, a more flexuose seta and a more turgid capsule) ; earth-bank. Fort Purandhar, Poona District, May, 1910 (nos. 87, 88, 117) ; face of a quarry, Purandhar, Oct. 1910 (no. 125) : all c. fr. ; leg Sedgwick. Var. SATARENSE Broth. & Dixon, var. nov. (Tab. 513, fig. 4.) Elatum, ad 2 cm. altum. Folia omnia angustiora, longiora, an- guste lanceolata vel linear i-lanceolata, angustissime temci-acumi- HYOPH[LOPSIS, A NEW GENUS OF POTTIACE^ 147 nata, una cum costa excurrente longe subulata ; areolatio angustior, elongata, juxta marginem ssepe in pluribus seriebus angustissime linearis. Hab. On a stone pillar, Mahableshwar (with the type, no. 61), Jan. 1909 (no. 62), c. fr. ; on earth-banks, Fort Purandhar, May, 1910 (no. 94), c. fr. ; leg. Sedgwick. Frequently found with the type. Strikingly different from the type in leaf-form and areolation, and it is indeed only the extreme variability of the typical form in these characters, together with the lack of any distinctive characters in the sporogonium, that has induced us to consider this plant as a variety only. It is quite possible, however, to find on the typical plant leaves with an approach to the outline and with almost the identical areolation of the variety. The fruiting characters present no differences. Derivation, from Satara, the district in which Mahableshwar is situated. Bryum ghatense is a very distinct species, at once separated from the allied species of Areodictyon by the slender habit and narrow, delicate leaves, with excurrent nerve ; the leaves for the most part at least tend to a narrow acuminate outline quite foreign to others of the section. B. sahyadrense, while resembling it in the leaf-form and structure, is a more robust plant, with larger leaves and a quite different peristome. While, however, it is easy to point out the difference from the allied species, B. ghatense presents many perplexities, and con- siderable doubt as to its true position. Mons. Cardot indeed considers it to be a Brachymeninm. The fact is that in its peristome (as well as in its vegetative characters) it exhibits a very wide range of variation, so much so that it is scarcely too much to say that, judged by the peristome alone, it would be often necessary to refer two plants from the same tuft, the one to Bryum, and the other to Brachymeninm, while a third plant might quite justly be placed in Haplodontium ! In fact, I have several times in the course of examination been driven to wonder if there were not two or three species inextricably mixed up, identical in all other characters, but differing only in peristome ! It is not only the endostome that varies in development, but the dorsal surface of the outer teeth presents notable differences in sculptur- ing. Usually they are papillose, with the papillae irregularly scattered (and it is this character principally which leads Mons. Cardot to place it in Brachyvienium) ; at other times, however, they are almost smooth, while in other cases, and usually where the endostome is most highly developed, tliey present a striolated appearance, more or less distinct, and curiously resembling, in the varying direction of the striaj, those of B. sahyadrense, already described (though in other points entirely distinct). I am there- fore inclined to consider this as indicative of the true position of the plant under Areodictyon. It is evidently a very plastic type, and one is inclined to hazard the suggestion as to whether we may not see in it a sort of starting point for tlie two groups of Brachymenium and Bryum ? Areodictyon. Apart altogether from 148 THE JOURNAL OF BOTANY the peristome, there is a very striking resemblance in many features between this (with B. sahyadrense) and certain South Indian species of Brachymenium. The vegetative characters are so closely alike that it would appear often quite impossible from sterile plants to tell to which genus a plant belonged. The cap- sules in Areoclictyon also frequently show a marked departure from the typical Bryum capsule, being inclined or suberect, nearly or quite symmetrical, and frequently also microstomous — a very usual character of Brachymenium — and it is indeed absolutely indeterminal)le from the general character of the sporogonium whether a plant belongs to Brachymenium or Areoclictyon. When to this is added that the sole peristome characters by which the § Areoclictyon can ultimately be separated from Brachymenium are striolate outer teeth, low basilar membrane and long narrow processes of the endostome, it will be obvious that a plant of the nature of B. gJiatense, showing at once striate and estriate teeth, a varying basal membrane, which may at times be almost obsolete, and processes which more frequently than not are altogether un- developed, offers a fruitful field to systematists, especially to the school addicted to what may be described as micro-taxonomy. In tab. 513 (fig. 3cl) I have figured the most highly developed (but less frequent) form of peristome. It is perhaps worth while to mention that there does not appear to be any correlation between these peristome distinctions and the vegetative characters of var. satarense. Bryum coronatum Schwaegr. Mauarghat, Malabar District, 500 ft. alt., Oct. 1910 (no. 1), c. fr. ; and Attapadi Hills, alt. 2300 ft., Oct. 1910 (no. 9), c. fr. ; leg. C. E. C. Fischer. B. argenteum L. Muttikulam bungalow floor, 3100 ft. alt., Attapadi Hills, Sept. 1910; leg. C. E. C. Fischer (no. 3), c. fr. A slender form, entirely different from the following. Var. AUSTRALE Rehm. MS. in sched. Syn. Bryum oranicum C. M. in Hedwigia, 1899, p. 68. Differs from the type in tlie very robust stems, turgidly julaceous with the densely imbricate, suborbicular, cochleariform leaves, the upper third or half hyaline, without chlorophyll. Hab. Castle Rock, Western Ghats, Oct. 1892 ; leg. G. A. Gammie, comm. Sedgwick (no. 100), st. Attapadi Hills, 5000 ft. alt., Oct. 1910 ; leg. C. E. C. Fischer (no. 10). c. fr. Of all the forms of B. arcjenteum I have seen, several of which have been elevated to specific position, this is by far the most marked, and 1 think well deserving of varietal rank. The seta is rather long in the Attapadi plant, but otherwise the fruit does not seem to offer any differences from that of B. arcjenteum. In searching through the material of B. argenteum at the British Museum, I have found nothing at all comparable to the plants in question except (a) Rehmann, M. austro-africani, no. 260. This C. Midler made the type of his Bryum oranicum, but I am fully in agreement with Brotherus (Engler &'Pva.r\t\, Pflanzenfam., Musci, p. 586) in retaining for it varietal rank only, in which case HYOPHILOPSIS, A NEW GENUS OP POTTIACEJi; 149 it seems desirable to retain the name given to it by Eehmann ; (b) two specimens from Southern India, viz. one from Ceylon, " Bry. argenteum L., Centr. Prov,, Ceylon, C. M. Ill " (i. e. Ceylon Mosses, No. Ill, issued by Thwaites) — the middle specimen only — the other a specimen from the Nilghiris, leg. Schmidt in Hampe's herbarium ; part of this is separated as having " theca subrotunda," evidently a "sport" or pathological condition only. Numerous specimens from the same localities are quite typical B. argenteum. The var. australe appears to be a well-marked variety with a wide but at present clearly defined area, viz. the South of India and the South and East Central Africa. It appears to occiu' on Mount Ruwenzori, for, as Mons. Cardot points out to me, the plant recorded by Dr. Negri, on p. 17 of his Memoir on the Mosses of Buivenzori of the Duke of the Abruzzi's Expedition, as B. ellipsi- folium C. M., probably belongs here ; the description, " Li dimen- sion! sono assai maggiori di quelle del B. argenteum, e le foglie caratteristicamente orbiculari," does not apply to C. Milller's species, but agrees well with the var. australe. The Indian plants are slightly more robust than Rehmann's, but in all other points they are identical. C. Miiller makes a purple nerve a strong character of his B. oranicum (" nervo angusto intense purpureo "), but the British Museum specimen of Eehm., no. 260, shows no trace of such a structure; the nerve at the most is reddish at the base of some of the leaves. B. PSEUDO-ALPiNUM R. & C. var. LATiFOLiuM Card. & Dixon, var. nov. Folia pro more latiora, 1*5-2 mm. longa, -G-"? mm. lata ; rete laxius, e cellulis rhomboideis, 50-60 ii latis, 10-13 /x latis, instructum. Hab. Purandhar Fort, Poona District, Oct. 1910 ; leg. Sedg- wick (no. 110), st. A rather strongly marked variety, bearing somewhat the same relationship to the type that B. Mildeanum has to B. alinnum. The leaf-apex is sharply denticulate, though the character is often masked by the strongly recurved margin. Mons. Cardot informs me that B. p)lumosiforme Ren. & Card, must be referred to B. pseudo-alinmLin. B. WiGHTii Mitt. {B. strigosumWih.). In lax carpet on rocks, Mahableshwar, Jan. 1909 ; leg. Sedgwick (no. 45), st. B. strigosuyn, it should be noted, is a nomen nudum, being un- accompanied by any description. Rhodobryum roseum Limpr. Ghoom, Eastern Himalayas, 1910 ; leg. Miss Craig (no. 1), st. Mnium rostratum Schrad. Naga Hills, Assam, comm. W. R. Sherrin (no. 7), st. Specimens sent to the British Museum from the above locality exhibit the " accessory leaves " described by me in Rev. Bry. 1909, p. 141, very well. Orthomnion crispum Wils. On twigs of trees in a damp place, Mahableshwar, Jan. 1909 ; leg. Sedgwick (no. 50), st. Cf. Rev. Bry., loc. cit. 150 THE JOURNAL OF BOTANY Explanation of Tab. 513. Fig. 1. Hyophilopsis entosthodontacea Card. & Dixon, a, plant, natural size, h, b', leaves, x 20. c, leaf apex, x 50. d, upper, e, basal cells, X 200. /, peristome viewed from within, x 200. g, stem section, x 200. Fig. 2. Orthotrichitm Griffithii Mitt. a, leaf, x 20. i), &', leaf apices, X 20. c, upper cells, x 200. d, base of capsule, x 20. e, peristome, x 100. f, stcma, front view, /', do., in profile, x 100. Fig. 3. Bryuvi ghatense Broth. & Dixon, a, plant, x 2. b, b', leaves, x 20. c, upper cells, x 200. d, peristome teeth, x 100 (dorsal view on left, ventral on right). Fig. 4. Do., var. satarense Broth. & Dixon, a, plant, x 2. b, b', leaves, x 20. c, upper cells, x 200. Fig. 5. Bryum sahyadrense Card. & Dixon, a, plant, x 2. b, leaf, X 20. c, upper cells, x 200. d, peristome, x 75 (dorsal vew to left, ventral to right). Fig. 6. Brachyvienium Fischeri Card. & Dixon, a, plant, x 2. b, leaf, X 20. c, upper cells, x 200. d, peristome, dorsal view, x 100. ALABASTKA DIVEKSA.— Part XX. By Spencer le M. Moore, B.Sc, F.L.S. 1. New or Bare Tropical African Plants." Bubiace^. Randia vestita, sp. nov. Frutex (vel arbuscula) spinis sat longis rectis instructis, ramis validis cortice cinereo rimoso ob- ductis efoliosis ramulos plures breves vel brevissimos foliosos sgepeque floriferos infra spines emittentibus, foliis parvulis oblongo- obovatis rarius obovatis obtusissimis basi in petiolum brevem gradatim attenuatis firme membranaceis utrinque griseo-velutinis subtus pallidioribus, stipulis ovatis sursum attenuatis apice aciitis velutinis, floribus pro rata submediocribus ssepissime solitariis subterminalibus breviter pedunculatis, ovario calycem excedente turbinate ut pedunculus et calyx campanulatus ore dentatus breviterve lebatus fulvo-velutine, cerellae tube late cylindrico extus inferne glabro superne albe-sericeo intus infra medium annuloso-villoso lobis suborbicularibus tubo duple longioribus, antheris breviter exsertis sessilibus anguste lineari-lanceolatis superne angustatis, style incrassate sub apice co'ntracto cerellae tube aBquilengo, stigmate alte bilobe. Hab. Salisbury ; Band, 1395. Bami 5-7 mm. diam. Spinse vulge 5-10 mm. long. ; ramuli ssepissime 5-15 mm. long., foliorum stipularumque reliquiis dense obtecti. Folia 1-5-2 cm. long., 7-12 (rare 15) mm. lat. ; cestae secundariae utrinque 4, nunc patulae nunc ascendentes, rectae levissimeve arcuatse. Stipulae summum 5 mm. long., subscariesae. Pedunculi circa 3 mm. long. Flores albi. Ovarium 6 mm. long. Calyx (anne siccatiene?) rare irregulariter ruptus, 4 mm. long., lebi dum adsint summum 1-5 mm. long. Corollae tubus 5-5 mm. long., 5 mm. lat. ; lobi 8-5 mm. long., supra medium 8 mm. lat. Antherae 5 mm. long., inferne 1 mm. superne "5 mm. lat. Stylus * The types of the new species here described are in the National Herbarium. ALABASTRA DIVERSA 151 5'5 mm. long. ; stigmatis lobi late oblongi, ol)tusissimi, pube- scentes, marginibus recurvis, 3'5 mm. long., fere 2 mm. lat. This differs from B. nilotica Stapf in the velvety leaves and flowers, the toothed or at most shortly lobed calyx, larger corolla- lobes, long and narrow anthers, and the longer turbinate ovary, which last character most probably points to difference in the shape of the berries. The spinose habit distinguishes it from the recently described B. lasio-phy lla K. Krause and B. torulosa K. Krause, both of which have larger differently clothed leaves, calyx longer than ovary, and shorter anthers. Moreover, the calyx of B. lasiophylla is lobed to the middle, and its ovary is hemi- spherical, while the other has, inter alia, corollas with a tube three or four times as long, and lobes half as long again. Near this and virtually indistinguishable, without careful examination, is Randia Taylori, sp. nov. Vestitu, ramis ramulisque, spinis, foliis, infiorescentia prsecedentis, stipulis abbreviatis ovatis ob- tusis vel acutiusculis circa 2 mm. long., ovario ovoideo segre 4 mm. long., calyce tubuloso ovarium excedente in toto 6 mm. long, lobis saepissime ambitu subsemicircularibus vulgo obtusis- simis basi baud contractis 2 mm. long., corollae 5-lob3e lobis 5 X 4-5 mm., antheris ovato-oblongis apice acutiusculis fere 3'5 mm. long. 1-5 mm. lat. Hab. German East Africa, between Zanzibar and Uyui ; Bev. W. E. Taylor. This differs from B. nilotica in the ovary and calyx and the smaller corollas. Oxyanthus Gossweileri, sp. nov. Suffrutex humilis crebro ramosus ramis validis cortice crasso badio obductis foliorum de- lapsorum cicatricibus signatis, foliis parvulis suborbicularibus coriaceis utrinque pube copiosa grisea onustis petiolis elongatis dense piloso-pubescentibus suffultis, stipulis , infiorescentiis subsessilibus bracteatis perpaucifloris, ovario calyce breviore cy- lindrico dense pubescente, calycis dense pubescentis lobis lineari- lanceolatis acuminatis tubum paullo excedentibus, corollae tubo limbum multoties excedente piloso-pubescente lobis oblongis acutis extus dense fulvo-pubescentibus, filamentis quam antherae oblongas obtusae apice brevissime apiculatae paullulum longioribus, ovario 2-loculari, stigmate clavato lobis brevibus deltoideis obtusis. Hab. Angola, bank of River Kiuvivi near Kassuango ; Goss- weiler, 3275. Folia modo l"5-2-5 cm. long, et lat. ; ordinis secundae costae utrinque circa 6, arrecto-ascendentibus, costulas quam se ipsae minus perspicuas frequenter gignentibus ; petioli circa 1"5 cm. long. Infiorescentiae axis dense pubescens, circa 5 mm. long. ; hujus bracteae ovato-lanceolatae, 3-4 mm. long. Flores dilutissime brunnei, suaveolentes. Ovarium 3 mm., calycis pars indivisa 2-5 mm. et lobi 3-3-5 mm. long. Corollae tubus vix 6 cm. long., circa 1'5 mm. lat., faucibus ipsis extus densius pubescentibus 5 mm. lat. ; lobi 1 cm. long. Filamenta 5 mm., antherae 4 mm. long. Stigmatis lobi 1x1 mm. 152 THE JOURNAL OF BOTANY A very remarkable plant on account of its lowly habit and small leaves. Canthium Randii, sp. nov. Arbor altitudine mediocri, ramulis ultimis foliosis cinereo- vel fusco-corticatis glabris novellis scabriuscule pubescentibus, foliis majusculis brevipetiolatis ovatis vel ovato-suborbicularibus apicem versus attenuatis obtusis basi late rotundatis nonnunquam leviter cordatis papyraceis supra scabriusculis subtus ad nervos scabriuscule pubescentibus in sicco utrobique viridibus, stipulis triangularibus superne in acumen sat longum productis mox dehiscentibus dorso scabriusculis aliquanto resinosis, floribus parvis 5-meris in cymis axillaribus pedunculatis laxis quam folia brevioribus scabriuscule pubescentibus digestis cymulis scorpioideis plurifloris, pedicellis ovario subaequilongis, ovario turbinato scabriuscule pubescente, calyce valde abbreviato ore truncate, corollae tubo late cylindrico lobis oblongo-ovatis acutiusculis aequilongo faucibus villosis, filamentis brevibus corollas ori affixis antheris leviter exsertis ovatis subito breviter acutatis, stylo crassiusculo sub stigmate attenuate glabro stigmate breviter exserto oblate sphseroideo utrinque truncate longitrorsum sulcato. Hab. Ehodesia, among granite kopjes near Salisbury ; Rand, 1393. Folia 7"5-ll cm. long., 5-9*5 cm. lat. ; costse secundariae utrin- que 5-6, pag. sup. planae, inf. prominulae etsi tenues, ascendentes vel ascendenti-patulse, paullulum arcuatae ; petioli 5-10 mm. long., scabriuscule pubescentes. Stipulse fuscae, 7-8 mm. long. Cymae bene evolutae 3-5 x 6 mm. Pedunculus + 1 cm. long. ; pedicelli circa 1 mm. Flores albi. Ovarium 1-2 mm., calyx -25 mm. long. Corolla tota 5 mm. long. ; tubus 2-5 x 2-5 mm. Filamenta -3 mm., antherae paullulum ultra 1 mm. long. Stylus 3 mm. long., stigma 1 X 1"25 mm. The inflorescence is much that of C. discolor Hiern, but the leaves and fiowers have many points of difference. Fadogia Kaessneri, sp. nov. Caulibus pluribus ascendenti- bus sesqui-bispithameis subdistanter foliosis tetragonis glabris, foliis plerumque ternatis brevissime petiolatis ovatis vel ovato- oblongis obtusis vel obtusissimis rare retusis basi rotundatis coriaceis utrinque glabris, stipulis a basi lata subulatis intus pubescentibus, floribus pro rata parvis in fasciculis axillaribus pedunculatis paucifloris digestis, pedicellis ovarium plane exce- dentibus ut pedunculus foliis brevier glabris, ovario subhemi- sphaerico 3-loculo glabro, calycis limbo ovarfo breviore brevissime 5-dentato, corollas extus glabrae tubo sursum levissime ampli- ficato intus villoso lobis 5 oblongo-lanceolatis acuminatis tubo circa aequilongis intus minute sericeo-tomentosis, antheris ori insertis, stylo exserto dimidio inf. incrassato superne attenuate glabro, stigmate pileato. Hab. N.W. Ehodesia, Malanguala Kiver, under trees; Kdss- ner, 2064. Folia 2-5 -3-5 cm. long., 12-15 mm. lat., in sicco utrobique brunnea ; ord. sec. costae utrinque 4, baud sine difficultate cer- nendae, anfractuosae ; reticulum satis laxum ; petiolus summum ALABASTRA DIVERSA 153 1"5 mm. long. Stipulge circa 4 mm. long. Fasciculi plerique 3-4-floii, in toto 2-5 cm. long. ; pedunculus saepius 1-1-5 cm. long. ; pedicelli + 5 mm. Ovarium 2 mm. long., calyx 1-5 mm. Corollas tubus 4'5 mm. long., basi 3 mm., ore 4 mm. diam. ; lobi 5 mm. long. Antherae oblongge, acutae, 2-5 mm. long. Stylus aegre 8 mm. long. ; stigma 1'25 x 1'25 mm. This has the appearance of F. obovata N. E. Br., but is quite different in the flower. The affinity is with F. glaberrivia Schweiuf. Fadogia salictaria, sp. nov. Satis elata, erecta, sursum ramosa, ramis g)'acilibus tetragonis glabris, foliis ternatis raro oppositis lineari-lanceolatis obtusis basi in petiolum brevissimum nee semper aspectabilem gradatim extenuatis membranaceis utrin- que glabris, stipulis subulatis puberulis, floribus pedicellatis pro rata parvis in axillis fasciculatis fasciculis sessilibus breviterve pedunculatis saepius 2-3-floris, ovario hemisphterico glabro, calyce ovario breviore denticulato-undulato, corollse fere usque medium divisae extus glabrae tubo cylindrico-infundibulari intus villoso lobis anguste ovato-oblongis acutis intus minute sericeis, antheris ori insertis oblongo-ovatis obtusis, stylo exserto superne attenuato glabro, stigmate pileato. Hab. Congo Free State, under trees at Kipaila ; Kcissner, 2536. Folia solemniter 4-5 cm. long., juxta medium 5-8 mm. lat. ; costae secundarias utrinque 6-8, subtus prominulae ; reticulum obscurum, maxime laxum ; petioli summum 2 mm. long. StipulaB circa 1-5 mm. long. Pedunculi 0-6 mm. long., pedicelli 1-3 mm. Ovarium 1 mm., ealycis limbus '5 mm. long. Corollae tubus 3-5 mm. long., ima basi 1-5 mm. ore 3 mm. lat. ; lobi 4 mm. long. Antherge 2 mm. long. Stylus 6 mm. long. ; stigma 1-25 x 1-25 mm. Differs from F. stenophylla Welw., inter alia in habit, usually ternate leaves, and shape of corolla-tube. Kassner's no. 2350, from Kitimbo, Congo Free State, seems from the description to be F. arenicola K. Schum. & K. Krause (Engler, Jahrb. xxxix. p. 544), and the Museum spec-mens have been provisionally so named. Campanulace^. Wahlenbergia saginoides, sp. nov. Annua, subspithamea, caulibus strictis tenuibus sca))riusculis inferne foliis imbricatis onustis superne foliis sparsis minoribus, foliis parvis sessilibus anguste linearibus acutis margine cartilagineis raridenticulatisque vel integris, cymis paniculatis laxis crebro ramosis bracteatis plurifloris, floribus parvulis pedicellatis, ovario 3-loculo henii- sphserico quam ealycis lobi subulato-lineares breviore, corolUe vix usque medium divisae tubo calycem breviter superante lato superne leviter dilatato lobis ovatis obtusis, stigmate breviter exserto 3-ramoso. Hab. Rhodesia, Victoria ; C. F. H. Monro, 649. Planta circa 12 cm. lat. Folia inferiora + 5 mm. long., circa Journal of Botany. — Vol. 49. [M.\y, 1911.] m 154 THE JOURNAL OF BOTANY 1 mm. lat., superiora + 2-5 mm. long. Paniculae 6-8 cm. long., usque ad 10 cm. vel etiam magis diam. ; harum bracteae anguste lineares, + 2 mm. long. Pedicelli + 5 mm. long., ascendenti- patentes. Flores verisimiliter albi. Ovarium 1 mm. long. Calycis lobi 1-6-2 mm. long. CorollcB tubus 2-5 mm. long., ima basi 1-5 mm. ore 2-5 mm. diam. ; lobi 2 mm. long. Stylus 3-5 mm. long. ; stigmatis rami '75 mm. long. Nearest W. leucantha Engl. & Pilg., which is a stronger grow- ing plant 30 cm. high, with a longer and relatively narrower corolla, lobed to only one -third of its length. Its habit is that of Sagina nodosa L. Sapotace^. Mimusops (QuATERNARiA § Integr^) Monroi, sp. nov. Ea- mulis teretibus cito glabris cortice cinereo longitrorsum rimoso mox circumdatis frequenter foliosis (novellis arete minute ferru- gineis), foliis lanceolatis vel lanceolato-oblongis basi apiceque obtusis margine cartilagineis coriaceis supra glabris subtus primo minute ferrugineis dein glabris costis secundariis utrinque circa 10 supra vix subtus baud sine difficultate cernendis petiolis 4-5-plo brevioribus fultis, fioribus pro rata submediocribus in axillis 2-3-nis, pedicellis petiolos saepissime excedentibus minute ferru- gineis, calycis segmentis 8 lanceolatis obtuse acutis extus ferru- gineis, corollse lobis 8 lineari-lanceolatis acuminatis, appendicibus corollas lobis aequilongis iisque similibus, antheris lanceolatis apiculatis quam filamenta pauUo longioribus, staminodiis lineari- lanceolatis caudato-acuminatis quam appendices paullo breviori- bus dorso villosis, ovario subgloboso sericeo, stylo crassiusculo ima basi sericeo ceterum glabro. Hab. Ehodesia, Victoria ; Monro, 761. Foliorum limbus solemniter 5-9 cm. long., l-5-2'8 cm. lat., pag. sup. olivaceo-badius, pallidissime nitens, pag. inf. viridis ; costa media supra plana, subtus maxime eminens ; reticulum segre aspectabile ; petioli 1-2 cm. long., canaliculati, ferruginei cito glabri. Pedicelli plerique 2 cm. long. Calycis segmenta 7 mm. long. Corollae tubus 1 mm., lobi 6 mm. long. Filamenta 2 mm., antherae 3 mm. long. Staminodia 5 mm. lon.g. Ovarium 2 mm., stylus 7 mm. long. To be inserted next M. marginata N. E. Br., which has terminal inflorescences, longer pedicels, and staminodes only half as long as the corolla-lobes. Mimusops (QuATERNARiA § Laciniat^) decorifolia, sp. nov. Ramulis subteretibus dense etsi minute ferrugineis deinde glabris (novellis ferrugineis), foliis quam petioli 3|-plo longioribus ob- longo-ovatis vel obovato-oblongis obtusissimis apice ipso saepius leviter retusis basi obtusis margine cartilagineis tenuiter coriaceis supra glabris subtus in nervis sparsissime ferrugineis alibi minute puberulis costis secundariis utrinque circa 15 (additis aliis ord. tertiae his aegre asquiprominentibus) patentibus sub margine eximie arcuatis, fioribus pro rata parvis, pedicellis in axillis saepius 1-2 quam petioli brevioribus ut calycis segmenta minute ALABASTRA DIVERSA 165 ferrugineis, calycis segmenta 8 oblongo-ovatis acutis coroUse circa sequilongis, corollas segmentis 8 oblongo-ovatis acutis minute denticulatis, appendicibus 8 petalis oppositis iisque sequilongis ad medium usque vel magis alte 2-4 lobis segmentis oblongis vel oblongo-linearibus acutis, staminibus corollse segmentis paullo brevioribus antheris ovato-lanceolatis acuminatis quam filamenta plane longioribus, staminodiis staminibus circa »quilongis anguste lineari-lanceolatis acuminatis dorso villosis, ovario subgloboso sericeo, stylo crassiusculo glabro. Hab. Ehodesia, Victoria ; Monro, 811. Foliorum limbus 6-5-7 cm. long., 3-3-3-8 cm. lat., in sicco pag. sup. olivaceus subtus dilute griseo-viridis ; costa media supra plana subtus eminens ; costulae ut reticulum laxiusculum utro- bique £eque aspectabiles ; petioli minute sed dense ferruginei, canaliculati, 1-8-2 cm. long. Pedicelli circa 1-5 cm. long. Flores pansi circiter 1 cm. diam. Calycis segmenta 6 mm. long. CorolltB tubus 1-5 mm., lobi segve 5 mm. long. Filamenta 1-75 mm. long., antherae 2-5 mm., staminodia 4 mm. Ovarium 2 mm. long., stylus 4-5 mm. Near ill. PoJiLii Engl., which has differently shaped leaves, longer pedicels, lanceolate corolla-lobes with irregularly laciniate and denticulate appendices. ASCLEPIADE^. Asclepias Goss-weileri, sp. nov. Planta tuberosa caulibus simplicibus ascendentibus subspithameis plurifoliosis strigoso- pilosis, foliis omnibus oppositis lanceolato-oblongis obtusis basi in petiolum brevem angustatis pergamaceis utrobique praesertim vero pag. inf. pilis strigosis sparsissime indutis, umbellis pro caule 3 laterahbus pedunculatis corymbosis 3-5 fioris, pedunculis in- ferioribus folia excedentibus omnibus pilosis, bracteis filiformibus pilosis quam pedicelli brevioribus, calycis segmentis lineari- lanceolatis acutis patentibus dorso marginibusque pilis strigillosis sparsim onustis, corollae alte partitas segmentis ovato-oblongis obtusissimis quam calycis segmenta paullo longioribus, coronas phyllis e basi columnas stamineas ortis et eandem paullulum exce- dentibus erectisbasisolummodo cucullatis superne oblongo-lanceo- latis obtusis leviterque concavis utrinque basi dente arrecto obtuso antheris applicato et una cum parte cucullata harum dimidium inferius occultante instructis cucullo intus apicem versus pulvino pubescente induto, antherarum alls prominentibus appendicibus ovatis supra stigma inflexis. Hab. Angola, Camoma, Ambaka ; Gosstveiler, 4557. Folia in sicco glauca, 3-4 cm. long., 7-10 mm. lat. ; petioli circa 3 mm. long. Pedunculi inferiores fere 5 cm. long., rehqui 1-3 cm. ; bracteae adusque 5 mm., pedicelli 10 mm. long. Flores monente cl. detectori fusco-purpurei. Calycis segmenta 8 mm. long. Corollae tubus 2 mm. long., segmenta 10 X 4-5-5 mm. Coronas phylla 5-5 mm. long. ; dentes basales 1-5 mm. long. Columna staminea 4-5 mm. long., antherarum ala3 eegre 2 mm. M 2 156 THE JOURNAL OF BOTANY This is very near A. pulcliella N. E. Br., its chief pecuHarities being the shorter pedicels to the dusky purple (not rose-coloured) flowers, the less hairy calyx, the longer and relatively narrower segments of the corolla, and the longer coronal lobes with their prominent basal teeth, which reach much higher up the staminal column. I have to thank Mr. N. E. Brown for kindly examining a flower, and confirming my suspicion of its being distinct from A. imlchella. The tuber is noticed in Mr. Gossweiler's note, but the only speci- men to hand is without this. Kanahia Monroi, sp. nov. Fruticosa, glabra, ramis strictis subteretibus crebro foliosis, foliis sessilibus anguste lineari-lanceo- latis apice acutis pungentilDUsque crassiusculis, floribus pro rata magnis, racemis brevibus longipedunculatis paucifloris pluri- bracteatis, bracteis pedicellis multo brevioribus lanceolatis acutis diutule persistentibus, pedicellis calyces 2-3-plo excedentibus, calycis segmentis lanceolatis acutis quam corollge segmenta fere duplo brevioribus, corollae satis alte partitae segmentis oblongo- ovatis obtusis intus minute puberulis margine ciliatis patentibus, coronge phyllis ex columna staminea paullo supra basin ortis superne complicatis a latere visis subquadratis apice in dentes duo fere horizontales columnam stamineam incumbentibus desi- nentibus, columna staminea coronse phyllis sequialta, folliculis abbreviatis ovoideis apice breviter necnon obtusissime umbonatis in sicco longitrorsum rugulatis glabris. Hab. Ehodesia, Victoria; G. H. F. Monro, 1100. Caulis circa 4 mm. diam., ad nodos levissime tumidos pilis brevibus rigidis albis instructus. Eolia pleraque 10 cm. long., juxta medium 5-7 mm. lat., in sicco Igete viridia. Pedunculi 4-5 cm. long., erecti. Pedicelli 1-5-2 cm., bracteae + 5 mm. long. Flores profecto pansi 2-5 cm. diam. Calycis segmenta 7 mm. long. Corolla tubus 3 mm., segmenta 12 mm. long. Columnae stamineae basis nuda 2 mm. long. Corollae phylla 4 mm. long., horum dentes obtusse, 1-5 mm. long. Antherarum alae basi baud divaricatse, 2-5 mm. long. ; appendices triangulares, obtusae, stigmati convexiusculo impendentis ; glandula ovata, >-4 mm. long. ; caudiculge -3 mm. long. ; pollinia oblongo-pyriformia, 1*5 mm. long. Folliculi (umbone 4 mm. long, inclusa) 3 cm. long., vix 2 cm. lat. Semina 3-4 mm. long., coma circa 8 mm. A very distinct species, differing from K. laniflora Br. and K. glaherrima N. E. Br. inter alia in the short follicles, and from K. consimilis N. E. Br. in the flowers larger in all their parts, the coronal lobes reaching the same height as the staminal column, the anther-wings parallel throughout, and not divaricate at the base, &c. SCROPHULARIACE^. Manulea rhodesiana, sp. nov. Herba erecta, satis elata, ramosa, paucifoliosa, minute papilloso-puberula, foliis caulinis (radicalibus baud visis) parvis linearibus basi levissime dilatatis margine sparsim denticulatis subcoriaceis, floribus permultis secus ALABASTEA DIVEESA 157 ramulos abbreviates dense vel longiores laxiuscule digestis pani- culam thyrsoideam elongatam referentibus, bracteis linearibus calycem circa aequantibus, calycis alte partiti papillosi segmentis, oblongo-spathulatis crassiusculis, corollae tubo calycem multo ex- cedente gracili faucibus ipsis dilatato crassiusculo extus papilloso lobis inter se fere aequalibus (2 posticis paullo angustioribus) oblongo-obovatis obtusissimis intus glabris, stylo incluso, capsula obovoidea apice emarginata calyci aequilonga glabra. Hab. Ehodesia, Victoria ; Monro, 924. Planta saltern trispithamea. Rami tetragon! vel subtetragoni, longitrorsum multistriati, ut folia infiorescentiaque papillis brevis- simis albidis muniti. Folia circa 1-2 cm. long., et 2 mm. lat. Bracteae 1-2 mm. long. Flores verisimiliter aurantiaci vel am-an- tiaco-pmiicei. Calyx totus 2 mm. long. ; lobi 1-8 mm. Corollae tubus 8 mm. long., -5 mm. lat., faucibus 1 mm. lat. ; lobi patentes, vix 2 mm. long. Antherae superiores brevissime exsertae, fertiles etsi immunitae, '25 mm. long., barum filamenta '75 mm. long. ; antberae inferiores oblongae, obtusae, 1 mm. long., barum filamenta •2 mm. long. Stylus circa 7 mm., capsula 2 mm. long. Near the South African JSI. parviflora Benth., but different in the narrow leaves, the smaller calyx with spathulate segments, and the longer corollas. This is the second known tropical species of this largish South i\.frican genus. Stemodiopsis linearis, sp. nov. Suffrutex parvus, glaber, caule verisimiliter repente subtereti nudo, ramulis gracillimis fere a basi foliosis, foliis sessilibus anguste linearibus obtuse acutis margine revolutis, floribus in axillis soiitariis, pedicellis foliis bre- vioribus supra medium bracteolis duabus parvulis filiformibus onustis ascendentibus sub fructu decurvis, calycis segmentis lineari-lanceolatis acuminatis quam corolla tubus brevioribus, corollae tubo superne levissime amplificato labio postico ovato bifido antico postico paullulum breviore intus minute pubescente palato valde intruso, staminum anticorum filamentis juxta medium tortis, staminodio filiform! a staminibus remoto, capsula decurva ovoidea obtusa breviterve rostrata glabra. Hab. Congo Free State, Mt. Penga, on rocks ; Kcissner, 2964. Tota planta circiter spithamea. Foha + 2 cm. long, (saepis- sime 1-5-2-5 cm.), circa -5 mm. lat. vel etiam minus. Pedicelli plerique 7-12 mm. long. ; horum bracteolae 1-5 mm. Calyx 3-25 mm. long. Corollae tubus 5 mm. long., inferne 1-5 mm. ore 2-5 nnii. lat. ; labium posticum intus sparsim piloso-puberulum, margine ciliolatum aegre 3 mm. long.; labium anticum 25 mm. long., hujus lobi abbi'eviati, rotundati, vix 1 mm. long. Staminodium corollae tubi basin versus insertum 2 mm. long. ; staminum anticorum filamenta 3 mm., posticorum 2 mm. long. ; antherarum loculi divergentes, anguste ovato-ol)longi, -8 mm. long. Ovarium 2 mm. long., stylus 3 mm. Capsula 7 mm. long. The linear leaves serve to distinguish this from its few con- geners. 158 THE JOURNAL OF BOTANY Ilysanthes micrantha, sp. nov. Annua, parva, glabra, caule gracili simplici erecto sursum sparsissime ramose distanter folioso, foliis sessilibus anguste linearibus obtusis integris, floribus paucis axillaribus vel pseudoterminalibus, pedicellis sat longis filiformibus erectis sub fructu patenti-nutantibus, calycis abbreviati campanu- lati adusque medium divisi lobis lanceolatis acuminatis, corollae parvas tubo calycem paullulum excedente lato superne brevissime amplificato labio postico anticum breviter excedente late oblongo apice rotundato ipso emarginato labii antici lobis inter se sub- sequalibus rotundatis, staminum posticorum loculo altero minimo casso loculis ambobus basi barbatis, staminodiis magnis superne clavatis necnon longe ciliatis, capsula oblonga acuminata calycem multo excedente. Hab. Ehodesia, Victoria ; C. F. H. Monro, 1031. Planta summuni 6 cm. alt. Folia + 5 mm. long., circa •5 mm. lat. Pedicelli solemniter 1-1'5 cm. long., sub flore tenuissimi. Calyx totus 2 mm. long. ; tubus 1 mm. long., ore totidem lat. ; lobi 1 mm. long. Corollge tubus 2-5 mm. long., faucibus segre 2 mm. diam. ; labium posticum 4 x 2'5 mm. ; labii antici lobus intermedins 3x4 mm. Stamina corollge faucibus infra os fere 1 mm. inserta ; filamenta complanata, vix 2'5 mm. long. ; antheras fertiles ovatte, -75 mm. long. Staminodia leviter cohserentia, in- curva, fere 3 mm. long., horum clava 1 mm. long. Stylus 3 mm. long. ; stigma '5x1 mm. Capsula 7-9 mm. long. The plant is remarkable for the large club-shaped, coherent, ciliated staminodes and the rudimentary second cell of its stamens. (To be continued.) NOTES ON EPILOBIUM HYBKIDS. By E. H. Compton. The study of naturally occurring hybrids is at^ the present day beset with difficulties, the chief of which is the lack of ex- perimental data. The process of identification of a supposed hybrid is in nearly all cases based on the assumption that the characters exhibited are intermediate between those shown by the parents. _ The species selected as the supposed parents are usually those which differ from the plant under consideration by about an equal amount on either side. As a rule, no experimental evidence is forthcoming, and the chief criterion of hybrid-nature is intermediateness between a pair of species selected to account for the facts. This assumption is carried still further in the case of "tertiary hybrids," which are supposed, on grounds of inter- mediateness, to have arisen from the union of a hybrid with a pure species. In fact, it appears that the words " hybrid " and "intermediate" are sometimes used indiscriminately — a deplor- able, though almost logical, result of the method of identification adopted in the past. When we read in a Flora, " A large number NOTES ON EPILOBIUM HYBEIDS 159 of hybrids occur in this genus, therefore many plants will yield mixed characters," though we may accept the statement without demur, it must not blind us to the fact that the historical process would be more fitly summed up in the inverse form of words, thus : — " A large number of plants yielding mixed characters have been found in this genus, therefore extensive hybridization has been assumed." Now it is quite clear that this method, though in the hands of skilled observers it may often yield accurate results, is essentially faulty. Eecent experimental w^ork has demonstrated that a cross- bred is by no means necessarily intermediate between its parents ; it may even be indistinguishable from one of them, and show no trace of the other. The phenomena of dominance and the Men- delian segregation of unit-characters are of the greatest impor- tance to students of naturally occurring hybrids ; but the general acceptance of these facts by systematists has been tardy, and in only a few cases has a sense of their importance been expressed." Whatever may be the exact applicability of the newly discovered phenomena to the mutual relationships of the highly complex congeries of characters which separate naturally occurring species, the experimental method is now seen to be indispensable for the solution of problems of hybridity. The mode of work should be changed, from the almost mathematical search for extremes to suit a given mean, to the emasculation and artificial-pollination processes of the experimental garden. To take one or two examples. Writing of Epilobium roseum, Sir James Smith remarked in 1800 : "Is it possible to have arisen from seeds of E. tetragonum impregnated by the pollen of E. mon- tanum ? " But it was not till 1842 that Bell Salter made the suggested cross, t and found that the result was not identical with E. roseuvi. (See below.) Again, I grew plants in my garden which were absolutely indistinguishable from typical Epilobnnn hirsutum. These had, however, been grown from seed produced by fertilising the occasionally occurring white (or rather, very pale pink-tinged) variety with pollen taken from the typical red-flowered plant. They were thus true cross-breds ; and when artificially self- fertilized they set seed which produced, out of fifteen plants, eleven having the full red flower of the type hirsutum, and four having the pale pink flower of the variety, the result approximating to the well-known Mendelian ratio 3 : 1. There are thus two tests for hybridity : (1) The sowing of artificially self-pollinated seeds and the observation of the off-spring ; (2) The reconstruction of the hybrid by the union of selected parents. The first test is often inapplicable, may- be because of sterility, or the complexity of the characters involved ; but the second is usually practicable, and should * See, for instance, Moss, "The Pimpeinels," Journ. But. I'Jll, p. 44. t FInjtologist, 1852, p. 737. 160 THE JOUENAL OF BOTANY be considered essential in any study of the phenomena of hybridity/'^ The sure method of identifying a naturally occurring plant sup- posed to be a hybrid is the following. The supposed pure parental species are cultivated in a garden and crossed artificially. The re- sulting offspring (if any) are grown in the garden side by side with the wild plant, and the comparison is made. The usual experimental precautions must be taken ; and the plant should be observed over a series of years, in order to allow for the differences in perennials between the first year's growth and that of succeeding years. The experimental method is cumbrous and lengthy, but con- clusive. Its adoption would obviate such a controversy as that between Eev. E. S. Marshall and Mr. C. B. Clarke over the existence and determination of wild hybrids {Journal of Botany, xxix. 1891, pp. 225, 296 ; xxx. 1892, pp. 78, 106). Mr. Clarke stated that " the makers of hybrids often go no further than the diagnostic characters of systematists ; their hybrids are not hybrids between any two plants that ever lived, either species, crosses, or individuals, but hybrids between two of the hybrid- monger's own diagnoses." Mr. Marshall termed this "caricature, pure and simple" ; but, like all caricature, it contains an element of truth. The fact on which it is founded is that systematists have almost exclusively explained ignotium per ignotiis, and sub- stituted questionable hypotheses for experimental proofs. Though Mr. Marshall succeeded in proving his case so far as it went, the lack of experimental evidence is manifest, and it was this lack that made the controversy possible. The genus E^nlobium is remarkable for the frequency with which interspecific hybrids are produced ; and as there also exist a great number of forms and varieties, the genus is a difficult one to study. The size and detail of Haussknecht's magnificent monograph is witness to the complexity that exists in the genus. Naturally occurring " hybrids " have been repeatedly observed, and many such have been described with great care by Hauss- knecht and others ; but it seems important for the better knowledge of the genus that artificial hybrids should be made experimentally between species and varieties of constant type (as tested by raising from self-fertilized seed).f With this object I have produced * In the case of a supposed wild hybrid the speciesin whose company it occurs may give vahiable hints as to its possible parentage ; and collateral evidence is often furnished by a study of its geographical distribution with respect to that of the supposed parent species. But such ex j^ost facto evidence cannot have the cogency of direct experimental results. t The following are the Epilobium hybrids which are recorded in the literature (so far as I know it) as having been made by design : — E.Lamyi x montanum: Haussknecht made the cross both ways, and found the hybrids indistinguishable [Monograph, p. 27). E. montanum x parvijioriim : both reciprocal crosses were made by Hauss- knecht (loc. cit.), and found almost identical. E. montanum, female x roseum, male: Focke [Pfianzenmischlinge, p. 528). E. montanum, female x obscurum, male: Focke (loc. cit.). E. montanum x '' tetragonum": Bell Salter (loc. cit.) made the reciprocal crosses, and found the products identical. (See below.) NOTES ON EPILOBIUM HYBRIDS 161 certain hybrid -willow-herbs by the ordinary experimental method of emasculating the flower before the anthers dehisce, pollinating from the other species, and covering with a bag until the fruit begins to set and the corolla falls off. Two such hybrids have this year been distributed through the British Botanical Exchange Club, and it is hoped that more will be available in the coming years. The following are notes upon the two hybrids distributed so far. E. adnatum Griseb., forma stenophylln Hausskn., male x E. hirsutum Linn., female. Both parents were identified with certainty by Eev. E. S. Marshall : they were both grown in Cam- bridge gardens, as were also the hybrids (open situation in light soil). Seven hybrids were grown in 1909, and all were absolutely uniform in appearance, k plant found once only by Haussknecht at Greussen, in Thuringia (1857), was identified by him as E. adnatum x hirsutum, and was carefully described {Monograph, p. 103). From this description my artificial hybrid differs in several points, while agreeing in the main. The discrepancies do not appear to be fully explained by the use of the " forma stenophylla," '^' and future experiment can alone determine the reason for them. The following are the chief points in which my hybrid differs from Haussknecht's description : — (1) The stem is quite terete towards the top. (2) The leaves, though they half clasp the stem below, do not do so at the top. (3) The upper leaves are but little more hairy than the lower (which are almost glabrous), and cannot be described as "allmah- lich beiderseits angedrilckt grau behaart." (4) There are no long erect hairs, such as are present in hirsutum; the hairs are longer than in adnatum, &nd are obliquely directed upwards, i. e. their position is intermediate between the closely appressed down of adnatum and the erect hairs of hir- sutu)n. There are no swollen tips to the hairs, such as are found in the shorter hairs of hirsutum. (5) The pedicel of the capsule is usually about 1-5 cm. long, and the capsule itself never grows to more than 3-4 cm. (Hauss- knecht's figures are: pedicel I'O cm., occasionally reaching 1-5 cm. ; capsule 5-6 cm. long.) The hybrid lacks the long runners of hirsutum ; autumnal rosettes are formed close to the plant, their leaves being long and strap-shaped, much as in adnatum f. stenophijlla, but larger. The flowers are like those of adnatum in size, colour, and general appearance ; the stigma is clavate, slightly notched, and * I am inclined to agree with Mr. Marshall that the forma steuophylla deserves varietal rank ; several plants preserved their proper characters when raised from seed and f,'rown in my garden at Tewkesbury side by side with typical adnatum. Mr. Marshall noticed the narrowness of the leaves in this hybrid as being " greater than one would prima facie expect from a cross with the type " adnatum. 162 THE JOURNAL OP BOTANY larger than in adnatum, being about 3 mm. long ; it frequently protrudes at the tip of the flower before anthesis. The anthers seem to be usually contabescent. Flowering goes on until very late in the autumn. In no case was seed set in 1909-1910, though superficially the ovules appeared well-formed ; and attempts to fertilize the flowers with pollen from adnatum f. stenopliylla and from hirstctum were also fruitless. While most of the characters of the hybrid are intermediate between those of the parents, it may be remarked that the pre- sence of long runners, of erect, long, shaggy hairs, of terminal glands on the shorter hairs, and of large flowers (all characters possessed by hirsutum), are all apparently recessive. The study of their exact hereditary behaviour is, however, checked by sterility. E. adnatum Griseb., female X montanum Linn., male. The parent stocks originated from plants self-sown in my garden at Tewkesbury, and came true from seed. Both were quite typical, agreeing well with Haussknecht's descriptions. I raised twenty- five plants of the hybrid, and all were absolutely uniform in appearance ; they correspond closely on the whole with Hauss- knecht's descriptions of wild specimens. The following points of interest may be added to his remarks (see Monograph, p. 104) :— (1) The upper part of the stem is practically terete, only the lower part showing decurrent lines. (2) A transverse commissure joins the bases of the opposite leaves, this being a character derived from montanum. (3) The stigma is very shortly four-lobed, this condition differ- ing from that found in adnatum i. stenophylla x hirsutwn (above). Thus the results of crossing § Synstigma and § Schizostigma are not always identical as regards the characters of the stigma of the hybrid. (4) The capsules appear well-formed, but the seeds seem to be always imperfect and are without embryos. This ^vas also ob- served by Haussknecht, who states that the seeds are totally sterile. The last point is of interest with regard to Bell Salter's artifi- cial hybrid between montanum and " tetragonum " {loc. cit.). He says: " I saved seeds from the original hybrids, and sowed them. The second race was indistinguishable from the first. The seeds of these I again saved and sowed, and still no difference could be detected ; and so on to four turns, when, being satisfied of the reproductive powers of the hybrids and the permanence of the form, I discontinued the experiment." It is difficult to know what value to set upon these remarkable results, in the absence of information as to what exactly was the plant used by Bell Salter under the name tetragonum. If it were adnatum Griseb., his results conflict with Haussknecht's and mine as regards the complete sterility of the hybrid. If it were obscurum Schreb., NOTES ON LEPIDIUM 163 they conflict with Haussknecht's description of the supposed wild viontanum x obscurum, which states that " the seeds are some- times completely shrivelled, at other times only partly developed and sterile." "■' It is just possible, though extremely improbable, that Bell Salter's .plant was E. Lamyi Schultz, in which case there is a very slight chance of escape from the dilemma, for Haussknecht remarks that E. Lamyi x viontanum (as found wild and as produced artificially) has " seeds for the most part un- developed." If this means that some good seeds are produced, this gives a loophole of escape for Bell Salter's results ; but the odds against their correctness are enormous. They certainly cannot be accepted at present without hesitation, and further experimental work is necessary to reconcile the various conflicting statements which have been made on this subject. In conclusion, I desire to express my thanks to Dr. C. E. Moss and to Eev. E. S. Marshall for valuable assistance. NOTES ON LEPIDIUM. By C. E. Salmon, F.L.S. In this Journal for 1910 (pp. 17, 162), mention is made of the variety longistylum of Leindium camjjestre, and some doubt is expressed as to what this really is. By the kindness of Miss M. C. Knowles, of the Dublin Museum, the Director allowed me to see Mr. A. G. More's Irish specimens upon which the variety was based, and I was able to obtain the opinion of Dr. A. Thellung, author of the Monograph of Lepidiwm, published in 1906. The sheet contains two specimens, both with rather young fruit, and both consisting of a single stem branched solely near the summit ; the label reads — "■ Leindium campestre var. longi- styla. Cultivated field, Loughgall, Co. Armagh. Coll. A. G. More. May, 1854." Contrary to his first opinion that Mr. More's plant might come under L. heterophyllum, Dr. Thellung named the examples L. campestre, with the following note: — "forma stylo quam in forma vulgari vix longiore, varietatem non constituens. In L. campestri ' styli pars libera cum maximum ^ mm. longa ' (Thell. Lepid. Mon. 1906, 76), in L. heterophyllo ' plerumque 1 mm. longa.' " I gathered a plant exactly matching Mr. More's speci- men as regards length of style near Godstone Station, Surrey, in 1908, and it is probably not rare. Mr. More admitted years ago tliat he did not consider the " foi'm of any importance as a variety," but it remains in our handbooks and lists to the present day. * Focke's artificial montanum x obscurum had "some good seeds and also barren crumpled seeds in considerable numbers " {loc. cit.). It seems probable that his "good seeds" correspond to the "partly developed" seeds of Hauss- knecht's wild hybrid ; such seeds strongly contrast with the aborted ovules, and only dissection or sowing reveals their sterility. 164 THE JOURNAL OF BOTANY Dr. Thellung (/. c. p. 93) includes under L. campestre proper the var. foliosum Rouy & Fouc. mentioned in Journ. Bot. 1910, p. 17. In looking over other examples of Lepidium, Dr. Thellung noted the following from my herbarium : — L. heterophyllwn (DC.) Benth. var. leiocarpitm Thell. Near Montrose, Forfar, W. Gardi- ner, Aug. 1843. L. densijiorum Schrad. Among garden rubbish, Rugby, June, 1898, S. T. Dunn. Waste ground. West Dock Reservation, Hull, v.-c. 61, Sept. 1902, G. Waterfall. L. neglectum Thell. Gravel pits near Burford Bridge, Surrey, Aug. 1905, G. E. S. L. densiflorum is close to L. virginicum L., as indeed both examples from the above localities had been named by the gather- ers (see Watson Bot. Ex. Club Reports, 1898-99, p. 5, and 1902-3, p. 8). It may be separated from that plant by its compact spike of flowers and smaller seed vessels. L. neglectum Thell. is near L. ruderale (as I had called my Burford Bridge plant), but is easily distinguished by its winged seeds and more orbicular silicle. I have since noted the plant from the following localities: — Waste ground. Putney, Surrey, July, 1881, G. G. Druce d W. W. Neioboidd (Hb. Brit. Mus.). Side of Port Meadow, Oxon., June, 1882, G. G. Druce (Hb. Brit. Mus.). Earlston, Wigton, v.-c. 74, 7, 89, coll. Mc Andrew (Hb. A. Bennett). In every case the plant was named L. ruderale. Both L. densijiorum and L. neglectum are N. American plants, adventive in Europe, and Dr. Thellung wrote on both my speci- mens : — " In Anglia nondum indicatum " ; in his Monograph, however, under L. densijiorum, one finds the record, which I print as given, — " England : East Greenwich, R. J. Mickfud ['?] R. J., 1878, C. W. Congdon— Herb. Reut. & Barb." FLOWERS AND INSECTS. By the Rev. E. Adrian Woodruffe-Peacock, F.L.S. The following notes are the outcome of a correspondence with Mr. Charles Nicholson, of Chingford, who is much interested in the cross-fertilization in the long and short styletl forms of Prim- rose and Polyanthus. Observers usually do not begin to collect the facts they require early enough in the year. The hive-bee (A2)is melUjica) and flies (Diptera) were out this year in January ; beetles soon followed ; Lepidoptera and other orders have not been taken yet (March 18, 1911) as visitors to flowers. As a result of forty years' interest in this subject, I find insects move pollen from fiower to flower under the following circum- stances : in seeking (1) honey, or (2) pollen ; in (3) sunning (or taking daylight rest '?) ; seeking (4) rest at night or during dull or cold days (sleep '?) ; while taking temporary (5) shelter during FLOWERS AND INSECTS 165 passing rain-storms ; seeking (6) victims ; while moving from flower to flower in (7) copulation ; also by (8) violence. In this list, by 3 and 4 I mean something quite difl'erent. Sunning or daylight rest is quite difl'erent from the heaviness of the evening or dark hours of night. I suggest sleep because insects are not on the alert as when sunning or resting in broad daylight ; they are much more easily taken in the glass tubes in which I capture specimens for identification. Butterfly or sweeping nets are much too inclusive for this special work. On 8 a few words must be said. Pollination in many so-called anemophilous species is caused by disturbance from insects flying through grass and other vegetation, and shaking out the pollen by the violence of their movements. They carry, I find, a proportion of the disturbed pollen on their bodies, to be scattered and left when they come in contact with other plants — often, in the case of grasses, of the same species. This form of pollination is quite independent of the help they give to wind drift. The hive-bee has been working in this parish (Cadney-cum- Howsham) part of ten days since January last. Galanthus was the first species I saw visited ; then in order came Eranthis, Crocus, Helleborus fcetidus, Privmla vulgaris, Cheiranthus, Dcvpline mezereum (but not D. Laureola, though it is in full flower), and Viola odorata. In every case the bees were collecting pollen. During four days bees have visited Primroses and Polyanthuses in three gardens here. Our local Polyanthuses are only the caulescent variety of the Primrose, without the least admixture, so far as I can discover. Now, though I am miles from the nearest clump of wild primroses, there have been quantities of garden ones flowering here since early February. During the opening days of March bees were steadily gather- ing pollen on all the flowers during the midday hours. All were visited, but the short-styled were the favourites. Every flower — and there are hundreds— must have been visited in the Manor House garden here. The bees seemed to prefer plants approxi- mating in colour to the wild form, but in the end all were visited. I noticed also the same preference for the yellow-coloured forms in the school-house garden at Howsham, two miles away. All the Primroses left in my garden are oft'-coloured, there are few plants, and, so far as I and my little sons have been able to observe, they have not been visited. I have taken for my type collection of insect visitors specimens of hive-bees with Primrose pollen-masses on their thighs. I find that these masses often fall off their legs after the bees are cyanided, thougli the tubes pre- serve them. Bees carry the pollen of the short-styled form on the hairs of the body, ligula, &c., and leave a certain proportion of it on the long-styled pistil in searching this form. The bee carefully inserts its ligula into the tube of each flower ; it, however, only delays if there is pollen on the lower side of the pistil, corolla-mouth, or on the petal disc. It leaves the pollen of the short-styled form, as a rule, only on the first and second flowers visited. In a few 166 THE JOURNAL OF BOTANY minutes another worker comes along, and leaves short-styled pollen on the long-styled pistils, not hitherto cross-fertilized though perhaps visited, by beginning its work with another flower and visiting them in another order. Until March 11th I could not fully make out how the pollen from long-styled was transferred regularly to short-styled flowers. After watching some hours about midday I discovered it was brought about in the following manner, thanks to the breeze blowing at the time : the lashing of the flowers on their stalks in the spring wind and the whipping caused by the visits of the bees shake the ripe pollen from the stamens on to the under side of the long-styled pistil and to the corolla-mouth. In rarer cases it appears even to reach the petal disc, for I have observed bees collecting it there. Two days later the rough westerly sleet- showers filled the petal discs of the long-styled upright flowers in my garden with water. In almost every case the pollen of the long-styled form was floating on the water. Yet I have every reason to believe that the plants in my garden have remained un- visited. It is quite unusual to find long-styled pollen on the discs of long-styled plants in the Manor House garden when the bees are working. From the shape of the stigma and its place in regard to the opening of the corolla in the long-styled form, and the closely felted hairs of these parts, only in rarely exceptional cases have a few globules of water entered into the flowers, but the pollen escapes all the same. When a bee's hgula is inserted into a long-styled flower a certain proportion of pollen is withdrawn. It may come from the lower side of the stigma or the corolla-mouth side, or it may be gathered from the petal disc, if it has been thrown out by the whipping of the flowers. The bee's next visit may be to a short- styled form. Its first action is to force its ligula amongst the stamens, and by doing this a certain proportion of the long-styled pollen reaches the sliort-styled pistil buried in the corolla. The bee then withdraws its ligula, and begins to collect the pollen on its legs. No doubt the thrusting in of the ligula when it first arrives greatly aids in the dispersal of the pollen in the short- styled form. A certain proportion of the long-styled pollen even reaches the short-styled stigma, for so far as I can learn, unless bees are very plentiful, the crossing from the short to the long is much more regular and bountiful than from the long to the short-styled form. As each flower in the. early days of the season is carefully and systematically worked at in most places by hive-bees, I have no doubt both forms are fully cross-fertilized. Later on, when hive-bees have a wider sweep of floral species, the Primrose is more or less neglected by them. This explains, no doubt, an observation I made as far back as 1877, that the seed from the early wild Primroses is more fertile than from the later flowers. While the bees were working the Primroses this season, other hive-bees were visiting the other plants named above, with the exception of Galanthus and Eranthis, which were past flower. In WILLIAM AMBEOSE CLAEKE 167 no case did I see a bee go from one species to another. In summer wild bumble-bees do cross from species to species. Here is the first record I can put my hand on : " Bombus lucorum, 23. 8. 1910, crossed from Centaurea nigra to Leontodon autmnnalis. It was seeking honey in both cases." Was it feeding itself rather than gathering food for others ? In my experience hive-bees do not do so ; neither early nor late in the season have I observed it in forty years. An old bee-keeper once told me that neither honey nor bee-bread would keep if they were to do so. Taking the flowering season throughout, I estimate that twenty- four flies (Diptera) visit the flowers of our whole flora for each insect of all the other orders put together, the Thysanoptera excepted. When it is on the wing in the summer heat Thrips is on every flower, even those of the simplest grasses, in countless numbers. The number of flies that visited flowers was most astonishing to me when I first discovered the fact. In mid- summer, when bees and beetles are most active and common, the proportion, though high, is nothing like one to twenty-four, but later in the season the notes soon run the other way. By mid- October the bees have practically ceased to work, but the flies continue as long as the flowers last, nearly up to Christmas in open sunny seasons. For example, on the 16th of last October I noted in a two-mile walk, on two soils, thirty-seven species in flower in this parish. A bee was on one species, a fly or flies on twenty-six, and there were as many as three species of flies on one flower, on which there were five altogether. Looking at the question of insect visitors in the light of a life-interest in the subject, I am inclined to agree with Andrew Knight (1799), that " in no plant does self-fertilization occur for an unlimited number of years"; not even in cleistogamous species as we call them. WILLIAM AMBEOSE CLAEKE. (1841-1911.) Another of the supporters of this Journal and of the friends of its Editor has been removed from among us in the person of William Ambrose Clarke, who died at Oxford on the 23rd of February. He was born at Hinckley, Leicestershire, on February 6th, 1841, his father, the Eev. T. A. Clarke, being then curate at Stapleton. He was articled to Mr. Peter Awdry at Chippenham and practised for about twenty years as a solicitor, chiefly at Chippenham, of which town he was Mayor in 1879. Here he became interested in botany, and was associated with the Eev. T. A. Preston, then at Marlborough, in investigating the county flora, his help l)eing duly acknowledged in the Flora of Wilis, published by Preston in 1888. His first communication to these pages was in 1887 — a brief note recox'ding certain additions to the county flora as then known. In January 1892 Clarke married Miss Emily S. Ward, daughter of the Vicar of Great Bedwyn, and in October of the same year 168 THE JOURNAL OF BOTANY he came to Oxfoi'cl, where he remained until his death. A devout and convinced Anglican, he was for fifteen years sidesman at the church of SS. Philip and James, of which he was also for a short time churchwarden. We had for some time hecome somewhat intimate as correspondents, and on his occasional visits to town he used to call upon me at the Natural History Museum ; it was on one of these visits that I suggested to him the compilation of a list of the records of the occurrence of the native plants of this country, which I had long thought would be of interest, and which his knowledge of botanical literature, tlien already considerable, rendered him qualified to undertake. This list was begun in the Journal for January 1892 and continued at somewhat irregular intervals until the end of 1906, when it was reissued as a volume entitled First Becords of British Flowering Plants-. Clarke was, however, never quite satisfied with the form in which the list appeared in the Journal, where exigencies of space demanded more strict limitations than he thought desirable ; and in 1900 he published a second edition, "revised aiid corrected," which gave him more satisfaction and is indeed an important contribu- tion to the history of British botany. In his introduction Clarke sets forth his views on nomenclature, which were sentimental rather than logical ; he had strong objections to certain conse- quences of the adoption of the Vienna Code, and these were often the subject of amusing discussion between us. Although possessed of a good knowledge of British plants, it was the bibliographical side of botany in which Clarke took most interest, and his proximity to the Bodleian and the Botanic Garden gave him every opportunity for following out his bent. WILLIAM AMBROSE CLARKE 169 His bibliographical note on the dates of Curtis's Flora Londinensis (Journ. Bot. 1895, 112) is a good example of his careful work, and sums up all that is known on the subject. In 1901 he pubhshed in the Journal (pp. 128-140) an interesting account of the pro- gress of British botany in the nineteenth century, which showed his familiarity with the literature of the period. One of his interesting discoveries was the account of Eriocaulon septangulare in an appendix to vol. ii. (p. 784 bis) of Withering's Arrangement (1776) ; this had been overlooked and the name cited in Index Kewensis (1796) and elsewhere as from the 3rd edition. The plant had been named Nasmythia articulata by Hudson in 1778, and Hudson's trivial, apparently the earliest, had been taken up by recent writers, who called the plant E. articulatum. The restora- tion of the name generally accepted gave Clarke much satisfaction. Clarke largely collaborated with the Eev. E. S. Marshall in the compilation of the tenth edition of the London Catalogue, and he transcribed for publication the reprint of Linneeus's Flora Anglica issued as a supplement to this Journal for 1909. During the later years of his life the state of his health com- pelled frequent visits to health resorts, where the local botany always interested him. His herbarium, though not large, con- tained specimens of most of our British plants, the majority of which he had seen in situ. An addition to these was Hypericum duhium, which, although far from uncommon, Clarke had not collected ; Mr. Druce took us to see it one Sunday afternoon last summer, during the little jaunt referred to by him in this Journal (p. 233). It was only during the last few years that I became personally intimate with Clarke. " Week-ending " with friends at Iffley, I had to go into Oxford to church, and afterwards not infrequently, though not as often as I think both of us would have liked, w^ent to him for luncheon. He was happy in the possession of a wife who took an inteUigent interest in his pursuits ; both had literary tastes (Clarke was a fine classical scholar), and an after- noon spent in their company was a pleasant experience to look back upon. Clarke was indeed one of the kindest of men, with a keen sense of justice ; this latter prompted his last note in this Journal (1909, 447) and he was not a little distressed at the storm in a tea-cup which was raised by his well-intentioned intervention. Clarke became a Fellow of the Linnean Society in 1890, but subsequently withdrew ; he rejoined the Society in 1909. A good and interesting correspondent, he will be missed by many botanical friends as well as by the writer of this notice. James Britten. SHORT NOTES. A Boy Botanist. — Under the heading of "Botany," the Seventy- seventh Annual Eeport of Bootham School (York) Natural His- tory, Literary, and Polytcclinic Society, says: — "A. \V. Graveson is again the botanist of the year, and his collection has now Journal of Botany. — Vol. 49. [May, 1911. j n 170 THE JOURNAL OP BOTANY reachocl the mai'vellous total of 1012 species [of British Flowering Plants presumably] as against 800 last year. This collection must surely be a record for any schoolboy, and we must again congratulate Graveson on his energetic and enthusiastic work." Thus Bootham School, whose Natural History Society is the oldest of any such school society in England, creates another record, and at the same time it is pleasant to realize that mere collecting is by no means the only thing encouraged at that school, which has produced, to mention a few of its botanists only, J. G. Baker, E. G. Baker, F. W. Oliver, A. H. Burtt, Miller Christy, H. Corder, H. Tuke Mennell, and the late James Back- house and Silvanus Thompson. Some otlier branches of Natural Science and Literature are equally well represented by distin- guished names among the alumni of this well-known Friends' School. — H. S. Thompson. Pyrus cordata Desv. in Monmouthshire (v.-c. 35). — Just within the county, close to the Wye, on a high rough bank, in the parish of Dixton, I found two plants of this species in 1910. It grows within a few miles in Gloucestershire, at Symonds Yat, and in Herefordshire, on the Doward. When I pointed out the trees to Mr. Ley, he agreed to the identification. — H. J. Eiddelsdell. REVIEWS. A Monograph of the British Lichens, being a Descriptive Catalogue of the Species in the Department of Botany, British Museum. Part II. By Annie Lorrain Smith, F.L.S. 8vo, cL, pp. 409 ; 59 plates. Price £1 Is. Many years have elapsed since the publication (in 1894) of the first volume of a Descriptive Catalogue of the British Lichens in the National Herbarium, compiled by the late Eev. J. M. Crombie. At the death of the author 66 genera of British Lichens out of a total of 93, according to the arrangement given in the Conspectus Generum of vol. i., had thus been published, including descrip- tions of 580 species, with numerous subspecies,' varieties, and forms. Lichenologists have been expecting eagerly, and now welcome joyfully, the appearance of vol. ii. of the Catalogue, in which Miss Lorrain Smith carries to completion the work pro- jected by Crombie. It would seem by reference to the Conspectus in vol. i. that the greater part of the work had already been accomplished in that volume, which leaves only 26 genera to be dealt with, accord- ing to the original arrangement. It happens, however, that these remaining genera include the largest, and some of the most per- plexing, groups of lichens, and therefore it is perhaps fair to say that the bulk of the work has fallen to Miss Smith. The 26 genei^a of Crombie have increased by additions and subdivisions to 55, including the large number of 858 species, as against 580 of the previous volume. The additions are due partly to the in- A MONOGRAPH OF THE BRITISH LICHENS 171 elusion among lichens of plants formerly regarded as algae {Conno- gonium, Bacodium), partly to the discovery in recent times of genera new to Britain (e. g. Gongylia). The subdivisions bring the book into line with Continental text-hooks, and cannot fail to be acceptable to students who have had to grapple hitherto with the unwieldy mass of incongruous species "lumped" together under such genera, for instance, as Lecidea and Verrucaria in the older manuals. It is, of course, quite possible to go to the other extreme by the process known as " splitting," but while this has been avoided, confusion is obviated by careful and systematic reference to modern synonyms. As stated in the introduction to this volume, " more importance is assigned in these days to the microscopic character of the apothecia than was allowed by Nylander and Crombie in their scheme of classification." Eelying on such natural distinctions, Miss Smith adopts 11 of the genera into which Lecidea has been divided, with three subgenera of the restricted genus— Psora, Biatora, Mycoblastus. Of the old genus Graphis, as described in Leighton's Lichen-Flora of Great Britain and L-eland, only four species are retained under that generic name, the rest being assigned to Phceographis and Graphina. We venture to think that the author might have gone a step further, and adopted Aulacographa for G. elegans and the allied species with a furrowed proper margin to the fruit. Out of Verrucaria of the old lichenologists, 13 genera are taken, which yet leaves 48 species in the type-genus. A systematic and modern revision of natural affinities is in- dicated by the removal in the scheme of this volume of the orders Gladoniacece and Gyrophoracece to the subtribe Lecideei, and of Dirinacea and Boccellacece to the Graphidei. Strigula, Endo- coccus, and Myriangium, regarded by Crombie and other lichen- ologists of his time as belonging to lichens (see vol. i.), are here relegated to fungi pure and simple. This fact is mentioned by Miss Smith (p. 263), though the names are not to be found in the index, whilst Agyrium and Odontotrema, presumably also fungi (vol. i. p. 15), are left without reference. Since in these and other details the systematic arrangement preferred by Miss Smith differs rather widely from that jpro- pounded by Crombie in the introduction to vol. i., we are inclined to think that a new Conspectus Generum, following the lines adopted in this volume, might have been desirable. The meagre glossary of technical terms prefixed to vol. i. has expanded in this to one that is copious as well as clear, while the admirable index to the entire work will be especially hailed as a boon by those students whose good resolutions of indexing the previous volume for themselves have never been carried out. After the small figures of microscopic details in the first part of the work, the fifty-nine full-page illustrations of the second part come as a pleasing surprise. These plates, ranging from a view of the plant as it appears in situ to higlily magnified points of structure, are attractive and, so far as we have had an opportunity of testing them, accurate. 172 THE JOURNAL OP BOTANY In looking through this manual we are reminded of the general acceptance at the present time- of the once much controverted doctrine of symbiosis in lichens. Ignored by Crombie in vol. i., and indignantly repudiated by Nylander and Leighton, it is assumed as an undoubted fact by the author of the present volame. Indeed, not the least interesting and valuable feature of the book is the mention under each genus of the particular algal form that constitutes the " symbiont," or messmate, in the com- position of the lichen. It is hardly necessary to say that this manual is indispensable to the student of British Lichens. While no futile attempts at " popularizing " them are made by the sacrifice of technical ter- minology, the somewhat pedantic and deterrent style of preceding Lichen-Floras has been happily avoided, and thus it may well be hoped that recruits will be enlisted for the practical study of these interesting plants. That increased activity of this kind has manifested itself of late is proved by the success of the recently established Exchange Club of British Lichens, which already can boast of members from all parts of the United King- dom, and has resulted in the discovery of a considerable number of species new to the British Flora, or even to science, all of which are described in the addenda to this volume. The few misprints observed are mainly in local names, and are no doubt due to defective handwriting on labels ; one contributor, at any rate, feels he must cry med culpa in this respect ! It need only be added that no more effective proof could be given of the rich- ness of the National Herbarium than the fact that a catalogue of species therein contained, described by an expert, becomes an admirable and exhaustive Manual of British Lichens. H. P. Keader. Flora of Jamaica : containing Descriptions of tlie Floivering Plants knoivn from the Island. By William Fawcett, B.Sc, F.L.S., and Alfred Barton Eendle, D.Sc, F.E.S. Vol. i. Orchidacese. Pp. xviii. 150, 32 plates. Price 10s. 6d. Dulau, Soho Square, W. This w^ell-printed and well-illustrated volume is the first instal- ment of an important undertaking for which the Trustees of the British Museum are responsible. That such a work was in pre- paration has been known to the readers of this Journal, in whicli the descriptions of the new species contained therein have from time to time been printed, but these in their necessary isolation from the text could give no notion of the extent of the complete enumeration. The combination of authors is a happy one, for Mr. Fawcett left the Museum to become Director of Public Gar- dens and Plantations in Jamaica and there had the opportunity for twenty-one years of studying its flora in situ, and was able to obtain the loan of all the material in the Jamaican Herbarium and of a fine series of drawings made at the Gardens, under his supervision, by Miss Helen A. Wood ; while Dr. Eendle's long connection with the Department of Botany has familiarized him MINNESOTA ALG^ 173 with the contents of the Herbarium and with the Hterature bear- ing on the subject. The Flora is intended for use in the field as well as in the herbarium, and care has been taken to make it of service to the botanist or collector who may wish to employ it in the island. Thus the descriptive portion of the book before us is prefaced by a general account of the Orcliidacea, followed by a very full key to the genera. The descriptions of the genera and species are also full and have evidently been drawn up with much care ; there is an excellent bibliography and synonymy for each species, and the distribution is fully detailed, the number of localities and collec- tors cited — among the latter Mr. Wilham Harris, Superintendent of the Gardens, is conspicuous, showing that the investigation of the island has been very thoroughly undertaken. The where- abouts of the type specimens is frequently indicated ; in the case of Swartz's specimens in the National Herbarium these should sometimes perhaps be considered as authentic examples, named as they are by Swartz himself, than actual types : the authors, by the way, have been fortunate in obtaining the loan, from Berlin, of a valuable collection of Swartz's unpublished drawings. The thirty-two excellent plates, largely taken from the drawings by Miss Wood already referred to, contain figures or details of a hundred and thirty species, and add greatly to the value of the work. Minnesota AlgcB. Vol. i. Josephine Tilden. Report of the Survey. Botanical Series, viii. Minneapolis, Minnesota. April 1, 1910. The title of Miss Tilden's book is somewhat misleading. One wonders why a book should be labelled " Minnesota Algae," when very few of the included species are found within hundreds of miles of Minnesota. There appears to be a newly-developed, but at the same time a widespread, ambition on the part of modern authors of works on the algae of small areas to include those of the whole world. This was first conspicuously manifested in Chodat's Algues Vertes de la Suisse, which was chiefly remarkal^le for the description (often very poor and accompanied by atrociously bad references) of those algae which do not occur in Switzerland, nor are ever likely to occur there. The first completed volume of Lemmer- mann's Algenflora von Brandenburg is of a similar nature, only, if we may so express it, much more so ! Lemmermann, by a wild flight of imagination, would have one believe that these plants are so cosmopolitan that all the algae of the world must sooner or later be found in Brandenburg ! Works of this kind are decidedly misleading, and give a false impression concerning the distribution of algae. Moreover, they scarcely serve a useful purpose when the geographical distribution is omitted. Miss Tilden's work, as stated on the title-page (possibly as an ex- planation of what is meant by "Minnesota Algae " !),is a synopsis of " the MyxophycecB of North America and adjacent regions, including Central America, Greenland, Bermuda, the West Indies, 174 THE JOURNAL OF BOTANY and Hawaii." As a compilation of previous records for the areas in question the work is good, a'tid will be decidedly useful. One would like, however, to see such compilations treated from a more critical standpoint, although such treatment obviously necessitates great personal experience and knowledge of the plants dealt with. The study of the subject would in this way be materially advanced and the work would be of immeasurably greater value. There appears to be an entire absence of critical observation from the whole of the volume under review, and this seriously detracts from its importance as an up-to-date record of the systematics of the Myxophycea. To give one instance : both Phormidium unci- natum and P. autumnale are fully described, with accompanying references and distribution, although Jobs. Schmidt, and others, have clearly shown that it is not possible to discriminate between them. The illustrations will no doubt be a useful feature of the book, but they are mostly copies of well-known figures by other authors. One would have imagined that a work of this nature would have given scope for many original drawings. The use of the term "plant" instead of "cell" for the single cells of the Goccogonece, is not a very happy expression. The author has also repeatedly used the name " De Toni " as a specific authority when it should have been " Forti." The compilation has been well done, and there are singularly few errors, but one's first impression on opening the book is distinctly not a good one, as it suffers like so many American publications from a lack of discrimination in the type. Specific names, authors' names, and geographical areas are all in the same type, and it is positively difficult to discover where one species ends and another begins. This entails a great waste of time in making use of the book for specific determinations. G. S. West. BOOK-NOTES, NEWS, dc. At the meeting of the Linnean Society on March 16th, Mrs. D, H. Scott gave a lantern exhibition of new species of the fossil genus Traquairia. She also exhibited the original diagram made by Dr. W. Carruthers, who first described the genus at a meeting of the British Association in 1872, in a paper entitled " Traquairia, a Eadiolarian rhizopod from the Coal-Measures." Count Solms- Laubach, Professors Schenk, Strassburger, and Zeiller considered it comparable to the massulge or sporocarps of Azolla. Professor Williamson (Phil. Trans. 1880) thought it the spore of a crypto- gam. He found a group of three Tmquairice in a sporangium of Lepidostrobus, and thought them three megaspores of a tetrad. The true megaspores are, however, now well known. Mrs. Scott defined Traquairia thus: — "Traquairia is a spherical organism, consisting of two parts each surrounded by a sharply defined membrane : an inner capsule, often containing spores, and an outer part, which is surrounded by a thick gelatinous envelope. BOOK-NOTES, NEWS, ETC. 175 In this are embedded numerous hollow spines. The apparent bases of these spines are produced into hollow anastomosing tubes, which spread over the surface of the sphere, forming a com- plicated network. The spines are hollow, and are perforated in every direction by projecting tubular pores. Emanating from these pores are delicate threads which appear to lose themselves in the gelatinous envelope. Sometimes the threads form a regular network in it. The inner capsule, a definite brown membrane, can only be observed in the more perfectly preserved specimens. Spores are generally present, which appear to produce small spores. The Traquairice. occur in groups in the decayed wood of Lepidodendron and other plants." She then exhibited T. Car- ruthersii, T. Spenceri, sp. nov., T. burntislandica, sp. nov., and T. stellata, sp. nov., and a species of an allied genus, S'porocariMn elegans. The most characteristic feature in the organisms de- scribed is the very complicated structure of the outer envelope with its elaborate system of anastomosing tubes connected with prominent spines, which are themselves very complex organs. Nothing parallel to this is known in the vegetable world. The presence of an "inner capsule" containing spores, in the interior of which small spores are produced, reminds one of Eadiolarians. These features are also common to Sporocarpon elegans, which with its long spines is very much like a Eadiolarian. At the same meeting, Mr. E. S. Adamson gave a lantern demonstration of his communication entitled " An Ecological Study of a Cambridgeshire Woodland," of which the following is an abstract : — The woodland investigated (Gamlingay Wood) is in the extreme West of Cambridgeshire, situated on Boulder Clay. This Boulder Clay gives rise to two soils — one a heavy fine-grained calcareous clay, and the other a non-calcareous loam. These soils are markedly different, especially in water-contents ; the calcareous clay has a large water-content all through the summer, while the loam never gets so wet in winter, and dries up very much in summer. Correlated with the differences in the soils, the two types of vegetation are quite different. Oak [QLtercm piedimciblata) is dominant all over. On the clay beneath the standard trees is a dense shrub-layer consisting mainly of Hazel and Ash, with nume- rous other species in less quantity. This layer is periodically coppiced and allowed to regenerate naturally ; on the loam, shrubs, as a distinct layer, are absent. The ground floras of the two soils are also quite different. Each can be divided into several societies depending on the interaction of several factors, of which soil- moisture and light-intensity are the most prominent. The chief plants of the clay flora are Spiraa Ulmaria, Deschampsia caspi- tosa, Mercuricdis perennis, and Primula elatior ; and of the loam, Pteris aquilina and Holcus mollis. That part of the wood on the clay is an (Ash)-Oak-Hazel Wood, and that on the loam a dry Oak Wood. In his introductory study of The Livcncoris, British and Foreign (Witherby & Co., 1911, pp. viii, 74, 49 figs., price 2s. 6^. net). Sir Edward Fry has produced a companion volume to his popular 176 THE JOURNAL OP BOTANY essay on the British Mosses. The plan of the book is briefly as follows. Two typical forms, one thallose {Pellia epiphylla), the other foliose {Diplophyllum albicans), are described clearly and in simple language. The principal groups of Hepaticae are then treated in turn — Ricciece, Monoclece, Antliocerea, Marchantiece, JungermanniecB. And finally such general matters as leaves, gemmas, modes of reproduction, alternation of generations, are discussed. The tables of unicellular and multicellular gemmae, their various forms and their place of reproduction in various genera of Jwigermannia, are helpful and instructive. Exception may be taken by systematists to the position given to Monoclecs, and especially to the inclusion of Galohryum in the group. There is a tendency nowadays to range Monoclea near to Targionia among the lower Marchantiece. ; and Galohryum is akin to Fossom- bronia. Again, the Anthocerece should hold a detached position right outside the Hepaticse. However, these questions of classifi- cation do not affect the real value of the book for the elementary student, for whom the author has provided an attractively written essay, containing a well-digested and simplified account of the morphology, anatomy and life-history of a difficult but interesting group of lowly plants. — A. G. We are somewhat surprised that Mr. Praeger's Open-Air Studies in Botany (Griffin), which we noticed at some length on its first appearance (Journ. Bot. 1897, 453), has only now reached a second edition. It is one of the best books introductory to a knowledge of wild flowers in their homes — indeed we might say the best, for we know of no other volume which tells so much about plants as they actually live, or which is so charmingly illus- trated by pictures from " photographs from nature." We cannot do better than repeat what we said in the notice referred to : "In a word, Open-Air Studies is the very thing to put into the hands of those who want in small compass an intelligible and accurate introduction to field botany." This second edition, which is in- scribed to the memory of the late S. A. Stewart, "a true field naturalist," has been revised throughout, though this has involved little alteration ; the nomenclature has been brought into line with the British Museum List of Seed-Plants. The publishers still disfigure the title-page of review copies with an ugly violet stamp, to which is added an indication in ink that the book costs 6s. net. John H. Hart, who was born in Suffolk in 1847 and died at Trinidad, Port of Spain, on the 20th of February, had been Super- intendent of the Botanic Gardens there since 1887 ; he had pre- viously been Director of the Jamaica Botanical Department. In 1908 he issued a list of the plants contained in the herbarumi of the Botanical Department at Trinidad, which includes the collec- tions by Lockhart, Purdie, Finlay and Prestoe, as well as more recent collections made by himself and others. An enumeration of the Ferns and Fern Allies of the British West Indies and Guiana, by G. S. Jenman, was edited by Hart and published in fragments with the Bulletin of the Botanical Department issued under his direction, and was subsequently issued as a volume. Journ. Bot. Tab. 514. E. W. Hunnybun del. West, Newman imp. SALICORNIA DISARTICULATA 2Ioss. 177 SOME SPECIES OF SALICORNIA. By C. E. Moss, D.Sc. (Plate 514.) The following is a preliminary account of an investigation, conducted by Dr. Ethel de Praine, Mr. E. J. Salisbury, and my- self, on some species of Salicornia which we have collected during the last three years. The investigation was commenced at Pro- fessor F. W. Oliver's ecological station on a sandy salt-marsh — the Bouche d'Erquy — on the north coast of Brittany. Tlie dried collections at the Linnean Society's rooms, at the British Museum (Natural History), at Kew, and at Cambridge have been examined, and also dried specimens kindly lent by Mr. C. Bailey, Mr. G. C. Druce, Professor C. Flahault, of Montpellier, Kev. E. F. Linton, Kev. E. S. Marshall, Mr. C. E. Salmon, Kev. C. H. Waddell, and Dr. E. de Wildeman, of Brussels. The full account of the investi- gation, under joint authorship, will be published elsewhere. SALICOENIA [Tournef.] Linn. Sp. PL 3 (1753), et Gen. PL ed. 5, 4 (1754) pro parte. Salicornia et Arthrocnemum Moq. Chen. Mon. Enum. (1840) et in DC. Prodr. xiii. 2 (1849). Salicornia Duval-Jouve in Bull. Soc. bot. France, 170 (1868). Bentham (Flo. Austral, v. 201, 1870) reduces tlie whole tribe Salicornieae to the genus Salicornia ; but this plan appears to be too drastic. Subgenus Anthrocnemum [Moq. oj). cit. pro gen.] . Salicornia sect. 1^, Duval-J. loc. cit. 1. Salicoenia glauca. S. virginica Forsk. Fl. M^.-Axdh. 2 (1775), non Linn. ("Vir- ginia") Sp. PL 4 (1753), nee Murray ("virginica"), Syst. Veg. 51 (1774). S. glauca Del. Des. de I'Egypte— Flo. ^gyptiacse— 49 (1813) ! S. mucronata Lag. Mem. PL Barill. 58 (1817), ex Ind. Kew. S. macrostachya Moric. Fl. Yen. i. 2 (1820). Arthrocnemtim fruticosum var. macrostachya Moq. op. cit. A. fruticosum var. glaucum Moq. op. cit. A. macrostachyum Mor. et Delponte in Ind. Sem. Hort. Bot. Taur. 35 (1854). S. macrostachya Duv.-J. oj). cit. 171 ! A. qlauciim Ung.-Sternb. in Atti Congr. Bot. Firenze, 283 (1874 ; 1876), ex Ind. Kew. S. fruticosa auct. mult., pro parte. The plant occurs in the salt-marslies, both inland and maritime, of the Mediterranean region from Portugal, Spain, and Morocco to Greece, Palestine, and Egypt. Records from India and Angola are based on misapprehensions. It is a small shrub, growing eventually to a height of about a metro. Plants I saw in Algeria in the spring of last year were Journal op Botany. — Vol. 49. [June, 1911.] o 178 THE JOURNAL OF BOTANY not quite half a metre high, but new growth had scarcely com- menced. It is branched and erect. It flowers in July and early August, and ripens its seeds in September. The terminal flowering- spikes-'- are stout, cylindrical, blunt, and yellowish, and may grow to be 10 or even 17 cm. long. The three flowers of the cymes are subequal in size ; and the central one reaches less than two-thirds of the way up the segment. The seed is black and tuberculate. Subgenus Eu-Sahcornia. Salicornia sect. 2'», Duv.-J. ojj. cit. 170. Salicornia auct. mult., sensu stricto. Section Perennes Duv.-J. loc. cit. "Flores laterales flore, medio disjuncti." 2. Salicoenia fruticosa. S. europcBCL v&r. fruticosa Linn. Sp. PI. 3 (1753). S. fruticosa Linn. Sp. PI. ed. 2, 5 (1762). Arthrocnemum fruticosmn Moq. op. cit., excl. vars. S. fruticosa Duv.-J. op. cit. 172 ! S. fruticosa auct. mult, (sed non auct. angl.) pro parte. In the Index Kewensis, S. arabica Linn. Sp. PI. 3 (1753) and S. anceps Lag. op. cit., are also cited as synonyms of this species. However, S. arabica Linn. herb. (!) is identical with S. arabica Pall. herb. (!), and is Kalidium arabicum ; and S. glauca Stokes (Bot. Mat. Med. i., 8, 1812) would appear to be referable to the same species. A specimen named S. anceps Lag. from Carthagena ("leg. Cadorini ") in herb. (!) C. Bailey would appear to belong to the section AnnucB, since the lateral flowers meet below the central one. S. fruticosa, sensu stricto, occurs on salt-marshes through- out the whole of the Mediterranean region, and is often a social plant. Duval-Jouve {op. cit., p. 173) says that specimens were sent to him from Brittany by Mons. J. Lloyd, and from Normandy by Mons. Lebel. However, in J. Lloyd's Flo. de rOuest Fr. (ed. 3, 262, 1876), the northern limit of the species is given as the mouth of the Loire; and I have seen no specimens from any station north of this limit. The floras add South Africa, Polynesia, and America ; but I am not certain that these localities yield the same plant. Plants collected by Major Wolley-Dod in South Africa (in herb. Brit. Mus !) are given varietal names. The plant is a tufted dwarf shrub, growing separately in rounded tufts or intermatted as a social plant. " It may attain a height of nearly a metre. The plant is erect, suberect, or ascend- ing. It flowers in September, considerably later therefore than S. glauca ; and perhaps therefore the two species do not hybridize, although it would appear that they frequently grow together. Its seeds ripen in November ; but ripe seeds appear to be very rare in herbarium specimens. The vegetative segments are in general longer than those of S. glauca, and the flowering-spikes shorter. In the lower part of the flowering-spikes, the central * It is important to note that, throughout this account, the flowering- spikes and not the fruiting-spikes are described. SOME SPECIES OF SALICORNIA 179 flower only reaches about one-third of the way up the segment, whilst in the upper half of the spike it reaches about half-way up. The seed is greyish, not black nor strictly tuberculate, but covered with short stout hairs. Eouy (op. cit. p. 60) gives a variety defiexa ; but perhaps this is not distinct or more than a mere state. 3. Salicornia pbeennis. S. ijerennis Mill. Gard. Diet. ed. 8, no. 2 (1768) ! S. radicans Sm. Eng. Bot. t. 1691 (1807) ! Arthrocnemum fniticosum var. radicans Moq. loc. cit. S. sarmentosa Duval-J. in Bull. Soc. bot. Fr. 174 (1868) ! S.fruticosa auct. mult., pro parte. S. fnUiculosa Tineo, Cat. Plant. Hort. Reg. Panorm. 280 (1827), might, judging simply by the description, be referred to this species ; but herbarium specimens seem to be distinct both from >S'. perennis and from S. fruticosa. S. perennis is with certainty only known to occur in England, France (including south France), Spain, and Algeria, though it may be ultimately proved to have a wider range than this. It is often a social plant, growing in matted clumps. Isolated tufts are sometimes a metre or more in diameter. At maturity, it is a tufted dwarf shrub, spreading centrifugally by means of freely rooting branches. The central flower of the cymes is larger than the lateral ones, which only reach about half-way up the segment. The seeds are nearly globular, and covered with slightly curved hairs, which are rather longer than those of the following species. Eouy (op), cit. p. 60) divides the species into three varieties, which are perhaps mere soil forms. 4. Salicornia lignosa. S. lignosa Woods, Bot. Gaz. 31 (1851) ! Giirke {Plantce EuropcBCB, II. i. 158, 1897) doubtfully refers this plant to the preceding species ; but the two, would appear to be quite distinct. If the two plants are to be referred to the same species, Woods's plant should be given a varietal name — S. perennis var. lignosa. Up to a year ago, this species was only known to occur in the south of England. In the spring of 1910, however, I found the plant on the shores of the small Sebka, near Oran, in north-west Algeria. Specimens from this locality are now in the herbarium at Cambridge ; and the plant was also collected in the same locality by Professor C. Schroter and Dr. E. Riibel, of Zurich. I submitted a specimen to Mr. A. Bennett, who wrote : — " Yes ! It seems to me to be exactly like Woods's plant." Doubtless it will prove to be spread over a wider area when the characters which distinguish it from S. fntticosa and S. perennis become more generally known. At maturity, it is a prostrate dwarf shrul), leaving the ground by a single stem, which grows mainly in a unilateral direction. The branches, although they lie flat on the ground, like those of 0 2 180 THE JOUBNAL OF BOTANY S. perennis, have no adventitious .roots. The three flowers of the cyme are roughly equal in size ; and the central one reaches two- thirds of the way up the segment. The seed is nearly like that of S. iJerennis, but its hairs are shorter. Section Annua Duval-J. op. cit. pp. 170 and 175. "Flores laterales flore medio contigui." Subsection Erecta. Plants erect ; stamens 2 or 1, if two, open- ing in succession. 5. Salicornia eueopjea. S. europcea Linn. Sp. PL 3 (1753), excl. var. ft. S. enrojJCBa var. herhacea L. Flo. Suec. ed. 2, 1 (1755). S. herbacea Linn. Sp. PI. ed. 2, 5 (1762). S. fruticosa Mill. Gard. Diet. ed. 8, no. 1 (1768), non Linn. S. annua Sm. Eng. Bot. t. 415 (1797) ! S. 2^rociLmhens Sm. Eng. Bot. t. 2475 (1813) ! S. acetaria Pall. 111. Plant. 7, t. 1, figs. 1 and 2, t. 2, fig. 1 (1803) ! S. radicans Mert. & Koch in Eoehl. Deutsclil. Flo. ed. 3, i. 292 (1823), ex Ind. Kew., non Sm. S. stricta Dum. in Bull. Soc. bot. Belg. 333 (1868) ! S. patula Duv.-J. op. cit. 175 (1868) ! S. Emerici Duv.-J. op. cit. 176 (1868) ! S. Jwrhacea Syme, E. B. 3rd ed. viii. 6 (1868), incl. var. S. herbacea Woods, loc. cit. ! S. herbacea auct mult., pro parte. In the salt-marshes (both maritime and inland) of Europe and the Mediterranean region this species is very abundant, and often social. Two forms, connected by intermediates (not hybrids), are distinguishable : — S. EUEOPiEA forma stricta, form. nov. {S. acetaria Pall., S. stricta Dum., S. Emerici Duv.-J.). Eouy (Fl. de Fr. xii. p. 59, 1910) uses the name "S. biennis Afz." for this form, which he designates a " race " ; but a reference to Smith's Flo. Brit. i. 2 (180()), cited by Rouy, shows it to be a mere MS. name, and one founded on a serious misconception. This form is usually a much branched and vigorous plant, with the branches fastigiate, and with the terminal spikes usually very long. It is the south European form of the species, and spreads northwards at least to Brittany, Belgium, and Essex. S. EUEOP^A forma patula, form. nov. {S.'europcea Linn, et S. herbacea Linn, in sensu stricto, S. anuria Sm., S. ptyocumbens Sm., S.p)atula Duv.-J. pro parte, S. herbaceava,r. jyatula Buchenau, Fl. Nordwestd. 192, 1894). This form has the branches spreading, that is, forming wide angles with the stem, and is usually a less branched plant than the preceding form. As a rule, its flowering- spikes are shorter. Being less vigorous than the preceding form, it frequently gets toppled over by the tides, and is then frequently misnamed S. procumbens. It is pre-eminently the north European form of the species. By Rouy {op. cit.) it is regarded as a "race"; but (according to his wont) he gives it a binomial — S. atimia — as if it were a species. It is the prevailing form in Scotland. The SOME SPECIES OF SALICORNIA 181 two forms, as will be seen from the above citations, have received several distinctive binomials ; but it is difficult to see how they can be separated specifically. Duval-Jouve {o}). cit., p. 176), how- ever, states that his S. Emerici has straight hairs, and he so figures it; but no such annual Salicornia has come under my notice as yet. S. europcsa is a comprehensive and very variable species ; but it must be emphasized that these variations remain perfectly dis- tinct from the other annual species. The latter are not comparable with " species " of Bitbits or Hieracium, and are so distinct (although local in their distribution), that to include them within the limits of S. europaa {" S. herhacea") seems most unreasonable. S. eiiropcea, usually bright green in colour, is easily distin- guished from the other species of the subsection by its long, often very long (up to 5 cm. or more) flowering-spikes, which are tapering and rather obtuse. Each spike has about 8-16 flowering- segments, as a rule. The flowers are nearly equal in size, and the central one reaches about two-thirds of the way up the segment. Usually it has only one stamen, but occasionally a second (which may or may not be rudimentary) is present. In addition to the European and Mediterranean localities, the floras give South x\frica, Siberia, Central Asia, East Indies, and America ; but some at least of these localities are quite untrust- worthy. It is the only species of the genus which can at present be recorded from Scotland and Ireland. In the British Isles, it occurs as far north as the Orkneys and the Shetlands, and as far south as the Isle of Wight. It occurs in the Channel Isles. It is, in general, the " Salicornia herhacea " of botanists, who appear to be content in this genus to distinguish subgenera and not species. [S. INTERMEDIA Woods {op. cit.), 30, 1851 ! Woods briefly described, under the name of S. intermedia, three erect Glass- worts said to be most abundant on the muddy salt-marshes of Sussex. One of the plants resembles S. pusilla, but has much longer and redder spikes. The second approaches the typical form of S. [europaa] herhacea in its yellowish green colour, hardly tinged with red, and in its cylindrical spikes an inch or more in length and eight or nine times their width ; but it has not more than eight or nine " sets of seeds." The third, in its bushy habit and colour, and in the form of its spikes, shows an affinity with S. ramosissima. It is obvious that one binomial cannot be retained for a medley of plants of such various origins and affinities ; and the name S. intermedia, if used at all, should be limited to the first of these three plants.] 6. Salicornia ramosissima. S. ramosissima Woods, op. cit., p. 29 (1851) ! This species occurs in the south of England, in the Channel Islands, and in Brittany. It is the prevailing species on the sandy salt-marsh at the Bouche d'Erquy. In a recent interesting paper by Miss Delf (Ann. Bot., April, 1911), it is confused with S. annua {— S. europcea). At Erquy, it is very polymorphic, in 182 THE JOURNAL OP BOTANY colour, size, and the amount of branching ; but it remains in all its forms quite distinct from S. europaa and the other annual species. In colour, S. ramosissima varies from apple-green to dingy red and crimson. Typically, it is much branched, often so much so as to be quite bushy ; but less branched and even branchless forms occur. On the Bouche d'Erquy, these forms are definitely related to variations in the amount of water and salt in the soil, and to the degree of exposure. When well-grown, *S'. ramosissima may be as much as 30 cm. in height. The vegetative segments are usually about 10 mm. long, rarely 20 mm. The terminal flowering spikes are not nearly so long as in typical forms of S. euroixza, and are usually not more than 5-10 (rarely up to 20, but perhaps these are hybrids with S. europcBo) mm. long : they are acute and tapering, and usually have only 4-6 flowering segments. The central flower is about twice as large as the lateral ones, and reaches about two- thirds of the way up the segment. S. ramosissima seems to be the only annual species which has constantly two stamens. 7. Salicornia pusilla. S. imsilla Woods, op. cit., p. 30 ! Giirke (op. cit.) erroneously cites this as synonymous with S. pygmcsa Pall. ; but as he also cites the latter as the original of Halopc])lis pygmaa, his statements carry little weight. S. jmsilla seems to be a very local plant, and can at present only be recorded from salt-marshes in Sussex, Hampshire, the Isle of Wight, and Dorset. It is grey-green in colour, and only grows to a height of about 12 cm. When well-grown, it is a branched plant with curved and graceful branches ; but less branched forms occur, and these seem to have furnished the plants on which Woods based his description. The vegetative segments are about 6-10 mm. long. The terminal flowering spikes are short (5-10 mm.), with only 2-4 flowering segments, which are almost globular in shape. The lateral flow^ers are only about a third as big as the central one, which reaches two-thirds of the way up the segment. ' The hairs of the seed are long, spirally coiled, and not crozier-shaped as in the other annual species. 8. Salicornia gracillima, sp. nov. S. p2isilla Woods var. qracillima Towns. Flo. Hampshire, ed. 2, 640 (1904) ! S. (gracillima), colore viridi, rubello, vel etiain rubro ; erecta, rigida, subracemosa, culmo circiter 12 cm. alto ; ramis ascen- dentibus, baud imparibus, parallelis ; horum sterihbus segmentis brevibus (circiter 5-8 mm. longis) ; spicis ol)tusis brevibus (cir- citer 6-12 mm. longis), 2-4 segmenta exhibentibus. Flores duos habet laterales, dimidio fere minores flore centrali, qui partem dimidiam segmenti certe superat ; stamine unico. Spicae etiam cellulas tum spirales cum spiculares continent. At present this plant is only definitely known from the south of England, e. g., Norfolk, Sussex, Hampshire, and the Isle of Wight. In this preliminary account, full details are not given of the SOME SPECIES OF SALICORNIA 183 occurrence and distribution of the strengthening spicular cells and the water-containing spiral cells which occur in the tissues of the species of Salicornia ; but it may here be mentioned that S. cjracil- linia and S. disarticulata are the only annual species in which spicular cells have been found. It is difficult to understand why Townsend {loc. cit.) placed the plant under S, pusilla. 9. Salicornia disarticulata, sp. nov. (Plate 514). S. (disarticulata), colore flavo-viridi, ramorum apicibus rube- scentibus ; erecta, rigida, ramosissima, culmo circiter 15-16 cm. alto ; segmentis brevibus, ad maturitatem facile disarticulantibus, (circiter 2-5 mm. longis) ; spicis terminalibus brevibus (circiter 2-6 mm. longis), segmenta 3-6 exhibentibus, spicis lateralibus brevissimis (circiter 1-3 mm. longis). Differt ab omni alia specie quippe quae spicarum segmenta singulos modo flores exhibent. Spicse etiam cellulas tum spirales cum spiculares continent. This species, distinguished by the ease with which the seg- ments disarticulate at or just before maturity, and by the solitary flowers, occurs in the south of England (Norfolk, Hampshire, and the Isle of Wight) and north France (the Bouche d'Erquy). It is characteristic of the drier parts of salt-marshes, and frequently grows among turf of Glyceria maritima. S. disarticulata is, in its anatomical characters, related to S. gracillima ; but its loosely attached segments and its constantly uniflorous "cymes" keep it separate from this and from all the other species of the genus. Subsection Prostrated. Plants prostrate or procumbent ; stamen 1 ; hairs of seeds crozier-shaped. The species of this subsection have been very greatly confused. 10. Salicornia Oliveri, sp. nov. S. (Oliveri), colore fusco-viridi, apud autumnum ad flavo- viridem tendente ; prostrata, culmo cix'citer 18 cm. longo, ramis duobus infimis in obtusos angulos pandentibus ; figurse triangu- lari approximans ; ramis paucis, nunquam non prostratis ; seg- mentis sterilibus circiter 8-15 mm. longis ; spicis cylindricis robustissimis, obtusis, circiter 25-40 mm. longis, 7-10 segmenta habentibus ; floribus subsequalibus — flore centrali partem segmonti dimidiam longe superante — cellulis cum spiralibus tum spiculari- bus nusquam apparentibus. At present this distinct and interesting species, named after its discoverer, Professor F. W. Oliver, is only known to occur on the Bouche d'Erquy. Here it is locally plentiful on un- stable sand which is subject to frequent tidal inundations. It grows with S. perennis, to which, unlike all the other annual species, it closely approximates in colour. It may yet be found on the south coast of England. 11. Salicornia Smithiana, sp. nov. S.procumhens auct., pro parte, non Sm. S. prostrata auct., pro parte, non Pall. S. (Smithiana), colore viridi, rubello, vel etiam rubro ; pro- cumbens, minime ascendens, culmo brevi, circiter 15 cm. longo, ramis paucis ascendentibus ; segmentis sterilibus brevibus (circiter 184 THiE JOURNAL OF BOTANY 10 mm. longis). Flores duos liabet laterales dimidio minores flore centrali, qui partem segment! dimidiam certe superat. Celluliae tum spirales cum spiculares nusquam apparent. This is tlie plant whicli English authorities have recently been iia,uiing '' S. 2)^'ocumbens Sm."; but Smith's description, figure, and specimen (!), and his account of its habitat prove that his plant was merely a form of S. europcBct toppled over by the tide. It occurs in southern England from Norfolk to Somerset. 12. Salicobnia peostrata. S. prostrata Pall. 111. Plant. 8, t. 3 (1803) ! S. prostrata auct., pro parte. S. promimhens auct., pro parte, non Sm. S. patula auct., pro parte. In addition to Eussia, the plant occurs in Belgium, and in north France on the Bouche d'Erquy. It should be found in the south of England, and may prove to be rather widespread in the salt-marshes of central and southern Europe. The plant is usually green in colour, prostrate, very much branched, the branches remaining quite prostrate, up to about 15- 20 cm. in length, and triangular in outline. This and S. ajjpressa differ from S. Oliveri and S. Smithiana in possessing very numer- ous secondary branches. From S. appressa it differs in its flower- ing spikes, which are cylindrical and obtuse. From S. Smithiana and S. ap)prcssa it differs in the much wider angles (about 90°) which the branches make with the main stem. The vegetative segments are about 8-10 mm. long, and the terminal flowering spikes about 6-12 mm. long. The lateral flowering spikes are about two-thirds as big as the central one which reaches nearly two-thirds of the way up the segment. 13. Salicornia appeessa. S. appressa Dum. in Bull. Soc. bot. Belg. 333 (1868) ! S. prostrata auct., pro parte, non Pall. S. procumbens auct., pro parte, non Sm. At present, this species is known from Belgium, the Bouche d'Erquy, and the south of England. It is given the rank of a species in the Index Kewensis. In colour it varies from a dingy red to a deep crimson. Like S. Oliveri and S. prostrata, it is a prostrate, triangular plant, but its branches are sometimes ascending at the tips. The main stem may grow to be 15 cm. or more in length. The vegetative seg- ments are about 7-9 mm. long. The terminal flowering spikes are short (10-12 mm.), tapering, and acute. The central flower is larger than the lateral ones, and reaches nearly to the top of the segment. Key to thu Species. I Perennial plants, small shrubs or dwarf shrubs ; lateral flowers ,) separated by the central one ; stigma bifid 2 1 Annual plants ; lateral flowers meeting below the middle one ; stigma \ much branched 5* * All the British species are included in 4 and 5. ALABASTRA DIVERSA 185 ;, (Seeds black, punctate, not hairy S. glauca. 1 Seeds greyish or greenish, hairy 3 o (Plant erect or suberect, ; hairs of seeds stout, conical... S. frtiticosa. (Mature plant prostrate ; hairs of seed very slender 4* [Branches freely rooting; lateral flowers smaller than the middle one J S. i^crennis. 1 Branches not rooting, plant leaving the ground by a single stem ; \ flowers nearly equal in size S. lignosa. (Plant erect (or pseudo-procumbent owing to being toppled over by 5- the tides, &c.) 6 (plant prostrate or procumbent 9 'Terminal flowering spikes long (up to 3-5 cm.) ; flowers nearly equal in size ; stamens visually 1, rarely 2 S. europcea. Branches fastigiate S. eurojpoia iorms, stricta. Branches spreading S. europcea iorma patuhi. I Terminal flowering spikes short (usually less than 1 cm.) ; lateral V flowers much smaller than the central one or absent 7 „ (Flowering spikes acute, tapering; stamens 2 S. ramosissima. (Flowering spikes blunt ; stamen 1 8 {Branches curved, graceful ; flowering segments almost globular ; hairs of seeds spirally coiled S. pusilla. Branches simple, subequal, parallel, stout, and rigid ...8. gracillima. Much branched ; segments freely disarticulating ; lateral flowers absent S. disarticulata. q (Plants simply branched 10 I Plants very freely branched 11 / Plant prosti'ate ; terminal flowering spikes cylindrical, stout, obtuse, long (25-40 cm.) ; flowers nearly equal in size S. Oliveri. 10 Plant procumbent or ascending; terminal flowering spikes short (1-2 cm. or less) ; lateral flowers about half as big as the V central one S. Smithiana. 1 Terminal flowering spikes cylindrical, obtuse ; branches making wide angles (about 'd(f) with the stem S. prostrata. Terminal flowering spikes tapering, very acute; branches making acute angles with the stem S. appressa. Description of Plate 514. Drawn by Mr. E. W. Ilunnybun for the New British Flora. (a) Salicornia disarticulata Moss (natural size), (b) Flowering spike (x 6). (c) Seeds (x 6). Specimens of this plant have been sent to the British and the Watson Exchange Clubs. ALABASTEA DIVERSA.— Part XX. By Spencer le M. Moore, B.Sc, F.L.S. (Concluded from p. 158.) 1. New or Eare Tropical African Plants. eubiace.e. Velvitsia calycina Hiern, Cat. Welw. PI. p. 771. In a large collection of Mr. Gossweiler's Angola plants received at the British Museum in 1907 are flowering and early fruiting specimens from Forte Princeza Amelia (no. 2124), which seem to be conspecific with Velvitsia calycina Hiern. Of this latter, a Welwitsch plant, 186 THE JOURNAL OF BOTANY the flowers have hitherto been known only by the description given of them in the great collector's accompanying note, all his specimens being in the advanced fruiting stage, at which period their large accrescent calyces lend them a peculiar appearance, well calculated to suggest the view that the plant bearing them belongs to a special generic type. In the Flora of Trojncal Africa (vol. iv. pt. 2, p. 362), Mr. Hemsley, still without opportunity of examining flowers, considers Velvitsia cahjcina to be a species of Melasma {M. calycimim Hemsl.). Flowers having now come to hand, it is possible to settle the question definitely. The flower of Gossweiler, 2124, is seen on dissection to have all the characters of Melasma, and to agree closely with Welwitsch's description above mentioned. True, the corolla-lobes Welwitsch states to be ovate, whereas Mr. Gossweiler's note speaks of them as orbicular ; iu effect, when moistened and laid out on the stage of the microscope, they seem to me to be neither the one shape nor the other, but suborbicular. This slight discrepancy may perhaps be accounted for by supposing that Welwitsch's flowers had already begun to shrivel up when his note was written. Then the stamens, says Welwitsch, are attached to the middle of the tube (" medio tubo fixa "), whereas, as Mr. Gossweiler writes, they are inserted not far from the base. The older collector gives white tending towards pale yellow with violet veins for the colour of the flowers, while Gossweiler describes them as " pure white, trans- lucent, puberulous, with longitudinal nervation in relief outside." In spite of these small points of difference, all the other characters seem so like in each as to leave no reasonable doubt that the two are conspecific. A few notes on the flower are appended : — Calyx florescens scabro-pubescens, 1 cm. long. ; lobi 4 mm. long. Corollge extus microscopice puberulae tubus 1 cm. long., ima basi 4 mm., superne 8 mm. lat. ; lobi inter se subaequales, suborbiculares, 8x7 mm. Stamina corollae tubo 2 vel 3 mm. supra basin inserta ; filamenta breviora 3 mm. longiora 5 mm. long. : antherae 2*5 mm. long. Ovarium ovoideum, 4 mm. long. ; stylus clavatus, ovarium duplo excedens, glaber; stigma hamatum, circa 3 mm. long. Gerardiina angolensis Engl, in Bot. Jahrb. xxiii. 507, var. nov. gracilis. Ob habitum gracilem, folia angusta (modo 1'5 mm. lat.), pedicellos cito 1-5-2 cm. long, a typo plane divergit sed vix species propria esse videtur. Hah. Angola, with GlumacecB and marsh herbs on the Wawaiela rivulet, Kaconda ; Gosstveiler, 3400. Sopubea laxior, sp. nov. Herba perennis metralis vel ultra, caule erecto tetragono sparsim vel frequentius ramoso glabro, foliis (internodiis satis distantibus) oppositis vel verticillatis vel etiam solitariis lineari-oblongis raro anguste oblanceolatis obtusis basi angustatis chartaceis pagina utraque scabridis, floribus medio- cribus pedicellatis in racemis laxis paniculatis satis elongatis digestis, bracteis linearibus pedicellis brevioribus, calycis scabridi lobis tubo brevioribus deltoideis acutiusculis margine breviter ALABASTRA DIVEESA 187 lanatis, corollas tubo subcylindrico lobis obovatis obtusissimis tubum paullulum excedentibus, antheris cohaerentibus exsertis. Hab. Angola, common in forests towards Catemba, near Malange, and between Kutchi and Kutelo ; Gossiceiler, 1095, 1096, 3168. Between Kuma and Kutsi ; Baum, 885. Folia pleraque 2-3-5 cm. long., 4-7 mm, lat. (summa minora, e. g. 15 X 3 mm.), margine breviter revoluta; costa centralis pag. inf. valde eminens. Inllorescentias 15- fere 30 cm. long. Bracteae + 6 mm., bracteolae 2 mm., pedicelli mox 5-7 mm. long. Fiores albi, violaceo-oculati. Calyx totus 4'5 mm. long., lobi 1*5 mm. Corollas tubus 4 mm. long., ima basi 2 mm., faucibus 4 mm. diam. ; lobi 5 mm. long, 4-4*5 mm. lat. Antherarum loculus fertilis 3 mm. long. Ovarium late ovoideum, vix 2 mm. long.; stylus 8 mm. long., glaber. Allied to *S'. ramosa Hochst., but differing from it chiefly in the sparser habit, the more scattered markedly broader leaves, the effuse inflorescences and longer pedicels to the flowers. Sopubea kacondensis, sp. nov. Suffrutex caule erecto ra- moso ramis erectis crebro foliosis tetragonis glabris, foliis pseudo- verticillatis aliis elongatis cum aliis multo brevioribus intermixtis linearibus obtusiusculis simplicibus vel trisectis rarius pinnatifidis supra scabriusculis subtus glabris, floribus mediocribus in spicam terminalem satis longam subdensifloram digestis, bracteis alabastra bene excedentibus vetustioribus foliis similibus junioribus gradatim imminutis lineari-lanceolatis acutis margine ciliato-lanata exempta glabris, bracteolis bracteis similibus nisi multo minoribus, calycis 5-goni usque ad medium divisi lobis lanceolatis ciliato-lanatis lobo postico ceteris latiore necnon bifido, corollas tubo calyce breviore superne ampliato lobis late obovatis obtusissimis tubura paullulum excedentibus, antheris cohaerentibus breviter exsertis. Hab. Angola, with Glumacece in moist meadows beside the Seculu river, near Kaconda ; Gossioeiler, 4312. Planta ex schedis cl. detectoris ultrametralis (4 ped. alt.). Folia in sicco fuscescentia, summum 5 cm. long., pleraque + 3 cm. ; segmenta foliorum minorum majoribus intermixtorum + 1 cm. long. Spicas circa 10 cm. long. Bracteas + 2 cm., bracteolas circa 6 mm. long. Fiores punicei. Calyx 10 mm. long. ; lobi •5 mm, long., lobus posticus binervosus. Corollas tubus 4-5 Uim. long., basi 3 mm. faucibus 5*5 mm. diam. ; lobi 5-5 mm. long. Antherarum loculus fertilis 4 mm. long. Ovarium ovoideum, 2 mm. long. ; stylus microscopice puberulus, 8 mm. long. This has the habit of »S'. lanata Engl., which is a woolly plant with entire leaves and pedicellate flowers. The affinity is with S. Welwitschii Engl., but the leaves of the new plant, when not simple, are divided close to the base, and not some distance up tlie leaf, and among other features, the calyx is glabrous except for the woolly ciliation of its lobes. Sopubea semula, sp. nov. Caulibus e rliizomate crasso pluri- bus erectis simplicil:)us superne densifoliis araneoso-villosulis delude glabrescentibus, foliis verticillatis rarius solitariis sessilibus anguste lineari-oblongis obtusis integris chartaceis utrinque pubescenti- 188 THE JOURNAL OP BOTANY bus, floribus sessilibus in spicam terminalem densam ovoideam aggregatis, bracteis anguste obovato-spathulatis apice obtusis ut bracteoltE sibi ipsis similes sed minores dorse saccatis extus araneoso-pubescentibus margine albis glandulis rubris crebro in- spersis, calycis 5-goni fusco-nitidi sursum puberuli lobis tubo brevioribus triangularibus acutis marginibus breviter albo-tomen- tosis, corollsB majusculae tubo superne amplificato lobis suborbicu- laribus tubo brevioribus, antheris inter se cohaerentibus breviter exsertis. Hab. Angola in shrubby pasturage at Kaconda ; Gossiveiler, 4240. Planta bispithaniea. Folia solemniter 2-2-5 cm. long., 2-4 mm. lat., summa breviora etsi baud angustiora in bracteas transeuntia. Spicffi nondum profecto evolutce 4-5 x 2-5 cm. Bracteae 9-12 mm. long., summum 5 mm. lat. ; bracteolae 5-6 mm. long. Calyx 7 mm. long. ; lobi 2-5 mm. long. Corollge tubus 10 mm. long. ; inferne 1-5 mm., faucibus 3*5 mm. diam. ; lobi circa 5x6 mm., margine crispuli. Antherarum loculus fertilis 3-5 mm. long. Ovarium ovoideum, 2 mm. long. Stylus villosus, apicem versus glaber, 13 mm. long. Nearest S. densiflora Skan, but showing too many differential features to render comparison necessary. Sopubea congensis, sp. nov. Verisimiliter planta circa bi- spithamea caulibus simplicibus erectis densifoliosis piloso-pube- scentibus deinde giabris, foliis sessilibus inferioribus sparsis superi- oribus verticillatis linearibus acutis integris membranaceis primo piloso-pubescentibus mox puberulis, floribus sessilibus confertis spicam brevem terminalem cylindricam referentibus, bracteis lineari-lanceolatis acuminatis calycem facile excedentibus glandulis paucis rubris instructis ut bracteoltB lineari-oblanceolatae calyxque albo-lanatis, calyce 5-gono ad medium usque diviso lobis triangu- lariljus obtusis, corollae mediocris tubo calyce vix longiore superne gradatim paullulum ampliato lobis suborbicularibus tubo aequi- longis, antheris inter se cohaerentibus plane exsertis'. Hab. Congo Free State, Kundelungu ; Kiissner, 2767. Folia + 2-5 cm. long., 1-2 mm. lat. Spicse 3-5 x 1'8 cm. Bracteae 8 mm. long., bracteolas 4 mm. Calyx 5 mm. long., summum totidem lat. ; lobi 2-5 x 2 mm. Corolla tubus 5-5 mm. long., inferne 1-75 mm., faucibus 2-5 mm. diam. ; lobi 6 x 5-5 mm., margine integri. Antherarum loculus fertilis fere 4 mm. long. Ovarium ovoideum, vix 2 mm. long.; stylus crassiusculus, inferne pubescens, circa 15 mm. long. The place of this is next keen enough after that date. . . . It is noteworthy, however, that it is due to his perseverance that a moss, Eurhyncliium ahbreviatum, was added to the list of British species. It had always been mistaken for E. Stvartzii. Ley would not admit the identity, and his friend Boswell, of Oxford, at last identified the species, which was well known on the Continent. Up to the last he was always interested in the mosses." As has been stated above, the Flora of Herefordshire was published in 1889. In pp. 217-220 of this Journal for that year appears an appreciative review by the Eev. E. S. Marshall. This work, the fruit of so many years' painstaking research, at once took a foremost place among our county floras, and has held it ever since. One of its conspicuous excellences is the frequency of brief critical notes, breaking the monotony of locality lists and proving of great help to less skilful botanists than the authors. AUGUSTIN LEY 205 The Flora was followed by the publication in this Journal of numerous supplementary papers by Ley, those in the 1894: and 1896 volumes being of especial interest. He also contributed the greater part of the botanical section to the Victoria History of the County of Herefordshire. His large and valuable herbarium now goes to the University of Birmingham. In addition to his frequent botanical rambles in most parts of Wales, and his very exhaustive exploration of his own county, Ley made expeditions from time to time to widely separated parts of the British Isles. His travels on the Continent — in Norway, the Tyrol, Switzerland, Normandy and Brittany, and the Eiviera — were frequent, though chiefly in earlier years. To Norway his first visit was as early as in 1863, when, with his father and brother, he explored neighbourhoods so unfrequented by tourists and under such primitive conditions that they were glad to eat the dog-biscuits which they had taken with them for their unfor- tunate dogs. But in his case, as of course with all vigorous natures, discomforts and hardships out of the common only added zest to the pleasure of exploration. In 1908, little more than two years before his death, conscious no doubt of somewhat failing health, he resigned his incumbency, and with his stepmother (always his zealous fellow-worker in the parish) removed to a cottage near Eoss. At the same time he placed himself on the Bishop of Hereford's Emergency List, and so to the end was constantly employed in helping his neighbours ; often taking the full charge of a parish, as he did last winter up to the middle of February. One of his most intimate botanical friends in recent years writes of him: "What a delightful companion he was! . . . not only a botanist, but a thoughtful and well-read man ; and men and women who knew nothing of botany always enjoyed his conversation." It could hardly have been otherwise. He was so frank, so refreshingly vigorous and unconventional, so out- spoken and yet considerate, and under all circumstances so full of cheery optimism, that none could altogether miss the charm of his companionship. He was so ready also to let you share in his ripe knowledge. When he could not agree with you, he lost no time in letting you see it, but always courteously and in such a way as to give you an impression of his readiness to learn any- thing that you might be able to teach, however contrary it might be to the conviction he had previously reached ; so that to many who knew him well intercourse with him was as cheering and bracing as a tonic. And all the more when one learned, as one soon did, that his breezy optimism was the reasoned outcome of candid consideration and real experience. It was not that he failed to see difficulties and dangers ahead in matters of deepest moment, but that he realized to the full how common a thing it is for the most satisfactory results to be thus reached. Hence the liopefulness which nothing damped, and the patience which often shamed one. "In many ways an ideal parish priest of a rural parish," ag 206 THE JOURNAL OF BOTANY was said of him at a large gathering of churchmen in Ross Chm'ch the day before his funeral, " kindly and generous-hearted, yet withal shrewd and practical, his was one of those lovable natures which attract to themselves all who come in contact with them." " The restored church at St. Weonard's," to quote yet one witness more, "and the bridge over the Wye built by his exertions ' to the honour of God, the lasting union of the two parishes ' — of Sellack and King's Capel — ' and the use of all,' testify to his care for the spiritual and general welfare of his parishioners; but those who knew him personally will be helped most by the recollection of his example of ' plain living and high thinking,' and of the purity and sunny simplicity of a life wholly dedicated to the service of the God of redemption and of nature." His body was laid to rest among his people in Sellack church- yard, and this was the comment of a friend who was present, " I never saw so many persons at a country funeral before." W. MoYLE Rogers. A REVISION OF THE GENUS HAM ELI A. By H. p. Wernham. This genus of Bubiacecs is confined to the New World, the earliest types described being the West Indian H. erecta and H. patens, upon which Jacquin (Stirp. Amer. 72, t. 50, 1763) founded the genus. Hamelia, comprising shrubs chiefly, with a few trees, is readily distinguishable as a genus by the regularly secund arrangement (except H. chrysantha) of the subsessile flowers in the cymes, and by the external appearance of the flowers themselves. The latter are usually tubular, often angled, with erect, inconspicuous corolla- lobes ; the stamens have long and linear anthers basi-fixed upon short membranous filaments inserted shortly above the base of the corolla-tube. A large floral disc is typically present, and this often persists as a large crown in the multilocular berry. The seeds are numerous, very rarely more than 1 mm. in their maximum dimension — usually much less — and more or less coarsely foveolate. Of the species dealt with in the present paper, H. patens Jacq. is by far the commonest and the most widely distributed, occurring as far north as Florida, and generally over Central America, the West Indies, Tropical South America and Brazil, and appearing south of the tropics, in Paraguay. H. lutea is next in order of wideness of distribution, although much rarer than H. patens, being found in various localities of the West Indies, in Central and South America, including Brazil. H. chrysantha is recorded as from Mexico, Jamaica, and Venezuela — widely separated dis- tricts. In all the other cases the distribution is, so far as our present knowledge goes, local only, in various parts of the West Indies, and Central and Tropical South America. A REVISION OF THE GENUS HAMELIA 207 Mexico is an important centre of distribution, with at least eleven species ; seven species occur in the remaining portion of Central America; seven in the West Indies; ten in Tropical South America, of which two only are recorded for Brazil. Specierum Clavis. Flowers with relatively long pedicels, terminating the branches of forking cymes. Leaves usually opposite, or sometimes 4-nate, glabrous 1. H. chrysantha. Flowers sessile, subsessile, or but shortly pedicellate, seciind upon the branches of forking cymes. Corolla 2-5 cm. or more in length (in H. magniloha barely 2"5 cm.), more or less widened above. Mature leaves glabrous, or slightly pubescent only on the veins beneath. Calyx-lobes not less than 5 mm. long ... 2. Zf. calycosa. Calyx-limb very shortly toothed. Corolla-lobes 8-9 mm. long, patent ... 3. H. magniloba. Corolla-lobes seldom exceeding 3 mm., erect. Leaves 20-25 cm. long; seeds distinctly pear- shaped 4. H. grandiflora. Leaves rarely exceeding 12 cm. ; seeds not pear-shaped. Berry subglobose, seeds 1 mm. and more across, with fine reticulum and dull surface 5. H. cuprcea. Berry oblong, seeds not exceeding \ mm., with coarse reticulum and glistening surface 6. U. ventricosa. Mature leaves with manifest indumentum, between as well as upon the veins beneath. Leaves ovate, 6-5-9 cm. x 4-5-5 cm., puberulous above 7. -H- liypomalaca. Leaves elliptic-oblong, 10-12-5 cm. x 4-5-5 cm., glabrous above 8. H. xorullensis. Corolla rarely exceeding 2 cm. in length. Leaves opposite. Leaves glabrous. Corolla campanulate above 9. H. lutea. Corolla tubular throughout. Leaves elhptic-lanceolate, flowers mostly ex- ceeding 1 cm. in length. Leaves 5-6 cm. long 10. H. axillaris. Leaves 15-24 cm. long 11. H.magnifolia. Leaves almost orbicular, flowers mostly less than 1 cm. in length 12. H. breviflora. Leaves not glabrous. Coroha and fruit glabrate 13. H. rostrata. CoroHa and fruit clothed with dense yellowish hairs 14. H. xerocarpa. Leaves whorled. Stamens reaching, or almost reaching, the mouth of the corolla, often exceeding it. Calyx-teeth relatively long, usually revolute. Corolla-tube glabrous 15. H. oyata. Corolla-tubo hairy on the outside... 16. H. lioviroscv. Calyx-teeth very short, deltoid, usually adpressed to corolla base. 208 THE JOURNAL OF BOTANY Leaves scabrid above with scattered papillae, often suborbiciilar 17. H. j^apMosa. Mature leaves glabrate, or with slight indu- mentum only on under midrib. Flowers mostly with manifest pedicels 18. H. pedicellata. Flowei'S almost or completely sessile. Leaves, petioles, and calyces pui'plish ; anthers with purple apical membra- nous process (ex descr.)... 19. H. suavcolens. Plant without purple tint; anthers without any conspicuous apical process. Leaves mostly 3-nate. Corolla-tube scarcely swollen at base 20. H. tuhiflora. Corolla-tube markedly ventricose at base 21. H. versicolor. Leaves mostly 4-nate. Cymes terminal, branches nodose 22. H. nodosa. Cymes borne on short axillary shoots ; branches not nodose ... 23. H. viridifoUa, Mature leaves with manifest indumentum, between as well as upon the veins, at least on the under side. Branches terete, vegetative parts generally of reddish tint, berry globose. Margin of calyx-limb almost entire (ex tab.) 24. H. sphcerocarpa. Branches sulcata, plant without red colour, berry not globose, calyx-limb distinctly toothed. Stem and inflorescence densely covered with ferruginous hairs ... 25. H. lamigmosa. Hairy covering not remarkably dense nor ferruginous 26. H. j^ateiis. Stamens not exceeding the middle of the corolla-tube 27. H. brachystcmon. 1. Hamelia chrysantha Swartz, Prodr. 46 (1788) ; DC. Prodr. iv. 442. Hab. Mexico : Hcsnke. Jamaica : Stvartz ! Harris, Fl. Jam. 9365 ! in Hb. Mus. Brit. Chilua : Hb. Moore ! in Hb. Kew. Tliis species is peculiar in the inflorescence ; the flowers have relatively long pedicels, and are not arranged in the manifestly secund manner of those of the other species. The flowers may attain a length of over 2 cm., and are sometimes appreciably widened above ; so that H. chrysantha may represent a transition to the large-flowered forms of the H. ventricosa type. The leaves are defined as opposite in the type description. One plant, how- ever, in the Kew Herbarium (Chilua: Hb. Moore) bears the leaves in whorls of four, but appears to differ in no other way from the type of H. chrysantha. This may be a variety ; but in the absence of further material, it may be well to refrain from any separate description. A REVISION OP THE GENUS HAMELIA 209 2. H. CALYCOSA J. D. Smith in Coulter Bot. Gaz. xii. 132 (1887). Hab. Pansamala, Guatemala : Tilrckheim, 454 ! Hb. Kew. 3. H. magniloba, sp. nov. Foliis nisi subtus super venam mediam glabratis verticillatis, corolla inter majores lobis pro genere magnis patentibus. Hab. Chontales, Nicaragua : Tate, 200 ! Hb. Kew. Arbor, ramulis complanatis glabrescentibus, foliis membranaceis ellipticis ca. 15 cm. x 7 cm. brevissime acuminatis apice subacute basi angustatis utrinque glabris nisi subtus super venam mediam et in axillis cum costis secundariis sparsissime puberulis petiolis 3- 3"5 cm. longis, cymis brevissime ferrugineo-pubescentibus floribus plerumque pedicellatis, calyce toto 3 mm. tubo subcylindrico vix pubescente limbo minutiusculo dentibus late deltoideis, corollas tubo 2-2 cm. longo sursum leniter ampliato (in ore ad 7 mm. lato) externe pubescente lobis patentibus ovatis acutis 9 mm. longis dorso pubescentibus, staminibus linearibus filamentis membra- naceis complanatis ca. 7 mm. longis antheribus 8 mm., stylo glabro complanato, bacca parvula, seminibus valde complanatis. This species is conspicuous for its large corolla-lobes and the approximate equality in length of anthers and filaments. 4. H. grandiflora Spruce MS. in Hbb. Mus. Brit, et Kew. Foliis glabris verticillatis, floribus magnis calycis et corollae lobis pro rata parvulis, seminibus pro rata magnis pyriformibus. Hab. Chimborazo : Spruce, 6193 ! Triana, 1759 ! x\rbor, ramulis glaberrimis complanatis vel angulatis, foliis pergamaceis verticillatis ellipticis vel elliptico-oblongis 20-25 cm. X 9-10 cm. breviter acuminatis subacutis basi brevissime angus- tatis vel subrotundatis utrinque glaberrimis nisi subtus super venas minutissime puberulis petiolis 5-9 cm. longis, stipulis lanceolatis obtusis coriaceis plus minus persistentibus 2-4 mm. longis, cymis glabratis terminalibus vel axillaribus in ramulorum furcis multifloris ramosissimis floribus plerumque manifesto pedi- cellatis, calycis tubo 4-5 mm. long, cylindrico valde sulcato glabro limbo 1-5 mm. long, dentibus brevissimis necnon subcarnosis con- caveis subacutis, corollae tubo 2-7 cm. long, sursum ad 1 cm. lat. ampliato glabro basi valde ventricoso lobis latis 2-5 mm. long, ovato-oblongis acutis, staminum linearum antheribus 2-1 cm. long, inclusis connectivo ad apicem in parvam laminam ovatam pro- ducto filamentis 5 mm. long. 4 mm. supra corolla? basin insertis, stylo filiforme 2-8 cm, long, stigmate anguste clavato subcostato, bacca ovato-lanceolata glabra 1 cm. long, x 5'5 mm. lat. disco ca. 2 mm. alto coronata, seminibus '8-1 mm. long, manifesto pyriformibus. 5. H. cuPR^A Grisebach, Fl. Brit. W. Ind. 320 (18G4). Hab. Jamaica : Marsh, 1772 ! Harm, Fl. Jam. 5271 ! 7705 ! 9343! 9635! Cuba: Wrifjht, 2664! Hb. Mus. Brit. ; Hb. Kew. 6. H. VENTRicosA Swartz, Prodr. 46 (1788). H. grandiflora L'Herit. (non Spruce) Sert. Angl. 4 ; H. pauciflora Willd. ex Spreng. Syst. i. 765. Hab. Jamaica: Alexander \ Macfadyen\ Distan\ Shakespearl 210 THE JOURNAL OP BOTANY Wright ! Srvartz ! Cuming, 18 ! 50 ! Faivcett, Fl. Jam. 8750 ! Harris, Fl. Jam. 9355 ! Hb. Mus. -Brit. ; Hb. Kew. 7. H. HYPOMALACA Eobinson in Proc. Am. Acad. xlv. 406 (1910). Hab. Mexico, Durango : Bose, 2304. 8. H. xoRULLENsis H. B. & K. Nov. Gen. et Sp. iii. 414 (1818). Hab. South Mexico, JoruUo : Humboldt (& Bonijland. Colima : Palmer, 1164 ! Hb. Kew. 9. H. LUTEA Kohr ex Smith in Eees, Cycl. v. 17, n. 4 (1811). H. chrysantha Jacq. Coll. iii. 204 (non Swartz). Hab. Ind. Occ. : Von Bohr ! Jamaica : Purdie ! Marsh, 627 ! Harris £ Britton, Fl. Jam. 10537 ! Thompson, Fl. Jam. 6473 ! Porto Eico : Sintenis, 1458 ! 2778 ! Eggers, 2816 ! S. Domingo : Tilrchheim, 2782 ! St. Thomas : Eggers, 351 ! Cuba : Wright, 234! 1265! Costa Eica: Tonduz, 10052! 70521 Panama: Hayes, 364 ! Nicaragua : Tate. 198 ! 199 ! Seemann, 114 I 136 ! 137! Colombia, S. Marta : Smith, 1818! Trinidad: Purdie \ Peru : Sjmice, 4231 ! Brasil : Poeppig, 2160; Ule, 5454 ! Hb. Mus. Brit. ; Hb. Kew ; Hb. Univ. Cambridge. This species is readily recognizable by its opposite leaves, small campanulate corolla, and longish subfiliform calycine teeth, often somewhat curled. 10. H. AXILLARIS Swartz, Prodr. 46 (1788) ; Fl. Ind. Occ. i. 443 ; DC. Prodr. iv. 442. H. appendiculata Gaertn. f. Fruct. iii. 64 ; H. lucida Desf . Tabl. ed. ii. 135. Hab. Jamaica : Swartz ! Wright ! Hb. Mus. Brit. ; Tond^iz, 9998 ! Hb. Kew. Var. /5 appendiculata DC. Prodr. iv. 442. I have had no opportunity of seeing this plant ; it is recorded as collected in Porto Eico. Grisebach (Fl. Br. W. Ind. 320) merges this species in H. lutea Eohr, describing it as "a form with the cymes exceeded by dichotomous branches." Swartz, however, describes the corolla as tubular, as is that of the specimen in the National Herbarium collected by him, which is quite distinct from H. lutea ; I see no reason, therefore, why H. axillaris should not be maiijtained. 11. H. magnifolia, sp. nov. Foliis magnis glaberrimis oppo- sitis, floribus in cymis glabris sessilibus, corolla inter minores tubulosa sursum vix ampliata. Hab. Costa Eica : Pittier & Durand, 10091! Hb. Mus. Brit. ; Pittier, 9904 ! Hb. Kew. Arbor glabra, ramulis complanatis, foliis firme pergamaceis oppositis elliptico-lanceolatis 19-24 cm. x 7-9-4 cm. leniter acu- minatis apice obtuso basin versus angustatis utrinque glaberrimis petiolis superne canaliculatis 2-5-4 cm. longis, stipulis subulatis truncatis mox deciduis vix 5 mm. longis, cymis verisimiliter termi- nalibus ramosissimis multifioris necnon laxiusculis floribus sessili- bus, calycis totius 3-3 mm. longi tubo cylindrico glabro valde sulcato lobis minutiusculis demum subcarnosis, corollge tubo sub- cylindrico 1-5 cm. longo sursum vix ampliato glabro ad basin valde ventricoso manifesto quinquecostato lobis latiusculis ca. A REVISION OP THE GENUS HAMELIA 211 1 mm. longis triangularibus obtusis, staminibus linearibus fila- mentis 3 mm. longis plus minus complanatis ca. 4-5 mm. supra corollae basin insertis antheribus 7'2 mm. longis nee exsertis, stylo filiforme complanato 1'3 cm. longo, bacca elliptico-oblonga glabra vix sulcata, seminibus manifesto foveolatis '5 mm. diam. A very distinct species, having affinities with H. lutea Eohr, but recognizable by its large leaves and small tubular flowers in large glabrous cymes. 12. H. BEEviFLORA A. Rich. Fl. Cub. ii. 17 (1853). Hab. Cuba, in montosis prope Santiago : Linden, 2048. Conspicuous for its broadly ovate, suborbicular leaves and small flowers. 13. H. EOSTEATA Bartlott ex DC. Prodr. iv. 442 (1830). Hab. Mexico, Acapulco : Hmnke. 14. H. XEROCAEPA Kuntze, Eev. Gen. PI. 284 (1891). Hab. Nicaragua : Seemann, 120 ! Hb. Mus. Brit. ; Tate, 222 ! Hb. Kew. 15. H. ovata, sp. nov. Foliis glabris verticillatis, floribus sessilibus secundis, corolla inter minores glabra staminibus corollae tubum subeequantibus, calycis dentibus longiuseulis. Hab. Venezuela : Funcke d; Schlim, 538 ! Hb. Mus. Brit. Frutex (?),ramulis compressis glabris, foliis verticillatis glabris ovatis 9-11 cm. x 5-6 cm. basi brevissime acutis vel subrotundis apice breviter acuminatis acutis, stipulis parvis crassiusculis del- toideis persistentibus, cymis terminalibus subglabris breviter pedicellatis, calyce infundibulari vix pubescente lobis oblongis truncatis ssepius revolutis 2-3 mm. longis, corolla tubulosa 1-4 cm. longa glabra basi ventricosa, staminibus linearibus corollre tubum subaequantibus, bacca . . . Related to H. Eovirosana, but readily distinguished by its almost complete glabrousness, and by the shape of the leaves. 16. H. Rovirosse, sp. nov. Foliis glabratis verticillatis, calycis lobis revolutis pro rata longiuseulis, corolla inter minores tubulosa extus pubescente-hirta. Hab. Marshes on banks of S. Sebastian River, Tabasco, Mexico, Hb. Bovirosa, 499 ! in Hb. Kew. VerisimiHter frutex ramulis teretiusculis sparse pilosis, foliis 3-nis ellipticis 5-10 cm. x 2-5-4 cm. utrinque angustatis apice acuto petiolo 5 mm.-l-4 cm. utrinque glaberrimis nisi subtus praesertim super venam mediam et in axillis cum costis secundariis sparse pilosis, stipulis deltoideis apice in filamentum ad 5 mm. longum acuminato, cymis terminalibus laxiusculis pubescentil)us breviter pedicellatis, calyce sparse piloso lobis lineari-oblongis apice trun- cate ad 3 mm. longis revolutis, corolla tubulosa basi vix ventricosa extus subhispidula, staminibus linearibus, stylo filiforme basin versus plus minus ampliato. I have not seen a fully opened flower of tliis species ; but it may be readily distinguished l)y the length and shape of the calyx- lobes and the indumentum of the corolla. 212 THE JOURNAL OP BOTANY 17. H. PAPiLLOSA Urban, Symb. Antill. v. 508 (1908). Hab. Jamaica : Harris, Fl. Jam. 8747 ! in Hb. Mus. Brit. ; Harris, Fl. Jam. 8957 ! Hb. Kew. 18. H. pedicellata, sp. nov. Foliis glabratis verticillatis, floribus plerumque pedicellatis secundis, corolla inter minores. Hab. Colombia, Santa Marta : Smith, 394 ! 1817 ! Von Bohr, 65 ! Trinidad : Fendler, 446 ! Dominica : Anderson ex herb. Forsyth in Hb. Kew ! St. Vincent's : Bev. L. Guilding in Hb. Kew ! Hb. Mus. Brit, and Hb. Kew. The specimen from the herbarium of C. M. Lemann, cultivated in the Serampore garden (Hort. Bot. Calc), and now preserved in the Cambridge University Herbarium, must also be referred to this new species. The native locality is given as " N. of S. America," and the plant is named Hamelia patens. Frutex (?), ramuHs glabrescentibus subfistulosis, foliis saepius 4-natis ovatis vel ellipticis 10-12 cm. x 4-5-2 cm. breviter acumi- natis acuto apice basi brevissime acutis vel vix rotundatis utrinque glabratis subtus cystolithis densiuscule distinctis petiolis 1-1-5 cm. longis, stipulis lanceolatis obtusis 2 mm, longis, cymis plerumque terminalibus multifloris minute et sparse pubescentibus floribus saepius pedicellatis, calycis tubo campanulato 2-5 mm. longo vix pubescente lobis brevissimis basi lato triangularibus, coroUae tubo 1-5 cm. longo sursum leniter sed paucissime ampliato glabro basi ventricoso lobis brevissimis truncatis ad 2 mm. latis, staminum linearum antheribus 7 mm. filamentis 5 mm. longis 4 mm. supra corollae basin insertis tubum vix excedentibus, stylo filiforme 1-5-1-6 cm. longo basi aliquantulum dilatato stamina subaequante, bacca glabra oblonga v. ovato-lanceolata 7-9 mm. x 3-4-5 mm. pedicella ad 3 mm. 19. H. suAVEOLENs H. B. & K. Nov. Gen. Am. iii. 414. H. odorata Willd. ex Eoem & Schult. Syst. v. 267. Hab. Near Garapatas, Badillas, &c., banks of E. Magdalena. 20. H. tubiflora, sp. nov. Foliis verticillatis glabratis nisi subtus nonnunquam super venam mediam obscure pubescentibus corolla inter minores tubulosa nee ventricosa glabra. Hab. Colombia, Barranquilla : Ifo^^ow! Santa Marta': Purdiel Hb. Kew. Frutex (?) ramulis glabris, foliis ternatim verticillatis ellipticis 6-12 cm. X 3-5 cm. glabratis saepius cystolithis dense distinctis breviter acuminatis subacutis basi brevissime- angustatis vel subrotundatis petiolis ad 2 cm. longis, stipulis deltoideis per- sistentibus vix 2 mm. longis, cymis terminalibus necnon axillaribus glabratis vel minutissime pubescentibus paucifloris, calycis tubo 1-5-2 mm. longo glabriusculo obconico lobis -8 mm. longis triangularibus basi lato et apice subacute, corollas totius 1-65 cm. longae tubo glabro cylindrico aliquantulum basi ampliato nee ventricoso lobis vix 1 mm. x 1 mm. ovato-oblongis apice obtuso, staminum linearum antheribus 1 cm. longis basi brevissime sagittatis filamentis 5-5 mm. longis 3-5 mm. supra corollae basin insertis, stylo 1-5 cm. longo filiforme stigmate parum manifesto, bacca parva glabra ovato-oblonga. A REVISION OF THE GENUS HAMELIA 213 Nearly related to the Mexican H. versicolor, but distinguished especially by the shape of the corolla-tube and the size and shape of the fruit. 21. H. VERSICOLOR Gray ex S. Wats, in Proc. Am. Acad. xxii. 416 (1886). Hab. Barranca, Mexico : Palmer, 125 ! Pringle, 2569 ! 9828 ! San Bias : Lay & Collie ! Hb. Mus. Brit. ; Hb. Kew. This species is notable for its relatively large globose berries, about 8 mm. in diameter. A curious inaccuracy appears in the original description of this species, the number of stamens being given as " saepius 7." The pentamerous condition of the androecium is so constant throughout the genus that the presence of a larger number of stamens in any flower under examination would reasonably arouse some suspicion of abnormality. I have, however, dissected several flowers of the type-plant itself {Palmer, 125), but in no case was the number of stamens other than 5 ; and this applies also to the other specimens examined. 22. H. NODOSA Mart. & Gal. Bull. Acad. Brux. xi. i. 234 (1844). Hab. South Mexico, near Mirador, at 3000 ft. : Galeotti, 2581. 23. H. viridifolia, sp. nov. Foliis glabratis verticillatis, cymis in ramulis brevibus lateralibus dispositis corolla inter minores, seminibus minutissimis vix foveolatis. Hab. Costa Eica : ToncUtz, 13,867 ! in Hb. Kew. Arbor (?) ramulis junioribus sparse pubescentibus sulcatis senioribus glabris obscure quadrangulatis, foliis 4-nis ellipticis 4-5 cm. X 2-2-5 cm. vix acuminatis apice obtusiusculo basin versus vix angustatis utrinque nisi immatims glabris crassiusculis superne olivaceo-viridis subtus multo Iffivioribus petiolis ad 5 mm. longis obscure pubescentibus et nonnunquam subalatis, stipulis crassis deltoideis mox deciduis, cymis ramulos axillares termi- nantibus laxiusculis pubescentibus floribus sessilibus, calycis tubo campanulato 2-3 mm. longo pubescente lobis minutis (-7 mm. longis) deltoideis, corollae tubo subcylindrico 1-7 cm. longo sursum parum ampliato glabro ad basin vix ventricoso lobis ca. 1 mm. longis late triangularibus subtruncatis, staminibus lineari-clavatis filamentis complanatis 3'8 mm. longis ca. 2"2 mm. supra corollye basin insertis antheribus 7"8 mm. longis ca. 2 mm. exsertis, stylo filiforme 1-2 cm. longo corollas tubum subaequanti, bacca elliptica tenuiter pubescente, seminibus pro genere minutis necnon vix foveolatis. This species seems to have affinities with the Mexican H. nodosa Mart. & Gal., but is distinct in the relative smallness of its leaves and flowers, the arrangement of the inflorescences, and the character of the seed. 24. H. SPH^ROCARPA Euiz & Pavon, Fl. Per. ii. 69, t. 221, fig. b (1799). Hab. Peru, woods, Chinchao Quebrada, near Challana : Ruiz & Pavon. Closely related, apparently, to H. iiatens ; according to tlio 214 THE JOURNAL OF BOTANY figure, the most notable differences are the shape of the fruit and the entire absence, in this species, of calycine teeth. 25. H. LANUGINOSA Mart, et Gal. in Bull. Acad. Brux. xi. 1, p. 233 (1844). Hab. South Mexico, Jalapa, Mirador, &c. : Galeotti, 2615. Pringle, 8150 ! and Botteri, 502 ! in Hb. Mus. Brit, seem to answer to the description of this species ; but in view of the extensive range of variability in the indumentum of H. patens, I am unable to decide the point without consulting the type of H. lamiginosa, which has hitherto been unavailable to me. 26. H. PATENS Jacq. Stirp. Amer. 72, t. 50 (1763) ; DC. Prodr. iv. 441 ; K. Schumann, Fl. Bras. vi. vi. 322. H. chrysantha Sieber ex Steud. Nom. ed. ii. i. 721 ; H. coccinea Swartz, Prodr. 46 ; H. erecta Jacq. Enum. PI. Carib. 16 ; H. latifolia Eeichb. ex DC. Prodr. iv. 442 ; H. verticillata M09. & Sesse ex DC. Prodr. iv. 442. To the list of synonyms must be added Schcenleinia thyrsoiclea Miers, in Proc. Eoy. Hort. Soc. iv. 186. The type of this plant was collected at Eio Magdalena, New Granada, by Weir (no. 26), and passed, with Miers's collection, into the National Herbarium. The type is accompanied by sketches of Miers's own examination, and, with one exception, they leave no doubt that the plant is a Hamelia. This exception is, that the ovary is figured as bilocular. Owing to the poorness of the material, I have been unable to ascertain definitely if five loculi are present, as typically in the genus ; but I find that there are certainly more than two. There is, moreover, a duplicate specimen in the Kew Herbarium bearing the words " Havielia, fide Triana" H. ixitens is a shrub which is cultivated extensively in the Old World tropics, particularly in India (Cooke, Fl. Bombay, i. 626 ; Prain, Bengal Plants, 563), where, Prain says {loc. cit.), it is " often subspontaneous near villages in Central Bengal." The plant collected in Bangkok by Zimmermann (no. 76 ! Hb. Mus. Brit.), bearing the name Hamelia ventricosa Sw., is a form of H. i^atens, interesting on account of its leaves, borne four in a whorl instead of the more usual number, three. Hab. Florida: Nash, 1278! Bugel, 311! 312! Pollard, 27! Curtiss, 1128! 5500! Mexico: Anclrieux, 329! 330! Liehmann, 55 ! 56 ! Gonin, s. n. ! Salle, 502 ! Palmer, 1277 ! Jurgensen, 1 ! Bourgeau, 1607 ! 2627 ! Coulter, 202 ! Guatemala : Salvin & Godman ! /. D. Smith, 1974 ! Yucatan : Gaumef ! Nicaragua : Seemann, 136 ; Tate, 198 ! Baker, 2002 ! 2478 ! Costa Pica : Endress, 19 ! Lehmann, 1778 ! Tonduz, 11,460 ! 12,813 ! Cooper ex herb. J. D. Smith, 5790! Polakowsky, 142! 241 ! Honduras, San Carlos, Gulf of Fonseca : Barclay, 2664 ! Panama: Hinds ! Hayes, s. n. ! Cuming, 1235 ! Colombia : Harttceg, 1054 ! Triana, 138 ! Holton, 439! Kalbreyer,2QQ\ 852\ Cmning, 12351 West Indies, Jamaica : Macfadyen ! Marsh, 636 ! Porto Rico: Heller, 3391 Sintenis, 65 ! 2851 ! Cuba : Comhs, 29 ! Wright, 227 ! TranqueviUe, 310 ! Bahamas : Northrop, 40 ! Martinique : Hahn, 428 ! Dominica : Hoskin, 14! Hispaniola : Sivartzl Trinidad: Fl Trin. 32! A REVISION OF THE GENUS HAMELIA 215 Venezuela : Moritz, 464 ! Fendler, 591 ! Ecuador : Spruce, 6226 ! Sinclair \ Peru: Ule, 6469! Spruce, 4188! Matthetvs, 1498! 3135! Brazil: Bieclel, 199! 423; Blanchet, 4, 101, 1359! Weir, 77 ! St. Hilaire ! Glocker, 148 ! Burchell, 7405 ! i?er/?jeZZ, 132* ! Neuivied, 84; Claussen, 272 a; Mendonca, 160; Schenck, 3741; Glaziou, 706 ; Mose?i, 1348, 4170 ; Pohl, 5134 ; Sello, 626, 1229, 1791. Bolivia: Banj/, 1222 ! BnVZ(/es, s. n. ! Paraguay: Hassler, 5212! 6854! 7773! 10,283! Fiebrig, 4260! Hb. Mus. Brit.; Hb. Kew ; Hb. Univ. Cambridge. This species, which has of all the widest distribution, is also so variable that the task of finding constant and critical characters is one of considerable difficulty. Some stress has hitherto been placed upon the number of leaves at each node ; and in so far as the distinction lies between the opposite and the whorled arrange- ment, this seems to be constant for the species respectively con- cerned. The numbers three or four, however, are rarely constant for any species. In H. pateyis the number three is certainly by far the commonest, and this number appears in previous descrip- tions ; but the 4-nate arrangement is to be found, e. g. in the herb. Miller specimen (Hb. Mus. Brit.); in /. D. Sfiiith, 1974; in Fl. Trinitatis, 32 ; in Barclay, s. n. (Hb. Kew.) ; in Spruce, 6226 ; in Zimmermann's plant from Siam cited above ; and in a specimen cultivated by Henslow in the Cambridge Botanic Garden, and preserved in the University Herbarium. The 5-verticillate arrangement occurs in the following plants gathered in two widely separated districts : — Costa Eica, Tonduz, 12,813; Cooper, herb. J. D. Smith, 5790: Honduras, Barclay, 2664 : Bolivia, Bang, 1222. In all the circumstances it seems doubtful whether De Candolle's variety y qidnifolia (Prodr. iv. 442), founded upon a plant collected in Mexico by Mo^ino & Sesse, should be maintained. I have not seen this plant ; but those cited above differ from the normal H. piatens in no way save the number of leaves in a whorl, and this cannot be regarded as a constant feature. The indumentum of the leaf, moreover, is very variable in extent. In some cases both sides of the leaf have a hairy covering, that of the upper being, however, much less conspicuous than that of the lower side ; the ventral surface, indeed, is often glabrescent and even quite glabrous. The dorsal indumentum shows every transition from an almost woolly character to minute pubescence confined largely to the veins. In the former case H. j^atens is nearly linked, apparently, with H. lanuginosa, a species which I have had no opportunity to examine. In the forms which approach glabrescence, H. ixitens reveals affinity with the glabrate group (species 14-22), typified in H. papillosa, H. pedicellata, H. versicolor, and H. viridifolia. The more sparsely pubescent forms of H. patens may be distinguished primarily from the members of this latter group by the presence of pubescence on the veins of the third order, as well as on the median and secon- dary veins, and, as a general rule, between the veins as well as upon them. 216 THE JOURNAL OF BOTANY In the inflorescence, H. patens displays the characteristic arrangement typical of the genus,- and I find no variability in this regard in the direction of the aberrant H. chrysantha (q. v. supra). The corolla is essentially of the smaller type, and is always tubular, being never appreciably inflated above. The calyx-teeth are of the very short, broad, erect type more usual in the genus ; the more conspicuous calyx-lobes of H. calycosa, H. lutea, H. axillaris, H. xerocarpa, H. ovata, and H. Bovirosce do not appear to be foreshadowed among the forms of H. patens. The following features, then, may be suggested as constant and critical for this species : — Leaves usually three, sometimes more, at each node, pubescent on lower surface, sometimes on the upper also, and generally on the secondary and tertiary veins and in the meshes of the reticulum as well as on the median vein ; the calyx has very short erect teeth with broad base ; the corolla is tubular, never exceeding 2 cm. in length, and is markedly ventricose at the base ; the stamens at their shortest almost reach the corolla-mouth, and sometimes exceed it. The following appears to be a well-marked variety of H. patens ; there seems to be no justification, at any rate, for regarding it as a distinct species : — ■ P AXiLLAROiDES, var. nov. Foliis angustis elliptico-oblongis sparse pubescentibus, cymis paucifloris axillaribus plerumque in ramulorum dichotomorum furcis apparentibus. Hab. Ind. Occ. : de PontJiieu ! in Hb. Mus. Brit. This variety bears a superficial resemblance to H. axillaris ; but it is readily separated therefrom by its ternately whorled pubescent leaves. 27v H. brachystemon, sp. nov. Foliis glabriusculis verticil- latis, corolla inter minores tubulosa, staminibus corollge medium vix excedentibus. Hab. New Grenada, Ocana : Schlim, 547 ! Hb. Kew. Frutex (?) ramulis gracilibus glabris subteretibus-, foliis ter- natim verticillatis ellipticis aliquanto asymmetricis 8-11 cm. x 4-5 cm. acuminatis acutis basi in tenerum petiolum^ ad 3 cm. longum angustatis supra glaberrimis subtus glabratis cystolithis frequentibus, stipulis triangularibus obtusis ad 3 mm. longis, cymis minute pubescentibus paucifloris, calycis totius 3-5 mm. longi tubo elliptico glabrato lobis latis necnon brevissimis, corollse tubo glabro 1-5 cm. longo parum ventricoso lobis 1 mm. longis x 2-1 mm. latis obtusis margine sinuato, staminum linearum antheri- bus 5 mm. filamentis 2 mm. longis 2-8 mm. supra corollae basin insertis, stylo filiforme stamina subaequante, bacca . . . This species is unique, among those described in the present paper, in the shortness of the stamens relatively to the corolla- tube, and is recognizable by its slender brauchlets and petioles, and somewhat asymmetric leaves. 217 LOPHOZIA SCHULTZII (Nees) Schiffn. var. nov. LAXA. By W. H. Burrell, F.L.S. Cffispitibus densis, caulibus ascendentibus, ad 6 cm. longis, 2-4 mm. latis, radiculosis ; foliis non latioribus quam longis ; cellulis circa 23 x 27 /* apud marginem apicis, usque ad 50 x 100 ^ versus basim mediam ; stipulis parvis lanceolatis, ciliatis, '25- •5 mm. longis, circa -09 mm. latis ; paroicis ; foliis perigonialibus transversis et gibbis, in amentulis ; calycibus subito in ore con- tractis, in rostrum prominens productis ; foliorum margine antica ad basim non dentata. In loco uliginoso, Flordon, Norfolk, 25 m. alto legit W. H. Burrell, vii. 1909. This plant was at first assumed to be Lophozia Miilleri (Nees) Dumortier, but Mr. W. E. Nicholson was not satisfied about its identity, the areolation being more lax than in the specimens with which he compared it, and in the autumn of 1910 he sent it to Prof. V. Schiffner, of Vienna, who detected the paroicous in- florescence, and placed it as a form of L. ScJmltzii, for which he proposed the varietal name " laxa." This determination did not pass unchallenged, and a gathering was sent to Herr Kaalaas, of Christiania, who was good enough to examine it. He agreed that the inflorescence was paroicous, and said that the plant could not be referred to L. Kaurini, but must either be considered a form or variety of L. Schultzii, or be set up as a proper though " small " species. L. ScJmltzii belongs to a polymorphous group of Hepatics having bifid leaves and smooth perianths, which Karl Mliller, in Eabenhorst's Kryptog amen- Flora, places in a section " Leiocolea," with L. Miilleri for the type. L. Schultzii is paroicous, with the alternate, succubous, subimbricate, obhquely inserted leaves and aromatic odour, which are more or less common to the group, and is restricted to lowland marshes. The principal characters which distinguish it are the tall erect habit, the inflorescence, and the large deeply segmented amphigastria. Considerable weight is attached by authors to these latter organs ; both Schiffner and Muller cite the large, much cut under leaves as a reliable character for distinguishing it from L. Hornschuchiana, the species with which, in the barren state, it is most likely to be confused. The perianth is described by Muller as cylindrical or clavate, with the mouth abruptly contracted and slightly beaked. In the Ottawa Naturalist for April, 1903, A. W. Evans describes L. Butlieana Limpr. (= L. Schultzii), from Yukon, as having the perianth gradually narrowed above the middle, not distinctly beaked nor contracted at the mouth. A tabular comparison of some of the characters of the Norfolk plant with those assigned to the species will show the chief points of agreement and divergence :— Journal of Botany.— Vol. 49. [July, 1911.] R 218 THE JOURNAL OF BOTANY Type Characters. Plant 3-4 mm. broad. Leaves broader than long, up to 3 mm. wide at base. One to three small teeth fre- quently occur on the antical margin near insertion. Leaf-cells 30 /x diam. near mar- gin ; 30 X 45 ft near middle of leaf. Amphigastria very distinct and large, divided to the base in lanceolate ciliated segments. Male bracts not gibbous (nicht bauchig geholt). Norfolk Plant. Some plants reach 4 mm., but 3 mm. is the more general maximum width. Slender growths of 2 mm. width are common. Breadth about equals length ; 1-5 to 2 mm. is an average length and breadth. Antical margin near insertion without teeth. At apex the cells average 23 x 37 //., increasing in size to- wards middle near insertion to 50 X 100 //. Amphigastria small and ob- scured by the radicles, though numerous and easily found when searched for ; lanceo- late, toothed, and ciliate at base, and here generally about five cells wide. Male bracts distinctly gibbous, insertion transverse, forming well-defined amentulae or Ahren. Bearing in mind the stress laid by authors on the large deeply cut amphigastria as a character of diagnostic value, I have had some difficulty in admitting that the Norfolk plant with these organs so small belongs here ; the perianth also offers a difficulty, being conspicuously contracted and beaked at the mouth. The habitat is the marshy bottom of a small valley in a chalky Boulder Clay area at 80 ft. O.D. A considerable amount of calcium carbonate is precipitated on the marsh floor from water which drains through the valley, and the submerged vegetation shows a white incrustation when dried. Although no other members of the Leiocolea section have been detected in the county, a large quantity of our plant occurs in this station, forming dense spongy tufts over a limited area, where it is associated with such hygro- phytes as Drosera, Pinguicula, Parnassia, Jimcus, Fissidens adi- antoides, Hypnum stellatum, H. molluscum, &c. The capsule is normally terminal, and takes two seasons to mature. When fertilization has been effected, elongation of the stem usually ceases until the spores are matured, although an innovation sometimes arises from below a fertile perianth. In a large number of cases the perianths are barren, and new growth pro- ceeds by innovation, stems frequently showing three annual seg- ments, each terminated by a small barren perianth. It follows that the maturing capsules are overtopped by the new growth of AFRICAN MILLETTIAS 219 the surrounding barren stems, and must be sought at the base of the tufts, where they are somewhat obscured by their brown covering leaves from which chlorophyll has disappeared. The known distribution of L. Schultzii is mainly subarctic ; several records exist for North Germany, Denmark, Norway, and Sweden ; it has been found near or within the Arctic Circle, at Sarek in Lapland, Yenisei in Siberia, and the Yukon territory in America. At Flordon in Norfolk, 52° 32' latitude, its first re- corded station for the British Islands, it reaches one of its most southerly points, being on practically the same parallel as Neumark and Mohrin in Brandenburg. I am indebted to Prof. V. Schiffner and Herr Kaalaas for their weighty assistance in determining this very interesting addition to the British Flora, and to Rev. D. Lillie, Messrs. D. A. Jones, S. M. Macvicar, W. E. Nicholson, and A. Wilson for their criticism and the loan of specimens of this species and its allies. AFRICAN MILLETTIAS. By S. T. Dunn, B.A., F.L.S. By the courtesy of the Director of the Royal Botanic Gardens, Kew, the writer has recently been afforded the opportunity not only of studying the rich collections of the genus Millettia pre- served in the herbarium of that establishment, but of supple- menting these with abundant material olitained on loan for him from Berlin and Paris. His thanks are especially due to Dr. Stapf, Dr. Harms, and M. Gagnepain for the friendly assistance that they have given him regarding the collections in the herbaria with which they are respectively connected. It will probably be a long time before so complete a mass of African Millettia material can be brought together again, and as full advantage as possible has been taken of the opportunity thus offered for revising that perplexing genus. It has been particu- larly fortunate that the types of Bentham's, Hooker's, and Baker's species among the older African collections at Kew, and seldom authentically represented in other herbaria, could be compared with the more recent collections of French and German explorers. Several doubtful points have been elucidated by reference to Welwitsch's beautiful series of types in the Herbarium of the British Museum (facilities for consulting which were kindly given by the Keeper). It has been necessary to distinguish many new species, speci- mens of which, bearing manuscript names, will be found in the four herbaria mentioned. Pendmg the appearance of a more elaborate description and arrangement of all the species, which is now in preparation, it will therefore be convenient to indicate as briefly as possible the relations of these new species with their allies, and the following keys have been constructed with tliat object :— R 2 220 THE JOURNAL OF BOTANY MiLLETTIA. Sect. Efulgentes Dunn, sect. nov. Foliola saepissime stipel- lata, subtus ssepius adpresse sericea, nitentes ; Acres subsessiles ; calyx dense velutinus ; corolla alba, glabra, vexillo auriculato. Clavis. Calycis dentes tubo bis breviores ad bis longiores ; foliola subtus efulgentia, ssepius minime stipellata. Foliola coriacea ; ramuli dense velutini coruscansDunn,sp. n. Foliola papyracea ; ramuli glabrescentes. Flores in nodis conferti Ulbrichiana Harms. Flores in ramulis floriferis distincti. Foliolorum venae 4-5-pares fulgens Dunn, sp. n. Foliolorum venge 8-pares vel plures. Bracteolae calycis tubo bis breviores lucens Dunn, sp. n. Bracteolge tubo 4-plo brevioi'es. StipellsB conspicuse hypolampra Harms. Stipellse obsoletae. Paniculi efoliati. Foliola terminalia cuneata subito acu- minata Lecomtei Dunn, sp. n. Foliola terminalia oblongo-lanceolata Dinklagei Harms. Paniculi foliosi Klainei Dunn, sp. n. Calycis dentes breves ; foliola subtus saepissime tenuiter sericea vel glabra, conspicue stipellata. Calyx purpureo-sericeus. Foliola caudata bipindensis Harms. Foliola acuta vel obtuse acuminata. Foliola subtus ferruginea Scott-Ellioti Dunn, sp. n. Foliola subtus glabra... porphyrocalyx Dunn, sp. n. Calyx brunneo-sericeus. Flores 20 mm. longi ; foliola subtus reticulata aureocalyx Dunn, sp. n. Flores 12-15 mm. longi ; foliolorum venae pri- mariae tantum prominentes. Foliola 2-5 cm. longa Warneckii Harms. Foliola 9-20 cm. longa. Foliola subtus dense sericea. Flores in nodis conferti ... ThoUoni Dunn, sp. n. Flores in ramulis floriferis distincti chrysophylla Dunn, sp. n. Foliola subglabra. Flores in nodis conferti macroura Harms. Flores in ramulis floriferis distincti viacrostachya Dunn, nom. n." Sect. CoMPRESso-GEMMAT53 Dunn, sect. nov. Arbores; axilla- rum gemmae compressse, lenticulares ; flores speciosi. • Lonchocarpus macrostachyus Hook. f. = Millettia macrostachya Dunn. AFRICAN MILLETTIAS . 221 Vexillum pilosum brevius quam 25 cm. longum. Foliola ovata. Foliola supra velutina, subtus moUiter hirsuta micans Taub. Foliola subglabra versicolor Welw. Foliola oblonga caffra Meissn. Vexillum tenuiter sericeum, 30-35 mm. longum Sacleuxii Dunn, sp. n. Millettia Stulilmanni Taub. and M. Laurentii De Wild., with glabrous corollas, also belong to this section. As Dr. Harms recently pointed out to the writer, the different forms which have usually t)een referred to M. ferruginea Baker and to M. leucantJia YB,tke, respectively, require revision, and after a careful study of the fine series sent back by German collectors from German East Africa, and by British collectors from Uganda, it appears convenient to recognize a small group of shrubby species related to them in the following manner : — Frutices erecti ; foliola multijuga, stipellata ; vexillum sericeum. Pedicelli calyce vix longiores ferruginea Baker. Pedicelli calyce multo longiores. Flores calyce 2-3 -plo longiores. Vexillum oblatum, carina valde falcata oblata Dunn,sp.n. Vexillum ovatum, carina haud falcata... dura Dunn, sp. n. Flores calyce 4-5-plo longiores ... leptocarpa Dunn, sp. n. Frutices erecti ; foliola paucijuga, stipellata ; vexillum pilosum. Paniculi rachis calyxque breviter adpresse sericei lasiantha Dunn, sp. n. Paniculi rachis calyxque dense velutini leucantha Vatke. The group of West African climbers, of which M. Barteri (Benth. suh Lonchocarpus) is the best known example, may be arranged as follows: — Foliolorum venae impressse. Bracteas longse lineares ; foliola 4 (-3)-juga, hirsuta ; stipellse nuUae vel minutae. Foliola subtus molliter ferruginea Soyauxii Taub. Foliola subtus tenuiter hirsuta bicolor Dunn, sp. n. Foliola subtus adpresse sericea Gagnepaineana Dunn, sp. n. Bracteae breves ; foliola 3-paria, subglabra. Foliola stipellata. Flores 15 mm. longi ; foliola obtusa Goetziana Harms. Flores 10 mm. longi ; foliola acuminata impressa Harms. Foliola exstipellata calabarica Dunn, sp. n. Foliolorum venae haud impressae Barteri Dunn, nom. n.''' • Millettia urophylla Welw. cannot be distinguished" from Lonchocarpus Barteri Benth., which has priority of pubUcation. 222 THE JOURNAL OF BOTANY NEW FOKMS OF MEBTENSIA FEOM THE FAR EAST. By H. Takeda. 1. Meetensia maeitima Don, subsp. nov. asiatica Takeda. Perennis. Caulis robustus, longe prostratus, plus minus ramosus. crassus, carnosus, glaberrimus, glaucus. Folia carnosa, intense glauca, supra punctato-scabra, subtus glaberrima, radicalia longe- petiolata, obovata, late obovatave, obtusa, plus minus mucronata, caulina ovata obovatave, inferiora mediaque in petiolum alatum decurrentia, superiora sessilia, basin versus attenuata, sparsa. Inflorescentia dichotomo-cymosa, initio densa, demum plus minus elongata, plurifiora, foliosa, basi foliis binis oppositis instructa. Flores magni, longe pedicellati, pedicellis iiore longiore, post anthesin nutantibus, inferioribus bracteatis, supremis ebracteatis, bracteis majusculis. Calyx 5 mm. longus, 5-partitus, lobis lineari-lanceolatis mox elongatis, deltoideis, 5-6 mm. longis, glaber. Corolla magna, plus 10 mm. longa, tubo calyce sequanti, limbo dimidio 5-lobo, lobis rotundato-quadrantibus. Stamina anthero 2 mm. longo, filamentis aequanti. Stylus inclusus, 8 mm. longus, staminibus superans. Nuculas 5 mm. longae, 3 mm. latae. Hab. Kamtschatka (Beechey ; Littledale) ; Ochotzk [Turczcmi- notu) ; coast of Manchuria (C. Wilford, 1859, n. 1090 ; Maximoioicz, 1860); Japan (/. Small); Yezo (Maries); Hakodate [Albrecht, 1861; Maximowicz, 1861); Mombetsu (C7. Faurie, 1887, n. 729); Kushiro {M. Uyeda); Kuriles ; Shakotan (H. Takeda, 1909); Eturup (C7. Faurie, 1890, n. 6793); Saghalien [Glehn, 1861). Planta nostra in littore maritimo regionis subarcticse Asije orientalis copiose crescit a planta typica europsea statura valde robustiore, floribus fructibusque majoribus, sty lis calyce duplo staminibusque superantibus, pedicellis fructiferis nutantibus di- versa est. 2. Mertensia rivularis DC. var. nova japonica Takeda. Caulis erectus usque ad 4 dm. altus, glaber, supremo hirtellus, ramosus. Folia basilaria plus minus longe petiolata (plerumque 10-20 cm.), late cordato-ovata, acuta acuminatave, 4-8 cm. longa, 3-7 cm. lata, subtus fere glaberrima, supra adpresse pilosula, margine adpresse ciliata, plus minus glauca, caulina inferiora breviter petiolata, superiora sessilia, ovata, acuminata, basi sub- cordato-rotundata, subtus fere glabra, supra hirtella. Flores longe pedicellati, pedicellis florendi tempore calyce duplo, demum 4-plo longioribus, hirsutis, eb];acteatis. Calyx 5-partitus, lobis lineari-lanceolatis, acutiusculis, trinervis, extus glabris, intus mar- gineque pilosis, 4^-6 mm. longis, post anthesin elongatis 7-8 mm. longis. Corolla magna, 15 mm. longa, tubo albo calyce sesqui- longiore, Hmbum cseruleum subiequanti. Stylus filiformis, corolla aequilongus, nee exsertus. Hab. Yezoj in summitate mentis Yubari [T. Ishikawa, 1896). Kuriles ;' in apice mentis Shakotanyama, insulse Shiko- tan (H. Takeda, 1909) ; Porosu, insulas Eturup [T. Katvakami, 1898), inter Porosu et Tore, ejusd. ins. (S. Yokoyavm, 1893), Otoi, GLOUCESTERSHIRE RECORDS 223 ejusd. ins. {Y. Tanaka, 1895), in apice mentis prope portum Moikeshi, ejusd. ins. {T. Ishikawa, 1890), in montanis inter Naipo et Tore, ejusd. ins. {M. Kambe, 1890) ; Yoshinohama, ins. Urup (Jf. Uckida, 1891), Suisanwan, ejusd. ins. {K. Jimbo, 1891). Nostra planta differt a typica foliis latioribus, subtus glabris nee pilosis, calyce majore, est plerumque in insulis Kurilis nee non in Yezo distributa. GLOUCESTEKSHIRE RECORDS. By THE Rev. H. J. Riddelsdell. The preparation of material for the projected Flora of Glou- cestershire has called my attention to the unsatisfactory nature of Watson's dividing line. The county is split into two vice-counties, East Gloucester and West Gloucester, under the respective Watsonian numbers 33 and 34. It is rather strange that there should be any difficulty with the division, for Watson was by no means a stranger to the county. His herbarium records show that he visited it more than once, notably in the early sixties : though there is no evidence, I believe, that he was in any other part of it than the neighbourhood of the Severn, from Tewkesbury down to Bristol. The difficulty is not that the division is of an arbitrary nature : to some degree, arbitrariness must come in, if it is to be practically useful. But Watson does not appear to have referred either to a large scale map, or to have consulted local botanists on the subject. (The alteration of the county boundaries since his time has made no difference in all this, I believe.) His rule is : " East and West Gloucester are separated by the Thames and Severn Canal, and by the river Severn from the point of junction up to Tewkesbury " {Top. Bot. ed. 2, p. xliii.). Now the Thames and Severn Canal begins by Lechlade, in Gloucestershire, and keeps in the county for a few miles, along a course about east and west. It is never more than about half a mile distant from the Wilts border, and sometimes is within a hundred or two hundred yards of it. About a mile west of Kempsford Church it enters Wilts, and stays there for six or seven miles, though at one point it seems (my evidence is the one inch reduction of the ordnance map) actually to touch the Gloucester- shire boundary. It finally re-enters this county some way beyond Cerney Wick, close to " Wildmoorway Bridge" (of the ordnance map). All goes well now till it joins the Severn at Upper Frami- lode. The division is now marked by the river. But difficulties arise at Gloucester. There the Severn runs in two channels, which convey about equal volumes of water. The question is, Which channel is Watson's line to follow ? It is an important matter, for interesting records are at stake. And once again : about a mile above Haw Bridge the river becomes the county boundary. For a mile or so, Worcester county appears on the right bank, but then comes the parish of Forthampton, on the 224 THE JOURNAL OF BOTANY same bank, and this is Gloucestershire ; and so it goes on till close to Tewkesbury. Here the Watsonian difficulties cease, for the Severn once again becomes the boundary. As work at the Flora has made the question one of practical importance, I propose (1) to disregard the Thames and Severn Canal until it enters Gloucestershire near Cerney Wick. Thus, all that part of the county about Lechlade and Kempsford will be in v.-c. 33 ; it obviously belongs to E. Gloucestershire, and is better there. (2) To take the division along the western channel of the Severn about Gloucester. In Watson's day the traffic ran that way. Now in these latter times the eastern channel has become the chief waterway, as I understand from Mr. A. J. Stephens, of Gloucester. Thus the Hams and Alney Island remain in v.-c. 33. (3) To reckon Forthampton parish in West Gloucester. I would rather it were put to East Gloucester, as it is closely associated with Tewkesbury in record work. But it seems better to deal with Watson's division faithfully. The following list — very imperfect, but I hope fairly accurate — serves to show several omissions from Top. Bot. which are not easy to understand. Publications and other sources of information that are far from obscure seem not to have been quite exhausted. Nowadays so many different publications are used for the record of plants seen — some of them with only a very remote bearing on tiie subject of topographical botany— that it becomes easier to miss information. Both Watson and others — including the present writer — have made slips in the assignment of records between the two vice-coun- ties. A locality " Minchinhampton Common above Brimscombe " was authoritatively placed in East Gloucester. I have seen a statement made by a botanist well acquainted with the county that the Gloucester and Berkeley Canal was in v.-c. 34, whereas the canal is cut into two not very unequal lengths by the Thames and Severn Canal; one — and the larger — part is, therefore, in v.-c. 33. Again, a record from a spot between Bath and Bristol was put to v.-c. 33 by Watson. Necessary corrections in such matters are made in the following list. Dr. G. O. St. Brody, of Gloucester, made a very valuable collection of Gloucestershire plants from 1863 to 1871. The herbarium is now in good order, and safely housed at the Glou- cester Museum. Eeferences to him in the following list are numerous ; for a considerable number of records rest on the authority of his specimens : but there is good reason to accept his bona fides. St. Brody, however, had the disconcerting habit of preserving sheets for species which he only knew of as reported from the county ; the report being usually traceable to some old publication. He seems, in such cases, to have got specimens elsewhere and put them on his sheets. At least, that appears to me the most probable interpretation of the facts : sometimes, for instance, he mentions a species as " reported " from Gloucestershire, but has no specimen in illustration. I do not believe his language can GLOUCESTERSHIRE RECORDS 225 mean that the specimens were sent by people who had found them and sent him a " report " along with them. Mr. White assents to this treatment of St. Brody's " records ": so that some of those given in Journ. Bot. 1907, pp. 407-8, as constituting new records for Gloucestershire have to be cancelled. My own finds are marked with ! Eecords which do not appear in Top. Bot. ed. 2, or Supplement of 1905, are starred. This, however, does not imply that such records are absolutely new ; but there is an obvious convenience in using such a well-known standard of reference. For Bubus, the standard taken is Rogers's Handbook, together with the supplementary list published in Journ. Bot. 1909, pp. 310 and 340. For Hieracium, Linton's Handbook is the standard. Plants probably not native are marked with a dagger (f). Contractions: — Hb. Brody = St. Brody's Herbarium. Hb. Sessions = Sessions' Herbarium, in the Gloucester Museum. Hb. Glos. = Other collections preserved in the Gloucester Museum. Hb. Witts = Herbarium of the Rev. E. F. Witts, kept at Slaughter. The thanks due to various well-known experts for their obliging help are indicated in the course of the list. ThaUctrum minus L. -^-33. Crickley Hill, Hb. Brocly A Hb. Br. Mus. Mr. Linton refers it to var. odoratum (Gren. & Godr.). 34. Summit of great quarry facing the Gully, St. Vincent's Rocks, same variety, Hb. Brocly. — T. flavum L. var. spharocarimm (Lej. & Court.). 33. Banks of the Severn ! Kempsford! \ Anemone ranunoidoides 1j. 33. Haresfield, "near the Vicarage, habitat known for fifty years," Hb. Brody, 1868. — f^- apennina L. 33. Waste ground, Whitcomb, Hb. Brody, Hb. Glos. i Adonis annua L. 33. Specimen from Painswick in Hb. Brody. Banunculus keterophyllus Weber. " Apparently 33 or 34," Top. Bot. Certainly ''34, from Shirehampton and Avonmouth (Mr. White's specimens, see Watson E. C. Bep. 1900-2, and Hb. Br. Mils.). Cerney Wick! "Var. submersus (Hiern) we are inclined to think," H. £ J. Groves. "33. Ponds near Gloucester, Hb. Brody. — B. acris L. var. Borcsanus (Jord.). 33. Moreton in Marsh to Todenham ; "discovered by F. Townsend " {W. W. Neivbould's notes). — Var. rectus (Jord.). 33. Kempsford! 34. Drybrook ! Lydney ! Caltha palustris L. var. Guerangerii (Bor.). 33. Aston-sub- Edge, /. B. Neve. 34. Broadmoor, near Cinderford ! Wood- chester, H. P. Beader. \Eranthis hyemaUs Salisb. 33. Painswick, Painsivick Parish Magazi7ie. 34. Minchinhampton, Mrs. Foord-Kelcey. Aquilegia vulgaris L. =''33. Sheepscombe, Hb. Glos. ; Frith Wood, Hb. Brody ; Pitchcombe Wood, H. P. Beader. Over a small area of wild, remote down, Bourton Downs, near the top ! Barrington Bushes, Hb. Witts. Colesborne Forest, Hb. Trump. Clearly frequent and native in E. Gloucester. \Aconitum Napellus L. 33. Hucclecote, not native, Hb. Brody, Hb. Glos. 34. Wood on May Hill, Hb. Sessions ; much more likely native, but further investigation is needed. 226 THE JOURNAL OP BOTANY \Papaver somniferuvi L. 33. Near Gloucester, Hb. Brody, var. hisindum H. C. Watson. Pields and waste near Upton, Journ. Bot. iv. 121. Kempsford ! Charlton Kings, Hh. Tnmi'p ; smooth-ped. form. 34. Eyton ! Near Newent ! Alvington ! — IP. Bhceas L. var. Pryorii Druce. 33. Kempsford ! 34. Near Newent! — Y&r. strigosuvi (Boenn.). 33. Kempsford! — P. Lecoqii Lamotte. *34. Arlingham, E. M. Day. Cornfields, near Chep- stow, Hb. Brody. Woodchester, H. P. Reader. Newent ! But is this a native species ? Glaucium flavwn Crantz. 34. Beach, New Passage, Hb. Brody. Very abundant there 1910, E. M. Day. \Corydalis solida Hook. 33. Koadside near Bibury Court, /. G. d- E. G. Bake)-. — C. claviculata DC. *34. Walls near Chepstow, 1868, Hb. Brody. I have seen an old specimen, dating from the forties, in the herbarium of the Rev. G. W. Sandys, gathered near Stroud (apparently). Fumaria capreolata L. "33. Fields near Frith Wood, St. Brody in B. E. C. Bep. 1864. Why was this not included in Top. Bot. ? It was passed in the Report. There is no specimen of a Fumaria from this spot in Hb. Brody, and the record cannot be further checked, it seems.— i^. purpurea Pugsley. '■'34. Fields near Beachley, and Lancaut, Hb. Brody ; neither typical, but both named by H. W. P. — F. Borcei Jord. *33. Waste, near Glou- cester, Hb. Brody ; this is var. muraliformis Clav., according to H.W. P.—F. Bastardi Bor. [33. Garden at Campden ! but, I believe, introduced. *34. Lydney, W. H. Purchas, in Hb. Br. Mus. — F. officinalis L. Rampant form at 33, Kempsford ! 34. Nailsworth ! Bromsberrow ! — F. Vaillantii Lois. *33. Kemps- ford ! Fields on Leckhampton Hill, Hb. Brody. St. Brody records both this and F. parviflora Lam. from Leckhampton Hill ; but his herbarium shows that F. Vaillantii was the plant in both cases. F. parviflora must therefore come out of Top. Bot., unless other- wise vouched for. Badicula N-aquaticum Rendle and Britten var. siifolia Druce. 34. Frampton on Severn! Slimbridge H. P. Beader. — Var. microphylla Druce. 34. Broadmoor. Barbarea vulgaris Ait. var. decipiens Druce. 33. Hailes ! and not uncommon. 34. By the River Leadon at Dymock and Upleadon ! — B. stricta Andrz. 34. Once, by the River Leadon, between Upleadon and the River Ell. The Sharpness record {Top. Bot. Supp.) is suspect ; it comes from "the Docks. — \B. intermedia Bor. 33. Ballast heaps, Gloucester, Hb. Brody. jArabis alpina L. 34. Near the White House, one and a half miles from Bristol, going to Kingswood, Hb. Banks, 1773. Hb. Wats, has a specimen from the same spot, 1837. — iA. albida Stev. 33. Walls, Bowbridge, near Stroud! — A. glabra Bernh. " 33 or 34 " ? Top. Bot. •■-34 certainly. At Bromsberrow Lees, Bot. of Malvern Hills ; plenty there, H. H. Knight, 1910. Specimen in Hb. Br. Mus. 33 probably ; waste places near Gloucester, N. B. G. ; but perhaps not native there. Cardamine pratensis Li. wa,v ^ewtoto (Schultes.). 34. Lancaut! GLOUCESTERSHIRE RECORDS 227 — C. flexuosa With. "33. Eoadside marsh, Stonehouse ! Ando- versford ! Cirencester, W. J. Greemvoocl, spn. ! — C. hirsuta x lyratensis [Hayneana Welw.). 34. Field at Anwards Farm, Tidenham, 1908 ! I" have cultivated the plant, and found it pretty constant ; it spreads freely, not by seeds, but by broken bits of leaves, &c., which take root where they fall. I think the name must stand. -\Alyssu7n vindobonense Beck. 33. Waste ground, Gloucester Docks, Hb. Brocly. St. Brocly named it A. incanum L. ; but I am satisfied that it must go to A. vindobonense. — f^. alyssoides L. 33 or 34. Fields near Stonehouse, Hb. Brody. 34. Fields near Dursley, St. Brody, in Journ. Bot. iv. 121. — ]'A. maritimum Lam. 33. Waste ground, Gloucester Docks, Hb. Brody. Draba muralis L. 34. Quarry at Henbury, Hb. Brody ; see Phytologist v. 215. Perhaps the character of the locality made Watson suspect the record ; but the species occurs in Som. N. Erophila pracox DC. *33. Bibury, /. G. d E. G. Baker. Fairford ! Witcombe, Hb. Glos. Cochlearia officinalis L. 34. Shore at Beachley, Hb. Brody. St. Brody calls it anglica, but wrongly. — C. danica L. *34. Sharpness Point, 1864, Hb. Brody. W. A. S. has it also from the opposite bank at Lydney. Sisymbrium officinale L. var. leiocarpiim DC. 34. Arlingham 1 — f S. orientale L. 33. Cheltenham, Trumj}, spn. ! Gloucester Docks, 3Irs. Foord-Kelcey, spn. ! Fairford, J. Taylor, spn. ! StanwayHill! 34. Lydney! — \S. pannonicicm J acq. 33. Glou- cester Docks, Hb. Brody. Andoversford, Trump, spn. ! Fairford, Hb. Br. Mus. \ErysMmim orientale Mill. 33. Gloucester Docks, Hb. Brody. Kempsford, Bev. W. Butt. 34. Near Tetbury, Mrs. Griffith.— \E. repandum L. 33. Near the Docks, Gloucester, Hb. Brody. — ^E. hieracifolium L. 33. Llanthony Priory, Gloucester, K. B. Black- burn, spn. ! \Gamelina saliva Crantz. 33. Docks, Gloucester, with var. fietida (Fr.) Hb. Brody. 34. Newent ! Lydney Canal!— fC. sylvestris Wallr. 33. Docks, Gloucester, Hb. Brody. Brassica arvensis 0. Kuntze var. orientalis (L.). 33. The common form about Ford and Guiting on the top of the Cots- wolds ! — t J5. PoUchii Sch. & Spenn. 33 or 34. Gravel pit, Stroud, Hb. Br. Mus. — iB. dissecta Boiss. 33. Docks, Gloucester, Hb. Brody. Diplotaxis tenuifolia DC. 33. Old walls, Campden, /. B. Neve, and wall near Cathedral, Gloucester, Phytologist iv. 1053. — D. ■muralis DC. var. Babingtonii Syme. 33. Gloucester! Chelten- ham ! Leckhampton Station, Mrs. Foord-Kelcey, spn. I 34. Newent! Over! Lydney! Tutshill ! Eyeford ! Nailsworth, Mrs. Foord-Kelcey, spn. ! Coronopus didymus Sm. "33. Kempsford ! Duntisbourne Abbots, Bev. W. Butt. Lcpidium latifolium L. •■-34. Westbury on Severn, Hb. Brody ; and E. M. Day ! In Hb. Br. Mus. is a specimen from Garden 228 THE JOURNAL OP BOTANY Cliff, Gloucestershire, but I do not know where this locality is. — f L. sativum L. 33. Weston-sub-Edge ! 33 or 34. Waste, about Stroud, H. P. Beader. — L. hetero2:)hylkwi Benth. var. canescens Gren. & Godr. ^'34. And certainly native, among gorse on a sandy heath at Bromsberrow ! Lydney ! Kocks, Pighole, Tiden- ham, Hb. Brody. — ]L. perfoUatum L. 33. Cheltenham, Trump, spn. ! Docks, Gloucester, Hh. Brody. Canal, near Gloucester, K. B. Blackburn, spn. ! — f L. Draha L. 33. Maisemore Ham and Sandhurst ! Cheltenham ! Gloucester Docks ! Leckhamp- ton Hill, Trump, spn. ! 34. Lydney ! Thlaspi perfoUatum L. =''=34. The Tetbury Road locality, as described in Journ. Bot. ix. 17, is partly in this division ; verbal descriptions received from observers on the spot support this conclusion. (I have received specimens from v.-c. 34, since writ- ing the above.) Iberis amara L. ^:'33. Cornfields, Northleach, Hb. Brody ; perhaps native here. Leckhampton Station, Mrs. Foord-Kelcey , spn. ! 34. Minchinhampton Common, H. M. Middleton, spn. ! Hutchinsia petrcea Br. 34. Penmoel cHffs, Tutshill ! Cold- well Rocks, Hb. Watson. Cakile maritima Scop. *34. Sand near Beachley, Hb. Brody. It occurred as a casual at 33. Kempsford, in a garden. ■\Bapliamis sativus L. 33. Canal bank, Gloucester, K. B. Blackburn, spn. ! \Neslia 2)aniculata Desv. 33. Docks, Gloucester, Hb. Brody. \Cleome serrulata Pursh. 33. Canal side, Gloucester, K. B. Blackburn, spn. ! \Beseda alba L. 33. Docks, Gloucester, Hb. Brody. Viola odorata L. var. imberbis Leight. 34. Highnam, Hb. Brody. — Var. subcarnea (Jord.). Seems frequent in 33, from Cheltenham to Kempsford ! Bibury, J. G. d- E. G. Baker. — Var. dumetorum (Jord.). Bibury, J. G. ct E. G. Baker. Fairford, G. C. Druce.— V. Jiirta h. var. propera Gillot. 33. Near Chalford ! with var. inconcinna Briq. — Var. Foudrasi (Jord.). 33. Fairford, G. C. Druce. — V. hirta x odorata {V. permixta Jord.). 33. Birdlip Wood, Hb. Brody. Fairford, G. G. Druce. — V. arvensis Murr. var. segetalis (Jord.). 33. Fields, near Gloucester, Hb. Brody. Near Ford ! PolygalaoxypteraB.eichh. '■■34. Pauntley ! Woolaston Chase ! Edge End ! Milkwall ! Gatcombe Wood, W. M. Bogers. Dianthus Armeria L. =''33. Field between Cirencester Road and Charlton Common, Cheltenham, L. Bichardson, spn. ! Waste ground on canal, Gloucester, Hb. Brody. \ Saponaria Vaccaria L. 33. Near Bourton on Hill, quarry, L. Mott, spn. ! Waste ground, Gloucester, Hb. Brody. 34. Sharpness, Hb. Trump ! Silene latifolia Rendle & Britten var. puberula Jord. 33. On the hills near Cheltenham, Buckman's Flora of Cheltenham. Snowshill, on downs ! Bourton Downs ! 34. Lydney ! Nails- worth ! Frampton Mansell ! Near Tetbury ! — S. maritima With. "33. Docks, Gloucester, Hb. Brody. •'34. Rocks by the Wye at GLOUCESTERSHIRE RECORDS 229 Lancaut ! — .S'. latifolia x viaritima. 33. Almost certainly one of the specimens on St. Brody's sheet of S. maritima at Gloucester is this hybrid, which I thought to have found with the type at 34. Lancaut; but this is far less certain. — \S. dicliotoma Ehrh. 33. Near Guiting hill Farm, in two forms, large and small flowered. — S. anglica L. *33. Campden, J. B. Neve. "■'34. Lydney, scarce ! — Var. quinquevulnera (L.). 33. Brockworth, Hb. Glos. : near Docks, Gloucester, Hb. Brody. — S. noctiflora L. "'■■33. Kempsford ! Docks, Gloucester, Hb. Brody. ••■34. Cornfields, Woodchester, H. P. Beader. Lychnis alba x dioica. 33. Campden, /. B. Neve. Guiting ! 84. Lydney ! Hewellsfield ! Moenchia erecta Gaertn. Mey. & Schreb. 34. Chase End Hill ! Broadmoor ! Serridge Green ! Probably fairly common in open grassy pa.rts of Forest of Dean. Stellaria neglecta Weihe. "33. Var. decipiens E. S. M. at Haresfield, Hb. Br. Mus. 34. Newent, Upleadon, Pauntley, Bromsberrow, frequent ! — S. iMlustris Eetz. "34. Marsh, near Mitcheldean, 1864, Hb. Brody. This is correctly named. Arenaria tenuifolia L. var. hybrida (Vill.). 34. Minchinhamp- ton Common, native ! — A. leptoclados Guss. Frequent in 33 and 84 ! — A. peploides L. "34. Sand near Beachley, Hb. Brody. Sagina maritima Don. "34. Banks, New Passage, Hb. Brody. — S. ciliata Fr. *34. Chase End Hill !—